rjones (Thu, 13 Apr 2017 18:34:56 GMT):
nage

nage (Thu, 13 Apr 2017 18:54:17 GMT):
Welcome back to the #indy channel, and thanks @rjones for fixing the join issues!

drummondreed (Thu, 13 Apr 2017 19:18:25 GMT):
Has joined the channel.

drummondreed (Thu, 13 Apr 2017 19:18:46 GMT):
Great, thanks to @rjones for the fixes

rjones (Thu, 13 Apr 2017 19:19:03 GMT):
sure thing!

anttikettunen (Thu, 13 Apr 2017 19:50:17 GMT):
Has joined the channel.

stevetolman (Thu, 13 Apr 2017 19:59:57 GMT):
Has joined the channel.

windley (Thu, 13 Apr 2017 20:18:04 GMT):
Has joined the channel.

devin-fisher (Thu, 13 Apr 2017 20:25:17 GMT):
Has joined the channel.

danielhardman (Fri, 14 Apr 2017 02:01:36 GMT):
Has joined the channel.

farooq_m_khan (Fri, 14 Apr 2017 05:37:44 GMT):
Has joined the channel.

peacekeeper (Fri, 14 Apr 2017 12:28:34 GMT):
Has joined the channel.

TelegramSam (Fri, 14 Apr 2017 18:20:05 GMT):
Has joined the channel.

JonathanLevi (Fri, 14 Apr 2017 20:43:33 GMT):
Has joined the channel.

sstone1 (Mon, 17 Apr 2017 08:41:50 GMT):
Has joined the channel.

dhuseby (Mon, 17 Apr 2017 14:40:44 GMT):
Has joined the channel.

VipinB (Tue, 18 Apr 2017 14:33:51 GMT):
Has joined the channel.

cbruguera (Thu, 20 Apr 2017 21:49:15 GMT):
Has joined the channel.

amundson (Mon, 24 Apr 2017 17:20:23 GMT):
Has joined the channel.

tkuhrt (Tue, 25 Apr 2017 15:23:15 GMT):
Has joined the channel.

stanliberman (Wed, 26 Apr 2017 04:19:25 GMT):
Has joined the channel.

LoveshHarchandani (Thu, 27 Apr 2017 06:46:33 GMT):
Has joined the channel.

darrell.odonnell (Mon, 01 May 2017 12:25:48 GMT):
Has joined the channel.

rjones (Mon, 01 May 2017 18:32:03 GMT):
Has left the channel.

prashiyn (Tue, 02 May 2017 06:45:00 GMT):
Has joined the channel.

SarahMills (Tue, 02 May 2017 15:51:40 GMT):
Has joined the channel.

rahulhegde (Wed, 03 May 2017 02:04:28 GMT):
Has joined the channel.

VipinB (Wed, 03 May 2017 14:04:07 GMT):
@here Identity WG call today at 12 EDT. Details in wiki link https://wiki.hyperledger.org/groups/identity/identity-wg. Paper link: https://docs.google.com/document/d/1ExFNRx-yYoS8FnDIUX1_0UBMha9TvQkfts2kVnDc4KE/edit?usp=sharing Agenda: Call for volunteers Splitting up workload and discuss timelines: a.Bridge to architecture WG paper – Nathan George (since he had already started on this) b.TEEs & HSMs – Steven from Rivetz c.Jan Camenisch (already provided some feedback)- ZK proofs and CL signatures- maybe something more concrete d. Benjamin Bollen – Monax-Identity e. Leonard Edwin maybe Iroha-Link is f. Daniela Merella - She can look at the paper and contribute g. Fabric CA team to provide some input on Fabric CA h. STL EPR – (End Point Registry) https://docs.google.com/document/d/1gWlbsKtgApOTm1E5XJsw2J3U6AvIRxujWiG-vm6ELOM/edit?usp=sharing EPID as identity repository. i. Sovrin-Identity Model- Drummond j. DiD->DDO model-Drummond k. X.509 (Version 3.0), X.525 ****-need volunteer l. Verifiable claims - Nathan/Sean/Drummond m. Sketch out Identity Interface- Vipin We have to recognize the fact that this conference season and people are very busy. In fact this can be a good thing, since the latest ideas garnered at conferences can be incorporated into our paper.

himansri (Wed, 03 May 2017 14:38:36 GMT):
Has joined the channel.

schwentker2 (Wed, 03 May 2017 19:17:32 GMT):
Has joined the channel.

turmewr3ck (Wed, 03 May 2017 21:31:34 GMT):
Has joined the channel.

snowy13 (Thu, 04 May 2017 16:38:39 GMT):
Has joined the channel.

JoVercammen (Thu, 04 May 2017 20:18:26 GMT):
Has joined the channel.

sandroku63 (Fri, 05 May 2017 09:01:52 GMT):
Has joined the channel.

abaumhof (Fri, 05 May 2017 12:26:29 GMT):
Has joined the channel.

paulmersky (Sat, 06 May 2017 07:26:04 GMT):
Has joined the channel.

cbf (Sun, 07 May 2017 19:48:03 GMT):
Has joined the channel.

mpiekarska (Mon, 08 May 2017 09:40:21 GMT):
Has joined the channel.

tyler (Mon, 08 May 2017 18:56:15 GMT):
Has joined the channel.

nage (Mon, 08 May 2017 19:57:37 GMT):
We have been tracking Sovrin Development Events on a shared google calendar here: https://www.google.com/calendar/embed?src=c292cmluLm9yZ18zcjdoa3V1bzZnc2puMmNwdGNkY2dubG9rMEBncm91cC5jYWxlbmRhci5nb29nbGUuY29t

nage (Mon, 08 May 2017 19:58:42 GMT):
We put conferences where you can meet and talk with TGB members, maintainers and developers here when they are representing Sovrin in some way, along with development events where folks are hacking on Indy. Let me know if you'd like to add something (a bunch of folks at sovrin.org have rights to add things to this calendar)

peacekeeper (Mon, 08 May 2017 22:11:11 GMT):
Hello @nage I travel quite often to events where I can speak about Sovrin and meet other people interested in it. I'd be happy to just announce it here and you can add it to the calendar?

peacekeeper (Mon, 08 May 2017 22:11:50 GMT):
E.g. this week a number of us are at the European Identity & Cloud Conference with several Sovrin-related talks and sessions: https://www.kuppingercole.com/events/eic2017/

nage (Mon, 08 May 2017 22:19:31 GMT):
We should get those on the calendar. I suggest we add this as an agenda item for TGB meetings so we don't lose track of it.

avkrishnan (Tue, 09 May 2017 04:49:16 GMT):
Has joined the channel.

theofilis (Tue, 09 May 2017 20:37:48 GMT):
Has joined the channel.

theofilis (Tue, 09 May 2017 20:43:23 GMT):
Has left the channel.

drummondreed (Tue, 09 May 2017 21:09:38 GMT):
+1

tiennv (Fri, 12 May 2017 17:22:03 GMT):
Has joined the channel.

stephencleary (Fri, 12 May 2017 20:26:02 GMT):
Has joined the channel.

HansDeLeenheer (Mon, 15 May 2017 09:48:49 GMT):
Has joined the channel.

peterfr (Mon, 15 May 2017 11:25:12 GMT):
Has joined the channel.

czar0 (Mon, 15 May 2017 20:08:57 GMT):
Has joined the channel.

Donald Liu (Wed, 17 May 2017 01:59:53 GMT):
Has joined the channel.

amigus (Wed, 17 May 2017 23:39:30 GMT):
Has joined the channel.

dhuseby (Thu, 18 May 2017 14:14:05 GMT):
@drummondreed who is working on setting up the Indy wiki page on wiki.hyperledger.org and starting the CII badge application?

drummondreed (Thu, 18 May 2017 14:14:59 GMT):
Sean and Nathan and Tyler would be coordinating that. Do you want to ping them directly or do you want to DM them here?

danielhardman (Thu, 18 May 2017 15:52:51 GMT):
^^^ @nage

nage (Thu, 18 May 2017 15:53:56 GMT):
@Sean_Bohan is working on the wiki pages (at least it is on his task list as conference season settles down)

Sean_Bohan (Thu, 18 May 2017 15:53:56 GMT):
Has joined the channel.

nage (Thu, 18 May 2017 15:55:20 GMT):
Our migration plan right now is to move Jira tickets over to Hyperledger first, then move github repositories and get through any CLA work before starting in on the CII badge issues (I think we already meet most of the CII requirements, which is why I haven't prioritized it). I'm happy to move it up the priority list if it is helpful to you @dhuseby

dhuseby (Thu, 18 May 2017 15:57:23 GMT):
Thanks. Just FYI, I don't think you have to switch to gerrit and Jira if you don't want too. Iroha uses github for all of that and they were just approved for graduation out of incubation by the technical steering committee this morning.

dhuseby (Thu, 18 May 2017 15:57:41 GMT):
@Sean_Bohan ^^^

dhuseby (Thu, 18 May 2017 15:58:12 GMT):
I take their graduation as tacit approval of using github.

dhuseby (Thu, 18 May 2017 15:59:39 GMT):
I certainly have no security concerns related to using github. But I do have concerns about gerrit. It is turning out to be an anti-pattern. Gerrit doesn't do branching well and that's the main resistance to the adoption of my proposed branching strategy.

dhuseby (Thu, 18 May 2017 16:00:06 GMT):
Iroha uses my proposed branching strategy more or less. And they are on github.

nage (Thu, 18 May 2017 16:16:19 GMT):
We are already using Jira internally, so we're thinking it makes sense to open that up to the community using the hyperledger Jira server. We are also planning on sticking with github, rather than move over to gerrit (our community is already very familiar with github).

Balneum (Fri, 19 May 2017 06:48:56 GMT):
Has joined the channel.

SanketPanchamia (Mon, 22 May 2017 04:14:56 GMT):
Has joined the channel.

alex.nagelberg (Tue, 23 May 2017 16:11:14 GMT):
Has joined the channel.

danielhardman (Tue, 23 May 2017 19:37:21 GMT):
I thought the community would be interested in the security review of Sovrin, being done by a third-party consultancy. http://engageidentity.com/sovrin-foundation-partnership/

danielhardman (Tue, 23 May 2017 19:37:33 GMT):
^^ @dhuseby

dhuseby (Wed, 24 May 2017 06:53:16 GMT):
Oooh yes. Thanks!!

ashcherbakov (Wed, 24 May 2017 09:49:54 GMT):
Has joined the channel.

mzk-vct (Wed, 24 May 2017 09:57:01 GMT):
Has joined the channel.

alexander.shekhovcov (Wed, 24 May 2017 09:57:20 GMT):
Has joined the channel.

andrey.goncharov (Wed, 24 May 2017 10:02:40 GMT):
Has joined the channel.

andkononykhin (Wed, 24 May 2017 10:50:19 GMT):
Has joined the channel.

krw910 (Wed, 24 May 2017 13:58:41 GMT):
Has joined the channel.

krw910 (Wed, 24 May 2017 17:56:02 GMT):
@LoveshHarchandani here is a capture of the error: rin_node[8930]: await self._serviceStack(self.age) rin_node[8930]: File "/usr/lib/python3.5/asyncio/coroutines.py", line 105, in __next__ rin_node[8930]: return self.gen.send(None) rin_node[8930]: File "/usr/local/lib/python3.5/dist-packages/stp_zmq/zstack.py", line 639, in _serviceStack rin_node[8930]: self._receiveFromListener(quota=self.listenerQuota) rin_node[8930]: File "/usr/local/lib/python3.5/dist-packages/stp_zmq/zstack.py", line 599, in _receiveFromListener rin_node[8930]: self._verifyAndAppend(msg, ident) rin_node[8930]: File "/usr/local/lib/python3.5/dist-packages/stp_zmq/zstack.py", line 580, in _verifyAndAppend rin_node[8930]: self.rxMsgs.append((msg.decode(), ident)) rin_node[8930]: UnicodeDecodeError: 'utf-8' codec can't decode byte 0xae in position 2: invalid start byte

krw910 (Wed, 24 May 2017 17:56:32 GMT):
Sorry it did not capture everything. Just log into one of the nodes and run status on sovrin-node

LoveshHarchandani (Wed, 24 May 2017 17:56:42 GMT):
@krw910 Thanks

LoveshHarchandani (Wed, 24 May 2017 18:04:09 GMT):
The problem is that we assume that incoming data is going to be utf-8, we can make a rule that we only accept utf-8, wondering how to let the client know, using reqnack wont work since the code will not reach at that point, so for now just logging an error message as the node after handling such errors

devin-fisher (Wed, 24 May 2017 18:40:34 GMT):
@here Many of the developers at Evernym are participating in a shakedown of the Indy Node. So, there should be a fair amount activity around this. Additionally, there will be a lot of activity around this in the Indy Jira around the shakedown. Hopefully this will expose a lot bugs before the Sovrin network goes live with Indy. https://en.wikipedia.org/wiki/Shakedown_(testing)

LoveshHarchandani (Wed, 24 May 2017 18:57:00 GMT):
@krw910 Its fixed, here is the fix https://github.com/evernym/stp/commit/fd68242d94e86d958d94108747a9b4581421d0bd#diff-0236f66d0eed2d1c7c0fdf4394ac91efR580 and here is the test https://github.com/evernym/stp/commit/fd68242d94e86d958d94108747a9b4581421d0bd#diff-7ffd3624d890e8f700882833e216fc4cR74

nage (Wed, 24 May 2017 19:16:55 GMT):
Minimum Go-Live shakedown

andrey.goncharov (Wed, 24 May 2017 20:29:44 GMT):
@here SOV-980 resolved. Found an intermittent test issue for plenum. Logged a ticket for it https://jira.hyperledger.org/browse/INDY-11

stevetolman (Wed, 24 May 2017 20:44:52 GMT):
Nice work

stevetolman (Wed, 24 May 2017 20:45:43 GMT):
Is that new issue a result of the fix, or is it something that was pre-existing?

devin-fisher (Wed, 24 May 2017 22:16:57 GMT):
@stevetolman I logged INDY-21. I think it low priority but it was causing me issues for awhile.

stevetolman (Wed, 24 May 2017 22:20:50 GMT):
Okay, thanks

stevetolman (Wed, 24 May 2017 22:48:51 GMT):
@nage @devin-fisher - checkpoint 3 time: https://zoom.us/j/404583509

nage (Thu, 25 May 2017 01:55:44 GMT):
@here for anyone interested in the Sovrin Ledger work, we have an Indy Node WG meeting tomorrow (Thursday May 25th at 9:00 Mountain Time (UTC-6). We are going to be discussing input validation, alpha/provisional network status, the shakedowns and hardening efforts we are doing (and how you could help). Join us at https://zoom.us/j/730665163 if you're interested.

farooq_m_khan (Thu, 25 May 2017 10:14:11 GMT):
We should probably have a concept of Named Pools. so a single file “pool_transactions_sandbox” can contain details of multiple different sovrin networks with one of them as default so a user can easily change from one sovrin pool to another, I am not sure if shakedown Pool1 has been fixed so I wanted to switch to pool2 and this cannot be done without some manual file editing which can be made more usable

farooq_m_khan (Thu, 25 May 2017 10:14:11 GMT):
We should probably have a concept of Named Pools. so a single file “pool_transactions_sandbox” can contain details of multiple different sovrin networks with one of them as default so a user can easily change from one sovrin pool to another, I am not sure if shakedown Pool1 has been fixed so I wanted to switch to pool2 and this cannot be done without some manual file editing which can be made more usable, if i could just edit the file pool_transactions_sandbox and add details of second network give it a name and use that when launching the sovrin cli

farooq_m_khan (Thu, 25 May 2017 12:27:51 GMT):
Shakedown Pool1 and Pool2 both seem to be down connection attempts dont succeed

krw910 (Thu, 25 May 2017 12:47:23 GMT):
Ok, I will take a look and reset Pool 1. Has anyone been working on Pool 2 and Pool 3 that might know why they are down?

krw910 (Thu, 25 May 2017 13:03:00 GMT):
Pool 3 is working

krw910 (Thu, 25 May 2017 13:03:23 GMT):
Make sure you are using the correct pool_transactions_sandbox information

KhageshSharma (Thu, 25 May 2017 13:06:59 GMT):
Has joined the channel.

ashcherbakov (Thu, 25 May 2017 13:13:26 GMT):
A pattern to be used when we close bugs/tickets: Problem reason: - list of reasons Changes: - list of changes Committed into: github commit ref build version Risk factors: Nothing is expected. Risk: Low Covered with tests: Test ref on github Recommendations for QA (optional):

krw910 (Thu, 25 May 2017 14:04:45 GMT):
As mentioned Pool 3 is running. With Pool 1 and Pool 2 being looked into I will be getting a Pool 4 ready as a backup to Pool 3.

devin-fisher (Thu, 25 May 2017 14:09:55 GMT):
What version of the code is shakedown running against? stable or master?

devin-fisher (Thu, 25 May 2017 14:22:33 GMT):
^^^ @krw910

krw910 (Thu, 25 May 2017 14:23:02 GMT):
We are only running against master

jlaw (Thu, 25 May 2017 14:25:23 GMT):
Has joined the channel.

devin-fisher (Thu, 25 May 2017 14:27:15 GMT):
cool

jlaw (Thu, 25 May 2017 14:29:02 GMT):
@ashcherbakov Are you suggesting that template be filled out and put as comment of a Jira bug?

nage (Thu, 25 May 2017 14:51:04 GMT):
Ledger WG meeting in 10 minutes https://zoom.us/j/730665163

ashcherbakov (Thu, 25 May 2017 14:54:32 GMT):
@jlaw 1 Yes, everytime one closes a ticket in Jira, a comment with the information above should be pasted

nage (Thu, 25 May 2017 15:00:46 GMT):
Ledger/Indy-Node WG starts now!

devin-fisher (Thu, 25 May 2017 15:25:15 GMT):
@krw910, how much memory do the nodes have?

krw910 (Thu, 25 May 2017 15:33:17 GMT):
@devin-fisher they should each be m4.large in Amazon which is: Model vCPU Mem (GiB) SSD Storage (GB) Dedicated EBS Bandwidth (Mbps) m4.large 2 8 EBS-only 450

krw910 (Thu, 25 May 2017 15:33:51 GMT):
@devin-fisher CPU 2 -- Mem 8GB -- Bandwith 450 Mbps

devin-fisher (Thu, 25 May 2017 15:34:04 GMT):
thanks

krw910 (Thu, 25 May 2017 15:34:39 GMT):
This is close to what the governance board requires.I think the memory is correct for what we are requiring.

devin-fisher (Thu, 25 May 2017 15:37:34 GMT):
ok, I'm going to try and send a single very large txn. Anyone object let me know.

krw910 (Thu, 25 May 2017 15:38:01 GMT):
going to pool 3?

devin-fisher (Thu, 25 May 2017 15:38:04 GMT):
yeah

krw910 (Thu, 25 May 2017 15:38:34 GMT):
ok

smithsamuelm (Thu, 25 May 2017 15:46:49 GMT):
Has joined the channel.

smithsamuelm (Thu, 25 May 2017 15:54:14 GMT):
Hi just joining.

nage (Thu, 25 May 2017 16:01:50 GMT):
Welcome @smithsamuelm!

krw910 (Thu, 25 May 2017 16:18:48 GMT):
We have a Pool 4 now if Pool 3 goes down. I am resetting Pool 1 and Pool 2

denofernandes (Thu, 25 May 2017 18:10:27 GMT):
Has joined the channel.

jlaw (Thu, 25 May 2017 19:17:52 GMT):
FYI, on a call with Lovesh re: election protocol on INDY-13 in case anyone wants to listen in.

krw910 (Thu, 25 May 2017 19:40:44 GMT):
We are working out of Pool 3, and I have reset Pools 1, 2, and 4. I have setup the GST agents Faber, Acme, and Thrift for each of them so if you go through the GST you don't have to setup the agents. I have also turned on debug level in the logs on all the nodes for all the pools. If Pool 3 fails move to Pool 4 then Pool 1. Please let me know if the pool fails and you know why. We need to be able to log the steps in the ticket.

devin-fisher (Thu, 25 May 2017 20:00:05 GMT):
@alexander.shekhovcov You still online?

devin-fisher (Thu, 25 May 2017 21:30:41 GMT):
@krw910 Can one access one of the machines in the test pool. Mostly I want to look at htop to see if there is a spike in activity.

krw910 (Thu, 25 May 2017 22:01:43 GMT):
Sorry I missed your message. Yes in the shakedown folder I have a AWS spreadsheet that has all the node information and a zip file that has all the .pem and .ppk files by region, since they are all named the same. Let me slack you the location. Use "ubuntu@" to log in.

prashiyn (Sat, 27 May 2017 09:53:53 GMT):
Hi all, joined the channel a few days back. Where can I start reading up about Indy and setting it up on my dev?

apoikola (Sat, 27 May 2017 15:26:51 GMT):
Has joined the channel.

compleatang (Sun, 28 May 2017 16:44:41 GMT):
Has joined the channel.

srottem (Mon, 29 May 2017 20:59:17 GMT):
Has joined the channel.

axel (Tue, 30 May 2017 09:15:52 GMT):
Has joined the channel.

danielhardman (Tue, 30 May 2017 19:07:50 GMT):
@prashyin: I apologize for the reply being delayed by Memorial Day weekend.

danielhardman (Tue, 30 May 2017 19:07:50 GMT):
@prashiyn : I apologize for the reply being delayed by Memorial Day weekend.

danielhardman (Tue, 30 May 2017 19:17:38 GMT):
Can you share with us your OS and dev platform, so we can provide some specifics?

danielhardman (Tue, 30 May 2017 19:20:00 GMT):
You may find these instructions helpful, in the meantime. They are written from the perspective of Sovrin rather than Indy, but that's just a branding issue. Most of the instructions are just as relevant. https://docs.google.com/document/d/1CyggP4nNPyx4SELNZEc2FOeln6G0F22B37cAVtB_FBM/edit

prashiyn (Wed, 31 May 2017 04:40:31 GMT):
@danielhardman I would build it on ubuntu 14.04 or a 16.04. Will build a bare bones ubuntu machine on azure and start from scratch :)

prashiyn (Wed, 31 May 2017 04:41:21 GMT):
Will get build an anaconda python environment. We use anaconda across our organization

turmewr3ck (Wed, 31 May 2017 07:22:46 GMT):
@danielhardman, so that instruction is partially based on Python virtual environment, which I've not tried yet. For the Sovrin client I've so far just used Docker to build a runnable image. Some hacks are required on the Dockerfile to get it working ... so I was wondering: Is the the Python virtual environment you prefer currently, or will Docker eventually be supported as well? (Perhaps I ought to send patches to Dockerfile?) :)

turmewr3ck (Wed, 31 May 2017 07:22:46 GMT):
@danielhardman, so that instruction is partially based on Python virtual environment, which I've not tried yet. For the Sovrin client I've so far just used Docker to build a runnable image. Some hacks are required on the Dockerfile to get it working ... so I was wondering: Is the the Python virtual environment what you prefer currently, or will Docker eventually be supported as well? (Perhaps I ought to send patches to Dockerfile?) :)

turmewr3ck (Wed, 31 May 2017 08:08:21 GMT):
On a different topic: A week ago there was the announcement of the Sovrin Foundation and Engage Identity partnership, with the purpose of conducting a technical review of the current security architecture, and a report coming this summer. Will Indy benefit from this somehow? More specifically, will it involve secure coding practices for the source code base? Can you already now talk about which secure coding practices you use for the Python code base (or plan for the future), as this concerns me a bit ... simply just because my lack of knowledge of what best practices exist for crypto development in Python. http://www.windley.com/archives/2017/05/sovrin_in-depth_technical_review.shtml

danielhardman (Wed, 31 May 2017 14:57:23 GMT):
@turmewr3ck, re docker: we want to support docker but haven't had the bandwidth to look at it for a while. Patches would be welcome. Re. Engage Identity, this work is absolutely going to benefit Indy, although some of what they're analyzing (e.g., governance) is specific to Sovrin's instantiation of Indy code. Here are a couple links that will show you some early thinking on these topics. https://docs.google.com/document/d/1Kv2SCsd-24iQPXjzutCQluTOGWWvbCKdUOqe-cISfq0/edit and https://docs.google.com/document/d/1kRCbPk3zB0qR5WOkgjUJXwD1pzYZCI0jDjqiwKhZkxs/edit?usp=sharing

gdinhof (Fri, 02 Jun 2017 06:02:31 GMT):
Has joined the channel.

tkuhrt (Fri, 02 Jun 2017 19:04:48 GMT):
Hi, Indy team. I created a shared calendar for all Hyperledger community meetings. You can find it here: http://bit.ly/HyperledgerCal. Use the &ctz=_TZ_ option to specify your time zone. See the TZ column for list of options: https://en.wikipedia.org/wiki/List_of_tz_database_time_zones#List. I would really love to add any Indy meetings to this calendar.

drummondreed (Fri, 02 Jun 2017 23:42:11 GMT):
Thanks, that calendar is going to be VERY useful.

VipinB (Sun, 04 Jun 2017 13:21:02 GMT):
@here we need to improve the Indy landing page from HL wiki. Today it takes you to a page under construction. Let me know if I can help in any way.

Captain63Dragon (Sun, 04 Jun 2017 19:08:09 GMT):
Has joined the channel.

dsurnin (Mon, 05 Jun 2017 09:49:03 GMT):
Has joined the channel.

ashcherbakov (Mon, 05 Jun 2017 09:49:36 GMT):
https://jira.hyperledger.org/browse/INDY-147

dsurnin (Mon, 05 Jun 2017 10:10:00 GMT):
https://jira.hyperledger.org/browse/INDY-148

aleksey-roldugin (Mon, 05 Jun 2017 10:14:49 GMT):
Has joined the channel.

aleksey-roldugin (Mon, 05 Jun 2017 16:48:35 GMT):
https://jira.hyperledger.org/browse/INDY-153

aleksey-roldugin (Mon, 05 Jun 2017 17:16:02 GMT):
https://jira.hyperledger.org/browse/INDY-154

avkrishnan (Tue, 06 Jun 2017 09:56:41 GMT):
https://jira.hyperledger.org/browse/INDY-155

vladimir.shishkin (Tue, 06 Jun 2017 12:48:09 GMT):
Has joined the channel.

vladimir.shishkin (Tue, 06 Jun 2017 12:48:42 GMT):
https://jira.hyperledger.org/browse/INDY-159

spivachuk (Tue, 06 Jun 2017 15:42:50 GMT):
Has joined the channel.

spivachuk (Tue, 06 Jun 2017 15:42:55 GMT):
https://jira.hyperledger.org/browse/INDY-158

dinghaiyang (Wed, 07 Jun 2017 02:26:23 GMT):
Has joined the channel.

spivachuk (Wed, 07 Jun 2017 11:13:00 GMT):
https://jira.hyperledger.org/browse/INDY-172

sapetani (Wed, 07 Jun 2017 11:55:06 GMT):
Has joined the channel.

vladimir.shishkin (Wed, 07 Jun 2017 14:52:00 GMT):
https://jira.hyperledger.org/browse/INDY-175

nage (Wed, 07 Jun 2017 16:22:05 GMT):
@avkrishnan @spivachuk and @vladimir.shishkin In the future would you post some context when you paste links here? That way folks who weren't privy to what caused you to share them here can follow along?

stevetolman (Thu, 08 Jun 2017 02:50:27 GMT):
sovrin

avkrishnan (Thu, 08 Jun 2017 03:44:08 GMT):
@nage The idea was that all issues found in the shakedowns would be reported on this channel. Perhaps a summary of the issue could be posted as well. And interested folks could also go to the provided Jira link for more info.

nage (Thu, 08 Jun 2017 03:45:09 GMT):
That is good. It is just confusing to have a stream of links without and context (perhaps a short message saying, here is what is going on would suffice).

nage (Thu, 08 Jun 2017 03:45:09 GMT):
That is good. It is just confusing to have a stream of links without any context (perhaps a short message saying: "here is what is going on" would suffice).

avkrishnan (Thu, 08 Jun 2017 08:39:10 GMT):
Okay, will do this ...

danielhardman (Thu, 08 Jun 2017 13:35:06 GMT):
It could be something very simple, such as, "Found a timing issue with consensus: "

nage (Thu, 08 Jun 2017 14:52:11 GMT):
The Indy-node WG starts in 9 minutes (see #indy-node for call details and notes).

drummondreed (Thu, 08 Jun 2017 15:21:24 GMT):
Link for the Sovrin Provisional Trust Framework—the legal framework that will govern the first stage of the Sovrin Provisional Network (the first global deployment of Indy): http://forum.sovrin.org/t/sovrin-provisional-trust-framework-approved-22-march-2017/221

farooq_m_khan (Thu, 08 Jun 2017 18:14:20 GMT):
@stevetolman I reported these defects since yesterday: INDY-171, INDY-173, INDY-178, INDY-179, INDY-188

farooq_m_khan (Thu, 08 Jun 2017 18:14:35 GMT):
INDY-173 was a very simple fix, I have also fixed it

farooq_m_khan (Thu, 08 Jun 2017 18:15:43 GMT):
@stevetolman do you want me to capture these numbers somewhere in the MGL Sheet?

stevetolman (Thu, 08 Jun 2017 18:32:05 GMT):
Yes please - in the triage section

peacekeeper (Thu, 08 Jun 2017 18:48:52 GMT):
Sometimes I'm receiving timeouts when using indy-sdk to send commands to the ledger. I mean I call the indy-sdk API and I never receive the callback. I don't have any more information at this point and can't tell whether it's an issue with my node, with indy-sdk, or with my java wrapper. I'll try to narrow down the problem, or maybe it's solved already with more recent versions. Just wanted to mention this, I'll let you know if I find out more.

peacekeeper (Thu, 08 Jun 2017 18:48:52 GMT):
Sometimes I'm getting timeouts when using indy-sdk to send commands to the ledger. I mean I call the indy-sdk API and I never receive the callback. I don't have any more information at this point and can't tell whether it's an issue with my node, with indy-sdk, or with my java wrapper. I'll try to narrow down the problem, or maybe it's solved already with more recent versions. Just wanted to mention this, I'll let you know if I find out more.

SarahMills (Thu, 08 Jun 2017 19:01:48 GMT):
Has left the channel.

Calvin_Heo (Fri, 09 Jun 2017 09:00:31 GMT):
Has joined the channel.

aleksey-roldugin (Fri, 09 Jun 2017 17:39:18 GMT):
New case about problems with catch up after stop-start sovrin-node.service on Stable build. https://jira.hyperledger.org/browse/INDY-191

bmiller59 (Sat, 10 Jun 2017 18:32:15 GMT):
Has joined the channel.

Wenlin (Mon, 12 Jun 2017 18:41:40 GMT):
Has joined the channel.

farooq_m_khan (Mon, 12 Jun 2017 20:50:38 GMT):
I have reported a defect: https://jira.hyperledger.org/browse/INDY-204

farooq_m_khan (Mon, 12 Jun 2017 20:51:31 GMT):
its a low Priority right now, I have started a discussion with Dmitry to evaluate if this could be a security vulnerability if yes this could be come a High priority

lm_nop (Mon, 12 Jun 2017 23:15:26 GMT):
Has joined the channel.

farooq_m_khan (Tue, 13 Jun 2017 10:37:07 GMT):
created a backlog story https://jira.hyperledger.org/browse/INDY-207 on suggestion from @jamesmonaghan

jamesmonaghan (Tue, 13 Jun 2017 10:37:07 GMT):
Has joined the channel.

alexander.shekhovcov (Tue, 13 Jun 2017 11:26:25 GMT):
*An ATTRIB transaction with an empty JSON* https://jira.hyperledger.org/browse/INDY-206

alexander.shekhovcov (Tue, 13 Jun 2017 11:26:25 GMT):
"An ATTRIB transaction with an empty JSON" https://jira.hyperledger.org/browse/INDY-206

farooq_m_khan (Tue, 13 Jun 2017 18:15:09 GMT):
reported a defect "Sovrin-Node: Python interpreter is not executed in Optimise mode" https://jira.hyperledger.org/browse/INDY-211

alain2sf (Wed, 14 Jun 2017 07:48:23 GMT):
Has joined the channel.

aleksey-roldugin (Wed, 14 Jun 2017 17:42:28 GMT):
New issue related to node adding with 'services': [] https://jira.hyperledger.org/browse/INDY-228

aleksey-roldugin (Wed, 14 Jun 2017 17:42:28 GMT):
New issue related to node adding with 'services': [] https://jira.hyperledger.org/browse/INDY-228

aleksey-roldugin (Wed, 14 Jun 2017 17:42:28 GMT):
New issue related to node adding with 'services': [] https://jira.hyperledger.org/browse/INDY-228

nage (Thu, 15 Jun 2017 04:56:14 GMT):
Just a reminder of the Indy Agent WG tomorrow at 9 am mountain time here https://zoom.us/j/232861185. The information on the Hyperledger Calendar is incorrect (the zoom link there is for the Indy Node WG meeting)

aleksey-roldugin (Thu, 15 Jun 2017 16:08:47 GMT):
Issue found during RC build testing related node promotion https://jira.hyperledger.org/browse/INDY-236

aleksey-roldugin (Thu, 15 Jun 2017 16:08:47 GMT):
Issue found during RC build testing related to node promotion https://jira.hyperledger.org/browse/INDY-236

aleksey-roldugin (Thu, 15 Jun 2017 18:20:59 GMT):
ADD GENESIS TRANSACTION command does not work https://jira.hyperledger.org/browse/INDY-238

xzib (Sat, 17 Jun 2017 11:02:04 GMT):
Has joined the channel.

aleksey-roldugin (Mon, 19 Jun 2017 14:20:49 GMT):
Validation error for send NYM command has poorly readable text https://jira.hyperledger.org/browse/INDY-261

VipinB (Mon, 19 Jun 2017 15:01:46 GMT):
@hartm asks Does anyone know if there is a Hyperledger Indy logo yet?

hartm (Mon, 19 Jun 2017 15:01:46 GMT):
Has joined the channel.

nage (Mon, 19 Jun 2017 15:09:51 GMT):
@Sean_Bohan ^^^

aleksey-roldugin (Mon, 19 Jun 2017 18:14:12 GMT):
[send ATTRIB] Validation error is unclear in case "endpoint" has not JSON value https://jira.hyperledger.org/browse/INDY-264

Honglei (Tue, 20 Jun 2017 01:26:17 GMT):
Has joined the channel.

mvsjober (Tue, 20 Jun 2017 13:26:48 GMT):
Has joined the channel.

nage (Thu, 22 Jun 2017 03:21:51 GMT):
@here just a reminder there is an Indy Node WG call tomorrow (see #indy-node for details).

wightman (Thu, 22 Jun 2017 15:43:44 GMT):
Has joined the channel.

aleksey-roldugin (Fri, 23 Jun 2017 14:19:13 GMT):
[POOL_UPGRADE] Node1 was broken after sending upgrade command https://jira.hyperledger.org/browse/INDY-324

aleksey-roldugin (Sat, 24 Jun 2017 13:15:26 GMT):
[SEND ATTRIB] CLI may crash in case of unexpected symbols in raw variable https://jira.hyperledger.org/browse/INDY-327

aleksey-roldugin (Sat, 24 Jun 2017 14:25:52 GMT):
[Enhancement][GET_NYM] Command should not require key presence in keyring https://jira.hyperledger.org/browse/INDY-328

gdinhof (Mon, 26 Jun 2017 05:47:08 GMT):
Has left the channel.

Tiitus (Mon, 26 Jun 2017 12:53:53 GMT):
Has joined the channel.

gudkov (Wed, 28 Jun 2017 15:58:48 GMT):
Has joined the channel.

jljordan_bcgov (Thu, 29 Jun 2017 03:26:09 GMT):
Has joined the channel.

agslc (Thu, 29 Jun 2017 15:32:29 GMT):
Has joined the channel.

btuttle (Thu, 29 Jun 2017 15:45:32 GMT):
Has joined the channel.

fabienpe (Fri, 30 Jun 2017 12:02:43 GMT):
Has joined the channel.

jellevdp (Fri, 30 Jun 2017 13:20:36 GMT):
Has joined the channel.

jkumar (Sat, 01 Jul 2017 08:55:46 GMT):
Has joined the channel.

stedwms (Sat, 01 Jul 2017 22:11:10 GMT):
Has joined the channel.

stedwms (Sat, 01 Jul 2017 22:31:51 GMT):
Greetings, I installed the system on a Vagrant box, Ubuntu 16.04, and am playing with the command line. Could someone please advise... what/where is the *most recent* pool_transactions_sandbox file? ... or similar... to best seed the connections to federated servers? Hope I'm on the right track.

farooq_m_khan (Tue, 04 Jul 2017 07:59:05 GMT):
@stedwms The Vagrant based system allows you to create a test network of 4 nodes and 4 agents

farooq_m_khan (Tue, 04 Jul 2017 07:59:59 GMT):
You can play with it as you wish and maybe even try to do some destruction operations

farooq_m_khan (Tue, 04 Jul 2017 08:00:42 GMT):
AFAIK You cannot get onto the Alpha network however until you are a steward

mindraker (Tue, 04 Jul 2017 10:49:50 GMT):
Has joined the channel.

andrewtan (Wed, 05 Jul 2017 00:39:17 GMT):
Has joined the channel.

BryanSparks (Thu, 06 Jul 2017 17:20:47 GMT):
Has joined the channel.

tkuhrt (Thu, 06 Jul 2017 19:10:35 GMT):
I am looking for a list of emails for the maintainers of Hyperledger Indy. I did not see anything in the source code. Does this exist somewhere that you can point me to? I would like to send an email to the maintainers of Hyperledger Indy without using the mailing list at lists.hyperledger.org.

jleders (Thu, 06 Jul 2017 20:37:13 GMT):
Has joined the channel.

nage (Fri, 07 Jul 2017 02:07:46 GMT):
I don't think we have such a list as text in the code (yet). I have logged it as https://jira.hyperledger.org/browse/INDY-387

nage (Fri, 07 Jul 2017 02:09:45 GMT):
For now you can send email to @danielhardman, @stevetolman, @nage or @farooq_m_khan and we'll make sure it gets to the right folks.

Mike_ATB (Fri, 07 Jul 2017 14:32:15 GMT):
Has joined the channel.

tkuhrt (Fri, 07 Jul 2017 17:22:29 GMT):
Thanks, @nage. @farooq_m_khan could you DM your email address to me?

farooq_m_khan (Sat, 08 Jul 2017 11:33:20 GMT):
@tkuhrt I am assuming you have obtained our email ids, I see a email from you regarding the Hyperledger Indo logo stuff

tkuhrt (Sat, 08 Jul 2017 13:36:23 GMT):
Yes. All good, Farooq

Audrius (Wed, 12 Jul 2017 08:49:01 GMT):
Has joined the channel.

nage (Wed, 12 Jul 2017 15:03:11 GMT):
There is a new #indy-pr-review channel where code owners and maintainers should hang out for responding to and reviewing pull requests (see you there!)

tharmon (Wed, 12 Jul 2017 15:43:23 GMT):
Has joined the channel.

LoveshHarchandani (Wed, 12 Jul 2017 16:25:54 GMT):
@channel We will be implementing a very basic mechanism (we don't want to spend more than 1.5 dev days to finish this) to store timestamps in the ledger. The ticket for that is https://jira.hyperledger.org/browse/INDY-353, it mentions the design that we will be implementing (it's same as bitcoin), please raise any concerns if you disagree with the design

LoveshHarchandani (Wed, 12 Jul 2017 16:25:54 GMT):
@channel We will be implementing a very basic mechanism (we don't want to spend more than 1.5 dev days to finish this) to store timestamps in the ledger. The ticket for that is https://jira.hyperledger.org/browse/INDY-353, it mentions the design that we will be implementing (it's same as bitcoin), please raise any concerns if you disagree with the design

LoveshHarchandani (Wed, 12 Jul 2017 16:25:54 GMT):
@here We will be implementing a very basic mechanism (we don't want to spend more than 1.5 dev days to finish this) to store timestamps in the ledger. The ticket for that is https://jira.hyperledger.org/browse/INDY-353, it mentions the design that we will be implementing (it's same as bitcoin), please raise any concerns if you disagree with the design

Rouven (Thu, 13 Jul 2017 12:25:20 GMT):
Has joined the channel.

nage (Thu, 13 Jul 2017 15:04:04 GMT):
Just a reminder that the Indy Agent WG call is starting now here https://zoom.us/j/232861185

andrey.goncharov (Thu, 13 Jul 2017 15:11:07 GMT):
@channel ATTENTION! Our DEB (and pypi) packages were renamed! python3-plenum -> indy-plenum python3-anoncreds -> indy-anoncreds sovrin-node -> indy-node A new package 'sovrin' was introduced. It depends on indy-node and just replaces indy-node's genesis transaction files with sovrin ones. Our final deliverable is sovrin. Please update your environments to install sovrin package instead of sovrin-node.

andrey.goncharov (Thu, 13 Jul 2017 15:11:07 GMT):
@here ATTENTION! Our DEB (and pypi) packages were renamed! python3-plenum -> indy-plenum python3-anoncreds -> indy-anoncreds sovrin-node -> indy-node A new package 'sovrin' was introduced. It depends on indy-node and just replaces indy-node's genesis transaction files with sovrin ones. Our final deliverable is sovrin. Please update your environments to install sovrin package instead of sovrin-node.

danielhardman (Thu, 13 Jul 2017 19:56:10 GMT):
Cross-posting from #indy-agent: we've been invited to create a logo for indy. I am going to attach the guidelines and then one pair of candidate logos (.svg, .png). Feel free to comment or to run wild with your own ideas.

danielhardman (Thu, 13 Jul 2017 19:56:22 GMT):

Message Attachments

danielhardman (Thu, 13 Jul 2017 19:56:36 GMT):

Message Attachments

danielhardman (Thu, 13 Jul 2017 19:56:48 GMT):

Message Attachments

alexeykoren (Fri, 14 Jul 2017 09:58:41 GMT):
Has joined the channel.

jljordan_bcgov (Wed, 19 Jul 2017 20:04:01 GMT):
I'm trying to get the demo script going based on docker images at https://github.com/evernym/sovrin-environments/tree/master/docker and I haven't figured out where to go for help ... the startup script ./pool_start.sh is getting stuck at Step 9/12 where it appears to be trying to RUN useradd -ms /bin/bash -u $uid sovrin

jljordan_bcgov (Wed, 19 Jul 2017 20:44:27 GMT):
Bit more info if it helps ... and this is the correct place CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES 52f517b3be05 699dec8c958e "/bin/sh -c 'usera..." 49 seconds ago Exited (0) 47 seconds ago adoring_liskov

faisal00813 (Thu, 20 Jul 2017 06:40:54 GMT):
Has joined the channel.

Dpkkmr (Thu, 20 Jul 2017 11:53:54 GMT):
Has joined the channel.

nage (Thu, 20 Jul 2017 14:59:47 GMT):
The Indy Node WG call starts now here https://zoom.us/j/730665163

stevetolman (Thu, 20 Jul 2017 15:08:29 GMT):
https://jira.hyperledger.org/secure/RapidBoard.jspa?rapidView=133&view=planning&selectedIssue=INDY-346

ryanmarsh (Thu, 20 Jul 2017 15:22:51 GMT):
Has joined the channel.

pwnintended (Fri, 21 Jul 2017 13:13:27 GMT):
Has joined the channel.

tgehrke (Sat, 22 Jul 2017 05:12:36 GMT):
Has joined the channel.

tmenner (Sun, 23 Jul 2017 20:14:44 GMT):
Has joined the channel.

rock_martin (Tue, 25 Jul 2017 10:13:22 GMT):
Has joined the channel.

Sean_Bohan (Tue, 25 Jul 2017 15:07:16 GMT):
This Thurs 7/27 we are doing the Agent WG call. Details here: http://forum.sovrin.org/t/agent-working-group-call-7-27-agenda/353 Agenda: We would love to have a focused discussion on the following topics: Our Agenda: 1. Product roadmap for Hyperledger Indy 2. We want to share the current needs for Agents 3. We want your input and thoughts on what is there, what you think is needed 4. “Trailhead” for Hyperledger Indy Agents 5. Agent Extensible API 6. Growing the Agent community (invite a friend to take a look at our GitHub or join a WG call!) THANKS!

srottem (Tue, 25 Jul 2017 18:27:41 GMT):
Is there information anywhere that clearly describes how the anoncreds functionality works? I've been working on the .NET wrapper and porting tests so I think I ]]

srottem (Tue, 25 Jul 2017 18:27:41 GMT):
Is there information anywhere that clearly describes how the anoncreds functionality works? I've been working on the .NET wrapper and porting tests so I think I'm starting to get the idea, but it would be nice if there were some tutorial information explaining how each of the functions exposed by the SDK are intended to be used together.

srottem (Tue, 25 Jul 2017 18:35:16 GMT):
I'm also pretty unclear on what the various elements of the JSON used in claim offers, claim attributes, claim schema, etc. represent (some are clear, but others are pretty opaque. Any pointers to where this information can be uncovered would be very welcome.

jbesan (Wed, 26 Jul 2017 10:56:05 GMT):
Has joined the channel.

sanbachs (Wed, 26 Jul 2017 17:51:56 GMT):
Has joined the channel.

Sean_Bohan (Wed, 26 Jul 2017 18:16:09 GMT):
@srottem - I am following up on these 2 questions for ya

Sean_Bohan (Thu, 27 Jul 2017 14:06:27 GMT):
@srottem we dont have a defined tutorial but check out the docs in the anoncreds repo - https://github.com/hyperledger/indy-anoncreds/tree/master/docs Still following up on the claim docs for ya We also have a bi-weekly Agent and Ledger call (today is Agent, next Thurs is Ledger - details above your post)

nage (Thu, 27 Jul 2017 14:47:12 GMT):
A reminder that the Agent WG call starts in 13 minutes here https://zoom.us/j/232861185

nage (Thu, 27 Jul 2017 14:47:12 GMT):
A reminder that the Indy Agent WG call starts in 13 minutes here https://zoom.us/j/232861185

edge (Thu, 27 Jul 2017 15:32:29 GMT):
Has joined the channel.

Terminalman90 (Thu, 27 Jul 2017 15:51:19 GMT):
Has joined the channel.

vidalalcala (Fri, 28 Jul 2017 22:47:56 GMT):
Has joined the channel.

skibz (Wed, 02 Aug 2017 10:33:35 GMT):
Has joined the channel.

Sean_Bohan (Wed, 02 Aug 2017 13:52:14 GMT):
Folks: Agenda for the Ledger call (tomorrow Aug 3): Quick note to share the agenda of the next Ledger Working Group and get the community thinking about new topics and discussions. Details: When: The next Ledge WG meeting will be held Thursday, August 03, 2017 8amPT, 9amMT, 11amET, 4pm BST. Where: Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/232861185 Or iPhone one-tap (US Toll): +16465588656,232861185# or +14086380968,232861185# Or Telephone: Dial: +1 646 558 8656 (US Toll) or +1 408 638 0968 (US Toll) Meeting ID: 232 861 185 International numbers available: https://zoom.us/zoomconference?m=a0jD_rTMnh0ZYGQDOKPCNrK_0dP7WPfp Agenda: We would love to have a focused discussion on the following topics: 1. The Sovrin Provisional Network is LIVE! 2. Product roadmap for Hyperledger Indy We want to share the current needs for Ledger We want your input and thoughts on what is there, what you think is needed “Trailhead” for Hyperledger Indy participation 3. Growing the Ledger community (invite a friend to take a look at our GitHub or join a WG call!) See you all then/there

masonhensley (Thu, 03 Aug 2017 03:37:52 GMT):
Has joined the channel.

marek.dapps (Thu, 03 Aug 2017 08:44:00 GMT):
Has joined the channel.

masonhensley (Thu, 03 Aug 2017 15:13:05 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=CQgjN8cEPf2bBBwQY) @Sean_Bohan Was this cancelled?

masonhensley (Thu, 03 Aug 2017 15:14:03 GMT):
@Sean_Bohan Was the call cancelled?

nage (Thu, 03 Aug 2017 15:34:41 GMT):
not cancelled, but the appointment info here pointed to the wrong zoom room (that zoom room is for the Agent WG)

bendon (Fri, 04 Aug 2017 08:04:35 GMT):
Has joined the channel.

bendon (Fri, 04 Aug 2017 08:13:39 GMT):
Hi there. As Steward for Sovrin, the company I'm working for is willing to put me on the task to contribute to https://jira.hyperledger.org/browse/INDY-580

bendon (Fri, 04 Aug 2017 08:14:24 GMT):
Could the first step be having a JIRA account?

bendon (Fri, 04 Aug 2017 08:33:19 GMT):
I've figured I had to ask @nage for this...

Sean_Bohan (Fri, 04 Aug 2017 12:04:41 GMT):
Hi Brendon. You would need a Linux Foundation ID to access Jira - you should be able to create an account here: https://identity.linuxfoundation.org/

bendon (Fri, 04 Aug 2017 12:34:21 GMT):
Hi. Got one. I believe I had to do so in order to signup in this chat.

bendon (Fri, 04 Aug 2017 12:34:51 GMT):
However, I'm still not asked for any Linux Foundation ID on JIRA...

bendon (Fri, 04 Aug 2017 12:35:42 GMT):
"Not a member? To request an account, please contact your JIRA administrators."

bendon (Fri, 04 Aug 2017 12:37:31 GMT):
I feel like I'm missing something there...

bendon (Fri, 04 Aug 2017 12:38:00 GMT):
I've signed up my Linux Foundation ID using Github.

bendon (Fri, 04 Aug 2017 12:38:32 GMT):
But now I wonder if I don't need to create a separate password!?

bendon (Fri, 04 Aug 2017 12:41:07 GMT):
now got a password for Linux F. ID, but can not pass the auth on JIRA with that...

Sean_Bohan (Fri, 04 Aug 2017 13:11:11 GMT):
Let me check...

nage (Mon, 07 Aug 2017 14:48:22 GMT):
@bendon yes, you need to go to the identity link Sean shared earlier and add a password from the LF side in order to log into Jira (id doesn't use single sign on the same way some of the other services do)

bendon (Mon, 07 Aug 2017 14:50:44 GMT):
ThX @nage, but I had trouble as I could not define a password at all. Only way I found was to change it, but this was requiring the old one which I never had! Even the reset procedure was requesting me the old password! Did not make sense to me.

bendon (Mon, 07 Aug 2017 14:51:33 GMT):
However, I did retry to reset by never defiend password today from the login page, and it did not request my old one.

bendon (Mon, 07 Aug 2017 14:51:46 GMT):
After that, I've been able to log in JIRA.

bendon (Mon, 07 Aug 2017 14:53:21 GMT):
This could be a bug/feature in LF id portal, or between my chair and my keyboard.

bendon (Mon, 07 Aug 2017 14:53:31 GMT):
I may try to reproduce it later...

nage (Mon, 07 Aug 2017 14:54:28 GMT):
Glad you made it through the issue

ckpilling (Wed, 09 Aug 2017 11:36:11 GMT):
Has joined the channel.

skibz (Thu, 10 Aug 2017 09:47:27 GMT):
Has left the channel.

nage (Thu, 10 Aug 2017 15:01:45 GMT):
Indy Agent WG call starting here https://zoom.us/j/232861185

lohan.spies (Thu, 10 Aug 2017 20:02:31 GMT):
Has joined the channel.

DibbsZA (Sun, 13 Aug 2017 06:25:35 GMT):
Has joined the channel.

ehanyegy (Tue, 15 Aug 2017 07:03:15 GMT):
Has joined the channel.

troyronda (Tue, 15 Aug 2017 21:39:18 GMT):
Has joined the channel.

Suedonym (Thu, 17 Aug 2017 17:12:00 GMT):
Has joined the channel.

RBowron (Fri, 18 Aug 2017 00:20:49 GMT):
Has joined the channel.

dazzaji (Fri, 18 Aug 2017 14:47:14 GMT):
Has joined the channel.

jamessheen (Mon, 21 Aug 2017 14:59:55 GMT):
Has joined the channel.

swcurran (Mon, 21 Aug 2017 19:48:59 GMT):
Has joined the channel.

MaheshBiradar (Tue, 22 Aug 2017 07:00:05 GMT):
Has joined the channel.

maheshkr81 (Tue, 22 Aug 2017 09:47:47 GMT):
Has joined the channel.

sklump (Tue, 22 Aug 2017 11:37:06 GMT):
Has joined the channel.

downTheFallLine (Wed, 23 Aug 2017 02:17:04 GMT):
Has joined the channel.

lovesh (Wed, 23 Aug 2017 09:21:06 GMT):
Has joined the channel.

OlufAndrews (Wed, 23 Aug 2017 14:24:55 GMT):
Has joined the channel.

VonNeumanniac (Wed, 23 Aug 2017 14:44:33 GMT):
Has joined the channel.

Othman.Darwish (Thu, 24 Aug 2017 07:48:04 GMT):
Has joined the channel.

darrell.odonnell (Sun, 27 Aug 2017 16:19:04 GMT):
Folks - just a heads up that I created a Wiki page on the HyperLedger Wiki: https://wiki.hyperledger.org/projects/indy

guidol (Wed, 30 Aug 2017 13:29:27 GMT):
Has joined the channel.

tkuhrt (Wed, 30 Aug 2017 23:30:53 GMT):
@darrell.odonnell : Project is in "Incubation", right?

tkuhrt (Wed, 30 Aug 2017 23:31:04 GMT):
Looks like the Wiki page says it is "Active"

leogzl (Thu, 31 Aug 2017 05:52:03 GMT):
Has joined the channel.

bjornhelge (Thu, 31 Aug 2017 09:46:37 GMT):
Has joined the channel.

darrell.odonnell (Thu, 31 Aug 2017 11:23:55 GMT):
@tkuhrt I just copied some starting text from another of the projects - please update the status if Active is incorrect.

benjaminbollen (Thu, 31 Aug 2017 14:43:16 GMT):
Has joined the channel.

nage (Thu, 31 Aug 2017 15:11:01 GMT):
@tkuhrt yes, we are in Incubation

tkuhrt (Thu, 31 Aug 2017 15:11:33 GMT):
Great...I can update the Wiki

tkuhrt (Thu, 31 Aug 2017 15:12:12 GMT):
Done

nage (Thu, 31 Aug 2017 15:12:25 GMT):
@tkuhrt I announced the anti-trust policy on the Indy Node developer call today and am recording the call, is there a Hyperledger location where I should upload those?

tkuhrt (Thu, 31 Aug 2017 15:12:44 GMT):
Yes...let me send you an invite for the location.

tkuhrt (Thu, 31 Aug 2017 15:13:52 GMT):
Do you prefer evernym or sovrin email?

nage (Thu, 31 Aug 2017 15:14:10 GMT):
Sovrin

tkuhrt (Thu, 31 Aug 2017 15:15:20 GMT):
Invite sent. People will be able to find meeting recordings at: https://drive.google.com/open?id=0B_NJV6eJXAA1UlJZMXd3cm1zNDQ

nage (Thu, 31 Aug 2017 18:26:09 GMT):
I have uploaded the recording of today's Indy Node call to the folder, thanks @tkuhrt !

pacog 1 (Thu, 31 Aug 2017 19:30:52 GMT):
Has joined the channel.

lmrln (Fri, 01 Sep 2017 10:09:27 GMT):
Has joined the channel.

conroydave (Sat, 02 Sep 2017 03:20:21 GMT):
Has joined the channel.

conroydave (Sat, 02 Sep 2017 03:20:25 GMT):
hey hey

conroydave (Sat, 02 Sep 2017 03:20:52 GMT):
bouncing around between sovrin and evernym and indy

conroydave (Sat, 02 Sep 2017 03:21:19 GMT):
got the demos working on my macbook. documenation needs a bit of updating.

conroydave (Sat, 02 Sep 2017 03:21:30 GMT):
is this the place to discuss sovrin, or should I be on there slack?

ft-wanghaibo (Mon, 04 Sep 2017 05:43:57 GMT):
Has joined the channel.

darrell.odonnell (Tue, 05 Sep 2017 12:09:03 GMT):
(my opinion) if it is technical and really about the codebase and the contents of Indy I think this is the place. Now, the #indy-sdk or #indy-agent channels here on Rocket.Chat may be more focused?

darrell.odonnell (Tue, 05 Sep 2017 12:09:30 GMT):
if the discussion is about the governance, organization, etc. - then perhaps Slack is a better place?

bendon (Tue, 05 Sep 2017 13:00:29 GMT):
Hi there. I've managed to compile and package indy-node and its deps for CentOS 7. Now I'm trying to figure if it works as expected using https://github.com/hyperledger/indy-node/blob/master/docs/Sovrin_Running_Locally.md

bendon (Tue, 05 Sep 2017 13:01:57 GMT):
But I can attache the faber agent, some thing about the wrong key I guess: client request invalid: CouldNotAuthenticate('Can not find verkey for DID...

bendon (Tue, 05 Sep 2017 13:01:57 GMT):
But I can NOT attach the faber agent, some thing about the wrong key I guess: client request invalid: CouldNotAuthenticate('Can not find verkey for DID...

bendon (Tue, 05 Sep 2017 13:03:22 GMT):
I'm not sure if my test or the setup is wrong...

bendon (Tue, 05 Sep 2017 13:05:59 GMT):
Only thing I'm concerned is the nodes (Running Sovrin 1.1.33) are using libsodium-1.0.5 (libsodium.so.13.3.0). No idea if/why 1.0.8+ is required.

Sean_Bohan (Wed, 06 Sep 2017 12:39:42 GMT):
@bendon - let me see if I can get some help

Sean_Bohan (Wed, 06 Sep 2017 17:02:36 GMT):
Hyperledger Indy Agents WG Call tomorrow 9/8 8amPT, 9amMT, 11amET, 4pm BST. Where: Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/232861185 Or iPhone one-tap (US Toll): +16465588656,232861185# or +14086380968,232861185# Or Telephone: Dial: +1 646 558 8656 (US Toll) or +1 408 638 0968 (US Toll) Meeting ID: 232 861 185 International numbers available: https://zoom.us/zoomconference?m=a0jD_rTMnh0ZYGQDOKPCNrK_0dP7WPfp1

Sean_Bohan (Wed, 06 Sep 2017 17:04:07 GMT):
@conroydave - generally ledger discussions happen in Indy-node, agent discussions in indy-agent. We still have discussions running in the Slack as well as doing a weekly WG call on Thursdays (see time/place above)

vladimir.shishkin (Thu, 07 Sep 2017 09:03:21 GMT):
@bendon > But I can NOT attach the faber agent, some thing about the wrong key I guess: client request invalid: CouldNotAuthenticate('Can not find verkey for DID... Make sure that agents' scripts are ran as sovrin user from /.sovrin/ and there are all genesis txn files in /.sovrin/.

bendon (Thu, 07 Sep 2017 09:46:25 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=ADBt8umhDM4HXmWc9) @vladimir.shishkin Checked and checked: running as sovrin and ~/.sovrin/domain_transactions_sandbox_genesis and ~/.sovrin/pool_transactions_sandbox_genesis are populated

darrell.odonnell (Fri, 08 Sep 2017 20:56:42 GMT):
can someone review the pull requests in #indy-pr-review ?

bdot (Sat, 09 Sep 2017 10:00:26 GMT):
Has joined the channel.

klcompt (Mon, 11 Sep 2017 01:02:33 GMT):
Has joined the channel.

skbodwell (Mon, 11 Sep 2017 20:47:59 GMT):
Has joined the channel.

slender (Tue, 12 Sep 2017 03:13:50 GMT):
Has joined the channel.

nage (Tue, 12 Sep 2017 19:48:59 GMT):
@darrell.odonnell this should be happening all along the way (perhaps we should ask folks to acknowledge them when they do so?) Is there a particular request that isn't getting love? (that way I can take it up with the right maintainer(s)?)

srottem (Wed, 13 Sep 2017 12:02:27 GMT):
Just watched Phil Windley's talk at Blockchain Utah and he mentioned premium claims and consent receipts. Is there detail on how these work on the Sovrin network?

drummondreed (Wed, 13 Sep 2017 12:33:02 GMT):
Premium claims are a branch of verifiable claims for which there will be economic exchange. We're deep in the process of figuring that out. Consent receipts are also still deep in the design process—we want to harmonize with the Kantara consent receipt work.

srottem (Wed, 13 Sep 2017 12:58:13 GMT):
Thanks Drummond.

bendon (Thu, 14 Sep 2017 10:20:17 GMT):
I've managed to replay the Getting Started demo about Alice Transcript. But now I wonder how connections, claims and proof can be controlled afterwards. As I don't see any remove/delete/change/revoke commands documented in the CLI help.

bendon (Thu, 14 Sep 2017 10:20:52 GMT):
Are those feature already implemented in the nodes and just not yet on the client?

bendon (Thu, 14 Sep 2017 10:20:52 GMT):
Are those features already implemented in the nodes and just not yet on the client?

SethiSaab (Thu, 14 Sep 2017 21:35:07 GMT):
Has joined the channel.

nchangfong (Sun, 17 Sep 2017 22:58:09 GMT):
Has joined the channel.

HintikkaKimmo (Mon, 18 Sep 2017 16:32:38 GMT):
Has joined the channel.

darrell.odonnell (Wed, 20 Sep 2017 11:27:45 GMT):
@bendon that leads into a topic we're pondering about on a project. Here's my take on your question: The DIDs and DDOs that are written to the ledger are immutable (i.e. they can't be changed). That means you need to be able to build your system so it can look at a history and I'll use a Verifiable Claim as an example. Let's say a company was issued a water permit. We have an Issuer (government), an Issuee (the company), and a DDO version of a claim that can be verified as signed by the Issuer and held by the Issuee.

darrell.odonnell (Wed, 20 Sep 2017 11:28:35 GMT):
That can't be changed. But a future revocation could occur. Let's say the company violates a rule and the government needs to issue a revocation of the original claim. That's where it's currently stumping us.

darrell.odonnell (Wed, 20 Sep 2017 11:29:04 GMT):
For connections (e.g. a DID-to-DID relationship) it is easier - just delete them from your wallet and don't respond on the endpoint that is provided.

darrell.odonnell (Wed, 20 Sep 2017 11:29:04 GMT):
For connections (e.g. a DID-to-DID relationship) it is easier - just delete them from your wallet and don't respond on the endpoint that is provided. The record will always be there but if you use a unique DID every time there is no correlation risk.

darrell.odonnell (Wed, 20 Sep 2017 11:30:03 GMT):
So - back to the Verifiable Claim Revocation. I am curious where the best current thinking is here? We're going to likely build a small open-source Python library to handle parts of this. @drummondreed @windley @nage

darrell.odonnell (Wed, 20 Sep 2017 11:31:46 GMT):
I'm happy to start sharing some how-to and guidance as we build it out.

darrell.odonnell (Wed, 20 Sep 2017 11:31:46 GMT):
I'm happy to start sharing some how-to docs, examples, and guidance as we build it out.

sklump (Wed, 20 Sep 2017 11:45:47 GMT):
At present, claim revocation is not implemented. There are a few ways to tackle this kind of thing, for different requirements: (1) make a claim that such and such was permitted at such and such a time, and then the client side policy is responsible for maximum age. E.g., this company is LEED certified. If the company fails to keep it up, don't re-issue. (2) make rotating short-duration claims; again, it is up to the client side policy what to accept. E.g., a passport is fine as ID until expiry, but country XYZ requires it to be valid 6 months after arrival for entry (3) have claim issuers available for agents to make claim requests in the moment (4) revocation of claims, as yet not implemented. #1-#2 are like the attribute certificate Push model, and #3 is like OCSP or the attribute certificate Pull model. #4 is not a great fit for claims whose truth values toggle back and forth such as suspension or duty rotation. Note that there are way more claim definitions available than attributes (because of predicates) and way more attributes than certificates. In this chaos menagerie, each claim can have its own meaning and own freshness/validity in client policy, and so some kind of profile will be advantageous. This effort risks mimicking the generational struggle to define a decent role engineering framework, which in real systems becomes a rat's nest in a hurry. The above is a hasty and partial consideration of a difficult and multivariate problem.

thomas_p (Wed, 20 Sep 2017 11:56:19 GMT):
Has joined the channel.

nage (Wed, 20 Sep 2017 12:44:27 GMT):
Good summaries. Revocation work has designed and some signs of it are in the code base. Classic revocation involves a pointer in the claim that a verifier can check to see if the claim is in a list of revoked credentials. Unfortunately this can expose who has been revoked and when they were revoked. The accumulator-based revokation approach outlined in the Anoncreds paper instead adds a number on the ledger, that along with a nonrevocation proof allows the holder to prove nonrevocation without disclosing which claim belongs to them. Mechanically the trickiest part of this revocation scheme is how to replicate the "tails files" from the issuer to the holder, as this tails list is usually very large.

nage (Wed, 20 Sep 2017 12:48:47 GMT):
as for approach #3 from @sklump, if you allow direct contact with the issuer, you create correlation exposure that reveals when and where a credential is being used, and in some cases this traffic can be observed even outside the relationships (for example, the DMV should not know when your drivers license is used to check your age). The other approaches should be leveraged as appropriate, I especially agree that well crafted expiry criteria should be used to avoid he expense of revocation where possible. @danielhardman any more thoughts?

nage (Wed, 20 Sep 2017 12:48:47 GMT):
as for approach #3 from @sklump, if you allow direct contact with the issuer, you create correlation exposure that reveals when and where a credential is being used, and in some cases this traffic can be observed even outside the relationships (for example, the DMV should not know when your drivers license is used to check your age). The other approaches should be leveraged as appropriate, I especially agree that well crafted expiry criteria should be used to avoid the expense of revocation where possible. @danielhardman any more thoughts?

nage (Wed, 20 Sep 2017 12:50:23 GMT):
@Sean_Bohan and @sklump this may make an interesting discussion for the Agent WG meeting Thursday

sklump (Wed, 20 Sep 2017 13:30:15 GMT):
Revocation of a claim is not tantamount to revocation of a 'who' - maybe one of the predicates in the claim is no longer true; e.g., favourite drink changed. In any case, I don't think revocation of a claim is sensitive info in and of itself. Disclaimer - I come from PKI land, if revocation lists were confidential then they would be encrypted and not just signed. Libindy revocation as roughed in at present does not carry a reason - heck, the issuer node itself could be shut down on compromise, hence all claims revoked and no slight to entities in claims. I don't understand what it is to which "tails files" refer in this context.

anttikettunen (Thu, 21 Sep 2017 13:13:10 GMT):
What's the best place to dig into Indy's crypto? We have crypto-professionals who want to explore the innards of how cryptography is implemented in Indy. If there's a crypto validation documentation of any kind, please point there. Also, would be great to have links to code, where to look into.

Sean_Bohan (Thu, 21 Sep 2017 13:30:31 GMT):
Hyperledger Indy Agents WG Call TODAY 9/21 8amPT, 9amMT, 11amET, 4pm BST. Where: Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/232861185 Or iPhone one-tap (US Toll): +16465588656,232861185# or +14086380968,232861185# Or Telephone: Dial: +1 646 558 8656 (US Toll) or +1 408 638 0968 (US Toll) Meeting ID: 232 861 185 International numbers available: https://zoom.us/zoomconference?m=a0jD_rTMnh0ZYGQDOKPCNrK_0dP7WPfp1TODAY

Sean_Bohan (Thu, 21 Sep 2017 13:30:31 GMT):
Hyperledger Indy Agents WG Call tomorrow 9/21 8amPT, 9amMT, 11amET, 4pm BST. Where: Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/232861185 Or iPhone one-tap (US Toll): +16465588656,232861185# or +14086380968,232861185# Or Telephone: Dial: +1 646 558 8656 (US Toll) or +1 408 638 0968 (US Toll) Meeting ID: 232 861 185 International numbers available: https://zoom.us/zoomconference?m=a0jD_rTMnh0ZYGQDOKPCNrK_0dP7WPfp1

Sean_Bohan (Thu, 21 Sep 2017 13:30:43 GMT):
actually today 9/21

Dan (Thu, 21 Sep 2017 13:38:08 GMT):
Has joined the channel.

Sean_Bohan (Thu, 21 Sep 2017 16:09:08 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=ny62bS9N4CSqiapoe) @anttikettunen Let me ask @danielhardman

danielhardman (Thu, 21 Sep 2017 17:19:17 GMT):
The indy-anoncreds repo is the richest place to find info. See, for example, https://github.com/hyperledger/indy-anoncreds/blob/master/docs/anoncred-usecase0.pdf. The code in this repo implements math in the paper. There is also some crypto work in indy-sdk (which is just now being moved into its own repo, indy-crypto). The indy-sdk/indy-crypto version of the code is newer and uses AMCL/milagro instead of charm as its foundation; eventually it will replace the python code. It also uses type 3 pairings instead of the older type 1 math. Regarding crypto validation: I recommend that you contact Dmitry Khovratovich at the University of Luxembourg, who wrote the papers. I know that Jan Camenisch (his collaborator and the father of many of the techniques embodied by this code) has some correctness proofs for some of his work, but I don't know if it applies to all, part, or none of the work in this specific code. Dmitry will know, or Jan will.

danielhardman (Thu, 21 Sep 2017 17:20:45 GMT):
Jan and others on the IBM research team in Zurich are just preparing to collaborate on the indy-crypto repo, now that it's broken out into its own place and framed in a way that could be used more broadly within Hyperledger.

danielhardman (Thu, 21 Sep 2017 17:21:17 GMT):
^^ @anttikettunen

danielhardman (Thu, 21 Sep 2017 17:24:20 GMT):
@srottem You are the 4th community member in the past week who has asked for info about claim offers, claim attributes, claim schema, etc. I am desperately trying to produce helpful info, but my time is so fragmented that it's slow going. I am about to submit a paper to Rebooting Web of Trust that should be very helpful. Please give me 24 more hours.

nage (Thu, 21 Sep 2017 19:04:00 GMT):
https://github.com/evernym/sovrin-environments/blob/master/vagrant/training/vb-multi-vm/TestSovrinClusterSetup.md

nage (Thu, 21 Sep 2017 19:04:15 GMT):
https://github.com/hyperledger/indy-node/blob/master/getting-started.md

lehors (Thu, 21 Sep 2017 19:04:27 GMT):
Has joined the channel.

danconway (Thu, 21 Sep 2017 19:05:11 GMT):
Has joined the channel.

jrosmith (Thu, 21 Sep 2017 19:05:32 GMT):
Has joined the channel.

ScottMorris (Thu, 21 Sep 2017 19:08:07 GMT):
Has joined the channel.

nage (Thu, 21 Sep 2017 19:27:55 GMT):
https://zoom.us/j/422273191

nage (Thu, 21 Sep 2017 19:28:28 GMT):
the zoom link for the developer onboarding discussion going on at the Hyperledger Hackfest

lehors (Thu, 21 Sep 2017 19:28:38 GMT):
requesting key 68DB5E88 from hkp server keyserver.ubuntu.com fails for me, the server can't be accessed

lehors (Thu, 21 Sep 2017 19:28:43 GMT):
what port is used?

mgbailey (Thu, 21 Sep 2017 19:48:11 GMT):
Has joined the channel.

tkuhrt (Thu, 21 Sep 2017 19:48:52 GMT):
@lehors : This is the same error I was getting initially...must be the ssh port that is not open on the guest network

mgbailey (Thu, 21 Sep 2017 19:49:13 GMT):
indy (sovrin) did not install

tkuhrt (Thu, 21 Sep 2017 19:49:32 GMT):
Can I run a command to do the install now?

tkuhrt (Thu, 21 Sep 2017 19:49:38 GMT):
see what the error is?

mgbailey (Thu, 21 Sep 2017 19:49:43 GMT):
try "sudo apt install sovrin"

mgbailey (Thu, 21 Sep 2017 19:50:56 GMT):
unusual that this did not run properly

tkuhrt (Thu, 21 Sep 2017 19:51:55 GMT):
I will check my other agents too...thanks

mgbailey (Thu, 21 Sep 2017 19:58:59 GMT):
DEBIAN_FRONTEND=noninteractive apt-get install -y debsigs debsig-verify apt-transport-https dialog figlet python-pip python3-pip python3.5-dev libsodium18 unzip make screen sovrin tmux vim wget

mgbailey (Thu, 21 Sep 2017 20:00:06 GMT):
sudo apt install debsigs debsig-verify apt-transport-https dialog figlet python-pip python3-pip python3.5-dev libsodium18 unzip make screen sovrin tmux vim wget

Dan (Thu, 21 Sep 2017 20:00:25 GMT):
Thanks, Mike!

Dan (Thu, 21 Sep 2017 20:00:40 GMT):
Love the CLI btw

nage (Thu, 21 Sep 2017 20:03:23 GMT):
@krw910 folks are asking about non-vagrant mechanisms for setting up a test network? We have cloud formation scripts, but do we have anything else?

Dan (Thu, 21 Sep 2017 20:09:25 GMT):
Mike can you see my screen?

mgbailey (Thu, 21 Sep 2017 20:09:45 GMT):
It looks like things are going onto the ledger fine. I suspect that the problem is with the actual line you were executing.

mgbailey (Thu, 21 Sep 2017 20:10:10 GMT):
I would hope that an extra space would not cause an issue, but maybe check that ?

Dan (Thu, 21 Sep 2017 20:10:39 GMT):
With the faber module?

mgbailey (Thu, 21 Sep 2017 20:12:35 GMT):
Looks like the error is "new key" with 2 spaces between the words

anttikettunen (Thu, 21 Sep 2017 20:16:07 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=znmyZWdRG2bqX2GE5) Thanks @danielhardman, that's awesome stuff!

darkcrux (Fri, 22 Sep 2017 19:52:28 GMT):
Has joined the channel.

jljordan_bcgov (Fri, 22 Sep 2017 21:27:16 GMT):
We have the Alice scenario up and running in an OpenShift environment thanks to @WadeBarnes ... I think we will be offering a contribution (pull) to the sovrin-environments repo

WadeBarnes (Fri, 22 Sep 2017 21:27:16 GMT):
Has joined the channel.

jljordan_bcgov (Fri, 22 Sep 2017 21:29:11 GMT):
I'm not sure how much different that is from a Kubernetes deployment environment but we think it will help us with our own development .. particularly with agents and their related web apps

tbrooke (Mon, 25 Sep 2017 18:35:01 GMT):
Has joined the channel.

DibbsZA (Tue, 26 Sep 2017 06:09:27 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=YEF4ijNwP5aSgKYGw) @nage Any possibility of a recording of that?

nage (Tue, 26 Sep 2017 11:57:56 GMT):
We didn’t record it, but just used it to share screens with the front of the room

DibbsZA (Tue, 26 Sep 2017 12:07:13 GMT):
@nage Ah. Ok thanks.

nicolapaoli (Tue, 26 Sep 2017 13:44:02 GMT):
Has joined the channel.

WadeBarnes (Tue, 26 Sep 2017 17:31:04 GMT):
Not sure this is the best place to ask, but since yesterday I've been having trouble with the core.ubuntu.dockerfile build from https://github.com/evernym/sovrin-environments. The build fails due to deps when installing sovrin. Anyone have any recommendations or know of issues introduced in the last few days?

danielhardman (Wed, 27 Sep 2017 09:51:32 GMT):
@mgbailey ^^

neeps (Wed, 27 Sep 2017 12:20:49 GMT):
Has joined the channel.

neeps (Wed, 27 Sep 2017 12:35:46 GMT):
Are there any ETA on when Indy will be "released"?

neeps (Wed, 27 Sep 2017 16:53:58 GMT):
+ is indy by itself a complete blockchain product? or is it primarily for delivering identity components to existing blockchain projects?

sklump (Wed, 27 Sep 2017 18:00:49 GMT):
indy implements a distributed ledger, which nodes on a/the 'plenum' read and write. A distributed ledger is not a blockchain: it requires authorization to write, rather than proof of work.

jstela (Wed, 27 Sep 2017 18:04:23 GMT):
Has joined the channel.

Suedonym (Wed, 27 Sep 2017 18:08:29 GMT):
Anyone here going to the Blockchain for Good Hackathon in Dublin this weekend?

downTheFallLine (Wed, 27 Sep 2017 22:27:36 GMT):
Nathan, are you on?

dmitow (Thu, 28 Sep 2017 09:01:18 GMT):
Has joined the channel.

nage (Thu, 28 Sep 2017 14:45:11 GMT):
@downTheFallLine I am on this morning, sorry I missed your comment last night (Nathan doesn't alert my client)

toddinpal (Thu, 28 Sep 2017 14:59:15 GMT):
Has joined the channel.

danielhardman (Thu, 28 Sep 2017 17:10:55 GMT):
@neeps: Indy has already been released once, in a sense: there is a production network running it (the Sovrin network, which launched July 31, 2017). This was the "Aires" release of Sovrin, and Indy's version numbers were incremented to 1.0 at that time, after a year of preparation and growth. There is likely to be another watershed moment for Indy in a few months, when Sovrin achieves other maturation goals. I don't know how Indy version numbers will change at that time.

danielhardman (Thu, 28 Sep 2017 17:11:36 GMT):
In about 1-2 months, the Indy SDK (not Indy proper, but the libraries that make it easy to use Indy) are hoping for an "Argon" refresh.

ashcherbakov (Thu, 28 Sep 2017 17:17:52 GMT):
@WadeBarnes It's a known issue we've been working on. The current build in master is broken becauase of new dependencies for indy-crypto. It will be fixed soon. I will let you know once it's fixed (should be really soon, waiting for PRs being built) Sorry for inconvenience

ashcherbakov (Thu, 28 Sep 2017 17:17:52 GMT):
@WadeBarnes The master build should work now, but there is a known problem that it uses wrong version of indy-crypto dependency (it will be fixed in 5 minutes). Also, as a workaround, one can use a Stable version (stable repo instead of master)

ashcherbakov (Thu, 28 Sep 2017 18:22:07 GMT):
The new build indy-node 1.1.149 (master) should be fine

WadeBarnes (Thu, 28 Sep 2017 18:32:42 GMT):
@ashcherbakov Thanks!

drewB (Fri, 29 Sep 2017 14:23:54 GMT):
Has joined the channel.

farskipper (Fri, 29 Sep 2017 17:17:50 GMT):
Has joined the channel.

Sean_Bohan (Sat, 30 Sep 2017 07:46:33 GMT):
folks, have a question here about connecting to the test network - I have been able to SSH to the server, but I am unable to utilize vagrant to connect to the cli01. I get the following error: agent@agent01:~$ vagrant ssh cli01 The program 'vagrant' is currently not installed. You can install it by typing: sudo apt install vagrant This is the first step under the Setting Up a CLI Client and Configuring Agents in the Sovrin Cluster section of the GSG. What do I need to do differently?

Sean_Bohan (Sat, 30 Sep 2017 08:03:18 GMT):
Welcome Dublin Blockchain for Good Hackathon Participants! You can find some resources for Hyperledger Indy here: https://wiki.hyperledger.org/projects/oct_2017_dublin_hackathon

lovesh (Sat, 30 Sep 2017 08:03:40 GMT):
@tkuhrt so what happens after you install vagrant?

tkuhrt (Sat, 30 Sep 2017 08:42:14 GMT):
@lovesh : I did not try. I was told to start at the section that does the connect. I did not want to do anything before finding out whether I should try that. So it sounds like the test network machines are not set up yet. Is that correct?

tkuhrt (Sat, 30 Sep 2017 08:42:14 GMT):
@lovesh : I did not try. I was told to start at the section that does the vagrant command. I did not want to do anything before finding out whether I should try that. So it sounds like the test network machines are not set up yet. Is that correct?

lovesh (Sat, 30 Sep 2017 08:43:08 GMT):
Ok, you need to install vagrant by `sudo apt install vagrant`

lovesh (Sat, 30 Sep 2017 08:44:10 GMT):
Are you following this https://github.com/evernym/sovrin-environments/blob/master/vagrant/training/vb-multi-vm/TestSovrinClusterSetup.md ?

tkuhrt (Sat, 30 Sep 2017 08:47:17 GMT):
Okay. I was told to start here: https://github.com/evernym/sovrin-environments/blob/master/vagrant/training/vb-multi-vm/TestSovrinClusterSetup.md#setting-up-a-cli-client-and-configuring-the-agents-in-the-sovrin-cluster. Can you provide the link for where I should start for these test machines?

tkuhrt (Sat, 30 Sep 2017 08:47:52 GMT):
So same document, but part way through. Should I start at the beginning?

lovesh (Sat, 30 Sep 2017 08:49:58 GMT):
yes, i would like to give you an overview of what is happening, you are trying to connect to a machine running the cli which will talk to some other machines running agents and validators, but first you need all those machines set up, doing `vagrant up` does that

tkuhrt (Sat, 30 Sep 2017 08:55:03 GMT):
Okay. That is great. I can definitely do that. Just didn't realize I needed to. Thanks for your help. I am sure there will be a ton of interest on this channel today from the folks here at the Hackathon. There is a lot of interest in identity

tkuhrt (Sat, 30 Sep 2017 08:56:00 GMT):
Bringing this back to the front for the Hackathon participants: https://chat.hyperledger.org/channel/indy?msg=NxkgBcNDfFPwT5sop

lovesh (Sat, 30 Sep 2017 08:57:42 GMT):
Your welcome Tracy.

lovesh (Sat, 30 Sep 2017 08:57:42 GMT):
You're welcome Tracy.

Artemkaaas (Sat, 30 Sep 2017 09:05:41 GMT):
Has joined the channel.

Riussi (Sat, 30 Sep 2017 12:15:04 GMT):
Has joined the channel.

kitty888 (Sat, 30 Sep 2017 12:22:37 GMT):
Has joined the channel.

alwaysgaryorourke (Sat, 30 Sep 2017 12:24:53 GMT):
Has joined the channel.

Benudek (Sat, 30 Sep 2017 13:48:45 GMT):
Has joined the channel.

Benudek (Sat, 30 Sep 2017 13:48:52 GMT):
GM

Sean_Bohan (Sat, 30 Sep 2017 14:01:44 GMT):
hi Benudek

tkuhrt (Sat, 30 Sep 2017 14:02:42 GMT):
@Sean_Bohan : @Benudek will be using the first agent machine that you set up

tkuhrt (Sat, 30 Sep 2017 14:02:58 GMT):
He will definitely have some questions here on next steps after setting up the agent machine

Benudek (Sat, 30 Sep 2017 14:03:06 GMT):
@vikasdanny

vikasdanny (Sat, 30 Sep 2017 14:03:06 GMT):
Has joined the channel.

Benudek (Sat, 30 Sep 2017 14:03:48 GMT):
Vikas is working on it, we are looking for some guidance then how to get into the development process. we wan to share healthrecords of part of members identity.

tkuhrt (Sat, 30 Sep 2017 14:06:00 GMT):
Besides the command line, do we have any examples of application usage of the SDK for Indy that will help the Hackathon participants get up more quickly?

dsurnin (Sat, 30 Sep 2017 14:08:37 GMT):
check the "sample" directory

Sean_Bohan (Sat, 30 Sep 2017 14:11:17 GMT):
I am runnning to get some wifi - back on in a few

vikasdanny (Sat, 30 Sep 2017 14:17:49 GMT):
Hey everyone

vikasdanny (Sat, 30 Sep 2017 14:18:44 GMT):
Are there any ready to clone projects for Hyperledger Indy just like in Composer? Any links would be greatly appreciated.

Benudek (Sat, 30 Sep 2017 14:49:32 GMT):
which sample directory? on github?

Sean_Bohan (Sat, 30 Sep 2017 14:51:21 GMT):
@vikasdanny - I don't believe there are, but let me check

vikasdanny (Sat, 30 Sep 2017 14:52:03 GMT):
@Sean_Bohan Thank you.

dsurnin (Sat, 30 Sep 2017 14:53:01 GMT):
yes? github https://github.com/hyperledger/indy-sdk/tree/master/samples

dsurnin (Sat, 30 Sep 2017 14:53:34 GMT):
tests could also be usfull

dsurnin (Sat, 30 Sep 2017 14:53:34 GMT):
tests could also be usefull

Benudek (Sat, 30 Sep 2017 14:55:48 GMT):
which tests? I checked the pythin example: is there somewhere a functional description, what it does? And how would we deploy it? Vikas has access to a machine

dsurnin (Sat, 30 Sep 2017 14:57:49 GMT):
you can check the "doc" dir

dsurnin (Sat, 30 Sep 2017 14:58:14 GMT):
python tests are here https://github.com/hyperledger/indy-sdk/tree/master/wrappers/python/tests

Sean_Bohan (Sat, 30 Sep 2017 14:58:32 GMT):
@Benudek you can check the getting started guide - https://github.com/hyperledger/indy-node/blob/master/getting-started.md as well as the 2 videos linked here (more descriptive of what INdy is/does and how DIDs and verifiable claims works) found here at the bottom of the page - https://wiki.hyperledger.org/projects/oct_2017_dublin_hackathon

Sean_Bohan (Sat, 30 Sep 2017 14:58:32 GMT):
Benudek you can check the getting started guide - https://github.com/hyperledger/indy-node/blob/master/getting-started.md as well as the 2 videos linked here (more descriptive of what INdy is/does and how DIDs and verifiable claims works) found here at the bottom of the page - https://wiki.hyperledger.org/projects/oct_2017_dublin_hackathon

Sean_Bohan (Sat, 30 Sep 2017 14:59:29 GMT):
@vikasdanny we don't have ready-to-clone projects but I will put that on the to-do list for future sprints

Benudek (Sat, 30 Sep 2017 14:59:53 GMT):
https://github.com/hyperledger/indy-sdk/tree/master/wrappers/python/tests/demo

Benudek (Sat, 30 Sep 2017 15:00:06 GMT):
what about this demo?

Benudek (Sat, 30 Sep 2017 15:02:38 GMT):
@vikasdanny https://github.com/hyperledger/indy-sdk/blob/master/wrappers/python/README.md

dsurnin (Sat, 30 Sep 2017 15:09:49 GMT):
also, for the fast test deployment you check the docker file in "ci" directory

dsurnin (Sat, 30 Sep 2017 15:09:49 GMT):
also, for the fast test deployment you can check the docker file in "ci" directory

Pheobe-Sun (Sat, 30 Sep 2017 15:15:48 GMT):
Has joined the channel.

Sean_Bohan (Sat, 30 Sep 2017 15:54:27 GMT):
@Benudek can you come back to the Zoom room - I have a couple engineers there

Sean_Bohan (Sat, 30 Sep 2017 15:54:31 GMT):
?

Sean_Bohan (Sat, 30 Sep 2017 15:54:35 GMT):
happy to help

Benudek (Sat, 30 Sep 2017 15:54:38 GMT):
sure,

Benudek (Sat, 30 Sep 2017 15:54:42 GMT):
one second

vikasdanny (Sat, 30 Sep 2017 15:54:43 GMT):
Awesome. :)

Benudek (Sat, 30 Sep 2017 15:54:46 GMT):
thanks so much !

Sean_Bohan (Sat, 30 Sep 2017 15:54:51 GMT):
vikasdanny - you are welcome too

Benudek (Sat, 30 Sep 2017 15:55:32 GMT):
one second

Sean_Bohan (Sat, 30 Sep 2017 15:58:19 GMT):
https://zoom.us/j/6466466446 password 10dublin01

Sean_Bohan (Sat, 30 Sep 2017 16:30:14 GMT):
@Benudek I sent you a DM

Sean_Bohan (Sat, 30 Sep 2017 16:30:46 GMT):
Dublin Hackathon Participants - Start Here: https://wiki.hyperledger.org/projects/oct_2017_dublin_hackathon

gudkov (Sat, 30 Sep 2017 17:03:38 GMT):
@Benudek @Sean_Bohan How to start working with indy sdk on MacOS (if you have no access to ubuntu or windows environment) with python: 1. Understand that libindy python package is actually a wrapper for the native libindy library. For MacOS we don’t have a binary package and you need to build libindy from source code. 2. Clone «rc» branch from https://github.com/hyperledger/indy-sdk/tree/rc 3. Build libindy.dylib (native lindy-sdk library) with the following instruction: https://github.com/bdot/indy-sdk/blob/cd0df217fcb4f366143816c8b023410382213c74/doc/mac-build.md 4. Make sure that processes can load ibindy.dylib. Usually, you need to set DYLD_LIBRARY_PATH env var looking to the path that contains libindy.dylib (you can find it in libindy/target/debug) 5. Make sure you have access to Indy Pool. Actually, you have 2 options: - Run pool (on 127.0.0.1 ip address) in Docker with https://github.com/hyperledger/indy-sdk/blob/rc/ci/indy-pool.dockerfile docker file: docker build -f ci/indy-pool.dockerfile -t indy_pool . docker run -itd -p 9701-9709:9701-9709 indy_pool Note that we bind 9701-9709 ports from docker to local ports. I have no direct experience with Docker on Mac, so make sure that it works this way. - Use our test pool. See #gid=0" target="_blank" rel="noopener noreferrer">https://docs.google.com/spreadsheets/d/1Xo3Q43TDrrPxwWds7ZtwgPHjU2SfLlMpj9VdKXYck1o/edit#gid=0 4. Look to python sample: https://github.com/hyperledger/indy-sdk/tree/rc/samples/python. This sample uses python3-indy package from pypi. Make sure that you use the latest stable version without rc or build number suffixes. 5. Connection to pool requires genesis txn file that describes some initial nodes transaction. This sample is already pre-configured to work with pool in docker. If you need to connect to the different pool you need to change node transactions in the method «pool_genesis_txn_data» in https://github.com/hyperledger/indy-sdk/blob/rc/samples/python/src/utils.py file. 6. This sample explains the main libindy usage workflows. So i suggest to look to main.py and find demos to all API parts. For Windows and Ubuntu you can find libindy build instructions here: https://github.com/hyperledger/indy-sdk/blob/master/README.md. Also, there are pre-built packages for windows and ubuntu in repo: https://repo.evernym.com/libindy. Use stable versions.

gudkov (Sat, 30 Sep 2017 17:03:38 GMT):
@Benudek @Sean_Bohan How to start working with indy sdk on MacOS (if you have no access to ubuntu or windows environment) with python: 1. Understand that libindy python package is actually a wrapper for the native libindy library. For MacOS we don’t have a binary package and you need to build libindy from source code. 2. Clone «rc» branch from https://github.com/hyperledger/indy-sdk/tree/rc 3. Build libindy.dylib (native lindy-sdk library) with the following instruction: https://github.com/bdot/indy-sdk/blob/cd0df217fcb4f366143816c8b023410382213c74/doc/mac-build.md 4. Make that processes can load ibindy.dylib. Usually, you need to set DYLD_LIBRARY_PATH env var looking to the path that contains libindy.dylib (you can find it in libindy/target/debug) 5. Make sure you have access to Indy Pool. Actually, you have 2 options: - Run pool (on 127.0.0.1 ip address) in Docker with https://github.com/hyperledger/indy-sdk/blob/rc/ci/indy-pool.dockerfile docker file: docker build -f ci/indy-pool.dockerfile -t indy_pool . docker run -itd -p 9701-9709:9701-9709 indy_pool Note that we bind 9701-9709 ports from docker to local ports. I have no direct experience with Docker on Mac, so make sure that it works this way. - Use our test pool. See #gid=0" target="_blank" rel="noopener noreferrer">https://docs.google.com/spreadsheets/d/1Xo3Q43TDrrPxwWds7ZtwgPHjU2SfLlMpj9VdKXYck1o/edit#gid=0 4. Look to python sample: https://github.com/hyperledger/indy-sdk/tree/rc/samples/python. This sample uses python3-indy package from pypi. Make sure that you use the latest stable version without rc or build number suffixes. 5. Connection to pool requires genesis txn file that describes some initial nodes transaction. This sample is already pre-configured to work with pool in docker. If you need to connect to the different pool you need to change node transactions in the method «pool_genesis_txn_data» in https://github.com/hyperledger/indy-sdk/blob/rc/samples/python/src/utils.py file. 6. This sample explains the main libindy usage workflows. So i suggest to look to main.py and find demos to all API parts. For Windows and Ubuntu you can find libindy build instructions here: https://github.com/hyperledger/indy-sdk/blob/master/README.md. Also, there are pre-built packages for windows and ubuntu in repo: https://repo.evernym.com/libindy. Use stable versions.

Sean_Bohan (Sat, 30 Sep 2017 17:03:52 GMT):
cc @Benudek

Benudek (Sat, 30 Sep 2017 17:19:02 GMT):
thx, let me try

Benudek (Sat, 30 Sep 2017 17:34:23 GMT):
carbo build

Benudek (Sat, 30 Sep 2017 17:34:24 GMT):
"-l" "crypto" "-l" "sqlite3" "-l" "sodium" "-l" "zmq-pw" "-l" "c++" "-l" "System" "-l" "resolv" "-l" "pthread" "-l" "c" "-l" "m" "-dynamiclib" "-Wl,-dylib" = note: Undefined symbols for architecture x86_64: "_je_malloc_usable_size", referenced from: heapsize::heap_size_of_impl::hd63dce4a1e7de8ce in libheapsize-81e9311798c7d6fb.rlib(heapsize-81e9311798c7d6fb.0.o) ld: symbol(s) not found for architecture x86_64 clang: error: linker command failed with exit code 1 (use -v to see invocation) error: aborting due to previous error error: Could not compile `indy`.

Benudek (Sat, 30 Sep 2017 17:34:46 GMT):
can you help @gudkov @Sean_Bohan ?

nage (Sat, 30 Sep 2017 17:38:09 GMT):
@Benudek Are you trying to build for OSX or iOS?

Benudek (Sat, 30 Sep 2017 17:39:51 GMT):
hm? trying this link: https://github.com/bdot/indy-sdk/blob/cd0df217fcb4f366143816c8b023410382213c74/doc/mac-build.md

Benudek (Sat, 30 Sep 2017 17:40:08 GMT):
I do not get this comment: '4. Make sure that processes can load ibindy.dylib. Usually, you need to set DYLD_LIBRARY_PATH env var looking to the path that contains libindy.dylib (you can find it in libindy/target/debug)'

nage (Sat, 30 Sep 2017 17:40:17 GMT):
Ok, looking now.

Benudek (Sat, 30 Sep 2017 17:40:30 GMT):
I cannot see a file liblindy.lib in target/debug

nage (Sat, 30 Sep 2017 17:40:52 GMT):
What do you see under that libindy folder?

gudkov (Sat, 30 Sep 2017 17:42:05 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=9QkESZvor24hkwahK) @Benudek It should be libindy.dylib. Did you ged cargo build without errors?

gudkov (Sat, 30 Sep 2017 17:42:05 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=9QkESZvor24hkwahK) @Benudek It should be libindy.dylib. Did you get cargo build without errors?

Benudek (Sat, 30 Sep 2017 17:42:06 GMT):
drwxr-xr-x 10 bherudek wheel 340 30 Sep 17:29 tests drwxr-xr-x 8 bherudek wheel 272 30 Sep 17:29 src drwxr-xr-x 11 bherudek wheel 374 30 Sep 17:29 include -rw-r--r-- 1 bherudek wheel 313 30 Sep 17:29 docker-compose.yml drwxr-xr-x 9 bherudek wheel 306 30 Sep 17:29 debian drwxr-xr-x 9 bherudek wheel 306 30 Sep 17:29 ci -rw-r--r-- 1 bherudek wheel 976 30 Sep 17:29 build.rs -rw-r--r-- 1 bherudek wheel 1368 30 Sep 17:29 build-libindy-ios.sh -rw-r--r-- 1 bherudek wheel 1913 30 Sep 17:29 Cargo.toml -rw-r--r-- 1 bherudek wheel 36807 30 Sep 17:29 Cargo.lock drwxr-xr-x 3 bherudek wheel 102 30 Sep 18:30 target bherudek-ltm:libindy bherudek$ pwd /Users/bherudek/indy-sdk/libindy

nage (Sat, 30 Sep 2017 17:42:17 GMT):
And I’m guessing the link error came from the cargo build command, correct?

Benudek (Sat, 30 Sep 2017 17:42:26 GMT):
yes

Benudek (Sat, 30 Sep 2017 17:42:29 GMT):
last command

Benudek (Sat, 30 Sep 2017 17:42:31 GMT):
cargo build

Benudek (Sat, 30 Sep 2017 17:42:38 GMT):
I just followed all steps listed before

nage (Sat, 30 Sep 2017 17:44:09 GMT):
@gudkov does this like look like it has a typo to you? export RUST_LOG=indy=trace

Benudek (Sat, 30 Sep 2017 17:44:27 GMT):
Compiling indy v1.0.0 (file:///Users/bherudek/indy-sdk/libindy) note: link against the following native artifacts when linking against this static library note: the order and any duplication can be significant on some platforms, and so may need to be preserved note: library: ssl note: library: crypto note: library: sqlite3 note: library: sodium note: library: zmq-pw note: library: c++ note: library: System note: library: resolv note: library: c note: library: m error: linking with `cc` failed: exit code: 1 | = note: "cc" "-m64" "-L" "/Users/bherudek/.rustup/toolchains/stable-x86_64-apple-darwin/lib/rustlib/x86_64-apple-darwin/lib" "/Users/bherudek/indy-sdk/libindy/target/debug/deps/indy.0.o" "-o" "/Users/bherudek/indy-sdk/libindy/target/debug/deps/libindy.dylib" "/Users/bherudek/i

Benudek (Sat, 30 Sep 2017 17:44:34 GMT):
this is the beginning of the error log

Benudek (Sat, 30 Sep 2017 17:44:40 GMT):
cc linking

Benudek (Sat, 30 Sep 2017 17:46:17 GMT):
= note: Undefined symbols for architecture x86_64: "_je_malloc_usable_size", referenced from: heapsize::heap_size_of_impl::hd63dce4a1e7de8ce in libheapsize-81e9311798c7d6fb.rlib(heapsize-81e9311798c7d6fb.0.o) ld: symbol(s) not found for architecture x86_64 clang: error: linker command failed with exit code 1 (use -v to see invocation)

Benudek (Sat, 30 Sep 2017 17:46:36 GMT):
is this a var name as intended: '_je_malloc_usable_size' ?

gudkov (Sat, 30 Sep 2017 17:46:44 GMT):
What rust and OSX versions do you use?

gudkov (Sat, 30 Sep 2017 17:52:59 GMT):
> _je_malloc_usable_size Looks like you have no jemalloc on your system. I suggest the following: 1. Check that you installed rust with instruction from #1 point 2. You can try to disable jemalloc and use system default allocator isntead. Just add the following lines to the top of libindy/lib.rs file: #![feature(alloc_system)] extern crate alloc_system;

gudkov (Sat, 30 Sep 2017 17:52:59 GMT):
> _je_malloc_usable_size Looks like you have no jemalloc on your system. I suggest the following: 1. Check that you installed rust with instruction from #1 point 2. You can try to disable jemalloc and use system default allocator isntead. Just add the following lines to the top of libindy/src/lib.rs file: #![feature(alloc_system)] extern crate alloc_system;

gudkov (Sat, 30 Sep 2017 17:53:26 GMT):
@Benudek <<

Benudek (Sat, 30 Sep 2017 17:54:00 GMT):
ok, I installed rust with your link: https://www.rust-lang.org/en-US/install.html

Benudek (Sat, 30 Sep 2017 17:54:06 GMT):
let me try what you suggest, thx !

Benudek (Sat, 30 Sep 2017 17:54:49 GMT):
drwxr-xr-x 10 bherudek wheel 340 30 Sep 17:29 tests drwxr-xr-x 8 bherudek wheel 272 30 Sep 17:29 src drwxr-xr-x 11 bherudek wheel 374 30 Sep 17:29 include -rw-r--r-- 1 bherudek wheel 313 30 Sep 17:29 docker-compose.yml drwxr-xr-x 9 bherudek wheel 306 30 Sep 17:29 debian drwxr-xr-x 9 bherudek wheel 306 30 Sep 17:29 ci -rw-r--r-- 1 bherudek wheel 976 30 Sep 17:29 build.rs -rw-r--r-- 1 bherudek wheel 1368 30 Sep 17:29 build-libindy-ios.sh -rw-r--r-- 1 bherudek wheel 1913 30 Sep 17:29 Cargo.toml -rw-r--r-- 1 bherudek wheel 36807 30 Sep 17:29 Cargo.lock drwxr-xr-x 3 bherudek wheel 102 30 Sep 18:30 target bherudek-ltm:libindy bherudek$ pwd /Users/bherudek/indy-sdk/libindy

Benudek (Sat, 30 Sep 2017 17:55:06 GMT):
you mean build.rs? I dont have lib.rs

gudkov (Sat, 30 Sep 2017 17:55:44 GMT):
https://github.com/hyperledger/indy-sdk/blob/rc/libindy/src/lib.rs

gudkov (Sat, 30 Sep 2017 17:58:01 GMT):
You can try to force install jemalloc with the following instruction: https://gist.github.com/zonomasa/481b7534594414b47723 But it is much better to install latest stable Rust compiler and cargo with official instruction.

gudkov (Sat, 30 Sep 2017 18:01:15 GMT):
@Benudek Also you can try to rebut your system.

gudkov (Sat, 30 Sep 2017 18:01:15 GMT):
@Benudek Also you can try to reboot your system.

Benudek (Sat, 30 Sep 2017 18:01:34 GMT):
I did install rust and cargo following latest & greatest?

Benudek (Sat, 30 Sep 2017 18:01:35 GMT):
Compiling indy v1.0.0 (file:///Users/bherudek/indy-sdk/libindy) error: an inner attribute is not permitted in this context --> src/lib.rs:16:3 | 16 | #![feature(alloc_system)] | ^ | = note: inner attributes and doc comments, like `#![no_std]` or `//! My crate`, annotate the item enclosing them, and are usually found at the beginning of source files. Outer attributes and doc comments, like `#[test]` and `/// My function`, annotate the item following them.

Benudek (Sat, 30 Sep 2017 18:01:42 GMT):
tried different positions in the file

gudkov (Sat, 30 Sep 2017 18:02:26 GMT):
Just add this to the top of the file

gudkov (Sat, 30 Sep 2017 18:04:35 GMT):
error[E0554]: #[feature] may not be used on the stable release channel

gudkov (Sat, 30 Sep 2017 18:04:51 GMT):
Looks like this feature isn't supported by stable rust

gudkov (Sat, 30 Sep 2017 18:05:55 GMT):
So we still have the following options: 1. Reboot 2. Try to re-install rust 3. try to install jemalloc with the instruction https://gist.github.com/zonomasa/481b7534594414b47723 4. Check the Rust version

Benudek (Sat, 30 Sep 2017 18:06:12 GMT):
ok, I did (3) and reopened the shell, just trying

Benudek (Sat, 30 Sep 2017 18:06:49 GMT):
still fails

Benudek (Sat, 30 Sep 2017 18:06:56 GMT):
let me reboot, back in some minutes

gudkov (Sat, 30 Sep 2017 18:09:14 GMT):
What is the output of command "rustc --version" on your system?

Benudek (Sat, 30 Sep 2017 18:15:42 GMT):
1 sec

Benudek (Sat, 30 Sep 2017 18:15:57 GMT):
bherudek-ltm:libindy bherudek$ rustc --version rustc 1.20.0 (f3d6973f4 2017-08-27)

gudkov (Sat, 30 Sep 2017 18:16:48 GMT):
1.20 is a correct versiobn

gudkov (Sat, 30 Sep 2017 18:16:48 GMT):
1.20 is a correct version

gudkov (Sat, 30 Sep 2017 18:18:16 GMT):
Did reboot help you?

Benudek (Sat, 30 Sep 2017 18:18:20 GMT):
no

Benudek (Sat, 30 Sep 2017 18:18:38 GMT):
I did this force install jemalloc

gudkov (Sat, 30 Sep 2017 18:19:26 GMT):
Could you just clean target folder and try "cargo build" again?

gudkov (Sat, 30 Sep 2017 18:19:41 GMT):
What OSX version do you use?

Benudek (Sat, 30 Sep 2017 18:20:30 GMT):
ok, one sec

Benudek (Sat, 30 Sep 2017 18:21:05 GMT):
bherudek-ltm:libindy bherudek$ sw_vers -productVersion 10.12.6 bherudek-ltm:libindy bherudek$

gudkov (Sat, 30 Sep 2017 18:22:39 GMT):
Actually i have the same. Will try to build on my home PC

Benudek (Sat, 30 Sep 2017 18:23:31 GMT):
ok, I cleaned the target folder, same error

Benudek (Sat, 30 Sep 2017 18:23:37 GMT):
hhm

Benudek (Sat, 30 Sep 2017 18:24:24 GMT):
we had issues with the test environment, so reverting to a local docker seems like our best option. but we need it for that mac ...

Benudek (Sat, 30 Sep 2017 18:24:32 GMT):
I can work on the other set up steps

Benudek (Sat, 30 Sep 2017 18:24:45 GMT):
but I guess if we dont work this out, we cant demo

Benudek (Sat, 30 Sep 2017 18:25:04 GMT):
you think we can fix this? its actually probably a small stupid thing

Benudek (Sat, 30 Sep 2017 18:25:32 GMT):
do I need some envs vars set for this _je_malloc_usable_size from .bash_profile?

gudkov (Sat, 30 Sep 2017 18:27:17 GMT):
I will try to repeat steps on my Mac. Could you get PC with ubuntu or windows 10 with local docker service?

Benudek (Sat, 30 Sep 2017 18:28:30 GMT):
no, I have a mac and a google notebook

Benudek (Sat, 30 Sep 2017 18:28:35 GMT):
https://gist.github.com/zonomasa/481b7534594414b47723

Benudek (Sat, 30 Sep 2017 18:28:58 GMT):
make install fails:

Benudek (Sat, 30 Sep 2017 18:28:59 GMT):
bherudek-ltm:jemalloc bherudek$ make install /usr/bin/install -c -d /usr/local/bin /usr/bin/install -c -m 755 bin/jemalloc-config /usr/local/bin /usr/bin/install -c -m 755 bin/jemalloc.sh /usr/local/bin /usr/bin/install -c -m 755 bin/jeprof /usr/local/bin /usr/bin/install -c -d /usr/local/include/jemalloc /usr/bin/install -c -m 644 include/jemalloc/jemalloc.h /usr/local/include/jemalloc /usr/bin/install -c -d /usr/local/lib /usr/bin/install -c -m 755 lib/libjemalloc.2.dylib /usr/local/lib ln -sf libjemalloc.2.dylib /usr/local/lib/libjemalloc.dylib /usr/bin/install -c -d /usr/local/lib /usr/bin/install -c -m 755 lib/libjemalloc.a /usr/local/lib /usr/bin/install -c -m 755 lib/libjemalloc_pic.a /usr/local/lib /usr/bin/install -c -d /usr/local/lib/pkgconfig /usr/bin/install -c -m 644 jemalloc.pc /usr/local/lib/pkgconfig /usr/bin/install -c -d /usr/local/share/doc/jemalloc /usr/bin/install -c -m 644 doc/jemalloc.html /usr/local/share/doc/jemalloc install: doc/jemalloc.html: No such file or directory make: *** [install_doc_html] Error 71

Benudek (Sat, 30 Sep 2017 18:29:08 GMT):
not sure if .html matters

gudkov (Sat, 30 Sep 2017 18:32:28 GMT):
I am trying to have the same environment as yours.

Benudek (Sat, 30 Sep 2017 18:32:33 GMT):
https://github.com/paritytech/parity/issues/4836

Benudek (Sat, 30 Sep 2017 18:33:26 GMT):
https://github.com/paritytech/parity/issues/5964

Benudek (Sat, 30 Sep 2017 18:33:33 GMT):
this error seemingly happened before

gudkov (Sat, 30 Sep 2017 18:36:48 GMT):
just executes "Mac-mini:~ vaceslavgudkov$ rustup update"

gudkov (Sat, 30 Sep 2017 18:36:55 GMT):
So now i have rust 1.20 too

gudkov (Sat, 30 Sep 2017 18:37:14 GMT):
Mac-mini:~ vaceslavgudkov$ rustc --version rustc 1.20.0 (f3d6973f4 2017-08-27)

Benudek (Sat, 30 Sep 2017 18:38:21 GMT):
tried this: brew install jemalloc

Benudek (Sat, 30 Sep 2017 18:38:29 GMT):
no difference, still erroring out

Benudek (Sat, 30 Sep 2017 18:39:10 GMT):
ok and can you build it?

Benudek (Sat, 30 Sep 2017 18:39:50 GMT):
its probably about some versions of these tools indeed

gudkov (Sat, 30 Sep 2017 18:41:03 GMT):
in progress

gudkov (Sat, 30 Sep 2017 18:41:24 GMT):
just installed pre-requirements. Updated some to have last versions

gudkov (Sat, 30 Sep 2017 18:41:46 GMT):
clonned rc with Mac-mini:devel vaceslavgudkov$ git clone -b rc https://github.com/hyperledger/indy-sdk.git

gudkov (Sat, 30 Sep 2017 18:42:41 GMT):
Will put results soon

Benudek (Sat, 30 Sep 2017 18:45:07 GMT):
ok

Benudek (Sat, 30 Sep 2017 18:45:44 GMT):
just seeing thisL git clone -b rc https://github.com/hyperledger/indy-sdk.git

Benudek (Sat, 30 Sep 2017 18:45:48 GMT):
rc

Benudek (Sat, 30 Sep 2017 18:45:56 GMT):
maybe that was my fault, let me try !

gudkov (Sat, 30 Sep 2017 18:47:31 GMT):
My results

gudkov (Sat, 30 Sep 2017 18:47:34 GMT):
note: library: ssl note: library: crypto note: library: sqlite3 note: library: sodium note: library: zmq-pw note: library: c++ note: library: System note: library: resolv note: library: c note: library: m Finished dev [unoptimized + debuginfo] target(s) in 148.51 secs Mac-mini:libindy vaceslavgudkov$

gudkov (Sat, 30 Sep 2017 18:47:49 GMT):
So it was compiled without errors

gudkov (Sat, 30 Sep 2017 18:48:31 GMT):
And corresponded binaries are here: Mac-mini:libindy vaceslavgudkov$ ls ./target/debug/ build deps examples incremental libindy.a libindy.dylib libindy.rlib native Mac-mini:libindy vaceslavgudkov$

Benudek (Sat, 30 Sep 2017 18:49:51 GMT):
yes, for me to o!

Benudek (Sat, 30 Sep 2017 18:50:09 GMT):
I jst followed to blindly the doc, you said here though and in the email take that rc branch

Benudek (Sat, 30 Sep 2017 18:50:12 GMT):
so that step works now

Benudek (Sat, 30 Sep 2017 18:50:19 GMT):
will look now at other steps

Benudek (Sat, 30 Sep 2017 18:50:24 GMT):
THANKS A LOT !!!

Benudek (Sat, 30 Sep 2017 18:50:36 GMT):
:smiley: very appreciated

Benudek (Sat, 30 Sep 2017 18:50:52 GMT):
are you available tomorrow to support? If necessary over a call ?

gudkov (Sat, 30 Sep 2017 18:51:36 GMT):
I will be available to support 13-00-15-00 Moskov Time

Benudek (Sat, 30 Sep 2017 18:52:29 GMT):
ok, its 8pm here and moscoe 10pm, that would be 11h our time. great

Benudek (Sat, 30 Sep 2017 18:52:36 GMT):
we just wanna get some demo to work really

Benudek (Sat, 30 Sep 2017 19:38:48 GMT):
docker build -f ci/indy-pool.dockerfile -t indy_pool .

Benudek (Sat, 30 Sep 2017 19:38:52 GMT):
doesnt seem working

Benudek (Sat, 30 Sep 2017 19:40:06 GMT):
bherudek-ltm:libindy bherudek$ cd ci bherudek-ltm:ci bherudek$ ls -ltr total 48 -rwxr-xr-x 1 bherudek wheel 2375 Sep 30 19:46 ubuntu.dockerfile -rwxr-xr-x 1 bherudek wheel 879 Sep 30 19:46 libindy-win-zip-and-upload.sh -rwxr-xr-x 1 bherudek wheel 822 Sep 30 19:46 libindy-rpm-build-and-upload.sh -rwxr-xr-x 1 bherudek wheel 629 Sep 30 19:46 libindy-deb-build-and-upload.sh -rw-r--r-- 1 bherudek wheel 2302 Sep 30 19:46 indy-sdk.spec.in -rwxr-xr-x 1 bherudek wheel 1368 Sep 30 19:46 amazon.dockerfile bherudek-ltm:ci bherudek$

Benudek (Sat, 30 Sep 2017 19:40:27 GMT):
I cant find this file: ci/indy-pool.dockerfile

Benudek (Sat, 30 Sep 2017 19:43:33 GMT):
I created the file from github manually locally

Benudek (Sat, 30 Sep 2017 19:43:34 GMT):
bherudek-ltm:libindy bherudek$ docker build -f ci/indy-pool.dockerfile -t indy_pool . Sending build context to Docker daemon 358.5MB Step 1/25 : FROM ubuntu:16.04 Get https://registry-1.docker.io/v2/library/ubuntu/manifests/16.04: unauthorized: incorrect username or password bherudek-ltm:libindy bherudek$

wightman (Sat, 30 Sep 2017 20:00:31 GMT):
looks like you might need to login to docker: https://github.com/docker/hub-feedback/issues/935

Benudek (Sat, 30 Sep 2017 20:20:13 GMT):
ok, that one I solved.

Benudek (Sat, 30 Sep 2017 20:20:15 GMT):
E: Version '1.1.112' for 'indy-plenum' was not found E: Version '1.0.25' for 'indy-anoncreds' was not found E: Version '1.1.119' for 'indy-node' was not found The command '/bin/sh -c apt-get update -y && apt-get install -y indy-plenum=${indy_plenum_ver} indy-anoncreds=${indy_anoncreds_ver} indy-node=${indy_node_ver}' returned a non-zero code: 100

Benudek (Sat, 30 Sep 2017 20:20:37 GMT):
it seems to me I need to first install all of this? somehow maybe your version numbers are not correct here?

Benudek (Sat, 30 Sep 2017 20:20:50 GMT):
so am thinking of installing them all manually

devin-fisher (Sat, 30 Sep 2017 20:31:13 GMT):
I'm coming into this behind but do you know what repository you are targeting? Those version number don't look right for stable. They may be right for master? Is that what you want?

Benudek (Sat, 30 Sep 2017 20:31:49 GMT):
I am using rc

Benudek (Sat, 30 Sep 2017 20:32:12 GMT):
How to start working with indy sdk on MacOS (if you have no access to ubuntu or windows environment) with python: 1. Understand that libindy python package is actually a wrapper for the native libindy library. For MacOS we don’t have a binary package and you need to build libindy from source code. 2. Clone «rc» branch from https://github.com/hyperledger/indy-sdk/tree/rc 3. Build libindy.dylib (native lindy-sdk library) with the following instruction: https://github.com/bdot/indy-sdk/blob/cd0df217fcb4f366143816c8b023410382213c74/doc/mac-build.md 4. Make that processes can load ibindy.dylib. Usually, you need to set DYLD_LIBRARY_PATH env var looking to the path that contains libindy.dylib (you can find it in libindy/target/debug) 5. Make sure you have access to Indy Pool. Actually, you have 2 options: - Run pool (on 127.0.0.1 ip address) in Docker with https://github.com/hyperledger/indy-sdk/blob/rc/ci/indy-pool.dockerfile docker file: docker build -f ci/indy-pool.dockerfile -t indy_pool . docker run -itd -p 9701-9709:9701-9709 indy_pool Note that we bind 9701-9709 ports from docker to local ports. I have no direct experience with Docker on Mac, so make sure that it works this way. - Use our test pool. See #gid=0" target="_blank" rel="noopener noreferrer">https://docs.google.com/spreadsheets/d/1Xo3Q43TDrrPxwWds7ZtwgPHjU2SfLlMpj9VdKXYck1o/edit#gid=0 4. Look to python sample: https://github.com/hyperledger/indy-sdk/tree/rc/samples/python. This sample uses python3-indy package from pypi. Make sure that you use the latest stable version without rc or build number suffixes. 5. Connection to pool requires genesis txn file that describes some initial nodes transaction. This sample is already pre-configured to work with pool in docker. If you need to connect to the different pool you need to change node transactions in the method «pool_genesis_txn_data» in https://github.com/hyperledger/indy-sdk/blob/rc/samples/python/src/utils.py file. 6. This sample explains the main libindy usage workflows. So i suggest to look to main.py and find demos to all API parts. For Windows and Ubuntu you can find libindy build instructions here: https://github.com/hyperledger/indy-sdk/blob/master/README.md. Also, there are pre-built packages for windows and ubuntu in repo: https://repo.evernym.com/libindy. Use stable versions.

Benudek (Sat, 30 Sep 2017 20:32:50 GMT):
following these instructions and getting stuck at step 1st command

Benudek (Sat, 30 Sep 2017 20:32:58 GMT):
docker build -f ci/indy-pool.dockerfile -t indy_pool

Benudek (Sat, 30 Sep 2017 20:33:22 GMT):
but not I created that file manually (downloaded): https://github.com/hyperledger/indy-sdk/blob/rc/ci/indy-pool.dockerfile

Benudek (Sat, 30 Sep 2017 20:33:37 GMT):
and then run docker build -f ci/indy-pool.dockerfile -t indy_pool .

Benudek (Sat, 30 Sep 2017 20:37:48 GMT):
1. bherudek-ltm:indy-sdk bherudek$ ls -ltr ci total 8 -rwxr-xr-x 1 bherudek wheel 2752 Sep 30 19:46 indy-pool.dockerfile bherudek-ltm:indy-sdk bherudek$ docker build -f ci/indy-pool.dockerfile -t indy_pool .

Benudek (Sat, 30 Sep 2017 20:38:22 GMT):
---> Running in df250e15174e Get:1 http://security.ubuntu.com/ubuntu xenial-security InRelease [102 kB] Hit:2 http://archive.ubuntu.com/ubuntu xenial InRelease Get:3 http://archive.ubuntu.com/ubuntu xenial-updates InRelease [102 kB] Get:4 http://archive.ubuntu.com/ubuntu xenial-backports InRelease [102 kB] Ign:5 https://repo.sovrin.org/deb xenial InRelease Get:6 https://repo.sovrin.org/deb xenial Release [46.4 kB] Get:7 https://repo.sovrin.org/deb xenial Release.gpg [819 B] Get:8 https://repo.sovrin.org/deb xenial/stable amd64 Packages [12.5 kB] Fetched 366 kB in 1s (198 kB/s) Reading package lists... Reading package lists... Building dependency tree... Reading state information... E: Version '1.1.112' for 'indy-plenum' was not found E: Version '1.0.25' for 'indy-anoncreds' was not found E: Version '1.1.119' for 'indy-node' was not found The command '/bin/sh -c apt-get update -y && apt-get install -y indy-plenum=${indy_plenum_ver} indy-anoncreds=${indy_anoncreds_ver} indy-node=${indy_node_ver}' returned a non-zero code: 100

devin-fisher (Sat, 30 Sep 2017 20:39:26 GMT):
ok, yeah the docker file looks wrong. It points to the stable deb repo but the versions don't exists in the repo.

Benudek (Sat, 30 Sep 2017 20:39:54 GMT):
ok, can you give a new one soon? or just tell me what to change? I can edit locally

devin-fisher (Sat, 30 Sep 2017 20:40:11 GMT):
I can give you the latest version in stable?

Benudek (Sat, 30 Sep 2017 20:40:21 GMT):
if that works, great ... :-)

devin-fisher (Sat, 30 Sep 2017 20:41:18 GMT):
indy-anoncreds_1.0.10 indy-node_1.1.37 indy-plenum_1.1.27

devin-fisher (Sat, 30 Sep 2017 20:42:09 GMT):
Should be able to edit the docker file to these versions. lines 27-29.

Benudek (Sat, 30 Sep 2017 20:44:43 GMT):
ok, let me check

Benudek (Sat, 30 Sep 2017 20:46:32 GMT):
Building dependency tree... Reading state information... E: Version '' for 'indy-plenum' was not found E: Version '' for 'indy-anoncreds' was not found E: Version '' for 'indy-node' was not found

Benudek (Sat, 30 Sep 2017 20:48:16 GMT):
can we give a tag like give the latest stable version instead?

devin-fisher (Sat, 30 Sep 2017 20:49:50 GMT):
Sorry, I don't work with this code base. Let me try running it on my mac and see what I get.

Benudek (Sat, 30 Sep 2017 20:51:01 GMT):
ok

devin-fisher (Sat, 30 Sep 2017 21:11:17 GMT):
alright I modified the docker file and it worked for me. But I can't get rocket chat to send it to you. Do you have another way I can get it to you?

Benudek (Sat, 30 Sep 2017 21:12:34 GMT):
benedikt.herudek@gmail.com

devin-fisher (Sat, 30 Sep 2017 21:12:41 GMT):

Message Attachments

Benudek (Sat, 30 Sep 2017 21:13:06 GMT):
am btw also trying following the getting started guide with setting up a local network with vagrant, one should work I guess ... :-) I prefer the docker, it looks easier

devin-fisher (Sat, 30 Sep 2017 21:13:15 GMT):
rocket chat don't like .dockerfile I guess. Added .txt at the end and it took it.

Benudek (Sat, 30 Sep 2017 21:16:03 GMT):
yes, ran it seems fine. continuing the steps, thanks a lot !

devin-fisher (Sat, 30 Sep 2017 21:16:25 GMT):
yeah no problem. Sorry its been a pain.

Benudek (Sat, 30 Sep 2017 21:17:30 GMT):
no worries, its great software

Benudek (Sat, 30 Sep 2017 21:17:30 GMT):
- Use our test pool. See #gid=0" target="_blank" rel="noopener noreferrer">https://docs.google.com/spreadsheets/d/1Xo3Q43TDrrPxwWds7ZtwgPHjU2SfLlMpj9VdKXYck1o/edit#gid=0 4. Look to python sample: https://github.com/hyperledger/indy-sdk/tree/rc/samples/python. This sample uses python3-indy package from pypi. Make sure that you use the latest stable version without rc or build number suffixes. 5. Connection to pool requires genesis txn file that describes some initial nodes transaction. This sample is already pre-configured to work with pool in docker. If you need to connect to the different pool you need to change node transactions in the method «pool_genesis_txn_data» in https://github.com/hyperledger/indy-sdk/blob/rc/samples/python/src/utils.py file. 6. This sample explains the main libindy usage workflows. So i suggest to look to main.py and find demos to all API parts.

Benudek (Sat, 30 Sep 2017 21:17:40 GMT):
I cant make too much out of these things here though ... :-)

Benudek (Sat, 30 Sep 2017 21:17:58 GMT):
why would I use the test pool if I have a local dockerfile?

Benudek (Sat, 30 Sep 2017 21:19:20 GMT):
and how can I use the python now? I think I install it locally and then via pool_genesis_txn_data it connects the running docker image?

devin-fisher (Sat, 30 Sep 2017 21:24:26 GMT):
sorry, I came into the middle of the conversation. What is your end goal right at the moment. The docker file stands up a local indy network. A network you can connect to with the indy-sdk. But it looks like based on the spreed sheet that someone setup a indy network for the hackfest?

Benudek (Sat, 30 Sep 2017 21:25:33 GMT):
yes, but we didnt het the vagrant up

Benudek (Sat, 30 Sep 2017 21:25:48 GMT):
thats why we diverted to docker, so not sure why this file is in here/

Benudek (Sat, 30 Sep 2017 21:26:22 GMT):
I just need instructions how to now interact with that docker image, which is running. have this pythin code local and that connects to my docker then?

devin-fisher (Sat, 30 Sep 2017 21:28:37 GMT):
you should be ready to connect to the network.

Benudek (Sat, 30 Sep 2017 21:32:11 GMT):
I know, but how :-)

Benudek (Sat, 30 Sep 2017 21:32:30 GMT):
with this from the guide? sovrin> prompt ALICE

devin-fisher (Sat, 30 Sep 2017 21:34:18 GMT):
Are wanting to use the python version of libindy?

vikasdanny (Sat, 30 Sep 2017 21:35:23 GMT):
python

vikasdanny (Sat, 30 Sep 2017 21:35:31 GMT):
Me and

vikasdanny (Sat, 30 Sep 2017 21:35:43 GMT):
@Benudek are working together

Benudek (Sat, 30 Sep 2017 21:36:09 GMT):
yes, python

devin-fisher (Sat, 30 Sep 2017 21:37:32 GMT):
you should be good to go. Have you pip installed the python3-indy module?

Benudek (Sat, 30 Sep 2017 21:39:44 GMT):
we are trying to follow this guide:

Benudek (Sat, 30 Sep 2017 21:39:45 GMT):
4. Look to python sample: https://github.com/hyperledger/indy-sdk/tree/rc/samples/python. This sample uses python3-indy package from pypi. Make sure that you use the latest stable version without rc or build number suffixes. 5. Connection to pool requires genesis txn file that describes some initial nodes transaction. This sample is already pre-configured to work with pool in docker. If you need to connect to the different pool you need to change node transactions in the method «pool_genesis_txn_data» in https://github.com/hyperledger/indy-sdk/blob/rc/samples/python/src/utils.py file. 6. This sample explains the main libindy usage workflows. So i suggest to look to main.py and find demos to all API parts.

Benudek (Sat, 30 Sep 2017 21:40:02 GMT):
installed python3?

Benudek (Sat, 30 Sep 2017 21:40:11 GMT):
pythin3-indy module? how?

Benudek (Sat, 30 Sep 2017 21:40:58 GMT):
how can we use the commandline? https://github.com/hyperledger/indy-node/blob/master/getting-started.md

devin-fisher (Sat, 30 Sep 2017 21:43:08 GMT):
pip3 install python3-indy

Benudek (Sat, 30 Sep 2017 21:43:53 GMT):
ok, that works. And then?

devin-fisher (Sat, 30 Sep 2017 21:45:18 GMT):
have you down loaded the sample/python code?

devin-fisher (Sat, 30 Sep 2017 21:46:30 GMT):
move into the src directory and run the main.py file. Should be: python3 main.py

Benudek (Sat, 30 Sep 2017 21:47:38 GMT):
ok, thx trying

devin-fisher (Sat, 30 Sep 2017 21:48:55 GMT):
cool let me know how it goes.

Benudek (Sat, 30 Sep 2017 21:49:41 GMT):
bherudek-ltm:src bherudek$ ls -ltr total 48 -rw-r--r-- 1 bherudek wheel 1999 Sep 30 22:48 utils.py -rw-r--r-- 1 bherudek wheel 2075 Sep 30 22:48 signus.py -rw-r--r-- 1 bherudek wheel 358 Sep 30 22:48 main.py -rw-r--r-- 1 bherudek wheel 2702 Sep 30 22:48 ledger.py -rw-r--r-- 1 bherudek wheel 3808 Sep 30 22:48 anoncreds.py -rw-r--r-- 1 bherudek wheel 3783 Sep 30 22:48 agent.py bherudek-ltm:src bherudek$ python3 main.py Traceback (most recent call last): File "main.py", line 4, in from src import anoncreds, signus, ledger, agent ModuleNotFoundError: No module named 'src' bherudek-ltm:src bherudek$ pwd /Users/bherudek/indy-sdk/indy-sdk/samples/python/src bherudek-ltm:src bherudek$

devin-fisher (Sat, 30 Sep 2017 21:53:47 GMT):
ok, thats a python path issue. Let me try on my end.

Benudek (Sat, 30 Sep 2017 21:54:32 GMT):
ah wait I will look at that

Benudek (Sat, 30 Sep 2017 21:54:39 GMT):
if its my local thing

devin-fisher (Sat, 30 Sep 2017 22:01:21 GMT):
If you have pycharm, it might be easier to get the python path correct

Benudek (Sat, 30 Sep 2017 22:07:05 GMT):
have it, so you want me to open main in the IDE?

devin-fisher (Sat, 30 Sep 2017 22:07:34 GMT):
I'm not totally sure how they wanted it run but if you add this enviroment variable it seems to get past your issue.

devin-fisher (Sat, 30 Sep 2017 22:07:36 GMT):
export PYTHONPATH=/home/devin/Downloads/indy-sdk-rc/samples/python/

devin-fisher (Sat, 30 Sep 2017 22:07:51 GMT):
change to the directory to match your system.

Benudek (Sat, 30 Sep 2017 22:08:28 GMT):
not so far but let me open it in pycharm

Benudek (Sat, 30 Sep 2017 22:08:28 GMT):
bherudek-ltm:src bherudek$ python3 main.py Traceback (most recent call last): File "main.py", line 4, in from src import anoncreds, signus, ledger, agent ModuleNotFoundError: No module named 'src' bherudek-ltm:src bherudek$

devin-fisher (Sat, 30 Sep 2017 22:09:27 GMT):
did you add the environment variable?

Benudek (Sat, 30 Sep 2017 22:09:35 GMT):
yes

devin-fisher (Sat, 30 Sep 2017 22:09:51 GMT):
run echo "$PYTHONPATH"

devin-fisher (Sat, 30 Sep 2017 22:09:57 GMT):
what do you get?

Benudek (Sat, 30 Sep 2017 22:11:30 GMT):
bherudek-ltm:src bherudek$ echo "$PYTHONPATH" /home/devin/Downloads/indy-sdk-rc/samples/python/

devin-fisher (Sat, 30 Sep 2017 22:12:46 GMT):
you got to change the command I gave you. export PYTHONPATH=

devin-fisher (Sat, 30 Sep 2017 22:13:26 GMT):
that was directory for my machine. Point it to your location on your machine.

Benudek (Sat, 30 Sep 2017 22:13:28 GMT):
aah

Benudek (Sat, 30 Sep 2017 22:13:30 GMT):
sure

Benudek (Sat, 30 Sep 2017 22:13:40 GMT):
just copying doing this start guide in parallel

devin-fisher (Sat, 30 Sep 2017 22:13:48 GMT):
no worries

Benudek (Sat, 30 Sep 2017 22:15:04 GMT):
bherudek-ltm:src bherudek$ python3 main.py INFO:src.anoncreds:Anoncreds sample -> started ERROR:indy.libindy:_load_cdll: Can't load libindy: dlopen(libindy.dylib, 6): image not found Traceback (most recent call last): File "main.py", line 15, in loop.run_until_complete(main()) File "/Library/Frameworks/Python.framework/Versions/3.6/lib/python3.6/asyncio/base_events.py", line 467, in run_until_complete return future.result() File "main.py", line 8, in main await anoncreds.demo() File "/Users/bherudek/indy-sdk/samples/python/src/anoncreds.py", line 18, in demo await wallet.create_wallet(pool_name, wallet_name, None, None, None) File "/Library/Frameworks/Python.framework/Versions/3.6/lib/python3.6/site-packages/indy/wallet.py", line 52, in create_wallet create_wallet.cb) File "/Library/Frameworks/Python.framework/Versions/3.6/lib/python3.6/site-packages/indy/libindy.py", line 24, in do_call err = getattr(_cdll(), name)(command_handle, File "/Library/Frameworks/Python.framework/Versions/3.6/lib/python3.6/site-packages/indy/libindy.py", line 87, in _cdll _cdll.cdll = _load_cdll() File "/Library/Frameworks/Python.framework/Versions/3.6/lib/python3.6/site-packages/indy/libindy.py", line 118, in _load_cdll raise e File "/Library/Frameworks/Python.framework/Versions/3.6/lib/python3.6/site-packages/indy/libindy.py", line 113, in _load_cdll res = CDLL(libindy_name) File "/Library/Frameworks/Python.framework/Versions/3.6/lib/python3.6/ctypes/__init__.py", line 348, in __init__ self._handle = _dlopen(self._name, mode) OSError: dlopen(libindy.dylib, 6): image not found bherudek-ltm:src bherudek$

Benudek (Sat, 30 Sep 2017 22:15:15 GMT):
journey continues ... ;-)

devin-fisher (Sat, 30 Sep 2017 22:16:19 GMT):
did you set the DYLD_LIBRARY_PATH to point to the location of libindy.dylib that you built?

Benudek (Sat, 30 Sep 2017 22:19:22 GMT):
dont think so, not sure it was mentioned in setup steps

devin-fisher (Sat, 30 Sep 2017 22:21:06 GMT):
4. Make that processes can load ibindy.dylib. Usually, you need to set DYLD_LIBRARY_PATH env var looking to the path that contains libindy.dylib (you can find it in libindy/target/debug)

Benudek (Sat, 30 Sep 2017 22:21:25 GMT):
ah yes, I had asked what that would mean ... :-)

devin-fisher (Sat, 30 Sep 2017 22:23:17 GMT):
export DYLD_LIBRARY_PATH=/path/to/imagemagick/library:${DYLD_LIBRARY_PATH}

devin-fisher (Sat, 30 Sep 2017 22:23:34 GMT):
again change the directory to the location of the libindy.dylib file.

Benudek (Sat, 30 Sep 2017 22:24:10 GMT):
so the pythin script seems to work now

Benudek (Sat, 30 Sep 2017 22:24:18 GMT):
do you have a gmail address?

devin-fisher (Sat, 30 Sep 2017 22:24:25 GMT):
cool

devin-fisher (Sat, 30 Sep 2017 22:24:33 GMT):
yeah

Benudek (Sat, 30 Sep 2017 22:24:42 GMT):
I try to create a short google doc with the setup, this is a lot of details over different chats ... :-)

devin-fisher (Sat, 30 Sep 2017 22:24:52 GMT):
yeah

devin-fisher (Sat, 30 Sep 2017 22:25:00 GMT):
dev.fisher@gmail.com

Benudek (Sat, 30 Sep 2017 22:26:32 GMT):
https://docs.google.com/presentation/d/1bo67xxDXMtRrY610qHkzhFYere9bDSZcwgRGKWYJKFg/edit?usp=sharing

devin-fisher (Sat, 30 Sep 2017 22:27:54 GMT):
Cool

devin-fisher (Sat, 30 Sep 2017 22:28:26 GMT):
I've got to get off. Sorry again that it has been a real pain.

Benudek (Sat, 30 Sep 2017 22:29:12 GMT):
thanks for your help

devin-fisher (Sat, 30 Sep 2017 22:30:48 GMT):
Hope you get chance to see a bit Ireland while you are there. That is if you are not from there.

devin-fisher (Sat, 30 Sep 2017 22:32:53 GMT):
Good luck, I'll check in hour or so and see if I can help.

Benudek (Sun, 01 Oct 2017 10:00:02 GMT):
hey there,

Benudek (Sun, 01 Oct 2017 10:00:10 GMT):
good morming from beautiful dublin!

Benudek (Sun, 01 Oct 2017 10:00:25 GMT):
working on the python API

spivachuk (Sun, 01 Oct 2017 10:00:25 GMT):
Hi Benudek!

Benudek (Sun, 01 Oct 2017 10:00:40 GMT):

Message Attachments

Benudek (Sun, 01 Oct 2017 10:00:44 GMT):
and was wondering if there is additional explanatin for these examples!

Benudek (Sun, 01 Oct 2017 10:01:01 GMT):
especially for anoncreds

Benudek (Sun, 01 Oct 2017 10:01:27 GMT):
We want to write a little example, where a refugee can share identity and health records.

Benudek (Sun, 01 Oct 2017 10:01:35 GMT):
Verifier would be an NGO

Benudek (Sun, 01 Oct 2017 10:02:05 GMT):
and he would need that statement, when visiting authorities and hospitals in the country where (s)he seeks asylum.

Benudek (Sun, 01 Oct 2017 10:02:30 GMT):
any tips as to that? Will just 'start hacking' ...:-) Any pointers welcome

gudkov (Sun, 01 Oct 2017 10:07:56 GMT):
@Benudek Hi, do you have any knowledge in zero knowledge proofs?

Benudek (Sun, 01 Oct 2017 10:09:50 GMT):
aah, conceptually yes!

Benudek (Sun, 01 Oct 2017 10:10:10 GMT):
I know the concept, I dont get the crypto stuff yet ... :-)

Benudek (Sun, 01 Oct 2017 10:10:30 GMT):
would I need that here?

gudkov (Sun, 01 Oct 2017 10:11:02 GMT):
https://github.com/hyperledger/indy-sdk/blob/master/doc/libindy-anoncreds.puml There is a diagram that explains anoncreds API workflow

Benudek (Sun, 01 Oct 2017 10:11:16 GMT):
could we have a necessary like: refugee gets a claim ' yes, I was vaccinated so I can pass the border' but he will NOT reveal his entire health records to the authorities?

Benudek (Sun, 01 Oct 2017 10:11:20 GMT):
that would be cool

Benudek (Sun, 01 Oct 2017 10:11:27 GMT):
it would show how indy can make a difference

gudkov (Sun, 01 Oct 2017 10:11:58 GMT):
Do you know how to render plant uml?

Benudek (Sun, 01 Oct 2017 10:12:12 GMT):
no, was just gonna ask ... :-)

gudkov (Sun, 01 Oct 2017 10:13:50 GMT):
http://www.plantuml.com/plantuml/uml/SyfFKj2rKt3CoKnELR1Io4ZDoSa70000

gudkov (Sun, 01 Oct 2017 10:14:44 GMT):
You can render it online here or install plugin for ide

Benudek (Sun, 01 Oct 2017 10:16:12 GMT):
ok, all I get though is this generic example?

Benudek (Sun, 01 Oct 2017 10:16:16 GMT):

Message Attachments

Benudek (Sun, 01 Oct 2017 10:17:46 GMT):
one sec

Benudek (Sun, 01 Oct 2017 10:17:50 GMT):
wrong button .. :-)

gudkov (Sun, 01 Oct 2017 10:19:52 GMT):
http://www.plantuml.com/plantuml/svg/jLPjRzis4FwkNt58Fgm5qZWlwxOQiORKSJi13YguhVnXA0HAecqYakYGj1q3-V4xlaYMOjbRMoo01OMyU-xkkUT8oGTMBeaW1OH4A9Qo9IbLIBACbNEukl1alV-UFpNMFSLKJc7C4bPcMo0bBrD1C-d9bE3wnVaxp_CI_WGdWPGhAaxWTCpc2_K-Nr5lkGi59rEIvrJH32f38Y6OUYHLNBWri-JHCaD0XneAlaPZPsR_qsRX0N1dRqocDPJrZgXWCsRz8wyDCARPtFcDiQAIAmecLZ0zWp6SRXPCKG_mBoDoWO6847mZ-q1WJR1Mc0bUW4M1WcmXeP2IL5k-OYLiJ7v_lZO8DucVy3m3HP_KJI_n-pjQ8NT4tuO1-kUvTPXnp37bEHS3jFBnrx_fwKKVMAceA4cEsqMnAbbA5ECOaAc5Yuefq1BRkDt3cWjAIzyyQpovIWgcUdEgGF6UgKhIT2MeGzi3IR1wNp-EGLXYZ6-Ia_IX_oBTXAeDPk5qqBDTx5KDytGZTKkmCuF108Q24kpFC2UiW3iwOoKpA8E15y0R7tlMroHVdeFcCcO6D4wDLqpAB9Oe1ngdsZ_EwCmBfl3cxrD3n3VhRYpWWMwbsSH5XAmTpg1-ZLShblcU5o20Pk6RFqPlp1U6uzlWFhuRVG3rL7hZUhz6WgnENVuYqUanBhc6QjHmvMgmvCVI_BHTFECYDjBho7wrp3H1TP8EkyNUXAwvOtz2vqmgiTK4jg9V_mk3GcFWSzwHWsv_pLyNkSOPAcVBS0ru1lNhOOOde-ZJC9wCFdOmNBiVOlRNds4MqPS6lOlPUo9n9lNO2EH0SnhgkA6D-_RSX5qQYLUK5iObjYvUs0eRremLJM0RyhDHcb7JfzaCDrqBDHKuPzhTaMKbnxLHAn_upH7XCchEXhC-a4u3KRfi5dpJp4OYCdmDzZZulSK1AgKPJtO4r69gXGkDPC9wAEsoxw9WXZeJheFRp2RqSKMbmfFMNK0sJQbJ2w5I3lfMsleu-XBl9bZGHs_SD0ixXjXl5_b7LwDVdZuQkBlSThEFzg16QJA2QpQXdHI2bXI4fQGz5Uyw9zuuVD-Cwwsk2TyDaDtoprp2VgrRFSAmD-Fnr_3kzil8ZFfXWRTRS7bnVa3QHpLMCNEKz9UAbrPkJb2x52xRurMB_QZCzmH-HDgXe5Y_SdEdE5ndvYDl2wym9fzLvRpMKnXrCTSoyvwVdziLN7LFgpG7fV7j4g04t6v3TzPGBz6SfbNvnbSjgFMlKkgUqi_3lqRtjzqZAl4Bh23-pusula6FGEnkeKw8-YBozvczxZeYRCb7nD5WsjZwJNviPwDcPyrszBtqHGtweX-WBshJ5_q0VT4-VH6uTMIlki7NxZFHlcmTWUxON_E7vbdV-ZvyspxiX37-Nm-x_tp4_GC0

Benudek (Sun, 01 Oct 2017 10:22:39 GMT):
ok, seeing the diagram but not fully sure what is functionally happening here. Is there a textual description

gudkov (Sun, 01 Oct 2017 10:30:26 GMT):
I will check for available high level documents. also in about 10 minutes there will be Alex. He has deep expirience in anoncreds.

gudkov (Sun, 01 Oct 2017 10:33:39 GMT):
So i suggest to start with definition of roles and mapping of this to our api entities.

ashcherbakov (Sun, 01 Oct 2017 10:36:41 GMT):
Hi @Benudek

ashcherbakov (Sun, 01 Oct 2017 10:37:12 GMT):
So, would you like a high level description of anoncreds protocol^

ashcherbakov (Sun, 01 Oct 2017 10:37:12 GMT):
So, would you like a high level description of anoncreds protocol?

Benudek (Sun, 01 Oct 2017 10:38:32 GMT):
yes, actually ideally we would get some help to adjust anoncreds to a situation where a refugee claims to the country where he seeked asylum that he has a certain vaccination. NGO approves the claim and the refugee gets that claim to pass the border without revealing the entire health history

Benudek (Sun, 01 Oct 2017 10:38:44 GMT):
is a zoom call an option?

Benudek (Sun, 01 Oct 2017 10:38:48 GMT):
just need to find a room

ashcherbakov (Sun, 01 Oct 2017 10:39:17 GMT):
yes, one momnet

ashcherbakov (Sun, 01 Oct 2017 10:41:23 GMT):
zoom 6466466446 password 10dublin01

ashcherbakov (Sun, 01 Oct 2017 10:41:54 GMT):
@Benudek ^

Benudek (Sun, 01 Oct 2017 10:43:00 GMT):
ok, 1 minute pls

Benudek (Sun, 01 Oct 2017 10:43:01 GMT):
coming

Benudek (Sun, 01 Oct 2017 10:44:41 GMT):
dilaing in

ashcherbakov (Sun, 01 Oct 2017 11:01:36 GMT):
@Benudek are you here?

Benudek (Sun, 01 Oct 2017 11:01:43 GMT):
coming back

Benudek (Sun, 01 Oct 2017 12:15:41 GMT):
date":"2","vaccinationOnRabisDate":"20"},"schema_seq_no":1,"issuer_did":"NcYxiDXkpYi6ov5FcYDi1e"}]},"predicates":{"predicate1_uuid":[{"claim_uuid":"claim::305cad52-204d-48d6-80f1-3e333dd66d55","attrs":{"vaccinationOnRabis":"1","personid":"2823423432","location":"123123","diseases":"23","name":"1139481716457488690172217916278103335","sex":"0","age":"52","birthdate":"2","vaccinationOnRabisDate":"20"},"schema_seq_no":1,"issuer_did":"NcYxiDXkpYi6ov5FcYDi1e"}]}}' INFO|command_executor | src/commands/mod.rs:91 | AnoncredsCommand command received INFO|anoncreds_command_executor | src/commands/anoncreds/mod.rs:48 | Prover command received INFO|prover_command_executor | src/commands/anoncreds/prover.rs:131 | CreateProof command received INFO|anoncreds_service | src/services/anoncreds/prover.rs:283 | Prover create proof -> start INFO|anoncreds_service | src/services/anoncreds/prover.rs:390 | Prover init primary proof -> start ERROR|indy::errors::indy | src/errors/indy.rs:63 | Casting error to ErrorCode: Invalid structure: Value by key 'vaccinationOnRabisDate' has invalid format WARNING:indy.libindy:_indy_loop_callback: Function returned error 113 Traceback (most recent call last): File "main.py", line 17, in loop.run_until_complete(main()) File "/Library/Frameworks/Python.framework/Versions/3.6/lib/python3.6/asyncio/base_events.py", line 467, in run_until_complete return future.result() File "main.py", line 10, in main await healthclaim.demo() File "/Users/bherudek/indy-sdk/indy-sdk/samples/python/src/healthclaim.py", line 141, in demo 'master_secret', claim_defs_json, revoc_regs_json) File "/Library/Frameworks/Python.framework/Versions/3.6/lib/python3.6/site-packages/indy/anoncreds.py", line 640, in prover_create_proof prover_create_proof.cb) indy.error.IndyError: ErrorCode.CommonInvalidStructure

Benudek (Sun, 01 Oct 2017 12:22:06 GMT):
THANKS A LOT !!!

ashcherbakov (Sun, 01 Oct 2017 12:28:27 GMT):
Welcome! We are really happy that you enjoed the technology. Feel free to reach us if you have any questions/problems

Benudek (Sun, 01 Oct 2017 13:02:23 GMT):
its great stuff, really impressed and happy to learn about it

mzk-vct (Sun, 01 Oct 2017 13:04:31 GMT):
:slight_smile: we are glad that you like it!

EarlyR4 (Mon, 02 Oct 2017 17:26:59 GMT):
Has joined the channel.

bdot (Tue, 03 Oct 2017 09:22:52 GMT):
Hi Guys, a newbie question. Need some help in understanding few things. If we want to use customer's fingerprint to authenticate user on Indy. Do we need to manage an intermediate system to maintain a link b/w Sovrin identity & Fingerprint or can we store fingerprint itself to Indy/Sovrin ledger. Thanks for your time.

bendon (Tue, 03 Oct 2017 09:30:51 GMT):
I'll give a newbie answer: I would sue who ever store my fingerprint even on a private server :-p

bendon (Tue, 03 Oct 2017 09:32:00 GMT):
Now, I bet there are ways to not store the fingerprint data itself, but something that prooves I've got this fingerprint under my finger :-p

bendon (Tue, 03 Oct 2017 09:32:54 GMT):
But I'm curious about a better and more accurate answer to your question

bdot (Tue, 03 Oct 2017 10:06:40 GMT):
@bendon Ha ha, Agree with you. I know it's not a good to store finger print.

bdot (Tue, 03 Oct 2017 10:06:45 GMT):
But what if we can have finger print as private key and store public key of it in ledger. Then how can we make sure others can not correlate my identity based on public key. I don't know, hope one of the experts can help.

lovesh (Tue, 03 Oct 2017 11:33:39 GMT):
@bdot You can use the hash of the biometric + nonce to generate the keypair where the nonce can be treated as an application specific id, so for app1, Alice uses `hash(biometric||app1_id)` (as seed) to generate keypair and for app2, Alice uses `hash(biometric||app2_id)` to generate keypair. When authenticating from app1, the app takes Alice's biometric, hashes it with app id and does the signature which can be verified against the key on the ledger. Thus the ledger has these 2 pieces of data for Alice, keys generated using `hash(biometric||app1_id)` and `hash(biometric||app2_id)` but nobody looking at the ledger can guess that biometric in both keys is same. This works nicely in theory but the reality is that biometrics are "noisy" and hash functions do not tolerate noise

lovesh (Tue, 03 Oct 2017 11:33:39 GMT):
@bdot You can use the hash of the biometric + nonce to generate the keypair where the nonce can be treated as an application specific id, so for *app1*, Alice uses `hash(biometric||app1_id)` (as seed) to generate keypair and for *app2*, Alice uses `hash(biometric||app2_id)` to generate keypair. When authenticating from *app1:, the app takes Alice's biometric, hashes it with app id and does the signature which can be verified against the key on the ledger. Thus the ledger has these 2 pieces of data for Alice, keys generated using `hash(biometric||app1_id)` and `hash(biometric||app2_id)` but nobody looking at the ledger can guess that biometric in both keys is same. This works nicely in theory but the reality is that biometrics are "noisy" and hash functions do not tolerate noise

lovesh (Tue, 03 Oct 2017 11:33:39 GMT):
@bdot You can use the hash of the biometric + nonce to generate the keypair where the nonce can be treated as an application specific id, so for *app1*, Alice uses `hash(biometric||app1_id)` (as seed) to generate keypair and for *app2*, Alice uses `hash(biometric||app2_id)` to generate keypair. When authenticating from *app1*, the app takes Alice's biometric, hashes it with app id and does the signature which can be verified against the key on the ledger. Thus the ledger has these 2 pieces of data for Alice, keys generated using `hash(biometric||app1_id)` and `hash(biometric||app2_id)` but nobody looking at the ledger can guess that biometric in both keys is same. This works nicely in theory but the reality is that biometrics are "noisy" and hash functions do not tolerate noise

bdot (Tue, 03 Oct 2017 13:16:36 GMT):
Thanks @lovesh That makes sense. Your point about noisy biometrics is noted.

bendon (Tue, 03 Oct 2017 13:16:39 GMT):
ThX Lovesh

bdot (Tue, 03 Oct 2017 13:17:45 GMT):
Is anyone here experimenting with biometric based sovrin identity?

jamesmonaghan (Tue, 03 Oct 2017 14:20:16 GMT):
yeah there are some folks doing that. the guys at iRespond, for example

jamesmonaghan (Tue, 03 Oct 2017 14:24:07 GMT):
in general i'd say you don't want to derive identifiers or keys directly from a biometric, because you cant change it in future

nage (Tue, 03 Oct 2017 14:34:01 GMT):
Yes, biometrics have some special security concerns, so t is usually best to use them as a secondary factor (like a seed to generate a key or a recovery mechanism for unlocking a recovery key). Indy has he advantage of being able to isolate these biometrics to just your identity wallet and agents, which prevents folks from assembling a database of biometrics to compromise, but it is very important to be careful with this data (even templated biometrics), as you cannot rotate these secrets if they are compromised.

nage (Tue, 03 Oct 2017 14:34:01 GMT):
Yes, biometrics have some special security concerns, so it is usually best to use them as a secondary factor (like a seed to generate a key or a recovery mechanism for unlocking a recovery key). Indy has the advantage of being able to isolate these biometrics to just your identity wallet and agents, which prevents folks from assembling a database of biometrics to compromise, but it is very important to be careful with this data (even templated biometrics), as you cannot rotate these secrets if they are compromised.

bdot (Tue, 03 Oct 2017 15:45:57 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=mTbWvdBJBcBZGi9yy) @Benudek Thanks for the doc, dockerfile saved lot of time!

tabzio (Tue, 03 Oct 2017 15:50:51 GMT):
Has joined the channel.

bdot (Tue, 03 Oct 2017 16:21:24 GMT):
@nage @jamesmonaghan Understand your point. Thanks.

darrell.odonnell (Tue, 03 Oct 2017 18:28:30 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=fjaJGKuZ3ERBEroGL) @jamesmonaghan such as @ewelton

ewelton (Tue, 03 Oct 2017 18:28:30 GMT):
Has joined the channel.

bryanhuang (Wed, 04 Oct 2017 02:40:03 GMT):
Has joined the channel.

cre8bidio (Wed, 04 Oct 2017 14:17:49 GMT):
Has joined the channel.

jamesmonaghan (Wed, 04 Oct 2017 14:48:22 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=dAK4Hm87J9x8xNBJ7) @gudkov this diagram is wonderful!

gudkov (Thu, 05 Oct 2017 13:12:34 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=irzrPafk22qFXpexF) @jamesmonaghan Thanks for Alex )

bendon (Fri, 06 Oct 2017 13:05:29 GMT):
Do similar diagrams exist for other scenarios? I'm looking for the sequence of "Alice gets a transcript" for instance?

Sean_Bohan (Fri, 06 Oct 2017 13:08:04 GMT):
@bendon - we have some - let me look around

bendon (Fri, 06 Oct 2017 13:10:38 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=8HmvofjxpbXDeX5Zf) @Sean_Bohan That would be awesome. I need to present the whole Sovrin/Indy thing (Starting from the demo) to my colleagues and the corresponding sequence diagram(s) will be helpful to dive in.

Sean_Bohan (Fri, 06 Oct 2017 14:30:06 GMT):
still looking (and I will probably have to clean one up for ya) any chance we could get you to share what you did on the Indy Ledger WG call next week?

Sean_Bohan (Fri, 06 Oct 2017 14:30:09 GMT):
Thursday

bendon (Fri, 06 Oct 2017 14:43:21 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=WhxZfNiocEuccMzeo) @Sean_Bohan If your question is for me: can you be more specific?

Sean_Bohan (Fri, 06 Oct 2017 15:06:49 GMT):
hey #indy channel - Hyperledger is a sponsor of Internet Identity Workshop this year and there is a discount code https://www.eventbrite.com/e/internet-identity-workshop-xxv-25-2017b-tickets-30179727377?discount=INDY_XXV_20

swcurran (Fri, 06 Oct 2017 16:15:05 GMT):
A question about the plantuml use cases (sequence diagrams) that were posted recently. Our assumption is that a common use case for something like a proof request/response is that it will be an online web interaction - e.g. on the order of seconds with the user typically waiting. In our toy sandbox, we are seeing times from the unit tests that are MUCH longer than that - 60/90 seconds in some cases. Is it safe for us to assume that's an anomaly, and that we need not build in complication for handling an elongated User Experience based on what we are seeing right now? When we get into the "real world", all will be Fast (or at least, fast enough)?

darrell.odonnell (Fri, 06 Oct 2017 17:16:36 GMT):
@nage any thoughts on @swcurran 's comments here?

nage (Fri, 06 Oct 2017 19:11:58 GMT):
There is some computation involved in calculating and issuing claims, but I don't think it should take quite that long. @ashcherbakov what would be a good "time to generate a claim" and "time to generate a proof" guideline? Are we aware of bugs in that code or tickets around optimization? (I know we have been working to move the python implementation over to the SDK in order to use better, more efficient libraries).

nage (Fri, 06 Oct 2017 19:12:30 GMT):
_notes that Alex probably won't see my reply until his Monday morning_

nage (Fri, 06 Oct 2017 19:12:30 GMT):
_notes that Alex probably won't see this reply until his Monday morning_

swcurran (Fri, 06 Oct 2017 19:30:12 GMT):
I'm less concerned about current performance than about expectations for Sovrin in production. Do we think (vs. hope) that HL-Indy/Sovrin will have "online response" times or should we be designing user interactions to assume delays?

non 22 (Mon, 09 Oct 2017 02:18:14 GMT):
Has joined the channel.

dckc (Mon, 09 Oct 2017 03:39:21 GMT):
Has joined the channel.

dckc (Mon, 09 Oct 2017 03:40:07 GMT):
and what about "time to finality"? What's the consensus algorithm?

ashcherbakov (Mon, 09 Oct 2017 07:46:50 GMT):
@nage @swcurran @darrell.odonnell The only thing that may take quite long time is generation of keys (claim defs) by the issuer. That's because of we use some special primes, and calculation of these primes may take time. I think it's ok for this generation, because it's on the Issuer side only, and should be done only once before we isssue any claims (no request-response involved). All other things such as claim request / issue claim, or proof request / generate proof / verification of proof should work pretty fast. And they do work pretty fast in all our tests (both libindy and python anoncreds). Do you face the problems with generation of Issuer keys, or with other parts of the workflow?

nage (Mon, 09 Oct 2017 12:43:13 GMT):
@swcurran beyond Alex’s question, it is a good idea to expect agent interactions to take time and be asynchronous. This allows agents to tolerate downtime, making them more resilient in important ways, but also allows the humans behind them to stay involved with permissioning and authorizing responses where it is important to them. I know that isn’t really directly related to your question, but from a philosophy perspective the ideal would enable two agents on bad “remote location” connections to interact without timeouts and intermittent connectivity preventing any progress (it would be slow and painful, but the hope is it would still work)

nage (Mon, 09 Oct 2017 12:44:47 GMT):
It also lowers the minimum uptime requirement so that it is realistic to host an agent in your home, or somewhere remote in cases where trust issues prevent doing otherwise.

darrell.odonnell (Tue, 10 Oct 2017 11:26:29 GMT):
@nage agreed on the asynch approach. We're just starting synchronous to allow one of the systems that we have to run synch for now. As a longer-term approach we totally agree.

Raffael 5 (Tue, 10 Oct 2017 15:06:49 GMT):
Has joined the channel.

mredmundto (Wed, 11 Oct 2017 02:57:03 GMT):
Has joined the channel.

jflack1970 (Wed, 11 Oct 2017 11:46:57 GMT):
Has joined the channel.

FranciscoReyes (Wed, 11 Oct 2017 12:38:28 GMT):
Has joined the channel.

mpiekarska (Wed, 11 Oct 2017 16:15:00 GMT):
@here we just cross signed with DIF, so now both organizations are associate members of each other. Thats a great step of bringing various pieces of the puzzle together.

darrell.odonnell (Wed, 11 Oct 2017 16:15:33 GMT):
@mpiekarska meaning HyperLedger did this?

lehors (Wed, 11 Oct 2017 16:15:49 GMT):
or LF?

Sean_Bohan (Wed, 11 Oct 2017 16:18:44 GMT):
DIF announced yesterday that Hyperledger & DIF were cross signing

Sean_Bohan (Wed, 11 Oct 2017 16:19:01 GMT):
think it was Hyperledger @lehors not LF (need to check)

mpiekarska (Wed, 11 Oct 2017 16:27:21 GMT):
hyperledger

drummondreed (Thu, 12 Oct 2017 02:55:13 GMT):
Huge! Congratulations!

Paratus (Thu, 12 Oct 2017 05:43:27 GMT):
Has joined the channel.

qiangminqian (Thu, 12 Oct 2017 08:15:01 GMT):
Has joined the channel.

Sean_Bohan (Thu, 12 Oct 2017 14:12:56 GMT):
Hyperledger Indy Ledger WG call is today

Sean_Bohan (Thu, 12 Oct 2017 14:14:18 GMT):
Hyperledger Indy Ledger WG Call TODAY 8amPT, 9amMT, 11amET, 4pm BST. Where: Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/232861185 Or iPhone one-tap (US Toll): +16465588656,232861185# or +14086380968,232861185# Or Telephone: Dial: +1 646 558 8656 (US Toll) or +1 408 638 0968 (US Toll) Meeting ID: 232 861 185 International numbers available: https://zoom.us/zoomconference?m=a0jD_rTMnh0ZYGQDOKPCNrK_0dP7WPfp1 Agenda: 1. Housekeeping (Sean) 2. Status (Slava and Alex) 3. Indy Roadmap (draft) Discussion (all)

peacekeeper (Fri, 13 Oct 2017 07:20:18 GMT):
hello what's currently the recommended way to install the sovrin/indy CLI ?

ashcherbakov (Fri, 13 Oct 2017 08:30:49 GMT):
As of now CLI is part of indy-node (and sovrin). So, you shoule install sovrin: apt-key adv --keyserver keyserver.ubuntu.com --recv-keys 68DB5E88 apt-key adv --keyserver keyserver.ubuntu.com --recv-keys BD33704C echo "deb https://repo.evernym.com/deb xenial stable" >> /etc/apt/sources.list echo "deb https://repo.sovrin.org/deb xenial stable" >> /etc/apt/sources.list apt-get update -y && apt-get install -y sovrin Please note that this CLI doesn't use indy-sdk yet.

bendon (Fri, 13 Oct 2017 13:36:31 GMT):
Got a question about the claim/proof schema: in which format are they exchanged with/from the ledger? I've seen a comment about "json in the style of schema.org, json-ld, etc" in the [code]|(https://github.com/hyperledger/indy-node/blob/f16beadc9e5477718334cfbc5033ece0a0dd9c87/indy_client/client/lib/src/lib.rs#L176) But also some reference to "JSON schemas, RDF ontologies, or XDI dictionaries,.." in Sovrin Tech. Foundations...

bendon (Fri, 13 Oct 2017 13:36:31 GMT):
Got a question about the claim/proof schema: in which format are they exchanged with/from the ledger? I've seen a comment about "json in the style of schema.org, json-ld, etc" in the code: https://github.com/hyperledger/indy-node/blob/f16beadc9e5477718334cfbc5033ece0a0dd9c87/indy_client/client/lib/src/lib.rs#L176 But also some reference to "JSON schemas, RDF ontologies, or XDI dictionaries,.." in Sovrin Tech. Foundations...

mgbailey (Fri, 13 Oct 2017 14:08:50 GMT):
The sovrin installation no longer requires the evernym repo, so the second key can be omitted, and the repo.evernym.com does not need to be added.

redpath (Fri, 13 Oct 2017 15:41:30 GMT):
Has joined the channel.

redpath (Fri, 13 Oct 2017 15:41:39 GMT):
Can anyone explain the following: 1) I have the Indy source installed on two Ubuntu machines. I can run a node like this $ init_indy_keys --name Alpha [--seed 111111111111111111111111111Alpha] [--force] Running Node $ start_indy_node Alpha 9701 9702 So why are their two port parameters, I understand the names are called nodestack and clientstack but have no idea what intrinsically that means? 2) On the other machine I can start Indy client. #indy Indy-CLI (c) 2017 Evernym, Inc. Type 'help' for more information. Running Indy 1.1 indy> status Not connected to Indy network. Please connect first. Usage: connect indy> Thats great. But how does it know of any nodes and connections? Pretty basic questions but the read on the github seems to have part of the conversation missing.

mgbailey (Fri, 13 Oct 2017 16:20:43 GMT):
I am more familiar with the python Indy installation, but the concepts should be the same. The "Alpha" argument is an alias (name) that the node is known as on the ledger. It will be used in logging, etc. 9701 is the (configurable) port that will be bound to for incoming communications from other validators, consensus messages and the like. 9702 is the (configurable) port that agents and client use to connect to the Indy node.

redpath (Fri, 13 Oct 2017 16:22:11 GMT):
so Essentially the node has two ports to speak

jenmcgrath01 (Fri, 13 Oct 2017 16:23:55 GMT):
Has joined the channel.

brycecurtis (Fri, 13 Oct 2017 17:39:27 GMT):
Has joined the channel.

ramnath00 (Mon, 16 Oct 2017 07:10:29 GMT):
Has joined the channel.

hidde-jan (Mon, 16 Oct 2017 09:47:55 GMT):
Has joined the channel.

redpath (Mon, 16 Oct 2017 15:43:53 GMT):
Does anyone have any documentation how to setup this infrastructure, since I can create Indy Nodes and I can run an Indy-CLI. All I am finding is doc that says here is a test example download (Vagrant) all of this and you can do Alice commands and such to speak the story that is documented. Seems quite vague at this point how these running pieces know of each other. Sure I want a Fabric College (node?) and sure I want graduate of that College (Client) to enable a future employer (client) to verify the credentials for the applicant future employee. Anything would be helpful, I find nothing really on Sovrin and Indy GIT areas that is of any help in this area. Not being negative here just trying to figure it out from nothing of documentation other than walk this mile of a demo and then what

the_identity_guy (Tue, 17 Oct 2017 04:14:03 GMT):
Has joined the channel.

the_identity_guy (Tue, 17 Oct 2017 04:24:29 GMT):
Hello Indy friends, Could someone assist me with telling me how a DID points to a DDO, if DID resides on blockchain and DDO resides on user device?

the_identity_guy (Tue, 17 Oct 2017 04:24:29 GMT):
Hello Indy friends, Could someone tell me how a DID points to a DDO, if DID resides on blockchain and DDO resides on user device?

srottem (Tue, 17 Oct 2017 08:33:07 GMT):
DDOs and DIDs are registered on the ledger. A DDO can be retrieved using DID is was created for.

bendon (Tue, 17 Oct 2017 08:36:11 GMT):
@redpath I'm also having hard time to assemble pieces of the puzzle, but I've found the Running_Locally guide in indy-node/docs a good starting point

bendon (Tue, 17 Oct 2017 08:38:13 GMT):
And I'm spending some time to summarize my understanding of the complete Alice demo scenarios in a sequence diagram. I hope people can help me to improve/correct it: http://www.plantuml.com/plantuml/svg/xLPDRziu43rRlsBWKueWnrwApG9fqW01YY10iyKUT1bHOvcmHAea_RM_laDzIwOSRx70ja1lalKqy-RnpPCJVO1bhFwoP9RH0eNJ1kvoAIXQb6YS5B94vU0-bwHSj4ZHOOAMOFbYv0uTnUeBlu8PBG4j-1-R-o3sCwKPDsgXzMr4AeL47wAavdANSIkx71DOFc929bRtEi-fwTPSJu1tek1EdueoryUcLtipGRwkZLovM7r4jQbm_cBAvacbHtWcjzTcinmDjKrOf_5CswGPQI4yD-6aLXQ-EDgZIRlAqLgc12kjdR-p6wcO5XRmqfDxmJILAeFtWySVzQ7_ERgzXReNF2df9ERo7_JjuFOsQXh2pSsVRNFu08hsi4UcvNh0y3aTn1fLHl3mz10hPB8xCXiw3j7TX8pV6-a8NhJEuVNlv_Ee1rsWL6_ZxhLQoMmEOnqh15zvLZu6RIeHAhUmc_HEMXxVgyCAP4PlItXaCnEATGsA5erXVDdsSYHB3uZZWJ3M7noB0fKch0XSqS6HKPY3bqnH_YvQp70klKti6fwFnOX02CDzIYrv77JEo6JhoCALB3-KshXtRxujSfI51IlML21SlHfKLXXP-bUvybGJl_BZYQkzR-SLckUg30o6lct9kk3Klh0_XZeW0bwiYth5QyH8SCwRAXdQV6amVi0-IJmZt16my7Q0QDaE3oPAQLhoj7ypsLppFsv6wWfqOdspBOUb-pq_fSCUTLiR85Afr8lgTEV1e4ee19MNaiortl3PTlwvhcswChhm-WneTlLvj-Ng62ihCyMC3SsAxyzhKgHR7lysLSjnhRJYhqRFjmdduCxLpLjZHMyOswzEi4o1PSriy72i7wMZamXmwl0xb3acEJLxTDQBiUhN8-01dWahwfFsZH5E3xlas3gp3F-HtgF33aZG5AyBXsJxTovgrIWr3pUp3rxgsgPdby6kzVvKdy086Ras01KMTCty_5SXwcqrW-uJDp21BsKlEydMqXUJhJMlF7T-kVkD7e5_RAUhG9tMQSSL_YDiYYh2BviwV3xzDAFUXZ2TxRt3V4kNnMVxh-xNoy-E-MYX6-jSrz9qnBi_hzrROVBLQ_LNi4EEUqrITuYQYzsPIctm3X-VJTxJ8Vwdx1s7pdaB1LqT7E2dZPrvjhzpvwBS-Gu0

bendon (Tue, 17 Oct 2017 08:39:32 GMT):
I'm ready to PR the puml file in indy-node/docs if somebody can help finilizing it ?

srottem (Tue, 17 Oct 2017 08:41:10 GMT):
The DID/DDO story is really the base of everything and a brief summary might be the DID is a unique identifier while the DDO is a record of information about that identifier. DDOs are persisted somewhere central that they can be accessed and are indexed by the DID they were created for. DIDs can be recorded locally and can be exchanged with others - if I give you my DID you can look up my DDO from where it's stored and get information about my DID that allow you to determine where you can send me messages, what public key you can encrypt them with and information to allow you to verify that messages send to you were actually sent by the party that controls that DID . In the Sovrin environment DIDs and DDOs are stored on the Sovrin distributed ledger.

srottem (Tue, 17 Oct 2017 08:41:10 GMT):
The DID/DDO story is really the base of everything and a brief summary might be the DID is a unique identifier while the DDO is a record of information about that identifier. DDOs are persisted somewhere central that they can be accessed and are indexed by the DID they were created for. DIDs can be recorded locally and can be exchanged with others - if I give you my DID you can look up my DDO from where it's stored and get information about my DID that allows you to determine where you can send me messages, what public key you can encrypt them with and information to allow you to verify that messages send to you were actually sent by the party that controls that DID . In the Sovrin environment DIDs and DDOs are stored on the Sovrin distributed ledger.

srottem (Tue, 17 Oct 2017 08:42:03 GMT):
Does that help?

bendon (Tue, 17 Oct 2017 08:45:34 GMT):
Sometime, it helps to think about DID/DDO lookup like resolving DNS name into an IP + a signature that only the owner of the name would have...

bendon (Tue, 17 Oct 2017 08:45:34 GMT):
Sometime, it helps to think about DID/DDO lookup like resolving DNS name into an IP + a signature that only the owner of the name would be able to produce...

bendon (Tue, 17 Oct 2017 08:45:57 GMT):
But of course, it a bit more complicated :-p

bendon (Tue, 17 Oct 2017 08:49:45 GMT):
It is also kind of help to see how could safe client/server connections be established via Indy without relying on DNS or SSL CA (only need IPs and pub keys of the node)

bendon (Tue, 17 Oct 2017 08:49:45 GMT):
It also does kind of help to see how could safe client/server connections be established via Indy without relying on DNS or SSL CA (only need IPs and pub keys of the node)

bendon (Tue, 17 Oct 2017 08:56:54 GMT):
@redpath And about your question about how the client knows about the node, I think the answer is the pool and domain files containing those ip and pub keys.

bendon (Tue, 17 Oct 2017 08:56:54 GMT):
@redpath And about your question about how the client knows about the nodes, I think the answer is the pool and domain files containing those ip and pub keys.

srottem (Tue, 17 Oct 2017 09:16:44 GMT):
If you want details (and lots of them) about the DIDs and DDOs see the spec: https://w3c-ccg.github.io/did-spec/

redpath (Tue, 17 Oct 2017 17:10:42 GMT):
Thanks bendon I will take a look

the_identity_guy (Wed, 18 Oct 2017 00:34:22 GMT):
@srottem thank you for the explanation. The reason I asked is I wanted to know where the self-sovereign identity is then stored? If DID is on blockchain, and DDO has public info about the user and thats also on blockchain. Does DDO contain the IP address of the service that provide access to self-sovereign identity attributes?

the_identity_guy (Wed, 18 Oct 2017 00:34:22 GMT):
@srottem thank you for the explanation. The reason I asked is I wanted to know where the self-sovereign identity is going to be accessed? If DID is on blockchain, and DDO has public info about the user and thats also on blockchain. Does DDO contain the IP address of the service that provide access to self-sovereign identity attributes?

srottem (Wed, 18 Oct 2017 07:14:48 GMT):
Yes - the DDO can contain information service endpoints the indicate how to connect to an identity. See here - the 'service' key in the JSON document is describing the endpoints: https://w3c-ccg.github.io/did-spec/#example-dids-and-ddos

srottem (Wed, 18 Oct 2017 07:14:48 GMT):
Yes - the DDO can contain information about service endpoints that indicate how to connect to an identity. See here - the 'service' key in the JSON document is describing the endpoints: https://w3c-ccg.github.io/did-spec/#example-dids-and-ddos

nage (Thu, 19 Oct 2017 00:09:41 GMT):

9AA0EE8B-56C7-4A91-BCA5-24DFAE453A42.jpeg

nage (Thu, 19 Oct 2017 00:10:10 GMT):
From IIW

nage (Thu, 19 Oct 2017 00:10:40 GMT):
(internet identity workshop 25)

bendon (Thu, 19 Oct 2017 10:36:42 GMT):
Does anybody know how/where I can retrieve the to send GET_CLAIM_DEF. For instance, for the Transcript issued by Faber College in Alice demo? I've been hexdumping client, agent and node files without any luck. I see a "ref" field related to the claim_def of Transcript in domain_transactions_sandbox/000003.log, but can not get anything else than "Claim def not found" while using decimal conversion from what I guess are those ref number...

ashcherbakov (Thu, 19 Oct 2017 10:45:42 GMT):
is a sequence number (unique id) of a txn as it exusts in the Ledger. You can get it by examining the corresponding SCHEMA txn in the ledger (there seqNo field there). CLI can also be used to get this information: `send GET_SCHEMA dest= name= version=`

bendon (Thu, 19 Oct 2017 11:24:15 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=xLp5FocZSP6zB8A8e) @ashcherbakov Here is what I get from the client (1.1.33)

bendon (Thu, 19 Oct 2017 11:24:17 GMT):
> send GET_SCHEMA dest=ULtgFQJe6bjiFbs7ke3NJD name=Transcript version=1.2 Getting schema ULtgFQJe6bjiFbs7ke3NJD Found schema {'attr_names': ['student_name', 'ssn', 'degree', 'year', 'status'], 'version': '1.2', 'origin': 'ULtgFQJe6bjiFbs7ke3NJD', 'name': 'Transcript'}

bendon (Thu, 19 Oct 2017 11:24:52 GMT):
I don't see any seq number there!?

bendon (Thu, 19 Oct 2017 11:49:39 GMT):
No clue how I can read this seqNo directly from the ledger: in which file should I look at (if not domain_transactions_sandbox/000003.log)?

bendon (Thu, 19 Oct 2017 12:09:08 GMT):
Found seqNo=18 on client side, in .sovrin/data/clients//Requests/, and it sounds like matching the ref field I fount on the ledger side (ref=0x12). But still not getting any output: Alice@test> send GET_CLAIM_DEF ref=18 signature_type=CL Getting claim def 18 Claim def not found

bendon (Thu, 19 Oct 2017 12:09:08 GMT):
Found seqNo=18 on agent side, in .sovrin/data/clients//Requests/, and it sounds like matching the ref field I found on the ledger side (ref=0x12). But still not getting any output: Alice@test> send GET_CLAIM_DEF ref=18 signature_type=CL Getting claim def 18 Claim def not found

ashcherbakov (Thu, 19 Oct 2017 12:44:18 GMT):
On a Node, you can run `read_ledger --type=domain` (run this script from 'indy' user (or 'sovrin', depending on whether you use a version without rebranding chnages)

ashcherbakov (Thu, 19 Oct 2017 12:44:42 GMT):
I was thinking that GET_SCHEMA returns seqNo. If not, then probably it's a missing featyre that needs to be fixed

ashcherbakov (Thu, 19 Oct 2017 12:44:42 GMT):
I was thinking that GET_SCHEMA returns seqNo. If not, then probably it's a missing feature that needs to be fixed

ashcherbakov (Thu, 19 Oct 2017 12:46:01 GMT):
I will create a ticket for this

bendon (Thu, 19 Oct 2017 12:46:56 GMT):
Thx: I'll try to read_ledger to check I've got the right seqNo. Meanwhile: do you know if GET_CLAIM_DEF works as expected?

ashcherbakov (Thu, 19 Oct 2017 12:48:28 GMT):
Yes, it should work

bendon (Thu, 19 Oct 2017 12:50:51 GMT):
Aside of this, I bet it would also make sense if "show connection " would also display the version of the available claims (only the name shows up in 1.1.33). Because version is required for the client to GET_SCHEMA.

bendon (Thu, 19 Oct 2017 12:51:53 GMT):
Without this, user needs to look directly in the client wallet to get the version

bendon (Thu, 19 Oct 2017 13:04:15 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=bGJ8iWLhJmLC95vAf) @ashcherbakov read_ledger works, and give me the transaction ordered in sequence I guess (starting from 10). But GET_CLAIM_DEF gives nothing using either the seqNo of the claim_def or the related shema. Neither using the reqId of those both!

ashcherbakov (Thu, 19 Oct 2017 13:19:45 GMT):
@bendon ok, I think I see the issue. As of now, GET_CLAIM_DEF command can return CLAIM_DEF issued by the sender (current DID in the CLI) only. This also needs to be fixed. There is ORIGIN field in GET_CLAIM_DEF request, and CLI always uses current sender's DID for it. In fact, ORIGIN must be provided explicitly. I will create a ticket for this as well.

ashcherbakov (Thu, 19 Oct 2017 13:20:21 GMT):
In the meantime, if you can `use DID` of CLAIM_DEF's issuer in CLI, and send GET_CLAIM_DEF, then it should return correct result

bendon (Thu, 19 Oct 2017 13:24:57 GMT):
Good catch: it works when using the right DID

bendon (Thu, 19 Oct 2017 13:25:01 GMT):
ThX

ashcherbakov (Thu, 19 Oct 2017 14:00:26 GMT):
@bendon created https://jira.hyperledger.org/browse/INDY-913 and https://jira.hyperledger.org/browse/INDY-914

bendon (Thu, 19 Oct 2017 15:27:44 GMT):
@ashcherbakov Can/should I create one to print the version of the schema of the available claim(s) for a connection?

ashcherbakov (Thu, 19 Oct 2017 15:28:37 GMT):
yes, sure, you can create it

bendon (Thu, 19 Oct 2017 15:32:39 GMT):
@ashcherbakov created https://jira.hyperledger.org/browse/INDY-916

ashcherbakov (Thu, 19 Oct 2017 15:33:51 GMT):
thanks

bendon (Thu, 19 Oct 2017 15:41:00 GMT):
thanks to you: I now have the answers to some rather painful questions :-p

jexek (Fri, 20 Oct 2017 10:04:12 GMT):
Has joined the channel.

mredmundto85 (Sat, 21 Oct 2017 20:44:13 GMT):
Has joined the channel.

BobFranceschini (Tue, 24 Oct 2017 16:04:28 GMT):
Has joined the channel.

codenamedmitri (Tue, 24 Oct 2017 18:01:33 GMT):
Has joined the channel.

ChristianSmith (Tue, 24 Oct 2017 18:06:45 GMT):
Has joined the channel.

gregorypro (Wed, 25 Oct 2017 04:02:58 GMT):
Has joined the channel.

gregorypro (Wed, 25 Oct 2017 04:03:37 GMT):
How do I find any info on Indy? the site says "Coming soon..." How soon? Is there code to look at? Anything?

bendon (Wed, 25 Oct 2017 06:56:44 GMT):
@gregorypro I guess it's "comming soon" becasue still in incubation. It looks like the recommended landing page for info would be the Hyperledger Wiki: https://wiki.hyperledger.org/projects/indy

bendon (Wed, 25 Oct 2017 06:56:44 GMT):
@gregorypro I guess it's "comming soon" because still in incubation. It looks like the recommended landing page for info would be the Hyperledger Wiki: https://wiki.hyperledger.org/projects/indy

mgbailey (Wed, 25 Oct 2017 15:54:38 GMT):
@gregorypro Since what is now indy originally was developed as Sovrin, take a look at the Sovrin Foundation's web page. Most of what is said about

mgbailey (Wed, 25 Oct 2017 15:54:55 GMT):
Sovrin applies to indy. www.sovrin.org

Sean_Bohan (Wed, 25 Oct 2017 20:49:34 GMT):
Hi @gregorypro - I am working with the Hyperledger team to get that page updated with links. Check out the wikipage @bendon mentioned (thanks Bendon!) the forums on Sovrin are also active

panickervinod (Thu, 26 Oct 2017 06:29:05 GMT):
Has joined the channel.

Sean_Bohan (Thu, 26 Oct 2017 14:39:21 GMT):
Hyperledger Indy LEDGER WG Call TODAY 8amPT, 9amMT, 11amET, 4pm BST. Where: Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/232861185 Or iPhone one-tap (US Toll): +16465588656,232861185# or +14086380968,232861185# Or Telephone: Dial: +1 646 558 8656 (US Toll) or +1 408 638 0968 (US Toll) Meeting ID: 232 861 185 International numbers available: https://zoom.us/zoomconference?m=a0jD_rTMnh0ZYGQDOKPCNrK_0dP7WPfp1 Agenda: 1. Housekeeping (Sean) 2. Status (Alex) 3. Posible Demo (Alex 4. Observations from IIW (Darren, JohnJ, Nathan)

Suedonym (Thu, 26 Oct 2017 17:06:42 GMT):
We are planning to delete the Sovrin Google calendar and just use the Hyperledger calendar for all community meetings and working groups. If you have any reason for this NOT to happen, please let me know before before EOD Friday, [10:51] https://calendar.google.com/calendar/embed?src=bGludXhmb3VuZGF0aW9uLm9yZ19uZjl1NjRnOWs5cnZkOWY4dnA0dnVyMjNiMEBncm91cC5jYWxlbmRhci5nb29nbGUuY29t

nbrempel (Thu, 26 Oct 2017 22:26:33 GMT):
Has joined the channel.

calh (Fri, 27 Oct 2017 17:56:12 GMT):
Has joined the channel.

Captain63Dragon (Fri, 27 Oct 2017 20:34:51 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=ht3g6yzRegqDunxGz) @Suedonym You posted a recording of the previous Thursday indy-node meeting. Do you have another one for the latest meeting?

Suedonym (Fri, 27 Oct 2017 21:56:21 GMT):
Unfortunately I was unable to attend the last meeting, so it did not get recorded. The notes from the meeting should be in the #indy-node channel though.

Suedonym (Fri, 27 Oct 2017 21:56:21 GMT):
@Captain63Dragon Unfortunately I was unable to attend the last meeting, so it did not get recorded. The notes from the meeting should be in the #indy-node channel though.

redpath (Mon, 30 Oct 2017 15:42:35 GMT):
I have the Sovrin Alice demo working nicely but just had a minor source question for what is most recent. I am using code git clone https://github.com/hyperledger/indy-node.git git clone https://github.com/hyperledger/indy-anoncreds.git All work fine and this code has recent dates shown when viewing the list. I noticed there is a https://github.com/evernym/sovrin-environments.git with code shown with dates 1 month and some 5 days ago and a https://github.com/evernym/sovrin.git which has code dates of a year old. So the question is, the code I am using is that correct and appropriate? Seems most up to date.

schristie (Mon, 30 Oct 2017 18:16:51 GMT):
Has joined the channel.

Suedonym (Mon, 30 Oct 2017 19:07:06 GMT):
@redpath Here is the wiki page to all Indy related links, etc. https://wiki.hyperledger.org/projects/indy

redpath (Mon, 30 Oct 2017 19:30:55 GMT):
thanks

apoikola (Mon, 30 Oct 2017 21:35:19 GMT):
I am looking for some explanatory use cases that would show how the identity bootstrapping and transactions work... any links?

vinomaster (Tue, 31 Oct 2017 13:11:56 GMT):
Has joined the channel.

Eelke (Wed, 01 Nov 2017 08:17:23 GMT):
Has joined the channel.

redpath (Wed, 01 Nov 2017 14:41:05 GMT):
Who do I need to contact to create a Repo under Hyperledger project called hyperledger/indy-tutorial-sandbox The purpose of this Repo is to provide A turn-key docker based sandbox that enables quick and easy exploration of Hyperledger Indy concepts. This devops repo can be used to gather hands-on experience of Indy basics using the scenarios outlined in the [Sovrin’s Getting Started Guide](https://github.com/hyperledger/indy-node/blob/stable/getting-started.md). Or better yet how do I join so I can create this Repo.

nbrempel (Wed, 01 Nov 2017 16:04:14 GMT):
Hi @redpath there are a few examples available in this repo: https://github.com/evernym/sovrin-environments. I'm currently working on adding another example that requires less setup and can start the Alice example in one command

nbrempel (Wed, 01 Nov 2017 16:05:12 GMT):
But the existing examples there are currently the closest turn-key Alice example that I know of

redpath (Wed, 01 Nov 2017 17:01:47 GMT):
I am aware of that. Ours is simple run this makefile and its ready, very nice anyway thanks

nbrempel (Wed, 01 Nov 2017 17:56:43 GMT):
that's great, I'm looking forward to checking it out

Dan (Wed, 01 Nov 2017 21:45:44 GMT):
Hey Indy crew, does your consensus model enforce an agreement on transaction ordering or state or both or something else? Thanks in advance.

Subramanyam (Thu, 02 Nov 2017 09:46:57 GMT):
Has joined the channel.

Sean_Bohan (Thu, 02 Nov 2017 14:08:15 GMT):
Folks: Hyperledger Indy Agents WG Call TODAY 8amPT, 9amMT, 11amET, 4pm BST. Where: Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/232861185 Or iPhone one-tap (US Toll): +16465588656,232861185# or +14086380968,232861185# Or Telephone: Dial: +1 646 558 8656 (US Toll) or +1 408 638 0968 (US Toll) Meeting ID: 232 861 185 International numbers available: https://zoom.us/zoomconference?m=a0jD_rTMnh0ZYGQDOKPCNrK_0dP7WPfp1 Today's Agenda: Housekeeping (Sean) Development Status (Alex) Universal Resolver discussion with Markus (@peacekeeper) Indy Roadmap discussion (Nathan, Sean)

ashcherbakov (Thu, 02 Nov 2017 14:59:15 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=oahvYRSbCNFFEtX2i) @Dan The main purpose of agreement is transaction ordering. The Primary rules the ordering, and other nodes come to consensus if they agree (in particular, if they ledgers and states are equal, that is roots of Merkle Trees (ledger) are equal and roots of Patricia Trie (state) are equal)

jrosmith (Thu, 02 Nov 2017 15:01:41 GMT):
Has left the channel.

Sean_Bohan (Thu, 02 Nov 2017 17:20:28 GMT):
DRAFT DRAFT DRAFT Roadmap - https://drive.google.com/open?id=0ByaxIFopqmUVRkhyRWZ1X0RpQ1U Please take a look. We will be discussing on the Indy Node call next Thursday A. Steve Tolman will be sharing the changes to JIRA workflow B. Roadmap Discussion ( see here, already shared on Agent call): https://drive.google.com/open?id=0ByaxIFopqmUVRkhyRWZ1X0RpQ1U ) C. Special Guest talking about what they are working on

JohnHClark3 (Thu, 02 Nov 2017 17:29:48 GMT):
Has joined the channel.

Dan (Thu, 02 Nov 2017 17:47:31 GMT):
@ashcherbakov thanks. So it's basically both then. I gather the leader's patricia root is in the block header? Or if you have a link to the block structure that would probably answer my questions too.

sidhujag (Thu, 02 Nov 2017 22:30:25 GMT):
Has joined the channel.

sidhujag (Thu, 02 Nov 2017 22:31:54 GMT):
hey guys just read how sovrin works paper, "Jane uses a master secret, a special key, to create the disclosure using a zero-knowledge proof algorithm." Is this using zkSNARK using libsnark or somet other zkp algo?

sidhujag (Thu, 02 Nov 2017 22:32:38 GMT):
also int he trust framework paper: "This preserves Anonyms as the default behavior, makes Privacy by Design the default choice, and means correlation, when needed, may be provided intentionally using Zero Knowledge Proofs"

sidhujag (Thu, 02 Nov 2017 22:33:10 GMT):
just wondering about the context of proving attestations about your identity and how they would be done via ZKP, rather than you know decrypt some info, encrypt to somebody using a timelock the old fashion way

ashcherbakov (Fri, 03 Nov 2017 07:42:45 GMT):
@Dan https://github.com/hyperledger/indy-plenum/blob/master/plenum/common/messages/node_messages.py#L129 This is how PRE-PREPARE message looks like (PRE_PREPARE is sent by the Leader and the first message in 3 Phase-Commit together with PREPARE and COMMIT)

Lavanya5896 (Fri, 03 Nov 2017 09:46:12 GMT):
Has joined the channel.

Dan (Fri, 03 Nov 2017 14:41:14 GMT):
cool. thanks @ashcherbakov

drummondreed (Fri, 03 Nov 2017 19:07:49 GMT):
@sidhujag I am not an expert, but the ZKP tech used for exchanging verifiable claims with Indy is based on Camenisch-Lysyanskaya signatures. http://groups.csail.mit.edu/cis/pubs/lysyanskaya/cl02b.pdf

sidhujag (Fri, 03 Nov 2017 19:08:34 GMT):
yes that is what I have read, from IBM. I wonder what that difference between that and using a trusted library like libsnark would be

sidhujag (Fri, 03 Nov 2017 19:08:47 GMT):
very cool that it is integrated though!

drummondreed (Fri, 03 Nov 2017 19:08:55 GMT):
If you want to go deep on the topic, read Jan Camenisch's papers on ABC for Trust. https://abc4trust.eu/

drummondreed (Fri, 03 Nov 2017 19:09:33 GMT):
You ask a very good question and that would require one of the LibIndy architects to answer. I'll try to point them at this thread.

sidhujag (Fri, 03 Nov 2017 20:01:58 GMT):
thanks will check that paper out

notOccupanther (Sat, 04 Nov 2017 18:09:29 GMT):
Has joined the channel.

CabMorris14 (Sun, 05 Nov 2017 21:13:53 GMT):
Has joined the channel.

the_identity_guy (Mon, 06 Nov 2017 17:44:48 GMT):
Hello I have a quick question, when developing a solution with Indy, are privacy policies for a given business use case setup using the schema that the node provides?

drummondreed (Mon, 06 Nov 2017 20:04:14 GMT):
That's a great question. Since the verifiable claims will often be exchanged agent-to-agent, it's the privacy policies of the Verifier that normally reply. The Issuer of the claim usually can't dictate the privacy policies of the Verifier (and the Holder of the claim usually can't either). See the Verifiable Claims Primer, https://github.com/WebOfTrustInfo/rebooting-the-web-of-trust-fall2017/blob/master/draft-documents/verifiable-claims-primer.md.

vinomaster (Mon, 06 Nov 2017 21:10:23 GMT):
I have been reading the Indy Contribution Guide and is does not state the process to get a new repo created. Anyone have any pointers for eager submitters? As @redpath eluded to in a prior post herein we would like to create a new hyperledger/indy-tutorial-sandbox repo that enables quick an easy deployment of a docker based indy environment. https://docs.google.com/presentation/d/1zbb1-4TjH1-iYN7LnFZyPWvB1w-YcoUwKYCfByugSE0/edit#slide=id.g23d83a3667_0_5

vinomaster (Mon, 06 Nov 2017 21:10:23 GMT):
I have been reading the Indy Contribution Guide and it does not state the process to get a new repo created. Anyone have any pointers for eager submitters? As @redpath eluded to in a prior post herein we would like to create a new hyperledger/indy-tutorial-sandbox repo that enables quick an easy deployment of a docker based indy environment. https://docs.google.com/presentation/d/1zbb1-4TjH1-iYN7LnFZyPWvB1w-YcoUwKYCfByugSE0/edit#slide=id.g23d83a3667_0_5

drummondreed (Tue, 07 Nov 2017 01:32:55 GMT):
@nage @MALodder @daniel.wu ^^^

MALodder (Tue, 07 Nov 2017 01:32:56 GMT):
Has joined the channel.

daniel.wu (Tue, 07 Nov 2017 01:32:56 GMT):
Has joined the channel.

grkvlt (Tue, 07 Nov 2017 13:38:24 GMT):
Has joined the channel.

nage (Tue, 07 Nov 2017 16:46:54 GMT):
The way Hyperledger process works (at least from what we've observed so far) it is a lot easier to add to an existing repo than to have a new one created. In order to help the release branches of the code in sync it may be better to add it to an existing repo (that way each branch can track changes to keep things compatible), but that isn't always possible. Is there a place in the current repo that would make sense? We are also looking at consolidating the sovrin-environments code into the indy-node and indy-sdk code bases, so perhaps these sorts of examples and tutorials could be placed in-line as well? I'm looking to the fabric repository as an example of this, but we're open to other suggestions. @tkuhrt do you have any recommendations on what works best?

aaronr (Tue, 07 Nov 2017 18:25:48 GMT):
Has joined the channel.

redpath (Tue, 07 Nov 2017 18:27:41 GMT):
Well maybe give it a try as a repo it is very small though quite elegant, simply run a makefile and your ready to go with a Indy demo that works for you without all the setup processes.

nbrempel (Tue, 07 Nov 2017 18:46:11 GMT):
@redpath do you have the tutorial repo hosted anywhere currently? I'd love to take a look at it for my own learning.

aaronr (Tue, 07 Nov 2017 19:01:30 GMT):
Reading the Alice scenario in the Getting Started guide I ran up against a question that I haven't been able to find an answer too. In the guide it says, "Trust anchors provide a way for DIDs to be added to to the ledger. They are generally organizations but can be persons as well. Faber College is a trust anchor, and if its connection request is accepted, will write Alice’s DID to the ledger." So from what I've been reading, only Validator Nodes can write to the ledger and they are run by the stewards on the network. If this wording means that Fabor is a Trust Anchor and Trust Anchor = Steward, then enough said. But if not, I'm curious to know the flow from Faber knowing that they have a DID to write to the ledger and it actually getting written to the ledger.

aaronr (Tue, 07 Nov 2017 19:04:44 GMT):
For example, when a Trust Anchor is brought onboard (I assume by a validator) it could be that the anchor will get access to APIs with which to make requests of the validator node. Is that what happens? And if so, does the Trust Anchor have an established relationship with just that one Validator Node? Or are requests to Validator Nodes load balanced in some way so that the request could potentially be handled by any of the Validator Nodes?

mgbailey (Tue, 07 Nov 2017 21:42:48 GMT):
@aaronr - You are right that only validator nodes can write to the ledger, which they themselves maintain. This is the nature of distributed ledgers. Such writes are done only when consensus is achieved among validator nodes, with 2/3 * N +1 of the validators agreeing to the transaction. What triggers the transaction? A request to the validator pool, via a client or an agent. All such requests must be signed by the requesting party. The signature is verified by the validators, and if the signature matches the DID and verification key of someone who is already on the ledger, who has a role that is sufficient to authorize that the transaction be placed onto the ledger, then the validators will perform the write. A trust anchor is a role that has sufficient privileges to sign a transaction to place a customer's DID onto the ledger.

redpath (Tue, 07 Nov 2017 22:27:55 GMT):
@nbrempel Let me see into that

aaronr (Tue, 07 Nov 2017 22:49:40 GMT):
@mgbailey - thank you for the great explanation. The trust anchor part confuses me a little, however. So any agent or client that has a sufficient role to successfully request writing to the ledger is a Trust Anchor? Or is Trust Anchor a role above that? So writing a DID to the ledger takes an even more specialized role? I assume the Stewards determine who can be a Trusted Authority. Is that true of all roles that a client or agent has? On a similar note, are read requests handled in a similar manner from the observer nodes? It is a pool of nodes that requests are made of by qualified agents or clients?

aaronr (Tue, 07 Nov 2017 22:51:10 GMT):
c /Trusted Authority/Trust Anchor

mgbailey (Tue, 07 Nov 2017 22:57:12 GMT):
All roles can read without restriction. To understand roles better, take a look at the "Sovrin Roles" appendix of [this document](https://docs.google.com/document/d/1kQcdICAYmSqbk4d5lUtFhk2L55dKJMfyM0JbZPeG55s/edit?usp=sharing)

nage (Wed, 08 Nov 2017 01:22:31 GMT):
@vinomaster is it out in a git repository now? I think there is support for sharing it, and once people have a chance to take a look we can figure out what to do next

smithbk (Wed, 08 Nov 2017 07:17:59 GMT):
Has joined the channel.

anikitinDSR (Wed, 08 Nov 2017 09:04:08 GMT):
Has joined the channel.

marc4gov (Wed, 08 Nov 2017 10:29:46 GMT):
Has joined the channel.

aaronr (Wed, 08 Nov 2017 15:48:10 GMT):
@mgbailey I guess my team is looking for more specific details. In the "real world", if I wanted to become an Trust Anchor of the production Sovrin network and be able to make a write request, how do I get the information that I need to do so? Is all of this (validator pool address, API set) hidden from view in a CLI? If so, is that CLI the standard CLI, or specialized software that a Steward would send to a Trust Anchor out of band as part of an onboarding process? Or is it some other flow other than CLI to make a write request to the ledger?

brycecurtis (Wed, 08 Nov 2017 15:56:31 GMT):
@ngage @mgbailey @nbrempel Here is the repo link to our simple demo setup that @redpath mentioned. Comments are welcome. https://github.com/brycecurtis/indy-tutorial-sandbox

brycecurtis (Wed, 08 Nov 2017 15:56:31 GMT):
@nage @mgbailey @nbrempel Here is the repo link to our simple demo setup that @redpath mentioned. Comments are welcome. https://github.com/brycecurtis/indy-tutorial-sandbox

mgbailey (Wed, 08 Nov 2017 16:26:28 GMT):
@aaronr Say "Acme GIzmos" wants to be a trust anchor and to establish its agent to be able to function in the Sovrin ecosystem. Acme would need to reach out to a steward, who would vet Acme. After agreeing that Acme should be a trust anchor, the steward would request a DID and verkey from Acme. Acme would use a tool such as the Sovrin CLI to generate the DID and verkey, along with the corresponding private key. It would store the private key in a secure location, and send the DID and verkey to the steward. The steward would write Acme's DID, verkey and the role (trust anchor) into a transaction on the ledger.

mgbailey (Wed, 08 Nov 2017 16:27:25 GMT):
@brycecurtis , thanks for sharing this. I will give it a spin.

aaronr (Wed, 08 Nov 2017 17:04:10 GMT):
@mgbailey and then Acme would use the Sovrin CLI to "write to the ledger" (which would make the request to the pool of validator nodes that would then be voted on, authorized, and then written by a validator node). The "write to the ledger" would be a command, or series of commands, like "add keys...." and "submit pending". And the address of the pool and the actual api to request the writing is hidden from the CLI user. Do I have that right?

nbrempel (Wed, 08 Nov 2017 17:12:56 GMT):
thank you very much @brycecurtis @redpath, I'm sure this will be very helpful!

mgbailey (Wed, 08 Nov 2017 17:47:15 GMT):
@aaronr most likely Acme will set up an agent application to automate interactions with the ledger, rather than doing most actions with the CLI. The CLI is handy for demonstration purposes, though. You are correct that Acme would then be able to submit transactions to the ledger such as adding a NYM (i.e. DID and verkey) for a customer. The validators will receive the transaction, see that it is signed by a valid trust anchor, vote on it, and put it on the ledger.

BruceMacArthur (Wed, 08 Nov 2017 23:08:46 GMT):
Has joined the channel.

notOccupanther (Thu, 09 Nov 2017 04:52:41 GMT):
Hi All. Newb here. I'm getting through the EdX course on Hyperledger and am in my final year of college and am looking for my final year to create some blockchain-based college accreditation platform. Something akin to BlockCert and what Malta are doing, but smaller scale and using HyperLedger Indy as the platform.... so would anyone have tips on a decent source I could get stuck into? Many thanks

nage (Thu, 09 Nov 2017 08:23:02 GMT):
@notOccupanther this is certainly an interesting use case. @mhailstone at BYU and a few others have been looking at university identity use cases for Verifiable Claims. The BYU-OIT folks have a github repository with some interesting things to take a look at here https://github.com/BYU-OIT. There are also some interesting things to compare on BlockCerts schemas that might be adapted into the Getting Started Guide's Alice example to produce a proof of concept.

mhailstone (Thu, 09 Nov 2017 08:23:02 GMT):
Has joined the channel.

AndrewTobin (Thu, 09 Nov 2017 10:34:46 GMT):
Has joined the channel.

AndrewTobin (Thu, 09 Nov 2017 10:38:46 GMT):
@notOccupanther we're getting increasing interest from universities for things like qualifications, which would simply be verifiable claims. We can be MUCH more creative with student identity though, as universities could bootstrap students with a host of verifiable claims to help them in their future life. Student identity itself, faculty membership, course credits, exam results, library access, system access etc. All can be verifiable claims. So it's much more than the relatively simple "degree certificates on a blockchain" that blockcerts does. You could use Sovrin to create a portable student passport usable anywhere.

grkvlt (Thu, 09 Nov 2017 13:50:20 GMT):
hi. i have a use case where i want to track blockchain users that have gone through a KYC or anti-money laundering process - only once they pass that are they trusted to perform transactions. is that the sort of thing Indy could help with? I'm using Sawtooth as my blockchain infrastructure, so it needs to be compatible with that...

grkvlt (Thu, 09 Nov 2017 13:50:59 GMT):
(KYC = Know Your Customer, in depth authentication used by banks and financial institutions, usually involves checking documents are valid etc.)

jamesmonaghan (Thu, 09 Nov 2017 13:52:04 GMT):
yes, very much so

jamesmonaghan (Thu, 09 Nov 2017 13:53:31 GMT):
the Sovrin ledger gives you a place to register decentralised identifiers (DIDs) for the users and the issuers of credentials, an the IndySDK gives you what you need to issue and validate Verifiable Claims to those users, digitally signed by the issuing party

jamesmonaghan (Thu, 09 Nov 2017 13:55:15 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=KrwF5GfC9dhv9xxSq) @drummondreed you _could_ do this with an information sharing agreement as part of the claims exchange, which the downstream parties had to adhere to

VipinB (Thu, 09 Nov 2017 14:29:20 GMT):
Meeting today?

VipinB (Thu, 09 Nov 2017 14:29:48 GMT):
I see it on node

Sean_Bohan (Thu, 09 Nov 2017 21:06:47 GMT):
VIpin I missed this, YES we had the WG call today. I will make sure to post the agenda and call in information here as well in the future

ianco (Thu, 09 Nov 2017 21:44:48 GMT):
Has joined the channel.

Sean_Bohan (Thu, 09 Nov 2017 23:33:49 GMT):
hey @scottmackenzie - spoke with one of our DevOps folks and he suggested you try stuff from the `stable` branch instead of `master`

scottmackenzie (Thu, 09 Nov 2017 23:33:49 GMT):
Has joined the channel.

Sean_Bohan (Thu, 09 Nov 2017 23:34:10 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=cg2okJDSsBpEmEGik) Hi guys, not sure if this is the right chat for this question, but I’m following the indy install process, as documented here (https://github.com/evernym/sovrin-environments/blob/master/vagrant/training/vb-multi-vm/TestIndyClusterSetup.md). When I come to “Setting Up a CLI Client and Configuring the Agents in the Indy Cluster” and use vagrant to ssh into the cli01 host, and run “indy”, the “indy” binary is not found. Infact, the “indy” binary itself doesn’t seem to be installed on cli01 or any of the validator nodes. Other indy related artifacts are though. Perhaps I’m missing something or the guide is out of date (although it was updated 11 days ago). Can anyone point me in the right direction?

Sean_Bohan (Thu, 09 Nov 2017 23:34:57 GMT):
https://github.com/evernym/sovrin-environments/blob/stable/vagrant/training/vb-multi-vm/TestSovrinClusterSetup.md

notOccupanther (Fri, 10 Nov 2017 17:05:37 GMT):
@AndrewTobin I agree with your points there. I just intended this project to do two things. 1) Allow me to submit an interesting Thesis as my final year project (none of the contet was on the course, apart from some Python and Java, but the lecurers are ok with me working on it), and 2) get me started with a project that is Blockchain based, to learn as I go kind of thinng. So in reality, if I ciould even just get a single transaction working, from DIT to Student and from student to claim his/her degree... that would be cool. My project would still involve fleshing out the Software Engineering angle with Workflows, State Diags, legal considerations and implications etc ...

notOccupanther (Fri, 10 Nov 2017 17:05:37 GMT):
@AndrewTobin I agree with your points there. I just intended this project to do two things. 1) Allow me to submit an interesting Thesis as my final year project (none of the contet was on the course, apart from some Python and Java, but the lecurers are ok with me working on it), and 2) get me started with a project that is Blockchain based, to learn as I go kind of thing. So in reality, if I could even just get a single transaction working, from DIT to Student and from student to claim his/her degree... that would be cool. My project would still involve fleshing out the Software Engineering angle with Workflows, State Diags, Case Diags, the Legal angle & Considerations, Project Management and some Visualization Analysis ... and Presentation / Demo thereof ... so not looking for anything enterprise-scale just yet :~)

notOccupanther (Fri, 10 Nov 2017 17:05:37 GMT):
@AndrewTobin I agree with your points there. I just intended this project to do two things. 1) Allow me to submit an interesting Thesis as my final year project (none of the contet was on the course, apart from some Python and Java, but the lecurers are ok with me working on it), and 2) get me started with a project that is Blockchain based, to learn as I go kind of thing. So in reality, if I could even just get a single transaction working, from DIT Uni to Student and from DIT Student to claim his/her degree... that would be cool. My project would still involve fleshing out the Software Engineering angle with Workflows, State Diags, Case Diags, the Legal angle & Considerations, Project Management and some Visualization Analysis ... and Presentation / Demo thereof ... so not looking for anything enterprise-scale just yet :~)

notOccupanther (Fri, 10 Nov 2017 17:06:12 GMT):
@nage thanks for the tip. will have a look at BYU .. many thanks!

swlcanderson (Fri, 10 Nov 2017 21:21:15 GMT):
Has joined the channel.

nengine (Sun, 12 Nov 2017 00:34:20 GMT):
Has joined the channel.

nengine (Mon, 13 Nov 2017 01:23:01 GMT):
Hello, I am trying to setup test network for indy-node, and followed to setup dev from https://github.com/hyperledger/indy-node/blob/master/docs/setup-dev.md and https://github.com/hyperledger/indy-node/blob/master/docs/indy-running-locally.md but when I run "generate_indy_pool_transactions --nodes 4 --clients 5 --nodeNum 1" I am getting "The 'indy-plenum-dev==1.2.171' distribution was not found and is required by indy-node-dev" message. Any help is appreciated.

rojanjose (Mon, 13 Nov 2017 04:11:37 GMT):
Has joined the channel.

andkononykhin (Mon, 13 Nov 2017 10:46:20 GMT):
@nengine Hello. Seems the docs are not good for all cases. We will update them asap. For now, I can suggest the following steps to fix that - run `sudo indy-node/scripts/create_dirs.sh` this will create file structure necessary for further pool initialization routine (as it is mentioned in https://github.com/hyperledger/indy-node/blob/master/docs/indy-running-locally.md#requirements using virtual machine is highly recommended) - use scripts from `indy-node/scripts` directory instead of available from the $PATH (i.e. `indy-node/scripts/generate_indy_pool_transactions`, `indy-node/scripts/start_indy_node`)

andkononykhin (Mon, 13 Nov 2017 10:46:20 GMT):
@nengine Hello. Seems the docs are not good for some cases. We will update them asap. For now, I can suggest the following steps to fix that - run `sudo indy-node/scripts/create_dirs.sh` this will create file structure necessary for further pool initialization routine (as it is mentioned in https://github.com/hyperledger/indy-node/blob/master/docs/indy-running-locally.md#requirements using virtual machine is highly recommended) - use scripts from `indy-node/scripts` directory instead of available from the $PATH (i.e. `indy-node/scripts/generate_indy_pool_transactions`, `indy-node/scripts/start_indy_node`)

dselman (Mon, 13 Nov 2017 13:07:25 GMT):
Has joined the channel.

sydi (Mon, 13 Nov 2017 13:07:32 GMT):
Has joined the channel.

andkononykhin (Mon, 13 Nov 2017 14:35:40 GMT):
@nengine Actually https://github.com/hyperledger/indy-node/blob/master/docs/indy-running-locally.md outdated a bit and as alternative you could try vargrant. If you want to develop indy-node you need just https://github.com/hyperledger/indy-node/blob/master/docs/setup-dev.md Or if you want to develop indy-sdk please, take a look

andkononykhin (Mon, 13 Nov 2017 14:35:40 GMT):
@nengine Actually https://github.com/hyperledger/indy-node/blob/master/docs/indy-running-locally.md outdated a bit and as alternative you could try vargrant. If you want to develop indy-node you need just https://github.com/hyperledger/indy-node/blob/master/docs/setup-dev.md Or if you want to develop indy-sdk please, take a look at indy-sdk docs: https://github.com/hyperledger/indy-sdk/tree/master/doc

nengine (Mon, 13 Nov 2017 15:20:58 GMT):
Thank you @andkononykhin . I am trying to setup a test indy-node Network with 4 nodes on Ubuntu 16.04 so that I follow a "Alice Gets a Transcript" tutorial from here https://github.comhyperledger/indy-node/blob/stable/getting-started.md. I believe it also depends on indy-plenum. Could you guide how to do it now? Or I should wait until documentation is complete?

andkononykhin (Mon, 13 Nov 2017 15:46:13 GMT):
@nengine You can try local pool setup with workarounds that I suggested. Hopefully, it will work. As an option you could try vagrant way mentioned in getting started tutorial. BTW consider to check docs and code from master branch https://github.com/hyperledger/indy-node/blob/master/getting-started.md. It includes a lot of fixes and more up-to-date docs and links. We will also try to make updates related to the issue you mentioned.

drummondreed (Mon, 13 Nov 2017 15:48:42 GMT):
@andkononykhin Good answer!

mgbailey (Mon, 13 Nov 2017 16:18:38 GMT):
@nengine Are you able to run the getting started? If not, for the stable branch, you can set up a vagrant VM cluster to go through it pretty easy using the instructions at https://github.com/evernym/sovrin-environments/blob/stable/vagrant/training/vb-multi-vm/TestSovrinClusterSetup.md. If you want to work with the master branch (which can be less stable, but is working well right now), @redpath @brycecurtis and @aaronr have contributed a nice and easy to use docker setup at https://github.com/brycecurtis/indy-tutorial-sandbox

vinomaster (Mon, 13 Nov 2017 17:37:56 GMT):
@mgbailey How can we get the indy-tutorial-sandbox contribution formally into indy?

mgbailey (Mon, 13 Nov 2017 17:41:12 GMT):
I don't have the permissions to create a new repo in hyperledger, and I think we need one for artifacts like this. I have begun the discussion with others who can do it.

vinomaster (Mon, 13 Nov 2017 17:59:54 GMT):
@mgbailey Thank You! Keep me informed.

Sean_Bohan (Mon, 13 Nov 2017 18:53:40 GMT):
THIS week we have the Indy Agent WG Call on Thursday November 16, 2017 Agenda: A. Housekeeping (Nathan and Sean) B. Development Achievements (Alex and Slava) C. Report from W3C TPAC (Nathan and Drummond) D. Report from Hyperledger Community Summit (Phil Windley, Sovrin Foundation) E. Dave Huesby from Hyperledger on American Fuzzy Lop and Fuzzing in Hyperledger F. Wrap Up Call details: Hyperledger Indy Agent (SDK) WG Call When: November 16, 2017 8amPT, 9amMT, 11amET, 4pm BST Where: Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/232861185 Or iPhone one-tap (US Toll): +16465588656,232861185# or +14086380968,232861185# Or Telephone: Dial: +1 646 558 8656 (US Toll) or +1 408 638 0968 (US Toll) Meeting ID: 232 861 185 International numbers available: https://zoom.us/zoomconference?m=a0jD_rTMnh0ZYGQDOKPCNrK_0dP7WPfp1 Please note: This meeting will be recored Bring a friend. It will be fun. - Hyperledger Indy Community Hyperledger Community Indy Rocketchat: https://chat.hyperledger.org/channel/indy https://chat.hyperledger.org/channel/indy-agent https://chat.hyperledger.org/channel/indy-node Indy Wiki: https://wiki.hyperledger.org/projects/indy

Suedonym (Mon, 13 Nov 2017 18:55:45 GMT):
Bring two friends! The more the merrier!!

nengine (Mon, 13 Nov 2017 20:08:47 GMT):
hi @mgbailey I was at the instructions, but found that it was refering to "Sovrin " network node. If I setup vagrant, is to going to use code hyperledger/indy-node github repository? thank a lot.

mgbailey (Mon, 13 Nov 2017 20:16:12 GMT):
@nengine If you are on the stable branch, it will say "sovrin". If you are on the master branch, "sovrin" has been re-branded as "indy". This confusion will die out soon, when we are ready to do the next release, and the master branch is promoted to stable.

nengine (Mon, 13 Nov 2017 20:29:17 GMT):
@mgbailey Ok I will give it a try. I am also interested in compiling hyperledger indy-node and indy-plenum and setup a node from scratch on Ubuntu machine without using vigrant and VM. Is it already possible to do this? Thanks.

theruss (Mon, 13 Nov 2017 20:29:18 GMT):
Has joined the channel.

mgbailey (Mon, 13 Nov 2017 20:31:54 GMT):
@nengine This is not the way I usually do it, but the devs do it this way all the time. I think there are instructions in the readme's.

theruss (Mon, 13 Nov 2017 21:42:11 GMT):
I've been reading around Indy for a day or two and roughly speaking I have wrapped my head around some aspects of it, but one question I need answering is this: Is Indy a DLT implementation itself, or rather does it require the services of a DLT as a "backend" for example Fabric?

Sean_Bohan (Mon, 13 Nov 2017 22:28:49 GMT):
hi @theruss Indy Node is the codebase for a DLT, the first instantiation of it is the Sovrin DLT (currently provisional network).

theruss (Mon, 13 Nov 2017 22:30:19 GMT):
Hi @Sean_Bohan and thanks.

theruss (Mon, 13 Nov 2017 22:31:33 GMT):
So if I, or my company, had a project that fit the bill of decentralised identity nearly perfectly, then am I right in thinking that an Indy Node setup, properly specced (in terms of project-specific business logic in chain code and the rest of it) would be another such instantiation?

theruss (Mon, 13 Nov 2017 22:31:33 GMT):
So if I, or my company, had a project that fit the bill of decentralised identity nearly perfectly, then am I right in thinking that an Indy Node setup, properly specced (in terms of project-specific business logic in chain code and the rest of it) would be another such "instantiation"?

drummondreed (Mon, 13 Nov 2017 22:54:41 GMT):
Yes, you could always instantiate a new ledger using Indy code. Or your needs might just be served by an existing ledger that is able to provide the decentralized identity functionality you need, since if you want to preserve privacy, the actual on-ledger component of decentralized identity (the DIDs and DID docs and claims schemas) is pretty thin.

nengine (Mon, 13 Nov 2017 23:31:55 GMT):
@andkononykhin I did vagrant setup and I was able to ssh into cli01 and when I run "vagrant@cli01:~$ indy" it says "-bash: indy: command not found".

theruss (Mon, 13 Nov 2017 23:45:54 GMT):
@drummondreed thanks. My trouble in almost (but not quite) "getting it" for Indy, is that I've been doing the EDX course and so far it would seem that for Fabric, Sawtooth and Iroha, for a client project, one would "simply" (tm) spin-up a network of nodes within a cloud where say for argument's sake, each node were physically installed on a single VM and where the network comprise N such VMs.

theruss (Mon, 13 Nov 2017 23:45:54 GMT):
@drummondreed thanks. My trouble in almost (but not quite) "getting it" for Indy, is that I've been doing the EDX course and so far it would seem that for Fabric, Sawtooth and Iroha, for a client project, one would "simply" (tm) spin-up a network of nodes within a cloud where say for argument's sake, each node were physically installed on a single VM and where the network comprised N such VMs.

theruss (Mon, 13 Nov 2017 23:47:06 GMT):
It might be worth noting that my company has its own cloud running OpenStack and it would be there, our POC project would be hosted and developed

drummondreed (Mon, 13 Nov 2017 23:48:20 GMT):
@theruss You could definitely do the same with Indy. @mgbailey is the guy to advise you more about that.

mgbailey (Mon, 13 Nov 2017 23:51:41 GMT):
@nengine, what was 'sovrin' is in the process of being rebranded to 'indy'. If you are working in the stable branch, make sure that that is the branch you are using everywhere, including repo checkouts and instructions. Then your terminology and commands will be consistent.

mgbailey (Mon, 13 Nov 2017 23:52:31 GMT):
Alternatively, for the newest (and perhaps less stable) stuff, use the master branch for everything

mgbailey (Mon, 13 Nov 2017 23:55:15 GMT):
Actually, I just found an error in a script that will make the master branch stuff fail...

mgbailey (Mon, 13 Nov 2017 23:59:23 GMT):
I am testing a fix to the vagrant scripts for the master branch, and will push it to the repo upon success

nengine (Tue, 14 Nov 2017 00:30:32 GMT):
ok, yes i was on stable and now switched to master and doing "vagrant up" again. but will wait for your news changes. thanks a lot.

dv29 (Tue, 14 Nov 2017 08:56:25 GMT):
Has joined the channel.

NiallC (Tue, 14 Nov 2017 19:11:59 GMT):
Has joined the channel.

theruss (Tue, 14 Nov 2017 21:27:27 GMT):
I had to dig quite deep to understand that Indy itself only deals with the immutibility of transactions, DID references to DDOs and the management thereof. However; my remaining issue is with the actual storage of PPI (personal) data.

theruss (Tue, 14 Nov 2017 21:28:30 GMT):
In that for now obvious reasons, storing PPI data on a ledger, any ledger is bad idea. So the problem of "security" and immutibility of PPI data goes off-chain and becomes a problem of re-centralisation does it not?

theruss (Tue, 14 Nov 2017 21:28:30 GMT):
In that for now obvious reasons, storing PPI data on a ledger, any ledger, is bad idea. So the problem of "security" and immutibility of PPI data goes off-chain and becomes a problem of re-centralisation does it not?

theruss (Tue, 14 Nov 2017 21:30:03 GMT):
Given that PPI data would probably need to be stored in a document DB or RDBMS somewhere, which is subject to all the problems that decentralisation promises to move us away from, how are the two states-of-affaird reconciled?

theruss (Tue, 14 Nov 2017 21:30:03 GMT):
Given that PPI data would probably need to be stored in a document DB or RDBMS somewhere, which is subject to all the problems that decentralisation promises to move us away from, how are the two states-of-affairs usually reconciled?

theruss (Tue, 14 Nov 2017 21:31:00 GMT):
Off the top of my head, the DID and a public key could be stored against records, so what's the "reccomend" way to achieve this? How does Sovrin's implementation solve this?

theruss (Tue, 14 Nov 2017 21:31:00 GMT):
Off the top of my head, the DID and a public key could be stored against a DB record, but what's the "reccomended" way to achieve this? How does Sovrin's implementation solve this?

theruss (Tue, 14 Nov 2017 21:42:50 GMT):
I should add that the project I have in mind, will allow for frequent updating of PPI data such as name-changes, change of circumstamces (marriage, divorce etc). The DDO may not change too much, but the PPI data itself, howsoever it is sent over-the-wire over a dervice endpoint like a local hostname or IP, will be written-to and read from frequently for the N "Entity's" as network participants.

theruss (Tue, 14 Nov 2017 21:42:50 GMT):
I should add that the project I have in mind, will allow for frequent updating of PPI data such as name-changes, change of circumstamces (marriage, divorce etc). The DDO may not change too much, but the PPI data itself, howsoever it is sent over-the-wire over a dervice endpoint like a local hostname or IP, will be written-to and read from frequently for the N "Entity's" that comprise the network's participants.

jljordan_bcgov (Tue, 14 Nov 2017 23:02:33 GMT):
@theruss PII is stored in the edge ... in an end user's "wallet" where they control the private keys to access the PII

theruss (Tue, 14 Nov 2017 23:03:17 GMT):
Of course it is!

theruss (Tue, 14 Nov 2017 23:03:25 GMT):
Why did I not think of that?

theruss (Tue, 14 Nov 2017 23:03:44 GMT):
Thank you @jljordan_bcgov

jljordan_bcgov (Tue, 14 Nov 2017 23:04:17 GMT):
because this is quite a different way of thinking about data, where its stored, and how its handled ... which I struggle to keep in mind

theruss (Tue, 14 Nov 2017 23:04:40 GMT):
Right, so Trust Anchors and other web-of-trust network participants, "login to users" in effect!?

theruss (Tue, 14 Nov 2017 23:04:50 GMT):
I struggle also :-)

theruss (Tue, 14 Nov 2017 23:05:55 GMT):
So I imagine that user's will get some sort of alert within the wallet app (for example) that agency X requires to know A, B, and C about you, but not D thru Z

theruss (Tue, 14 Nov 2017 23:06:07 GMT):
...implementation specific of course

jljordan_bcgov (Tue, 14 Nov 2017 23:06:34 GMT):
yes that is the general idea ... as I understand it

theruss (Tue, 14 Nov 2017 23:06:58 GMT):
That's perfect, and is almost the missing pice of the puzzle for me (atm!)

jljordan_bcgov (Tue, 14 Nov 2017 23:06:59 GMT):
the SSI network enables secure peer to peer interactions

jljordan_bcgov (Tue, 14 Nov 2017 23:08:12 GMT):
peers are now any entity in the ecosystem ... end users (humans), services, things .. anything that can be represented by a DID, DID Doc ... whether or not the entity is able to act on its own behalf is another question :)

jljordan_bcgov (Tue, 14 Nov 2017 23:09:42 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=odLyW7Ry6pfB66sw8) @theruss great and we understand the (atm!) :)

theruss (Tue, 14 Nov 2017 23:10:27 GMT):
Awesome and thanks for taking the time to respond in detail.

jljordan_bcgov (Tue, 14 Nov 2017 23:25:28 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=wjhmSpxMCeXMJ9h7c) @theruss NP and have a great day

mountbranch (Wed, 15 Nov 2017 17:14:57 GMT):
Has joined the channel.

josepsan (Wed, 15 Nov 2017 17:31:36 GMT):
Has joined the channel.

jljordan_bcgov (Wed, 15 Nov 2017 18:19:15 GMT):
Community Question: In the What Goes on the Ledger document the following is stated "The consent receipt contains the two DIDs that shared the information, and the attribute names and data types that were shared (rather than the actual data itself), and is signed by both parties to provide non-repudiation. The proof which is stored on Sovrin is simply a cryptographic hash of this receipt." It there still the concept of a consent receipt (or link contract) in sovrin, and if so, is the proof stored on a ledger in the sovrin network?

drummondreed (Wed, 15 Nov 2017 18:22:51 GMT):
John, it's a great question. The short answer is "Yes". The longer answer is the classic "the devil is in the details". In other words, several formats for the consent receipt have been proposed, and what needs to be standardized is that format so it's clear what the hash on the ledger will prove.

jljordan_bcgov (Wed, 15 Nov 2017 18:56:15 GMT):
So is there code in Indy SDK that will support this function yet?

drummondreed (Wed, 15 Nov 2017 19:52:39 GMT):
I don't believe so, but @nage would know better.

drummondreed (Wed, 15 Nov 2017 19:53:13 GMT):
If not, I'd be willing to suggest a plan for how to arrive at a canonical format for the consent receipts.

jljordan_bcgov (Wed, 15 Nov 2017 19:54:10 GMT):
it is not an immediate requirement but I can see it would be useful in regulated use cases where one would want proof that credentials were provided and verified to a site inspector for instance

nage (Wed, 15 Nov 2017 21:55:58 GMT):
@jljordan_bcgov no, we don't have a specific consent receipt type of format (yet), the thought was to treat it like any other verifiable claim (now being called a verifiable credential by the W3C working group)

nage (Wed, 15 Nov 2017 21:56:18 GMT):
and when a standard emerges in practice we could promote that to standardize it across the ecosystem

theruss (Thu, 16 Nov 2017 03:36:39 GMT):
I see that Canada is now using Fabric via SecureKey & IBM for their identity solution. Can anyone answer the question, in what fundamental ways do Fabric and Indy differ such that Identity is do-able using both frameworks?

theruss (Thu, 16 Nov 2017 03:37:55 GMT):
I realise that Indy is not yet stable, and that would have factored into their decision-making, but myself and my team are weighing up which backend to use for our upcoming project, so any pointers are welcomed.

nengine (Thu, 16 Nov 2017 12:49:07 GMT):
Hi @mgbailey Pls let me know if was able to fix an error in vagrant script? Thanks.

Sean_Bohan (Thu, 16 Nov 2017 14:44:53 GMT):
Indy quarterly status can be found here: https://wiki.hyperledger.org/groups/tsc/project-updates/indy-2017-nov

mgbailey (Thu, 16 Nov 2017 14:52:29 GMT):
@nengine , did, but the changes have not been accepted into the repo yet. I will re-ping the approvers

nengine (Thu, 16 Nov 2017 17:01:19 GMT):
Great!

gokhankocak (Thu, 16 Nov 2017 17:14:18 GMT):
Has joined the channel.

jljordan_bcgov (Thu, 16 Nov 2017 17:50:31 GMT):
@theruss Fabric is a distributed ledger with the design intent of the network being a permission closed network and being able to support primarily business network (ie supply chain, etc) type problems. It has no native privacy by design identity capabilities built into it. The "identity" aspects that are there, membership services last time i looked, are about managing the permissions to the distributed ledger and not about providing identity capabilities to end users of the network. Securekey is a Canadian technology company that has added capabilities on top of Fabric to handle KYC type identity capabilities and is offering that as a service to the market. It is not generally available yet to my knowledge. Indy is designed to be operated as a permissioned open network with privacy by design identity primitive built into the core technology. So I would offer that these two Hyperledger Projects have different design intents and its important to understand what your business problem is so that you select the appropriate technology. The two could work together if the business problem required it. This is how I understand the current landscape.

theruss (Thu, 16 Nov 2017 19:09:52 GMT):
@jljordan_bcgov Thanks, very much appreciated. I understood the same wr/t to identity being a first-class citizen, and was just wishing to ensue I wasn't missing something.

theruss (Thu, 16 Nov 2017 19:09:52 GMT):
@jljordan_bcgov Thanks, very much appreciated. I understood the same wr/t to identity being a first-class citizen, and was just wishing to ensure I wasn't missing something.

theruss (Thu, 16 Nov 2017 19:10:08 GMT):
AFAIKT, Indy really does satisy the high level reqs of the project at this point, other than the truly "open" bit. It is highly likely we'd need to set this up as a permissioned network, even though the sensitive data would be held by the user, not on-chain, and regardless of the cryptographic security offered by Indy itself.

theruss (Thu, 16 Nov 2017 19:10:40 GMT):
It's the nature of the entities involved at this point.

theruss (Thu, 16 Nov 2017 19:10:40 GMT):
It's simply the nature of the entities involved at this point.

jljordan_bcgov (Thu, 16 Nov 2017 19:29:36 GMT):
if you run your own instance of Indy ... you decide on whether or not the ledger is public

jljordan_bcgov (Thu, 16 Nov 2017 19:30:02 GMT):
the nodes that write to the ledger are permissioned in both networks

jljordan_bcgov (Thu, 16 Nov 2017 19:30:50 GMT):
so the choice of public vs not-public in terms of access to the ledger is a choice

drummondreed (Fri, 17 Nov 2017 01:20:00 GMT):
@theruss Yes, @jljordan_bcgov is right on that the choice of a public vs. non-public ledger is up to the implementer. At one end you have Sovrin as a public permissioned ledger for a global public utility with governance by an international non-profit foundation and a global trust framework (https://sovrin.org/trust-framework/). But at the other end you could have a specific community running an Indy ledger with minimal governance. The full range is open.

theruss (Fri, 17 Nov 2017 01:20:42 GMT):
@drummondreed Understood, thank you.

theruss (Fri, 17 Nov 2017 01:22:34 GMT):
Forgive the following question; but how much of the Sovrin Trust Model (I've been going off this that you posted some time ago: http://forum.sovrin.org/t/recording-claims-on-behalf-of-another-agent/151/5) is "enshrined" within Indy at the protocol level, seeing as identity is this first-class citizen?

theruss (Fri, 17 Nov 2017 01:23:35 GMT):
...when I say "going off", I of course mean "working from" those emboldened definitions :-)

theruss (Fri, 17 Nov 2017 01:23:35 GMT):
...when I say "going off", I of course mean "working from" - those emboldened definitions :-)

drummondreed (Fri, 17 Nov 2017 05:32:10 GMT):
It's a good question @theruss. The answer is "quite a bit" and "more all the time", i.e., we are developing the Sovrin protocol to enable interoperation according to the Sovrin Trust Framework, not the other way around.

nitefrog (Fri, 17 Nov 2017 06:44:53 GMT):
Has joined the channel.

silasdavis (Fri, 17 Nov 2017 10:32:59 GMT):
Has joined the channel.

silasdavis (Fri, 17 Nov 2017 10:33:13 GMT):
Hi there, Silas Davis maintainer of Burrow here

silasdavis (Fri, 17 Nov 2017 10:33:27 GMT):
I was just wondering if anyone from the Indy project will be at hackfest?

silasdavis (Fri, 17 Nov 2017 10:33:38 GMT):
in Lisbon

silasdavis (Fri, 17 Nov 2017 10:37:32 GMT):
On an unrelated note does Indy/Sovrin have any thoughts about password-based identity authentication. In particular I have wanted to implement password based key exchange to give long-key-like security based on a relatively shorter password: https://en.wikipedia.org/wiki/Password-authenticated_key_agreement as a way of getting around key management (would require support for ephemeral key pairs on a DLT)

silasdavis (Fri, 17 Nov 2017 10:38:10 GMT):
could this fit in any way with the Indy model? I sort of assume not, and that Indy depends on you looking after your root private key matter

JasperR (Fri, 17 Nov 2017 14:26:42 GMT):
Has joined the channel.

Suedonym (Fri, 17 Nov 2017 18:16:49 GMT):
@nage :arrow_up:

nage (Fri, 17 Nov 2017 18:18:09 GMT):
@silasdavis we will have a large contingent there in Lisbon. I know there are plans to have some good hacking sessions around anonymous credentials and shared crypto code.

nage (Fri, 17 Nov 2017 18:23:26 GMT):
As for password-based identity authentication there is a lot of research in this area, and it would be good to reach out to @drummondreed about the distributed key management systems work that he is spearheading. As for having a password or seed to generate a key, there are a lot of tradeoffs to consider there, including what keys are the most important and how one key can be used to generate many others. The bitcoin and ethereum communities have pioneered a lot of interesting techniques and tried a lot of things that can now be compared. The open source Sovrin code can use a seed to generate keys, but how this seed is generated can affect the security of your key (it shouldn't be guessable or easily determined, and our human brains are bad at randomness).

ahp (Fri, 17 Nov 2017 19:56:05 GMT):
Has joined the channel.

Infodataway (Sat, 18 Nov 2017 10:34:28 GMT):
Has joined the channel.

curtissummers (Sat, 18 Nov 2017 15:13:03 GMT):
Has joined the channel.

theruss (Sun, 19 Nov 2017 20:32:19 GMT):
@drummondreed again, many thanks for your response. I'm still working through what is the "indy" part, and what is the "sovrin" part, where each are mentioned in the former's docs.

theruss (Sun, 19 Nov 2017 23:41:27 GMT):
@nage I was just looking at this at the weekend, and I came across the following which may be useful (2 of 3 blind key management) https://blog.gridplus.io/2-of-3-blind-key-multi-signature-security-965e90640b4

saadmir (Mon, 20 Nov 2017 02:59:05 GMT):
Has joined the channel.

nghia47 (Mon, 20 Nov 2017 07:40:14 GMT):
Has joined the channel.

cca88 (Mon, 20 Nov 2017 16:25:27 GMT):
Has joined the channel.

MarioPissardo (Mon, 20 Nov 2017 22:06:26 GMT):
Has joined the channel.

theruss (Tue, 21 Nov 2017 02:53:23 GMT):
Possibly a daft question but; Am I correct in thinking that there's nothing special about Indy that precludes BA's and dev's being able to model business logic and develop APIs for an Identify soluton using Hyperledger Composer?

prmdmshra (Tue, 21 Nov 2017 11:10:24 GMT):
Has joined the channel.

prmdmshra (Tue, 21 Nov 2017 11:13:22 GMT):
Hi All, I am trying to build a PoC on Birth/Death certificate registration using Hyperledger Indy. I am familiar with Hyperledger Composer/Fabric. Kindly suggest whether we have any sample application for Birth Certificate Registration in Indy.

drummondreed (Wed, 22 Nov 2017 07:38:06 GMT):
Hi @prmdmshra. Indy doesn't yet have any Composer-like tools for building verifiable claims or credentials like a birth certificate. You have to build the schema and claim definition manually at this point. I'm not actually a coder, so I'll leave it to the developers here to provide you more guidance.

jagansundar (Wed, 22 Nov 2017 08:57:45 GMT):
Has joined the channel.

jagansundar (Wed, 22 Nov 2017 08:58:01 GMT):
Hi All,

jagansundar (Wed, 22 Nov 2017 09:08:08 GMT):
Hi All, I am new to Idny and just started reading. I have few basic question on Indy 1) Is the data/transcript distributed across the Indy nodes? 2) Can we track the history of these data/transcript? If yes, how do we achieve this? 3) If I have hyperledger fabric blockchain network, can I integrate Indy in the same nodes? Or should Indy be totally a different network? 4) Do Indy support smart contract? Or it is only about the digital identity? Can you please help me in above questions? Thank You Jagan S

mountbranch (Wed, 22 Nov 2017 09:40:34 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=CeB2qv6WtR64ocriP) @prmdmshra Hi! I'd be very interested in your application! Are you publishing it on GitHub or could we talk some to discuss your strategy and problems?

lovesh (Wed, 22 Nov 2017 10:08:59 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=N7Tgx5KQPq5o3esGz) @jagansundar Only the Schema (structure of data/transcript) and Claim Definition (public keys and some other cryptographic material for the issuer) is distributed across the Indy nodes, the actual user data is not. Indy runs as a separate network than fabric. Indy does not support smart contract

jagansundar (Wed, 22 Nov 2017 10:35:45 GMT):
@lovesh Thank you. If only the schema is distributed then, if the node which has data is out of the network, then the data would be lost. How do other nodes get the data? For example if university is issuing degree certificate and if the university node is out of the network then how do the employer or bank node get the degree information of the student

mountbranch (Wed, 22 Nov 2017 10:41:11 GMT):
I think that a secure channel is opened between the applicant and the employer where the data is shared. The claim is verifiable because it can be verified by the employer through the ZKP that the University has published. Or maybe the applicant just publishes a ZKP that his/her certificate is part of the list of issued certificates by the Uni. But please correct me if I'm wrong ppl with more understanding! :D

panickervinod (Wed, 22 Nov 2017 13:54:43 GMT):
Is there a web3.js equivalent for indy?

panickervinod (Wed, 22 Nov 2017 13:55:53 GMT):
to interface with and electron app.

panickervinod (Wed, 22 Nov 2017 13:55:53 GMT):
to interface with an electron app.

conroydave (Wed, 22 Nov 2017 20:41:25 GMT):
hello Indy team

conroydave (Wed, 22 Nov 2017 20:42:00 GMT):
was wondering if there are any open source projects looking at biometrics for digital passports with indy

conroydave (Wed, 22 Nov 2017 20:42:33 GMT):
working on a project right now regarding refugees, would love to use Indy.. but i might have to use fabric/composer given the time window

theruss (Thu, 23 Nov 2017 01:33:08 GMT):
@conroydave I spotted this on my travels today with regard to the use of biometrics. The particular use-case of biometric data desrcibed, seems geared towards using biometric data to unlock a wallet app, rather than as the basis for a unique identifier. But devil's advocate; what does it matter the source of entropy or randomness, that generates a user's DID and public keys? https://sovrin.discoursehosting.net/t/convenient-authentication-on-self-sovereign-identity/39/2

theruss (Thu, 23 Nov 2017 01:35:59 GMT):
A couple of random questions I need help answering, for a proposal myself and my team are putting together. I see that back in May, Sovrin and Engage Identity were to work together to audit Indy's trust model (and assuming the codebase also). Is this process complete? What were the conclusions?

conroydave (Thu, 23 Nov 2017 01:38:49 GMT):
Here I was thinkin sovrin And Indy were the same

theruss (Thu, 23 Nov 2017 01:39:06 GMT):
Yeah, it's hard and on a technical level "they" are.

theruss (Thu, 23 Nov 2017 01:39:44 GMT):
You need to make the distinction between the "Sovrin Foundation" and the "Sovrin Project" (codebase) that has since been donated to the Hyperledger Foundation

theruss (Thu, 23 Nov 2017 01:39:44 GMT):
You need to make the distinction between the "Sovrin Foundation" and the "Sovrin Project" (codebase) that has since been donated to the Hyperledger Project (Under the Linux Foundation)

theruss (Thu, 23 Nov 2017 01:40:18 GMT):
....and donated as "Hyperledger Indy"

theruss (Thu, 23 Nov 2017 01:40:59 GMT):
Things are compounded (and conceded by maintainers) that the distinction is in flux with regard to how the docs are written.

theruss (Thu, 23 Nov 2017 01:41:49 GMT):
For all intents and purposes, unless you see the word "Organisation" or "Foundation" alongside the word "Sovrin", asume that "sovrin" refers to "Indy".

theruss (Thu, 23 Nov 2017 01:42:13 GMT):
If I'm wrong on any point there, mods, maintainers etc, please feel free to correct me :-)

theruss (Thu, 23 Nov 2017 01:43:05 GMT):
I've found many anwers to my Q's on the Sovrin Forum, and I just mentally replace "sovrin" with "indy" when I need to.

theruss (Thu, 23 Nov 2017 01:44:50 GMT):
The other question I wanted answered was around the process for divulging customer data to the police for example. I suppose there is no PII on-ledger, so there's not much to "divulge" if asked, but I'm going to be asked this, and I should be able to give an answer, even if it's just "It's up to the implementation".

swcurran (Thu, 23 Nov 2017 03:08:01 GMT):
@theruss - HL Indy is an piece of Open Source Software. Sovrin is a running a live instance of HL Indy. In theory, you could run your own instance of HL Indy as a network, independent of Sovrin. We are temporarily doing that during our development. But when we go live, we plan to be using Sovrin.

theruss (Thu, 23 Nov 2017 03:08:34 GMT):
@swcurran Understood.

theruss (Thu, 23 Nov 2017 03:09:07 GMT):
Myself and my team are planning an Indy implementation of our own (with other HL projects added)

theruss (Thu, 23 Nov 2017 03:09:07 GMT):
Myself and my team are planning an Indy implementation of our own (with other HL projects in concert)

theruss (Thu, 23 Nov 2017 03:09:20 GMT):
But still very much in the R&D phase right now.

theruss (Thu, 23 Nov 2017 03:11:53 GMT):
It hasn't escaped my understanding that the entities to be provisioned onto (our) network, could run their own Sovrin nodes.

theruss (Thu, 23 Nov 2017 03:12:50 GMT):
Regardless, the reading I've been doing has been invaluabe.

theruss (Thu, 23 Nov 2017 03:12:50 GMT):
Regardless, the reading I've been doing of late, has been invaluable.

swcurran (Thu, 23 Nov 2017 03:16:01 GMT):
There is no useful data on the Ledger that the node operators could divulge. All of the Verifiable Claim data that is exchanged may be held by the Issuers (as they see fit), and the Holder - usually the individual about which claims has been made. So a citizen would be holding claims they have received. What the police could do about that data is presumably up to the police in the jurisdiction of the citizen. The advantage over the current model is that the issuer (an online service, for example) need not hold onto the data that was used to identify the user. For example, they need not hold the name and address of the person locally - if they need that data, they can request it from the user. So their liability decreases, and there is less information that can be accessed outside of the control of the user (hacked, compelled by 3rd parties, etc.)

swcurran (Thu, 23 Nov 2017 03:18:50 GMT):
We are really interested in the global reach of networks like Sovrin, so if and when we go live, we definitely want to be part of those. We're experimenting now to ensure we understand the risks of using such a network vs. operating our own. In our use cases - our own network will lack the needed reach.

swcurran (Thu, 23 Nov 2017 03:18:50 GMT):
We are really interested in the global reach of networks like Sovrin, so if and when we go live, we definitely want to be part of those. We're experimenting now to ensure we understand the risks of using such a network vs. operating our own. In our use cases - our own network will lack the needed reach, and at this point, we don't see using the global solutions will add risk. More learning to come though!

darrell.odonnell (Thu, 23 Nov 2017 12:54:18 GMT):
@swcurran I think it's also worth mentioning that the request (for name & address) could also be done "just in time" so the liability can decrease even further. There is an interesting future use case where a vendor (e.g. Amazon) doesn't really need to know my address - they just need to be able to pass something on to the delivery service (e.g. UPS) that my Agent would tell where to ship - and that may be dynamic in time (e.g. hotel next week, home the week after, office this week).

lovesh (Thu, 23 Nov 2017 13:35:00 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=GaGXMbraajgDQPbXZ) @jagansundar Non ledger node has any identity-owner specific data, the ledger nodes records data that is used by relying parties to verify the issuer's attestation over the identity-owner's claims, in the university example, the university will put it's public keys (it is not an EDDSA key) on the ledger and when it issues the certificate to a student, the certificate data is signed by the university and given to the student using a secure channel, when the student uses the certificate at some relying party, the relying party verifies the university's signature using the public key on the ledger. The ledger is used for revocation too but it is important to note that any identity-owner specific data (say PII) is never stored on the ledger

lovesh (Thu, 23 Nov 2017 13:35:00 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=GaGXMbraajgDQPbXZ) @jagansundar No ledger node has any identity-owner specific data, the ledger nodes records data that is used by relying parties to verify the issuer's attestation over the identity-owner's claims, in the university example, the university will put it's public keys (it is not an EDDSA key) on the ledger and when it issues the certificate to a student, the certificate data is signed by the university and given to the student using a secure channel, when the student uses the certificate at some relying party, the relying party verifies the university's signature using the public key on the ledger. The ledger is used for revocation too but it is important to note that any identity-owner specific data (say PII) is never stored on the ledger

the_identity_guy (Sun, 26 Nov 2017 23:23:25 GMT):
hi is there any tutorials on how to create custom agents/nodes etc after completing the getting started guide on Indy/Sovrin?

the_identity_guy (Sun, 26 Nov 2017 23:47:47 GMT):
do I have to setup the Vagrant development environment to be able to create new agents/nodes and verification keys?

DerionSickRage (Mon, 27 Nov 2017 00:33:21 GMT):
Has joined the channel.

mgbailey (Mon, 27 Nov 2017 16:06:53 GMT):
@the_identity_guy, I have something that may help you at https://docs.google.com/document/d/1kQcdICAYmSqbk4d5lUtFhk2L55dKJMfyM0JbZPeG55s/edit?usp=sharing

pupeter (Mon, 27 Nov 2017 16:42:18 GMT):
Has joined the channel.

theruss (Mon, 27 Nov 2017 17:31:10 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=9qBZ6JSy9BYSinZcW) @swcurran While it's true that network participants _need not_ directly examine PII data, these data have to exist _somewhere_, and we cannot expect _every_ scrap of data from driver's licence to medical records to be stored on a customer's mobile device. There is also "current state" to be considered; which is to say the transitional period any municipality / region / country needs to go through when bringing such an SSI system online, in porting silo'd data "somewhere else". I envisage that there will be a decidely long time (~10y+) where current data silo's containing yours and my PII data persist within organisations who are simulateneously SSI system participants. We cannot expect with all the wherewithal in the world, that _all_ data from _all_ participating agencies' silos are ported to devices or hardware wallets or other "trusted" 3rd parties. Taking this as predicate then; it follows that such (toxic) data is still accessible by said agencies using standard processes such as intranet applications as executables, web-based query tools or whatever. The one saving grace being that in a SSI network context, assuming such "standard" (legacy) query methods are howsoever used, the claims data and silo'd PII data cannot be correlated.

swcurran (Mon, 27 Nov 2017 17:41:48 GMT):
@theruss - I think we are agreeing. The bulk of the data will remain with the service, but the data that can correlate it back to the users (that which makes it useful for illicit action) will NOT be stored with that bulk of data, but will remain in the control of the user. The speed of this will depend on the ease of the transition and the motivation to change. Rules like GDPR and hacks like Equifax will drive the stick motivation, and good software will provide the carrot. Go Sovrin!! One other factor makes the data less useful in an SSI-world. When it becomes "normal" that the only PII data accepted is that backed by a Verifiable Claim, then PII data in the wild becomes much less valuable. If you can't use a SSN as plain-text, but only as a Verifiable Claim, then having millions of SSNs will be much less valuable. I think this will take longer than de-toxifying enterprise data, but when it arrives it will be huge help in restricting the current type of break-ins we are seeing.

the_identity_guy (Mon, 27 Nov 2017 17:44:34 GMT):
thank you @mgbailey

theruss (Mon, 27 Nov 2017 17:44:47 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=kCy84mSnKBa7qJb2C) @swcurran Cheers - great to bounce my team's current thinking from others.

theruss (Mon, 27 Nov 2017 17:44:47 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=kCy84mSnKBa7qJb2C) @swcurran Cheers - great to bounce my team's current thinking off others.

jljordan_bcgov (Mon, 27 Nov 2017 18:54:15 GMT):
@swcurran @theruss If it becomes practical and true that unverifiable PII would not be accepted then this creates a very strong economic incentive for services requiring PII (e.g. banks, insurance, utilities, govt, commerce) to adopt Verifiable Claims based technology as a strong risk mitigation measure against fraud ... quite interesting

theruss (Mon, 27 Nov 2017 19:00:06 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=85xEcS5tBEw7W64fC) @jljordan_bcgov Yes indeed, this is the next logical step. The whole world of DLT / Blockchain is interesting!

mpiekarska (Tue, 28 Nov 2017 13:59:22 GMT):
Dear Indy Community! We are coorganizing IEEE Security and Privacy on Blockchain workshop as part of the Euro S&P this year in London: https://www.ieee-security.org/TC/EuroSP2018/ . We are seeking interested and experienced people who can volunteer as members of Program Committee. Please email me privately if you would like to participate: mpiekarska@linuxfoundation.org

jww (Wed, 29 Nov 2017 15:53:54 GMT):
Has joined the channel.

Sean_Bohan (Wed, 29 Nov 2017 17:58:46 GMT):
folks: Notes from this week's IdentityWG Call: https://docs.google.com/document/d/1nVkjhSlghjc4KFWEYDU2B0ztWCFh1CK8a7VmqwFsnKY/edit#

Sean_Bohan (Wed, 29 Nov 2017 18:04:47 GMT):
ndy Agent Folks: TOMORROW is the Indy Agent WG Call - Thursday November 30, 2017 Agenda: A. Housekeeping (Nathan and Sean) B. Development Achievements (Alex and Slava) C.DEMO and discussion with the team from British Columbia: @jljordan_bcgov and @swcurran D. Wrap Up Call details: Hyperledger Indy Agent (SDK) WG Call When: November 30, 2017 8amPT, 9amMT, 11amET, 4pm BST Where: Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/232861185 Or iPhone one-tap (US Toll): +16465588656,232861185# or +14086380968,232861185# Or Telephone: Dial: +1 646 558 8656 (US Toll) or +1 408 638 0968 (US Toll) Meeting ID: 232 861 185 International numbers available: https://zoom.us/zoomconference?m=a0jD_rTMnh0ZYGQDOKPCNrK_0dP7WPfp1 Please note: This meeting will be recored Bring a friend. It will be fun. - Hyperledger Indy Community Hyperledger Community Indy Rocketchat: https://chat.hyperledger.org/channel/indy https://chat.hyperledger.org/channel/indy-agent https://chat.hyperledger.org/channel/indy-node Indy Wiki: https://wiki.hyperledger.org/projects/indy

wittrock (Wed, 29 Nov 2017 18:55:24 GMT):
Has joined the channel.

nbrempel (Wed, 29 Nov 2017 19:01:47 GMT):
Hi all, not sure what the best channel to ask this in: It looks like all nodes listen for traffic on different ports starting from 9700: https://github.com/sovrin-foundation/sovrin/blob/master/sovrin/pool_transactions_live_genesis Is there a reason for this? Each IP could listen on 2 dedicated ports for client and node connections. We have an environment where outbound ports are not open by default. Requesting an arbitrary number of outbound ports may not be feasible. Is it possible to predict which outbound ports we need to open to have an agent connect to the node pool?

sklump (Wed, 29 Nov 2017 19:53:38 GMT):
If I understand it correctly, the genesis transaction(s) specify these; e.g, ``` { "data": { "alias": "Node1", "blskey": "4N8aUNHSgjQVgkpm8nhNEfDf6txHznoYREg9kirmJrkivgL4oSEimFF6nsQ6M41QvhM2Z33nves5vfSn9n1UwNFJBYtWVnHYMATn76vLuL3zU88KyeAYcHfsih3He6UHcXDxcaecHVz6jhCYz1P2UZn2bDVruL5wXpehgBfBaLKm3Ba", "client_ip": "10.0.0.2", "client_port": 9702, "node_ip": "10.0.0.2", "node_port": 9701, "services": [ "VALIDATOR" ] }, "dest": "Gw6pDLhcBcoQesN72qfotTgFa7cbuqZpkX3Xo6pLhPhv", "identifier": "Th7MpTaRZVRYnPiabds81Y", "txnId": "fea82e10e894419fe2bea7d96296a6d46f50f93f9eeda954ec461b2ed2950b62", "type": "0" }```

sklump (Wed, 29 Nov 2017 19:54:24 GMT):
So if you can run a pool off multiple IP addresses, there's nothing stopping you from defining each node on its own IP address, with each node using the same two ports?

compleatang (Wed, 29 Nov 2017 19:56:14 GMT):
Has left the channel.

nbrempel (Wed, 29 Nov 2017 20:22:48 GMT):
I'm more concerned that the provisional network running right now uses arbitrary ports. ``` {"data":{"alias":"ev1","client_ip":"54.207.36.81","client_port":"9702","node_ip":"18.231.96.215","node_port":"9701","services":["VALIDATOR"]},"dest":"GWgp6huggos5HrzHVDy5xeBkYHxPvrRZzjPNAyJAqpjA","identifier":"J4N1K1SEB8uY2muwmecY5q","txnId":"b0c82a3ade3497964cb8034be915da179459287823d92b5717e6d642784c50e6","type":"0"} {"data":{"alias":"zaValidator","client_ip":"154.0.164.39","client_port":"9702","node_ip":"154.0.164.39","node_port":"9701","services":["VALIDATOR"]},"dest":"BnubzSjE3dDVakR77yuJAuDdNajBdsh71ZtWePKhZTWe","identifier":"UoFyxT8BAqotbkhiehxHCn","txnId":"d5f775f65e44af60ff69cfbcf4f081cd31a218bf16a941d949339dadd55024d0","type":"0"} {"data":{"alias":"danube","client_ip":"128.130.204.35","client_port":"9722","node_ip":"128.130.204.35","node_port":"9721","services":["VALIDATOR"]},"dest":"476kwEjDj5rxH5ZcmTtgnWqDbAnYJAGGMgX7Sq183VED","identifier":"BrYDA5NubejDVHkCYBbpY5","txnId":"ebf340b317c044d970fcd0ca018d8903726fa70c8d8854752cd65e29d443686c","type":"0"} {"data":{"alias":"royal_sovrin","client_ip":"35.167.133.255","client_port":"9702","node_ip":"35.167.133.255","node_port":"9701","services":["VALIDATOR"]},"dest":"Et6M1U7zXQksf7QM6Y61TtmXF1JU23nsHCwcp1M9S8Ly","identifier":"4ohadAwtb2kfqvXynfmfbq","txnId":"24d391604c62e0e142ea51c6527481ae114722102e27f7878144d405d40df88d","type":"0"} {"data":{"alias":"digitalbazaar","client_ip":"34.226.105.29","client_port":"9701","node_ip":"34.226.105.29","node_port":"9700","services":["VALIDATOR"]},"dest":"D9oXgXC3b6ms3bXxrUu6KqR65TGhmC1eu7SUUanPoF71","identifier":"rckdVhnC5R5WvdtC83NQp","txnId":"56e1af48ef806615659304b1e5cf3ebf87050ad48e6310c5e8a8d9332ac5c0d8","type":"0"} {"data":{"alias":"OASFCU","client_ip":"38.70.17.248","client_port":"9702","node_ip":"38.70.17.248","node_port":"9701","services":["VALIDATOR"]},"dest":"8gM8NHpq2cE13rJYF33iDroEGiyU6wWLiU1jd2J4jSBz","identifier":"BFAeui85mkcuNeQQhZfqQY","txnId":"825aeaa33bc238449ec9bd58374b2b747a0b4859c5418da0ad201e928c3049ad","type":"0"} {"data":{"alias":"BIGAWSUSEAST1-001","client_ip":"34.224.255.108","client_port":"9796","node_ip":"34.224.255.108","node_port":"9769","services":["VALIDATOR"]},"dest":"HMJedzRbFkkuijvijASW2HZvQ93ooEVprxvNhqhCJUti","identifier":"L851TgZcjr6xqh4w6vYa34","txnId":"40fceb5fea4dbcadbd270be6d5752980e89692151baf77a6bb64c8ade42ac148","type":"0"} {"data":{"alias":"DustStorm","client_ip":"207.224.246.57","client_port":"9712","node_ip":"207.224.246.57","node_port":"9711","services":["VALIDATOR"]},"dest":"8gGDjbrn6wdq6CEjwoVStjQCEj3r7FCxKrA5d3qqXxjm","identifier":"FjuHvTjq76Pr9kdZiDadqq","txnId":"6d1ee3eb2057b8435333b23f271ab5c255a598193090452e9767f1edf1b4c72b","type":"0"} {"data":{"alias":"prosovitor","client_ip":"138.68.240.143","client_port":"9711","node_ip":"138.68.240.143","node_port":"9710","services":["VALIDATOR"]},"dest":"C8W35r9D2eubcrnAjyb4F3PC3vWQS1BHDg7UvDkvdV6Q","identifier":"Y1ENo59jsXYvTeP378hKWG","txnId":"15f22de8c95ef194f6448cfc03e93aeef199b9b1b7075c5ea13cfef71985bd83","type":"0"} {"data":{"alias":"iRespond","client_ip":"52.187.10.28","client_port":"9702","node_ip":"52.187.10.28","node_port":"9701","services":["VALIDATOR"]},"dest":"3SD8yyJsK7iKYdesQjwuYbBGCPSs1Y9kYJizdwp2Q1zp","identifier":"JdJi97RRDH7Bx7khr1znAq","txnId":"b65ce086b631ed75722a4e1f28fc9cf6119b8bc695bbb77b7bdff53cfe0fc2e2","type":"0"} ``` In the future, if we want to connect, we'll need to allow outbound traffic for every port that validator nodes use. Right now it seems they choose ports arbitrarily between 9700 and 9769.

nage (Wed, 29 Nov 2017 22:00:21 GMT):
This is because of how both RAET and CurveZMQ (based on CurveCP) handle keys on a per port basis (the ledger was never moved to pairwise CurveZMQ like we temporarily tried with agents in indy-sdk). @lovesh or @ashcherbakov may be able to point us to more recent information or documentation

nage (Wed, 29 Nov 2017 22:01:51 GMT):
@mgbailey is there a "best practice" on port usage or documentation on what to expect from a firewall perspective?

mgbailey (Wed, 29 Nov 2017 22:15:38 GMT):
@nbrempel As a "diversity" feature, stewards are allowed to select their own client and node (inter-validator) ports within the 9700-9799 range. Stewards are instructed to either globally enable outgoing traffic from their validator to this range, or to enable outgoing to the specific ip addresses and ports that are actually in use by the validators in the pool. The latter approach is much more administrator time-intensive, since validators are added and removed from the pool on an ongoing basis, and the required ip address/port combinations will need to be upgraded promptly. The [Sovrin Provisional Network networking diagram](https://drive.google.com/file/d/0B8BSiLQLtlY8Zzd1UG9iNG1RTTA/view?usp=sharing) includes a listing of the current ip addresses and node ports.

Russell-Columbia (Wed, 29 Nov 2017 23:02:45 GMT):
Has joined the channel.

nbrempel (Wed, 29 Nov 2017 23:47:35 GMT):
Thank you

arjanvaneersel (Thu, 30 Nov 2017 07:36:39 GMT):
Has joined the channel.

lovesh (Thu, 30 Nov 2017 09:45:15 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=HXqys8ZwLDbrm9SzD) @nbrempel There is no reason why each validator cannot use the same 2 ports for node-to-node and node-to-client traffic, the reason why it used to be an arbitrary choice was that very early pools had all machines running on the same machine, so the first node would use 9701,9702, 2nd node would use 9703,9704 and so on, but it is not relevant anymore

genggjh (Thu, 30 Nov 2017 14:02:33 GMT):
Has joined the channel.

chriswinc (Thu, 30 Nov 2017 14:52:48 GMT):
Has joined the channel.

theruss (Thu, 30 Nov 2017 20:37:50 GMT):
Hi folks, I have been reading a lot of the disparate docs around Indy, Sovrin Trust, DID Records and the rest of it, but I am yet to find a decent (actually - any) description of how "physically" a trust-anchor, or any suitable qualified role, pushes ("provisions"?) a new DID+DDO pair into the ledger. Can someone point me to a resource that will help me figure out this misisng piece? Thanks!

theruss (Thu, 30 Nov 2017 20:37:50 GMT):
Hi folks, I have been reading a lot of the disparate docs around Indy, Sovrin Trust, DID Records and the rest of it, but I am yet to find a decent (actually - any) description of how "physically" a trust-anchor, or any suitably qualified role, pushes ("provisions"?) a new DID+DDO pair into the ledger. Can anyone point me to a resource that will help me figure out this missing piece please? Thanks!

nbrempel (Thu, 30 Nov 2017 21:39:08 GMT):
We've been working on this. We have an early test implementation here: https://github.com/PSPC-SPAC-buyandsell/von_agent/blob/master/test/test_agents.py#L94

nbrempel (Thu, 30 Nov 2017 21:39:08 GMT):
@theruss We've been working on this. We have an early test implementation here: https://github.com/PSPC-SPAC-buyandsell/von_agent/blob/master/test/test_agents.py#L94

nbrempel (Thu, 30 Nov 2017 21:40:44 GMT):
Here is a sequence diagram of what's happening: https://github.com/PSPC-SPAC-buyandsell/demo-agent/blob/master/doc/pic/boot.tag.png

alkopnin (Mon, 04 Dec 2017 07:45:16 GMT):
Has joined the channel.

alkopnin (Mon, 04 Dec 2017 07:47:08 GMT):
Hi folks, I have had the similar issue on my Mac. Do you have any recommendations how to fix it? https://github.com/rust-lang/rust/issues/45629 https://github.com/hyperledger/indy-sdk/issues/295

gudkov (Mon, 04 Dec 2017 08:57:22 GMT):
@alkopnin > https://github.com/hyperledger/indy-sdk/issues/295 This issue already fixed current stable and master.

alkopnin (Mon, 04 Dec 2017 08:58:55 GMT):
Okay, sounds good. Than you! Hence I should work with the master not with tags, is it right?

gudkov (Mon, 04 Dec 2017 09:03:31 GMT):
It depends on your needs. The latest stable can be build without _je_malloc_usable_size error too.

gudkov (Mon, 04 Dec 2017 09:04:10 GMT):
stable branch always corresponds to the latest release tag now.

alkopnin (Mon, 04 Dec 2017 09:11:21 GMT):
I see, you are talking about jemalloc bug. I'm curious about _crypto_stream_aes128ctr_xor one.

alkopnin (Mon, 04 Dec 2017 09:11:38 GMT):
comment below in the same ticket

alkopnin (Mon, 04 Dec 2017 09:12:04 GMT):
by giulianograssi 17 days ago

gudkov (Mon, 04 Dec 2017 09:30:31 GMT):
> I'm curious about _crypto_stream_aes128ctr_xor one. To solve this you need to downgrade libsodium version. Released sodiumoxide crate isn't compatible with the latest libsodium. Mac build instruction describes how to install correct version.

alkopnin (Mon, 04 Dec 2017 14:22:53 GMT):
thanks, it builds

alkopnin (Mon, 04 Dec 2017 14:22:53 GMT):
thanks, it works

alkopnin (Mon, 04 Dec 2017 14:57:36 GMT):
one more question from my side, in demo you have dependensi

Daniela_Barbosa (Tue, 05 Dec 2017 03:24:52 GMT):
Has joined the channel.

JOYELIN (Tue, 05 Dec 2017 07:04:07 GMT):
Has joined the channel.

jeffblack360 (Tue, 05 Dec 2017 15:36:18 GMT):
Has joined the channel.

riichard (Tue, 05 Dec 2017 17:54:52 GMT):
Has joined the channel.

ronniemh (Tue, 05 Dec 2017 20:35:04 GMT):
Has joined the channel.

gudkov (Wed, 06 Dec 2017 12:52:37 GMT):
We have a pool request to Indy SDK related to usage of Indy Crypto as backed for anoncreds protocol. There are no changes in API, but It will cause changing of serialization of some anoncreds entities. As result claims from previous Indy SDK version will be incompatible with the new code after this commit. Please let me known if it is problem for someone. We will try to align this. PR https://github.com/hyperledger/indy-sdk/pull/401

lenardb79 (Wed, 06 Dec 2017 22:15:16 GMT):
Has joined the channel.

hendry19901990 (Wed, 06 Dec 2017 22:39:15 GMT):
Has joined the channel.

jagansundar (Thu, 07 Dec 2017 11:55:44 GMT):
We have tried the getting started guide. Is there any other document/ tutorial available for next steps? In the getting started guide, the CLI is used for Alice and when I prompted/changed to Fabber then there I don't see the connections. Do we need to do something else to see the connection in Fabber and Banker's CLI? Where do I see the request and response of Alice's in Acme and Banker's agent/CLI?

jagansundar (Thu, 07 Dec 2017 12:00:12 GMT):
As I understand Indy is a permission public blockchain network. In Etherium, anyone can be part of the blockchain network, by using the Etherium platform/application. In Indy how do the end user join the Indy's network? Is there any application like Etherium for Indy through which we can join to the Indy's network?

swcurran (Thu, 07 Dec 2017 19:59:00 GMT):
@jagansundar - for running the demo, use this this repo. You need a bash shell, make and docker (easy on all but windows) and it starts all of the pieces you need for the full demo - nodes, command line and the agent code - Faber, Acme and Thrift. https://github.com/brycecurtis/indy-tutorial-sandbox Let me know if you are on windows. I'm going to be doing a pr shortly to put a fix in for Windows.

swcurran (Thu, 07 Dec 2017 20:04:44 GMT):
@jagansundar - HL-Indy is just the code, not a running instance. A running instance of HL-Indy (such as Sovrin) decides who will run the nodes that operate the network. With Sovrin those are called "Stewards" and they have to sign a legal agreement to become a node operating the Ledger (blockchain). Read up on "Trust Anchors" in the HL-Indy. They are the entities in Indy that are permitted to write to the Ledger. All Stewards are Trust Anchors, and they can make other entities Trust Anchors through a development of trust. Individuals use Trust Anchors to have their decentralized IDs written to the ledger. That's a start - gotta run. See the documents on sovrin.org site, under "Library" - https://sovrin.org/library/

jagansundar (Fri, 08 Dec 2017 04:14:05 GMT):
@swcurran Thank you

jagansundar (Fri, 08 Dec 2017 05:45:33 GMT):
@swcurran :Thank you for the clarification. If we have to develop an app using HL-Indy Sovrin network, do we need to be “Stewards“? or if “Stewards”, makes us a Trust Anchor, will be able to onboard individuals (Non Anchors)? In the getting started example, “Faber college”, “Acme corp” and “Thrift bank” are the Trust Anchors. Does this mean they are Stewards?, If they are different Stewards then in this examples, how are the proofs and claims are synchronized? My assumption of the getting started example is that an organization(XYZ) build an app which involves, college, students, hiring companies and bank. Here the organization (XYZ) is a Steward and that makes college, hiring companies and bank as “Trust Anchor” and the attributes in the claim and proof should be same Is HL-Indy GA available? If not any idea on tentative timeline? Cn you please clarify all above point? Thank you .

Audrius (Fri, 08 Dec 2017 12:01:29 GMT):
@swcurran Is it any rules on how often (as per releases, etc) Indy code is updated on Sovrin network nodes?

mgbailey (Fri, 08 Dec 2017 14:54:21 GMT):
There are no hard and fast rules on upgrade cadence on the nodes. The expectation is that when a release is promoted to the stable branch, after about week of testing it will be put onto the live network. In practice, this has been every month so.

swcurran (Fri, 08 Dec 2017 17:05:10 GMT):
@jagansundar - an app you would build might (not required) be an "Agency", which is an aggregation of agents that interact on behalf of end users and the Sovrin network. In that case, you would want to be a Trust Anchor, as you will be onboarding end users by writing to the network. You would not have to be a Steward - that would be a choice you would have. You could also write software that enables an organization to be "Sovrin-Enabled" - to allow a Service (bank, college) to interact with Sovrin. Again, the likely scenario is the operator of the Service running your app (e.g. the bank) would be Trust Anchor. We're building that type of component. In the Alice example, none of the organizations (Faber, etc.) are Stewards, because they don't operate the nodes of the network. The Stewards are unnamed in that example (just IP addresses), but they are the 4 (in the example) processes that are running the Ledger. The agents are writing to the Ledger (by sending messages to one or more of the nodes) and the nodes are doing the blockchain operations - distributing transactions, creating updates to the ledger, reaching consensus on the evolving state of the Ledger. Note that the claims are NOT written to the ledger - they are signed using the keys that are referenced on the ledger, but the claims and proofs are shared between the identities - e.g. Faber to Alice, Alice to Acme, etc. On your assumption - XYZ need not be a Steward, but would be Trust Anchor. Again - from an interacting with the Sovrin Ledger - you only need to be a Trust Anchor. You can be a Steward, but it's not necessary.

slafranca (Fri, 08 Dec 2017 18:37:16 GMT):
Has joined the channel.

Sean_Bohan (Fri, 08 Dec 2017 19:53:02 GMT):
Next week is the Indy Agent WG Call NEXT WEEK we have the Indy Agent WG Call - Thursday December 14, 2017 Agenda: A. Housekeeping (Nathan and Sean) B. Development Achievements (Alex and Slava) C. Dave Huesby (@dhuesby ) and Nathan George (@nage ) will share their experiences at the Hyperledger Hackfest D. Community Input discussion (Nathan) D. Wrap Up Call details: Hyperledger Indy Agent (SDK) WG Call When: December 14, 2017 8amPT, 9amMT, 11amET, 4pm BST Where: Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/232861185 Or iPhone one-tap (US Toll): +16465588656,232861185# or +14086380968,232861185# Or Telephone: Dial: +1 646 558 8656 (US Toll) or +1 408 638 0968 (US Toll) Meeting ID: 232 861 185 International numbers available: https://zoom.us/zoomconference?m=a0jD_rTMnh0ZYGQDOKPCNrK_0dP7WPfp1 Please note: This meeting will be recored Bring a friend. It will be fun. - Hyperledger Indy Community Hyperledger Community Indy Rocketchat: https://chat.hyperledger.org/channel/indy https://chat.hyperledger.org/channel/indy-agent https://chat.hyperledger.org/channel/indy-node Indy Wiki: https://wiki.hyperledger.org/projects/indy Indy WG Agendas: https://drive.google.com/open?id=1wNnp1pPS6-1Y4B2oygfI6tOU4eHGmpRH Indy WG Videos: https://drive.google.com/open?id=1Z8-jR7hnXb57fufE0OXxIfeUn05zNmRt Indy WG Meeting Recordings: https://drive.google.com/open?id=0B_NJV6eJXAA1UlJZMXd3cm1zNDQ

Sean_Bohan (Fri, 08 Dec 2017 19:53:02 GMT):
Next week is the Indy Agent WG Call TOMORROW is the Indy Agent WG Call - Thursday December 14, 2017 Agenda: A. Housekeeping (Nathan and Sean) B. Development Achievements (Alex and Slava) C. Dave Huesby (@dhuesby ) and Nathan George (@nage ) will share their experiences at the Hyperledger Hackfest D. Community Input discussion (Nathan) D. Wrap Up Call details: Hyperledger Indy Agent (SDK) WG Call When: December 14, 2017 8amPT, 9amMT, 11amET, 4pm BST Where: Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/232861185 Or iPhone one-tap (US Toll): +16465588656,232861185# or +14086380968,232861185# Or Telephone: Dial: +1 646 558 8656 (US Toll) or +1 408 638 0968 (US Toll) Meeting ID: 232 861 185 International numbers available: https://zoom.us/zoomconference?m=a0jD_rTMnh0ZYGQDOKPCNrK_0dP7WPfp1 Please note: This meeting will be recored Bring a friend. It will be fun. - Hyperledger Indy Community Hyperledger Community Indy Rocketchat: https://chat.hyperledger.org/channel/indy https://chat.hyperledger.org/channel/indy-agent https://chat.hyperledger.org/channel/indy-node Indy Wiki: https://wiki.hyperledger.org/projects/indy Indy WG Agendas: https://drive.google.com/open?id=1wNnp1pPS6-1Y4B2oygfI6tOU4eHGmpRH Indy WG Videos: https://drive.google.com/open?id=1Z8-jR7hnXb57fufE0OXxIfeUn05zNmRt Indy WG Meeting Recordings: https://drive.google.com/open?id=0B_NJV6eJXAA1UlJZMXd3cm1zNDQ

JeremyStanton (Fri, 08 Dec 2017 22:01:09 GMT):
Has joined the channel.

bjain (Sun, 10 Dec 2017 11:46:29 GMT):
Has joined the channel.

Audrius (Sun, 10 Dec 2017 12:55:06 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=R4o7Xm2qq2m3KWKmc) @mgbailey Thank you.

alexeykoren (Mon, 11 Dec 2017 16:01:22 GMT):
Has left the channel.

alexeykoren (Mon, 11 Dec 2017 16:01:39 GMT):
Has joined the channel.

Manreet (Tue, 12 Dec 2017 23:32:53 GMT):
Has joined the channel.

BrandonKlotz (Wed, 13 Dec 2017 16:34:48 GMT):
Has joined the channel.

nuxibyte_old (Thu, 14 Dec 2017 09:35:49 GMT):
Has joined the channel.

calvinx (Thu, 14 Dec 2017 09:43:11 GMT):
Has joined the channel.

arjanvaneersel (Thu, 14 Dec 2017 12:52:09 GMT):
Hello. I'm doing to getting started guide (https://github.com/hyperledger/indy-node/blob/stable/getting-started.md) with vagrant. However after setting up vagrant I don't have a prompt saying indy>, but sovrin>. Am I doing something wrong or is this the same?

ronniemh (Thu, 14 Dec 2017 13:54:22 GMT):
Hello. I want to create an actual Developer Environment connected to a sandbox (https://github.com/hyperledger/indy-node/blob/stable/getting-started.md) but the link that asks us to go does not exist (404) help me please

arjanvaneersel (Thu, 14 Dec 2017 14:15:01 GMT):
@ronniemh I think this is what you are looking for: https://github.com/evernym/sovrin-environments/tree/stable/vagrant/sandbox/DevelopmentEnvironment

arjanvaneersel (Thu, 14 Dec 2017 14:15:34 GMT):
There are indeed a few dead links in the documentation.

ronniemh (Thu, 14 Dec 2017 14:22:40 GMT):
@arjanvaneersel yes, thanks!

mgbailey (Thu, 14 Dec 2017 15:24:02 GMT):
@arjanvaneersel There has recently been rebranding of "sovrin" to "indy" The stable branch is still sovrin. The master branch is "indy". Hopefully the stable branch will be updated soon.

arjanvaneersel (Thu, 14 Dec 2017 15:25:19 GMT):
@mgbailey Thanks. I'm also having a problem with the sample files on vagrant. When I do "show sample/faber-request.indy", I get an error that the file doesn't exist

mgbailey (Thu, 14 Dec 2017 15:51:06 GMT):
@arjanvaneersel - same problem. Try "show sample/faber-request.sovrin" instead.

arjanvaneersel (Thu, 14 Dec 2017 15:51:32 GMT):
That one does exist

balmeida (Thu, 14 Dec 2017 16:20:26 GMT):
Has joined the channel.

gvvishwanath (Fri, 15 Dec 2017 05:21:22 GMT):
Has joined the channel.

calvinx (Fri, 15 Dec 2017 07:49:13 GMT):
New to indy and managed to get a test cluster of 4 nodes and 5 clients running and also load up the sample faber-request.indy files with Alice example. Used this simple-to-run docker tutorial - https://github.com/brycecurtis/indy-tutorial-sandbox

calvinx (Fri, 15 Dec 2017 07:49:59 GMT):
So how do I actually persist transactions in this sandbox cluster? What's the persistence mechanism used by the nodes?

calvinx (Fri, 15 Dec 2017 07:49:59 GMT):
So how do I actually persist wallets and transactions etc in this sandbox cluster? What's the persistence mechanism used by the nodes?

calvinx (Fri, 15 Dec 2017 09:28:56 GMT):
Also, is it possible to use a mobile client to interact with the nodes? I don't quite understand how I can have a mobile client transact with the cluster of nodes in my indy deployment if there are no REST APIs (mapped to a domain name) to call.

calvinx (Fri, 15 Dec 2017 09:28:56 GMT):
Also, is it possible to use a mobile client to interact with the nodes? I don't quite understand how I can have a mobile client transact with the cluster of nodes in my indy deployment if there are no REST APIs (mapped to a domain name) to call, which of course is the "traditional/centralized service approach". So, I must be missing a concept here.

calvinx (Fri, 15 Dec 2017 09:28:56 GMT):
Also, is it possible to use a mobile client to interact with the nodes? I don't quite understand how I can have a mobile client transact with the cluster of nodes in my indy deployment if there are no REST APIs (mapped to a domain name) to call, which of course is the "traditional/centralized service approach". So, I must be missing a concept somewhere. Would anyone be able to clarify how that is supposed to work? A mobile app client transacting with the Indy/Sovrin cluster of nodes.

calvinx (Fri, 15 Dec 2017 13:42:47 GMT):
And what do I need before I can connect to the live network?

mgbailey (Fri, 15 Dec 2017 15:16:04 GMT):
Hi @calvinx . In the demo code, your wallet is persisted in a hidden .indy-cli directory. Transactions are written to the distributed ledger, and can be seen on your validator nodes (which are docker images on your system) using the "sudo read_ledger --type domain" command from the command line. There is an libindy project that is creating a Rust CLI with wrappers for for Python, Java, and other languages. This can be used to write agents or clients. Communications are done using this CLI. They are via a messaging queue technology rather than Rest.

mgbailey (Fri, 15 Dec 2017 15:16:46 GMT):
excuse me, I meant a Rust API, not CLI

calvinx (Fri, 15 Dec 2017 15:16:57 GMT):
I see. I am reading through the rust cli wrapper now.

calvinx (Fri, 15 Dec 2017 15:16:57 GMT):
@mgbailey I see. I am reading through the rust cli wrapper now. re connection to live network, why do I get `Do not have information to connect to live` ?

calvinx (Fri, 15 Dec 2017 15:17:32 GMT):
re connection to live network, why do I get `Do not have information to connect to live` ?

calvinx (Fri, 15 Dec 2017 15:19:38 GMT):
(right rust API)

calvinx (Fri, 15 Dec 2017 15:19:38 GMT):
(Right, rust API)

calvinx (Fri, 15 Dec 2017 15:23:43 GMT):
any idea why I might be getting the "Do not have information to connect to live" problem? `connect test` works fine. `connect live` gives me this message and connection fails

mgbailey (Fri, 15 Dec 2017 15:23:59 GMT):
There are genesis files that are used to bootstrap clients and and agents with the connectivity information needed to initially contact networks. It sounds like your setup does not have the live file genesis file. It can be found at https://github.com/sovrin-foundation/sovrin/tree/master/sovrin, named "pool_transactions_live_genesis". This should be placed in your .indy-client directory. However it will be of little use to you without credentials that are enabled to put transactions onto the ledger.

mgbailey (Fri, 15 Dec 2017 15:25:07 GMT):
There is also a Sovrin Test Network (STN) that I could get you TRUST_ANCHOR credentials for, if you desire.

calvinx (Fri, 15 Dec 2017 15:25:53 GMT):
O yes, that's what I need. Is there any documentation on how to place `pool_transactions_live_genesis` in my .indy-client directory (if you could get me TRUST_ANCHOR credentials)

mgbailey (Fri, 15 Dec 2017 15:26:03 GMT):
Are you with an organization that is considering becoming a Sovrin Steward?

calvinx (Fri, 15 Dec 2017 15:26:20 GMT):
Yes. But we are a super new and super small team of 3.

calvinx (Fri, 15 Dec 2017 15:26:28 GMT):
:)

mgbailey (Fri, 15 Dec 2017 15:26:43 GMT):
For the STN you will want the pool_transactions_live_genesis file instead.

mgbailey (Fri, 15 Dec 2017 15:27:11 GMT):
oops pool_transactions_sandbox_genesis

calvinx (Fri, 15 Dec 2017 15:27:33 GMT):
right. I was confused there for a second.

calvinx (Fri, 15 Dec 2017 15:28:24 GMT):
STN <=> `pool_transactions_sandbox_genesis` TRUST_ANCHOR <=> `pool_transactions_live_genesis`

calvinx (Fri, 15 Dec 2017 15:28:54 GMT):
So what are the criteria in order to apply or qualify as a Sovrin Steward?

mgbailey (Fri, 15 Dec 2017 15:29:26 GMT):
Trust anchor is the name of a role on any indy ledger that has the ability to put NYMs onto the ledger

mgbailey (Fri, 15 Dec 2017 15:29:43 GMT):
There is info on sovrin.org.

mgbailey (Fri, 15 Dec 2017 15:31:55 GMT):
Ping support@sovrin.org to get trust anchor credentials on the STN

calvinx (Fri, 15 Dec 2017 15:33:04 GMT):
Ok. Thank you @mgbailey Will email support@sovrin.org

swcurran (Fri, 15 Dec 2017 17:16:27 GMT):
@calvinx - it' getting easier to spin up your own network. We did that for our testing. You could take a look at this repo that we have used to spin up a network - https://github.com/nrempel/von-network It's got a web server that allows you to see the nodes, the ledger, the genesis file and let's you create a DID. We use it to connect our own agents to it.

swcurran (Fri, 15 Dec 2017 17:16:56 GMT):
We have an instance running on Digital Ocean - http://138.197.170.136/

swcurran (Fri, 15 Dec 2017 17:20:17 GMT):
Nick has our other related repos - von-agent and von-connector. These are currently tightly coupled to our use current use case, but the von-agent code might be generally useful. Here's a background on our starting point - https://bcgov.github.io/TheOrgBook/

binarycrayon (Sat, 16 Dec 2017 00:37:25 GMT):
Has joined the channel.

calvinx (Sun, 17 Dec 2017 14:56:12 GMT):
@swcurran looks great! Thanks for your example and `TheOrgBook`. Do you know how I can connect to the `live` network? Apparently, I need the genesis files to do so?

mountbranch (Sun, 17 Dec 2017 18:52:40 GMT):
Hi! Does anyone have any info on scalability of Indy? I've read this: https://home.kpmg.com/content/dam/kpmg/pdf/2016/06/appendix3-kpmg-blockchain-paper.pdf and this: https://github.com/WebOfTrustInfo/ID2020DesignWorkshop/blob/master/topics-and-advance-readings/scaling-a-bft-consensus-protocol-for-identity.md, both of which describe theoretical scalability to some degree. Does anyone know of experimental values?

mountbranch (Sun, 17 Dec 2017 22:11:34 GMT):
Anyway, I found some details on performance of verifying ECDSA's and a quote from guys of Evernym. I was researching for a research proposal that I submitted at Lykke Streams. It's about Voting using blockchain and I wrote up my understanding ofhow it could be done using Indy. If you're interested I have a look here: streams.lykke.com/Project/ProjectDetails/blockchain-voting

marc0o (Mon, 18 Dec 2017 00:30:04 GMT):
Has joined the channel.

calvinx (Mon, 18 Dec 2017 06:29:11 GMT):
I am still confused about how to connect to the live Sovrin network. Is there any documentation on how a client can create a DID on Sovrin (live)?

Sean_Bohan (Mon, 18 Dec 2017 14:27:43 GMT):
calvinx - send a note to support@sovrin.org to get trust anchor credentials on the STN

Harmannz (Mon, 18 Dec 2017 23:56:53 GMT):
Has joined the channel.

hanumankumar (Tue, 19 Dec 2017 06:18:35 GMT):
Has joined the channel.

hanumankumar (Tue, 19 Dec 2017 06:27:35 GMT):
hi how to create hyperledger sawtooth lake instance on aws?

hanumankumar (Tue, 19 Dec 2017 06:28:05 GMT):
1:45 AM hi how to create hyperledger indy instance on aws?

kunxian (Tue, 19 Dec 2017 06:38:44 GMT):
Has joined the channel.

C0rnelius (Tue, 19 Dec 2017 14:06:47 GMT):
Has joined the channel.

nuxibyte_old (Tue, 19 Dec 2017 15:09:34 GMT):
Hi, I created this gist to help get a 4 node cluster up and running to go through the walkthrough. https://gist.github.com/nuxibyte/208326d6f7910a4fe2089fa515855411

nuxibyte_old (Tue, 19 Dec 2017 15:11:48 GMT):
I wanted something simple that didn't involve having to install and configure dev dependencies for python etc.

WadeBarnes (Tue, 19 Dec 2017 15:23:24 GMT):
@nuxibyte_old, We have something similar here; https://github.com/bcgov/von-network Which adds a bit of a ledger browser. Running instance can be found here; http://138.197.170.136/

WadeBarnes (Tue, 19 Dec 2017 15:23:55 GMT):
All in Docker/DOcker-Compose and sripted.

WadeBarnes (Tue, 19 Dec 2017 15:23:55 GMT):
All in Docker/DOcker-Compose and scripted.

nuxibyte_old (Tue, 19 Dec 2017 15:27:26 GMT):
@WadeBarnes great. only just saw that.

nuxibyte_old (Tue, 19 Dec 2017 15:30:15 GMT):
@WadeBarnes I spent the weekend hitting many dead-ends and feeling like I was wading through treacle so wanted to share where I ended up.

WadeBarnes (Tue, 19 Dec 2017 15:31:38 GMT):
Thanks awesome! We've been down the same road and we'd like to share everything we've learned with as many people as possible.

WadeBarnes (Tue, 19 Dec 2017 15:31:38 GMT):
Thanks, awesome! We've been down the same road and we'd like to share everything we've learned with as many people as possible.

WadeBarnes (Tue, 19 Dec 2017 15:34:10 GMT):
There are a lot of people in the same boat. It would be nice if someone had the time to accumulate all of the great work and document links to it all.

nuxibyte_old (Tue, 19 Dec 2017 15:36:09 GMT):
I agree. I think there are a several 'similar' pages with instructions that have instructions that say 'go here and do this first' but some of the steps on those other pages are not entirely necessary. I am sure it will come together though.

Sean_Bohan (Tue, 19 Dec 2017 17:39:21 GMT):
@WadeBarnes & @nuxibyte_old - we are working on streamlining and upgrading the documentation over the holidays

PabloBascunana (Wed, 20 Dec 2017 08:48:42 GMT):
Has joined the channel.

the_identity_guy (Wed, 20 Dec 2017 14:53:35 GMT):
Hi, is it possible to access sovrin.slack.com without an evernym account?

Sean_Bohan (Wed, 20 Dec 2017 16:41:38 GMT):
@the_identity_guy check out this page for signup https://sovrin-slack-signup.herokuapp.com/

the_identity_guy (Wed, 20 Dec 2017 16:47:53 GMT):
thanks @Sean_Bohan

adamtilton (Wed, 20 Dec 2017 21:00:43 GMT):
Has joined the channel.

calvinx (Thu, 21 Dec 2017 08:08:59 GMT):
@WadeBarnes @nuxibyte_old so I am trying to understand what a "Trust Anchor" is and what is the difference between a "Trust Anchor" and a "Steward" ?

calvinx (Thu, 21 Dec 2017 08:41:32 GMT):
Actually, never mind the question, Trust Anchor is just an existing Sovrin identity owner who is specifically designated so. @WadeBarnes in the case of your bcgov/von-network example, who is the Trust Anchor?

calvinx (Thu, 21 Dec 2017 09:54:45 GMT):
O, you have a "Trustee1" agent which acts as the trust anchor. I see.

Sean_Bohan (Thu, 21 Dec 2017 15:56:43 GMT):
In 5 min - a Holiday (shortened) Indy Ledger call! https://zoom.us/j/232861185

ashcherbakov (Thu, 21 Dec 2017 16:00:19 GMT):
https://docs.google.com/spreadsheets/d/1-8mv5dokDPd5Laqur7EZtrgxmln0sY5CiJuSsL1Rg9Q/edit#gid=1918048553

Sean_Bohan (Thu, 21 Dec 2017 16:06:03 GMT):
Indy WG call for 12/21 Development Achievements: https://docs.google.com/spreadsheets/d/1-8mv5dokDPd5Laqur7EZtrgxmln0sY5CiJuSsL1Rg9Q/edit#gid=1918048553

swcurran (Thu, 21 Dec 2017 16:26:50 GMT):
Here is the link to the basic BCovrin Network browser: http://138.197.170.136/

rtemprano (Thu, 21 Dec 2017 16:28:34 GMT):
Has joined the channel.

Sean_Bohan (Thu, 21 Dec 2017 18:36:05 GMT):
That looks awesome @swcurran - thank you for sharing it on the call and here on Rocketchat

Sean_Bohan (Thu, 21 Dec 2017 18:36:35 GMT):
INDY FRIENDS HAPPY HOLIDAYS DON'T FORGET TO JOIN THE MAILING LIST https://lists.hyperledger.org/mailman/listinfo/hyperledger-indy

mgbailey (Thu, 21 Dec 2017 19:23:02 GMT):
@calvinx, This document has an appendix that explains roles, including Trust Anchor: https://docs.google.com/document/d/1kQcdICAYmSqbk4d5lUtFhk2L55dKJMfyM0JbZPeG55s/edit?usp=sharing

calvinx (Fri, 22 Dec 2017 04:08:02 GMT):
@mgbailey thanks.

calvinx (Fri, 22 Dec 2017 05:38:15 GMT):
@WadeBarnes in your von-network and associated example repositories, is there a provisioning example for a new client (identity owner) who wants to create an identity for the first time?

calvinx (Fri, 22 Dec 2017 05:38:15 GMT):
@WadeBarnes in your von-network and associated example repositories, is there a provisioning example for a new client (identity owner) who wants to create an identity for the first time? I suppose this is the provisioning example? https://github.com/nrempel/von-network/blob/master/server/server.py#L223 I am trying to reconcile that with what is mentioned in "First-Time Provisioning" section on page 14 of the "White Paper: The Technical Foundations of Sovirn".

wolili (Fri, 22 Dec 2017 11:26:46 GMT):
Has joined the channel.

wolili (Fri, 22 Dec 2017 11:47:13 GMT):
Hi, since there is no indy-plenum chat, I hope this is the correct place to address this problem. I would like to get a deeper understanding of the Plenum protocol and tried to execute the `indy-plenum/tutorial/tutorial.py` script. But it seems to exit with `AttributeError: 'Node' object has no attribute 'startKeySharing'`. Is there something wrong with my virtualenv setup? Is the file deprecated? Is there a way to have an instance of Plenum with several nodes running?

ashcherbakov (Fri, 22 Dec 2017 12:02:35 GMT):
@wolili Please have a look at indy-node project and docs there (in particlar, docker images in https://github.com/evernym/sovrin-environments). As of now, indy-node is actually Plenum with some custom txns support. We don't have specific environment to start just Plenum (not indy-node) pool. As of `tutorial.py` looks like it's brokem now, thanks for noticing it. Can you please log a ticket in Hyperledger INDY Jira, so we can take care and fix it?

wolili (Fri, 22 Dec 2017 12:16:10 GMT):
@ashcherbakov okay i will take a look into the docker images, thank you

WadeBarnes (Fri, 22 Dec 2017 15:24:34 GMT):
@calvinx, @nbrempel may be able to answer your question better than I.

agawish (Sun, 24 Dec 2017 20:37:57 GMT):
Has joined the channel.

agawish (Sun, 24 Dec 2017 22:51:21 GMT):
Hey everyone, does anyone have an example of a GDPR complaint project done with Indy?

davux (Tue, 26 Dec 2017 19:34:14 GMT):
Has joined the channel.

davux (Tue, 26 Dec 2017 23:09:44 GMT):
hi! I'm confused about the link (or absence thereof) between the different HL projects. For instance, is Indy based on Fabric?

davux (Tue, 26 Dec 2017 23:10:10 GMT):
(my understanding is that the low-level internode distributed ledger is Fabric)

tkuhrt (Tue, 26 Dec 2017 23:28:39 GMT):
@davux : Hyperledger Indy currently has its own distributed ledger that is not based on Fabric. You can see more about Indy here: https://github.com/hyperledger/indy-node#indy-node-repository-structure. Currently the different HL projects are mostly standalone with only some initial interaction between projects (like SETH, the work of Sawtooth and Burrow communities coming together to bring EVM to Sawtooth). I would expect more of this sort of cross-project collaboration to happen in the 2018 timeframe. I will let the Hyperledger Indy community discuss further any plans they might have to work with some of the other HL ledgers/frameworks.

agawish (Tue, 26 Dec 2017 23:49:17 GMT):
Hi @tkuhrt it would be interesting to know if there are any projects (use cases) that are running indy / sovrin on production and how they utilise it to work with their other Enterprise software

davux (Tue, 26 Dec 2017 23:49:50 GMT):
thanks @tkuhrt

SaiChaitanya (Wed, 27 Dec 2017 16:29:51 GMT):
Has joined the channel.

Sean_Bohan (Wed, 27 Dec 2017 17:11:37 GMT):
Hi @agawish - check out what the BC.gov team is doing with the Org Book (https://bcgov.github.io/TheOrgBook/), there are a number of companies and orgs working with Sovrin (ATB in Canada, MyData in Finland, Illinois Blockchain Initiative). I am working on a reboot of the Indy documentation and it will include a section for Use Cases. Also take a look at Sovrin.org. I will post here when ready

agawish (Wed, 27 Dec 2017 17:26:26 GMT):
Thanks @Sean_Bohan, we are currently having a talk with UK HMRC and they seems to be interested but skeptical about the use case uses, I believe seeing other governments could speed things up for them, also GDPR compliance is one of the key reasons as well

Sean_Bohan (Thu, 28 Dec 2017 14:52:08 GMT):
No INDY WG call today!@

rohanagarwal (Fri, 29 Dec 2017 03:21:21 GMT):
Has joined the channel.

charyorde (Fri, 29 Dec 2017 11:50:35 GMT):
Has joined the channel.

charyorde (Fri, 29 Dec 2017 11:51:28 GMT):
Hello everyone, a quick question:

charyorde (Fri, 29 Dec 2017 11:51:37 GMT):
As it stands today, Can identities running on Indy be referenced in Fabric blockchain (like in a smart contract)?

charyorde (Fri, 29 Dec 2017 12:34:43 GMT):
Also, is it possible to use Indy as bitcoin and Ether Wallet? I want to maintain Bitcoin Wallet and Ether Wallet as Indy identities.

chadgates (Fri, 29 Dec 2017 16:47:16 GMT):
Has joined the channel.

swcurran (Fri, 29 Dec 2017 18:14:11 GMT):
If you are looking for working code of either example, I'm pretty sure the answer is no. I'm not aware of either. The first is presumably possible, since a reference to an Indy ID is a reference to a DID, which is an extended instance of a URN (see the DID spec - https://w3c-ccg.github.io/did-spec/). So to be used in a Fabric smart contract, there would need to be in Fabric the ability to reference a DID (easy - it's a string) and reference the related identity - trickier - must interact with an instance of HL-Indy, such as Sovrin. It's not really an Indy thing, although presumably as this becomes real, there will be some Indy effort put forth. The second I'd have to think about. Not sure about how an Indy key could be related to a Bitcoin wallet. It's an interesting use case.

charyorde (Fri, 29 Dec 2017 18:35:36 GMT):
Thanks @swcurran

charyorde (Fri, 29 Dec 2017 18:36:38 GMT):
Regarding the second, please share what you find. I'm doing some experiments as well. but nothing concrete yet.

charyorde (Fri, 29 Dec 2017 18:37:40 GMT):
I'm thinking of a "semi-hybrid" blockchain implementation.

charyorde (Fri, 29 Dec 2017 18:50:18 GMT):
Okay. i've been able to spin up agent01, 02 and 03

charyorde (Fri, 29 Dec 2017 18:51:20 GMT):
Now trying to add one of the agent trust anchor to the ledger but I'm getting this error:

charyorde (Fri, 29 Dec 2017 18:51:33 GMT):
`indy@sandbox> new key with seed Faber000000000000000000000000000 Key created in wallet Default-23a020 DID for key is ULtgFQJe6bjiFbs7ke3NJD Verification key is ~5kh3FB4H3NKq7tUDqeqHc1 Current DID set to ULtgFQJe6bjiFbs7ke3NJD indy@sandbox> send NYM dest=ULtgFQJe6bjiFbs7ke3NJD role=TRUST_ANCHOR verkey=~5kh3FB4H3NKq7tUDqeqHc1 Adding nym ULtgFQJe6bjiFbs7ke3NJD Error: client request invalid: CouldNotAuthenticate('Can not find verkey for DID ULtgFQJe6bjiFbs7ke3NJD',)`

charyorde (Fri, 29 Dec 2017 18:51:33 GMT):
```indy@sandbox> new key with seed Faber000000000000000000000000000 Key created in wallet Default-23a020 DID for key is ULtgFQJe6bjiFbs7ke3NJD Verification key is ~5kh3FB4H3NKq7tUDqeqHc1 Current DID set to ULtgFQJe6bjiFbs7ke3NJD indy@sandbox> send NYM dest=ULtgFQJe6bjiFbs7ke3NJD role=TRUST_ANCHOR verkey=~5kh3FB4H3NKq7tUDqeqHc1 Adding nym ULtgFQJe6bjiFbs7ke3NJD Error: client request invalid: CouldNotAuthenticate('Can not find verkey for DID ULtgFQJe6bjiFbs7ke3NJD',)```

charyorde (Fri, 29 Dec 2017 18:52:22 GMT):
I can see the 3 agents as VMs in virtualbox

charyorde (Fri, 29 Dec 2017 18:56:22 GMT):
@swcurran An Indy key could be related to a Bitcoin Wallet because an Indy Wallet can hold record isn't it. So how about having BitcoinWallet and EtherWallet objects with attributes peculiar to Bitcoin and Ethereum?

charyorde (Fri, 29 Dec 2017 19:54:03 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=d2MB8P6oN5cir8zS6) Nevermind. Forgot to load sample/faber-request.indy

charyorde (Fri, 29 Dec 2017 20:44:41 GMT):
Sync "Fabric College" is not doing anything: ``` indy@sandbox> sync "Faber College" Synchronizing... ```

charyorde (Fri, 29 Dec 2017 20:45:09 GMT):
What logs is available to me to view please?

charyorde (Sat, 30 Dec 2017 06:29:14 GMT):
@ashcherbakov @LoveshHarchandani Can you point me to the PR that adds plugin support for payments please?

jbool24 (Sat, 30 Dec 2017 23:01:02 GMT):
Has joined the channel.

supernus (Mon, 01 Jan 2018 12:41:57 GMT):
Has joined the channel.

MohdShadab (Mon, 01 Jan 2018 18:17:08 GMT):
Has joined the channel.

MohdShadab (Mon, 01 Jan 2018 18:17:37 GMT):
is there an architecture document for Indy?

mgbailey (Tue, 02 Jan 2018 16:31:16 GMT):
@charyorde Are you following the getting started guide at https://github.com/hyperledger/indy-node/blob/master/environment/vagrant/sandbox/DevelopmentEnvironment/Virtualbox/Vagrantfile? In what you posted, you appear to be attempting to use Faber as a steward, when it is not. It is intended to be a trust anchor.

mgbailey (Tue, 02 Jan 2018 16:32:31 GMT):
Excuese me, the correct link is https://github.com/hyperledger/indy-node/blob/master/environment/vagrant/training/vb-multi-vm/TestIndyClusterSetup.md

charyorde (Tue, 02 Jan 2018 16:32:56 GMT):
@mgbailey How do I make it a trust anchor?

mgbailey (Tue, 02 Jan 2018 16:33:14 GMT):
Follow the instructions in the above link

charyorde (Tue, 02 Jan 2018 16:33:51 GMT):
I think I resolved that bi

charyorde (Tue, 02 Jan 2018 16:33:51 GMT):
I think I resolved that bit

charyorde (Tue, 02 Jan 2018 16:34:16 GMT):
I was stuck with synchronizing...

charyorde (Tue, 02 Jan 2018 16:36:39 GMT):
It was failing to synchronize and there's no log I know about to check what's happening.

charyorde (Tue, 02 Jan 2018 16:37:55 GMT):
The `CouldNotAuthenticate` error was because I hadn't sent the connection request for Faber earlier. I did that then was stuck at "sync Faber"

mgbailey (Tue, 02 Jan 2018 16:39:35 GMT):
The log for the CLI is at .indy-cli/cli.log.

ronniemh (Tue, 02 Jan 2018 17:43:37 GMT):
good afternoon, can you please explain how to use the Indy SDK? and how to implement it please, if I must first use for example this implementation of Docker https://github.com/hyperledger/indy-node/blob/master/environment/docker/pool/README.md and then use the SDK or are they projects that work without depending on each other? since I want to implement a JAVA EE application to interact with the user.

ronniemh (Tue, 02 Jan 2018 17:43:37 GMT):
good afternoon, can you please explain how to use the Indy SDK? and how to implement it please, if I must first use for example this implementation of Docker https://github.com/hyperledger/indy-node/blob/master/environment/docker/pool/README.md and then use the SDK or are they projects that work without depending on each other? since I want to implement a JAVA EE application to interact with the user.another question, How and in what type of database does the information keep?

ronniemh (Tue, 02 Jan 2018 17:43:37 GMT):
good afternoon, can you please explain how to use the Indy SDK? and how to implement it please, if I must first use for example this implementation of Docker https://github.com/hyperledger/indy-node/blob/master/environment/docker/pool/README.md and then use the SDK or are they projects that work without depending on each other? since I want to implement a JAVA EE application to interact with the user. Another question, How and in what type of database does the information keep?

elias_p (Tue, 02 Jan 2018 22:55:32 GMT):
Has joined the channel.

mountbranch (Wed, 03 Jan 2018 09:31:22 GMT):
Hi everyone! I drew up a quick concept for how voting could be done using indy and submitted it to a contest by Lykke. I'd be happy to get your feedback and thoughts on it!

mountbranch (Wed, 03 Jan 2018 09:31:52 GMT):

trustless-scalable-censorship-Jonatan-Bergquist.pdf

mountbranch (Wed, 03 Jan 2018 09:32:50 GMT):
And here is the contest (now closed) https://streams.lykke.com/Project/ProjectDetails/blockchain-voting

Infitude (Thu, 04 Jan 2018 03:34:07 GMT):
Has joined the channel.

Sean_Bohan (Thu, 04 Jan 2018 14:36:28 GMT):
please note there is no Indy call today! We will reconvene next Thursday 1/11/18

smithbk (Thu, 04 Jan 2018 14:38:21 GMT):
What are recommended puml viewers? I can't seem to get the chrome plugin to work and https://www.planttext.com doesn't display all flows.

mgbailey (Thu, 04 Jan 2018 16:36:29 GMT):
@smithbk I use http://www.plantuml.com/plantuml/uml. For larger diagrams I click on "View as SVG" to see it all.

smithbk (Thu, 04 Jan 2018 17:39:45 GMT):
@mgbailey Thanks

darrell.odonnell (Thu, 04 Jan 2018 21:16:45 GMT):
@smithbk I use the PlantUML viewer that embeds into WebStorm (JetBrains IDE - works in all of them). It works very well.

smithbk (Thu, 04 Jan 2018 22:35:48 GMT):
@darrell.odonnell Thanks

balmeida (Fri, 05 Jan 2018 11:35:51 GMT):
Hi guys! Happy new year for all of you!

balmeida (Fri, 05 Jan 2018 11:36:23 GMT):
The Indy demo will be on the 11th?

tkuhrt (Fri, 05 Jan 2018 18:13:24 GMT):
@balmeida : The Hyperledger Indy community hosts developer calls every Thursday at 9:00 am US Mountain Time (the international time can fluctuate based on daylight savings time both in Utah and your local time). Meetings alternate between our Agent Working Group which concerns itself with the off-ledger interactions using Sovrin Identities, and the Ledger or Indy-Node Working Group which talks about the ledger itself. You can find out more details at the community calendar : https://calendar.google.com/calendar/embed?src=linuxfoundation.org_nf9u64g9k9rvd9f8vp4vur23b0%40group.calendar.google.com&ctz=UTC

the_identity_guy (Mon, 08 Jan 2018 18:39:21 GMT):
Hi, I have a question about how proof responses are generated (ex. Alice responding Acme Corp's proof request).. is the proof response also a verifiable claim? and how does Alice pick and choose the claims she wants to share (or derivative of the claims ex. age above 19) with Acme corp without invalidating the signature that initially came with the verifiable claim from Faber college?

ashcherbakov (Tue, 09 Jan 2018 13:34:58 GMT):
@the_identity_guy The proof request response is a zero-knowledger proof. This is in fact an aggregated proof created for all claims needed to fulfill the proof request. There is a call in indy to get all claims matching given proof request (so it can be shown on UI for example). So, Alice can select whta claims she wants to use for proof creation and what data she wants to disclose.

ashcherbakov (Tue, 09 Jan 2018 13:35:28 GMT):
You can have a look at https://github.com/hyperledger/indy-sdk/blob/master/doc/libindy-anoncreds.svg

ashcherbakov (Tue, 09 Jan 2018 13:36:15 GMT):
@ronniemh Please have a look at https://github.com/hyperledger/indy-sdk It already has Java wrapper support. It also contains docs about how to use SDK

Sean_Bohan (Tue, 09 Jan 2018 17:05:39 GMT):
THIS WEEK we have the Indy Agent WG Call - Thursday January 11, 2018 Agenda: A. Housekeeping (Nathan and Sean) B. Development Achievements (Alex and Slava) C. Evolving the Getting Started Guide (Nathan and Slava) D. Demo/Discussion with IBM D. Wrap Up Please note: This meeting will be recored Bring a friend. It will be fun. Call details: Hyperledger Indy Agent (SDK) WG Call When: January 11, 2018 8amPT, 9amMT, 11amET, 4pm BST Where: Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/232861185 Or iPhone one-tap (US Toll): +16465588656,232861185# or +14086380968,232861185# Or Telephone: Dial: +1 646 558 8656 (US Toll) or +1 408 638 0968 (US Toll) Meeting ID: 232 861 185 International numbers available: https://zoom.us/zoomconference?m=a0jD_rTMnh0ZYGQDOKPCNrK_0dP7WPfp1 Hyperledger Indy Community Indy Rocketchat: https://chat.hyperledger.org/channel/indy https://chat.hyperledger.org/channel/indy-agent https://chat.hyperledger.org/channel/indy-node Indy Wiki: https://wiki.hyperledger.org/projects/indy Indy WG Agendas: https://drive.google.com/open?id=1wNnp1pPS6-1Y4B2oygfI6tOU4eHGmpRH Indy Videos: https://drive.google.com/open?id=1Z8-jR7hnXb57fufE0OXxIfeUn05zNmRt Indy WG Meeting Recordings: https://drive.google.com/open?id=0B_NJV6eJXAA1UlJZMXd3cm1zNDQ

gbarcomu (Wed, 10 Jan 2018 11:46:07 GMT):
Has joined the channel.

gbarcomu (Wed, 10 Jan 2018 11:51:14 GMT):
Hi, is there any API to interact with Hyperledger Indy?

sklump (Wed, 10 Jan 2018 13:10:14 GMT):
https://github.com/hyperledger/indy-sdk

jsm (Wed, 10 Jan 2018 13:21:21 GMT):
Has joined the channel.

cchalc (Wed, 10 Jan 2018 17:48:08 GMT):
Has joined the channel.

Sean_Bohan (Wed, 10 Jan 2018 20:30:15 GMT):
thanks @sklump !

Sean_Bohan (Wed, 10 Jan 2018 20:32:49 GMT):
Indy folks. The Indy Glossary (still work in progress) can be found here: https://wiki.hyperledger.org/projects/indy_glossary

tkuhrt (Wed, 10 Jan 2018 22:09:56 GMT):
@Sean_Bohan : Can we move that link so that it is under the projects:indy namespace, not under the projects namespace?

xixuejia (Thu, 11 Jan 2018 01:47:57 GMT):
Has joined the channel.

Sean_Bohan (Thu, 11 Jan 2018 04:03:20 GMT):
We can and that is my mistake

JeroenDePrest (Thu, 11 Jan 2018 09:46:04 GMT):
Has joined the channel.

JeroenDePrest (Thu, 11 Jan 2018 09:48:02 GMT):
anyone here have some good resources for understanding Decentralized Identifiers (DIDs) and Self-Sovereign Identity

Sean_Bohan (Thu, 11 Jan 2018 15:36:55 GMT):
there is a W3C Working Group and Decentralized Identity Foundation: https://www.w3.org/2017/vc/WG/ http://identity.foundation/ We will also have an explainer video out in a couple of days

ashcherbakov (Thu, 11 Jan 2018 16:15:21 GMT):
FYI: Environment (docker, vagrant, etc.) scripts were moved from sovrin-environment to https://github.com/hyperledger/indy-node/tree/master/environment

jljordan_bcgov (Thu, 11 Jan 2018 16:18:59 GMT):
This is a great setting started helper by the folks at IBM https://github.com/brycecurtis/indy-tutorial-sandbox

lehors (Thu, 11 Jan 2018 16:31:14 GMT):
@JeroenDePrest to clarify, DID is not part of the W3C Verifiable Claims WG, it is part of the Credential Community Group - difference is in level of endorsement by W3C (less)

lehors (Thu, 11 Jan 2018 16:31:28 GMT):
https://w3c-ccg.github.io/did-spec/

lehors (Thu, 11 Jan 2018 16:32:48 GMT):
https://www.w3.org/community/credentials/

nage (Thu, 11 Jan 2018 17:02:10 GMT):
More information about Estonia's voting scheme here https://drive.google.com/file/d/0B_luqCyRPBXjQlZRMlVPWUtPX3RMal9FZWNHUzNkU0RzN3NF/view (see page 3)

mountbranch (Thu, 11 Jan 2018 17:03:46 GMT):
Thanks! @nage Here is a link to the paper: https://www.dropbox.com/s/mc9004cx3xry6v1/trustless-scalable-censorship-Jonatan-Bergquist.pdf?dl=0

hawkmauk (Thu, 11 Jan 2018 17:15:36 GMT):
Has joined the channel.

Sean_Bohan (Thu, 11 Jan 2018 20:05:32 GMT):
@lehors that is correct

tkuhrt (Thu, 11 Jan 2018 23:53:36 GMT):
Are there any good Intro to Indy slide decks out there?

Derashe (Fri, 12 Jan 2018 06:37:24 GMT):
Has joined the channel.

TiaanOtto (Fri, 12 Jan 2018 08:30:25 GMT):
Has joined the channel.

tyrellperera (Fri, 12 Jan 2018 09:33:12 GMT):
Has joined the channel.

Skoya (Fri, 12 Jan 2018 12:12:33 GMT):
Has joined the channel.

Skoya (Fri, 12 Jan 2018 12:15:56 GMT):
hi all, I am from UBS and am attempting to setup some nodes. I have the inital cluster of validators up and am now attempting to understand the roles and responsiblities of each process. Is there an dummies definition of steward, validator, agent and if possible what they are responsible for... (source.... https://github.com/hyperledger/indy-node/blob/master/environment/vagrant/training/vb-multi-vm/TestIndyClusterSetup.md)

ashcherbakov (Fri, 12 Jan 2018 12:21:55 GMT):
@Skoya Please have a look at https://wiki.hyperledger.org/projects/indy_glossary and https://github.com/hyperledger/indy-node/blob/master/getting-started.md

Skoya (Fri, 12 Jan 2018 12:28:36 GMT):
thanks... is there also a newer or alternate version of this diagram to help understand the comms flow? https://sovrin.discoursehosting.net/uploads/db4005/original/1X/2fb95c87709c9170e3fbe506d6ef2d50c92756f1.jpg

Skoya (Fri, 12 Jan 2018 12:28:52 GMT):
looking for something that shows the messages being passed

Skoya (Fri, 12 Jan 2018 12:29:07 GMT):
i.e. one level lower

ashcherbakov (Fri, 12 Jan 2018 13:20:08 GMT):
@Skoya If you are interested in Verifiable Clams (anoncreds) workflow, you can have a look at https://github.com/hyperledger/indy-sdk/blob/master/doc/libindy-anoncreds.svg

Skoya (Fri, 12 Jan 2018 13:21:03 GMT):
great, much appreciated

ashcherbakov (Fri, 12 Jan 2018 13:21:30 GMT):
We're working on documentation improvements. You can find documentation on Hyperledger Indy Wiki as well as in `docs` folder in all repos: - https://wiki.hyperledger.org/projects/indy_documentation - https://github.com/hyperledger/indy-plenum/tree/master/docs - https://github.com/hyperledger/indy-node/tree/master/docs - https://github.com/hyperledger/indy-sdk/tree/master/doc

Skoya (Fri, 12 Jan 2018 13:25:07 GMT):
understood thanks again

Skoya (Fri, 12 Jan 2018 13:25:57 GMT):
sequence flow diagrams would be fantastic inside the docs

ashcherbakov (Fri, 12 Jan 2018 15:16:24 GMT):
agree. there are already some sequence diagrams there. We're working on adding more. BTW it could be a good excersice for anyone who wants to learn Indy to create a diagram (if there is no any) as a result of learning and send a PR.

Skoya (Sat, 13 Jan 2018 12:49:37 GMT):
I would have to double check that I am allowed to commit PRs. If not I can send something via email to you

the_identity_guy (Mon, 15 Jan 2018 16:06:41 GMT):
Is the 'did' keyword omitted from the DIDs stored on the ledger to improve efficiency?

jjensenevernym (Mon, 15 Jan 2018 19:20:04 GMT):
Has joined the channel.

rajivgandhi2010 (Tue, 16 Jan 2018 07:01:58 GMT):
Has joined the channel.

rajivgandhi2010 (Tue, 16 Jan 2018 07:02:07 GMT):
Hi

rajivgandhi2010 (Tue, 16 Jan 2018 07:02:51 GMT):
Am new to Hyperledger Indy.. is it possible to install and develop Indy applications in windows environment?

ashcherbakov (Tue, 16 Jan 2018 09:06:06 GMT):
Hi @rajivgandhi2010 You can install and work on Windows with https://github.com/hyperledger/indy-sdk (the API and SDK for applications dealing with Indy). So, you can developer applications based on Indy on Windows. As for the Indy Pool and Ledger (https://github.com/hyperledger/indy-node), it works on Linux only as of now. So, if you want to contribute to the Indy Pool (blockchain level), you have to do it on Linux.

WadeBarnes (Tue, 16 Jan 2018 16:09:05 GMT):
Hey @ashcherbakov, We have some projects you may want to look at. I develop on Windows and we have an indy agent implementation that works on Windows using python. Some info about libindy on Windows here; https://github.com/WadeBarnes/TheOrgBook/tree/master/tob-api#libindy-and-the-von-agent Projects: https://github.com/bcgov/TheOrgBook https://github.com/bcgov/von-connector https://github.com/bcgov/von-network https://github.com/bcgov/permitify Running instance of our development pool complete with ledger browser; http://138.197.170.136/ Running instance(s) of TheOrgBook and Permitify services can be accessed here; https://devex-von-dev.pathfinder.gov.bc.ca/roadmap - Here you can register a business and apply for various claims, complete with verified proofs along the way.

WadeBarnes (Tue, 16 Jan 2018 16:09:05 GMT):
Hey @ashcherbakov, We have some projects you may want to look at. I develop on Windows and we have an indy agent implementation that works on Windows using python. Some info about libindy on Windows here; https://github.com/WadeBarnes/TheOrgBook/tree/master/tob-api#libindy-and-the-von-agent Projects: https://github.com/bcgov/TheOrgBook https://github.com/bcgov/von-connector https://github.com/bcgov/von-network https://github.com/bcgov/permitify Running instance of our development pool complete with ledger browser; http://138.197.170.136/ Running instance(s) of TheOrgBook and Permitify services can be accessed here; https://devex-von-dev.pathfinder.gov.bc.ca/roadmap - Here you can register a business and apply for various claims, complete with verified proofs along the way. All provisional development implementations, so no fear something real will get messed up.

WadeBarnes (Tue, 16 Jan 2018 16:09:05 GMT):
Hey @ashcherbakov, @rajivgandhi2010, We have some projects you may want to look at. I develop on Windows and we have an indy agent implementation that works on Windows using python. Some info about libindy on Windows here; https://github.com/WadeBarnes/TheOrgBook/tree/master/tob-api#libindy-and-the-von-agent Projects: https://github.com/bcgov/TheOrgBook https://github.com/bcgov/von-connector https://github.com/bcgov/von-network https://github.com/bcgov/permitify Running instance of our development pool complete with ledger browser; http://138.197.170.136/ Running instance(s) of TheOrgBook and Permitify services can be accessed here; https://devex-von-dev.pathfinder.gov.bc.ca/roadmap - Here you can register a business and apply for various claims, complete with verified proofs along the way. All provisional development implementations, so no fear something real will get messed up.

WadeBarnes (Tue, 16 Jan 2018 16:11:34 GMT):
Most of the time we're working in containers (Docker/Docker Compose/OpenShift) so the platform does not matter.

nij458 (Tue, 16 Jan 2018 17:34:21 GMT):
Has joined the channel.

Rickzan (Tue, 16 Jan 2018 19:23:13 GMT):
Has joined the channel.

numical (Wed, 17 Jan 2018 09:44:17 GMT):
Has joined the channel.

nij458 (Wed, 17 Jan 2018 12:16:31 GMT):
Hi all, I am new to Indy and I am trying to build the indy-sdk on my MacOS using the guide (https://github.com/hyperledger/indy-sdk/blob/master/doc/mac-build.md). I have followed the guide without issue up until running "cargo build". The guide (https://github.com/hyperledger/indy-sdk/blob/master/wrappers/java/README.md) says that I should expect to see a file "libindy.so" after "cargo build" finishes executing and move it to "./lib/", "cargo build" runs successfully yet I cannot find the file "libindy.so" anywhere in the directory. Have you come across this issue before? Is there a step I am missing that isn't in the guides? Thanks, Nij

brycecurtis (Wed, 17 Jan 2018 14:47:09 GMT):
slack

sklump (Wed, 17 Jan 2018 17:48:34 GMT):
Has left the channel.

Sean_Bohan (Wed, 17 Jan 2018 19:19:57 GMT):
Nij - will check with the guys

rajivgandhi2010 (Thu, 18 Jan 2018 05:22:01 GMT):
Can anyone help me to resolve the below issue bc_registries_1 | ---------------------------------------------------------------------------------- bc_registries_1 | No command line parameters were provided to the entry point. bc_registries_1 | Using the values specified for the environment, or defaults if none are provided. bc_registries_1 | bc_registries_1 | TEMPLATE_NAME: bc_registries bc_registries_1 | APPLICATION_IP: 0.0.0.0 bc_registries_1 | APPLICATION_PORT: 8080 bc_registries_1 | ---------------------------------------------------------------------------------- bc_registries_1 | ================================================================================ bc_registries_1 | Using template: bc_registries bc_registries_1 | Cmd: python manage.py runserver 0.0.0.0:8080 bc_registries_1 | ================================================================================ bc_registries_1 | Directory /app/site_templates/bc_registries doesn't exist. docker_bc_registries_1 exited with code 1

wolili (Thu, 18 Jan 2018 12:52:09 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=d5FhDXT8Qan9WhNLH) @ashcherbakov as recommended I created a sequence diagram regarding the parties involved in the creation of a new Identity. Since I'm new, I would apprciate if someone could take a look at the diagrams and correct me if I missunderstood something. Thanks in advance! PNG: https://github.com/wolilo/share/blob/master/DID_create_request.png PUML: https://github.com/wolilo/share/blob/master/DID_create_request.puml

balmeida (Thu, 18 Jan 2018 15:06:39 GMT):
Hi guys

balmeida (Thu, 18 Jan 2018 15:06:59 GMT):
Can you confirm that the today's meeting will have a demo session?

ashcherbakov (Thu, 18 Jan 2018 15:27:28 GMT):
@wolili Thank you, it looks really good. The only thing I'm not so sure is that we should mention sending request by email (I belive we do not restrict or require only emails). @nage @Sean_Bohan ?

ashcherbakov (Thu, 18 Jan 2018 15:30:21 GMT):
@wolili Can you please create a PR to indy-node repo and put the diagrams into docs folder (https://github.com/hyperledger/indy-node/tree/master/docs)? You can see instructions on how to send a PR: https://github.com/hyperledger/indy-node/blob/master/README.md#how-to-send-a-pr the only thing you should prpbably tak into account is DCO signature required for Hyperledger projects.

Sean_Bohan (Thu, 18 Jan 2018 16:00:27 GMT):
Indy WG Call is happening now - https://zoom.us/j/232861185

nage (Thu, 18 Jan 2018 21:08:02 GMT):
@wolili we talk about key exchange being “out of band” rather than by email. You could use any other means for that step of exchange. It is usually accompanied by some type of identity validation or vetting (call center authentication steps, in person validation, notarizing-like processes, etc. depending on what is at stake for that identity/key).

wolili (Fri, 19 Jan 2018 10:42:48 GMT):
I mentioned the email, because I have seen a few people in the chat asking how to access the Sovrin Network "in production" mode. And the answer given was, to contact the the Sovrin team personally via Email. But probably this not is more confusing than it helps in this context.

PraveenKumar0815 (Fri, 19 Jan 2018 11:59:37 GMT):
Has joined the channel.

PraveenKumar0815 (Fri, 19 Jan 2018 12:01:38 GMT):
Hi Everybody new to the thread I was trying to workout uPort as a self soverign login system for a custom appication but could not suceed with it CAN any one suggest a good demo for indy / How do i get started with it

hawkmauk (Fri, 19 Jan 2018 13:27:55 GMT):
@PraveenKumar0815 Hi, I'm new to the community too but ave found that the wiki is a hub for the documentation that there is on Indy https://wiki.hyperledger.org/projects/indy/documentation. There is a lot of activity around creating documentation at the moment so I'd expect this to get updated on a regular basis

hawkmauk (Fri, 19 Jan 2018 13:27:55 GMT):
@PraveenKumar0815 Hi, I'm new to the community too but have found that the wiki is a hub for the documentation that there is on Indy https://wiki.hyperledger.org/projects/indy/documentation. There is a lot of activity around creating documentation at the moment so I'd expect this to get updated on a regular basis

hawkmauk (Fri, 19 Jan 2018 13:31:36 GMT):
I'd be happy to organise a UK Indy meetup if there is anyone else that would be interested?

hawkmauk (Fri, 19 Jan 2018 13:31:36 GMT):
Hi All I'd be happy to organise a UK Indy meetup if there is anyone else that would be interested?

Sean_Bohan (Fri, 19 Jan 2018 14:14:01 GMT):
hey @PraveenKumar0815 You can check out a couple of things. The Getting Started Guide is getting revised but the current version is here: https://github.com/hyperledger/indy-node/blob/stable/getting-started.md#getting-started-with-indy You can check out Sovrin.org, one of the first instantiations of the Indy Code And look at what the BC.gov crew is up to with the Org Book: https://bcgov.github.io/TheOrgBook/

Sean_Bohan (Fri, 19 Jan 2018 14:14:01 GMT):
hey @PraveenKumar0815 You can check out a couple of things. The Getting Started Guide is getting revised but the current version is here: https://github.com/hyperledger/indy-node/blob/stable/getting-started.md#getting-started-with-indy You can check out Sovrin.org, one of the first instantiations of the Indy Code @hawkmauk an Indy meetup would be awesome (and we have some team members who are local who can join). There is a hyperledger meetup in London too (I need to find the details) And look at what the BC.gov crew is up to with the Org Book: https://bcgov.github.io/TheOrgBook/

Sean_Bohan (Fri, 19 Jan 2018 14:27:40 GMT):
@hawkmauk an Indy meetup would be awesome (and we have some team members who are local who can join). There is a hyperledger meetup in London too (I need to find the details)

Sean_Bohan (Fri, 19 Jan 2018 14:33:27 GMT):
Thanks to all who joined yesterday's Ledger WG call. Here are the links from the call: 1. Agenda: https://docs.google.com/presentation/d/1PCJua6VssoRLkuGCrse9y7E92f0HTvZaqSRd5VDgA8E/edit#slide=id.g2c72e85c7b_0_49 2. Dev Status (note - tabs at bottom for each week's status): https://docs.google.com/spreadsheets/d/1-8mv5dokDPd5Laqur7EZtrgxmln0sY5CiJuSsL1Rg9Q/edit#gid=970515928 3. Indy Wiki: https://wiki.hyperledger.org/projects/indy/ 4. Hyperledger Hackfest LA in Feb: https://www.hyperledger.org/event/hyperledger-hackfest-february-2018 5. Hyperledger Identity WG Call next week: https://wiki.hyperledger.org/groups/identity/identity-wg 6. Indy Mailing List: https://lists.hyperledger.org/mailman/listinfo/hyperledger-indy 7. WG Call Recording: https://drive.google.com/open?id=1ubhdivWsnXfe-GY5-bxDPizG4HTAvdst

Sean_Bohan (Fri, 19 Jan 2018 14:33:27 GMT):
Thanks to all who joined yesterday's Ledger WG call. Here are the links from the call: 1. Agenda: https://docs.google.com/presentation/d/1PCJua6VssoRLkuGCrse9y7E92f0HTvZaqSRd5VDgA8E/edit#slide=id.g2c72e85c7b_0_49 2. Dev Status (note - tabs at bottom for each week's status): https://docs.google.com/spreadsheets/d/1-8mv5dokDPd5Laqur7EZtrgxmln0sY5CiJuSsL1Rg9Q/edit#gid=970515928 3. Indy Wiki: https://wiki.hyperledger.org/projects/indy/ 4. Hyperledger Hackfest LA in Feb: https://www.hyperledger.org/event/hyperledger-hackfest-february-2018 5. Hyperledger Identity WG Call next week: https://wiki.hyperledger.org/groups/identity/identity-wg 6. Indy Mailing List: https://lists.hyperledger.org/mailman/listinfo/hyperledger-indy

Sean_Bohan (Fri, 19 Jan 2018 14:34:09 GMT):
PLEASE - if you have an idea for a topic for next week's WG call or want to demo your work, let me know.

viraj 8 (Sat, 20 Jan 2018 10:00:35 GMT):
Has joined the channel.

mastersingh24 (Mon, 22 Jan 2018 17:00:12 GMT):
Has joined the channel.

bhavneesh (Mon, 22 Jan 2018 17:15:31 GMT):
Has joined the channel.

bhavneesh (Mon, 22 Jan 2018 17:15:35 GMT):
Developers needed; https://paeanproject.org Email me at bhavneesh@vasudacapitalmanagement.com

akhihaki (Mon, 22 Jan 2018 18:12:14 GMT):
Has joined the channel.

rogerso (Tue, 23 Jan 2018 02:34:59 GMT):
Has joined the channel.

tobowers (Tue, 23 Jan 2018 13:39:40 GMT):
Has joined the channel.

tobowers (Tue, 23 Jan 2018 13:41:02 GMT):
hiyo

tobowers (Tue, 23 Jan 2018 13:42:41 GMT):
I'm new to the community. I notice that in some white papers there is a mention of "issuer" but doesn't seem to exist when talking about indy in reality. Do we expect individual persons to create their identities using "personal" software and then use those in various situations or do we think there will be "issuing authorities" that will help users create sovrin identities? Is anyone working on making the project easy to use for devs/users? (mobile apps, etc)

ashcherbakov (Tue, 23 Jan 2018 15:58:32 GMT):
Hi @tobowers Please have a look at this doc: https://github.com/hyperledger/indy-crypto/blob/master/libindy-crypto/docs/anoncreds-design.md and https://wiki.hyperledger.org/projects/indy/glossary to find out what we mean by the "Issuer"

drummondreed (Tue, 23 Jan 2018 15:59:40 GMT):
"Issuers" issue verifiable claims using DIDs (decentralized identifiers) registered on a distributed ledger that supports DIDs (such as Sovrin). Claims can be self-issued or issued by third party issuers, the same way identity works in the real world. I recommend the new Sovrin white paper for a full explanation of what a verifiable claims infrastructure will look like. https://sovrin.org/library/sovrin-protocol-and-token-white-paper

JustinBoyer (Tue, 23 Jan 2018 22:58:50 GMT):
Has joined the channel.

tobowers (Wed, 24 Jan 2018 07:39:14 GMT):
ok thanks. I understand issuers now. So Sovrin expects users to create their own digital identities using a piece of software that is not yet defined, and use that for identity at everything that supports it. Is anyone working on the "create identity" part or a usable oauth/mobile app solution?

tobowers (Wed, 24 Jan 2018 07:45:25 GMT):
I'm very excited about indy. It solves an entire class of problems in my software that I solved in similar (but not as robust) ways. I'm very interested in making indy usable ASAP and contributing where I can. Is this on track? https://drive.google.com/file/d/0ByaxIFopqmUVRkhyRWZ1X0RpQ1U/view I'm also wondering if I could run my own internal indy sooner than a public release... where "could" is defined as "contributors would feel fairly confident". I've joined the mailing list, here, and I'm about to go look at the jira. Is anyone working on indy in Berlin? I will buy you a beer/coffee/acaibowl if we can meetup and chat about this stuff :).

tobowers (Wed, 24 Jan 2018 08:46:04 GMT):
found this: https://docs.google.com/presentation/d/1zbb1-4TjH1-iYN7LnFZyPWvB1w-YcoUwKYCfByugSE0/edit#slide=id.p

zhigunenko.dsr (Wed, 24 Jan 2018 09:19:53 GMT):
Has joined the channel.

tobowers (Wed, 24 Jan 2018 10:01:12 GMT):
hmm... so if this is really decentralized, why do we need Trust Anchors? Shouldn't we allow anyone to create an ID (on the internet, no one knows your a dog)?

tobowers (Wed, 24 Jan 2018 10:21:13 GMT):
oh is that because only the DIDs are on ledger?

tobowers (Wed, 24 Jan 2018 14:38:48 GMT):
looks like trust anchors are a sovrin thing and not an indy thing in terms of joining the network

AlanDickman (Wed, 24 Jan 2018 19:43:20 GMT):
Has joined the channel.

the_identity_guy (Wed, 24 Jan 2018 20:01:24 GMT):
Hi, Is the concept/roles of trustee and steward specific to Sovrin or are they treated as a standard roles within Indy as well

cockayne (Wed, 24 Jan 2018 20:34:45 GMT):
Has joined the channel.

AlanDickman (Wed, 24 Jan 2018 21:19:46 GMT):
Is there a general overview deck on Project Indy? I've been searching for one, but have only found low level documents.

Sean_Bohan (Wed, 24 Jan 2018 21:31:32 GMT):
Alan - I am working on a high level product view of Indy and am trying to get it out this week

AlanDickman (Wed, 24 Jan 2018 21:34:38 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=n2Hzpa7sNgkuW999T) @Sean_Bohan There's some interest within the Hyperledger group in Colorado around the topic of DID so would love to see it when it's available!

Sean_Bohan (Wed, 24 Jan 2018 21:36:42 GMT):
@tobowers - couple answers (sorry I am late with these - buried today) 1. we would love your help 2. The roadmap is being revised and we are planning on sharing tomorrow as a Work IN Progress tomorrow during the WG call (will share call-in details in a moment 3 We will probably be at the Hyperledger Hackfest in Amsterdam in June and can grab a coffee then (if one of us doesn't end up in Germany before then!) 4. the contributor guidelines are being revised (into a whole section called "HOW TO INDY" which is a hub for everything contribution, but that deck is still valid 5.

Sean_Bohan (Wed, 24 Jan 2018 21:37:30 GMT):
@AlanDickman that sounds awesome and we have teammates in SLC who might be interested in stopping by :)

Sean_Bohan (Wed, 24 Jan 2018 21:39:20 GMT):
@the_identity_guy Trustees and Stewards are specific to Sovrin and it's governance (trust framework). If you are building your own Indy you can call the nodes whatever you want.

jtsiros (Wed, 24 Jan 2018 22:50:58 GMT):
Has joined the channel.

jtsiros (Wed, 24 Jan 2018 22:58:17 GMT):
Hello! I need some help with the indy-crypto library: Currently, I'm trying to call the indy-crypto library functions as a C library from Golang. I've created the header files for Golang along with importing the dylib into my project. ``` extern indy_crypto_error_t indy_crypto_bls_sign_key_new(unsigned char* seed, size_t seed_len, const void* sign_key_p); ``` Here is my Go calling code: ``` func NewSigningKey() { var cInstance interface{} ptr := unsafe.Pointer(&cInstance) e := C.indy_crypto_bls_sign_key_new(nil, C.size_t(0), ptr) fmt.Println("new sign key initialized result:", e) C.indy_crypto_bls_generator_free(ptr) } ``` Here are the directives I'm setting at the top: ``` /* #cgo CFLAGS: -I../dyLib #cgo LDFLAGS: -L../dylib -lindy_crypto #include */ import "C" ``` The issue is that I'm getting a malloc error message when attempting to free that pointer that was allocated: ``` new sign key initialized result: 0 didnet(5608,0x7fffaec0d3c0) malloc: *** error for object 0xc4200104f0: pointer being freed was not allocated *** set a breakpoint in malloc_error_break to debug SIGABRT: abort PC=0x7fffa5e1ed42 m=0 sigcode=0 signal arrived during cgo execution goroutine 1 [syscall, locked to thread]: runtime.cgocall(0x40c9670, 0xc420055ba8, 0x4028fd6) /usr/local/Cellar/go/1.9.2/libexec/src/runtime/cgocall.go:132 +0xe4 fp=0xc420055b78 sp=0xc420055b38 pc=0x4003b94 ***/rp/didnet/crypto._Cfunc_indy_crypto_bls_generator_free(0xc4200104f0, 0xc400000000) ```

jtsiros (Wed, 24 Jan 2018 22:59:29 GMT):
Has left the channel.

isi8787 (Thu, 25 Jan 2018 08:12:23 GMT):
Has joined the channel.

tobowers (Thu, 25 Jan 2018 08:43:14 GMT):
@Sean_Bohan thanks for the info! I noticed that Sovrin launched its "provisional" network back in August and promised more info "over the coming weeks." Did something happen to prevent that? Also, I think we're probably time-shifted so no worries about "late" replies... just happy to hear info as I can.

tobowers (Thu, 25 Jan 2018 09:57:49 GMT):
in the DID document on authorization ( https://w3c-ccg.github.io/did-spec/#authorization ) - it mentions that the authorizationCapability field will be removed: "The current consensus is to use object capabilities when possible to express authorization to modify a DID Document." Is ld-ocap what that means? ( https://w3c-ccg.github.io/ld-ocap/ ) and would the minimum signatures part of the existing authorizationCapability be handled as a caveat?

tobowers (Thu, 25 Jan 2018 10:30:53 GMT):
does indy/sovrin currently support the authorizationCapability for updating DID docs and is there already a plan for what it will move to?

tobowers (Thu, 25 Jan 2018 10:36:07 GMT):
evernym seems to claim to have a mobile app, but connect.me is broken and there's nothing on the app store... is it just in private beta?

nage (Thu, 25 Jan 2018 13:41:44 GMT):
@tobowers no, we continue to make steady progress on the technical underpinnings of the network, it’s governance and the business of setting up stewards and onboarding participants. Take a look at the Trust Framework working group over on Sovrin’s slack if you want to see more about what is going on.

nage (Thu, 25 Jan 2018 13:43:25 GMT):
As for DID spec discussions, the right way to follow along there is through the W3C Credentials Community Group. There are active discussions there, some of which we don’t agree with, but our intent is to support the specification that comes out of that process.

tobowers (Thu, 25 Jan 2018 13:43:46 GMT):
ah I didn't know about sovrin's slack

nage (Thu, 25 Jan 2018 13:45:17 GMT):
We try to keep development discussions here, where the devs hang out, and have governance and Sovrin operation discussions there, but obviously things bleed across both ways from time to time.

nage (Thu, 25 Jan 2018 13:45:56 GMT):
https://sovrin-slack-signup.herokuapp.com/

tobowers (Thu, 25 Jan 2018 13:47:52 GMT):
is this the w3c group you mean? https://www.w3.org/community/credentials/

nage (Thu, 25 Jan 2018 15:30:28 GMT):
yes, that is the one.

nage (Thu, 25 Jan 2018 15:31:13 GMT):
@drummondreed is heading up the DID specification efforts if you want to discuss it here or in @idenity-wg (there are a number of folks involved in the spec that follow Hyperledger)

nage (Thu, 25 Jan 2018 15:31:13 GMT):
@drummondreed is heading up the DID specification efforts if you want to discuss it here or in @idenity-wg (there are a number of folks involved in the spec that follow Hyperledger developments)

drummondreed (Thu, 25 Jan 2018 15:32:23 GMT):
Yes, if you are interested in DID, I recommend joining the W3C Credentials Community Group that Nathan linked to above. It's free and not a high volume list.

Sean_Bohan (Thu, 25 Jan 2018 15:46:36 GMT):
THIS WEEK we have the Indy Agent WG Call - Thursday January 25, 2018 Agenda: Housekeeping (Sean) Community Appreciation (Sean) Development Status (Alex and Slava) Roadmap Update/Discussion (Sean) Entity Relationship Diagram Discussion (Alex, Slava, Nathan) Wrap Up (Sean) Please note: This meeting will be recored Bring a friend. It will be fun. Call details: Hyperledger Indy Agent (SDK) WG Call When: January 25, 2018 8amPT, 9amMT, 11amET, 4pm BST Where: Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/232861185 Or iPhone one-tap (US Toll): +16465588656,232861185# or +14086380968,232861185# Or Telephone: Dial: +1 646 558 8656 (US Toll) or +1 408 638 0968 (US Toll) Meeting ID: 232 861 185 International numbers available: https://zoom.us/zoomconference?m=a0jD_rTMnh0ZYGQDOKPCNrK_0dP7WPfp1 Hyperledger Indy Community Indy Rocketchat: https://chat.hyperledger.org/channel/indy https://chat.hyperledger.org/channel/indy-agent https://chat.hyperledger.org/channel/indy-node Indy Wiki: https://wiki.hyperledger.org/projects/indy Indy WG Agendas: https://drive.google.com/open?id=1wNnp1pPS6-1Y4B2oygfI6tOU4eHGmpRH Indy Videos: https://drive.google.com/open?id=1Z8-jR7hnXb57fufE0OXxIfeUn05zNmRt Indy WG Meeting Recordings: https://drive.google.com/open?id=0B_NJV6eJXAA1UlJZMXd3cm1zNDQ25

Sean_Bohan (Thu, 25 Jan 2018 15:46:36 GMT):
THIS WEEK we have the Indy Agent WG Call - Thursday January 25, 2018 Agenda: Housekeeping (Sean) Community Appreciation (Sean) Development Status (Alex and Slava) Roadmap Update/Discussion (Sean) Entity Relationship Diagram Discussion (Alex, Slava, Nathan) Wrap Up (Sean) Please note: This meeting will be recored Bring a friend. It will be fun. Call details: Hyperledger Indy Agent (SDK) WG Call When: January 11, 2018 8amPT, 9amMT, 11amET, 4pm BST Where: Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/232861185 Or iPhone one-tap (US Toll): +16465588656,232861185# or +14086380968,232861185# Or Telephone: Dial: +1 646 558 8656 (US Toll) or +1 408 638 0968 (US Toll) Meeting ID: 232 861 185 International numbers available: https://zoom.us/zoomconference?m=a0jD_rTMnh0ZYGQDOKPCNrK_0dP7WPfp1 Hyperledger Indy Community Indy Rocketchat: https://chat.hyperledger.org/channel/indy https://chat.hyperledger.org/channel/indy-agent https://chat.hyperledger.org/channel/indy-node Indy Wiki: https://wiki.hyperledger.org/projects/indy Indy WG Agendas: https://drive.google.com/open?id=1wNnp1pPS6-1Y4B2oygfI6tOU4eHGmpRH Indy Videos: https://drive.google.com/open?id=1Z8-jR7hnXb57fufE0OXxIfeUn05zNmRt Indy WG Meeting Recordings: https://drive.google.com/open?id=0B_NJV6eJXAA1UlJZMXd3cm1zNDQ

panickervinod (Thu, 25 Jan 2018 16:12:49 GMT):
Thanks Sean, This is excellent!!

panickervinod (Thu, 25 Jan 2018 16:13:53 GMT):
Please do include the sprint tracker link as well

drew 41 (Thu, 25 Jan 2018 16:24:23 GMT):
Has joined the channel.

darrell.odonnell (Thu, 25 Jan 2018 16:37:45 GMT):
Folks - the BC Government has put out 2 bounty projects that should be looked at by the group here: Indy-SDK Corporate Scale Wallet: https://bcdevexchange.org/opportunities/opp-enhancements-to-hyperledger-indy-sdk--wallet--and-implementation-of-a-large-scale-claims-store DID-Auth: https://bcdevexchange.org/opportunities/opp-initial-reference-implementation-of-decentralized-authentication--did-auth--and-authorization-mechanisms

Sean_Bohan (Thu, 25 Jan 2018 16:45:43 GMT):
+1 @darrell.odonnell

Sean_Bohan (Thu, 25 Jan 2018 16:46:13 GMT):
@panickervinod great point - I will add a link to that doc as well

panickervinod (Thu, 25 Jan 2018 16:47:13 GMT):
@Sean_Bohan thank you.

tobowers (Thu, 25 Jan 2018 19:11:28 GMT):
thanks guys! i've joined all the groups I think... except i'm not positive what @identity-wg is

drew 41 (Thu, 25 Jan 2018 19:27:55 GMT):
@tobowers I just joined as well, so idk but my guess is it would a working group based on identity.

drew 41 (Thu, 25 Jan 2018 19:29:15 GMT):
@tobowers More information available on the wiki - https://wiki.hyperledger.org/groups/identity/identity-wg

AnabelTunala (Thu, 25 Jan 2018 19:49:50 GMT):
Has joined the channel.

AnabelTunala (Thu, 25 Jan 2018 20:02:17 GMT):
Hi, can you help me, I already did the ALICE tutoring, but I want to learn more, where to start ?.

Sean_Bohan (Thu, 25 Jan 2018 20:24:03 GMT):
@tobowers @drew 41 the Identity WG is focused on decentralized identity, issues and implementations. It meets every other week (next call is 2/7). Interoperability of identity is a big topic and we spend a lot of time talking about it across hyperledger projects and non-HL DLTs (we have folks from Ethereum and Bitcoin communities join the calls sometimes as well as DIF) You can check out the charter here: https://docs.google.com/document/d/1NZIOmxBSKCPkJ2OW9bPdUgH7T4JVSP4zbaumXtsyOpg/edit

pl30 (Fri, 26 Jan 2018 02:36:39 GMT):
Has joined the channel.

tobowers (Fri, 26 Jan 2018 07:18:57 GMT):
thanks!

tkuhrt (Fri, 26 Jan 2018 12:40:01 GMT):
SeanBohan_Sovrin

panickervinod (Fri, 26 Jan 2018 12:42:51 GMT):
@darrell.odonnell Are the challenges open only to Residents of BC or anyone can participate?

tobowers (Fri, 26 Jan 2018 14:12:48 GMT):
cross-posting from the sovrin slack: "in the new spec in “real world examples” there is both a publicKey field and a “authentication” field. Is that just a mistake? https://pr-preview.s3.amazonaws.com/w3c-ccg/did-spec/pull/41.html#authentication"

tobowers (Fri, 26 Jan 2018 14:14:42 GMT):
or is that because the authentication section "links" to the public keys?

AnabelTunala (Fri, 26 Jan 2018 14:54:24 GMT):
What is the consensus algorithm used by indy?

redpath (Fri, 26 Jan 2018 15:00:36 GMT):
So maybe I am asking for too much too early, but lets assume I am an institution with 1000 employees and I want to add them as Sovrin Entities with a claim that I know who they are, I assume I am a Sovrin Trust Anchor who has thoroughly vetted the owners. Do I setup four nodes (as a Alice demo) and use some API (of which I have no idea where) to programmatically add these people to the Sovrin network? I assume I would know the DIDs afterward. Then I can run a test program to check if they are there. Most of any code found is over a year old. Wouldn't it be in the best interest to have such a sample? How about this simple example Imagine lets say I am a University with a list of graduates that could be added as Sovrin Entities. Certainly each granduate would not on their own add themselves but the University certainly would use the public resource to provide public information about a graduate without investing in resources out of their pocket. The University has an obligation to the graduate to assist in proving they graduated to those that need to know. Sovrin it is a decentralized public repository for information and hence this is perfect for the University to use without financial internal resources needed. This scenario from the Sovrin Trusted Anchor can certainly flush out things, such as an API that works to add Entities programmatically from a Trusted source that has done the obligation of vetting legally the owner to be added. Or maybe I am wrong about a Sovrin Trusted Anchor, but the reality is, there has to be an end to end sample for an Institution which already has an obligation to vet owners legally with public information of that owner that the institution is obligated to prove graduation (Degree) for the benefit of the owner.

tobowers (Fri, 26 Jan 2018 15:07:25 GMT):
@AnabelTunala I believe plenum https://github.com/hyperledger/indy-plenum

ashcherbakov (Fri, 26 Jan 2018 15:52:02 GMT):
@AnabelTunala Conensus algorithm is plenum which is based on RBFT: https://pakupaku.me/plaublin/rbft/5000a297.pdf https://github.com/hyperledger/indy-node/blob/master/README.md#docs-and-links

Sean_Bohan (Fri, 26 Jan 2018 15:53:42 GMT):
@redpath welcome! we are in the middle of refactoring the Getting Started Guide for Indy to help with some of this and it should be coming out soon. Indy is the codebase for the Sovrin DLT (Sovrin Foundation donated the code to Hyperledger to incubate the Indy project last year). Sovrin-specific questions can be asked over at Sovrin.org in the forums or the Sovrin slack. In your first para: 1. 1000 employees would need an edge agent on their device, and have an out of band connection request (QR Code, etc.) from the institution 2. Check out the Indy SDK - https://github.com/hyperledger/indy-sdk 3. We are working on upgrading the documentation across the board, including PUML diagrams and more code samples. More stuff to come 2nd para: 1. The university wouldn't necessarily need a blockchain for a public list of graduates - a database could do it. 2 . A university would give a grad a verifiable claim that could include graduation date, degree, years attended, grades, etc,. That student could share the claim or a subset of the data within with anyone they wanted - another school, potential employer, etc. Verifiers of the claim wouldn't need to call the university as the Claim is signed and not revoked 3. Sovrin describes itself as a global public utility for self sovereign identity. It isn't a decentralized public repository - no personal data is actually on the ledger. 4th para: 1. Public information is a really interesting point. Check out the awesome work the BC.gov team are doing with the org book: https://bcgov.github.io/TheOrgBook/ Instead of private, personal data in a self-sovereign mode, they are using Indy in a B2G context 2. We want to filter up more use cases, demos and samples. I think the Getting Started Guide might cover your request for an end to end sample (Alice gets a transcript). 3. As Indy is open source in active development, anyone can build on it and if you would want to contribute that kind of a sample as an explainer (especially if I am wrong and the GSG isn't exactly what you are looking for), it would be awesome

Sean_Bohan (Fri, 26 Jan 2018 15:53:42 GMT):
@redpath welcome! we are in the middle of refactoring the Getting Started Guide for Indy to help with some of this and it should be coming out soon. Indy is the codebase for the Sovrin DLT (Sovrin Foundation donated the code to Hyperledger to incubate the Indy project last year). Sovrin-specific questions can be asked over at Sovrin.org in the forums or the Sovrin slack. In your first para: 1. 1000 employees would need an edge agent on their device, and have an out of band connection request (QR Code, etc.) from the institution 2. Check out the Indy SDK - https://github.com/hyperledger/indy-sdk 3. We are working on upgrading the documentation across the board, including PUML diagrams and more code samples. More stuff to come 2nd para: 1. The university wouldn't necessarily need a blockchain for a public list of graduates - a database could do it. 2 . A university would give a grad a verifiable claim that could include graduation date, degree, years attended, grades, etc,. That student could share the claim or a subset of the data within with anyone they wanted - another school, potential employer, etc. Verifiers of the claim wouldn't need to call the university as the Claim is signed and not revoked 3. Sovrin describes itself as a global public utility for self sovereign identity. It isn't a decentralized public repository - no personal data is actually on the ledger. 4th para: 1. Public information is a really interesting point. Check out the awesome work the BC.gov team are doing with the org book: https://bcgov.github.io/TheOrgBook/ Instead of private, personal data in a self-sovereign mode, they are using Indy in a B2G context 2. We want to filter up more use cases, demos and samples. I think the Getting Started Guide covers your request for an end to end sample (Alice gets a transcript). 3. As Indy is open source in active development, anyone can build on it and if you would want to contribute that kind of a sample as an explainer (especially if I am wrong and the GSG isn't exactly what you are looking for), it would be awesome

Sean_Bohan (Fri, 26 Jan 2018 16:43:27 GMT):
Indy Channel: Thanks for joining us on the WG call yesterday. Followup: 1. Hyperledger Hackfest Los Angeles February: https://www.hyperledger.org/event/hyperledger-hackfest-february-2018 2. Indy Mailing List (architecture discussions, proposals, in-depth discussions) https://lists.hyperledger.org/mailman/listinfo/hyperledger-indy 3. Indy Wiki: https://wiki.hyperledger.org/projects/indy 4. Indy Folder (Agendas, Videos, Roadmaps): https://drive.google.com/drive/u/1/folders/0B_NJV6eJXAA1UWtBSmRNLTR3UjQ 5. Developer Status: https://docs.google.com/spreadsheets/d/1-8mv5dokDPd5Laqur7EZtrgxmln0sY5CiJuSsL1Rg9Q/edit#gid=970515928 6. Entity Relationship Diagram: https://github.com/hyperledger/indy-node/blob/master/docs/relationship-diagram.png 7. Draft roadmap: https://drive.google.com/open?id=12qYcIxwqzQaB9Bri1c2DyGInejzS8Dwx 8. WG Call recording: https://drive.google.com/open?id=1c3elWEP-2qG9xMKFULBFX52BBMifBJJU 9. BC Dev Exchange Bounties for Indy and DID Auth: https://www.bcdevexchange.org/projects/prj-verifiable-organizations-network---theorgbook 10. Indy Agenda Suggestions: https://docs.google.com/document/d/15CxUMxaUB3yZ9PSNEX87FUUf1YqV0zEGzQJFb0TX4Tc/edit?usp=sharing

silasdavis (Fri, 26 Jan 2018 17:50:34 GMT):
Has left the channel.

gaijinviki (Mon, 29 Jan 2018 02:50:20 GMT):
Has joined the channel.

gaurang (Mon, 29 Jan 2018 08:17:49 GMT):
Has joined the channel.

ajayatgit (Mon, 29 Jan 2018 19:48:25 GMT):
Has joined the channel.

swcurran (Mon, 29 Jan 2018 21:46:12 GMT):
@panickervinod - in response to your question about the BC Gov bounties - they are open to anyone in the world. They are not limited to residents of BC.

nmellal (Mon, 29 Jan 2018 23:47:29 GMT):
Has joined the channel.

Dimple-Kanwar (Tue, 30 Jan 2018 12:39:22 GMT):
Has joined the channel.

RichLevinson (Tue, 30 Jan 2018 16:34:19 GMT):
Has joined the channel.

dilipsam (Wed, 31 Jan 2018 16:16:25 GMT):
Has joined the channel.

gen_el (Thu, 01 Feb 2018 09:22:05 GMT):
Has joined the channel.

panickervinod (Thu, 01 Feb 2018 15:53:50 GMT):
@swcurran Thank you for clarifying

pimotte (Fri, 02 Feb 2018 07:30:05 GMT):
Has joined the channel.

jljordan_bcgov (Fri, 02 Feb 2018 18:13:04 GMT):
I posted in #indy-sdk as well ... Hi All. We are happy to announce we have selected a dev for the BC Government Code With Us opportunity we offered. @ianco will be working with us to complete the work over the next few months. See https://github.com/bcgov/TheOrgBook/issues/164 for more as well as this is related to https://jira.hyperledger.org/browse/IS-486. Please welcome Ian to the community 🙂

nage (Fri, 02 Feb 2018 21:32:54 GMT):
Welcome @ianco !

ianco (Fri, 02 Feb 2018 22:01:56 GMT):
@nage thanks!

jljordan_bcgov (Sat, 03 Feb 2018 06:26:35 GMT):
Hi All. We are happy to announce we have selected a lead architect/dev for the BC Government Code With Us opportunity to work on an interoperable DID Auth protocol and implementation. @peacekeeper will be working with ourselves and the community to complete the work over the next few months. See https://github.com/bcgov/TheOrgBook/issues/165 for the opportunity details. We are excited to get going on this important initiative!

atonit (Sat, 03 Feb 2018 12:27:38 GMT):
Has joined the channel.

ajithjosek (Sun, 04 Feb 2018 20:19:22 GMT):
Has joined the channel.

salmanbaset (Sun, 04 Feb 2018 22:17:23 GMT):
Has joined the channel.

chokha (Mon, 05 Feb 2018 14:38:35 GMT):
Has joined the channel.

Jasdog (Wed, 07 Feb 2018 06:17:50 GMT):
Has joined the channel.

drummondreed (Wed, 07 Feb 2018 17:33:14 GMT):
@jljordan_bcgov Congrats to BC Gov on these two awards. This is really great news!

wangdong (Thu, 08 Feb 2018 08:00:44 GMT):
Has joined the channel.

gudkov (Thu, 08 Feb 2018 10:47:47 GMT):
Just merged new Getting Started Guide for Indy SDK https://github.com/hyperledger/indy-sdk/blob/master/doc/getting-started/getting-started.md We will appreciate any feedback.

panickervinod (Thu, 08 Feb 2018 14:38:30 GMT):
@gudkov Looks Good. Will try this out..

panickervinod (Thu, 08 Feb 2018 14:41:16 GMT):
@ashcherbakov & @Sean_Bohan Is there a plan to move http://github.com/brycecurtis/indy-tutorial-sandbox to https://github.com/hyperledger/indy-node/tree/master/environment/docker/getting_started_turnkey

panickervinod (Thu, 08 Feb 2018 15:43:53 GMT):
@gudkov Would be nice if you have a pictorial view upfront in the "Introducing Alice Section" and Numbered steps for all the Alice, Faber, Acme,Thrift & government interactions.

jljordan_bcgov (Thu, 08 Feb 2018 16:54:37 GMT):
yay @nage!

nage (Thu, 08 Feb 2018 17:07:39 GMT):
From the WG call today: As of this week I have left Evernym to pursue a new opportunity as the CTO of the Sovrin Foundation. Besides immediately improving our maintainer diversity to extend to another organization beyond Evernym, this means I will be acting as a technical voice in Indy directly in support of the Sovrin Trust Framework without any corporate conflicts of interest. It also makes me more free to pursue additional contributors and support development efforts to grow the Sovrin network (and thus the Indy development community) without creating any conflicts with Evernym’s priorities or interests. Thanks again to the kind words and well wishes on the call and over chat. I look forward to all the great things we get to build with Indy!

panickervinod (Thu, 08 Feb 2018 17:50:54 GMT):
Congrats @nage !!

siddesh_sangodkar (Fri, 09 Feb 2018 07:01:46 GMT):
Has joined the channel.

siddesh_sangodkar (Fri, 09 Feb 2018 07:11:05 GMT):
Hi! i took a quick look at Getting Started Guide: https://github.com/hyperledger/indy-sdk/blob/master/doc/getting-started/getting-started.md I did not find any information regarding how the identity provided by INDY can be used in other ecosystem like fabric project. Is this possible currently? If yes. Any guide on this is available?

AndrewHughes3000 (Fri, 09 Feb 2018 09:16:32 GMT):
Has joined the channel.

veralimita (Fri, 09 Feb 2018 12:13:37 GMT):
Has joined the channel.

veralimita (Fri, 09 Feb 2018 12:14:44 GMT):
@siddesh_sangodkar +1

alain2sf (Fri, 09 Feb 2018 12:40:24 GMT):
Congrats @nage and best wishes in your new endeavour!

landgraf.paul (Fri, 09 Feb 2018 13:18:12 GMT):
Has joined the channel.

grkvlt (Fri, 09 Feb 2018 15:29:18 GMT):
I'm also interested in how Indy might be used with other Hyperledger projects, for example as a source of identities for Sawtooth?

wangdong (Fri, 09 Feb 2018 16:20:46 GMT):
@grkvlt I would like to know this too. If any info, would you please share it?

jon_s (Fri, 09 Feb 2018 16:36:02 GMT):
Has joined the channel.

trbouma (Sat, 10 Feb 2018 14:12:27 GMT):
Has joined the channel.

ianco (Sat, 10 Feb 2018 15:54:08 GMT):
Hi Folks, just wondering if anyone has done any thinking on the concept of Guardianship in Indy/Sovrin, and how this would be implemented? I noticed there are a few JIRA tickets (e.g. https://jira.hyperledger.org/browse/INDY-624) and they have linked docs (that I don't have access to). I'd be interested in any work/thinking that has gone into this (and especially how it affects the wallet). Thanks in advance :-)

bmiller59 (Sat, 10 Feb 2018 23:23:50 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=ffeDEijXjhoKgXoCp) @grkvlt Yes, I am also interested in this.

wangdong (Mon, 12 Feb 2018 14:47:11 GMT):
pip install indy-crypto

wangdong (Mon, 12 Feb 2018 14:47:37 GMT):
but the libindy-crypto.so is not built

wangdong (Mon, 12 Feb 2018 14:47:42 GMT):
is it a bug?

wangdong (Mon, 12 Feb 2018 14:48:18 GMT):
I think it is supposed to be built and copied to /usr/lib

ashcherbakov (Mon, 12 Feb 2018 14:48:56 GMT):
libindy-crypto needs to be installed separately before installing python wrapper from pypi

wangdong (Mon, 12 Feb 2018 14:49:16 GMT):
OK

wangdong (Mon, 12 Feb 2018 14:49:36 GMT):
when I install it manually and copy the .so to /usr/lib

wangdong (Mon, 12 Feb 2018 14:49:46 GMT):
everything works well

wangdong (Mon, 12 Feb 2018 14:49:50 GMT):
and the node is up

wangdong (Mon, 12 Feb 2018 14:50:14 GMT):
@ashcherbakov thanks for reply

wangdong (Mon, 12 Feb 2018 14:50:27 GMT):
At last I can run the node

wangdong (Mon, 12 Feb 2018 14:53:00 GMT):
How can I verify the network is OK?

ashcherbakov (Mon, 12 Feb 2018 15:37:05 GMT):
You can try running a CLI. Also you can just make sure that tests pass (no need to run the nodes for this, as tests start a test pool (at the same process) automatically to run the test cases).

the_identity_guy (Mon, 12 Feb 2018 18:53:05 GMT):
Hi is the claim definition transaction going to contain the revocation material, or will that be a separate transaction on the Ledger called REVOC_REG ?

the_identity_guy (Mon, 12 Feb 2018 18:53:05 GMT):
Hi, is the claim definition transaction going to contain the revocation material, or will that be a separate transaction on the Ledger called REVOC_REG ?

NickThomas (Mon, 12 Feb 2018 23:12:39 GMT):
Has joined the channel.

rajivgandhi2010 (Tue, 13 Feb 2018 06:15:46 GMT):
Has left the channel.

ashcherbakov (Tue, 13 Feb 2018 09:46:09 GMT):
The CLAIM_DEF txn contains the main Revocation Key, but there will also be anotehr entity called REVOC_REG_DEF containing the accumulator (revoc reg) key

ashcherbakov (Tue, 13 Feb 2018 09:46:09 GMT):
@the_identity_guy The CLAIM_DEF txn contains the main Revocation Key, but there will also be anotehr entity called REVOC_REG_DEF containing the accumulator (revoc reg) key You can find more details in https://github.com/hyperledger/indy-node/blob/master/design/anoncreds.md

Sean_Bohan (Tue, 13 Feb 2018 19:30:42 GMT):
THIS THURSDAY we have the Indy Ledger WG Call for Thursday February 15, 2018 Agenda: A. Housekeeping (NathanG and SeanB) B. Development Achievements (Alex and Slava) C. Wallets and Hyperledger Indy (DanielH and IanC) D. Opening Up Indy: Sprints, Dev, etc. (NathanG and SeanB) E. Wrap Up (SeanB) Please note: This meeting will be recorded Bring a friend. It will be fun. Call details: Hyperledger Indy Node (Ledger) WG Call When: Feb 15, 2018 8amPT, 9amMT, 11amET, 4pm BST Where: Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/232861185 Or iPhone one-tap (US Toll): +16465588656,232861185# or +14086380968,232861185# Or Telephone: Dial: +1 646 558 8656 (US Toll) or +1 408 638 0968 (US Toll) Meeting ID: 232 861 185 International numbers available: https://zoom.us/zoomconference?m=a0jD_rTMnh0ZYGQDOKPCNrK_0dP7WPfp1 Hyperledger Indy Community Indy Rocketchat: https://chat.hyperledger.org/channel/indy https://chat.hyperledger.org/channel/indy-agent https://chat.hyperledger.org/channel/indy-node Indy Wiki: https://wiki.hyperledger.org/projects/indy Indy WG Agendas: https://drive.google.com/open?id=1wNnp1pPS6-1Y4B2oygfI6tOU4eHGmpRH Indy WG Meeting Recordings: https://drive.google.com/open?id=0B_NJV6eJXAA1UlJZMXd3cm1zNDQ

prathyushpv (Wed, 14 Feb 2018 06:17:59 GMT):
Has joined the channel.

RameshKunhiraman (Wed, 14 Feb 2018 14:29:13 GMT):
Has joined the channel.

hawkmauk (Wed, 14 Feb 2018 16:10:40 GMT):
I've been doing some swatting up on identity management systems as I've only had experience as a consumer. I was surprised to see that 'Active Directory Federated Services' and SAML seem to already provide a solution to delegated trust and Indy follows a similar design. Forgive me if this is a naive question but is there a compelling reason why Indy would be favoured over these existing solutions?

hawkmauk (Wed, 14 Feb 2018 16:10:40 GMT):
I've been doing some swatting up on identity management systems as I've only had experience as a consumer. I was surprised to see that 'Active Directory Federated Services' and SAML seem to already provide a solution to delegated trust and Indy follows a similar design. Forgive me if this is a naive question but is there a compelling reason why Indy would be favoured over existing solutions?

Dan (Wed, 14 Feb 2018 19:52:42 GMT):
howdy which aggregate signature scheme are you guys using?

mohdhafeezaj (Wed, 14 Feb 2018 21:31:28 GMT):
Has joined the channel.

mjmckean (Wed, 14 Feb 2018 22:22:04 GMT):
Has joined the channel.

AjeetBansal (Wed, 14 Feb 2018 23:26:41 GMT):
Has joined the channel.

gudkov (Thu, 15 Feb 2018 04:18:06 GMT):
Bls based multisignature

rishabhmthakur (Thu, 15 Feb 2018 08:25:58 GMT):
Has joined the channel.

daveryIBM (Thu, 15 Feb 2018 13:44:25 GMT):
Has joined the channel.

nage (Thu, 15 Feb 2018 16:20:25 GMT):
@hawkmauk certainly patterns for delegates credentials are helpful for all types of systems, so my hope is that over time centralized systems would use these patterns more and more. As for Indy (and the Sovrin Network), these systems enable you to interact in a peer to peer manner, without needing an identity provider to mediate the interaction. This can radically simplify the type of integration required for interoperability and data sharing. As for systems like active directory, I believe they could make some of the very best credential issuers—they wouldn’t have to handle online credentials verification anymore and could leverage known keys beyond their own root of trust off the public ledger (some of the stickiest integration problems in terms of delegating authority and preserving privacy).

nage (Thu, 15 Feb 2018 16:20:25 GMT):
@hawkmauk certainly patterns for delegated credentials are helpful for all types of systems, so my hope is that over time centralized systems would use these patterns more and more. As for Indy (and the Sovrin Network), these systems enable you to interact in a peer to peer manner, without needing an identity provider to mediate the interaction. This can radically simplify the type of integration required for interoperability and data sharing. As for systems like active directory, I believe they could make some of the very best credential issuers—they wouldn’t have to handle online credentials verification anymore and could leverage known keys beyond their own root of trust off the public ledger (some of the stickiest integration problems in terms of delegating authority and preserving privacy).

twshelton (Thu, 15 Feb 2018 16:52:28 GMT):
Has joined the channel.

nage (Thu, 15 Feb 2018 17:00:20 GMT):
Email list for wallet discussion https://lists.hyperledger.org/mailman/listinfo/hyperledger-indy

PJUllrich (Fri, 16 Feb 2018 09:05:04 GMT):
Has joined the channel.

hawkmauk (Fri, 16 Feb 2018 09:20:16 GMT):
@nage thanks for the feedback which does clarify the advantages and where Indy sits in the ecosystem. I've been surprised at the amount of industry investment into the area of identity management with the recent blog post from Microsoft[1] and the growing memberhip of the ID2020[2] initiative so it's exciting to have the opportunity to contribute and watch as the space evolves. There will be a steep learning curve for mass adoption so technology development will need to be matched with development of educational material. [1] https://cloudblogs.microsoft.com/enterprisemobility/2018/02/12/decentralized-digital-identities-and-blockchain-the-future-as-we-see-it/ [2] http://markets.businessinsider.com/news/stocks/The-ID2020-Alliance-Announces-New-Partners-in-Digital-Identity-Initiative-1013613037

hawkmauk (Fri, 16 Feb 2018 09:20:16 GMT):
@nage thanks for the feedback which does clarify the advantages and where Indy sits in the ecosystem. I've been surprised at the amount of industry investment into the area of identity management with the recent blog post from Microsoft[1] and the growing membership of the ID2020[2] initiative so it's exciting to have the opportunity to contribute and watch as the space evolves. There will be a steep learning curve for mass adoption so technology development will need to be matched with development of educational material. [1] https://cloudblogs.microsoft.com/enterprisemobility/2018/02/12/decentralized-digital-identities-and-blockchain-the-future-as-we-see-it/ [2] http://markets.businessinsider.com/news/stocks/The-ID2020-Alliance-Announces-New-Partners-in-Digital-Identity-Initiative-1013613037

hawkmauk (Fri, 16 Feb 2018 11:16:23 GMT):
Is there anyone from the Indy community that will be attending the Hyperledger Hackathon in Amsterdam?

nage (Fri, 16 Feb 2018 13:23:19 GMT):
Yes. I am planning on attending. Anyone else?

ashcherbakov (Fri, 16 Feb 2018 13:54:27 GMT):
I would like to attend as well

Sean_Bohan (Fri, 16 Feb 2018 16:00:59 GMT):
Hey @hawkmauk re: Amsterdam, right now it is myself, Nathan, Alex and possibly 1 or two others

hawkmauk (Fri, 16 Feb 2018 16:12:02 GMT):
@nage @ashcherbakov @Sean_Bohan I'm booked up to attend too so it'd be good to meet everyone

Sean_Bohan (Fri, 16 Feb 2018 16:12:15 GMT):
outstanding

nage (Fri, 16 Feb 2018 16:43:48 GMT):
I have just sent out an email to the list about the indy-rfc repository I've been talking about with some of the maintainers for a while. Take a look and offer feedback. I'm hoping to have some content in there to discuss at the LA Hackfest next week.

Drew 42 (Fri, 16 Feb 2018 21:35:58 GMT):
Has joined the channel.

SwePalm (Sun, 18 Feb 2018 19:06:22 GMT):
Has joined the channel.

Harmannz (Mon, 19 Feb 2018 19:59:16 GMT):
Hi there, One question that I've been getting asked when explaining Self-sovereign identity is how can a verifier be confident that an issuer has performed proper checks to issue a claim for a user. This level of assurance question is particularly important for usecases where legislation's require organisations to have a high level of confidence that the users they interact with are who they claim to be. For example, AML (anti-money laundering) legislation in my country requires organisations to follow strict procedures (such as non-photographic identification + driver license) for onboarding and performing 'high-risk' transactions. If a user was to transfer large sum of money, they would have to bring their verifying documents to the bank and be verified for AML purposes. Self-sovereign Identity can streamline a lot of these issues. For example, once a user is given a verified claim he/she can use that claim to verify the transfer of large funds from their account. A user is no longer required to go to the bank and provide proof of their ID again and again. The only issue is that organisations need to have a high-level of confidence that the issuer of a claim has performed the necessary checks, with respect to the current legislation, to provide the claim and that the claim is up-to-date. Reading the newly released whitepaper, it seems that insurers could provide a solution to this problem? Is there documentation that describes how insurers work in detail? Do you have any opinions on how to solve this problem?

jljordan_bcgov (Tue, 20 Feb 2018 01:06:01 GMT):
@Harmannz it is an important and totally relevant question. The answer is not a technical one but a process one. The assurance level of a claim is related to the processes by which the claim was issued. This should generally be the domain of a trust framework. In Canada we are working on a Pan Canadian Trust Framework (https://diacc.ca/pan-canadian-trust-framework/) which would be an example of such a framework. The main point is to have a set of common processes and conformance criteria that issuers (in our case government identity issuers) follow and comply with such that the claims they issue can be considered trusted and used within and across jurisdictions for services that require stronger levels of assurance. There could be other mechanisms such as insurers as well I would imagine.

Harmannz (Tue, 20 Feb 2018 04:10:15 GMT):
@jljordan_bcgov Thanks for your reply. You are right. I was trying to avoid mentioning trust framework to see if Indy authors are planning to introduce any mechanisms to make this easier. A trust framework is something we are currently investigating, but developing it will require collaboration from multiple public and private sector organisations which may be the next step for us. Realistically, there isn't really a need for insurers if there is a trust framework in place and all participating parties are legally bound to it.   I have been following the Pan Canadian Trust Framework and I am interested to see how the final document will look like. I have read the Verified Person Component (https://github.com/canada-ca/PCTF). Do you guys have a timeline on when the PCTF will be publically available?

jljordan_bcgov (Tue, 20 Feb 2018 14:44:59 GMT):
My understanding is the focus is on Verified Person and Verified Login right now. My main area of activity is Organizations ... we have our work in this area available at http://von.pathfinder.gov.bc.ca .. there are links there to demos, github repos and some documentation

MoeAfilal (Tue, 20 Feb 2018 19:27:53 GMT):
Has joined the channel.

malvikam (Tue, 20 Feb 2018 21:05:27 GMT):
Has joined the channel.

abdulbasitg (Wed, 21 Feb 2018 14:32:54 GMT):
Has joined the channel.

Dan (Wed, 21 Feb 2018 16:55:10 GMT):
@nage I put a sticky note on the planning board here at the hackfest to do the app dev

Dan (Wed, 21 Feb 2018 16:55:10 GMT):
@nage I put a sticky note on the planning board here at the hackfest to do the hac

nage (Wed, 21 Feb 2018 16:56:39 GMT):
Excellent

Dan (Wed, 21 Feb 2018 16:56:39 GMT):
@pschwarz will lead

pschwarz (Wed, 21 Feb 2018 16:56:40 GMT):
Has joined the channel.

kdenhartog (Wed, 21 Feb 2018 18:38:46 GMT):
Has joined the channel.

nickris (Wed, 21 Feb 2018 19:14:08 GMT):
Has joined the channel.

mdoan (Wed, 21 Feb 2018 19:35:47 GMT):
Has joined the channel.

mdoan (Wed, 21 Feb 2018 19:37:26 GMT):
Hi all, if my business use indy or any other hyperledger project do i need to create my own network or my business network will be shared with other in the permissioned network - because if it deploys it within my own servers it won't be transparent to users

swcurran (Wed, 21 Feb 2018 20:02:14 GMT):
@mdoan - you could create your own network, but the intention of most is that the Sovrin foundation will be operating the "World Wide" instance of Indy that we will all use. If your organization wants, it would apply to be a Sovrin Steward and operate a node. Most of us are running a private test instance of Indy nodes while we learn and create code, but in the big picture use, I think we all expect to be part of the Sovrin network.

mdoan (Wed, 21 Feb 2018 20:03:59 GMT):
@swcurran why Sovrin Steward ? not hyperledger?

swcurran (Wed, 21 Feb 2018 20:07:57 GMT):
Hyperledger owns the open source indy code, but the plan is that Sovrin will operate an instance of the Indy code around the world. Hyperledger is not going to operate an instance of an HL-Indy network. Others could operate an instance of the code as a network, but given the decentralized nature of the Sovrin Foundation and their Trust Framework, if you buy into the concepts in the Indy code, you likely buy into Sovrin as the operator of the network. See sovrin.org for more about the network the Foundation operates.

swcurran (Wed, 21 Feb 2018 20:07:57 GMT):
Hyperledger owns the open source indy code, but Sovrin already operates an instance of the Indy code around the world. Hyperledger is not going to operate an instance of an HL-Indy network. Others could operate an instance of the code as a network, but given the decentralized nature of the Sovrin Foundation and their Trust Framework, if you buy into the concepts in the Indy code, you likely buy into Sovrin as the operator of the network. See sovrin.org for more about the network the Foundation operates.

swcurran (Wed, 21 Feb 2018 21:45:52 GMT):
One more on this - you don't have to have a Steward on the Sovrin network to use it. Any identity can use the network. A Steward is just if you want to operate a part of the network.

Er1ck (Wed, 21 Feb 2018 21:48:46 GMT):
Has joined the channel.

Sean_Bohan (Thu, 22 Feb 2018 00:56:36 GMT):
Folks: THIS THURSDAY (TOMORROW!) we have the Indy Agent WG Call for Thursday February 22, 2018 Agenda: A. Housekeeping (Nathan and Sean) B. Development Achievements (Alex and Slava) C. Hyperledger Indy SDK - Wallet Discussion Part 2 with Ian & Daniel D. Wrap Up Please note: This meeting will be recorded Bring a friend. It will be fun. Call details: Hyperledger Indy Agent (SDK) WG Call When: Feb 22, 2018 8amPT, 9amMT, 11amET, 4pm BST Where: Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/232861185 Or iPhone one-tap (US Toll): +16465588656,232861185# or +14086380968,232861185# Or Telephone: Dial: +1 646 558 8656 (US Toll) or +1 408 638 0968 (US Toll) Meeting ID: 232 861 185 International numbers available: https://zoom.us/zoomconference?m=a0jD_rTMnh0ZYGQDOKPCNrK_0dP7WPfp1 Hyperledger Indy Community Indy Rocketchat: https://chat.hyperledger.org/channel/indy https://chat.hyperledger.org/channel/indy-agenthttps://chat.hyperledger.org/channel/indy-node Indy Wiki: https://wiki.hyperledger.org/projects/indy Indy WG Agendas: https://drive.google.com/open?id=1wNnp1pPS6-1Y4B2oygfI6tOU4eHGmpRH Indy Videos: https://drive.google.com/open?id=1Z8-jR7hnXb57fufE0OXxIfeUn05zNmRt Indy WG Meeting Recordings: https://drive.google.com/open?id=0B_NJV6eJXAA1UlJZMXd3cm1zNDQ

mdoan (Thu, 22 Feb 2018 01:08:33 GMT):
@swcurran can i define the spec of the identity data model in the network?

swcurran (Thu, 22 Feb 2018 01:09:47 GMT):
You mean the schema of the Verifiable Claim - what data it contains? Yes. That gets published by an issuer - any issuer - to the ledger and anyone can use it.

swcurran (Thu, 22 Feb 2018 01:09:47 GMT):
You mean the schema of the Verifiable Claim - what data fields it contains? Yes. That gets published by an issuer - any issuer - to the ledger and anyone can use it.

the_identity_guy (Thu, 22 Feb 2018 02:21:59 GMT):
@Sean_Bohan Is it possible to get the recording of the Thursday Feb 15 meeting, I cant find it on https://drive.google.com/drive/folders/0B0ChWSmqhJemS21ldjNsYVBtNUk, thank you

the_identity_guy (Thu, 22 Feb 2018 02:21:59 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=c7PeSFPPyDZJ4AuTX) @Sean_Bohan Is it possible to get the recording of this meeting, I cant find it on https://drive.google.com/drive/folders/0B0ChWSmqhJemS21ldjNsYVBtNUk, thank you

mattsmithies (Thu, 22 Feb 2018 14:09:13 GMT):
Has joined the channel.

dkulic (Thu, 22 Feb 2018 16:19:48 GMT):
Has joined the channel.

jankokrstic (Thu, 22 Feb 2018 16:56:58 GMT):
Has joined the channel.

danielhardman (Thu, 22 Feb 2018 17:04:43 GMT):
Here is a link to the wallet notes doc I shared on today's call: https://docs.google.com/document/d/1uvoZkMdZz-TZKrTYbyLXSa-kDueIex69BBy2SGPgQ2s/edit

the_identity_guy (Thu, 22 Feb 2018 17:09:13 GMT):
thank you, could you also share the github wiki that has the VON details

Sean_Bohan (Thu, 22 Feb 2018 17:31:05 GMT):
start here @the_identity_guy https://www.bcdevexchange.org/projects/prj-verifiable-organizations-network---theorgbook

ianco (Thu, 22 Feb 2018 17:42:30 GMT):
Here's the link for the doc I shared in today's call: https://github.com/ianco/indy-sdk/blob/master/doc/wallet/enterprise-wallet-design.md

shicholas (Thu, 22 Feb 2018 17:53:41 GMT):
Has joined the channel.

hawkmauk (Thu, 22 Feb 2018 18:47:30 GMT):
Thanks for another informative call which helped to further my understanding of the role of anoncreds and wallets in Indy.

peter-choe (Thu, 22 Feb 2018 19:42:05 GMT):
Has joined the channel.

anaik (Thu, 22 Feb 2018 19:43:59 GMT):
Has joined the channel.

kdenhartog (Thu, 22 Feb 2018 20:00:13 GMT):
http://redhat.slides.com/npmccallum/devconf17#/

kdenhartog (Thu, 22 Feb 2018 20:00:22 GMT):
https://archive.fosdem.org/2017/schedule/event/securing_automated_decryption/

Sean_Bohan (Thu, 22 Feb 2018 21:47:21 GMT):
Folks: Below, please find links to the docs shared in the Hyperledger Indy WG call today (2/22/2018) Video recording of the meeting: https://drive.google.com/open?id=1QCHycndmn597KAa2WS4DJLZuhpCaoQSa Agenda of the call: https://docs.google.com/presentation/d/1SQfjirAYC58w1diLbFX5cVnYhJFtaAkcXm0_ojkIRi8/edit#slide=id.g32810f0578_0_13 Daniel's notes about Wallets: https://docs.google.com/document/d/1uvoZkMdZz-TZKrTYbyLXSa-kDueIex69BBy2SGPgQ2s/edit Ian's doc about Wallets https://github.com/ianco/indy-sdk/blob/master/doc/wallet/enterprise-wallet-design.md Slava's deck on Anoncreds API changes: https://docs.google.com/presentation/d/1whbMoVhw8oxK_CasOBfs0e5Yyyp3KKPKgMnzhRv1UCo/edit#slide=id.g306237b895_0_0 The Org Book from BC.gov team: https://www.bcdevexchange.org/projects/prj-verifiable-organizations-network---theorgbook https://bcgov.github.io/TheOrgBook/ Hyperledger Hackfest: Amsterdam (June 27-29) https://www.hyperledger.org/event/hyperledger-hackfest-june-2018 Hyperledger IDENTITY Working Group (Feb 21, 2018: https://wiki.hyperledger.org/groups/identity/identity-wg Check out the Hyperledger Indy Wiki: https://wiki.hyperledger.org/projects/indy Join the Mailing List: https://lists.hyperledger.org/mailman/listinfo/hyperledger-indy Internet Identity Workshop April 2-5, Mountainview, CA - Hyperledger Discount Code: https://iiw26.eventbrite.com?discount=HL_XXVI_20

VolodymyrRudyi (Sun, 25 Feb 2018 10:26:49 GMT):
Has joined the channel.

anttikettunen (Sun, 25 Feb 2018 22:50:06 GMT):
Hi everyone @here. I've felt for a long time that Indy's community should grow faster and deeper. One crucial step in community growth is to have good documentation, which helps onboarding of new users. I got fed up with Indy's documentation being scattered and outdated at times, so I decided to take example from our distributed brothers at Fabric & Corda. They have taken use of "Read the Docs" / Sphinx for documentation. I've created a starting point for creating better Indy/Sovrin documentation. I've just created a structure for now, but the plan would be to incorporate all the documentation created thus far and curate it into the documentation structure so that it would serve new users as well as the developers working on indy/Sovrin. To avoid any overlapping initiatives, I'd like to get a) info if anybody else is working on similar things and b) your feedback on what would be a useful structure. Also, I'd be interested to know, if Hyperledger has a logical place to host this documentations, or should it rather be at sovrin.org? I've thought that having it in the indy codebase, would make it most natural. I have the docs in a forked indy-node -repo in github. The current version of the docs can be found at http://ledger.sovrinsandbox.pub.tds.tieto.com/docs/index.html Your thoughts?

anttikettunen (Sun, 25 Feb 2018 22:50:06 GMT):
Hi everyone @here. I've felt for a long time that Indy's community should grow faster and deeper. One crucial step in community growth is to have good documentation, which helps onboarding of new users. I got fed up with Indy's documentation being scattered and outdated at times, so I decided to take example from our distributed brothers at Fabric & Corda. They have taken use of "Read the Docs" / Sphinx for documentation. I've created a starting point for creating better Indy/Sovrin documentation. I've just created a structure for now, but the plan would be to incorporate all the documentation created thus far and curate it into the documentation structure so that it would serve new users as well as the developers working on indy/Sovrin. To avoid any overlapping initiatives, I'd like to get a) info if anybody else is working on similar things and b) your feedback on what would be a useful structure. Also, I'd be interested to know, if Hyperledger has a logical place to host this documentations, or should it rather be at sovrin.org? I've thought that having it in the indy codebase, would make it most natural. I have the docs in a forked indy-node -repo in github. Your thoughts?

Harmannz (Mon, 26 Feb 2018 00:41:14 GMT):
@anttikettunen This is awesome. I was confused when I first started because I kept traversing between sovrin and indy getting started guides. Indy has been undergoing restructuring so things may not be so tidy right now, but there are plans to collect all indy codebase into just two repositories (indy-sdk and indy-node) so that should also ease the strain for developers to get acclimated with the library. (My opinion)

Harmannz (Mon, 26 Feb 2018 00:41:14 GMT):
@anttikettunen This is awesome. I was confused when I first started because I was traversing between sovrin and indy getting started guides. Indy has been undergoing restructuring so things may not be so tidy right now, but there are plans to collect all indy codebase into just two repositories (indy-sdk and indy-node) so that should also ease the strain for developers to get acclimated with the library. (My opinion)

vinomaster (Mon, 26 Feb 2018 14:52:00 GMT):
@danielhardman @swcurran Gents just trying to catch upon the Wallet design discussions. If storage of SSI metadata and claims will be separate where will the correlation between DID and Claims (per connection) be maintained?

nage (Mon, 26 Feb 2018 15:41:12 GMT):
@anttikettunen I agree we need to do more. I have asked the conversation from Sovrin Slack to move here, lets see if it follows us.

nage (Mon, 26 Feb 2018 15:44:42 GMT):
Basically we have three things that have been proposed: 1) docs folders in each repo for user-facing documentation (getting started guides, how to build/test/contribute, and similar user-facing docs) 2) The new indy-rfc repository for Indy Improvement Initiatives. This is an equivalent to Rust's RFCs, Python PEPs and Bitcoin BIPs. The idea here is for protocol issues, architecture, recommendations and modifications to the community standard to be proposed and discussed in the open, that way we are left with documentation of the architecture with version control, along with discussions and artifacts that help us point new contributors in the direction of folks that can help them get started with things that want to influence and change. 3) This new readthedocs repository that would be a compiled version, presumably of the documentation in location 1 & 2.

TechWritingWhiz (Mon, 26 Feb 2018 15:47:02 GMT):
Has joined the channel.

nage (Mon, 26 Feb 2018 15:50:23 GMT):
I have a task this week to get the initial structure of indy-rfc setup so that we can get a few process documents underway (Improvement Initiative process iii-0001, an architecture intro iii-0101, a document on basic architecture (repo and code organization) iii-0005, documentation on privacy by design architecture iii-0006, documentation of the anoncreds scheme iii-0007, documentation of the agent messaging protocol iii-0011 -- notice I'm skipping some numbers to leave room for documents that we haven't started yet).

anttikettunen (Mon, 26 Feb 2018 16:05:12 GMT):
@nage I had a thought that the sphinx source docs & html would replace the current docs-folder in indy-node (perhaps partially in indy-sdk as well). We could link or build the relevant iii's into the developer guides there. This would give us a single place to look for the info, instead of having it scattered around. What's your take on where the live built sphinx site would/should/could be hosted?

anttikettunen (Mon, 26 Feb 2018 16:13:35 GMT):
Btw. I'm happy to do some coordination and curation work to get at least the first generation of documentation running.

nage (Mon, 26 Feb 2018 17:04:52 GMT):
I was hoping to use the github.io site that matches the indy-rfc repo to start with (keep it simple)

nage (Mon, 26 Feb 2018 17:05:12 GMT):
but I haven't looked at what tools or process we want to use to do that yet

nage (Mon, 26 Feb 2018 17:05:45 GMT):
I'm open to suggestions (and volunteering to do the work usually means your suggestion will win ;) ).

Sean_Bohan (Mon, 26 Feb 2018 18:12:24 GMT):
are we not going to use the wiki as the hub going forward?

agunde (Mon, 26 Feb 2018 18:52:37 GMT):
Has joined the channel.

TechWritingWhiz (Mon, 26 Feb 2018 21:30:54 GMT):
@nage doesn't github.io have a like a 3 gig storage limit though?

TechWritingWhiz (Mon, 26 Feb 2018 21:31:12 GMT):
That may not be enough.

nage (Mon, 26 Feb 2018 22:07:05 GMT):
These should be small .md files, and we can reference other resources elsewhere if anything gets gnarly, but yes, I hope the RFCs stay well below 3G

nage (Mon, 26 Feb 2018 22:07:20 GMT):
If it isn’t enough, I wouldn’t be sad to have that problem

nage (Mon, 26 Feb 2018 22:07:50 GMT):
(So much documentation we don’t know where to put it)

mawi (Tue, 27 Feb 2018 08:08:49 GMT):
Has joined the channel.

anttikettunen (Tue, 27 Feb 2018 08:43:36 GMT):
@nage was your idea, that III's would be presented like the Rust RFC's here: http://rust-lang.github.io/rfcs/ ? Could the III process be roughly the following: 1) Create an III for an improvement at the Github issue queue. 2) The III is worked on in the issue, until it's done and marked as 'approved' 3) The approved (or also drafts?) are listed at the github.io page, similar to the Rust RFC page above. 4) Once an III is approved, it should be modified to a documentation-proper format (for example .md or .rst for Sphinx), and added to the documentation in indy repos (also referencing the original III issue). I see the issue queue as a working area and history for the III's, but not necessarily a reference documentation, that we should use, since there will be no logical sorting of the improvements, or a larger picture at hand. The actual documentation then would be more flexible for changes and explaining related things.

sergey.khoroshavin (Tue, 27 Feb 2018 10:26:59 GMT):
Has joined the channel.

smcnamara (Tue, 27 Feb 2018 11:49:49 GMT):
Has joined the channel.

nage (Tue, 27 Feb 2018 12:46:42 GMT):
Yes, though I wasn’t requiring step 4 yet, as we don’t have many contributors focusing on documentation, so I’m not sure if we can support moving them to yet another location unless it is part of the build process of the repo somehow.

hawkmauk (Tue, 27 Feb 2018 12:55:07 GMT):
@nage @anttikettunen Count me in for contributing to documentation

Toktar (Tue, 27 Feb 2018 14:15:02 GMT):
Has joined the channel.

LeBaton (Tue, 27 Feb 2018 14:26:55 GMT):
Has joined the channel.

dsurnin (Tue, 27 Feb 2018 14:43:45 GMT):
Hi! We are going to improve our node monitoring tool - validator_info. Some design notes are listed here https://github.com/hyperledger/indy-node/pull/584 Please feel free to post your comments or ideas

VikasJakhar (Tue, 27 Feb 2018 20:04:45 GMT):
Has joined the channel.

SergeyPalamarchuk (Wed, 28 Feb 2018 07:59:59 GMT):
Has joined the channel.

arjanvaneersel (Wed, 28 Feb 2018 12:19:52 GMT):
I went through the documentation and the Alice/Faber demo several times now, but one thing still isn't clear to me. When is something written on the Indy blockchain and which details are exactly written on the blockchain. It's clear to be that personal details, like the SSN of Alice will never be written on the blockchain, but I also have the impression that the blockchain in Indy is more than just an index of public keys.

ashcherbakov (Wed, 28 Feb 2018 14:03:34 GMT):
@arjanvaneersel I believe a pretty good description of this can be found in https://www.evernym.com/wp-content/uploads/2017/07/What-Goes-On-The-Ledger.pdf

arjanvaneersel (Wed, 28 Feb 2018 14:04:35 GMT):
@ashcherbakov Thanks, that looks like a very detailed answer to my question. Going to read that document.

ashcherbakov (Wed, 28 Feb 2018 15:50:44 GMT):
@anttikettunen @nage @Sean_Bohan @hawkmauk I've created a ticket regarding documentation: https://jira.hyperledger.org/browse/INDY-1202 This is about having more diagrams (sequence, etc.) to understand the system better from the technical point of view. Feel free to contribute, it would be very appreciated.

mboyd (Wed, 28 Feb 2018 16:55:40 GMT):
Has joined the channel.

mawi (Wed, 28 Feb 2018 17:29:52 GMT):
Hi everyone, I have a question regarding connection requests; is the DID embedded in a connection request the verinym of the trust anchor requesting the connection, or is it already a pairwise (pseudonymous) DID? In the Alice/Faber demo, when Alice downloads the connection request, she signs the nonce and returns it to Faber. Upon receiving the signed connection request, does Faber then create a pairwise DID or not?

the_identity_guy (Wed, 28 Feb 2018 17:43:16 GMT):
Is Indy a proper implementation of DPKI? is there any articles about this

nage (Thu, 01 Mar 2018 05:07:01 GMT):
mawi: in the past the getting started guide did not implement full pairwise DIDs to help simplify the tutorial, but the intent is that you would bootstrap a pairwise connection so that there was a unique id for each side. It turns out that those DIDs do not need to be anchored to the ledger unless they are used to resolve public service endpoints or have reputations for credentials (a verinym to use the term you used from the Sovrin Trust Framework).

nage (Thu, 01 Mar 2018 05:07:52 GMT):
the_identity_guy: we'd like to think so, but there may be differing opinions on what constitutes a "proper implementation of DPKI", so we should talk about your requirements and expectations.

nage (Thu, 01 Mar 2018 05:09:47 GMT):
A rough draft of the III process for the indy-rfc repository readme can be found here https://github.com/nage/indy-rfc/blob/master/README.md I was hoping to get through building 0000-template.md and starting empty text/####-topic.md files for any missing or broken links but haven't been able to get around to it yet. I'll be updating my fork of the indy-rfc repo to get to that initial pull request over the next few days.

hvandurme (Thu, 01 Mar 2018 08:00:22 GMT):
Has joined the channel.

lcinacio (Thu, 01 Mar 2018 08:17:37 GMT):
Has joined the channel.

mawi (Thu, 01 Mar 2018 08:19:40 GMT):
thanks @nage , that clarifies

drummondreed (Thu, 01 Mar 2018 08:45:56 GMT):
@the_identity_guy Yes, Indy is implementing DKMS, which is key management for DPKI. See this Rebooting the Web of Trust intro paper: https://github.com/WebOfTrustInfo/rebooting-the-web-of-trust-spring2017/blob/master/topics-and-advance-readings/dkms-decentralized-key-mgmt-system.md

drummondreed (Thu, 01 Mar 2018 08:46:21 GMT):
DKMS will be a major topic at next week's Rebooting the Web of Trust #6 in Santa Barbara

arjanvaneersel (Thu, 01 Mar 2018 08:57:44 GMT):
I'm doing the Alice/Faber demo at the moment and trying to simulate the situation that Faber would have had a data breach and Alice changes the DID for that connection. Intuitively I expected the command 'change current key' to do just that, but I seem to be wrong here or doing it in the wrong what. So what would be the proper way to do this?

arjanvaneersel (Thu, 01 Mar 2018 08:57:44 GMT):
I'm doing the Alice/Faber demo at the moment and trying to simulate the situation that Faber would have had a data breach and Alice changes the DID for that connection. Intuitively I expected the command 'change current key' to do just that, but I seem to be wrong here or doing it in the wrong way. So what would be the proper way to do this?

arjanvaneersel (Thu, 01 Mar 2018 08:57:44 GMT):
I'm doing the Alice/Faber demo at the moment and trying to simulate the situation that Faber would have had a data breach and Alice wants to change the DID for that connection. Intuitively I expected the command 'change current key' to do just that, but I seem to be wrong here or doing it in the wrong way. So what would be the proper way to do this?

ashcherbakov (Thu, 01 Mar 2018 10:23:36 GMT):
@arjanvaneersel I think you could use the following SDK calls: - indy_replace_keys_start (https://github.com/hyperledger/indy-sdk/blob/master/libindy/src/api/did.rs#L89) - send new keys (NYM) to the ledger (build_nym_request and sign_and_submit_request) - indy_replace_keys_apply (https://github.com/hyperledger/indy-sdk/blob/master/libindy/src/api/did.rs#L127) Please have a look at the test `replace_keys` (https://github.com/hyperledger/indy-sdk/blob/913a2c7702af14761f8dbea632cc17cde9003743/libindy/tests/utils/did.rs#L85)

arjanvaneersel (Thu, 01 Mar 2018 10:54:38 GMT):
@ashcherbakov I'm using the CLI and that doesn't seem to have that option

ashcherbakov (Thu, 01 Mar 2018 10:55:27 GMT):
@arjanvaneersel Are you using old (deprecated) CLI from indy-node, or a new one from indy-sdk?

arjanvaneersel (Thu, 01 Mar 2018 10:57:27 GMT):
Using indy-node/environment/docker/pool

arjanvaneersel (Thu, 01 Mar 2018 10:57:47 GMT):
Pulled the master branch from the repo yesterday

arjanvaneersel (Thu, 01 Mar 2018 10:58:31 GMT):
Those are the instructions I used: https://github.com/hyperledger/indy-node/tree/master/environment/docker/pool

wolili (Thu, 01 Mar 2018 12:36:23 GMT):
Hi, I have seen that Indy-Crypto is using BLS and Multisignatures. Are threshold signatures also supported? Is there a way to use indy-nodes to produce a Threshold signature?

wolili (Thu, 01 Mar 2018 12:36:23 GMT):
Hi, I have seen that Indy-Crypto is using BLS and Multisignatures. Are threshold signatures also supported? Is there a way to use indy-nodes to produce a threshold signature?

Sean_Bohan (Thu, 01 Mar 2018 15:17:51 GMT):
TODAY we have the Indy Ledger WG Call for Thursday Mar 1, 2018 Agenda: A. Housekeeping (Nathan and Sean) B. Development Achievements (Alex and Slava) C. Developer Experience and Demo (DanielH) D. Revocation (Nathan, Alex, Slava) E. Wrap Up Please note: This meeting will be recorded Bring a friend. It will be fun. Call details: Hyperledger Indy Agent (SDK) WG Call When: Mar 1, 2018 8amPT, 9amMT, 11amET, 4pm BST Where: Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/232861185 Or iPhone one-tap (US Toll): +16465588656,232861185# or +14086380968,232861185# Or Telephone: Dial: +1 646 558 8656 (US Toll) or +1 408 638 0968 (US Toll) Meeting ID: 232 861 185 International numbers available: https://zoom.us/zoomconference?m=a0jD_rTMnh0ZYGQDOKPCNrK_0dP7WPfp1 Hyperledger Indy Community Indy Rocketchat: https://chat.hyperledger.org/channel/indyhttps://chat.hyperledger.org/channel/indy-agenthttps://chat.hyperledger.org/channel/indy-node Indy Wiki: https://wiki.hyperledger.org/projects/indy Indy WG Agendas: https://drive.google.com/open?id=1wNnp1pPS6-1Y4B2oygfI6tOU4eHGmpRH Indy Videos: https://drive.google.com/open?id=1Z8-jR7hnXb57fufE0OXxIfeUn05zNmRt Indy WG Meeting Recordings: https://drive.google.com/open?id=0B_NJV6eJXAA1UlJZMXd3cm1zNDQ

ashcherbakov (Thu, 01 Mar 2018 16:07:36 GMT):
@arjanvaneersel Ok, for the old CLI there is a command `change current key` as you mentioned. I believe this is what you need.

nage (Thu, 01 Mar 2018 16:24:12 GMT):
@wolili I don't think we've pursued using our BLS implementation for more generic threshold signature use case -- we use it to have validator nodes collectively sign state. But my understanding is that it is possible if someone wanted to pursue it. My hope is that moving it into the new shared crypto library would help attract these types of users and contributions (we'll wait until we have a reason to use a more generic threshold signature before signing off on that type of use). @ashcherbakov might have more to say about what is in the current implementation.

ashcherbakov (Thu, 01 Mar 2018 16:34:21 GMT):
We have a quite abstarct BLS multi-signature implementation in Indy Crypto lib: https://github.com/hyperledger/indy-crypto/tree/master/libindy-crypto/src/bls Indy Plenum just uses this implementation and integrates it into RBFT consensus protocol.

danielhardman (Thu, 01 Mar 2018 16:50:10 GMT):
@jankokrstic and I just did a demo of the work we are doing to create short developer tutorials on various topics to make learning easier in the indy ecosystem. I wanted to post more info about it: 1. Link to the first written tutorial, so you can see the sort of experience we have in mind: Here is the link to the first written tutorial: https://github.com/dhh1128/indy-sdk/blob/master/doc/tutorials/python/send-secure-msg.md 2. Link to a PR that contains another 4 or 5 scripts around which we plan to write more tutorials: https://github.com/hyperledger/indy-sdk/pull/551 3. Link to a recording of @jankokrstic playing the role of a developer and showing the results of following the first written demo: https://drive.google.com/drive/u/0/folders/1CmY0FbgesVZ_ZXIpYOFuG9egWgae-2FI (The sharing on this is not yet world visible, but we will fix that momentarily)

jankokrstic (Thu, 01 Mar 2018 16:55:20 GMT):
Check this link: https://drive.google.com/file/d/1LGDF08R5fpx_4TaxZL4gqc0btqy4G-3n/view?usp=sharing

jankokrstic (Thu, 01 Mar 2018 16:55:55 GMT):
The recorded demo is there

jankokrstic (Thu, 01 Mar 2018 16:56:13 GMT):
One note - I made a mistake - after decrypting the verification key is printed, not DID

jankokrstic (Thu, 01 Mar 2018 16:59:17 GMT):
The code from the demo can be available soon, after I clean it up and document it more. As Daniel said, the idea is to have many of those byte-sized tutorials demonstrating different use cases and blocks for building stuff using indy-sdk

lcinacio (Fri, 02 Mar 2018 10:14:00 GMT):
@danielhardman @jankokrstic this is a great initiative! I just started working with Indy and tutorials like this are a great help! I ran the first tutorial and the only issue was that I had to add code to write msg to plaintext.txt before trying to read it in prep...

hawkmauk (Fri, 02 Mar 2018 12:42:29 GMT):
android

alexvicegrab (Fri, 02 Mar 2018 15:43:16 GMT):
Has joined the channel.

danielhardman (Fri, 02 Mar 2018 22:35:25 GMT):
Ooh, thanks for the bug report! I would be glad to fix my tutorial, but I'd love to have you submit a PR instead so I can make you an indy contributor... What is your preference?

Sakurann (Mon, 05 Mar 2018 04:42:43 GMT):
Has joined the channel.

mspatel (Mon, 05 Mar 2018 16:02:12 GMT):
Has joined the channel.

hendry19901990 (Mon, 05 Mar 2018 16:39:40 GMT):
is there a library client to connect to indy ?

ashcherbakov (Mon, 05 Mar 2018 17:04:34 GMT):
@hendry19901990 https://github.com/hyperledger/indy-sdk There is also a CLI: https://github.com/hyperledger/indy-sdk/tree/master/cli

hendry19901990 (Mon, 05 Mar 2018 17:26:19 GMT):
:thumbsup:

tomislav (Mon, 05 Mar 2018 19:10:21 GMT):
Has joined the channel.

olegwb (Wed, 07 Mar 2018 09:01:34 GMT):
Has joined the channel.

olegwb (Wed, 07 Mar 2018 10:40:36 GMT):
skIrock7

maxmijn (Wed, 07 Mar 2018 11:17:40 GMT):
Has joined the channel.

tobowers (Wed, 07 Mar 2018 21:07:30 GMT):
I'm having a bit of trouble getting go to interact with libindy-crypto... It's returning valid for invalid sigs

vishwasbalakrishna (Wed, 07 Mar 2018 21:44:43 GMT):
Has joined the channel.

ArthurManz (Thu, 08 Mar 2018 10:02:47 GMT):
Has joined the channel.

haowangbe (Thu, 08 Mar 2018 11:57:25 GMT):
Has joined the channel.

haowangbe (Thu, 08 Mar 2018 12:19:39 GMT):
Hello! I just joined the group and I have a rather general question about how the Evernym, Sovrin and Indy projects are related to each other at the moment. If I have an idea for an application based on Indy, who can I contact?

jimscarver (Thu, 08 Mar 2018 16:52:18 GMT):
Has joined the channel.

rado0x54 (Thu, 08 Mar 2018 18:57:58 GMT):
Has joined the channel.

malvikam (Thu, 08 Mar 2018 20:26:35 GMT):
Can someone confirm if this is the correct genesis txn file to connect to STN network in order to create DIDs? https://github.com/sovrin-foundation/sovrin/blob/master/sovrin/domain_transactions_sandbox_genesis

Drew 42 (Thu, 08 Mar 2018 23:28:41 GMT):
On the getting started guide, https://github.com/hyperledger/indy-node/blob/master/getting-started.md, it says "You can install a test network in one of several ways." Which includes Docker: Start Pool and Client with Docker. However, on the page for Docker it links to (https://github.com/hyperledger/indy-node/blob/master/environment/docker/pool/StartIndyAgents.md), it started by stating, "Once you have the Indy test nodes and client running..."

Drew 42 (Thu, 08 Mar 2018 23:29:07 GMT):
What steps am I missing for getting started, before running `docker exec -i -t indyclient bash`

Drew 42 (Thu, 08 Mar 2018 23:29:08 GMT):
?

Drew 42 (Thu, 08 Mar 2018 23:43:30 GMT):
Seems like setup dev https://github.com/hyperledger/indy-node/blob/master/docs/setup-dev.md is a better place to be starting :)

carter.willetts (Fri, 09 Mar 2018 00:54:41 GMT):
Has joined the channel.

carter.willetts (Fri, 09 Mar 2018 02:00:12 GMT):
Hi there! I am a software engineer working on the Identity & Access Management workspace. I am currently working on a possible integration project with Sovrin with existing identity platforms. My first step of business included installing a test network of Indy to better understand Sovrin and its capabilities. I soon began to use this guide to start up a test network with virtualbox (https://github.com/hyperledger/indy-node/blob/master/environment/vagrant/training/vb-multi-vm/TestIndyClusterSetup.md). However, everything went well until I got to the section about connecting indy to a sandbox. When running the command 'connect sandbox' via the indy prompt, I get back 'invalid command' or 'unknown environment'. I am not sure how to create a sandbox for this of application nor do I see any documentation of how to proceed. Would someone be willing to point me in the correct direction so that I can get the sandbox all setup and a full indy test network working properly? Any help would be greatly appreciated! Thanks!

pimotte (Fri, 09 Mar 2018 07:38:05 GMT):
@Drew 42 You might be looking for https://github.com/hyperledger/indy-node/blob/master/environment/docker/pool/README.md (Linked from https://github.com/hyperledger/indy-node README)

mawi (Fri, 09 Mar 2018 09:49:56 GMT):
Hi everyone, I have a question regarding Trust Anchors. The Trust Framework whitepaper states that: "A Trust Anchor Verinym is the DID of a Trust Anchor itself. This DID is required to have at least one Public Claim asserting the Trust Anchor’s Legal Identity". Is there a convention of what this public claim should look like? Maybe even a schema?

p6g (Fri, 09 Mar 2018 14:17:55 GMT):
Has joined the channel.

malo30 (Sat, 10 Mar 2018 14:16:46 GMT):
Has joined the channel.

laoqui (Sat, 10 Mar 2018 19:13:06 GMT):
Has joined the channel.

laoqui (Sat, 10 Mar 2018 19:30:41 GMT):
hi all, is there a place that documents the genesis block and genesis transaction more in-depth? i am trying to understand how to create a genesis block and what each field in there means. thanks!

maddydev (Sat, 10 Mar 2018 23:29:28 GMT):
Has joined the channel.

JulianGordonHK (Sun, 11 Mar 2018 01:03:16 GMT):
Has joined the channel.

Srijan_Foundation (Sun, 11 Mar 2018 04:51:20 GMT):
Has joined the channel.

Srijan_Foundation (Sun, 11 Mar 2018 04:52:47 GMT):
@Sean_Bohan @Sean_Bohan Do you have a recording (that is shareable) for the the March 01 WebEx call on Indy

Srijan_Foundation (Sun, 11 Mar 2018 04:56:42 GMT):
I am looking to utilize HyperLedger implementation for tractability and visibility of charity/donations. Due to volatility of BitCoin/ Crypto, I am looking for an alternate approach where folks are able to donate directly from credit card/ bank account (payment gateway). Would be interested to collaborate with folks who may be working on similar use cases or wanting to learn together :-)

drummondreed (Mon, 12 Mar 2018 00:35:01 GMT):
@mawi My apologies for the slow response—myself, @nage, and several others were at Rebooting the Web of Trust in Santa Barbara last week. On your question about Trust Anchors, two points: 1) that provision of the Sovrin Trust Framework is evolving somewhat as the public claims may be hosted directly at the Trust Anchor's public agent endpoint, and 2) in the V1 Trust Framework (called the Sovrin Provisional Trust Framework), we didn't do a technical definition of the schema, however in the V2 Trust Framework we will be doing that as part of fully defining the Sovrin Web of Trust model. If you are interested in participating in this work, see this post on the Sovrin Forum: https://sovrin.discoursehosting.net/t/joining-the-sovrin-trust-framework-working-group/502?u=drummond

mawi (Mon, 12 Mar 2018 07:49:03 GMT):
@drummondreed That's alright, thanks for your response. This clears things up, I'll definitely take a look at the WG

tiennv (Mon, 12 Mar 2018 08:14:04 GMT):
Hi there,

tiennv (Mon, 12 Mar 2018 08:14:41 GMT):
Is there any document that explains how indy works?

mdoan (Mon, 12 Mar 2018 08:21:53 GMT):
sovrin

mdoan (Mon, 12 Mar 2018 09:16:57 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=bcAB5tGBKbXMg6rjz) @swcurran what's the benefits of being Steward? in the case i'm building a business and would like store identity of the user in sorvin, probably becoming a trust anchor would be an option

lcinacio (Mon, 12 Mar 2018 09:52:21 GMT):
hi everyone. Could you point me to documentation describing in details the structure of the domain_transactions_genesis and pool_transaction_genesis files? For example, I would like to understand what role "0" and "2" mean in the domain_transactions_genesis file...

ashcherbakov (Mon, 12 Mar 2018 11:11:05 GMT):
Hi @lcinacio. You can have a look at https://github.com/hyperledger/indy-node/blob/master/docs/transactions.md. The description of roles: https://github.com/hyperledger/indy-node/blob/master/docs/transactions.md#nym

lcinacio (Mon, 12 Mar 2018 11:41:47 GMT):
@ashcherbakov Thanks! I am going over it now...

Skorpion7777 (Mon, 12 Mar 2018 13:06:26 GMT):
Has joined the channel.

galaxy5 (Mon, 12 Mar 2018 13:50:08 GMT):
Has joined the channel.

swcurran (Mon, 12 Mar 2018 15:42:56 GMT):
@mdoan - you can look at the info in the sovrin.org/library for details, but the main benefit is putting power behind the network. In theory there is an added benefit of having faster access to the data on the ledger (since your server is operating the ledger), but I don't think that is going to be a strong motivator. The main motivation is that it is in your interest to see the network grow, and being a Steward is demonstrating a commitment to that.

ydennisy (Mon, 12 Mar 2018 17:08:05 GMT):
Has joined the channel.

jsmitchell (Tue, 13 Mar 2018 18:07:31 GMT):
Has joined the channel.

jsmitchell (Tue, 13 Mar 2018 18:09:11 GMT):
Hi Indyists - is there a document/example of how ZKPs are used for verifiable claims as mentioned on the bottom of page 23 of https://sovrin.org/wp-content/uploads/Sovrin-Protocol-and-Token-White-Paper.pdf ?

carter.willetts (Tue, 13 Mar 2018 19:47:31 GMT):
Hi All - I am using https://github.com/brycecurtis/indy-tutorial-sandbox to create a test environment, when creating the Docker image I am given 'Cloning into 'indy-anoncreds'... git-remote-https: error while loading shared libraries: /usr/lib/x86_64-linux-gnu/libasn1.so.8: invalid ELF header"... any help?

ashcherbakov (Wed, 14 Mar 2018 07:47:29 GMT):
Hi @jsmitchell All ZKP math is in https://github.com/hyperledger/indy-crypto (https://github.com/hyperledger/indy-crypto/tree/master/libindy-crypto/src/cl) This is the main paper the verifable claims math is based on: https://github.com/hyperledger/indy-crypto/blob/master/libindy-crypto/docs/AnonCred.pdf

Dan (Wed, 14 Mar 2018 08:22:04 GMT):
@ashcherbakov I wonder if @jsmitchell meant an API example using indy (rather than how does the math work)?

ashcherbakov (Wed, 14 Mar 2018 08:23:19 GMT):
@Dan https://github.com/hyperledger/indy-crypto/tree/master/libindy-crypto/src/cl is exactly the API. It' used by libindy in https://github.com/hyperledger/indy-sdk/blob/master/libindy/src/api/anoncreds.rs

ashcherbakov (Wed, 14 Mar 2018 08:24:32 GMT):
A demo test showing the anoncreds (verifiable claims) API and usage (in Python for simplicity): https://github.com/hyperledger/indy-sdk/blob/master/wrappers/python/tests/demo/test_anoncreds.py

ashcherbakov (Wed, 14 Mar 2018 08:25:04 GMT):
Anocreds sequence diagram (showing the workflow and API calls): https://github.com/hyperledger/indy-sdk/blob/master/doc/libindy-anoncreds.svg

brian038 (Wed, 14 Mar 2018 15:31:22 GMT):
Has joined the channel.

carter.willetts (Wed, 14 Mar 2018 18:41:23 GMT):
Hi All - I am attempting to create an Indy Node test network but I am having trouble connecting the sandbox Can someone offer me some troubleshooting guidance?

mboyd (Wed, 14 Mar 2018 20:41:34 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=hQicnmoyTcCkbvxYZ) @carter.willetts Hi @carter.willetts, I've recently gone through the vagrant setup with a network of test VMs. I maybe be able to offer some feedback based on what worked and didn't work for me. Can you please describe the problems you're facing in more detail?

carter.willetts (Wed, 14 Mar 2018 20:46:45 GMT):
@mboyd Hi there! Following the VM and Vagrant guide found here (https://github.com/hyperledger/indy-node/blob/stable/environment/vagrant/training/vb-multi-vm/TestIndyClusterSetup.md) When I get to the section called 'Setting Up a CLI Client and Configuring the Agents in the Indy Cluster' Once i attempt to connect to the sandbox, my terminal returns 'unknown environement sandbox' and 'unknown command connect sandbox'

endity (Thu, 15 Mar 2018 03:34:43 GMT):
Has joined the channel.

mboyd (Thu, 15 Mar 2018 03:41:28 GMT):
@carter.willetts I also ran into this bug, and I remember that I actually switched to using a local test network on the cli01 vm instance instead of doing the entire vm test network, because I mostly wanted to get to the getting started started guide and didn't care too much about the test network's implementation. If you want to follow this route, a guide for setting up a local test network is here: (https://github.com/hyperledger/indy-node/blob/stable/docs/cluster-simulation.md). If that doesn't work out, you could also try this readme: (https://github.com/hyperledger/indy-node/blob/stable/docs/indy-running-locally.md). tomorrow morning I'm going to try to get through the full vagrant test vm network walkthrough, and see if I can figure out a way to associate 'sandbox' with the virtual network that we've created. I'll let you know how it goes!

pwalimbe (Thu, 15 Mar 2018 06:44:46 GMT):
Has joined the channel.

carter.willetts (Thu, 15 Mar 2018 14:03:08 GMT):
@mboyd Great! Thank you! I am excited to hear the results!

Sean_Bohan (Thu, 15 Mar 2018 14:24:36 GMT):
TODAY we have the Indy Ledger WG Call for Thursday Mar 15, 2018 Agenda: A. Housekeeping (Sean) B. Development Achievements (Alex and Slava) C. Rebooting Web of Trust Recap (Nathan and Kyle) D. Open Discussion E. Wrap Up Please note: This meeting will be recorded Bring a friend. It will be fun. Call details: Hyperledger Indy Ledger (Node) WG Call When: Mar 15, 2018 8amPT, 9amMT, 11amET, 4pm BST Where: Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/232861185 Or iPhone one-tap (US Toll): +16465588656,232861185# or +14086380968,232861185# Or Telephone: Dial: +1 646 558 8656 (US Toll) or +1 408 638 0968 (US Toll) Meeting ID: 232 861 185 International numbers available: https://zoom.us/zoomconference?m=a0jD_rTMnh0ZYGQDOKPCNrK_0dP7WPfp1 Hyperledger Indy Community Indy Rocketchat: https://chat.hyperledger.org/channel/indyhttps://chat.hyperledger.org/channel/indy-agenthttps://chat.hyperledger.org/channel/indy-node Indy Wiki: https://wiki.hyperledger.org/projects/indy Indy WG Agendas: https://drive.google.com/open?id=1wNnp1pPS6-1Y4B2oygfI6tOU4eHGmpRH Indy Videos: https://drive.google.com/open?id=1Z8-jR7hnXb57fufE0OXxIfeUn05zNmRt Indy WG Meeting Recordings: https://drive.google.com/open?id=0B_NJV6eJXAA1UlJZMXd3cm1zNDQ

Sean_Bohan (Thu, 15 Mar 2018 14:24:36 GMT):
TODAY we have the Indy Ledger WG Call for Thursday Mar 15, 2018 Agenda: A. Housekeeping (Sean) B. Development Achievements (Alex and Slava) C. Rebooting Web of Trust Recap (Nathan and Kyle) D. Open Discussion E. Wrap Up Please note: This meeting will be recorded Bring a friend. It will be fun. Call details: Hyperledger Indy Agent (SDK) WG Call When: Mar 15, 2018 8amPT, 9amMT, 11amET, 4pm BST Where: Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/232861185 Or iPhone one-tap (US Toll): +16465588656,232861185# or +14086380968,232861185# Or Telephone: Dial: +1 646 558 8656 (US Toll) or +1 408 638 0968 (US Toll) Meeting ID: 232 861 185 International numbers available: https://zoom.us/zoomconference?m=a0jD_rTMnh0ZYGQDOKPCNrK_0dP7WPfp1 Hyperledger Indy Community Indy Rocketchat: https://chat.hyperledger.org/channel/indyhttps://chat.hyperledger.org/channel/indy-agenthttps://chat.hyperledger.org/channel/indy-node Indy Wiki: https://wiki.hyperledger.org/projects/indy Indy WG Agendas: https://drive.google.com/open?id=1wNnp1pPS6-1Y4B2oygfI6tOU4eHGmpRH Indy Videos: https://drive.google.com/open?id=1Z8-jR7hnXb57fufE0OXxIfeUn05zNmRt Indy WG Meeting Recordings: https://drive.google.com/open?id=0B_NJV6eJXAA1UlJZMXd3cm1zNDQ

Jayamine (Thu, 15 Mar 2018 14:39:55 GMT):
Has joined the channel.

eramitg (Thu, 15 Mar 2018 15:09:07 GMT):
Has joined the channel.

ashcherbakov (Thu, 15 Mar 2018 15:18:28 GMT):
PR with transactions and requests refactoring: https://github.com/hyperledger/indy-node/pull/536

burin (Thu, 15 Mar 2018 15:19:48 GMT):
Has joined the channel.

burin (Thu, 15 Mar 2018 15:20:26 GMT):
i've gone through the getting started guide using docker and everything's looking good! #accomplished

hawkmauk (Thu, 15 Mar 2018 16:08:27 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=a82ptbxNwzRribFKz) @Sean_Bohan That confused me as we don't use BST until Sunday, 25 March this year!

hawkmauk (Thu, 15 Mar 2018 16:08:27 GMT):
@Sean_Bohan That confused me as we don't use BST until Sunday, 25 March this year!

mboyd (Thu, 15 Mar 2018 16:52:28 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=RntRMBNLyqip3oFBk) @burin which method did you use to set up your test network?

burin (Thu, 15 Mar 2018 17:09:20 GMT):
it was in one of the markdown files: i did a bunch of digging and have so many tabs open :joy: let me check

burin (Thu, 15 Mar 2018 17:13:18 GMT):
@mboyd 1) i got the docker containers set up and all the nodes running, 2) set up the agents 3) walked through Alice's interaction w/ all the businesses 1) https://github.com/hyperledger/indy-node/tree/cfd1fce091991c2f2bfd667e019edc72ff8e8a72/environment/docker/pool 2) https://github.com/hyperledger/indy-node/blob/master/environment/docker/pool/StartIndyAgents.md 3) https://github.com/hyperledger/indy-node/blob/master/getting-started.md

burin (Thu, 15 Mar 2018 17:15:36 GMT):
right after that, i had a question of "how do i see the transactions that occurred on the ledger(s)?" and answered that by finding the `read_ledger` helper script. i didn't know how to use the helper scripts, but then learned to do all them via the indyclient: ``` docker exec -i -t indyclient bash ``` then ``` read_ledger --type domain ```

burin (Thu, 15 Mar 2018 17:16:14 GMT):
i could then see the transactions being added, depending on the interaction Alice had

burin (Thu, 15 Mar 2018 17:17:39 GMT):
i played w/ hyperledger composer and a tool like that was awesome to demo what hyperledger fabric was to my team. i am working on doing something similar for my team for hyperledger indy. currently the demo is all CLI, but over the next few weeks i'll add more on top

burin (Thu, 15 Mar 2018 17:19:15 GMT):
my recent expertise is in web and mobile development, but i spent some time doing ruby on rails in the past. i'm just putting together the pieces to understand how everything works right now w/ indy

burin (Thu, 15 Mar 2018 17:19:51 GMT):
i just wanted to drop in and give props to the folks involved in writing the docs and making everything relatively easy to get set up :tada:

mboyd (Thu, 15 Mar 2018 17:23:03 GMT):
@burin awesome! nice work. I'll be going through the docker tutorial in a bit, looks like it went well

burin (Thu, 15 Mar 2018 17:23:48 GMT):
yeah, it's all kinda scattered throughout, but it's there.

burin (Thu, 15 Mar 2018 17:24:17 GMT):
confusing at some points because there are many options (vagrant, etc)

mboyd (Thu, 15 Mar 2018 17:30:33 GMT):
yeah, I've actually been asked to help make the developer onboarding experience easier for indy-node and indy-sdk. All the documentation is well done, but there are a lot of options to choose from. Does anyone have an opinion on which method to set up a test indy network is simplest for someone encountering the repository for the first time?

tomislav (Thu, 15 Mar 2018 18:22:20 GMT):
I have to say docker has worked very well for me getting into Sawtooth. Being able to spint a test network, build SDK and be able to use it right away was very smooth. I would love to be able to see the same for Indy. Run a node, build the SDK and map a local folder for dev work

kdenhartog (Thu, 15 Mar 2018 19:56:23 GMT):
I really like the idea of docker, it seems to be the most OS agnostic which should make for an easier time debugging when people come with issues

kdenhartog (Thu, 15 Mar 2018 19:57:53 GMT):
@burin when you setup through docker did you have to make any changes to the docker file to get it to run? On my mac I've had quite a bit of issues getting my docker to fully get the pool running.

burin (Thu, 15 Mar 2018 20:06:25 GMT):
@kdenhartog i didn't have to change any files. the only hiccup was I had to pass in a parameter for the client start script to know the IP address it should use (10.0.0.6 or something)

burin (Thu, 15 Mar 2018 20:07:37 GMT):
i used docker for hyperledger fabric + composer too. pretty helpful for just getting things set up and poking around. i could then go in later and change stuff. but a one command setup was awesome. indy is close

kdenhartog (Thu, 15 Mar 2018 20:16:01 GMT):
yup, I see us as being close as well which is where I want to see us going. I also realized that docker sets up quite a few docker images that aren't labeled properly so a few tweaks there would make it so I know what every image is doing.

burin (Thu, 15 Mar 2018 22:42:16 GMT):
i've been trying to understand what gets recorded to the ledger and what doesnt, and feel like i have a pretty good grasp now (looking at sequence diagrams, getting local nodes set up and running through some sequences while spying on the ledger) i still have some conceptual questions though:

burin (Thu, 15 Mar 2018 22:43:15 GMT):
there's this idea that people could know when businesses use/share their information (if that's even possible). would something like that be recorded on the ledger?

burin (Thu, 15 Mar 2018 22:44:58 GMT):
i see that a pairwise did is added to the ledger when you interact w/ a verifier/issuer. that's kinda like a chime when you walk in the door at a business. but is there a record of the "proof" happening?

amundson (Thu, 15 Mar 2018 23:27:37 GMT):
+1 for docker and docker-compose

kdenhartog (Thu, 15 Mar 2018 23:55:47 GMT):
@burin the majority of cases would only put a DID on the ledger and then when a claim is issued a number representing whether a claim has been revoked goes on the ledger. Then when a holder wants to prove to a verifier that they have a claim, they create a predicate zero knowledge proof indicating that they possess the claim AND that it hasn't been revoked (this is proven by verifying the number still exists within the accumulator on the ledger)

burin (Fri, 16 Mar 2018 00:00:09 GMT):
@kdenhartog so in the sequence of interacting w/ "faber college", there is the initial DID put on the ledger that represents the alice + faber relationship... and is there another one once the "transcript" claim is picked up?

burin (Fri, 16 Mar 2018 00:02:42 GMT):
ok i see those

burin (Fri, 16 Mar 2018 00:04:11 GMT):
so when i as a holder take that claim, use it to create a proof, and send it over to the verifier, no new entry is made on the ledger. they verify it, and can also hold onto that data. and look at it later?

burin (Fri, 16 Mar 2018 00:05:07 GMT):
i could revoke it, or faber college could revoke it. which means it's invalid, but the verifier could still hold onto the data and view/use it?

burin (Fri, 16 Mar 2018 00:05:53 GMT):
i'm trying to understand whether there is a mechanism built in that allows me as the holder to know whether the information sent in the proof is "accessed" beyond that initial interaction

kdenhartog (Fri, 16 Mar 2018 00:06:29 GMT):
correct, to revocation question.

kdenhartog (Fri, 16 Mar 2018 00:06:58 GMT):
I'm not sure what you mean by accessed, do you mean that a relying party has viewed it?

burin (Fri, 16 Mar 2018 00:07:07 GMT):
yes

burin (Fri, 16 Mar 2018 00:08:28 GMT):
i've heard some folks mention a benefit of "having identity on the blockchain" COULD be that "you could know who and when people are accessing your data"

burin (Fri, 16 Mar 2018 00:08:48 GMT):
indy's implementation may not be the case though, which i understand

burin (Fri, 16 Mar 2018 00:08:48 GMT):
indy's implementation may not support that though, which i understand

kdenhartog (Fri, 16 Mar 2018 00:10:55 GMT):
currently no this wouldn't be possible. I'm not sure how you could do what you mentioned because anyone could spin up their own node (in a permissionless system) and query their own node without writing to the ledger that they queried it.

burin (Fri, 16 Mar 2018 00:11:30 GMT):
yeah, that's what i'm confirming

kdenhartog (Fri, 16 Mar 2018 00:11:33 GMT):
I suppose there's ways that you could force all queries to be recorded on ledger, but it would slow down the system signaficantly I would presume.

burin (Fri, 16 Mar 2018 00:14:36 GMT):
i'm not sure if that's even a hole or issue that needs to be addressed. in the physical world, when you give someone your ID, you get it back. and if you watch them the whole time, you can see if they make a copy of it

burin (Fri, 16 Mar 2018 00:17:38 GMT):
for liability reasons the verifiers may not even want to hold onto that data ( i wouldn't want to hold onto it). i might just be thinking in old school ways, where you save the info so you can compare it again later. but w/ the proofs, you don't need to

kdenhartog (Fri, 16 Mar 2018 02:09:24 GMT):
With an Anonymous credential there's ways that the holder wouldn't have to reveal the information completely, but could reveal information about it. So for example, I can generate a proof that proves I'm over 21, but don't have to reveal the exact birthday. This allows for the holder to say they don't want to reveal the information the RP is asking for, but they will provide proof of something that is more appropriate. On the other hand, a RP who gathers this data will either 1 be kicked out of the market because they are asking for too much data (ideally) or they will adapt their practices.

burin (Fri, 16 Mar 2018 02:37:35 GMT):
@kdenhartog thanks for talking through this. helps to have someone to bounce ideas off of and confirm assumptions/doubts. definitely requires a change in thought!

kdenhartog (Fri, 16 Mar 2018 02:38:08 GMT):
no problem, I'm glad I could help! feel free to ask any questions in here any time and someone will help.

burin (Fri, 16 Mar 2018 02:39:52 GMT):
an airline for example, doesn't really need to ask for all the birthday, name, secure traveler type of info. (most just scan your boarding pass at the gate anyway). so in the future, they could sell a plane ticket with very minimal info

yopep (Fri, 16 Mar 2018 05:59:49 GMT):
Has joined the channel.

pimotte (Fri, 16 Mar 2018 07:50:31 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=bChzcY7sq5LPpSfAa) @mboyd I agree docker was the easiest to set up. Making absolutely clear that the ips of machines matter is also key for smooth onboarding, as that was the most non-trivial part for me.

burin (Fri, 16 Mar 2018 13:43:56 GMT):
i'm diagramming what the flow would be for what it would be like to have vaccination records for travel on indy. i have two diagrams, one shorter less technically accurate one, and a longer one. the shorter one is used to give a high level understanding of what's going on. anyone want to take a look?

burin (Fri, 16 Mar 2018 13:48:31 GMT):

short but not 100% accurate

burin (Fri, 16 Mar 2018 14:20:59 GMT):

longer version

ashcherbakov (Fri, 16 Mar 2018 15:00:43 GMT):
Hi @burin. Did you have a look at https://github.com/hyperledger/indy-sdk/blob/master/doc/libindy-anoncreds.svg? It looks quite similar to your diagrams, although this is rather abstract

burin (Fri, 16 Mar 2018 15:05:33 GMT):
@ashcherbakov yep, i dug through a lot of the diagrams. they're at varying levels of detail and really helpful. i wanted to confirm that my high level understanding was mostly correct by applying it to this specific scenario. it's definitely a rabbit hole 😂

burin (Fri, 16 Mar 2018 15:08:00 GMT):
@ashcherbakov the one you sent is good reference, because it includes the revocation stuff. which some diagrams don't have

ashcherbakov (Fri, 16 Mar 2018 15:09:27 GMT):
@burin 1) The current workflow (in master) assumes that Issuer needs to send 'Offer Credential' before the Traveler (Prover) sends 'Request Credential' 2) Traveler (Prover) NEVER send's credentials to the Verifier (Inspector), this is very essential. The Prover sends a *proof* created from credentials on the Prover side. Credntials are private and located in the Prover's wallet only.

ashcherbakov (Fri, 16 Mar 2018 15:09:27 GMT):
@burin 1) The current workflow (in master) assumes that Issuer needs to send 'Offer Credential' before the Traveler (Prover) sends 'Request Credential' 2) Traveler (Prover) NEVER send's credentials to the Verifier (Inspector), this is very essential. The Prover sends a *proof* created from credentials on the Prover side. Credntials are private and located in the Prover's wallet only.This is what Zero-knowledge proof is about.

burin (Fri, 16 Mar 2018 15:11:25 GMT):
ok good clarification

ashcherbakov (Fri, 16 Mar 2018 15:11:48 GMT):
3) Step 0: Issuer needs to save Public Keys (needed to sign claims, that's issue credentials) on the Blockchain

ashcherbakov (Fri, 16 Mar 2018 15:13:00 GMT):
Please note that this is not the same key as used for pairwise relationship (ed25519 keys are used for pairwise, while CL keys (our anoncreds math) are used for credentials)

burin (Fri, 16 Mar 2018 15:13:17 GMT):
the #1 thing: i see in the "getting started" diagram where "Offer Credential" happens before "Request Credential", but there's an implicit step where the Traveler would initiate. kinda like physically going into the office. then at the desk, the Issuer "Offers Credentials"

burin (Fri, 16 Mar 2018 15:14:55 GMT):
or is that basically the connection that's established?

burin (Fri, 16 Mar 2018 15:15:46 GMT):
i'm trying to relate these into concepts that are easy to understand, because i have to explain it to non-tech folks 😀

ashcherbakov (Fri, 16 Mar 2018 15:53:34 GMT):
I think it's still an open question whether Claim Offer is required or optional step. An example of Claim Offer: Alice graduates, and the University sends Claim Offer that her Certificate is ready to be issued. Alice replied (with a cred request) that yes, she wants to have her Certificate issued, The University issues the credential with Certificate. So, in this example the workflow is initialized by the Issuer (Claim Offer). If the workflow needs to be initialized by the Prover (Cred Holder), then the Prover (Alice) may send step-0 Request for Claim Offer.

burin (Fri, 16 Mar 2018 16:05:17 GMT):
that makes sense in the Vaccination example too: like if you had a vaccine many years ago, the Health Services org could send the Claim Offer for a vaccination record, similar to a transcript

burin (Fri, 16 Mar 2018 16:06:48 GMT):
if you JUST got the vaccine today, it could work the same way, just without the big gap in time before the Claim Offer is sent

mboyd (Fri, 16 Mar 2018 22:26:12 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=BAWYH7xzjB8RGeenr) @pimotte can you please send me the link of the readme that worked for you?

wangdong (Mon, 19 Mar 2018 01:47:58 GMT):
I try to run the test cases and I got some errors. So I run pytest --fixtures and got errors **ImportError: cannot import name 'signus'**

wangdong (Mon, 19 Mar 2018 01:48:30 GMT):
I installed libindy with compilation.

wangdong (Mon, 19 Mar 2018 01:48:55 GMT):
Maybe I missed something else?

wangdong (Mon, 19 Mar 2018 02:45:54 GMT):
E _pytest.config.ConftestImportFailure: ImportError("cannot import name 'signus'",) E File "/usr/local/lib/python3.5/dist-packages/_pytest/assertion/rewrite.py", line 213, in load_module E py.builtin.exec_(co, mod.__dict__) E File "/root/indy-node/acceptance/indy_acceptance/test/conftest.py", line 8, in E from indy_acceptance.test.client import Client E File "/root/indy-node/acceptance/indy_acceptance/test/client.py", line 3, in E from indy import ledger, signus

wangdong (Mon, 19 Mar 2018 02:45:54 GMT):
E _pytest.config.ConftestImportFailure: ImportError("cannot import name 'signus'",) E File "/usr/local/lib/python3.5/dist-packages/_pytest/assertion/rewrite.py", line 213, in load_module E py.builtin.exec_(co, mod.__dict__) E File "/root/indy-node/acceptance/indy_acceptance/test/conftest.py", line 8, in E from indy_acceptance.test.client import Client E File "/root/indy-node/acceptance/indy_acceptance/test/client.py", line 3, in E **from indy import ledger, signus**

dinesh.rivankar (Mon, 19 Mar 2018 04:33:00 GMT):
Has joined the channel.

wangdong (Mon, 19 Mar 2018 06:45:57 GMT):
And I did not find any module named signus in wrapper/python/indy

wangdong (Mon, 19 Mar 2018 06:46:31 GMT):
So I guess either I missed something or it is a bug?

wangdong (Mon, 19 Mar 2018 07:47:44 GMT):
it seems that module 'signus' replaced by 'did'

zhaoyulong (Mon, 19 Mar 2018 08:33:44 GMT):
Has joined the channel.

dinesh.rivankar (Mon, 19 Mar 2018 08:37:33 GMT):
Hi , we have installed Indy on Ubuntu system ,however getting below error generate_indy_pool_transactions: command not found

ashcherbakov (Mon, 19 Mar 2018 12:43:13 GMT):
@wangdong It looks like that the tests in acceptance folder are based on the stable version of libindy (which is quite old). These tests are not part of our CI. The tests that are part of the CI are the *test* folders in package (for example https://github.com/hyperledger/indy-plenum/tree/master/plenum/test). These tests are based on the master libindy, and yes, master libindy has some incompatible changes with the latest stable. In particular, `signus` was renamed to `did`.

wangdong (Mon, 19 Mar 2018 12:52:32 GMT):
I cloned the indy-node from github. If it is an old version, should I clone it as well as other repo with --recursive option?

wangdong (Mon, 19 Mar 2018 12:53:30 GMT):
I just want to verify if the test cases of the repos can work in our OS.

wangdong (Mon, 19 Mar 2018 12:57:57 GMT):
I wonder how I can find all the test cases of CI.

wangdong (Mon, 19 Mar 2018 12:58:00 GMT):
@ashcherbakov

wangdong (Mon, 19 Mar 2018 13:04:19 GMT):
CI test will include all the test cases in the independent repos(indy-node, indy

ashcherbakov (Mon, 19 Mar 2018 13:07:47 GMT):
As I said, you can have a look at all `test` folder in all submodules (a default path for tests): https://github.com/hyperledger/indy-plenum/tree/master/plenum/test https://github.com/hyperledger/indy-plenum/tree/master/stp_zmq/test https://github.com/hyperledger/indy-plenum/tree/master/state/test https://github.com/hyperledger/indy-plenum/tree/master/ledger/test The same for indy-node: https://github.com/hyperledger/indy-node/tree/master/indy_client/test https://github.com/hyperledger/indy-node/tree/master/indy_node/test https://github.com/hyperledger/indy-node/tree/master/indy_common/test

wangdong (Mon, 19 Mar 2018 13:11:54 GMT):
I think there are also https://github.com/hyperledger/indy-crypto/tree/master/wrappers/python/tests/bls and https://github.com/hyperledger/indy-anoncreds/tree/master/anoncreds/test

wangdong (Mon, 19 Mar 2018 13:12:06 GMT):
are also part of the CI

ashcherbakov (Mon, 19 Mar 2018 13:22:47 GMT):
I was talking about indy-node (and plenum) parts only. There are tests in indy-sdk and indy-crupto as well. indy-anoncreds is deprecated.

wangdong (Mon, 19 Mar 2018 13:35:56 GMT):
OK

wangdong (Mon, 19 Mar 2018 13:36:05 GMT):
I need them all.

wangdong (Mon, 19 Mar 2018 13:36:14 GMT):
thanks for your reply

kevinmcmahon (Mon, 19 Mar 2018 17:41:58 GMT):
Has joined the channel.

Skorpion7777 (Mon, 19 Mar 2018 17:53:07 GMT):
Hey everyone, since you helped me out a lot with better understanding claims I thought I might ask a question generall to permissioned chains. Does someone knows any papers on game theory analysis for permissioned chains? Sure you could say that in a blockchain its easily visible that someone is behaving faulty and in a permissioned network you can exclude that person - but if someone knows a more scientific approach to this topic that would be great!

ck0316 (Tue, 20 Mar 2018 03:44:05 GMT):
Has joined the channel.

dinesh.rivankar (Tue, 20 Mar 2018 05:17:24 GMT):
Hi Guys , i have already connect to the sandbox,however when i accept the connection request i get below response Dinesh@sandbox> accept request from Faber Expanding Faber to "Faber College" Request not yet verified. Connection not yet synchronized. Request acceptance aborted. Cannot sync because not connected. Please connect first. Usage: connect

dinesh.rivankar (Tue, 20 Mar 2018 05:17:24 GMT):
Hi Guys , i have already connect to the sandbox,however when i accept the connection request i get below response ALICE@sandbox> accept request from Faber Expanding Faber to "Faber College" Request not yet verified. Connection not yet synchronized. Request acceptance aborted. Cannot sync because not connected. Please connect first. Usage: connect

pimotte (Tue, 20 Mar 2018 07:57:42 GMT):
@mboyd I ended up using the indy-sdk ci setup. Bring up the docker pool with step 4: https://github.com/hyperledger/indy-sdk/blob/master/doc/ubuntu-build.md and then running the code samples worked

bafonins (Tue, 20 Mar 2018 13:37:54 GMT):
Has joined the channel.

andrewtan (Tue, 20 Mar 2018 14:29:05 GMT):
nstall

andrewtan (Tue, 20 Mar 2018 14:32:58 GMT):
Hi everyone, I’m following the indy install process, as documented here (https://github.com/hyperledger/indy-node/blob/master/environment/vagrant/training/vb-multi-vm/TestIndyClusterSetup.md). When I come to ssh into the cli01 host, and run “indy”, the “indy” binary is not found. In fact, the “indy” binary itself doesn’t seem to be installed on cli01 or any of the validator nodes. Appreciate if anyone can advice where I have gone wrong?

andrewtan (Tue, 20 Mar 2018 14:55:30 GMT):
In running the step Vagrant Up, have observed the following message.

andrewtan (Tue, 20 Mar 2018 14:55:37 GMT):

Clipboard - March 20, 2018 10:55 PM

ashcherbakov (Tue, 20 Mar 2018 15:16:23 GMT):
@andrewtan This is a bug that we are planning to fix this sprint. There is a workaround for now: https://forum.sovrin.org/t/error-when-running-vagrant-up/582

ShikarSharma (Tue, 20 Mar 2018 22:47:43 GMT):
Has joined the channel.

rekwet (Wed, 21 Mar 2018 08:45:36 GMT):
Has joined the channel.

secretsayan (Wed, 21 Mar 2018 12:50:25 GMT):
Has joined the channel.

the_identity_guy (Wed, 21 Mar 2018 15:52:37 GMT):
@andrewtan I am a beginner too but I had a similar issue sometime ago and had to repovision the vm to fix it, make sure all the dependencies and steps are completed..i.e. all the libraries are downloaded and no error is shown during provisioning

the_identity_guy (Wed, 21 Mar 2018 15:52:37 GMT):
@andrewtan I am a beginner too but I had a similar issue sometime ago and had to repovision the vm to fix it, make sure all the dependencies and steps are completed..i.e. all the libraries are downloaded and no error is shown during provisioning.. also make sure you have enough step and connection to download them

the_identity_guy (Wed, 21 Mar 2018 15:52:37 GMT):
@andrewtan I am a beginner too but I had a similar issue sometime ago and had to repovision the vm to fix it, make sure all the dependencies and steps are completed..i.e. all the libraries are downloaded and no error is shown during provisioning.. also make sure you have enough space and connection to download them

mboyd (Wed, 21 Mar 2018 19:42:59 GMT):
A pull request has been submitted for this issue, and was merged this morning. The vagrant VM test network should run smoothyl now

schristie (Wed, 21 Mar 2018 20:56:30 GMT):
why are pairwise DIDs published to the ledger? are there any documents explaining the logic behind that?

swcurran (Wed, 21 Mar 2018 23:11:50 GMT):
@schristie - I believe the plan is that they will not be; they will just be exchanged between the entities in the relationship.

kdenhartog (Wed, 21 Mar 2018 23:28:54 GMT):
@ashcherbakov what version of python3-indy should I be running the getting-started guide with. It seems to not work when I pull it down with pip, so I tried changing to 1.3.1.dev410 and that is missing some of the calls.

sergey.minaev (Thu, 22 Mar 2018 08:58:08 GMT):
Has joined the channel.

sergey.minaev (Thu, 22 Mar 2018 09:02:44 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=bcykc3SrqGexsgkxG) @kdenhartog 1) it's recommended to use docker-compose for getting-started guide, please see `doc/getting-started/run-getting-started.md` 2) in general, it's highly recommended to use same version of python3-indy wrapper and libindy itself

sergey.minaev (Thu, 22 Mar 2018 09:02:44 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=bcykc3SrqGexsgkxG) @kdenhartog 1) it's recommended to use docker-compose for getting-started guide, please see `doc/getting-started/run-getting-started.md` https://github.com/hyperledger/indy-sdk/blob/master/doc/getting-started/run-getting-started.md 2) in general, it's highly recommended to use same version of python3-indy wrapper and libindy itself

sergey.minaev (Thu, 22 Mar 2018 09:05:08 GMT):
As I can see from dockerfile for GSG last tested version is 1.3.1-dev-398 . Latest master (418) most probably should also work, but this fact isn't tested.

sergey.minaev (Thu, 22 Mar 2018 09:05:08 GMT):
As I can see from dockerfile for GSG last tested version is 1.3.1-dev-398 . Latest master (418) should also works, but this fact isn't tested.

ydennisy (Thu, 22 Mar 2018 09:34:05 GMT):
hello all, sorry new here so quite a basic and broad question... I was wondering if there are plans to integrate composer style functionality with indy? Also to have the auto created HTTP API same as with Fabric? Thanks in advance.

Sean_Bohan (Thu, 22 Mar 2018 13:11:17 GMT):
Hyperledger Indy WG Call AGENDA: Housekeeping (Sean) Development Status (Alex and Slava) European MyEnergyData Project Discussion (Harri & Peter) Opening Up the Sprint Process (Sean and Nathan) Wrap Up When: Mar 22, 2018 8amPT, 9amMT, 11amET, 3pm GMT Where: Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/232861185 Or iPhone one-tap (US Toll): +16465588656,232861185# or +14086380968,232861185# Or Telephone: Dial: +1 646 558 8656 (US Toll) or +1 408 638 0968 (US Toll) Meeting ID: 232 861 185 International numbers available: https://zoom.us/zoomconference?m=a0jD_rTMnh0ZYGQDOKPCNrK_0dP7WPfp1 Hyperledger Indy Community Indy Rocketchat: https://chat.hyperledger.org/channel/indyhttps://chat.hyperledger.org/channel/indy-agenthttps://chat.hyperledger.org/channel/indy-node Indy Wiki: https://wiki.hyperledger.org/projects/indy Indy WG Agendas: https://drive.google.com/open?id=1wNnp1pPS6-1Y4B2oygfI6tOU4eHGmpRH Indy Videos: https://drive.google.com/open?id=1Z8-jR7hnXb57fufE0OXxIfeUn05zNmRt Indy WG Meeting Recordings: https://drive.google.com/open?id=0B_NJV6eJXAA1UlJZMXd3cm1zNDQ

Sean_Bohan (Thu, 22 Mar 2018 15:27:24 GMT):
https://github.com/hyperledger/indy-sdk/tree/master/doc/how-tos/send-secure-msg/python

saholman (Thu, 22 Mar 2018 16:01:49 GMT):
Has joined the channel.

stevenmilstein (Thu, 22 Mar 2018 16:02:16 GMT):
Has joined the channel.

kdenhartog (Thu, 22 Mar 2018 17:00:21 GMT):
@sergey.minaev sounds good, I'll report back. I tried running it originally using jupyter, but encountered errors, so I tried setting it up using the python package and was running into issues still. Here's the stack trace of what I got in jupyter

kdenhartog (Thu, 22 Mar 2018 17:00:23 GMT):
--------------------------------------------------------------------------- RuntimeError Traceback (most recent call last) in () 802 if __name__ == '__main__': 803 loop = asyncio.get_event_loop() --> 804 loop.run_until_complete(run()) 805 time.sleep(1) # FIXME waiting for libindy thread complete /usr/lib/python3.5/asyncio/base_events.py in run_until_complete(self, future) 373 future.add_done_callback(_run_until_complete_cb) 374 try: --> 375 self.run_forever() 376 except: 377 if new_task and future.done() and not future.cancelled(): /usr/lib/python3.5/asyncio/base_events.py in run_forever(self) 338 self._check_closed() 339 if self.is_running(): --> 340 raise RuntimeError('Event loop is running.') 341 self._set_coroutine_wrapper(self._debug) 342 self._thread_id = threading.get_ident() RuntimeError: Event loop is running.

kdenhartog (Thu, 22 Mar 2018 17:30:41 GMT):
looks like error 306 means the config pool has already been setup. trying to figure out where the files are within jupyter so I can remove them.

matthewehoward (Thu, 22 Mar 2018 22:17:33 GMT):
Has joined the channel.

kdenhartog (Fri, 23 Mar 2018 00:51:36 GMT):
For those in this channel who are interested in the Hyperledger Indy internship: I've gotten quite a bit of questions about what the project will look like so they can address certain aspects of the application. I apologize for addressing these questions so late (1 day before application deadline), but I wanted to provide a little clarification. First, the goal of the IPython Notebooks is to build short use cases similar to the "getting started" guide which will walk people through understand concepts in Hyperledger Indy, and show some basic use cases such as how to create, manage, and update a DID. We ideally would like to see a candidate be able to create their own use cases for this in the application and the project would involve actual implementation of this with the help of your mentors. For the DID-Auth work, we are still in discussions as to what this project will look like, but ideally this will become a replacement to traditional authentication (username/passwords) or integrate with systems that are familiar (OAuth/SSO). DID-Auth is an integral idea that is in process, so we would expect for you to be first get an understanding of the ideas, and build upon the current implementations that have been completed. Links to work related to DID Auth will be provided in the next comment.

kdenhartog (Fri, 23 Mar 2018 00:54:14 GMT):
DID Auth implementation: https://github.com/bcgov/did-auth-extension DID-Auth related paper: https://github.com/WebOfTrustInfo/rebooting-the-web-of-trust-spring2018/blob/master/draft-documents/did_auth_draft.md

PrashantS (Fri, 23 Mar 2018 03:15:04 GMT):
Has joined the channel.

trthhrtz (Fri, 23 Mar 2018 03:33:01 GMT):
Has joined the channel.

sergey.minaev (Fri, 23 Mar 2018 11:43:46 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=6iw7XReW5bk2CFF6k) @kdenhartog yes, 306 error means attempt to create already existing pool configuration. You can clean-up libindy home directory. It's folder `.indy_client` in home user directory

kdenhartog (Fri, 23 Mar 2018 12:46:55 GMT):
Okay, I wasn't sure if the jupyter system used the local directory too, or had it's own somehow. I did try renaming just the pool and then got another error which is probably wallet already exists error. Thanks @sergey.minaev I'll give these fixes a try.

hkrichen (Sat, 24 Mar 2018 08:42:23 GMT):
Has joined the channel.

paul.sitoh (Sun, 25 Mar 2018 19:00:16 GMT):
Has joined the channel.

Vinaykumarpaspula (Sun, 25 Mar 2018 21:01:22 GMT):
Has joined the channel.

the_identity_guy (Sun, 25 Mar 2018 23:27:41 GMT):
Is there a doc on Indy's Idemix implementation and how it defers from the original IBM Idemix? Is the major difference between the two the revocation process based on accumulators?

the_identity_guy (Mon, 26 Mar 2018 02:57:08 GMT):
Hi, Is there any documentation on identity hubs?

pimotte (Mon, 26 Mar 2018 06:23:35 GMT):
@the_identity_guy There's a fairly low-level description at https://github.com/hyperledger/indy-crypto/blob/master/libindy-crypto/docs/AnonCred.pdf

pimotte (Mon, 26 Mar 2018 06:23:58 GMT):
If you find anything that actually describes the differences, I'd be interested as well

Kelattar (Mon, 26 Mar 2018 08:00:34 GMT):
Has joined the channel.

Kelattar (Mon, 26 Mar 2018 08:03:58 GMT):
Hello ! I am trying to install indy-dev-node in order to deploy indy nodes and test the platform but it has been a couple of days that I'm facing a problem while installing rocksdb, I get this error : "x86_64-linux-gnu-g++ -pthread -shared -Wl,-O1 -Wl,-Bsymbolic-functions -Wl,-Bsymbolic-functions -Wl,-z,relro -Wl,-Bsymbolic-functions -Wl,-z,relro -g -fstack-protector-strong -Wformat -Werror=format-security -Wdate-time -D_FORTIFY_SOURCE=2 build/temp.linux-x86_64-3.5/rocksdb/_rocksdb.o -lrocksdb -lsnappy -lbz2 -lz -llz4 -o build/lib.linux-x86_64-3.5/rocksdb/_rocksdb.cpython-35m-x86_64-linux-gnu.so /usr/bin/ld: cannot find -llz4 collect2: error: ld returned 1 exit status error: command 'x86_64-linux-gnu-g++' failed with exit status 1", "Command "/home/indy/.virtualenvs/venv-test/bin/python3.5 -u -c "import setuptools, tokenize;__file__='/tmp/pip-build-qe5rxchw/python-rocksdb/setup.py';f=getattr(tokenize, 'open', open)(__file__);code=f.read().replace('\r\n', '\n');f.close();exec(compile(code, __file__, 'exec'))" install --record /tmp/pip-agj1euyi-record/install-record.txt --single-version-externally-managed --compile --install-headers /home/indy/.virtualenvs/venv-test/include/site/python3.5/python-rocksdb" failed with error code 1 in /tmp/pip-build-qe5rxchw/python-rocksdb/", can anyone help me ?

DurgaRao641 (Mon, 26 Mar 2018 08:51:27 GMT):
Has joined the channel.

DurgaRao641 (Mon, 26 Mar 2018 08:51:53 GMT):
We have installed indy using vagrant, and tried to run "indy" command in a client and we found the following issue:Loading module /usr/local/lib/python3.5/dist-packages/config/config-crypto-example1.py Module loaded. Illegal instruction (core dumped) Please help me with this

saurabharora (Mon, 26 Mar 2018 09:13:50 GMT):
Has joined the channel.

pknowles (Mon, 26 Mar 2018 11:12:50 GMT):
Has joined the channel.

pknowles (Mon, 26 Mar 2018 11:13:10 GMT):
Hello Hyperledger Indy community - Following a "suggestion for progression" email from one of the Evernym directors, I'm bringing the following concept to the table. This concept has already been vetted by Evernym and we collectively believe that we’re looking at a game-changer in the world of data analytics. Creating a Sovrin-linked public permissionless ledger and accompanying cloud based data warehousing solution to store GDPR-compliant data tables aligned to Sovrin-built “ghost" schemas would enable developers to build algorithms and decentralised applications on pre-verified data without compromising self-sovereign identity. Imagine the benefit to a postdoctoral researcher conducting research for a groundbreaking thesis or enabling a startup company to build industry-specific algorithmic automation processes to better serve their community. All data stored in the schema-related data tables would already be verified by trusted stewards on the Sovrin platform. The linked ledger would aim to become the preferred DLT platform for worldwide societal analytics and applications. Programmers, coders and developers would be able to build open source statistics and analytics applications for societal, corporate and personal benefit. The concept deck has been drafted and I’m now bringing it to the Hyperledger Indy community to see what people think. This new proposed build would need to be built methodically and with real expert input from now until final implementation. From Evernym: "Hi Paul - this is starting to look good. My suggestion for progression is for you to get involved in the Sovrin community at sovrin.slack.com, or in the Hyperledger Indy rocketchat. You've got an all encompassing vision so when you introduce it to the community you might want to go back up to 50,000 feet to articulate the problem you're trying to solve, and see what people think.” The working title of the concept deck … "Building a Linked Public Permissionless Ledger - 'the future of data analytics’ " Please get back to me, if you're interested to hear more.

pknowles (Mon, 26 Mar 2018 12:20:58 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=vGN4bgJbBwS8hcs6N) Think Google Analytics but totally open-source. The foundation behind the initiative would be a private-sector, international non-profit.

pknowles (Mon, 26 Mar 2018 12:51:53 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=vGN4bgJbBwS8hcs6N) Think Google Analytics but totally open-source. The foundation behind the initiative would be a private-sector, international non-profit.

ashcherbakov (Mon, 26 Mar 2018 13:03:17 GMT):
@Kelattar @DurgaRao641 Sorry, we are in the middle of supporting rocksdb build on master. It was broken for a couple of days, it should be fine again today. As a workaround, you can use stable version of Indy-node for a while.

the_identity_guy (Mon, 26 Mar 2018 18:37:32 GMT):
Is there any documents on Indy Identity hubs and how they work?

nage (Mon, 26 Mar 2018 20:42:36 GMT):
We call them Agents, and there are several presentations and docs floating around, but much of this part of the system is still being decided (most developers are building their own agent on top of the SDK—but we all agree we want an open source agent we can share). Join the discussion in #indy-agent if you’d like to help.

the_identity_guy (Mon, 26 Mar 2018 23:29:17 GMT):
@nage there seems to be a distinction between agents and hubs. agents seem to represent an identity owner where as hubs store the identity data of many identity owners... This doc for example references both separately https://github.com/WebOfTrustInfo/rebooting-the-web-of-trust-spring2018/blob/master/draft-documents/Identity%20Hub%20Attestation%20Handling.md

pknowles (Tue, 27 Mar 2018 02:51:21 GMT):
I've been told that the channel is the best place to share new concepts for th is Hyperledger Indy community. Enjoy. "Building a Linked Public Permissionless Ledger" https://we.tl/9hzS84CMDq

pknowles (Tue, 27 Mar 2018 02:57:38 GMT):
I've been told that the purpose of the Hyperledger Indy community is to take great ideas and bring minds together to make new stuff work. On that note, here is a download link for the concept deck that I mentioned previously. I'm intrigued to know your thoughts. https://we.tl/9hzS84CMDq

sunilbarve (Tue, 27 Mar 2018 09:39:36 GMT):
Has joined the channel.

sunilbarve (Tue, 27 Mar 2018 09:43:01 GMT):
I have created locally 4 node cluster on Ubuntu machine. Also clone and run a 'getting_stated.py' from Indy_sdk/samples/python/src getting the error 'undefined symbol: indy_crypto_anon_crypt'.

sunilbarve (Tue, 27 Mar 2018 09:43:01 GMT):
I have created locally 4 node cluster on Ubuntu machine. Also clone and run a 'getting_stated.py' from Indy_sdk/samples/python/src getting the error 'undefined symbol: indy_crypto_anon_crypt'. can anyone help?

DurgaRao641 (Tue, 27 Mar 2018 09:43:53 GMT):
Thanks @ashcherbakov, In the mean while we have tried with the docker installation and we are able to create the four nodes and indyclient(node1, .. node4 and indyclient) with the given scripts. Now we are doing the indy tutorial(https://github.com/hyperledger/indy-node/blob/stable/environment/docker/pool/StartIndyAgents.md), In that when we run "indy" command in indyclient it is running and taking the commands but not able to see what we are typing. Also the document suggested to connect with the test network(connect test), when we tried that we got an issue "Unknown environment test". Where actually this test network will be created? Please help with this also. Thanks in advance.

sergey.minaev (Tue, 27 Mar 2018 10:04:25 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=TnKasar2DgvjCQrph) @sunilbarve please see my comment above

sergey.minaev (Tue, 27 Mar 2018 10:06:42 GMT):
@gudkov @ashcherbakov It's second request about how to run getting started. Looks like we have not obvious naming.

sergey.minaev (Tue, 27 Mar 2018 10:06:42 GMT):
@gudkov @ashcherbakov It's second request in a short time about how to run getting started. Looks like we have not obvious naming.

ashcherbakov (Tue, 27 Mar 2018 10:14:48 GMT):
@DurgaRao641 I believe you should type `connect sandbox` BTW the Getting Started you are trying is an old one, and will be deprecated soon (we have a number of tickets for mentioning this more explicitly). The new Getting Started is based on libindy: https://github.com/hyperledger/indy-sdk/tree/master/doc/getting-started

sunilbarve (Tue, 27 Mar 2018 10:23:38 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=uNdrY7iojFJsZ4ehs) @sergey.minaev

sunilbarve (Tue, 27 Mar 2018 10:23:38 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=uNdrY7iojFJsZ4ehs) @sergey.minaev thank you,i wl check that.

DurgaRao641 (Tue, 27 Mar 2018 10:31:24 GMT):
Thanks @ashcherbakov, Yes it is "connect sanbox". We will also follow your latest document. Thanks.

stevenmilstein (Tue, 27 Mar 2018 13:40:11 GMT):
Regarding Getting Started https://github.com/hyperledger/indy-node/blob/master/getting-started.md and the the Vagrant https://github.com/hyperledger/indy-node/blob/master/environment/vagrant/training/vb-multi-vm/TestIndyClusterSetup.md#setting-up-a-cli-client-and-configuring-the-agents-in-the-indy-cluster, how do I determine the ? Any help is greatly appreciated.

ashcherbakov (Tue, 27 Mar 2018 14:39:50 GMT):
@stevenmilstein The network name will be `sandbox`

ashcherbakov (Tue, 27 Mar 2018 14:39:58 GMT):
BTW the Getting Started you are trying is an old one, and will be deprecated soon (we have a number of tickets for mentioning this more explicitly). The new Getting Started is based on libindy: https://github.com/hyperledger/indy-sdk/tree/master/doc/getting-started

stevenmilstein (Tue, 27 Mar 2018 16:01:15 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=PvEwoMMgcncQnrKwE) @ashcherbakov Thanks for the quick reply. FYI. I tried `sandbox` & received ``` vagrant@agent01:~$ python /usr/local/lib/python3.5/dist-packages/indy_client/test/agent/faber.py --port 5555 --network sandbox File "/usr/local/lib/python3.5/dist-packages/indy_client/test/agent/faber.py", line 99 async def bootstrap_faber(agent): ^ SyntaxError: invalid syntax ``` I'll drop it and move to the new link. Thanks

sunilbarve (Wed, 28 Mar 2018 10:04:09 GMT):
Getting error in cargo build. error: failed to run custom build command for `zmq-sys v0.8.2` process didn't exit successfully: `/home/bcsadmin/indy-sdk/libindy/target/debug/build/zmq-sys-5993ac7ad3aa9f46/build-script-build` (exit code: 101) --- stderr thread 'main' panicked at 'Unable to locate libzmq: `"pkg-config" "--libs" "--cflags" "libzmq >= 3.2"` did not exit successfully: exit code: 1 --- stderr Package libzmq was not found in the pkg-config search path. Perhaps you should add the directory containing `libzmq.pc' to the PKG_CONFIG_PATH environment variable No package 'libzmq' found ', /root/.cargo/registry/src/github.com-1ecc6299db9ec823/zmq-sys-0.8.2/build.rs:31:16 note: Run with `RUST_BACKTRACE=1` for a backtrace. warning: build failed, waiting for other jobs to finish... error: build failed

nage (Wed, 28 Mar 2018 15:45:42 GMT):
Discussion, questions and support for the stable branch of Hyperledger Indy (for master branch development and feature proposals join the appropriate channels: #indy-node (ledger), #indy-sdk (client library), and #indy-agent (reference implementations).

mboyd (Wed, 28 Mar 2018 16:51:12 GMT):
@sunilbarve Based on responses from members of the community, depending on your OS environment, you need to first install `libzmq` in one of the following ways: *macOS:* `brew install zeromq` *Ubuntu 16:* `[sudo] apt-get install libzmq3-dev` *Windows 10:* haven't tried this yet (probably easier to just run a VM of ubuntu) but go here for installation instructions: http://zeromq.org/docs:windows-installations

KOttoni (Wed, 28 Mar 2018 16:54:06 GMT):
Has joined the channel.

bmul (Wed, 28 Mar 2018 19:16:27 GMT):
Has joined the channel.

ZPDCEO (Thu, 29 Mar 2018 06:27:58 GMT):
Has joined the channel.

pwalimbe (Thu, 29 Mar 2018 12:15:52 GMT):
@mboyd I tried this on MacOS and I am having same issues. The error I get is: error: linking with `cc` failed: exit code: 1 = note: Undefined symbols for architecture x86_64: "_crypto_stream_aes128ctr_xor", referenced from: sodiumoxide::crypto::stream::aes128ctr::stream_xor::h08077e296fddea4d in libsodiumoxide-3f089728ded0e527.rlib(sodiumoxide-3f089728ded0e527.sodiumoxide10.rcgu.o) sodiumoxide::crypto::stream::aes128ctr::stream_xor_inplace::hb08450f153a5d06c in libsodiumoxide-3f089728ded0e527.rlib(sodiumoxide-3f089728ded0e527.sodiumoxide10.rcgu.o) "_crypto_stream_aes128ctr", referenced from: sodiumoxide::crypto::stream::aes128ctr::stream::hd56e35330e724195 in libsodiumoxide-3f089728ded0e527.rlib(sodiumoxide-3f089728ded0e527.sodiumoxide10.rcgu.o) ld: symbol(s) not found for architecture x86_64 clang: error: linker command failed with exit code 1 (use -v to see invocation) error: aborting due to previous error error: Could not compile `indy`.

Sean_Bohan (Thu, 29 Mar 2018 15:14:12 GMT):
https://drive.google.com/file/d/1jC5YRTELUNsrwLlEqcMgmmkmpdzgq4oQ/view

jljordan_bcgov (Thu, 29 Mar 2018 15:54:54 GMT):
#teameffort Beverly Dicks our Assistant Deputy Minister of Service BC was given the award for Government Leader of the Year at the KNOW ID conference in Washington DC this week! The work and support of the SSI / Decentralized Identity community was a contributor to this recognition. She is the person in the middle of the group not wearing a boring dark jacket :slightly_smiling_face:. We are looking forward to more great stuff this coming year. Thanks everyone! (apologies for the cross posting for folks in Sovrin or DIF) https://twitter.com/1worldidentity/status/978769457131737090?s=21

Kelattar (Fri, 30 Mar 2018 13:42:42 GMT):
Hey! Thank's for your answer .. I managed to run all the demo and everything works locally .. Now I want to run a client in a different machine than the nodes, which is more logical in real use, is it possible ? I don't know how to make my client know about the network "sandbox" (the same as the one of the nodes) and connect to it .. Can you please help me with this ?

paradisialberto (Fri, 30 Mar 2018 18:17:56 GMT):
Has joined the channel.

tkuhrt (Fri, 30 Mar 2018 18:40:41 GMT):
I was looking for a list of "help wanted" or "good first bugs" for the Hyperledger Indy project. I found https://jira.hyperledger.org/secure/RapidBoard.jspa?rapidView=133&projectKey=INDY&view=planning.nodetail&quickFilter=445, but there is nothing in there. Does Indy have bugs that would be good for someone just getting started?

MikeCohen (Fri, 30 Mar 2018 18:50:20 GMT):
Has joined the channel.

the_identity_guy (Fri, 30 Mar 2018 19:36:52 GMT):
Is there any docs on the key differences between Indy and uPort other than the underlying ledger and existence of smart contracts?

VipinB (Sat, 31 Mar 2018 14:24:16 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=qe5RrGP9hs7ogCmmJ) @jljordan_bcgov Great job... Will look forward to talk/presentation in #identity-wg later by experts on BC, we had a presentation earlier, could we have some exposure into how the work is developing and progressing.

Othman.Darwish (Sun, 01 Apr 2018 06:32:56 GMT):
Has left the channel.

sunilbarve (Mon, 02 Apr 2018 09:11:41 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=ioxx9KCP2A97YccvK) @mboyd thank you..solved the issue.

sunilbarve (Mon, 02 Apr 2018 09:13:24 GMT):
Get the following error while running cargo test.

sunilbarve (Mon, 02 Apr 2018 09:13:24 GMT):
Get the following error while running cargo test. ``` failures: ---- ledger_demo_works stdout ---- thread 'ledger_demo_works' panicked at 'called `Result::unwrap()` on an `Err` value: Timeout', /checkout/src/libcore/result.rs:916:5 note: Run with `RUST_BACKTRACE=1` for a backtrace. failures: ledger_demo_works test result: FAILED. 10 passed; 1 failed; 0 ignored; 0 measured; 0 filtered out error: test failed, to rerun pass '--test demo'

dinesh.rivankar (Mon, 02 Apr 2018 11:31:07 GMT):
Has left the channel.

sergey.minaev (Mon, 02 Apr 2018 15:13:20 GMT):
@sunilbarve Timeout for `ledger_demo_works` test most probably means that libindy can't connect to test pool. Please see https://github.com/hyperledger/indy-sdk#how-to-start-local-nodes-pool-with-docker

wangdong (Tue, 03 Apr 2018 01:26:45 GMT):
I wonder which part includes the Zero knowledge work.

mboyd (Tue, 03 Apr 2018 02:40:34 GMT):
@the_identity_guy If you are willing to take the deep dive, the Sovrin website has some good whitepapers on how a public permissioned ledger works: https://sovrin.org/library/

mboyd (Tue, 03 Apr 2018 02:44:06 GMT):
@tkuhrt I think a big help for "first bugs" or "first PRs" could be centered on streamlining the onboarding process for people who encounter the Indy project for the first time. I know that for me, learning how self-sovereign identity works technically has been a challenge. The getting started guides are great, but I think there are still ways we can improve the experience. What are your thoughts?

mboyd (Tue, 03 Apr 2018 02:45:21 GMT):
@pwalimbe Are you still having issues, or were you able to iron them out?

richbl (Tue, 03 Apr 2018 03:55:51 GMT):
Has joined the channel.

Kelattar (Tue, 03 Apr 2018 07:19:34 GMT):
Hello .. I managed to run all the demo and everything works locally .. Now I want to run a client in a different machine than the nodes, which is more logical in real use, is it possible ? I don't know how to make my client know about the network "sandbox" (the same as the one of the nodes) and connect to it .. Can you please help me with this ?

sunilbarve (Tue, 03 Apr 2018 11:39:34 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=vSJKDkopafAgofut8) @sergey.minaev yes i have done ``` ``` docker build -f ci/indy-pool.dockerfile -t indy_pool . docker run -itd -p 9701-9708:9701-9708 indy_pool but on second command shows docker: Error response from daemon: driver failed programming external connectivity on endpoint goofy_nightingale (7b043063887fd815381b347df553e6a68dd863ba00f6df166ac14ffa27e750bc): Bind for 0.0.0.0:9708 failed: port is already allocated.

sunilbarve (Tue, 03 Apr 2018 11:39:34 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=vSJKDkopafAgofut8) @sergey.minaev yes i have done, docker build -f ci/indy-pool.dockerfile -t indy_pool . docker run -itd -p 9701-9708:9701-9708 indy_pool but on second command shows docker: Error response from daemon: driver failed programming external connectivity on endpoint goofy_nightingale (7b043063887fd815381b347df553e6a68dd863ba00f6df166ac14ffa27e750bc): Bind for 0.0.0.0:9708 failed: port is already allocated.

sunilbarve (Tue, 03 Apr 2018 11:39:34 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=vSJKDkopafAgofut8) @sergey.minaev yes i have done, -- docker build -f ci/indy-pool.dockerfile -t indy_pool . --docker run -itd -p 9701-9708:9701-9708 indy_pool but on second command shows docker: Error response from daemon: driver failed programming external connectivity on endpoint goofy_nightingale (7b043063887fd815381b347df553e6a68dd863ba00f6df166ac14ffa27e750bc): Bind for 0.0.0.0:9708 failed: port is already allocated.

sunilbarve (Tue, 03 Apr 2018 11:39:34 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=vSJKDkopafAgofut8) @sergey.minaev yes i have done, 1.docker build -f ci/indy-pool.dockerfile -t indy_pool . 2.docker run -itd -p 9701-9708:9701-9708 indy_pool but on second command shows docker: Error response from daemon: driver failed programming external connectivity on endpoint goofy_nightingale (7b043063887fd815381b347df553e6a68dd863ba00f6df166ac14ffa27e750bc): Bind for 0.0.0.0:9708 failed: port is already allocated.

yghazi (Tue, 03 Apr 2018 11:59:11 GMT):
Has joined the channel.

sergey.minaev (Tue, 03 Apr 2018 12:06:43 GMT):
@sunilbarve Without running pool the `ledger_demo_works` test will fail

sunilbarve (Tue, 03 Apr 2018 13:28:28 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=h3aEREnfR9S8mkMmm) @sergey.minaev passed that *ledger_demo_works*,Getting new error while running 'getting_started.py' IndyError(

sergey.minaev (Tue, 03 Apr 2018 13:40:30 GMT):
@sunilbarve what version do you use? master or rc/stable?

sunilbarve (Tue, 03 Apr 2018 13:41:41 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=zF5TQooyekyBjHX4k) @sergey.minaev stable

sergey.minaev (Tue, 03 Apr 2018 13:59:47 GMT):
@sunilbarve could you share more context please: what are you doing and in witch sequence? AFAIR getting_started.py presents only master, not stable or rc... So please share more details. Also I can recommend to start from https://github.com/hyperledger/indy-sdk/blob/master/doc/getting-started/getting-started.md And also [this way](https://github.com/hyperledger/indy-sdk/blob/master/doc/getting-started/run-getting-started.md) should be more user-friendly to run

sergey.minaev (Tue, 03 Apr 2018 13:59:47 GMT):
@sunilbarve could you share more context please: what are you doing and in witch sequence? AFAIR getting_started.py presents only master, not stable or rc... So please share more details. Also I can recommend to start from https://github.com/hyperledger/indy-sdk/blob/master/doc/getting-started/getting-started.md And [this way](https://github.com/hyperledger/indy-sdk/blob/master/doc/getting-started/run-getting-started.md) should be more user-friendly to run

sergey.minaev (Tue, 03 Apr 2018 13:59:47 GMT):
@sunilbarve could you share more context please: what are you doing and in witch sequence? AFAIR getting_started.py presents only in master, not stable or rc... So please share more details. Also I can recommend to start from https://github.com/hyperledger/indy-sdk/blob/master/doc/getting-started/getting-started.md And [this way](https://github.com/hyperledger/indy-sdk/blob/master/doc/getting-started/run-getting-started.md) should be more user-friendly to run

JasneetSingh (Tue, 03 Apr 2018 21:32:37 GMT):
Has joined the channel.

mohdhafeezaj (Tue, 03 Apr 2018 21:33:26 GMT):
Hi team.. I really like Indy and its potential.. I just wanted to know is Indy current build a production ready ? And can Indy talk to Sawtooth or fabric framework ... if not can Indy be used as a digital identity wallet and have sawtooth or fabric for other transaction ?

stefanopulzeic (Wed, 04 Apr 2018 13:13:40 GMT):
Has joined the channel.

ravipatisivaramaprasad (Thu, 05 Apr 2018 07:07:33 GMT):
Has joined the channel.

ravipatisivaramaprasad (Thu, 05 Apr 2018 07:07:44 GMT):
Hi I have a general question. Hyperledger-Indy is for identity. Hyperledger-fabric has its own identity mechanism. But Can we integrate Hyperledger-indy with hyperledger fabric for identity purpose? If yes, in which case we can integrate it.

ravig-kant (Thu, 05 Apr 2018 10:27:40 GMT):
Has joined the channel.

sunilbarve (Thu, 05 Apr 2018 11:54:45 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=JAAenioTjz5ozuXET) @sergey.minaev While reun *docker-compose up* get the permission error : Recreating libindy_indy-client-rust-test_1 Attaching to libindy_indy-client-rust-test_1 indy-client-rust-test_1 | error: failed to open: /home/indy/indy-client-rust/target/debug/.cargo-lock indy-client-rust-test_1 | indy-client-rust-test_1 | Caused by: indy-client-rust-test_1 | Permission denied (os error 13) libindy_indy-client-rust-test_1 exited with code 101

sunilbarve (Thu, 05 Apr 2018 11:54:45 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=JAAenioTjz5ozuXET) @sergey.minaev While run *docker-compose up* get the permission error : Recreating libindy_indy-client-rust-test_1 Attaching to libindy_indy-client-rust-test_1 indy-client-rust-test_1 | error: failed to open: /home/indy/indy-client-rust/target/debug/.cargo-lock indy-client-rust-test_1 | indy-client-rust-test_1 | Caused by: indy-client-rust-test_1 | Permission denied (os error 13) libindy_indy-client-rust-test_1 exited with code 101

kpkrish (Thu, 05 Apr 2018 12:58:41 GMT):
Has joined the channel.

Sean_Bohan (Thu, 05 Apr 2018 14:09:20 GMT):
Folks: INDY Call today, April 5, 2018: Agenda: Housekeeping (Sean) Development Status (Slava) Wallet Update (IanC) Benchmarking and Scaling Discussion (Slava and Vladimir) 8amPT, 9amMT, 11amET, 4pm BST Where: Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/232861185 Or iPhone one-tap (US Toll): +16465588656,232861185# or +14086380968,232861185# Or Telephone: Dial: +1 646 558 8656 (US Toll) or +1 408 638 0968 (US Toll) Meeting ID: 232 861 185 International numbers available: https://zoom.us/zoomconference?m=a0jD_rTMnh0ZYGQDOKPCNrK_0dP7WPfp1 Hyperledger Indy Community Indy Rocketchat: https://chat.hyperledger.org/channel/indyhttps://chat.hyperledger.org/channel/indy-agenthttps://chat.hyperledger.org/channel/indy-node Indy Wiki: https://wiki.hyperledger.org/projects/indy Indy WG Agendas: https://drive.google.com/open?id=1wNnp1pPS6-1Y4B2oygfI6tOU4eHGmpRH Indy Videos: https://drive.google.com/open?id=1Z8-jR7hnXb57fufE0OXxIfeUn05zNmRt Indy WG Meeting Recordings: https://drive.google.com/open?id=0B_NJV6eJXAA1UlJZMXd3cm1zNDQ

vatsal.shah (Thu, 05 Apr 2018 14:33:18 GMT):
Has joined the channel.

tkuhrt (Thu, 05 Apr 2018 21:13:55 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=rzDSH7956dxJGF5zw) @mboyd : This makes sense to me. Normally the fresh set of eyes is the one who can best improve documentation that we either don't look at anymore because we have advanced beyond it or we have our own version of it that we use to remind us the steps. We tend to struggle through getting started, but rarely update the material to make it better for the next person who has to use the material

mboyd (Thu, 05 Apr 2018 21:26:52 GMT):
@tkuhrt I'm happy to help refine the HL-Indy getting started process. Ill post a document at the beginning of next week with my process I've used and the suggestions to other people. Perhaps there are others who would also like to help with making developer onboarding a more enjoyable experience? Reach out to me if you would like to help, or have suggestions. One of my goals is to make a 15 minute interactive demo that helps people see a high level vision of Indy's potential, and another is to make a ~2 hr how-to tutorial for the indy-sdk so developers can see how this would be used in an application. I think we have the beginnings of this, but id like to refine it further.

mboyd (Thu, 05 Apr 2018 21:26:52 GMT):
@tkuhrt I'm happy to help refine the HL-Indy getting started process. Ill post a document at the beginning of next week with my process I've used and the suggestions to other people. Perhaps there are others who would also like to help with making developer onboarding a more enjoyable experience? out to me if you would like to help, or have suggestions. One of my goals is to make a 15 minute interactive demo that helps people see a high level vision of Indy's potential, and another is to make a ~2 hr how-to tutorial for the indy-sdk so developers can see how this would be used in an application. I think we have the beginnings of this, but id like to refine it further.

mboyd (Thu, 05 Apr 2018 21:26:52 GMT):
@tkuhrt I'm happy to help refine the HL-Indy getting started process. Ill post a document at the beginning of next week with my process I've used and the suggestions to other people. Perhaps there are others who would also like to help with making developer onboarding a more enjoyable experience? Reach to me if you would like to help, or have suggestions. One of my goals is to make a 15 minute interactive demo that helps people see a high level vision of Indy's potential, and another is to make a ~2 hr how-to tutorial for the indy-sdk so developers can see how this would be used in an application. I think we have the beginnings of this, but id like to refine it further.

tkuhrt (Thu, 05 Apr 2018 21:28:23 GMT):
@mboyd : I am really interested in your videos. We want to put more videos on the hyperledger.org site that are intended for developers, as a way of getting started.

tkuhrt (Thu, 05 Apr 2018 21:28:46 GMT):
I would love to use something that the community puts together instead of something that I do.

tkuhrt (Thu, 05 Apr 2018 21:29:33 GMT):
Also, count me in as a reviewer to the developer onboarding work that you want to do

niceoneallround (Thu, 05 Apr 2018 22:25:13 GMT):
Has joined the channel.

the_identity_guy (Thu, 05 Apr 2018 22:25:28 GMT):
Hi, what is the major differences between the two VON projects residing on: https://github.com/bcgov/von_agent and https://github.com/PSPC-SPAC-buyandsell/von_agent

nuxibyte_old (Thu, 05 Apr 2018 22:54:41 GMT):
Has left the channel.

DannyWong (Fri, 06 Apr 2018 02:42:38 GMT):
Has joined the channel.

ianco (Fri, 06 Apr 2018 03:07:05 GMT):
@the_identity_guy the bcgov project is just a fork of the PSPC-SPAC-buyandsell project, it' used to freeze dependencies for the TheOrgBook and permitify projects in bcgov

ianco (Fri, 06 Apr 2018 03:07:05 GMT):
@the_identity_guy the bcgov project is just a fork of the PSPC-SPAC-buyandsell project, it' used to freeze dependencies for the TheOrgBook and permitify projects in bcgov ... https://github.com/PSPC-SPAC-buyandsell/von_agent is the latest @ greatest

ibmamnt (Fri, 06 Apr 2018 05:07:24 GMT):
Has joined the channel.

MichaelStein (Fri, 06 Apr 2018 10:30:58 GMT):
Has joined the channel.

sergey.minaev (Fri, 06 Apr 2018 14:30:07 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=JfH4z9AB9KNwiFbRq) @Artemkaaas

brondera (Fri, 06 Apr 2018 15:10:16 GMT):
Has joined the channel.

the_identity_guy (Fri, 06 Apr 2018 18:50:35 GMT):
@ianco great thanks!

the_identity_guy (Mon, 09 Apr 2018 19:24:31 GMT):
is anonymous credentials a type of zero knowledge proof? or are these two related but not same?

drummondreed (Mon, 09 Apr 2018 19:38:26 GMT):
@the_identity_guy Yes, anoncreds is a name typically given to the exchange of digital credentials using ZKP. More specifically it is often used to refer to using Attribute-Based Credentials (ABC). See the ABC4Trust project for more details: https://abc4trust.eu/

esplinr (Mon, 09 Apr 2018 22:35:19 GMT):
Has joined the channel.

varinder (Tue, 10 Apr 2018 05:47:21 GMT):
Has joined the channel.

easeev (Tue, 10 Apr 2018 10:22:44 GMT):
Has joined the channel.

zacchariah (Tue, 10 Apr 2018 11:22:19 GMT):
Has joined the channel.

nicovalencia (Tue, 10 Apr 2018 16:52:39 GMT):
Has joined the channel.

SanketPanchamia (Wed, 11 Apr 2018 11:28:51 GMT):

Clipboard - April 11, 2018 4:58 PM

SanketPanchamia (Wed, 11 Apr 2018 11:28:53 GMT):
Hi. i am new to Indy and setting up the dev environment to run indy locally on my system. I was able to install all the dev tools using the setup-dev folder for ubuntu 16. However when i am trying to install indy-node-dev, i get the following error

eramitg (Wed, 11 Apr 2018 15:39:28 GMT):
Hi Folks , I am an Phd Candidate in www.nitrr.ac.in my Linkedind Profile is https://www.linkedin.com/in/eramitg/ for sake of earning an Phd Degree i was proposed Blockchain Technology research work area to my guide so oom I request all of you gyus ,please guide me and assign me some research oriented task so that we mutullay benifited research related to Hyperledger Umbrella Project , All of you feel free to catch me on twitter or skype to https://twitter.com/eramitg1 or amitg.iitb skype id also in Zoom to in Zoom ID 3649222703 or whatsapp +917773011100 Regards

esplinr (Wed, 11 Apr 2018 17:16:10 GMT):
@eramitg I see that this is a cross-post from the other channel. Are you specifically interested in Project Indy, or did you post in lots of projects?

esplinr (Wed, 11 Apr 2018 17:16:20 GMT):
Depending on your background and interests, we have lots of work that you can do.

esplinr (Wed, 11 Apr 2018 17:19:55 GMT):
Hello everyone. I recently joined Evernym and have been assigned to the Indy project. I don't know much yet, but I hope I can be a resource here in the future. Expect a lot of newbie questions as I get up to speed.

kdenhartog (Thu, 12 Apr 2018 00:25:52 GMT):
@SanketPanchamia we were getting a similar error today. The way we resolved it was to do "apt-get/brew install rocksdb" Also, if that doesn't specifically resolve it try version 5.8.8

NareshReddy (Thu, 12 Apr 2018 04:57:04 GMT):
Has joined the channel.

SanketPanchamia (Thu, 12 Apr 2018 06:11:47 GMT):
Hi, after setting up the dev tools, i am trying to build the indy node following the tutorial at https://github.com/hyperledger/indy-node/blob/master/docs/indy-running-locally.md I am getting an error at the generate scripts where it fails with NETWORK_NAME

SanketPanchamia (Thu, 12 Apr 2018 06:11:47 GMT):
Hi, after setting up the dev tools, i am trying to build the indy node following the tutorial at https://github.com/hyperledger/indy-node/blob/master/docs/indy-running-locally.md I am getting an error at the generate scripts where it fails with NETWORK_NAME

SanketPanchamia (Thu, 12 Apr 2018 06:12:10 GMT):

Clipboard - April 12, 2018 11:42 AM

MyMate (Thu, 12 Apr 2018 09:25:44 GMT):
Has joined the channel.

thesauri (Thu, 12 Apr 2018 11:49:54 GMT):
Has joined the channel.

thesauri (Thu, 12 Apr 2018 11:55:42 GMT):
Hello! In the tutorial on running an Indy cluster locally, we are both spawning four nodes using `start_indy_node Node1 9701 9702` etc. and `python /usr/local/lib/python3.5/dist-packages/indy_client/test/agent/faber.py --port 5555 --network `. Are the first set of 4 nodes validator nodes and the latter 3 observer nodes (Faber, Acme, and Thrift)?

mawi (Thu, 12 Apr 2018 12:01:49 GMT):
The latter three are agents, who do not necessarily have to be observer nodes as far as i know

johadahl (Thu, 12 Apr 2018 12:03:15 GMT):
Has joined the channel.

thesauri (Thu, 12 Apr 2018 12:08:07 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=rGmZ6NqKvXrauD2MN) @mawi Right, thanks

davidgsmits (Thu, 12 Apr 2018 13:50:33 GMT):
Has joined the channel.

Sean_Bohan (Thu, 12 Apr 2018 14:25:06 GMT):
Hyperledger Indy WG Call is TODAY AGENDA: Housekeeping (Sean) Revocation (Slava and Alex) Thoughts on April Internet Identity Workshop (Multiple) Open Chat (All) Wrap Up When: April 12, 2018 8amPT, 9amMT, 11amET, 3pm GMT Where: Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/232861185 Or iPhone one-tap (US Toll): +16465588656,232861185# or +14086380968,232861185# Or Telephone: Dial: +1 646 558 8656 (US Toll) or +1 408 638 0968 (US Toll) Meeting ID: 232 861 185 International numbers available: https://zoom.us/zoomconference?m=a0jD_rTMnh0ZYGQDOKPCNrK_0dP7WPfp1 Hyperledger Indy Community Indy Rocketchat: https://chat.hyperledger.org/channel/indyhttps://chat.hyperledger.org/channel/indy-agenthttps://chat.hyperledger.org/channel/indy-node Indy Wiki: https://wiki.hyperledger.org/projects/indy Indy WG Agendas: https://drive.google.com/open?id=1wNnp1pPS6-1Y4B2oygfI6tOU4eHGmpRH Indy Videos: https://drive.google.com/open?id=1Z8-jR7hnXb57fufE0OXxIfeUn05zNmRt Indy WG Meeting Recordings: https://drive.google.com/open?id=0B_NJV6eJXAA1UlJZMXd3cm1zNDQ

danielhardman (Thu, 12 Apr 2018 16:00:51 GMT):
Here are the developer tutorials done so far: https://github.com/hyperledger/indy-sdk/tree/master/doc/how-tos. Mixture of python and Java (some of each). The "prerequisites" link is currently broken but is essentially a placeholder for instructions on A) how to compile or install libindy; and B) how to start an indy validator pool in docker. Both of these are covered in the main readme at https://github.com/hyperledger/indy-sdk/

eramitg (Thu, 12 Apr 2018 16:05:14 GMT):
@esplinr just i have seen your message , my intrest area to do reserach oriented research work related to blockchain. i am much more intrested related to security issue but i am open for all type of scenario related to Blockchain project.

eramitg (Thu, 12 Apr 2018 16:06:41 GMT):
tha's a reason my qurey to all folk , please assign a challengin research oreinted task so that i can contribute my work and get some research exposer for my academic research

esplinr (Thu, 12 Apr 2018 16:29:17 GMT):
@eramitg What specifically do you mean when you say you are interested in "security issue"? Crypto? Identity? Pen testing?

the_identity_guy (Thu, 12 Apr 2018 19:08:54 GMT):
@eramitg thats great, do you have any research papers on blockchain security ?

eramitg (Thu, 12 Apr 2018 19:09:28 GMT):
Yes i have some research papers ..

eramitg (Thu, 12 Apr 2018 19:10:24 GMT):
but if your provide some your open problem and puzzles related to security issue it would be very much welcome.

eramitg (Thu, 12 Apr 2018 19:10:30 GMT):
@the_identity_guy

the_identity_guy (Thu, 12 Apr 2018 19:14:05 GMT):
I am a security researcher myself working on blockchain, currently working exploring smart contract security but thats not directly related to Indy

the_identity_guy (Thu, 12 Apr 2018 19:14:05 GMT):
I am a security researcher myself working on blockchain, currently exploring smart contract security but thats not directly related to Indy

eramitg (Thu, 12 Apr 2018 19:44:53 GMT):
@the_identity_guy i am also exploring in same topic in Blockchain with ethereum ,smart contract,cryptography such type of concepts and i am also not bounded myself for particular topic but indy project is so good resource for learner and researcher like me and am so thankfull to indy group too

SanketPanchamia (Fri, 13 Apr 2018 11:15:02 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=rKbytG8eHNyWMWnb7) Any one can help with this?

KevinFu (Fri, 13 Apr 2018 11:27:52 GMT):
Has joined the channel.

SanketPanchamia (Fri, 13 Apr 2018 11:42:55 GMT):
Hi, I am following the dev tutorial to setup indy on my local system and was on step 2 that says this *set Network name in config file* the location of the config depends on how a Node was installed. It's usually inside /etc/indy for Ubuntu. the following needs to be added: NETWORK_NAME={network_name} where {network_name} matches the one in genesis transaction files above I am no sure how to give this network name because i cannot initialize a node without a network name. Am i missing something?

icemanblue (Fri, 13 Apr 2018 15:58:22 GMT):
Has joined the channel.

the_identity_guy (Sat, 14 Apr 2018 03:03:01 GMT):
is there any documents around how Plenum validates signatures for write requests?

eramitg (Sat, 14 Apr 2018 17:01:27 GMT):
Hey gus how can i contribute or become a maintenaer for this group . i am using gerrit for fabric and cello project but for indy i am not able to figure out how to and where i ger gerrit setting to install indy ...i forked github indy project. but if i would like to contribute in gerrit how it could be able to do that>

eramitg (Sat, 14 Apr 2018 17:01:27 GMT):
Hey guys how can i contribute or become a maintenaer for this group . i am using gerrit for fabric and cello project but for indy i am not able to figure out how to and where i ger gerrit setting to install indy ...i forked github indy project. but if i would like to contribute in gerrit how it could be able to do that>

eramitg (Sat, 14 Apr 2018 17:01:27 GMT):
Hey guys how can i contribute or become a maintenaer for this group . i am using gerrit for fabric and cello project but for indy i am not able to figure out how to and where i ger gerrit setting to install indy ...i forked github indy project. but if i would like to contribute in gerrit how it could be able to do that?

eramitg (Sat, 14 Apr 2018 19:07:18 GMT):
when i fired this command i encounter error in indy_sdk /home/amit/indy-sdk/libindy/target/debug/deps/libindy.rlib

eramitg (Sat, 14 Apr 2018 19:07:59 GMT):

error.png

eramitg (Sat, 14 Apr 2018 19:08:53 GMT):

error-indy-sdk.png

eramitg (Sat, 14 Apr 2018 19:08:53 GMT):

error-indy-sdk.png

eramitg (Sat, 14 Apr 2018 19:10:59 GMT):

error2.png

eramitg (Sat, 14 Apr 2018 19:18:45 GMT):
Hey folks i solved this issue just firing this command sudo RUST_TEST_THREADS=1 cargo test

Steve-Boyd (Mon, 16 Apr 2018 01:47:56 GMT):
Has joined the channel.

ruchit 2 (Mon, 16 Apr 2018 02:05:19 GMT):
Has joined the channel.

DurgaRao641 (Mon, 16 Apr 2018 05:41:32 GMT):
Hi, Please clarify some of my questions in Indy:``` 1) Where the private key of Issuer will store? 2) Where the Wallet of each identity owner will store? 3) Pool data is stored in pool ledger which is immutable, What if a node went down or want to be removed? 4) Can we use Indy with the other frameworks like Sawtooth or Fabric? ```

ZahidZafar (Mon, 16 Apr 2018 07:10:12 GMT):
Has joined the channel.

DannyWong (Mon, 16 Apr 2018 10:05:22 GMT):
i am trying to convince my client to leverage the production Sovrin network

DannyWong (Mon, 16 Apr 2018 10:06:04 GMT):
is it possible to know the companies are serving as Identity Issuer / Identity verifier?

KevinFu (Mon, 16 Apr 2018 14:07:04 GMT):
Hi,I got 1 question,when prepare env, there have 1 step: Fork indy-plenum and indy-node, may i know what's this mean ?

SanketPanchamia (Tue, 17 Apr 2018 06:52:51 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=vn8THRmTuJFtKFgcM) @KevinFu Its like cloning the repository

KevinFu (Tue, 17 Apr 2018 10:05:29 GMT):
I see. thanks @SanketPanchamia

KevinFu (Tue, 17 Apr 2018 10:06:44 GMT):
do we have any infra document of Indy? and how business working in indy something, I review the fabric, it's very clear

CarlitoIBM (Tue, 17 Apr 2018 14:26:01 GMT):
Has joined the channel.

Mahmic404 (Tue, 17 Apr 2018 14:35:14 GMT):
Has joined the channel.

lebdron (Tue, 17 Apr 2018 15:43:25 GMT):
Has joined the channel.

mtzmlz (Tue, 17 Apr 2018 16:16:35 GMT):
Has joined the channel.

hendry19901990 (Tue, 17 Apr 2018 16:49:10 GMT):
is there a library to connect blockchain ?

smithbk (Tue, 17 Apr 2018 21:28:40 GMT):
The getting started tutorial at https://github.com/hyperledger/indy-sdk/blob/master/doc/getting-started/getting-started.md says ```In this tutorial we will describe the simple version of onboarding process. In our case, one party will always be the Trust Anchor. Real enterprise scenarios can use a more complex version.```Could someone describe what a real enterprise scenario would be? I'm guessing that if neither Alice nor Bob are trust anchors but they want to establish a relationship, then they would have to send their pair-wise DIDs to a trust anchor to write to the ledger, but then that trust anchor would know about Alice and Bob's relationship. Is this correct? Any info on enterprise scenarios is appreciated.

rjones (Tue, 17 Apr 2018 22:42:38 GMT):
Has joined the channel.

tomislav (Wed, 18 Apr 2018 01:24:19 GMT):
Alice and Bob can create pairwise DID's with the issuer and send them to the ledger themselves. They will also have pairwise created with each other and send that to the ledger. Enterprise scenarios can be for example Alice and Bob having a cloud client that does the interaction with the issuer on their behalf.

pknowles (Wed, 18 Apr 2018 09:17:12 GMT):
Due to the importance of Schema “Overlays”, we’ve set up a new channel on Slack (sovrin.slack.com) to cover this topic: #overlays-services In a nutshell, “Overlays” are acetate-type layers to help define schema usage, compliance practices and services that compliment what the base schema definitions provide as a foundation. The intro deck can be downloaded via the following link … https://we.tl/1KZX2WlIPp To get the ball rolling, there is a sandbox programming task defined in the deck for any keen coders out there!

pknowles (Wed, 18 Apr 2018 09:17:12 GMT):
Due to the importance of Schema “Overlays”, we’ve set up a new channel on Slack (sovrin.slack.com) to cover this topic: #overlays-services In a nutshell, “Overlays” are acetate-type layers to help define schema usage, compliance practices and services that compliment what the base schema definitions provide as a foundation. The intro deck can be downloaded via the following link … https://we.tl/1KZX2WlIPp To get the implementation ball rolling, there is a programming task defined in the deck for any keen coders out there!

DurgaRao641 (Wed, 18 Apr 2018 10:25:25 GMT):
Hi guys, Can we see any transactions in the ledger? Please tell me if there is any way.

smithbk (Wed, 18 Apr 2018 11:28:59 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=r756Kj4JEJLa4ZKQA) @tomislav https://docs.google.com/spreadsheets/d/1TWXF7NtBjSOaUIBeIH77SyZnawfo91cJ_ns4TR-wsq4/edit#gid=0 says the NYM transaction will be allowed by `Identity Owner` and `non-identity entity` later, so does this mean that Alice and Bob (say as identity owners) can't send DIDs to the ledger currently but will be able to in the future? If yes, any ETA? Thanks

jljordan_bcgov (Wed, 18 Apr 2018 21:09:16 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=fFJwZWrYH7gMYLPc7) @VipinB Sorry .. been a while since I checked in to Rocketchat .. for sure we can share where we are at ... http://von.pathfinder.bcgov.gc.ca has the latest working demos, code, some docs, links to repos and all that. Happy to talk about our work and partners on a call sometime

LoveshHarchandani_1 (Wed, 18 Apr 2018 21:35:23 GMT):
Has joined the channel.

Sean_Bohan (Wed, 18 Apr 2018 21:47:43 GMT):
This week (Thurs 4/19) on the Hyperledger Indy WG Call: Agenda: We will be talking about Wallets, Storage Layer and Export Formats 8amPT, 9amMT, 11amET, 3pm GMT Where: Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/232861185 Or iPhone one-tap (US Toll): +16465588656,232861185# or +14086380968,232861185# Or Telephone: Dial: +1 646 558 8656 (US Toll) or +1 408 638 0968 (US Toll) Meeting ID: 232 861 185 International numbers available: https://zoom.us/zoomconference?m=a0jD_rTMnh0ZYGQDOKPCNrK_0dP7WPfp1

nitin.agarwal (Thu, 19 Apr 2018 05:32:31 GMT):
Has joined the channel.

neewy (Thu, 19 Apr 2018 08:17:49 GMT):
Has joined the channel.

Kelattar (Thu, 19 Apr 2018 13:28:12 GMT):
How can an agent send a msg to another agent ? without going through the indy CLI

mhomaid (Thu, 19 Apr 2018 16:00:04 GMT):
Has joined the channel.

agunde (Thu, 19 Apr 2018 20:27:12 GMT):
Has left the channel.

vamsiraju (Thu, 19 Apr 2018 20:34:52 GMT):
Has joined the channel.

grice_32 (Thu, 19 Apr 2018 20:59:08 GMT):
Has joined the channel.

stanleyc-trustscience (Thu, 19 Apr 2018 22:27:08 GMT):
Has joined the channel.

srinivasanraghavan (Fri, 20 Apr 2018 03:47:10 GMT):
Has joined the channel.

NeerajKukreti (Fri, 20 Apr 2018 03:51:24 GMT):
Has joined the channel.

stanleyc-trustscience (Fri, 20 Apr 2018 05:33:41 GMT):
Hi, I'm reading through this getting-started guide, https://github.com/hyperledger/indy-node/blob/master/getting-started.md. But it says it's been depreciated. Why is it depreciated though?

stanleyc-trustscience (Fri, 20 Apr 2018 05:33:41 GMT):
Hi, I'm reading through this getting-started guide, https://github.com/hyperledger/indy-node/blob/master/getting-started.md. But it says it's been depreciated. Why is it depreciated though? The new guide doesn't seem to talk about indy CLI ...

stanleyc-trustscience (Fri, 20 Apr 2018 05:33:41 GMT):
Hi, I'm reading through this getting-started guide, https://github.com/hyperledger/indy-node/blob/master/getting-started.md. But it says it's been depreciated. Why is it depreciated though? The new guide doesn't seem to talk about indy CLI ... Is the CLI not that important to know about now?

stanleyc-trustscience (Fri, 20 Apr 2018 05:33:41 GMT):
Hi, I'm reading through this getting-started guide, https://github.com/hyperledger/indy-node/blob/master/getting-started.md. But it says it's been depreciated. Why is it depreciated though? The new guide doesn't seem to talk about indy CLI ... Is the CLI not that important to know about now?

srinivasanraghavan (Fri, 20 Apr 2018 06:02:34 GMT):
Hey I am new to INDY. I was trying go through the work flow. I followed through the steps in CLI and got stuck in "$connect sandbox" . Its not able to connect. Is there any other CLI documentation or CLI is deprecated

ashcherbakov (Fri, 20 Apr 2018 07:53:46 GMT):
@stanleyc-trustscience @srinivasanraghavan The new and actual Getting Started is in https://github.com/hyperledger/indy-sdk/tree/master/doc/getting-started It has documentation about how to run it (a local pool needs to be started, for example in docker, so that `connect sandbox` will work). The old CLI (in indy-node repo) is being deprecated, because old client code and old CLI (from indy-node) is being deprecated as well. We want to have just one official and supported client: indy-sdk. All client, CLI and getting started code will be removed soon from indy-node.

srinivasanraghavan (Fri, 20 Apr 2018 07:55:24 GMT):
Thank you @ashcherbakov

WadeBarnes (Fri, 20 Apr 2018 13:27:36 GMT):
For all of those that have heard of the work we're doing here in Canada, and especially those that have not, we've refined a Quick Start Guide to help you get a complete set of applications up and running on your local machine. https://github.com/bcgov/TheOrgBook#getting-started https://github.com/bcgov/von-network#quick-start-guide https://github.com/bcgov/permitify#quick-start-guide

Steve-Boyd (Fri, 20 Apr 2018 13:29:25 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=cTv85Xd9ub3QYC5j2) @WadeBarnes are you with the BC gov?

WadeBarnes (Fri, 20 Apr 2018 13:29:41 GMT):
Yes

WadeBarnes (Fri, 20 Apr 2018 13:29:41 GMT):
Yes, consultant for the BC Gov.

Steve-Boyd (Fri, 20 Apr 2018 13:31:11 GMT):
Ontario gov here. We are looking into doing some POC work. I'd love to pick your brain to see what you all are doing on the west coast :smiley:

WadeBarnes (Fri, 20 Apr 2018 13:33:11 GMT):
We are working with an Ontario government team. Would be good for you to connect with them too.

WadeBarnes (Fri, 20 Apr 2018 13:35:23 GMT):
@Steve-Boyd, you should connect with @mitovskaol, Olena is one of the members of that team.

mitovskaol (Fri, 20 Apr 2018 13:35:23 GMT):
Has joined the channel.

WadeBarnes (Fri, 20 Apr 2018 13:36:05 GMT):
Happy to help grow the community in Canada.

WadeBarnes (Fri, 20 Apr 2018 13:37:08 GMT):
You'll see an Ontario profile in TheOrgBook and Permitify repositories.

Steve-Boyd (Fri, 20 Apr 2018 13:50:30 GMT):
Very cool, thanks for the info @WadeBarnes!

WadeBarnes (Fri, 20 Apr 2018 13:53:25 GMT):
@Steve-Boyd, You can see a deployed instance of the Ontario profile here; https://ontvon-von-dev.pathfinder.gov.bc.ca/en/home And see the Ontario business (sample) registration process here; https://onbis-devex-von-permitify-dev.pathfinder.gov.bc.ca/

WadeBarnes (Fri, 20 Apr 2018 13:55:30 GMT):
The BC example is a bit further along. You can walk through a complete "Permitify Demo - Starting a Restaurant Recipe" here; https://devex-von-dev.pathfinder.gov.bc.ca/en/recipe/start_a_restaurant

WadeBarnes (Fri, 20 Apr 2018 13:58:25 GMT):
The links I provided above allow you to run all of the applications necessary for this on our local machine; a provisional ledger complete with ledger browser, TheOrgBook complete with all it's various components, and a full set of issuer services from Permitify.

Steve-Boyd (Fri, 20 Apr 2018 13:59:39 GMT):
Nice. Is this POC or something your team has/will be deploying into production?

WadeBarnes (Fri, 20 Apr 2018 14:00:56 GMT):
We're working toward a production BETA for the BC Registries.

WadeBarnes (Fri, 20 Apr 2018 14:03:21 GMT):
Feel free to play around with what is there now. It's all provisional. We'll be rolling out the real thing in a separate environment, these environments are used for demos, development, and experimentation.

WadeBarnes (Fri, 20 Apr 2018 14:06:37 GMT):
Pull requests are welcome.

Steve-Boyd (Fri, 20 Apr 2018 14:11:17 GMT):
Thanks, going to pass the link to my manager as I'm sure he will be interested to check it out.

esplinr (Fri, 20 Apr 2018 15:01:13 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=yRQeirtS2JoFCpWgE) @DurgaRao641 Did you get answers to your questions? I don't know all the answers, but this is what I can share: 3. The Pool Ledger is appended to reflect the state of the network. 4. They are different tools with different goals. In what way would you like to see them used together?

esplinr (Fri, 20 Apr 2018 15:02:31 GMT):
@eramitg Sorry about my delay in responding. Are you still interested in contributing to Hyperledger Indy? What sort of contributions are you interested in making? Did you look at the HelpWanted issues in JIRA? Feel free to submit pull requests.

eramitg (Fri, 20 Apr 2018 15:02:54 GMT):
Yes

eramitg (Fri, 20 Apr 2018 15:03:11 GMT):
i am doing research as Phd candidate in Blockchain era

eramitg (Fri, 20 Apr 2018 15:03:44 GMT):
and security & Privacy is my main motivation @esplinr

esplinr (Fri, 20 Apr 2018 15:04:10 GMT):
@ngeorge Would it be useful to create a new channel for developers working on Indy, rather than using Indy? Most of the conversation in the existing channels, #indy, #indy-agent, #indy-node, #indy-sdk, contain conversation with people using the technologies rather than the sprint teams doing development.

ngeorge (Fri, 20 Apr 2018 15:04:10 GMT):
Has joined the channel.

esplinr (Fri, 20 Apr 2018 15:04:26 GMT):
Specifically, we should discuss process questions like releases, versioning, etc.

esplinr (Fri, 20 Apr 2018 15:05:04 GMT):
@eramitg Oh yeah, I remember now. You should share a proposal for what you want to do, and get feedback on it.

eramitg (Fri, 20 Apr 2018 15:05:21 GMT):
i am able to use github indy sdk ,but if i would like to pull from github or

eramitg (Fri, 20 Apr 2018 15:05:51 GMT):
in JIRA not sdk for indy to commit

eramitg (Fri, 20 Apr 2018 15:06:02 GMT):
so i have to use github

eramitg (Fri, 20 Apr 2018 15:06:55 GMT):
yeah @ngeorge @esplinr ,please create channel as developer for indy sdk rather than using it

esplinr (Fri, 20 Apr 2018 19:19:48 GMT):
@eramitg I spoke with @ngeorge (he is in our office today). He is the official maintainer of the Indy project at Hyperledger. He is also the CTO at Sovrin, which is the expected to be the first production implementation.

esplinr (Fri, 20 Apr 2018 19:21:00 GMT):
He thinks that most usage oriented questions should be addressed here in #indy, and that development related questions should be in the other channels based on the team: #indy-agent , #indy-node , #indy-sdk . So there is no need for a separate channel.

esplinr (Fri, 20 Apr 2018 19:22:53 GMT):
@eramitg I'm new here, and am still learning about how the work gets organized. But I suggest that you proposed a project that meets your requirements for your PhD. You should also join the Thursday Indy Working Group call at 4PM BST. https://zoom.us/j/232861185

esplinr (Fri, 20 Apr 2018 19:23:30 GMT):
If you subscribe to the Hyperledger shared calendar, you'll see the appointment.

vgotcheva (Fri, 20 Apr 2018 23:29:16 GMT):
Has joined the channel.

kyogesh91 (Sat, 21 Apr 2018 20:11:49 GMT):
Has joined the channel.

kyogesh91 (Sat, 21 Apr 2018 21:24:03 GMT):

Screenshot from 2018-04-21 16-13-26.png

kyogesh91 (Sat, 21 Apr 2018 21:24:32 GMT):
any idea what would be the reason for this error? any idea what would be the reason for this error? any idea what would be the reason for this error? any idea what would be the reason for this error? Any idea what could be causing this issue?

kyogesh91 (Sat, 21 Apr 2018 21:24:32 GMT):
any idea what would be the reason for this error?

raccoonrat (Sun, 22 Apr 2018 06:28:24 GMT):
Has joined the channel.

KitHat (Mon, 23 Apr 2018 08:52:36 GMT):
Has joined the channel.

musquash (Mon, 23 Apr 2018 11:07:00 GMT):
Has joined the channel.

Toktar (Mon, 23 Apr 2018 12:01:22 GMT):
A new version of the base58 1.0.0 was released. In it, the format of the response of the b58encode () method was changed from string to bytes. We will make hotfix in the near future.

smithbk (Mon, 23 Apr 2018 20:10:28 GMT):
Hi, can someone point me in the right direction? How would I do the equivalent of PKI certificate chaining with indy? More specifically, suppose A issues a cred to B, B issues a cred to C, and D wants to verify that C’s cred was issued by someone with a cred from A. I assume B would need to generate a proof of its ownership of an A cred and include that in the cred that it issues to C. Is there any more to it than that, or a better way? Thanks

the_identity_guy (Mon, 23 Apr 2018 20:12:43 GMT):
Is it possible to get a copy of Kazue Sako's presentation or paper about zero knowledge proofs?

StevenCotes (Mon, 23 Apr 2018 20:25:45 GMT):
Has joined the channel.

mccown (Mon, 23 Apr 2018 20:32:47 GMT):
Has joined the channel.

VP18 (Tue, 24 Apr 2018 04:07:04 GMT):
Has joined the channel.

sukeshshetty (Tue, 24 Apr 2018 11:06:10 GMT):
Has joined the channel.

timms (Wed, 25 Apr 2018 07:35:32 GMT):
Has joined the channel.

ragpach2 (Wed, 25 Apr 2018 10:57:27 GMT):
Has joined the channel.

mjmckean (Wed, 25 Apr 2018 15:04:30 GMT):
@the_identity_guy you can find notes on a ZKP presentation that Kazue gave at IIW this year at https://github.com/afroDC/Personal/wiki/Kazui-Sako-Zero-Knowledge-Proof-101-Notes-from-IIW-26

tomislav (Wed, 25 Apr 2018 15:29:42 GMT):
This one? https://github.com/afroDC/Personal/wiki/Kazui-Sako-Zero-Knowledge-Proof-101-Notes-from-IIW-26

jackson18 (Wed, 25 Apr 2018 19:57:31 GMT):
Has joined the channel.

Sean_Bohan (Wed, 25 Apr 2018 21:16:50 GMT):
TOMORROW 4/26 - Hyperledger INDY WG Call AGENDA: Housekeeping (Sean) SDK 1.4 & Backwards Incompatible Wallet Changes (SlavaG) Indy Product Roadmap Collaboration (RichardE) Maintainers Meeting (RichardE, SeanB) Multiple Teams in Hyperledger JIRA? (All) Wrap Up When: April 26, 2018 8amPT, 9amMT, 11amET, 3pm GMT Where: Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/232861185 Or iPhone one-tap (US Toll): +16465588656,232861185# or +14086380968,232861185# Or Telephone: Dial: +1 646 558 8656 (US Toll) or +1 408 638 0968 (US Toll) Meeting ID: 232 861 185 International numbers available: https://zoom.us/zoomconference?m=a0jD_rTMnh0ZYGQDOKPCNrK_0dP7WPfp1 Hyperledger Indy Community Indy Rocketchat: https://chat.hyperledger.org/channel/indyhttps://chat.hyperledger.org/channel/indy-agenthttps://chat.hyperledger.org/channel/indy-node Indy Wiki: https://wiki.hyperledger.org/projects/indy Indy WG Agendas: https://drive.google.com/open?id=1wNnp1pPS6-1Y4B2oygfI6tOU4eHGmpRH Indy Videos: https://drive.google.com/open?id=1Z8-jR7hnXb57fufE0OXxIfeUn05zNmRt Indy WG Meeting Recordings: https://drive.google.com/open?id=0B_NJV6eJXAA1UlJZMXd3cm1zNDQ

Sean_Bohan (Wed, 25 Apr 2018 21:41:08 GMT):
Really excited to share some of the hard work that Daniel and Janko and the team put together for the community. CHECK OUT INDY HOW-TOs HERE: https://github.com/hyperledger/indy-sdk/tree/master/doc/how-tos This folder contains short tutorials demonstrating how to accomplish common tasks with the Indy SDK. For best results, proceed through these in order: Write a DID and Query Its Verkey Rotate a Key Save a Schema and Cred Def Issue a Credential Negotiate a Proof Send a Secure Message

IVictorFeng (Thu, 26 Apr 2018 02:31:48 GMT):
Has joined the channel.

jackson18 (Thu, 26 Apr 2018 14:00:38 GMT):
Sweet... good work Daniel and Janko!

arthurgreef (Thu, 26 Apr 2018 15:17:48 GMT):
Has joined the channel.

NihadOgresevic (Thu, 26 Apr 2018 15:30:43 GMT):
Has joined the channel.

nuxibyte (Thu, 26 Apr 2018 15:34:04 GMT):
Has joined the channel.

Wesley2018 (Fri, 27 Apr 2018 00:41:40 GMT):
Has joined the channel.

jastisriradheshyam (Fri, 27 Apr 2018 08:16:45 GMT):
Has joined the channel.

Susmit (Fri, 27 Apr 2018 09:45:51 GMT):
Has joined the channel.

Susmit (Fri, 27 Apr 2018 09:47:02 GMT):
Hi All, pretty new to Indy, require a quick clarification or feedback on the thought while using Indy

Susmit (Fri, 27 Apr 2018 09:47:13 GMT):
Indy seems more a DLT rather to a blockchain to hold data and share it. Still one can store the truth (i.e. hash) in blockchain and actual data safe in Indy’s wallet for sharing.

Susmit (Fri, 27 Apr 2018 09:47:59 GMT):
Something like using Fabric +Indy together

Susmit (Fri, 27 Apr 2018 09:49:20 GMT):
any views on the thought will be great

drummondreed (Mon, 30 Apr 2018 00:19:26 GMT):
@Susmit Indy is indeed a DLT for decentralized identity. Note that by *by itself* is it not designed to share identity data. Rather it is designed to hold the decentralized identifiers (DIDs) and DID documents containing the public keys and agent endpoints of the parties to a digital relationship (called a connection). Those parties can then exchang verifiable credentials to establish trust.

drummondreed (Mon, 30 Apr 2018 00:21:19 GMT):
So as a identity utility, Indy can really be used by any other application or system that needs decentralized identity, very much like any Web application can use DNS (this is not to say that Indy serves the same purpose as DNS—it's just an analogy).

drummondreed (Mon, 30 Apr 2018 00:22:19 GMT):
Certainly other blockchain applications based on Fabric that need identity can use Indy, so indeed I anticipate Fabric + Indy will be a common pattern.

sudomann (Mon, 30 Apr 2018 04:24:18 GMT):
Has joined the channel.

sudomann (Mon, 30 Apr 2018 04:25:32 GMT):
@drummondreed So this Sovrin I've seen mention of...built on top Hyperledger Indy for sharing this identity data for validation purposes, etc?

drummondreed (Mon, 30 Apr 2018 04:27:33 GMT):
@sudomann Yes, the simple way to put it is that Sovrin is a global public utility for self-sovereign identity in which all the validators of the Sovrin public ledger run Indy-Node and all the hosts of private identity agents run Indy-SDK (or the equivalent). For full details on Sovrin, see the Sovrin white paper: https://sovrin.org/library/sovrin-protocol-and-token-white-paper.

sudomann (Mon, 30 Apr 2018 04:27:57 GMT):
neat, thank you

Susmit (Mon, 30 Apr 2018 05:54:29 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=YMqYcbngnCifwsWHT) @drummondreed Thanks for the clarification! Believe there is no direct integration with other blockchain platform (i.e. like Fabric). But to integrate i guess one has to create identity in Indy and have a mapping against the blockchain identity by managing it via a custom application.

drummondreed (Mon, 30 Apr 2018 05:56:00 GMT):
@Susmit Yes, direct integration is definitely not required. But if necessary you could build it into the other blockchain, or into apps that use the other blockchain.

Susmit (Mon, 30 Apr 2018 07:48:32 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=MDYRTRDC9X8NN6xTS) @drummondreed Thanks for the clarification!

mbwhite (Mon, 30 Apr 2018 13:19:32 GMT):
Has joined the channel.

dbluhm (Mon, 30 Apr 2018 15:23:39 GMT):
Has joined the channel.

dbluhm (Mon, 30 Apr 2018 15:25:09 GMT):
Out of curiosity, are there any plans to eventually rewrite plenum in Rust? Once the async/await RFC makes it into stable, it doesn't seem like all that outlandish of an idea.

esplinr (Mon, 30 Apr 2018 23:51:01 GMT):
@dbluhm We've discussed it. It's a large project that would prevent us doing other things we want to do. We have been discussing how we can do it in pieces.

esplinr (Mon, 30 Apr 2018 23:51:18 GMT):
But it's not something we are considering for the short term.

esplinr (Mon, 30 Apr 2018 23:52:07 GMT):
@dbluhm When I say "we", I'm referring to the Evernym development team that is currently the largest contributor. We aren't against someone else doing the effort.

esplinr (Mon, 30 Apr 2018 23:58:57 GMT):
@sudomann The Sovrin network is likely to be the first production deployment of Hyperledger Indy code. The expectation is that there will eventually be other networks built with the Indy code in addition to Sovrin.

Susmit (Tue, 01 May 2018 08:07:59 GMT):
In context of GDPR's right to forget have a query. Once an identity data is shared by creating DIDs with any other party, is there a way that the access to that data could be revoked.

nuxibyte (Tue, 01 May 2018 10:07:34 GMT):
Hi guys, this might be a dumb question (possibly being asked in the wrong place) but am I right in thinking that at the moment I can run up indy in docker for myself and create DIDs and interact with it like the Alice sample (which I have done and was great) - but I can't then do that with Sovrin?

nuxibyte (Tue, 01 May 2018 11:30:01 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=ghH2pvauT2GNWJdnq) Does this mean that if were to create an indy network and make it public it would somehow connect to sovrin or would it be an individual network where users could just choose between it and sovrin? I feel like I'm missing the link between indy and sovrin and why you would want to run a separate indy node detached from sovrin.

nuxibyte (Tue, 01 May 2018 11:30:01 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=ghH2pvauT2GNWJdnq) Does this mean that if I were to create an indy network and make it public it would somehow connect to sovrin or would it be an individual network where users could just choose between it and sovrin? I feel like I'm missing the link between indy and sovrin and why you would want to run a separate indy node detached from sovrin.

esplinr (Tue, 01 May 2018 14:19:03 GMT):
@nuxibyte If you create an instance of an Indy network, it will run alongside the Sovrin network and be an alternative. Applications you write against that network should also be able to run on the Sovrin network.

esplinr (Tue, 01 May 2018 14:19:24 GMT):
@nuxibyte If you use the Indy SDK to develop an application, it should run on the Sovrin network.

esplinr (Tue, 01 May 2018 14:19:57 GMT):
And it isn't a dumb question, and it isn't in the wrong place. grin

esplinr (Tue, 01 May 2018 14:21:35 GMT):
The goal of the Sovrin Foundation is to build a network that people want to use as a global public utility such that they don't want to set up alternative networks. Running your own network will continue to be useful for development. But people might find some reason to run private identity networks.

nuxibyte (Tue, 01 May 2018 16:58:04 GMT):
@esplinr thanks.

nuxibyte (Tue, 01 May 2018 17:08:31 GMT):
@esplinr So if I want to create an iOS app that allows users to create a DID and 'connect' with other users my options are to develop it using my own dev Indy network, then release it on my production Indy network? There is no option for me to release an app that can connect to the Sovrin network at the moment. Is that right? I found some info on Trust Anchors. Is it possible to become a Trust Anchor on the Sovrin network instead of running my own network?

esplinr (Tue, 01 May 2018 17:09:04 GMT):
Sorry, that wasn't what I was trying to say.

esplinr (Tue, 01 May 2018 17:09:22 GMT):
The preferred approach would be to develop it against the Sovrin test network, and release it against the Sovrin production network (when it's eventually live).

esplinr (Tue, 01 May 2018 17:09:35 GMT):
But you might consider setting up a test network for certain development tasks.

esplinr (Tue, 01 May 2018 17:10:21 GMT):
And some people think it might be useful to have their own private Indy networks for specific production use cases, but those of us at Sovrin are skeptical that would be better than just using Sovrin.

esplinr (Tue, 01 May 2018 17:11:19 GMT):
It is possible to be a Trust Anchor on the Sovrin network. The goal of the Sovrin network is that you can issue claims for others using that network to verify.

esplinr (Tue, 01 May 2018 17:11:44 GMT):
@nuxibyte

esplinr (Tue, 01 May 2018 17:13:11 GMT):
The summary is that most people will want to learn how to use Indy SDK against Sovrin. It's going to be a smaller community that want to learn how to run Indy directly. But we don't want to discourage that community by talking like Sovrin is the only use case.

nuxibyte (Tue, 01 May 2018 17:15:06 GMT):
that makes sense. i didn't realise there was a test network. how is that accessed? is that currently available?

esplinr (Tue, 01 May 2018 17:16:12 GMT):
It's currently available, but I don't know all the policies. (I've only been on the job a couple of weeks.)

esplinr (Tue, 01 May 2018 17:16:20 GMT):
_goes hunting around to see what information is available_

nuxibyte (Tue, 01 May 2018 17:16:36 GMT):
:D :+1_tone3:

esplinr (Tue, 01 May 2018 17:17:50 GMT):
As an FYI, the forums specific to the Sovrin network are here: https://forum.sovrin.org/

esplinr (Tue, 01 May 2018 17:18:19 GMT):
https://forum.sovrin.org/t/testing-on-the-sovrin-test-network-stn/643

nuxibyte (Tue, 01 May 2018 17:19:19 GMT):
perfect. thanks.

lcinacio (Wed, 02 May 2018 07:12:49 GMT):
@esplinr Thanks for sharing the Sovrin test network link. Do you know where I can find its pool transactions genesis file?

lcinacio (Wed, 02 May 2018 07:33:44 GMT):
I read news about ING's improvements to zero-knowledge proof such as range proofs and set membership. Are there any plans to add these features to Indy?

the_identity_guy (Wed, 02 May 2018 18:22:45 GMT):
Is there any document or architectural diagram in regards to identity hubs?

SteveGoob (Wed, 02 May 2018 19:42:21 GMT):
Has joined the channel.

jljordan_bcgov (Wed, 02 May 2018 21:41:47 GMT):
@the_identity_guy https://github.com/decentralized-identity/hubs/blob/master/explainer.md

Susmit (Thu, 03 May 2018 04:12:03 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=fms4opB8sZj9FEJPX) @drummondreed If you could share your views will be quite helpful

hcsatish (Thu, 03 May 2018 09:41:41 GMT):
Has joined the channel.

LucW (Thu, 03 May 2018 13:36:56 GMT):
Has joined the channel.

LucW (Thu, 03 May 2018 13:38:57 GMT):
Hey guys, I am trying to install libindy 1.4.0 from the repo.sovrin.org website, but it uses libssl1.0.0 and other old, deprecated dependencies. Is it outdated or am I missing something ?

Sean_Bohan (Thu, 03 May 2018 13:50:22 GMT):
Hyperledger Indy WG Call AGENDA: Shared Roadmap (RichardE) Release Plan SDK 1.4 (Slava) Release Notes and Other Docs (Richard) Architectural/Maintainer Docs (Richard & Nathan) Open Conversation When: TODAY - May 3, 2018 8amPT, 9amMT, 11amET, 3pm GMT Where: Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/232861185 Or iPhone one-tap (US Toll): +16465588656,232861185# or +14086380968,232861185# Or Telephone: Dial: +1 646 558 8656 (US Toll) or +1 408 638 0968 (US Toll) Meeting ID: 232 861 185 International numbers available: https://zoom.us/zoomconference?m=a0jD_rTMnh0ZYGQDOKPCNrK_0dP7WPfp1 Hyperledger Indy Community Indy Rocketchat: https://chat.hyperledger.org/channel/indyhttps://chat.hyperledger.org/channel/indy-agenthttps://chat.hyperledger.org/channel/indy-node Indy Wiki: https://wiki.hyperledger.org/projects/indy Indy WG Agendas: https://drive.google.com/open?id=1wNnp1pPS6-1Y4B2oygfI6tOU4eHGmpRH Indy Videos: https://drive.google.com/open?id=1Z8-jR7hnXb57fufE0OXxIfeUn05zNmRt Indy WG Meeting Recordings: https://drive.google.com/open?id=0B_NJV6eJXAA1UlJZMXd3cm1zNDQ

esplinr (Thu, 03 May 2018 14:27:46 GMT):
@lcinacio We have discussed some potential improvements to zero-knowledge proofs, but I don't know about ING's specific work. If you create an improvement request JIRA we can discuss it in more concrete terms.

esplinr (Thu, 03 May 2018 14:28:28 GMT):
@lcinacio Unfortunately, I don't yet know details such as the genesis file.

esplinr (Thu, 03 May 2018 14:28:28 GMT):
@lcinacio Unfortunately, I don't yet know details such as the genesis file. I was hoping someone else would answer.

mboyd (Thu, 03 May 2018 16:34:02 GMT):
@LucW Can you describe to me what OS you're trying to install on, what errors are showing up, and where in your installation process you're getting them?

mboyd (Thu, 03 May 2018 16:35:09 GMT):
You've probably already seen it, but here is the link for installation instructions on different OS's: https://github.com/hyperledger/indy-sdk/#installation

the_identity_guy (Thu, 03 May 2018 16:57:40 GMT):
@jljordan_bcgov thank you for the link, its very useful. I have actually seen that one was hoping to find more documents on it. I like to develop a simple prototype for it

cadi 2 (Thu, 03 May 2018 17:53:47 GMT):
Has joined the channel.

srki (Thu, 03 May 2018 23:29:29 GMT):
Has joined the channel.

LucW (Fri, 04 May 2018 09:44:36 GMT):
@mboyd I have a Debian Stretch (the last one). The output, when I try to install the .deb package of libindy is : "libindy depends on libssl1.0.0" and "libindy depends on libsqlite0", however these required versions are too old, which is why my system won't install them thus downgrade the currently installed versions

srki (Fri, 04 May 2018 15:40:29 GMT):
Any idea how to troubleshoot java.util.concurrent.ExecutionException: org.hyperledger.indy.sdk.InvalidStateException: The SDK library experienced an unexpected internal error. on openPoolLedger?

srki (Fri, 04 May 2018 15:41:46 GMT):
api seems to work find with createWallet, openWallet, createAndStoreMyDid, but when it tries to open pool ledger, it throws this exception which seems to be coming from the native library.

srki (Fri, 04 May 2018 15:43:50 GMT):
I guess one thing to note is that I'm using Java API wrapper.

srki (Fri, 04 May 2018 15:43:50 GMT):
I guess one thing to note is that I'm using Java API wrapper and am on Mac OS X.

AxelNennker (Fri, 04 May 2018 17:57:29 GMT):
Has joined the channel.

samadsajanlal (Fri, 04 May 2018 18:37:44 GMT):
Has joined the channel.

esplinr (Fri, 04 May 2018 20:40:32 GMT):
@eramitg Did you end up picking a project for your dissertation? @ngeorge and @lovesh had some ideas for you.

eramitg (Sat, 05 May 2018 01:40:18 GMT):
hey @esplinr just exploring in health sector right now ...if you assign some work then it will be hihly appreciate to work.

eramitg (Sat, 05 May 2018 01:40:18 GMT):
hey @esplinr just exploring in health sector right now ...if you assign some work then it will be highly appreciate to work.

nathanaw (Sat, 05 May 2018 02:26:44 GMT):
Has joined the channel.

nathanaw (Sat, 05 May 2018 02:27:14 GMT):
Hi friends! I am Nathan Aw from Singapore. This is urgent, I need to do a hyperledger indy demo to students in Singapore to get them excited about indy. Any resources will be very useful, please? nathan.mk.aw@gmail.com

Aswath8687 (Sat, 05 May 2018 03:35:41 GMT):
Has joined the channel.

eramitg (Sat, 05 May 2018 06:09:52 GMT):
Hey @ngeorge and @lovesh ,Please share with me the idea which should i work related to blockchain for my research work.

kdenhartog (Sat, 05 May 2018 08:17:00 GMT):
In order to help people get setup and work through setup bugs with Indy, I've decided to start hosting weekly sessions for debugging. If you are having troubles getting something setup and would like some help DM me and we can schedule some time at night or on the weekends where I can help you. I am MST which is UTC-7.

kdenhartog (Sat, 05 May 2018 08:17:00 GMT):
In order to help people get setup and work through setup bugs with Indy, I've decided to start hosting video sessions by appointment for debugging. If you are having troubles getting something setup and would like some help DM me and we can schedule some time at night or on the weekends where I can help you. I am MST which is UTC-7. My only request is that in the future you will pay it forward in the future, by helping others debug as well.

kdenhartog (Sat, 05 May 2018 08:17:00 GMT):
In order to help people get setup and work through setup bugs with Indy, I've decided to start hosting video sessions by appointment for debugging. If you are having troubles getting something setup and would like some help DM me and we can schedule some time at night or on the weekends where I can help you. I am MST which is UTC-7. My only request is that in the future you will pay it forward in the future, by helping others debug as well.

rjones (Sat, 05 May 2018 08:26:48 GMT):
that is very generous of you, @kdenhartog

kdenhartog (Sat, 05 May 2018 08:30:02 GMT):
I figure it will help us to resolve common issues people run into, as well as help build the community. If you wouldn't mind could you pin this message so I don't have to constantly post it?

rjones (Sat, 05 May 2018 08:32:26 GMT):
I'll let @Sean_Bohan make that decision, I don't want to run his channel :)

kdenhartog (Sat, 05 May 2018 08:51:15 GMT):
Oh gotcha, I saw the admin role and jumped the gun. I'll talk with him about this in more detail because I'm sure him and @nage will have some suggestions on how to do this efficiently.

mboyd (Sat, 05 May 2018 16:48:59 GMT):
@nathanaw check out some of the resources in https://github.com/hyperledger/indy-sdk/

umtyzc (Sat, 05 May 2018 18:56:41 GMT):
Has joined the channel.

drummondreed (Mon, 07 May 2018 00:20:10 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=YsMwPWviHdkfEvLCk) @Susmit Sorry to be slow; was heads down last week. The short answer, using GDPR terminology, is that if a data subject shares data with a data controller using pairwise pseudonymous DIDs (which is the default in the Sovrin ecosystem), then the data subject can always: a) send a deletion request to the data controller via the private encrypted DID channel to delete any shared data, and b) revoke the pairwise pseudonymous DID assigned to the data controller. So IMHO it's a near-perfect way to implement the right to be forgotten.

uvaraj6 (Mon, 07 May 2018 05:57:28 GMT):
Has joined the channel.

kdenhartog (Mon, 07 May 2018 14:24:33 GMT):

jljordan_bcgov (Mon, 07 May 2018 16:14:44 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=58A8rXbnKoDRxzJ4s) @the_identity_guy Try @TelegramSam for more details and github code repors

jljordan_bcgov (Mon, 07 May 2018 16:14:44 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=58A8rXbnKoDRxzJ4s) @the_identity_guy Try @TelegramSam for more details and github code repos

JonGeater (Mon, 07 May 2018 20:52:15 GMT):
Has joined the channel.

JonGeater (Mon, 07 May 2018 20:55:20 GMT):
Hi guys, just chatting on the #crypto-lib channel and the topic of threshold signatures came up: I understand this is a need of the Indy project. I and my team have some experience of implementing TS and plumbing it behind standard crypto libs. Who ( @nage ? Anyone else?) can express the requirements and wants to chat about it? Would be good to see if we have a quick route to doing what’s needed.

esplinr (Mon, 07 May 2018 21:00:17 GMT):
@JonGeater If I understand correctly, threshold signatures would allow a definition that says that a credential is only valid if it is signed by a certain number of independent issuers?

JonGeater (Mon, 07 May 2018 21:05:16 GMT):
That rather depends on the original generation/distribution/sharing

JonGeater (Mon, 07 May 2018 21:07:46 GMT):
We use it in a scheme where the holder of the key can’t trust individual memories/processing units (because they’re really paranoid about rowhammers or spectre or something, or because they’re in cloud/leases infrastructure). So we share the key processing around to protect the material but allow all participants to play effectively under the orders of a single issuer.

JonGeater (Mon, 07 May 2018 21:08:41 GMT):
You can add AuthZ logic to the participants of course, at which point your assertion works, subject to control of the initial m-of-n quorum

JonGeater (Mon, 07 May 2018 21:10:05 GMT):
We (Thales, not #crypto-lib) are working in some tech to add provably secure generation/share selection, which would complete your needs I think

JonGeater (Mon, 07 May 2018 21:12:05 GMT):
What’s the concrete implementation you have in mind? What actor would issue the request to use the private part of the credential?

esplinr (Mon, 07 May 2018 21:37:45 GMT):
I don't have a concrete implementation in mind. I'm fairly new and trying to get educated.

esplinr (Mon, 07 May 2018 21:38:08 GMT):
The architects I work with are in Europe, so aren't online. But I'll ask them for more details.

esplinr (Mon, 07 May 2018 21:38:33 GMT):
@JonGeater I appreciate your offer to assist.

nage (Mon, 07 May 2018 21:48:57 GMT):
In addition to @ashcherbakov and @gudkov, @jlaw 1 @lovesh @danielhardman and @brentzundel all might have meaningful feedback on threshold signature schemes that would enable compatibility with anonymous credentials or broaden our support for HSMs. We’ve been think by about how the ZKLang work we’re doing would fit into the HL shared crypto library, which has raised these same kinds of questions.

nage (Mon, 07 May 2018 21:48:57 GMT):
In addition to @ashcherbakov and @gudkov, @jlaw 1 @lovesh @danielhardman and @brentzundel all might have meaningful feedback on threshold signature schemes that would enable compatibility with anonymous credentials or broaden our support for HSMs. We’ve been thinking about how the ZKLang work we’re doing would fit into the HL shared crypto library, which has raised these same kinds of questions.

brentzundel (Mon, 07 May 2018 21:49:01 GMT):
Has joined the channel.

JonGeater (Mon, 07 May 2018 21:50:02 GMT):
OK cool. I’m London timezone too, happy to have a quick call with them if it’s easier.

JonGeater (Mon, 07 May 2018 21:50:36 GMT):
We have a demonstration of what we’ve done (minus the cool generation stuff - that’s a paper pending) hosted in Amazon that we can walk through maybe

JonGeater (Mon, 07 May 2018 21:51:29 GMT):
(Also happy to be told to have the rest of this discussion on the crypto lib channel if that’s more appropriate)

nage (Mon, 07 May 2018 21:51:35 GMT):
Many of us are in the Mountain time zone. I will be in Munich for EIC next week and am scheduling a few meetings next Monday before the conference starts in earnest.

nage (Mon, 07 May 2018 21:51:35 GMT):
Many of us are in the US Mountain time zone. I will be in Munich for EIC next week and am scheduling a few meetings next Monday before the conference starts in earnest.

nage (Mon, 07 May 2018 21:53:40 GMT):
We currently use our threshold signatures to allow the validation pool to sign state proofs that allow read requests to be serviced by “observer nodes”

nage (Mon, 07 May 2018 22:17:58 GMT):
In terms of a signing API, I think more of the right folks are paying attention in the #crypto-lib channel. If we want to talk about extending indy's use of threshold signatures, lets talk here, if we are talking about a common shared API for those types of libraries, then #crypto-lib is the right home.

rjones (Mon, 07 May 2018 22:54:58 GMT):
Are there any PRs stuck waiting on DCO bot, and DCO bot is not firing? if so, could you please add links here? https://github.com/probot/dco/issues/69 thank you

LoveshHarchandani (Mon, 07 May 2018 23:52:48 GMT):
@JonGeater indy-crypto (used by Indy validators) does not really provide a threshold signature scheme but a signature aggregation scheme called BLS (https://www.iacr.org/archive/asiacrypt2001/22480516.pdf). Indy-node uses the aggregate signature and knowledge of participating validators to implement the threshold.

LoveshHarchandani (Mon, 07 May 2018 23:52:48 GMT):
@JonGeater indy-crypto (used by Indy validators) does not really provide a threshold signature scheme but a signature aggregation scheme called BLS (https://www.iacr.org/archive/asiacrypt2001/22480516.pdf). Indy-node uses the aggregate signature and knowledge of participating validators to implement the threshold. So it is up to the verifier of the signature to decide the threshold, the aggregate just specifies the signature and the participants Can you share some details of your implementation, maybe a paper?

LoveshHarchandani (Mon, 07 May 2018 23:52:48 GMT):
@JonGeater indy-crypto (used by Indy validators) does not really provide a threshold signature scheme but a signature aggregation scheme called BLS (https://www.iacr.org/archive/asiacrypt2001/22480516.pdf). Indy-node uses the aggregate signature and knowledge of participating validators to implement the threshold. So it is up to the verifier of the signature to decide the threshold, the aggregate just specifies the signature and the participants. Can you share some details of your implementation, maybe a paper?

vidor (Mon, 07 May 2018 23:55:41 GMT):
Has joined the channel.

danielhardman (Tue, 08 May 2018 01:48:43 GMT):
^^ @MALodder

Susmit (Tue, 08 May 2018 05:21:09 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=eJwFJ3gzvEERw2i4i) @drummondreed Thanks for your reply! So even though data is shared but still it is in control of the owner. This deletion request could be specific to individual data been shared or it revokes access to all the data been shared on that channel?

drummondreed (Tue, 08 May 2018 05:23:31 GMT):
It could be either one. The best part (IMHO) is that it could be automated by the individual's Sovrin identity agent. Assuming the agent has been tracking what the individual has shared, the agent could just show the individual a menu of what the individual can choose to delete, and then send the appropriate deletion request (authenticated and encrypted) to the data controller.

Susmit (Tue, 08 May 2018 05:24:23 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=2vRsxgBGnQscpHiPQ) @drummondreed That's great!

Rohit-Thukral (Tue, 08 May 2018 11:48:42 GMT):
Has joined the channel.

Rohit-Thukral (Tue, 08 May 2018 11:52:37 GMT):
Hi All i am trying to create pool in hyperledger indy cli using the following command "pool create sandbox gen_txn_file=/etc/sovrin/sandbox.txn" but not sure how to create that genesis transaction file. Also if anyone have commands sample then please share that as well. Any help would be appreciated

daveryIBM (Tue, 08 May 2018 15:07:27 GMT):
If anyone here hasn't seen it yet, the identity team at IBM (@vinomaster) recently published a simple demo that transforms the Indy Alice tutorial into a nice web UI. You can check it out at https://github.com/IBM-Blockchain-Identity/indy-ssivc-tutorial

daveryIBM (Tue, 08 May 2018 15:07:46 GMT):
Shout out to BC Gov (@jljordan_bcgov, @swcurran, @nbrempel) for letting us use their work around verifiable organizations as a starting point. It was super helpful. I think anyone trying to stand up their own pool of Indy nodes to play with should just use von-network. Von-web is too useful not to have.

vinomaster (Tue, 08 May 2018 15:17:01 GMT):
:thumbsup:

swcurran (Tue, 08 May 2018 15:17:37 GMT):
@daveryIBM - way cool! The video in the repo is great.

swcurran (Tue, 08 May 2018 15:17:37 GMT):
@daveryIBM - way cool! The video in the repo is great. Can we post that to YouTube? Or can you?

the_identity_guy (Tue, 08 May 2018 15:17:58 GMT):
thank you @jljordan_bcgov

daveryIBM (Tue, 08 May 2018 15:19:21 GMT):
thanks @swcurran. I'll check to see if we have a better, approved spot to stick that video

esplinr (Tue, 08 May 2018 20:14:06 GMT):
I see a Hyperledger channel on

esplinr (Tue, 08 May 2018 20:14:06 GMT):
I see a Hyperledger channel on YouTube, but I don't know who owns it.

esplinr (Tue, 08 May 2018 20:14:39 GMT):
There isn't yet an Indy playlist. @daveryIBM: your video would make a great anchor to that playlist.

esplinr (Tue, 08 May 2018 20:15:09 GMT):
Evernym has some content I would like to publish there too, but I'm not organized enough to have it ready yet.

tkuhrt (Wed, 09 May 2018 00:00:50 GMT):
@esplinr : I am guessing one of my counterparts at The Linux Foundation owns the Hyperledger YouTube channel. Let me reach out to them.

tkuhrt (Wed, 09 May 2018 00:03:47 GMT):
@daveryIBM : Would you be interested in writing a blog post for hyperledger.org on the repository?

tkuhrt (Wed, 09 May 2018 00:04:06 GMT):
We are always looking for more developer-type content

tkuhrt (Wed, 09 May 2018 00:04:49 GMT):
You can find guest blog posting guidelines at: https://docs.google.com/document/d/1ANQdRlk1Pv214a-jxk5Yw_IYIYiQhDB7zELrIzQlbrU/edit

tkuhrt (Wed, 09 May 2018 00:04:49 GMT):
You can find guest blog guidelines at: https://docs.google.com/document/d/1ANQdRlk1Pv214a-jxk5Yw_IYIYiQhDB7zELrIzQlbrU/edit

mawi (Wed, 09 May 2018 10:04:59 GMT):
Has anyone used TouchID for encrypting data inside a default wallet implementation? Or is it only possible to use a passphrase?

danielhardman (Wed, 09 May 2018 11:42:03 GMT):
I submitted an RFC for indy, proposing a standard notation for SSI concepts. This supersedes a google doc that some of you may have seen, that was hyperlinked from the "How DIDS, Keys, Credentials, and Agents..." doc. Would love comments; maybe we can discuss on the community call on Thursday? https://github.com/hyperledger/indy-rfc/blob/master/text/ssi-notation.md

NihadOgresevic (Wed, 09 May 2018 12:44:55 GMT):
Before I say anything I just want to thank you guys for all the effort that you are putting into hyperledger indy. The problems that I am facing is basically understanding how the system works overall especially with indy-plenum. Also, use cases are fairly complex and there is only a few of them. I already spent a couple of weeks and I still have no clear idea of how basic registration and login of a user would look like with indy. If could provide more and simpler use cases so that people could evolve that would be great.

NihadOgresevic (Wed, 09 May 2018 12:44:55 GMT):
Before I say anything I just want to thank you guys for all the effort that you are putting into hyperledger indy. The problems that I am facing is basically understanding how the system works overall especially with indy-plenum. Also, use cases are fairly complex and there is only a few of them. I already spent a couple of weeks and I still have no clear idea of how basic registration and login of a user would look like with indy. If you could provide more and simpler use cases so that people could evolve from that it would be great.

NihadOgresevic (Wed, 09 May 2018 12:44:55 GMT):
Before I say anything I just want to thank you guys for all the effort that you are putting into hyperledger indy. The problems that I am facing are basically understanding how the system works overall especially with indy-plenum. Also, use cases are fairly complex and there is only a few of them. I already spent a couple of weeks and I still have no clear idea of how basic registration and login of a user would look like with indy. If you could provide more and simpler use cases so that people could evolve from that it would be great.

NihadOgresevic (Wed, 09 May 2018 12:44:55 GMT):
Before I say anything I just want to thank you guys for all the effort that you are putting into hyperledger indy. The problems that I am facing are basically understanding how the system works overall especially with indy-plenum. Also, use cases are fairly complex and there is only a few of them. I already spent a couple of weeks and I still have no clear idea of how basic registration and login of a user would look like with indy. If you could provide more and simpler use cases so that people could evolve from that it would be great.

daveryIBM (Wed, 09 May 2018 13:53:00 GMT):
@tkuhrt I'll take a look at the guidelines. Thanks for passing them along. We're trying to get a blog about that repo out on IBM channels right now, actually.

guido.santos (Wed, 09 May 2018 14:16:52 GMT):
Has joined the channel.

kdenhartog (Wed, 09 May 2018 14:21:20 GMT):
@NihadOgresevic I'd be willing to setup a call to walk through these things with you. This is something I've notice quite a few people are running into so by helping you we gain very valuable insights on how to improve. DM me and we can set up a time.

NihadOgresevic (Wed, 09 May 2018 18:05:32 GMT):
@kdenhartog wow thats awesome. Thanks man

NihadOgresevic (Wed, 09 May 2018 18:05:32 GMT):
@kdenhartog wow thats awesome. Thanks man

NihadOgresevic (Wed, 09 May 2018 18:05:32 GMT):
@kdenhartog thats awesome. Thanks man

mboyd (Wed, 09 May 2018 18:57:26 GMT):
@kdenhartog @NihadOgresevic I am working to make it easier to understand Indy and help developers get started with the code faster. I would be interested in joining your call as well if you are willing. We just had a working group call yesterday about how to improve the developer experience. I'd love to hear your feedback and also point you to any resources I can that can help. We've started a new channel for discussions of this nature at #indy-outreach. So far, the video posted by @daveryIBM is probably one of the [most concise and concrete use case examples of Indy](https://github.com/IBM-Blockchain-Identity/indy-ssivc-tutorial/raw/master/docs/video/IndyWorldVideo2.mp4).

kdenhartog (Wed, 09 May 2018 18:58:25 GMT):
certainly, I'm going to be busy for the next week, so we scheduled it for next friday at 9AM. I'll add you to the calendar invite

kdenhartog (Wed, 09 May 2018 18:58:55 GMT):
Also, I'll plan on putting you as an optional invite for all of these that way you can join anytime

mboyd (Wed, 09 May 2018 19:02:09 GMT):
I will also be doing a ~10 min presentation tomorrow in the Indy working group call on How to Contribute to Indy. I will cover how to start a test pool of nodes, how to run the getting started guide, and the right steps to make your first commits and PR with a list of good first bugs.

kdenhartog (Wed, 09 May 2018 19:10:24 GMT):
awesome! I'll be sure to join tomorrow. @NihadOgresevic I'd recommend joining this call as well. @Sean_Bohan typically posts the details on how to connect Thursday morning before the meetings. Here's the link to join https://zoom.us/j/232861185 I believe it will be at 5pm your time.

NihadOgresevic (Wed, 09 May 2018 21:55:36 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=Fzx2iQyHFn2PqQpK2) @kdenhartog Will do, thanks guys

NihadOgresevic (Wed, 09 May 2018 23:15:56 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=pR9ctpQk2eu9jPQ5G) @mboyd o my...did not see that one before thanks

nhelmy (Thu, 10 May 2018 07:41:56 GMT):
Has joined the channel.

pd93 (Thu, 10 May 2018 08:21:06 GMT):
Has joined the channel.

LuisMarado (Thu, 10 May 2018 10:00:35 GMT):
Has joined the channel.

beingsumit (Thu, 10 May 2018 11:46:39 GMT):
Has joined the channel.

rfitzpatrick (Thu, 10 May 2018 14:58:00 GMT):
Has joined the channel.

nfarley (Thu, 10 May 2018 15:21:08 GMT):
Has joined the channel.

mboyd (Thu, 10 May 2018 16:07:07 GMT):
Links from my presentation on the working group call: * [Indy-SDK Contributor Resources](https://github.com/michaeldboyd/indy-sdk/blob/6a2a94a3c520c36c9c8b2521136facbd97a36cd6/doc/contributor-resources.md) * [How to start a local pool of nodes with docker](https://github.com/hyperledger/indy-sdk#how-to-start-local-nodes-pool-with-docker) * [Run the Alice demo with python](https://github.com/hyperledger/indy-sdk/tree/master/samples/python) * [Most current and updated documentation resource for indy](https://wiki.hyperledger.org/projects/indy/documentation): Please add helpful links and resources to this page as you see fit. * [Good first bugs](https://jira.hyperledger.org/secure/RapidBoard.jspa?rapidView=149&projectKey=IS&view=planning.nodetail&quickFilter=447): Scroll down. This page can also be used to see what tickets everyone is working on. * [Indy Reference Agent](https://github.com/hyperledger/indy-agent) * [IBM Alice Demo](https://github.com/IBM-Blockchain-Identity/indy-ssivc-tutorial) Please message me if you have any questions about how to starting using the Indy codebase.

mboyd (Thu, 10 May 2018 16:07:07 GMT):
Links from my presentation on How to Contribute to Indy on the working group call: * [Indy-SDK Contributor Resources](https://github.com/michaeldboyd/indy-sdk/blob/6a2a94a3c520c36c9c8b2521136facbd97a36cd6/doc/contributor-resources.md) * [How to start a local pool of nodes with docker](https://github.com/hyperledger/indy-sdk#how-to-start-local-nodes-pool-with-docker) * [Run the Alice demo with python](https://github.com/hyperledger/indy-sdk/tree/master/samples/python) * [Most current and updated documentation resource for indy](https://wiki.hyperledger.org/projects/indy/documentation): Please add helpful links and resources to this page as you see fit. * [Good first bugs](https://jira.hyperledger.org/secure/RapidBoard.jspa?rapidView=149&projectKey=IS&view=planning.nodetail&quickFilter=447): Scroll down. This page can also be used to see what tickets everyone is working on. * [Indy Reference Agent](https://github.com/hyperledger/indy-agent) * [IBM Alice Demo](https://github.com/IBM-Blockchain-Identity/indy-ssivc-tutorial) Please message me if you have any questions about how to starting using the Indy codebase.

daveryIBM (Thu, 10 May 2018 17:51:06 GMT):
@swcurran @esplinr finally got that video published to YT https://www.youtube.com/watch?v=cz-6BldajiA

danielhardman (Thu, 10 May 2018 19:47:57 GMT):
As promised, I cleaned up the two RFC proposals for the indy-rfc repo. They are now unmerged PRs, ready for comment and github issues. You can find them here: Credential Revocation -- https://github.com/hyperledger/indy-rfc/pull/8 SSI Notation -- https://github.com/hyperledger/indy-rfc/pull/9 I incorporated suggestions from @swcurran and @gudkov about a slight change to directory structure. Would love feedback!

swcurran (Thu, 10 May 2018 19:52:31 GMT):
Looks good - I like the use of README.

swcurran (Thu, 10 May 2018 19:58:36 GMT):
One other thought re: numbering. What about making the protocol that to submit an RFC, you create an issue. That gives you the number to use for the RFC, which (a) gives a unique number, and (b) allows setting the directory name so that as @gudkov says, they are unchanged after acceptance. If the RFC is not accepted (PR not merged) no great harm - a number is list in the sequence numbers, but who cares as long as the numbers are unique. It also gives a place for "pre-RFC" discussion about what the shape/scope of the RFC should be before it is formalized into a PR. That is friendlier for a newbie coming in with an idea that they aren't 100% sure about and don't want to start with an RFC PR.

swcurran (Thu, 10 May 2018 19:58:36 GMT):
One other thought re: numbering. What about making the protocol that to submit an RFC, you create an issue. That gives you the number to use for the RFC, which (a) gives a unique number, and (b) allows setting the directory name so that as @gudkov says, they are unchanged after acceptance. If the RFC is not accepted (PR not merged) no great harm - a number is skipped in the list, but who cares as long as the numbers are unique. It also gives a place for "pre-RFC" discussion about what the shape/scope of the RFC should be before it is formalized into a PR. That is friendlier for a newbie coming in with an idea that they aren't 100% sure about and don't want to start with an RFC PR.

danielhardman (Thu, 10 May 2018 20:05:07 GMT):
@swcurran I like that idea (issue number = github number). However, it would make for sparse numbering if we used issues to track bugs or updates in RFCs, in addition to RFCs themselves...

swcurran (Thu, 10 May 2018 20:06:35 GMT):
It is a repo for RFCs, so less likely, but even so, I don't think sparse numbering matters. People know what ERC-725 is, even if they don't know about any other RFC.

gudkov (Fri, 11 May 2018 11:02:37 GMT):
Why just don't use ordering starting from 1? RFCs can be incremental.

swcurran (Fri, 11 May 2018 14:49:26 GMT):
@gudkov - I think the pre-RFC issue is helpful for conversation. The numbering from one is fine as long as we're good with th issue of two (or more) people creating their own PR in parallel with the same number. Likely no big deal.

json (Fri, 11 May 2018 16:46:48 GMT):
Has joined the channel.

rjones (Fri, 11 May 2018 22:05:46 GMT):
@gudkov @swcurran why not have someone request a number? that's how IETF RFCs work, right? the Indy committer committee can parcel them out and avoid all of this.

rjones (Fri, 11 May 2018 22:06:21 GMT):
call them yyyy-serial or something

rjones (Fri, 11 May 2018 22:06:21 GMT):
~call them yyyy-serial or something~

rjones (Fri, 11 May 2018 22:07:49 GMT):
Here are some thoughts: https://tools.ietf.org/html/draft-claise-semver-02

swcurran (Fri, 11 May 2018 22:34:22 GMT):
Or you can have github give you a number and avoid all this :-). Ethereum starts with issue numbers and it works. I personally think the github issue to suggest an RFC with some discussion is a good idea - it's a more welcoming approach for a new community member. But at the end of the day, any and all of these approaches works, so I'm good to leave it to @nage and team to decide.

rjones (Fri, 11 May 2018 22:54:45 GMT):
@swcurran agreed. I think it's either neat or a bug that GitHub PRs and issues share a namespace. Which one I think it is depends on the side of bed I get up on :)

anupkumar2009 (Sat, 12 May 2018 12:45:09 GMT):
Has joined the channel.

DST 32 (Sat, 12 May 2018 15:21:20 GMT):
Has joined the channel.

ramanagak (Mon, 14 May 2018 06:19:28 GMT):
Has joined the channel.

ramanagak (Mon, 14 May 2018 06:26:21 GMT):
Hi, In the script `https://github.com/hyperledger/indy-sdk/blob/57674d75adf2af63ed81dc4fcc53f73ffed7a39f/doc/design/005-dkms/getting-started/getting-started.ipynb`, When Acme creating a job application proof request, How it will know about `faber_transcript_cred_def_id` which is created at faber. The example given is a single script so all variables available there so the script knows about the varialbe. What if I created a seperate scripts for each entity, How it Acme knows to get that id. Please suggest me.

ramanagak (Mon, 14 May 2018 06:26:21 GMT):
Hi, In the script `https://github.com/hyperledger/indy-sdk/blob/57674d75adf2af63ed81dc4fcc53f73ffed7a39f/doc/design/005-dkms/getting-started/getting-started.ipynb`, When Acme creating a job application proof request, How it will know about `faber_transcript_cred_def_id` which is created at faber. The example given is a single script so all variables available there so the script knows about the varialbe. What if I created a seperate scripts for each entity, How Acme knows to get that id. Please suggest me.

alek (Mon, 14 May 2018 07:27:49 GMT):
Has joined the channel.

pimotte (Mon, 14 May 2018 07:35:42 GMT):
@ramanagak It would need to be communicated by Faber to the world, or to Acme. Note that this means that Acme specifically only accepts those attributes from that credential issued by Faber, so there needs to be some information exchange there

ramanagak (Mon, 14 May 2018 08:20:20 GMT):
Thanks @pimotte. So Faber needs to advertise it's id whenever created to the world or Acme.

ramanagak (Mon, 14 May 2018 08:20:20 GMT):
Thanks @pimotte , So Faber needs to advertise it's id whenever created to the world or Acme.

alex873 (Mon, 14 May 2018 09:38:54 GMT):
Has joined the channel.

alex873 (Mon, 14 May 2018 09:44:13 GMT):
Hi,

alex873 (Mon, 14 May 2018 10:52:01 GMT):
Hi, when the private data is stored off-ledger, how can you guarantee that it won't be deleted? As far as I understood, the ledger acts like a look-up service, and has a link to the private data which is stored somewhere else. However, for my specific use case, it shouldn't be possible that the private data f

alex873 (Mon, 14 May 2018 11:02:15 GMT):
Hi :) when the private data is stored off-ledger, how can you guarantee that it won't be deleted? As far as I understood, the ledger acts like a look-up service, and has a link to the private data which is stored somewhere else. However, for my specific use case, I need the following: - Private data should never get deleted or manipulated. So is it possible to store the private data encrypted on the ledger as well? - It should be possible, that some claims (e.g. data from a sensor) can be "written" into someone's identity, without the identity holder requesting them Would that be fine with Indy?

MALodder (Mon, 14 May 2018 15:22:32 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=Lq5CEhfwubzfMc8Zc) @pwalimbe You are probably trying to compile against the latest version of libsodium. Try compiling with lib sodium 1.0.12

NihadOgresevic (Mon, 14 May 2018 16:51:58 GMT):
What could be the reason for the poolLedger timeout? This is the error that I am getting

NihadOgresevic (Mon, 14 May 2018 16:52:01 GMT):
von-web_1 | ERROR|indy::services::pool | src/services/pool/mod.rs:426 | Pool worker thread finished with error Timeout von-web_1 | ERROR|indy::errors::indy | src/errors/indy.rs:68 | Casting error to ErrorCode: Timeout von-web_1 | _indy_loop_callback: Function returned error 307 von-web_1 | Traceback (most recent call last): von-web_1 | File "server.py", line 295, in von-web_1 | loop.run_until_complete(boot()) von-web_1 | File "/usr/lib/python3.5/asyncio/base_events.py", line 387, in run_until_complete von-web_1 | return future.result() von-web_1 | File "/usr/lib/python3.5/asyncio/futures.py", line 274, in result von-web_1 | raise self._exception von-web_1 | File "/usr/lib/python3.5/asyncio/tasks.py", line 241, in _step von-web_1 | result = coro.throw(exc) von-web_1 | File "server.py", line 112, in boot von-web_1 | await pool.open() von-web_1 | File "/home/indy/.local/share/virtualenvs/server-8XoupS0v/lib/python3.5/site-packages/von_agent/nodepool.py", line 125, in open von-web_1 | self._handle = await pool.open_pool_ledger(self.name, None) von-web_1 | File "/home/indy/.local/share/virtualenvs/server-8XoupS0v/lib/python3.5/site-packages/indy/pool.py", line 82, in open_pool_ledger von-web_1 | open_pool_ledger.cb) von-web_1 | File "/usr/lib/python3.5/asyncio/futures.py", line 361, in __iter__ von-web_1 | yield self # This tells Task to wait for completion. von-web_1 | File "/usr/lib/python3.5/asyncio/tasks.py", line 296, in _wakeup von-web_1 | future.result() von-web_1 | File "/usr/lib/python3.5/asyncio/futures.py", line 274, in result von-web_1 | raise self._exception von-web_1 | indy.error.IndyError: ErrorCode.PoolLedgerTimeout

NihadOgresevic (Mon, 14 May 2018 16:53:09 GMT):
indy-ssivc-tutorial sample von_network

NihadOgresevic (Mon, 14 May 2018 16:56:42 GMT):
Also, the same error is happening with samples from indy-sdk repository

daveryIBM (Mon, 14 May 2018 17:31:33 GMT):
@NihadOgresevic Do you have any logs from the nodes? If they crashed or weren't able to start, then it makes sense for von-web to time out trying to reach them

daveryIBM (Mon, 14 May 2018 17:31:33 GMT):
@NihadOgresevic Do you have any logs from the nodes? If they crashed or weren't able to start, then it makes sense for von-web to time out trying to reach them. Same goes for the indy sdk sample code

NihadOgresevic (Mon, 14 May 2018 19:36:29 GMT):
@daveryIBM it seems as if nodes are running because they start listening the problem happens when the nodes try to establish connection.

NihadOgresevic (Mon, 14 May 2018 19:40:12 GMT):
Which log output? I only have console output.

daveryIBM (Mon, 14 May 2018 19:40:46 GMT):
you'd see logs tagged with `von-node_1` or something similar at the front

daveryIBM (Mon, 14 May 2018 19:41:08 GMT):
are you using a service called `firewalld` on your docker host?

daveryIBM (Mon, 14 May 2018 19:41:54 GMT):
that gave me a lot of problems. It seems that `firewalld` deletes the `DOCKER` iptable and prevents docker from being able to set up networking rules that allow container networking to function properly

daveryIBM (Mon, 14 May 2018 19:42:28 GMT):
`systemctl status firewalld`. If it's running, the best thing I know to do is `systemctl stop firewalld; systemctl restart docker`

daveryIBM (Mon, 14 May 2018 19:43:31 GMT):
that's just a guess based on the issues I've had though. You're issue may be something different

daveryIBM (Mon, 14 May 2018 19:44:21 GMT):
you could always try `docker exec -ti bash` to get into the container and pinging the other containers to see if you can reach them

daveryIBM (Mon, 14 May 2018 19:44:36 GMT):
just some stuff to try

NihadOgresevic (Mon, 14 May 2018 20:05:25 GMT):
Is there any known issues when trying to run this example on Mac

NihadOgresevic (Mon, 14 May 2018 20:05:25 GMT):
Is there any known issues when trying to run this example on Mac?

NihadOgresevic (Mon, 14 May 2018 20:06:20 GMT):
firewalld service is not installed

toddinpal (Mon, 14 May 2018 20:28:06 GMT):
Does anyone know of any good write-ups comparing/contrasting Indy with SecureKey?

NihadOgresevic (Mon, 14 May 2018 22:35:17 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=YbWayGSb4iZPSJ2CE) @daveryIBM It was an issues with docker-compose not really sure why...anyway thanks for the effort.

NihadOgresevic (Mon, 14 May 2018 22:35:17 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=YbWayGSb4iZPSJ2CE) @daveryIBM It was an issue with docker-compose not really sure why...anyway thanks for the effort.

kevin.chan (Tue, 15 May 2018 00:31:26 GMT):
Has joined the channel.

team-securekey (Tue, 15 May 2018 01:41:17 GMT):
Has joined the channel.

Neumann347 (Tue, 15 May 2018 18:35:49 GMT):
Has joined the channel.

am (Tue, 15 May 2018 19:10:00 GMT):
Has joined the channel.

Kelattar (Wed, 16 May 2018 13:08:39 GMT):
hello ! I am working on the python wrappers, but I have some questions : when we create a ledger configuration, what does it exactly mean ? more like a copy of the ledger or a copy of all information about nodes .. ?, and when the prover receives a claim from the issuer and stores it with the functions anoncreds.prover_store_credential where is it stored ? I can't find any of the exchanges done through python wrappers stored somewhere ? can you please help me figure this out ?

resndes (Wed, 16 May 2018 15:18:37 GMT):
Has joined the channel.

jonathanreynolds (Wed, 16 May 2018 17:36:07 GMT):
Has joined the channel.

mboyd (Wed, 16 May 2018 18:13:21 GMT):
When working with the python wrapper, how to we target compiled libindy.so that we make from the indy-sdk source code? Right now the wrapper is searching for the natively installed libindy package for ubuntu

ramanagak (Thu, 17 May 2018 05:30:34 GMT):
ramanagak

tusharg (Thu, 17 May 2018 11:02:40 GMT):
Has joined the channel.

ramanagak (Thu, 17 May 2018 12:21:38 GMT):
Hi, Can we use Indy to implement the login mechanism like social logins?

RomanaL 1 (Thu, 17 May 2018 13:41:59 GMT):
Has joined the channel.

kdenhartog (Thu, 17 May 2018 21:50:37 GMT):
@ramanagak this feature isn't built completely yet but there's a proof of concept that exists for it here https://github.com/bcgov/did-auth-extension

kdenhartog (Thu, 17 May 2018 21:51:31 GMT):
This was built by a @peacekeeper and funded by the guys at BCGov

kdenhartog (Thu, 17 May 2018 21:51:31 GMT):
This was built by @peacekeeper and funded by the guys at BCGov

Edelweiss (Fri, 18 May 2018 03:09:30 GMT):
Has joined the channel.

ramanagak (Fri, 18 May 2018 05:14:31 GMT):
Thanks @kdenhartog

Sarah.Conway (Fri, 18 May 2018 12:38:27 GMT):
Has joined the channel.

joshuajeeson (Fri, 18 May 2018 16:04:34 GMT):
Has joined the channel.

joshuajeeson (Fri, 18 May 2018 16:06:06 GMT):
i am trying to follow the instructions of getting started using docker containers

joshuajeeson (Fri, 18 May 2018 16:06:47 GMT):
when I run the command "./pool_start.sh 4 10.0.0.2,10.0.0.3,10.0.0.4,10.0.0.5 10 9701" but I end up with 4 containers that arnt running

joshuajeeson (Fri, 18 May 2018 16:07:15 GMT):
would help to understand what I am missing in the documentation

stanleyc-trustscience (Fri, 18 May 2018 17:17:50 GMT):
Hello, can someone help me understand the difference in roles? https://github.com/hyperledger/indy-node/blob/master/indy_common/roles.py ``` class Roles(Enum): # These numeric constants CANNOT be changed once they have been used, # because that would break backwards compatibility with the ledger # Also the numeric constants CANNOT collide with the roles in plenum TRUSTEE = Roles.TRUSTEE.value STEWARD = Roles.STEWARD.value TGB = "100" TRUST_ANCHOR = "101" ```

stanleyc-trustscience (Fri, 18 May 2018 17:26:37 GMT):
eg, Trustee vs Steward vs TGB vs TrustAnchor

kdenhartog (Fri, 18 May 2018 22:05:57 GMT):
Trustee is Defined in section 6.2 here https://github.com/hyperledger/indy-sdk/master/tree/doc/design/005-dkms/DKMS Design and Architecture V3.md

kdenhartog (Fri, 18 May 2018 22:06:51 GMT):
Steward is defined in Step 1 here https://github.com/hyperledger/indy-sdk/blob/master/doc/getting-started/getting-started.md

kdenhartog (Fri, 18 May 2018 22:06:51 GMT):
Trust Anchor is defined in Step 1 here https://github.com/hyperledger/indy-sdk/blob/master/doc/getting-started/getting-started.md

kdenhartog (Fri, 18 May 2018 22:16:04 GMT):
A list of all the definitions can more easily be found in the Sovrin glossary https://docs.google.com/document/d/1giOzpTFXypJ6bAUp_6g93kYOEiNa5eWI1KeIg6wb598/edit

kdenhartog (Fri, 18 May 2018 22:19:34 GMT):
For what more information about what each roles' capabilities are, see this link https://docs.google.com/spreadsheets/d/1TWXF7NtBjSOaUIBeIH77SyZnawfo91cJ_ns4TR-wsq4/edit#gid=0

stanleyc-trustscience (Fri, 18 May 2018 22:47:18 GMT):
@brendonakay Thank you

brendonakay (Fri, 18 May 2018 22:47:18 GMT):
Has joined the channel.

stanleyc-trustscience (Fri, 18 May 2018 22:47:27 GMT):
@kdenhartog Thank you

finkga (Sun, 20 May 2018 23:36:39 GMT):
Has joined the channel.

finkga (Sun, 20 May 2018 23:37:17 GMT):
I'm on a Mac:

finkga (Sun, 20 May 2018 23:37:17 GMT):
I'm on a Mac: $ uname -a Darwin we29277 17.5.0 Darwin Kernel Version 17.5.0: Fri Apr 13 19:32:32 PDT 2018; root:xnu-4570.51.2~1/RELEASE_X86_64 x86_64 and I think I successfully built Indy, but when I try to use it in Python the import fails. And when I run cargo test it fails in ledger_demo_works with, "thread 'ledger_demo_works' panicked at 'called `Result::unwrap()` on an `Err` value: Timeout', libcore/result.rs:945:5." I'm not familiar with Rust or with the Indy codebase. Anything obviously wrong I should note?

finkga (Sun, 20 May 2018 23:37:17 GMT):
Cargo test 'ledger_demo_works' fails. I'm on a Mac: $ uname -a Darwin we29277 17.5.0 Darwin Kernel Version 17.5.0: Fri Apr 13 19:32:32 PDT 2018; root:xnu-4570.51.2~1/RELEASE_X86_64 x86_64 and I think I successfully built Indy, but when I try to use it in Python the import fails. And when I run cargo test it fails in ledger_demo_works with, "thread 'ledger_demo_works' panicked at 'called `Result::unwrap()` on an `Err` value: Timeout', libcore/result.rs:945:5." I'm not familiar with Rust or with the Indy codebase. Anything obviously wrong I should note?

JulesMiller (Mon, 21 May 2018 01:42:56 GMT):
Has joined the channel.

KotsAshish (Mon, 21 May 2018 10:37:43 GMT):
Has joined the channel.

CabMorris2 (Mon, 21 May 2018 19:28:01 GMT):
Has joined the channel.

jayapalreddy (Tue, 22 May 2018 12:00:34 GMT):
Has joined the channel.

Sarah.Conway (Tue, 22 May 2018 12:20:27 GMT):
hi all. I am new to working on marketing/PR for Hyperledger. We are writing a follow up Consensus blog for the HL site that focuses on interoperability. Can someone share a few sentences, maybe 2-3, on what concrete progress we have made or are we planning to make with #indy on this front? Info. related to our own projects and other public blockchain platforms. Feel free to point me to some urls, PPTs, etc. Thanks!

nud3l (Tue, 22 May 2018 14:17:05 GMT):
Has joined the channel.

AxelNennker (Tue, 22 May 2018 15:36:06 GMT):
What to do about TypeError: validation error [LedgerStatus]: expected types 'str', got 'bytes' (merkleRoot=b'9dq5w8xqZAp41vnVKPaqoDcpSp7KpCcRYrBdtYE37Rtw') ?

AxelNennker (Tue, 22 May 2018 15:37:11 GMT):
hi indy-anoncreds 1.0.11 amd64 Anonymous credentials hi indy-node 1.3.55 amd64 Indy node hi indy-plenum 1.2.34 amd64 Plenum Byzantine Fault Tolerant Protocol ii libindy-crypto 0.4.0 amd64 This is the shared crypto libirary for Hyperledger Indy components. hi python3-indy-crypto 0.2.0 amd64 This is the official wrapper for Hyperledger Indy Crypto library (https://www.hyperledger.org/projects).

AxelNennker (Tue, 22 May 2018 15:37:11 GMT):
hi indy-anoncreds 1.0.11 amd64 Anonymous credentials hi indy-node 1.3.55 amd64 Indy node hi indy-plenum 1.2.34 amd64 Plenum Byzantine Fault Tolerant Protocol ii libindy-crypto 0.4.0 amd64 This is the shared crypto libirary for Hyperledger Indy components. hi python3-indy-crypto 0.2.0 amd64 This is the official wrapper for Hyperledger Indy Crypto library (https://www.hyperledger.org/projects). hi indy-anoncreds 1.0.11 amd64 Anonymous credentials hi indy-node 1.3.55 amd64 Indy node hi indy-plenum 1.2.34 amd64 Plenum Byzantine Fault Tolerant Protocol ii libindy-crypto 0.4.0 amd64 This is the shared crypto libirary for Hyperledger Indy components. hi python3-indy-crypto 0.2.0 amd64 This is the official wrapper for Hyperledger Indy Crypto library (https://www.hyperledger.org/projects).

AxelNennker (Tue, 22 May 2018 15:38:00 GMT):
hi sovrin 1.1.8 amd64 Sovrin node

AlBeruni (Wed, 23 May 2018 01:24:34 GMT):
Has joined the channel.

gudkov (Wed, 23 May 2018 07:36:30 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=7xTy69SyQds86tzDS) @finkga It just mean that test can't connect to nodes pool

gudkov (Wed, 23 May 2018 07:36:30 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=7xTy69SyQds86tzDS) @finkga It just means that test can't connect to nodes pool

username343 (Wed, 23 May 2018 11:51:06 GMT):
Has joined the channel.

jack21 (Wed, 23 May 2018 12:16:29 GMT):
Has joined the channel.

esplinr (Wed, 23 May 2018 14:34:07 GMT):
@Sarah.Conway We created an #indy-outreach channel for these sorts of conversations. Are you referring to interoperability with other Hyperledger projects, or with other ecosystems?

kbreite27 (Wed, 23 May 2018 14:37:56 GMT):
Has joined the channel.

Sarah.Conway (Wed, 23 May 2018 20:18:24 GMT):
@esplinr would be great to know about interoperability with other HL projects and other ecosystems. Thanks for the pointer on the #indy-outreach channel

esplinr (Wed, 23 May 2018 21:34:43 GMT):
@Sarah.Conway The Indy crypto library is being adopted by other HL projects (at least Sawtooth). We have also been evaluating using components from other projects for our needs.

esplinr (Wed, 23 May 2018 21:35:17 GMT):
I've seen people ask about deploying Indy and Fabric together, but no one has explained to me their use case so I don't understand their goals. I'm not aware of any roadblocks.

esplinr (Wed, 23 May 2018 21:36:04 GMT):
Indy has implemented a generic payment gateway, that is expected to be used with something like bitcoin to purchase credentials. It's brand new, and so I'm not aware of anyone actually trying it yet.

esplinr (Wed, 23 May 2018 21:36:24 GMT):
If there are other technologies you are interested in, you should ask about them specifically. It's hard to know what information you would find helpful.

esplinr (Wed, 23 May 2018 21:36:56 GMT):
Indy is a fairly new project, and so there aren't yet a lot of real world use cases showing how the different technologies come together.

rjones (Wed, 23 May 2018 22:27:58 GMT):
Has left the channel.

annap (Thu, 24 May 2018 03:31:50 GMT):
Has joined the channel.

bendon (Thu, 24 May 2018 08:41:12 GMT):
Hi there. Looking at the 'Indy Contributor Guidelines' I read (slide 4, point 4): "Write good code, including unit tests (TDD!). Observe best practices and style guidelines as documented in the relevant CODING-CONVENTIONS.md file." But I've not managed to find that md file in any repo so far... Any clue? I'm looking for this because I wonder if you guys would accept PR containing symbolic links (that some Windows developer may have hard time with)?

Sean_Bohan (Thu, 24 May 2018 14:08:13 GMT):
in one hour: the Hyperledger Indy WG Call Where: Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/232861185 Or iPhone one-tap (US Toll): +16465588656,232861185# or +14086380968,232861185# Or Telephone: Dial: +1 646 558 8656 (US Toll) or +1 408 638 0968 (US Toll) Meeting ID: 232 861 185 International numbers available: https://zoom.us/zoomconference?m=a0jD_rTMnh0ZYGQDOKPCNrK_0dP7WPfp1 Indy Rocketchat: https://chat.hyperledger.org/channel/indyhttps://chat.hyperledger.org/channel/indy-agenthttps://chat.hyperledger.org/channel/indy-node Indy Wiki: https://wiki.hyperledger.org/projects/indy Indy WG Agendas: https://drive.google.com/open?id=1wNnp1pPS6-1Y4B2oygfI6tOU4eHGmpRH Indy Videos: https://drive.google.com/open?id=1Z8-jR7hnXb57fufE0OXxIfeUn05zNmRt Indy WG Meeting Recordings: https://drive.google.com/open?id=0B_NJV6eJXAA1UlJZMXd3cm1zNDQ

Sean_Bohan (Thu, 24 May 2018 14:53:47 GMT):
Links from todays call: Hyperledger Hackfest: Amsterdam (June 27-29) https://www.hyperledger.org/event/hyperledger-hackfest-june-2018 Hyperledger Blockchain Showcase https://docs.google.com/forms/d/e/1FAIpQLScvr8gpCDOj5cYiptrCtzNpK5aXowbWWiBlrXnHUXNQX01-aw/viewform Webinars: May 31: Blockchain & the Enterprise: Blockchain and the enterprise. But what about Security https://event.on24.com/wcc/r/1668381/6D322A06EEB5311622FAFC1D79C09655?partnerref=hlmcomm June 12: Decentralized Identity Distilled https://event.on24.com/wcc/r/1678323/8CF92FCAAF2A657C0E912D756E4EE009?partnerref=hlmcomm WG Call Topics (add yours): https://docs.google.com/document/d/15CxUMxaUB3yZ9PSNEX87FUUf1YqV0zEGzQJFb0TX4Tc/edit

Sean_Bohan (Thu, 24 May 2018 14:53:47 GMT):
Links from todays call: Hyperledger Hackfest: Amsterdam (June 27-29) https://www.hyperledger.org/event/hyperledger-hackfest-june-2018 Hyperledger Blockchain Showcase https://docs.google.com/forms/d/e/1FAIpQLScvr8gpCDOj5cYiptrCtzNpK5aXowbWWiBlrXnHUXNQX01-aw/viewform Webinars: May 31: Blockchain & the Enterprise: Blockchain and the enterprise. But what about Security June 12: Decentralized Identity Distilled WG Call Topics (add yours): https://docs.google.com/document/d/15CxUMxaUB3yZ9PSNEX87FUUf1YqV0zEGzQJFb0TX4Tc/edit

Sean_Bohan (Thu, 24 May 2018 14:58:22 GMT):
Sprint Demo for 18.10: https://drive.google.com/drive/folders/1fftst8tmXcI8BvGwoDGKmZoyBsY56sJA

Sean_Bohan (Thu, 24 May 2018 15:13:17 GMT):
- CI/CD docs in indy-node repo: https://github.com/hyperledger/indy-node/blob/master/docs/ci-cd.md

 - CI pipeline: https://github.com/hyperledger/indy-node/blob/master/Jenkinsfile.ci
(dockerfiles for CI pipeline https://github.com/hyperledger/indy-node/tree/master/ci) 

- CD pipeline: https://github.com/hyperledger/indy-node/blob/master/Jenkinsfile.cd
(deb packaging build scripts: https://github.com/hyperledger/indy-node/tree/master/build-scripts/ubuntu-1604)

 - Hyperledger Jenkins indy jobs: https://jenkins.hyperledger.org/view/indy/
 - debian repo: https://repo.sovrin.org

KellyCooper (Thu, 24 May 2018 15:18:08 GMT):
Has joined the channel.

KellyCooper (Thu, 24 May 2018 15:29:45 GMT):
#training-and-education-wg includes volunteers available to edit, test and contribute to tutorials and learning materials. Feel free to post possible collaboration in the wg or @KellyCooper. We have many volunteers for beginning projects and some volunteers for advanced. One way we can support your efforts is as users/consumers to comment on clarity, strengths and gaps in your documentation/materials. Your request(s) will be viewed by the group every Monday and we can track/manage on our end. Thanks, Kelly

Sean_Bohan (Thu, 24 May 2018 15:31:20 GMT):
http://www.identitynorth.ca/events/identitynorth-2018/

stanleyc-trustscience (Thu, 24 May 2018 18:32:58 GMT):
A general question. How can one use DID to facilitate the user login action to a website?

mboyd (Thu, 24 May 2018 18:41:07 GMT):
@bendon You are correct, we don't have any repository specific coding conventions readme. Generally speaking, we follow the conventions defined by the languages that we use (mostly python and rust). If you have any insight on symbolic links for win developers, we would certainly welcome a pull request! Follow the steps in our [contributor resources page](https://github.com/michaeldboyd/indy-sdk/blob/e834e86d8b43fc28c668b6396ae3af3182d68285/doc/contributor-resources.md) to get up and running quickly

esplinr (Thu, 24 May 2018 19:27:51 GMT):
@bendon Thanks for pointing out the instructions that we need to update. @mboyd is that a task that you can handle?

esplinr (Thu, 24 May 2018 19:28:31 GMT):
@mboyd You should add the guidance that you just gave to the document @bendon mentioned points to the missing repo md files.

esplinr (Thu, 24 May 2018 19:28:39 GMT):
(If you are available to do that)

lxy (Fri, 25 May 2018 02:29:02 GMT):
Has joined the channel.

sndiaye2022 (Fri, 25 May 2018 04:20:55 GMT):
Has joined the channel.

jeremi 24 (Fri, 25 May 2018 07:04:34 GMT):
Has joined the channel.

huangliping (Fri, 25 May 2018 10:00:28 GMT):
Has joined the channel.

StanleyChen (Fri, 25 May 2018 16:51:39 GMT):
Has joined the channel.

StanleyChen (Fri, 25 May 2018 16:52:08 GMT):
Hi guys, I've been thinking about how to enable the website "sign-in" action using indy and DID

StanleyChen (Fri, 25 May 2018 17:00:51 GMT):
Here's my thought, and I hope I can get some feedback: Essentially, there is no sign-in. The pairwise DIDs that the website and the user used to communicate with each other. THAT alone is sufficient to authenticate a user. The pairwise DIDs are stored on a blockchain, and if no expiration is set on the DIDs, then they should last forever.. In order for the user to prove who he really is within the society (if this is something to be desired in the web app), then he then required to fill out the proofRequest from the website, using the claims stored in his wallet. However, this only needs to be done once. Once the web site receives the user's credential proof and verified against the indy blockchain, there is no need to do this every single time - sort of like the identity verification in Equifax except we're replacing it with indy's verifiable claim approach. The pairwise identifiers will serve as something like certificates from CA that we currently use for mutal authentication.

StanleyChen (Fri, 25 May 2018 17:00:51 GMT):
Here's my thought, and I hope I can get some feedback: Essentially, there is no sign-in. The pairwise DIDs that the website and the user used to communicate with each other - THAT alone is sufficient to authenticate a user, same also goes for the user. The pairwise DIDs are stored on a blockchain, and if no expiration is set on the DIDs, then they should last forever.. In order for the user to prove who he really is within the society (if this is something to be desired in the web app), then he then required to fill out the proofRequest from the website, using the claims stored in his wallet. However, this only needs to be done once. Once the web site receives the user's credential proof and verified against the indy blockchain, there is no need to do this every single time - sort of like the identity verification in Equifax except we're replacing it with indy's verifiable claim approach. The pairwise identifiers will serve as something like certificates from CA that we currently use for mutal authentication.

StanleyChen (Fri, 25 May 2018 17:00:51 GMT):
Here's my thought, and I hope I can get some feedback: Essentially, there is no sign-in. The pairwise DIDs that the website and the user used to communicate with each other - THAT alone is sufficient to authenticate a user, same also goes for the user. The pairwise DIDs are stored on a blockchain, and if no expiration is set on the DIDs, then they should last forever.. In order for the user to prove who he really is within the society (if this is something desired in the web app), then he then required to fill out the proofRequest from the website, using the claims stored in his wallet. However, this only needs to be done once. Once the web site receives the user's credential proof and verified against the indy blockchain, there is no need to do this every single time - sort of like the identity verification in Equifax except we're replacing it with indy's verifiable claim approach. The pairwise identifiers will serve as something like certificates from CA that we currently use for mutal authentication.

StanleyChen (Fri, 25 May 2018 17:17:09 GMT):
So the user account creation flow may look something like this: - User creates a user account within the website using his DID - After user account is created, the account is placed in a verification stage. The website sends users a proofRequest for the credentials that it cares about (eg, ssn number credential issued by US government) - Assuming the user has the required credentials, he/she then produce a credential proof using the claims stored in his wallet. - The website receive the credential proof and verified it against indy's blockchain. The result is positive. - The website then records that it has verified the user is valid by associating a "valid" status with the DID After some time, the same user tries to log into the website again - User makes a "sign-in" request with the pairwise DID. The request contains the user's DID and request body = "sign-in". There's really no sign-in here as there is no input of username/password. However, the website will check that the user account status by user's DID. And if the user is found and valid, then it allows user access to the web site, Seeing that the request body="sign-in", the web site directs the user to the home page or dashboard ..

StanleyChen (Fri, 25 May 2018 17:17:09 GMT):
So the user account creation flow may look something like this: - User creates a user account within the website using his DID - After user account is created, the account is placed in a verification stage. The website sends users a proofRequest for the credentials that it cares about (eg, ssn number credential issued by US government) - Assuming the user has the required credentials, he/she then produce a credential proof using the claims stored in his wallet. - The website receive the credential proof and verified it against indy's blockchain. The result is positive. - The website then records that it has verified the user is valid by associating a "valid" status with the DID or perhaps save the credential proof along with it After some time, the same user tries to log into the website again - User makes a "sign-in" request with the pairwise DID. The request contains the user's DID and request body = "sign-in". There's really no sign-in here as there is no input of username/password. However, the website will check that the user account status by user's DID. And if the user is found and valid, then it allows user access to the web site, Seeing that the request body="sign-in", the web site directs the user to the home page or dashboard ..

StanleyChen (Fri, 25 May 2018 17:31:45 GMT):
Or if someone can point me to some material relating to proper way of doing DID Auth for a website, that'd be great too

StanleyChen (Fri, 25 May 2018 17:31:45 GMT):
Or if someone can point me to some material relating to proper way of doing DID Auth using indy for a website, that'd be great too

swcurran (Fri, 25 May 2018 19:54:08 GMT):
@StanleyChen - you pretty much have the process. BC Gov sponsored some work that have now been moved to DIF that implement a couple of these - for example: https://github.com/decentralized-identity/did-auth-extension which is the closest to the website use case you are talking about. There was some preliminary discussion at the recent IIW conference to merge DID-Auth with FIDO2. One thing about your use case to keep in mind. A DID (or more specifically, a DID Doc) can be updated for key rotation (e.g. replacing a compromised keypair). That should be a signal to the other side to recheck the verifiable credentials, as that mechanism can be used to transfer ownership of the DID to another identity (which is OK - just be aware of it). As well, VerCreds may have a lifespan, and a service may (depending on its verification needs) want to re-verify the VerCreds from time to time.

swcurran (Fri, 25 May 2018 19:55:36 GMT):
One other thing. Much of the reason for collecting information from the user is to send notifications via email, so an email address is collected. However, since a pairwise DID relationship enables a communication path, there is no need to collect an email address for notifications. DIDs are fine. Cool!!

ColorfulZebra (Fri, 25 May 2018 21:56:17 GMT):
Has joined the channel.

vudathasaiomkar (Sun, 27 May 2018 06:48:31 GMT):
Has joined the channel.

udithj95 (Sun, 27 May 2018 08:25:51 GMT):
Has joined the channel.

Neumann347 (Sun, 27 May 2018 20:16:33 GMT):
@swcurran @StanleyChen Sorry, maybe I don't fully understand. The user wouldn't necessarily re-use a DID, they should create a new DID for this relationship with the website. More importantly, the order of the user account creation flow isn't how I would understand it

Neumann347 (Sun, 27 May 2018 20:16:33 GMT):
@swcurran @StanleyChen Sorry, maybe I don't fully understand. The order of the user account creation flow isn't how I would understand it ``` - Both the User and the website create new DIDs and uses them to create the pairwise connection to each other ```

Neumann347 (Sun, 27 May 2018 20:33:45 GMT):
@swcurran @StanleyChen I am a little confused about how a credential proof is verified against the blockchain. The credential isn't stored on the blockchain, only the credential schema (the fields of the credential) and the credential definition (the link between the credential schema and a specific issuer) are stored on the blockchain, right? Is the credential stored by both the issuer and the holder? Or is the credential re-verified each time is included in a proof request?

Neumann347 (Sun, 27 May 2018 20:33:45 GMT):
@swcurran @StanleyChen I am a little confused about how a credential proof is verified against the blockchain. The credential isn't stored on the blockchain, only the credential schema (the fields of the credential) and the credential definition (the link between the credential schema and a specific issuer) are stored on the blockchain, right? Is the credential stored by both the issuer and the holder? Or is the credential re-verified each time it is included in a proof request?

Neumann347 (Sun, 27 May 2018 20:33:45 GMT):
@swcurran @StanleyChen I am a little confused about how a credential proof is verified against the blockchain. The credential isn't stored on the blockchain, only the credential schema (the fields of the credential) and the credential definition (the link between the credential schema and a specific issuer) are stored on the blockchain, right? Is the credential stored by both the issuer and the holder? Or is the credential re-verified each time it is included in a proof request by the issuer who issued it?

swcurran (Sun, 27 May 2018 20:46:16 GMT):
The proof of a claim is not verified "against the blockchain". The blockchain provides some of the information necessary to complete the verification process as you indicate - and there is also a Revocation Registry. The credential itself (held by the Holder/Prover) contains sufficient information that via a Zero Knowledge Proof (ZKP): that it was issued by the issuer (crypto related to the CredDef and the DID of the issuer associated with the CredDef), that it was issued to the holder (crypto related to the "link secret" held by the Holder), that it hasn't been tampered with (signature) and that it hasn't been revoked - a ZKP process executed by the Holder that the Credential has not been revoked. Each of those mechanisms are, shall we say "tricky". I've managed to understand it enough that I believe it, without getting down into the Rust code and go line by line. :-). Reading on the pieces and looking at the data structures passed through to the Indy-SDK is helpful, if you want to dig deeper. Or you can trust that they have implemented those pieces.

swcurran (Sun, 27 May 2018 20:46:16 GMT):
The proof of a claim is not verified "against the blockchain". The blockchain provides some of the information necessary to complete the verification process as you indicate - and there is also a Revocation Registry on the ledger. The credential itself (held by the Holder/Prover) contains sufficient information that via a Zero Knowledge Proof (ZKP): that it was issued by the issuer (crypto related to the CredDef and the DID of the issuer associated with the CredDef), that it was issued to the holder (crypto related to the "link secret" held by the Holder), that it hasn't been tampered with (signature) and that it hasn't been revoked - a ZKP process executed by the Holder that the Credential has not been revoked without providing a piece of information linked directly to the Holder. Each of those mechanisms are, shall we say "tricky". I've managed to understand it enough that I believe it, without getting down into the Rust code and go line by line. :-). Reading on the pieces and looking at the data structures passed through to the Indy-SDK is helpful, if you want to dig deeper. Or you can trust that they have implemented those pieces.

Neumann347 (Sun, 27 May 2018 20:54:45 GMT):
That's what I thought. I looked at the Rust code...my coding skills require me to trust that those pieces are correct ;)

vd30992 (Tue, 29 May 2018 04:55:22 GMT):
Has joined the channel.

vd30992 (Tue, 29 May 2018 04:56:05 GMT):
Hello everyone, got issue while setting up Hyperledger Indy Agent. -------------------------------------------------------------------------------------------------------ERROR------------------------------------------------------------------------------------------------------- Agent startup failed: [cause : error occurred during operation: client request invalid: InvalidClientRequest('ULtgFQJe6bjiFbs7ke3NJD can have one and only one SCHEMA with name Transcript and version 1.2',)] -------------------------------------------------------------------------------------------------------ERROR------------------------------------------------------------------------------------------------------- Kindly provide me solution, Thank you.

vd30992 (Tue, 29 May 2018 04:56:05 GMT):
Hello everyone, got issue while setting up Hyperledger Indy Agent. ``` ``` ---------------------------------------------------------------------------------------------------*----ERROR------------------------------------------------------------------------------------------------------- Agent startup failed: [cause : error occurred during operation: client request invalid: InvalidClientRequest('ULtgFQJe6bjiFbs7ke3NJD can have one and only one SCHEMA with name Transcript and version 1.2',)] -------------------------------------------------------------------------------------------------------ERROR-------------------------------------------------------------------------------------------------------* Kindly provide me solution, Thank you.

vd30992 (Tue, 29 May 2018 04:56:05 GMT):
Hello everyone, got issue while setting up Hyperledger Indy Agent. ``` ``` Kindly provide me solution, Thank you.

vd30992 (Tue, 29 May 2018 04:56:05 GMT):
Hello everyone, got issue while setting up Hyperledger Indy Agent. ``` *-------------------------------------------------------------------------------------------------------ERROR------------------------------------------------------------------------------------------------------- Agent startup failed: [cause : error occurred during operation: client request invalid: InvalidClientRequest('ULtgFQJe6bjiFbs7ke3NJD can have one and only one SCHEMA with name Transcript and version 1.2',)] -------------------------------------------------------------------------------------------------------ERROR------------------------------------------------------------------------------------------------------- * ``` Kindly provide me solution, Thank you.

vd30992 (Tue, 29 May 2018 04:56:05 GMT):
Hello everyone, got issue while setting up Hyperledger Indy Agent. ``` -------------------------------------------------------------------------------------------------------ERROR------------------------------------------------------------------------------------------------------- Agent startup failed: [cause : error occurred during operation: client request invalid: InvalidClientRequest('ULtgFQJe6bjiFbs7ke3NJD can have one and only one SCHEMA with name Transcript and version 1.2',)] -------------------------------------------------------------------------------------------------------ERROR------------------------------------------------------------------------------------------------------- ``` Kindly provide me solution, Thank you.

stanleyc-trustscience (Tue, 29 May 2018 05:35:21 GMT):
Hi guys, I have a question about verifiable claim. Say, if I want to prove I have the credential `ssn: 1234-56-7890` issued by the US government. If this ssn 1234-56-7890 is stored in a claim, wouldn't be that storing personal data on the blockchain? And I understand that it is not recommended to store any personal info on the blockchain. But how'd be the user able to proof the ownership over the ssn, if it's not stored on the blockchain? I heard from Drummond about storing pointer, how exactly would that mechanism work?

stanleyc-trustscience (Tue, 29 May 2018 05:35:21 GMT):
Hi guys, I have a question about verifiable claim. Say, if I want to prove I have the credential `ssn: 1234-56-7890` issued by the US government. If this ssn 1234-56-7890 is stored in a claim, wouldn't be that storing personal data on the blockchain? And I understand that it is not recommended to store any personal info on the blockchain. But how'd be the user able to proof the ownership over the ssn?

alex873 (Tue, 29 May 2018 11:34:52 GMT):
Similar question here: If I store private data on the ledger, is that encrypted somehow, or is it readable by anyone, e.g. the stewards?

swcurran (Tue, 29 May 2018 15:44:31 GMT):
@stanleyc-trustscience - verifiable credentials are not stored on the Indy blockchain. Some of the crypto material necessary to prove the credential is pulled from the ledger, but the actual credential is sent directly from the issuer to the holder (usually the subject of the credential). The holder can then use the credential to prove claims to any number of verifiers, without having to contact the issuers.

stanleyc-trustscience (Tue, 29 May 2018 15:50:31 GMT):
@swcurran Thank you for answering all my question! ^^ You're right, I missed this part in the alice's example. I had thought that the follow code stores verifiable claim on the blockchain. Looks like it doesn''t do that. ``` transcript_cred_json, _, _ = \ await anoncreds.issuer_create_credential(faber_wallet, transcript_cred_offer_json, authdecrypted_transcript_cred_request_json, transcript_cred_values, None, None)``` The sending of credential is done with pairwise identifier and encrypted, so it's good ``` authcrypted_transcript_cred_json = await crypto.auth_crypt(faber_wallet, faber_alice_key, alice_faber_verkey, transcript_cred_json.encode('utf-8'))``` ` logger.info("\"Faber\" -> Send authcrypted \"Transcript\" Credential to Alice")`

stanleyc-trustscience (Tue, 29 May 2018 15:50:31 GMT):
@swcurran Thank you for answering all my question! ^^ You're right, I missed this part in the alice's example. I had thought that the follow code stores verifiable claim on the blockchain. Looks like it doesn''t do that. ` transcript_cred_json, _, _ = \ await anoncreds.issuer_create_credential(faber_wallet, transcript_cred_offer_json, authdecrypted_transcript_cred_request_json, transcript_cred_values, None, None)` The sending of credential is done with pairwise identifier and encrypted, so it's good ` authcrypted_transcript_cred_json = await crypto.auth_crypt(faber_wallet, faber_alice_key, alice_faber_verkey, transcript_cred_json.encode('utf-8'))` ` logger.info("\"Faber\" -> Send authcrypted \"Transcript\" Credential to Alice")`

stanleyc-trustscience (Tue, 29 May 2018 16:05:00 GMT):
@swcurran Is DID's key rotation done on a periodic basis or a on-demand basis? If it happens, how'd do I figure it out on a DID document?

swcurran (Tue, 29 May 2018 16:13:18 GMT):
DID rotation is done by the DID owner on an ad hoc basis - whenever they want. Figuring out that it was done is up to the user of the DID - do they always dereference the DID (e.g. get the DID Document) before every use or perhaps only on failure to see if they have the wrong key? Part of the Agent-to-Agent protocol (at the layer above the Indy-SDK) could be to notify the other side that a rotation has occurred.

siddareddy (Tue, 29 May 2018 18:44:42 GMT):
Has joined the channel.

burdettadam (Tue, 29 May 2018 19:13:54 GMT):
Has joined the channel.

drummondreed (Wed, 30 May 2018 05:22:33 GMT):
@swcurran Just wanted to say that I've been away from these channels for much of the last 3 weeks and you've given some really good answers to some really hard questions. Thank you!

vd30992 (Wed, 30 May 2018 06:43:21 GMT):
Hello everyone, wasted my much time on executing this (https://github.com/hyperledger/indy-node/blob/master/getting-started.md) example on deprecated cli version. Now working on new cli version (https://github.com/hyperledger/indy-sdk/tree/master/cli) but here Unable to run above example because commands are changed. Anyone please guide me that How to run ALICE example using new cli. Thanks for any help :-)

jwaup (Wed, 30 May 2018 13:57:55 GMT):
Has joined the channel.

adaml (Wed, 30 May 2018 16:39:29 GMT):
Has joined the channel.

im418 (Wed, 30 May 2018 19:01:39 GMT):
Has joined the channel.

json (Wed, 30 May 2018 19:23:25 GMT):
Hi everyone, I've been working thru the first indy Python dev tutorial (https://github.com/hyperledger/indy-sdk/blob/master/doc/how-tos/write-did-and-query-verkey/python/README.md). I've put in pull requests as I've found some errors. Currently stuck right now on this step: 2. Open ledger and get handle ←[0m _indy_loop_callback: Function returned error 307 Error occurred: ErrorCode.PoolLedgerTimeout

json (Wed, 30 May 2018 19:23:52 GMT):
Not sure why, I have the indy-pool running in docker as per the pre-reqs. Anyone seen this before?

im418 (Wed, 30 May 2018 19:52:07 GMT):
The IBM tutorial (indy-tutorial-sandbox) also seems to be failing. After building the docker and running 'make run-alice', all the nodes get built and the scripts run, but it never seems to fully connect to the sandbox -- it never says "Connected to sandbox." The scripts seem to add items to the default wallet, but fail when sending NYM, ATTRIB, etc. Instead of seeing positive results, I'm seeing things like "Request not yet verified." and "Connection not yet synchronized." When the scripts end, it sometimes returns to the "indy@sandbox" prompt, but sometimes fails out to the docker prompt "root@67870d10e0d0:/home/indy#" I'm using Docker v17.10.0-ce, btw. Any ideas?

PoojaVarshneya (Wed, 30 May 2018 19:55:57 GMT):
Has joined the channel.

azur3s0ng (Wed, 30 May 2018 22:44:59 GMT):
Has joined the channel.

StanleyChen (Thu, 31 May 2018 00:02:28 GMT):
Hi everyone, I'm wrapping my head around DID Auth and Sovrin https://github.com/WebOfTrustInfo/rebooting-the-web-of-trust-spring2018/blob/master/topics-and-advance-readings/DID%20Auth:%20Scope%2C%20Formats%2C%20and%20Protocols.md ``` The relying party resolves the identity owner's DID to a DID Document. The relying party attempts to authenticate the identity owner using the authentication object(s) found in the DID Document. The authentication object(s) may include or reference a publicKey object, in cases where the identity owner's proof is established as a cryptographic signature. ``` Looks like for Sovrin - Authentication is just using the pairwise DID established during enrollment. Well... there isn't really a need for it. Just talk over the pairwise DID directly, anytime.

AvikHazra (Thu, 31 May 2018 06:41:31 GMT):
Has joined the channel.

AvikHazra (Thu, 31 May 2018 06:42:39 GMT):
hello, is there any demo node js project hyperledger indy ? hello, is there any demo node js project hyperledger indy ? hello, is there any demo node js project hyperledger indy ? is there any node js demo project in hyperledger indy ?

AvikHazra (Thu, 31 May 2018 06:42:49 GMT):
hello, is there any demo node js project hyperledger indy ?

mdapps (Thu, 31 May 2018 07:58:12 GMT):
Has joined the channel.

kdenhartog (Thu, 31 May 2018 08:09:49 GMT):
@StanleyChen yeah we've considered a few ways to perform Auth. For Sovrin to Sovrin agentsw they'll run the Agent to Agent protocol which will perform authentication for each message. This paper was an outline of what is currently being done by a few projects and i need to add the Authenticated encryption setup that we use with paurwise DIDs to this doc. It will eventually lead into an interoperable DID-Auth solution where we can't assume all parties can use A2A communication, so we'll have to use other methods which are suggested in that paper.

sergey-shilov (Thu, 31 May 2018 08:21:01 GMT):
Has joined the channel.

peter.danko (Thu, 31 May 2018 09:15:30 GMT):
Has joined the channel.

lkolisko (Thu, 31 May 2018 12:41:58 GMT):
Has joined the channel.

swcurran (Thu, 31 May 2018 14:45:20 GMT):
@StanleyChen - true. However, you want other use cases supported - e.g. login to a website with a browser using did-auth, or authenticating a request to an API. Those get more interesting - more intertwined with other technologies over which we have little to no control.

uniwebb (Thu, 31 May 2018 15:35:30 GMT):
Has joined the channel.

vd30992 (Fri, 01 Jun 2018 07:17:58 GMT):
Hello everyone, anyone know how to create "*pool_transactions_genesis*" file from $ - pool create AS-0301-owner gen_txn_file=./pool_transactions_genesis this command? I tried to find location *"/var/lib/indy/sandbox/pool_transactions_genesis*" but only blank indy folder found. Thank you very much for any suggestions.

vd30992 (Fri, 01 Jun 2018 07:17:58 GMT):
Hello everyone, anyone know how to create *pool_transactions_genesis* file from $ - pool create AS-0301-owner gen_txn_file=./pool_transactions_genesis this command? I tried to find location *"/var/lib/indy/sandbox/pool_transactions_genesis*" but only blank indy folder found. Thank you very much for any suggestions.

vd30992 (Fri, 01 Jun 2018 07:17:58 GMT):
Hello everyone, anyone know how to create *pool_transactions_genesis* file from $ *pool create sandbox gen_txn_file=/var/lib/indy/sandbox/pool_transactions_genesis* this command? I tried to find location *"/var/lib/indy/sandbox/pool_transactions_genesis*" but only blank indy folder found. Thank you very much for any suggestions.

DeltaForce (Fri, 01 Jun 2018 09:01:55 GMT):
Has joined the channel.

DeltaForce (Fri, 01 Jun 2018 09:12:05 GMT):
Hi all, I tried starting the pool using docker. I executed the *pool_start.sh* ```Starting node node1 10.0.0.2 Unable to find image 'node1:latest' locally docker: Error response from daemon: pull access denied for node1, repository does not exist or may require 'docker login'. See 'docker run --help'. Error: No such container: node1 Node node1 started on 10.0.0.2 Starting node node2 10.0.0.3 Unable to find image 'node2:latest' locally docker: Error response from daemon: pull access denied for node2, repository does not exist or may require 'docker login'. See 'docker run --help'. Error: No such container: node2``` I am getting this error. Can some one help me ?

pathfinder2104 (Fri, 01 Jun 2018 09:55:55 GMT):
Has joined the channel.

ashcherbakov (Fri, 01 Jun 2018 14:01:53 GMT):
@vd30992 IF you would like to have a test network, you can try https://github.com/hyperledger/indy-node/blob/master/scripts/generate_indy_pool_transactions If you use Docker, then the nodes there already has generated genesis txns

StanleyChen (Fri, 01 Jun 2018 16:56:22 GMT):
HI guys, does anyone know why, for the Alice getting started example, government has to first publish a schema and then faber fetches the schema and publish a credential definition?

StanleyChen (Fri, 01 Jun 2018 16:56:50 GMT):
Or perhaps what's schema and what's credential definition

swcurran (Fri, 01 Jun 2018 16:59:58 GMT):
@StanleyChen - that sequence is to get across that the schema of the credential that Faber is publishing (which is really just a list of claim/field names) is a common one that every University might use. In theory, everyone that issues credentials can create their own schema, but that makes integration hard. In theory, we could limit the system to only "standard" schema, but that means we have to get everyone to agree on those standards. Indy is trying to walk the line between - allow schema to be published (put on the ledger) but encourage reuse so that defacto standards develop.

swcurran (Fri, 01 Jun 2018 16:59:58 GMT):
@StanleyChen - that sequence is to get across that the schema of the credential that Faber is publishing (which is really just a list of claim/field names) is a common one that every University might use. In theory, everyone that issues credentials can create their own schema, but that makes integration hard. In theory, we could limit the system to only "standard" schema, but that means we have to get everyone to agree on those standards and that's really hard. Indy is trying to walk the line between - allow schema to be published (put on the ledger) but encourage reuse so that defacto standards develop.

StanleyChen (Fri, 01 Jun 2018 17:00:32 GMT):
@swcurran So faber can add more fields to the schema published by the government?

StanleyChen (Fri, 01 Jun 2018 17:00:32 GMT):
@swcurran So faber can add more fields to the schema published by the government? and then publish them in its cred def

StanleyChen (Fri, 01 Jun 2018 17:00:45 GMT):
The schema published by government is more like a "base template",

swcurran (Fri, 01 Jun 2018 17:00:55 GMT):
No - if they wanted to do that, they would have to publish their own.

swcurran (Fri, 01 Jun 2018 17:01:26 GMT):
A Credential Definition points to a Schema. They cannot be layered or anything like that.

StanleyChen (Fri, 01 Jun 2018 17:04:42 GMT):
I see, I think I get it now... this is to encourage standard schemas to be reused by different insinuation. When an insituation reused a schema, they "announce" it by publishing a credential def on the blockchain

StanleyChen (Fri, 01 Jun 2018 17:04:42 GMT):
I see, I think I get it now... this is to encourage standard schemas to be reused by different institution. When an institution reused a schema, they "announce" it by publishing a credential def on the blockchain

StanleyChen (Fri, 01 Jun 2018 17:04:42 GMT):
I see, I think I get it now... this is to encourage standard schemas to be reused by different institution. When an institution reuses a schema, they "announce" it by publishing a credential def on the blockchain

StanleyChen (Fri, 01 Jun 2018 17:04:42 GMT):
I see, I think I get it now... this is to encourage standard schemas to be reused by different institution. When an institution reuses a schema, they "announce" or "declare" it by publishing a credential def on the blockchain

StanleyChen (Fri, 01 Jun 2018 17:04:42 GMT):
I see, I think I get it now... this is to encourage standard schemas to be reused by different organization. When an organization reuses a schema, they "announce" or "declare" it by publishing a credential def on the blockchain

swcurran (Fri, 01 Jun 2018 17:07:31 GMT):
That's the effect - a CredDef is used as the basis for issuing Credentials, and it contains: * a public key for the CredDef used in issuing * a DID for the issuer * a link to the schema that the Creds will use * an optional link to a Revocation Registry so that the Creds issued using this CredDef can be revoked.

StanleyChen (Fri, 01 Jun 2018 17:08:56 GMT):
@swcurran Got it. This is awesome. Thank you again. You've been so helpful! :)

StanleyChen (Fri, 01 Jun 2018 17:08:56 GMT):
@swcurran Hmm, I'm only seeing four lines, and they seem to be describing VALIDATOR ``` {"data":{"alias":"Node1","blskey":"4N8aUNHSgjQVgkpm8nhNEfDf6txHznoYREg9kirmJrkivgL4oSEimFF6nsQ6M41QvhM2Z33nves5vfSn9n1UwNFJBYtWVnHYMATn76vLuL3zU88KyeAYcHfsih3He6UHcXDxcaecHVz6jhCYz1P2UZn2bDVruL5wXpehgBfBaLKm3Ba","client_ip":"127.0.0.1","client_port":9702,"node_ip":"127.0.0.1","node_port":9701,"services":["VALIDATOR"]},"dest":"Gw6pDLhcBcoQesN72qfotTgFa7cbuqZpkX3Xo6pLhPhv","identifier":"Th7MpTaRZVRYnPiabds81Y","txnId":"fea82e10e894419fe2bea7d96296a6d46f50f93f9eeda954ec461b2ed2950b62","type":"0"} {"data":{"alias":"Node2","blskey":"37rAPpXVoxzKhz7d9gkUe52XuXryuLXoM6P6LbWDB7LSbG62Lsb33sfG7zqS8TK1MXwuCHj1FKNzVpsnafmqLG1vXN88rt38mNFs9TENzm4QHdBzsvCuoBnPH7rpYYDo9DZNJePaDvRvqJKByCabubJz3XXKbEeshzpz4Ma5QYpJqjk","client_ip":"127.0.0.1","client_port":9704,"node_ip":"127.0.0.1","node_port":9703,"services":["VALIDATOR"]},"dest":"8ECVSk179mjsjKRLWiQtssMLgp6EPhWXtaYyStWPSGAb","identifier":"EbP4aYNeTHL6q385GuVpRV","txnId":"1ac8aece2a18ced660fef8694b61aac3af08ba875ce3026a160acbc3a3af35fc","type":"0"} {"data":{"alias":"Node3","blskey":"3WFpdbg7C5cnLYZwFZevJqhubkFALBfCBBok15GdrKMUhUjGsk3jV6QKj6MZgEubF7oqCafxNdkm7eswgA4sdKTRc82tLGzZBd6vNqU8dupzup6uYUf32KTHTPQbuUM8Yk4QFXjEf2Usu2TJcNkdgpyeUSX42u5LqdDDpNSWUK5deC5","client_ip":"127.0.0.1","client_port":9706,"node_ip":"127.0.0.1","node_port":9705,"services":["VALIDATOR"]},"dest":"DKVxG2fXXTU8yT5N7hGEbXB3dfdAnYv1JczDUHpmDxya","identifier":"4cU41vWW82ArfxJxHkzXPG","txnId":"7e9f355dffa78ed24668f0e0e369fd8c224076571c51e2ea8be5f26479edebe4","type":"0"} {"data":{"alias":"Node4","blskey":"2zN3bHM1m4rLz54MJHYSwvqzPchYp8jkHswveCLAEJVcX6Mm1wHQD1SkPYMzUDTZvWvhuE6VNAkK3KxVeEmsanSmvjVkReDeBEMxeDaayjcZjFGPydyey1qxBHmTvAnBKoPydvuTAqx5f7YNNRAdeLmUi99gERUU7TD8KfAa6MpQ9bw","client_ip":"127.0.0.1","client_port":9708,"node_ip":"127.0.0.1","node_port":9707,"services":["VALIDATOR"]},"dest":"4PS3EDQ3dW1tci1Bp6543CfuuebjFrg36kLAUcskGfaA","identifier":"TWwCRQRZ2ZHMJFn9TzLp7W","txnId":"aa5e817d7cc626170eca175822029339a444eb0ee8f0bd20d3b0b76e566fb008","type":"0"} ```

the_identity_guy (Sat, 02 Jun 2018 21:46:12 GMT):
Has anyone else experienced trouble connecting to the Pool nodes, when using Docker and MacOS (default port and ips), following the instructions here: https://github.com/hyperledger/indy-sdk/blob/master/README.md#how-to-start-local-nodes-pool-with-docker I get a timeout error: indy.error.IndyError: ErrorCode.PoolLedgerTimeout , error code 307

the_identity_guy (Sat, 02 Jun 2018 21:46:12 GMT):
Has anyone else experienced trouble connecting to the Pool nodes, when using Docker on MacOS (using default port and ips), following the instructions here: https://github.com/hyperledger/indy-sdk/blob/master/README.md#how-to-start-local-nodes-pool-with-docker I get a timeout error: indy.error.IndyError: ErrorCode.PoolLedgerTimeout , error code 307

the_identity_guy (Sat, 02 Jun 2018 21:46:57 GMT):
my IP on utils.py is: pool_ip = environ.get("TEST_POOL_IP", "127.0.0.1") and the IP on docker file is also same

sativ (Tue, 05 Jun 2018 14:05:12 GMT):
Has joined the channel.

jayapalreddy (Wed, 06 Jun 2018 01:53:28 GMT):
Can Indy be deployed as private network?

K.Amine (Wed, 06 Jun 2018 02:55:40 GMT):
Has joined the channel.

manu (Wed, 06 Jun 2018 11:30:26 GMT):
Has joined the channel.

BreizhIndy (Wed, 06 Jun 2018 14:39:54 GMT):
Has joined the channel.

BreizhIndy (Wed, 06 Jun 2018 14:40:32 GMT):
I guess it can but it's not its purpose

StanleyChen (Wed, 06 Jun 2018 19:18:05 GMT):
Hi, I have a question about this from getting started guide `The test ledger we use was pre-configured to store some known Steward NYMs.` Where's this configuration?

StanleyChen (Wed, 06 Jun 2018 19:18:05 GMT):
Hi, I have a question about this from getting started guide https://github.com/hyperledger/indy-sdk/blob/master/doc/getting-started/getting-started.md `The test ledger we use was pre-configured to store some known Steward NYMs.` Where's this configuration?

swcurran (Wed, 06 Jun 2018 19:40:22 GMT):
Pretty sure that the Steward NYMs and a a Trust Anchor are in the genesis file. There are likely 5 entries in the file - 4 for the Stewards/nodes and one for a Trust Anchor.

swcurran (Wed, 06 Jun 2018 19:40:22 GMT):
Pretty sure that the Steward NYMs and a Trust Anchor are in the genesis file. There are likely 5 entries in the file - 4 for the Stewards/nodes and one for a Trust Anchor.

kdenhartog (Wed, 06 Jun 2018 23:56:24 GMT):
I'm working on build a FAQs guide right now to address common questions people encounter when wanting to learn more about Indy, SSI, and how they can develop using Indy, as well as what they need to know to start contributing back to Indy. If people have suggestions please add them in here.

AvikHazra (Thu, 07 Jun 2018 06:53:46 GMT):
hello, when I`m trying to run indy-plenum from the document https://github.com/hyperledger/indy-plenum It through the error Traceback (most recent call last): File "/Library/Frameworks/Python.framework/Versions/3.6/lib/python3.6/runpy.py", line 183, in _run_module_as_main mod_name, mod_spec, code = _get_module_details(mod_name, _Error) File "/Library/Frameworks/Python.framework/Versions/3.6/lib/python3.6/runpy.py", line 109, in _get_module_details _import_(pkg_name) File "/Applications/MAMP/htdocs/hyperledger_indy/indy-plenum/plenum/__init__.py", line 57, in setup_plugins() File "/Applications/MAMP/htdocs/hyperledger_indy/indy-plenum/plenum/__init__.py", line 51, in setup_plugins import plenum.common.messages.node_messages File "/Applications/MAMP/htdocs/hyperledger_indy/indy-plenum/plenum/common/messages/node_messages.py", line 10, in from plenum.common.messages.client_request import ClientMessageValidator File "/Applications/MAMP/htdocs/hyperledger_indy/indy-plenum/plenum/common/messages/client_request.py", line 5, in from plenum.common.messages.fields import NetworkIpAddressField, \ File "/Applications/MAMP/htdocs/hyperledger_indy/indy-plenum/plenum/common/messages/fields.py", line 393, in class RequestIdentifierField(FieldBase): File "/Applications/MAMP/htdocs/hyperledger_indy/indy-plenum/plenum/common/messages/fields.py", line 396, in RequestIdentifierField _idr_field = IdentifierField() File "/Applications/MAMP/htdocs/hyperledger_indy/indy-plenum/plenum/common/messages/fields.py", line 370, in _init_ super().__init__(byte_lengths=(16, 32), *args, *kwargs) File "/Applications/MAMP/htdocs/hyperledger_indy/indy-plenum/plenum/common/messages/fields.py", line 345, in _init_ self._alphabet = set(base58.alphabet.decode("utf-8")) AttributeError: 'str' object has no attribute 'decode'

AvikHazra (Thu, 07 Jun 2018 06:53:46 GMT):
hello, when I`m trying to run indy-plenum from the document https://github.com/hyperledger/indy-plenum It show the error Traceback (most recent call last): File "/Library/Frameworks/Python.framework/Versions/3.6/lib/python3.6/runpy.py", line 183, in _run_module_as_main mod_name, mod_spec, code = _get_module_details(mod_name, _Error) File "/Library/Frameworks/Python.framework/Versions/3.6/lib/python3.6/runpy.py", line 109, in _get_module_details _import_(pkg_name) File "/Applications/MAMP/htdocs/hyperledger_indy/indy-plenum/plenum/__init__.py", line 57, in setup_plugins() File "/Applications/MAMP/htdocs/hyperledger_indy/indy-plenum/plenum/__init__.py", line 51, in setup_plugins import plenum.common.messages.node_messages File "/Applications/MAMP/htdocs/hyperledger_indy/indy-plenum/plenum/common/messages/node_messages.py", line 10, in from plenum.common.messages.client_request import ClientMessageValidator File "/Applications/MAMP/htdocs/hyperledger_indy/indy-plenum/plenum/common/messages/client_request.py", line 5, in from plenum.common.messages.fields import NetworkIpAddressField, \ File "/Applications/MAMP/htdocs/hyperledger_indy/indy-plenum/plenum/common/messages/fields.py", line 393, in class RequestIdentifierField(FieldBase): File "/Applications/MAMP/htdocs/hyperledger_indy/indy-plenum/plenum/common/messages/fields.py", line 396, in RequestIdentifierField _idr_field = IdentifierField() File "/Applications/MAMP/htdocs/hyperledger_indy/indy-plenum/plenum/common/messages/fields.py", line 370, in _init_ super().__init__(byte_lengths=(16, 32), *args, *kwargs) File "/Applications/MAMP/htdocs/hyperledger_indy/indy-plenum/plenum/common/messages/fields.py", line 345, in _init_ self._alphabet = set(base58.alphabet.decode("utf-8")) AttributeError: 'str' object has no attribute 'decode'

AvikHazra (Thu, 07 Jun 2018 06:53:46 GMT):
hello, when I`m trying to run indy-plenum from the document https://github.com/hyperledger/indy-plenum It show the error Traceback (most recent call last): File "/Library/Frameworks/Python.framework/Versions/3.6/lib/python3.6/runpy.py", line 183, in _run_module_as_main mod_name, mod_spec, code = _get_module_details(mod_name, _Error) File "/Library/Frameworks/Python.framework/Versions/3.6/lib/python3.6/runpy.py", line 109, in _get_module_details _import_(pkg_name) File "/Applications/MAMP/htdocs/hyperledger_indy/indy-plenum/plenum/__init__.py", line 57, in setup_plugins() File "/Applications/MAMP/htdocs/hyperledger_indy/indy-plenum/plenum/__init__.py", line 51, in setup_plugins import plenum.common.messages.node_messages File "/Applications/MAMP/htdocs/hyperledger_indy/indy-plenum/plenum/common/messages/node_messages.py", line 10, in from plenum.common.messages.client_request import ClientMessageValidator File "/Applications/MAMP/htdocs/hyperledger_indy/indy-plenum/plenum/common/messages/client_request.py", line 5, in from plenum.common.messages.fields import NetworkIpAddressField, \ File "/Applications/MAMP/htdocs/hyperledger_indy/indy-plenum/plenum/common/messages/fields.py", line 393, in class RequestIdentifierField(FieldBase): File "/Applications/MAMP/htdocs/hyperledger_indy/indy-plenum/plenum/common/messages/fields.py", line 396, in RequestIdentifierField _idr_field = IdentifierField() File "/Applications/MAMP/htdocs/hyperledger_indy/indy-plenum/plenum/common/messages/fields.py", line 370, in _init_ super().__init__(byte_lengths=(16, 32), *args, *kwargs) File "/Applications/MAMP/htdocs/hyperledger_indy/indy-plenum/plenum/common/messages/fields.py", line 345, in _init_ self._alphabet = set(base58.alphabet.decode("utf-8")) AttributeError: 'str' object has no attribute 'decode' How to solve this error? Please help me. I`m very new to this.

AvikHazra (Thu, 07 Jun 2018 06:53:46 GMT):
hello, when I`m trying to run indy-plenum from the document https://github.com/hyperledger/indy-plenum It show the error Traceback (most recent call last): File "/Library/Frameworks/Python.framework/Versions/3.6/lib/python3.6/runpy.py", line 183, in _run_module_as_main mod_name, mod_spec, code = _get_module_details(mod_name, _Error) File "/Library/Frameworks/Python.framework/Versions/3.6/lib/python3.6/runpy.py", line 109, in _get_module_details _import_(pkg_name) File "/Applications/MAMP/htdocs/hyperledger_indy/indy-plenum/plenum/__init__.py", line 57, in setup_plugins() File "/Applications/MAMP/htdocs/hyperledger_indy/indy-plenum/plenum/__init__.py", line 51, in setup_plugins import plenum.common.messages.node_messages File "/Applications/MAMP/htdocs/hyperledger_indy/indy-plenum/plenum/common/messages/node_messages.py", line 10, in from plenum.common.messages.client_request import ClientMessageValidator File "/Applications/MAMP/htdocs/hyperledger_indy/indy-plenum/plenum/common/messages/client_request.py", line 5, in from plenum.common.messages.fields import NetworkIpAddressField, \ File "/Applications/MAMP/htdocs/hyperledger_indy/indy-plenum/plenum/common/messages/fields.py", line 393, in class RequestIdentifierField(FieldBase): File "/Applications/MAMP/htdocs/hyperledger_indy/indy-plenum/plenum/common/messages/fields.py", line 396, in RequestIdentifierField _idr_field = IdentifierField() File "/Applications/MAMP/htdocs/hyperledger_indy/indy-plenum/plenum/common/messages/fields.py", line 370, in _init_ super().__init__(byte_lengths=(16, 32), *args, *kwargs) File "/Applications/MAMP/htdocs/hyperledger_indy/indy-plenum/plenum/common/messages/fields.py", line 345, in _init_ self._alphabet = set(base58.alphabet.decode("utf-8")) AttributeError: 'str' object has no attribute 'decode' How to solve this error? Please help me.

ashcherbakov (Thu, 07 Jun 2018 07:06:00 GMT):
@AvikHazra What version of base58 do you have? It needs to be 1.0.0

ashcherbakov (Thu, 07 Jun 2018 07:07:09 GMT):
BTW for what purpose are you running plenum? Are you doing this for development? Otherwise, if you just need to have a running pool, you can use ready to use Docker: https://github.com/hyperledger/indy-node/tree/master/environment/docker/pool

StanleyChen (Thu, 07 Jun 2018 07:21:48 GMT):
@swcurran Hm I'm only seeing 4 lines ``` {"data":{"alias":"Node1","blskey":"4N8aUNHSgjQVgkpm8nhNEfDf6txHznoYREg9kirmJrkivgL4oSEimFF6nsQ6M41QvhM2Z33nves5vfSn9n1UwNFJBYtWVnHYMATn76vLuL3zU88KyeAYcHfsih3He6UHcXDxcaecHVz6jhCYz1P2UZn2bDVruL5wXpehgBfBaLKm3Ba","client_ip":"127.0.0.1","client_port":9702,"node_ip":"127.0.0.1","node_port":9701,"services":["VALIDATOR"]},"dest":"Gw6pDLhcBcoQesN72qfotTgFa7cbuqZpkX3Xo6pLhPhv","identifier":"Th7MpTaRZVRYnPiabds81Y","txnId":"fea82e10e894419fe2bea7d96296a6d46f50f93f9eeda954ec461b2ed2950b62","type":"0"} {"data":{"alias":"Node2","blskey":"37rAPpXVoxzKhz7d9gkUe52XuXryuLXoM6P6LbWDB7LSbG62Lsb33sfG7zqS8TK1MXwuCHj1FKNzVpsnafmqLG1vXN88rt38mNFs9TENzm4QHdBzsvCuoBnPH7rpYYDo9DZNJePaDvRvqJKByCabubJz3XXKbEeshzpz4Ma5QYpJqjk","client_ip":"127.0.0.1","client_port":9704,"node_ip":"127.0.0.1","node_port":9703,"services":["VALIDATOR"]},"dest":"8ECVSk179mjsjKRLWiQtssMLgp6EPhWXtaYyStWPSGAb","identifier":"EbP4aYNeTHL6q385GuVpRV","txnId":"1ac8aece2a18ced660fef8694b61aac3af08ba875ce3026a160acbc3a3af35fc","type":"0"} {"data":{"alias":"Node3","blskey":"3WFpdbg7C5cnLYZwFZevJqhubkFALBfCBBok15GdrKMUhUjGsk3jV6QKj6MZgEubF7oqCafxNdkm7eswgA4sdKTRc82tLGzZBd6vNqU8dupzup6uYUf32KTHTPQbuUM8Yk4QFXjEf2Usu2TJcNkdgpyeUSX42u5LqdDDpNSWUK5deC5","client_ip":"127.0.0.1","client_port":9706,"node_ip":"127.0.0.1","node_port":9705,"services":["VALIDATOR"]},"dest":"DKVxG2fXXTU8yT5N7hGEbXB3dfdAnYv1JczDUHpmDxya","identifier":"4cU41vWW82ArfxJxHkzXPG","txnId":"7e9f355dffa78ed24668f0e0e369fd8c224076571c51e2ea8be5f26479edebe4","type":"0"} {"data":{"alias":"Node4","blskey":"2zN3bHM1m4rLz54MJHYSwvqzPchYp8jkHswveCLAEJVcX6Mm1wHQD1SkPYMzUDTZvWvhuE6VNAkK3KxVeEmsanSmvjVkReDeBEMxeDaayjcZjFGPydyey1qxBHmTvAnBKoPydvuTAqx5f7YNNRAdeLmUi99gERUU7TD8KfAa6MpQ9bw","client_ip":"127.0.0.1","client_port":9708,"node_ip":"127.0.0.1","node_port":9707,"services":["VALIDATOR"]},"dest":"4PS3EDQ3dW1tci1Bp6543CfuuebjFrg36kLAUcskGfaA","identifier":"TWwCRQRZ2ZHMJFn9TzLp7W","txnId":"aa5e817d7cc626170eca175822029339a444eb0ee8f0bd20d3b0b76e566fb008","type":"0"} ```

StanleyChen (Thu, 07 Jun 2018 07:21:48 GMT):
@swcurran Hm I'm only seeing 4 lines in genesis file ``` {"data":{"alias":"Node1","blskey":"4N8aUNHSgjQVgkpm8nhNEfDf6txHznoYREg9kirmJrkivgL4oSEimFF6nsQ6M41QvhM2Z33nves5vfSn9n1UwNFJBYtWVnHYMATn76vLuL3zU88KyeAYcHfsih3He6UHcXDxcaecHVz6jhCYz1P2UZn2bDVruL5wXpehgBfBaLKm3Ba","client_ip":"127.0.0.1","client_port":9702,"node_ip":"127.0.0.1","node_port":9701,"services":["VALIDATOR"]},"dest":"Gw6pDLhcBcoQesN72qfotTgFa7cbuqZpkX3Xo6pLhPhv","identifier":"Th7MpTaRZVRYnPiabds81Y","txnId":"fea82e10e894419fe2bea7d96296a6d46f50f93f9eeda954ec461b2ed2950b62","type":"0"} {"data":{"alias":"Node2","blskey":"37rAPpXVoxzKhz7d9gkUe52XuXryuLXoM6P6LbWDB7LSbG62Lsb33sfG7zqS8TK1MXwuCHj1FKNzVpsnafmqLG1vXN88rt38mNFs9TENzm4QHdBzsvCuoBnPH7rpYYDo9DZNJePaDvRvqJKByCabubJz3XXKbEeshzpz4Ma5QYpJqjk","client_ip":"127.0.0.1","client_port":9704,"node_ip":"127.0.0.1","node_port":9703,"services":["VALIDATOR"]},"dest":"8ECVSk179mjsjKRLWiQtssMLgp6EPhWXtaYyStWPSGAb","identifier":"EbP4aYNeTHL6q385GuVpRV","txnId":"1ac8aece2a18ced660fef8694b61aac3af08ba875ce3026a160acbc3a3af35fc","type":"0"} {"data":{"alias":"Node3","blskey":"3WFpdbg7C5cnLYZwFZevJqhubkFALBfCBBok15GdrKMUhUjGsk3jV6QKj6MZgEubF7oqCafxNdkm7eswgA4sdKTRc82tLGzZBd6vNqU8dupzup6uYUf32KTHTPQbuUM8Yk4QFXjEf2Usu2TJcNkdgpyeUSX42u5LqdDDpNSWUK5deC5","client_ip":"127.0.0.1","client_port":9706,"node_ip":"127.0.0.1","node_port":9705,"services":["VALIDATOR"]},"dest":"DKVxG2fXXTU8yT5N7hGEbXB3dfdAnYv1JczDUHpmDxya","identifier":"4cU41vWW82ArfxJxHkzXPG","txnId":"7e9f355dffa78ed24668f0e0e369fd8c224076571c51e2ea8be5f26479edebe4","type":"0"} {"data":{"alias":"Node4","blskey":"2zN3bHM1m4rLz54MJHYSwvqzPchYp8jkHswveCLAEJVcX6Mm1wHQD1SkPYMzUDTZvWvhuE6VNAkK3KxVeEmsanSmvjVkReDeBEMxeDaayjcZjFGPydyey1qxBHmTvAnBKoPydvuTAqx5f7YNNRAdeLmUi99gERUU7TD8KfAa6MpQ9bw","client_ip":"127.0.0.1","client_port":9708,"node_ip":"127.0.0.1","node_port":9707,"services":["VALIDATOR"]},"dest":"4PS3EDQ3dW1tci1Bp6543CfuuebjFrg36kLAUcskGfaA","identifier":"TWwCRQRZ2ZHMJFn9TzLp7W","txnId":"aa5e817d7cc626170eca175822029339a444eb0ee8f0bd20d3b0b76e566fb008","type":"0"} ```

StanleyChen (Thu, 07 Jun 2018 07:23:00 GMT):
Looks like you don't need to be steward or trust anchor to create NYM txn at the moment ..

AvikHazra (Thu, 07 Jun 2018 07:32:56 GMT):
@ashcherbakov , Thanks for your replay. yes I`m already using base58- 1.0.0. Yes I`m trying to use this for development. Actually, I`m trying to use this for identity purpose.

AvikHazra (Thu, 07 Jun 2018 07:32:56 GMT):
@ashcherbakov , Thanks for your reply. yes I`m already using base58- 1.0.0. Yes I`m trying to use this for development. Actually, I`m trying to use this for identity purpose.

AvikHazra (Thu, 07 Jun 2018 07:32:56 GMT):
@ashcherbakov , Thanks for your response. yes I`m already using base58- 1.0.0. Yes I`m trying to use this for development. Actually, I`m trying to use this for identity purpose.

ashcherbakov (Thu, 07 Jun 2018 07:38:32 GMT):
Did you follow https://github.com/hyperledger/indy-node/blob/master/docs/setup-dev.md?

ashcherbakov (Thu, 07 Jun 2018 07:40:00 GMT):
@StanleyChen You are looking at the pool ledger gensis txns file. But NYMs are in Domain ledger (domain genesis txn file).

AvikHazra (Thu, 07 Jun 2018 08:33:12 GMT):
@ashcherbakov is it possible without virtual environment ?

AvikHazra (Thu, 07 Jun 2018 08:33:12 GMT):
@ashcherbakov is it possible without virtual environment in Mac ?

AvikHazra (Thu, 07 Jun 2018 08:41:11 GMT):
thank you @ashcherbakov

skynet (Thu, 07 Jun 2018 10:44:36 GMT):
Has joined the channel.

hanumankumarn (Thu, 07 Jun 2018 12:31:19 GMT):
Has joined the channel.

hanumankumarn (Thu, 07 Jun 2018 12:40:12 GMT):

Selection_039.png

hanumankumarn (Thu, 07 Jun 2018 12:40:34 GMT):
any solution for the above issue?

brycecurtis (Thu, 07 Jun 2018 13:35:59 GMT):
@hanumankumarn Please refer to the dependencies section of the readme for instructions on installing s2i. https://github.com/IBM-Blockchain-Identity/indy-ssivc-tutorial#dependencies

hanumankumarn (Thu, 07 Jun 2018 13:56:46 GMT):
@brycecurtis i have followed the url what you have sent me but still i am facing the same issue

brycecurtis (Thu, 07 Jun 2018 13:59:39 GMT):
Here is the s2i website installation url: https://github.com/openshift/source-to-image#installation . What OS are you using?

hanumankumarn (Thu, 07 Jun 2018 14:03:32 GMT):
ubuntu os

hanumankumarn (Thu, 07 Jun 2018 14:05:03 GMT):
ubuntu debian

brycecurtis (Thu, 07 Jun 2018 14:05:39 GMT):
Verify that s2i is in your path and make sure you open a new terminal window if it's being set in your profile.

hanumankumarn (Thu, 07 Jun 2018 14:07:20 GMT):
how to check the path? please can you tell steps

swcurran (Thu, 07 Jun 2018 14:07:47 GMT):
@StanleyChen - thanks - sorry about that.

brycecurtis (Thu, 07 Jun 2018 14:25:29 GMT):
@hanumankumarn https://linuxconfig.org/linux-path-environment-variable

mjmckean (Thu, 07 Jun 2018 14:53:04 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=7qsWFQLAoCQLnTiW3) @the_identity_guy Did you end up resolving this issue? I've run into the same thing

grice_32 (Thu, 07 Jun 2018 14:54:19 GMT):
Has anyone tried to use Fabric with a plugin to the Indy network?

the_identity_guy (Thu, 07 Jun 2018 14:56:02 GMT):
@mjmckean I did I think, I fixed it by making sure my Indy-node and Indy-SDK are same version and the genesis transactions have the correct IP addresses. In some places 127.0.0.1 is used as IP and in other places 10.20.30.201 format. To make sure you have the same version for everything, pull the Indy-SDK using the 1.4.0 Tag and for Indy-Node Vagrant (or Docker) make sure you use the 'stable' branch for add-apt-repository command anywhere, such as within agent.sh , validator.sh and lib_indy.sh files.. <- these are script files that are executed when VagrantFile is executed to create the VMs

the_identity_guy (Thu, 07 Jun 2018 14:56:02 GMT):
@mjmckean I did I think, I fixed it by making sure my Indy-node and Indy-SDK are same version and the genesis transactions have the correct IP addresses. In some places 127.0.0.1 is used as IP and in other places 10.20.30.201 format. So make sure you are using the correct IP and Ports Also to make sure you have the same version for everything, pull the Indy-SDK using the 1.4.0 Tag and for Indy-Node Vagrant (or Docker) make sure you use the 'stable' branch for add-apt-repository command anywhere, such as within agent.sh , validator.sh and lib_indy.sh files.. <- these are script files that are executed when VagrantFile is executed to create the VMs

the_identity_guy (Thu, 07 Jun 2018 14:56:02 GMT):
@mjmckean I did I think, I fixed it by making sure my Indy-node and Indy-SDK are same version and the genesis transactions have the correct IP addresses. In some places 127.0.0.1 is used as IP and in other places 10.20.30.201 format. So make sure you are using the correct IP and Ports Also to make sure you have the same version for everything, pull the Indy-SDK using the 1.4.0 Tag and for Indy-Node Vagrant (or Docker) make sure you use the 'stable' branch for add-apt-repository command everywhere that command appears.. such as within agent.sh , validator.sh and lib_indy.sh files.. <- these are script files that are executed when VagrantFile is executed to create the VMs

joeldudleyr3 (Thu, 07 Jun 2018 15:04:40 GMT):
Has joined the channel.

joeldudleyr3 (Thu, 07 Jun 2018 15:05:22 GMT):
Hey guys :) I had some questions about verinyms vs pseudonyms, not sure if this is the right place to ask?

bdeva (Thu, 07 Jun 2018 15:06:24 GMT):
Has joined the channel.

joeldudleyr3 (Thu, 07 Jun 2018 15:15:23 GMT):
Ah dammit, found the Trust Framework whitepaper :D

joeldudleyr3 (Thu, 07 Jun 2018 15:15:44 GMT):
The laughing smiley really doesn't look like it's smiling...

joeldudleyr3 (Thu, 07 Jun 2018 15:15:44 GMT):
The laughing smiley really doesn't look like it's laughing...

bdeva (Thu, 07 Jun 2018 15:22:02 GMT):
encoded

joeldudleyr3 (Thu, 07 Jun 2018 15:28:20 GMT):
@mjmckean While I'm here - I was getting that error too, just running the sample code in Java

mjmckean (Thu, 07 Jun 2018 15:47:19 GMT):
@the_identity_guy I'll try that out, thank you!

mjmckean (Thu, 07 Jun 2018 15:50:04 GMT):
@joeldudleyr3 If it works for me, I'll let you know

joeldudleyr3 (Thu, 07 Jun 2018 15:50:13 GMT):
Awesome :)

mjmckean (Thu, 07 Jun 2018 17:25:38 GMT):
@joeldudleyr3 I got it to work! I used Docker, and in my indy-sdk repo, I edited my pool's genesis file to use 127.0.0.1 as its client ip and node ip. Does that help?

swcurran (Thu, 07 Jun 2018 18:04:12 GMT):
@mjmckean @joeldudleyr3 - usiing docker and hard coding the ips to 127.0.0.1 will work on a Mac, won't on Windows. Not sure about linux. This command will work from a bash shell on all platforms to set DOCKERHOST so that you can use that in scripts. export DOCKERHOST=${APPLICATION_URL-$(docker run --rm --net=host codenvy/che-ip)}

swcurran (Thu, 07 Jun 2018 18:04:12 GMT):
@mjmckean @joeldudleyr3 - using docker and hard coding the ips to 127.0.0.1 will work on a Mac, won't on Windows. Not sure about linux. This command will work from a bash shell on all platforms to set DOCKERHOST so that you can use that in scripts. export DOCKERHOST=${APPLICATION_URL-$(docker run --rm --net=host codenvy/che-ip)}

swcurran (Thu, 07 Jun 2018 18:04:12 GMT):
@mjmckean @joeldudleyr3 - using docker and hard coding the ips to 127.0.0.1 will work on a Mac, won't on Windows. Not sure about linux. This command will work from a bash shell on all platforms to set a DOCKERHOST environment variable so that you can use that in scripts. export DOCKERHOST=${APPLICATION_URL-$(docker run --rm --net=host codenvy/che-ip)}

swcurran (Thu, 07 Jun 2018 18:05:23 GMT):
After you do that, you can just `echo $DOCKERHOST` to see the value, and use DOCKERHOST in your scripts. Some use that env variable already.

dbluhm (Thu, 07 Jun 2018 18:57:17 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=GMwM7rn7TmHZbtFEx) @swcurran It does work on linux

rogeriob2br (Thu, 07 Jun 2018 21:21:00 GMT):
Has joined the channel.

abraham (Fri, 08 Jun 2018 05:20:47 GMT):
Has joined the channel.

AvikHazra (Fri, 08 Jun 2018 06:38:03 GMT):
hello, I`m following https://github.com/hyperledger/indy-node/blob/master/docs/setup-dev.md then complete the setup. But how to start indy-node ? Can you please help me ?

AvikHazra (Fri, 08 Jun 2018 06:56:31 GMT):
When I`m useing generate_indy_pool_transactions --nodes=1 --clients=1 --nodeNum=1 --network=local its throwing generate_indy_pool_transactions: command not found

AvikHazra (Fri, 08 Jun 2018 06:56:31 GMT):
When I`m useing generate_indy_pool_transactions --nodes=1 --clients=1 --nodeNum=1 --network=local its throwing generate_indy_pool_transactions: command not found Please, anyone, help me. what i have done wrong ?

aKesav (Fri, 08 Jun 2018 07:32:58 GMT):
Has joined the channel.

dungnguyen116 (Fri, 08 Jun 2018 10:26:50 GMT):
Has joined the channel.

swcurran (Fri, 08 Jun 2018 12:57:43 GMT):
I'm not sure whether this is the error you are hitting, but you cannot have a network with less than 4 nodes, as it the consensus algorithm requires at least 4.

kdenhartog (Sat, 09 Jun 2018 02:33:56 GMT):
@swcurran how did you test this on windows? I want to make sure the systems I'm developing will work on there too (although I personally hate developing on windows)

kdenhartog (Sat, 09 Jun 2018 02:35:22 GMT):
ideally, I want this to be able to just run the same environment within a docker image so that everyone will encounter similar problems and we can start fixing them and in the short term developing work arounds to fix it

enjoythecode (Sat, 09 Jun 2018 06:18:03 GMT):
Has joined the channel.

SjirNijssen (Sat, 09 Jun 2018 07:33:50 GMT):
Has joined the channel.

AvikHazra (Sat, 09 Jun 2018 09:29:32 GMT):
hello, After running the demo, status always "Attempting connection to sandbox Indy network". Why this is happening ?

mgk (Sat, 09 Jun 2018 16:04:32 GMT):
Has joined the channel.

swcurran (Sat, 09 Jun 2018 16:45:35 GMT):
@kdenhartog - not sure what you mean by "this" :-). Agree on using docker for consistency - von-image (base docker images for nodes and clients) and von-network (4-node network) are good starting points. Having a "manage" script (see von-network) with docker-compose that works across platforms is also very helpful - handling things like the DOCKERHOST problem on all platforms. With "git bash" and Linux Subsystem for Windows (WSL - what a name...) and a good editor - VS-Code seems to be the best right now - makes Windows development reasonable. Although - I'm not a developer, so who am I to say. :-)

kdenhartog (Sat, 09 Jun 2018 19:52:48 GMT):
When I say this I mean the inconsistency of what's happening when people try to start things up. I'll definitely take a look at using a makefile as well. That's made it much easier for me to run docker.

swcurran (Sun, 10 Jun 2018 00:54:36 GMT):
We are running docker. As long as we get the dockerhost right, all good on Windows, however - Mac and Linux script builders tend to use ifconfig or hardcode the docker IP address, which doesn't work on Windows. That's why we are using the codenvy docker image to get the correct IP address.

swcurran (Sun, 10 Jun 2018 00:54:36 GMT):
We are running docker on Mac and Windows. As long as we get the dockerhost right, all good on Windows, however - Mac and Linux script builders tend to use ifconfig or hardcode the docker IP address, which doesn't work on Windows. That's why we are using the codenvy docker image to get the correct IP address.

kdenhartog (Sun, 10 Jun 2018 01:05:43 GMT):
Good to know. I want to try and schedule a call with you and Bryce from IBM to find out more about why you designed your setups the ways you did. We've built internal scripts that. change based upon the identified OS so we could probably do something similar.

SjirNijssen (Sun, 10 Jun 2018 17:55:58 GMT):
To indy experts Some history The Hyperledger T & E WG approved June 4th the following: "Proposed transaction: Let us select the 15 most essential concepts (is not terms) in Fabric, Sawtooth, Burrow and Iroha, and describe clearly, using illustrating relevant examples, understandable to end users and management, how they relate, overlap and where they are different; look forward to endorsements." On behalf of the WG I submit to you the first two concepts descriptions: A blockchain is an append-only data structure; a blockchain (instance) consists at any moment in time of a number of blocks. A block contains a number of transactions. The contents of a block is encrypted and each block contains a hash of major parts. It contains also a copy of the hash of the previous block and this is the reason for the term chain. A transaction proposal is a proposal by a client application or administrator to the blockchain network community. The network community can decide to make the transaction proposal a valid transaction, or is classified as an invalid transaction. Please let us know which improvements you want to make to these two concept definitions. Thanks Current situation • SjirNijssen 19:44 @KellyCooper @roleybrown @amundson @dan @yacovm @tkuhrt and members of the Training & Education WG, The feedback of amundson is incorporated, as well as Dan's and Tracy's. The feedback of yacov about the concept hash has hastened the introduction of concepts lists per audience; the feedback of Yacov was probably directed towards computer scientists or software engineers and I did not want to go that deep for decision makers and end users. So yacovm when we move up through the six audiances you suggestions will there get additional space. The response to the first two concept definitions has been encouraging. As a result there are now in the second increment 8 concept definitions. The proposal has been extended to make the concept list specific for 6 audiances, decision makers and end users, blockchain architects and blockchain consultants, admins, client application developers, smart contract developers and Hyperledger Platform developers. Everyone is invited to add his suggestions at https://wiki.hyperledger.org/groups/training-and-education/list-of-blockchain-core-concepts Thanks for your help. Please consider the list of 8 definitions and let us know whether we need to give additional attention to Indy concepts; your suggestions are very welcome. Thanks for your attention.

yacovm (Sun, 10 Jun 2018 17:55:59 GMT):
Has joined the channel.

roleybrown (Sun, 10 Jun 2018 17:56:00 GMT):
Has joined the channel.

AvikHazra (Mon, 11 Jun 2018 07:10:12 GMT):
hello, I am using Indy-Node docker. When I`m trying to "accept request from "Faber College"" then always returning Request not yet verified. Connection not yet synchronized. Request acceptance aborted. Cannot sync because not connected. Please connect first.

AvikHazra (Mon, 11 Jun 2018 07:10:12 GMT):
hello, I am using Indy-Node docker. When I`m trying to "accept request from "Faber College"" then always returning Request not yet verified. Connection not yet synchronized. Request acceptance aborted. Cannot sync because not connected. Please connect first.

AvikHazra (Mon, 11 Jun 2018 07:10:12 GMT):
hello, I am using Indy-Node docker. When I`m trying to "accept request from "Faber College"" then always returning Request not yet verified. Connection not yet synchronized. Request acceptance aborted. Cannot sync because not connected. Please connect first. Please help me. What is wrong ?

PhillipGibb (Mon, 11 Jun 2018 14:00:56 GMT):
Has joined the channel.

yacovm (Mon, 11 Jun 2018 14:12:37 GMT):
Has left the channel.

Vikrum (Mon, 11 Jun 2018 15:32:44 GMT):
Has joined the channel.

slr (Mon, 11 Jun 2018 16:55:06 GMT):
Has joined the channel.

slr (Mon, 11 Jun 2018 17:11:08 GMT):
Hi! I'm new to indy-node and I'm trying to implement a pretty simple use case to just create a new connection request. I've been looking for a while now and I'm not sure how to do that, changing/adding files in the sample folder doesn't seem to be working. Please help!

nage (Tue, 12 Jun 2018 13:44:54 GMT):
Are you just trying with indy-node, or are you working from the examples in the indy-sdk repository?

AvikHazra (Tue, 12 Jun 2018 13:45:51 GMT):
I`m working from indy-sdk example with docker

segevm (Tue, 12 Jun 2018 14:06:20 GMT):
Has joined the channel.

segevm (Tue, 12 Jun 2018 14:26:08 GMT):
Hi, I'm new to indy as well--just to clarify, I need to install both indy-node and indy-sdk in order to get a test network up and running and start tinkering with it right?

dbluhm (Tue, 12 Jun 2018 14:33:59 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=qmSCCWaYsmQYM7NMp) @segevm You don't have to install indy-node to get a test network going. In the Indy-sdk repo there instructions to start a test network using docker containers. Look through the main README and you should see those instructions 🙂

dbluhm (Tue, 12 Jun 2018 14:33:59 GMT):
@segevm You don't have to install indy-node to get a test network going. In the Indy-sdk repo there are instructions to start a test network using docker containers. Look through the main README and you should see those instructions 🙂

dbluhm (Tue, 12 Jun 2018 14:34:37 GMT):
From there, you will interact with the network through indy-sdk

segevm (Tue, 12 Jun 2018 14:36:42 GMT):
@dbluhm great, thank you!

dbluhm (Tue, 12 Jun 2018 14:38:22 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=aJ2ypdPnpJvwzqJPB) @segevm If you have any other questions, feel free to ask here or on #indy-sdk !

dbluhm (Tue, 12 Jun 2018 15:29:57 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=XhZDvjQEvqdpuecGh) @slr Welcome to the Indy community! I think your question may be easier to answer with some more information :slight_smile: First off, are you running the getting started guide from inside of a docker container or did you set it up to run natively?

kenty (Tue, 12 Jun 2018 15:43:43 GMT):
Has joined the channel.

slr (Tue, 12 Jun 2018 19:07:30 GMT):
@dbluhm thanks! I set it up with vagrant/virtualbox. I actually think I got that part working. I was wondering however if anyone else was having issues running pytest? I seem to run into a bunch of issues

mjmckean (Tue, 12 Jun 2018 20:53:04 GMT):
@slr I was able to successfully build on MacOS/Docker, but pytest also failed for me. @dbluhm didn’t you work on the python wrapper? Or was that someone else

AntoniosHadji (Tue, 12 Jun 2018 21:10:39 GMT):
Has joined the channel.

SherifMuhammed (Tue, 12 Jun 2018 21:41:29 GMT):
Has joined the channel.

devin-fisher (Tue, 12 Jun 2018 23:44:18 GMT):
Is there anyone online that can help me get a build started on the hyperledger jenkins? Looking for help with Indy-node.

shiyj (Wed, 13 Jun 2018 04:29:25 GMT):
Has joined the channel.

mboyd (Wed, 13 Jun 2018 13:18:04 GMT):
@dbluhm What is Steven's handle? He could probably help ^^

dbluhm (Wed, 13 Jun 2018 13:18:36 GMT):
@SteveGoob

haggs (Wed, 13 Jun 2018 13:50:51 GMT):
Has joined the channel.

SteveGoob (Wed, 13 Jun 2018 16:32:35 GMT):
@devin-fisher What are you looking to add to the jenkins pipeline? I'm not quite sure of the process, (I haven't need to add anything to it, yet), but I know the people who know, so we should be able to figure this out.

SteveGoob (Wed, 13 Jun 2018 16:32:35 GMT):
@devin-fisher What are you looking to add to the jenkins pipeline? I'm not quite sure of the process, (I haven't needed to add anything to it, yet), but I know the people who know, so we should be able to figure this out.

devin-fisher (Wed, 13 Jun 2018 17:38:04 GMT):
Thanks for replying @SteveGoob. I was not very clear. There is already jenkins pipeline for indy-node. But it was not being triggered for my PR. I got @ashcherbakov to help me. So, I'm good. He also explained why it was not triggering for me and I have fixed that too. Thanks for being welling to help.

Sean_Bohan (Wed, 13 Jun 2018 18:32:50 GMT):
folks: On Friday June 15, at 8amPT the BC.gov team and Sovrin Foundation are giving a webinar on their Verified Organizations Network. You can RSVP here: rsvp@sovrin.org

Sean_Bohan (Wed, 13 Jun 2018 18:32:58 GMT):

Clipboard - June 13, 2018 2:32 PM

Hangyu (Thu, 14 Jun 2018 04:07:52 GMT):
Has joined the channel.

LucW (Thu, 14 Jun 2018 09:06:15 GMT):
Hi guys. I am really interested in SSI, and I'd like to know if you have any relevant resources on trust in decentralized systems. I know that openbazaar has a nice concept https://www.openbazaar.org/blog/trust-is-risk-a-decentralized-trust-system/ but this would be relevant for decentralized marketplaces only.

Sean_Bohan (Thu, 14 Jun 2018 13:53:45 GMT):
Hi @LucW LucW. The Indy codebase was originally donated by the Sovrin Foundation, which has been doing a LOT of work on a Trust Framework for their network: https://sovrin.org/library/sovrin-trust-framework/ INDY does not have a trust framework (to be clear).

LucW (Thu, 14 Jun 2018 14:02:02 GMT):
Alright, thanks, I was indeed looking into Sovrin.

Sean_Bohan (Thu, 14 Jun 2018 14:24:11 GMT):
Today's Indy WG Call: Agenda: - Open Sourcing LibVCX (RichardE) - BYU Agent Demo (MatthewH) - Open Discussion Call-in: 8amPT, 9amMT, 11amET, 3pm GMT Where: Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/232861185 Or iPhone one-tap (US Toll): +16465588656,232861185# or +14086380968,232861185# Or Telephone: Dial: +1 646 558 8656 (US Toll) or +1 408 638 0968 (US Toll) Meeting ID: 232 861 185 International numbers available: https://zoom.us/zoomconference?m=a0jD_rTMnh0ZYGQDOKPCNrK_0dP7WPfp1 Hyperledger Indy Community Indy Rocketchat: https://chat.hyperledger.org/channel/indyhttps://chat.hyperledger.org/channel/indy-agenthttps://chat.hyperledger.org/channel/indy-node Indy Wiki: https://wiki.hyperledger.org/projects/indy Indy WG Agendas: https://drive.google.com/open?id=1wNnp1pPS6-1Y4B2oygfI6tOU4eHGmpRH Indy Videos: https://drive.google.com/open?id=1Z8-jR7hnXb57fufE0OXxIfeUn05zNmRt Indy WG Meeting Recordings: https://drive.google.com/open?id=0B_NJV6eJXAA1UlJZMXd3cm1zNDQ

tja 16 (Thu, 14 Jun 2018 17:05:13 GMT):
Has joined the channel.

gmillard (Thu, 14 Jun 2018 18:35:16 GMT):
Has joined the channel.

ibartley (Thu, 14 Jun 2018 18:35:37 GMT):
Has joined the channel.

slr (Thu, 14 Jun 2018 23:18:41 GMT):
I'm trying to get a hold of a public key for an agent I just created--not sure where to look for the json file/how to get a hold of it. Any help would be appreciated!

slr (Thu, 14 Jun 2018 23:38:11 GMT):
scratch that, sorry I think I got it!

OKUURAWataru (Fri, 15 Jun 2018 03:45:44 GMT):
Has joined the channel.

ashcherbakov (Fri, 15 Jun 2018 07:00:49 GMT):
As was announced, the next stable release 1.4 of Indy-Node (which will be issued in a week or so), contains some breaking changes related to transaction format. The changes are done for a better versioning support (that is to support compatibility in future) and more readable format of data (separating payload data, metadata and signature). There will a new stable release of LibIndy (1.5) which should be used to work with IndyNode 1.4. *General* - By default LibIndy 1.5 will be compatible with IndyNode 1.3, and not 1.4 - LibIndy 1.5 can become compatible with IndyNode 1.4 if `indy_set_protocol_version(2)` is called during app initialization. *Guideline for applications* - An application can freely update to LibIndy 1.5 and still use stable Node 1.3 - If an app wants to work with the latest master or Stable Node 1.4, then it needs to - support breaking changes (there are not so many, mostly a new reply for write txns as txn format is changed) There is a migration guide for Indy-Node 1.3 to 1.4: https://github.com/hyperledger/indy-node/blob/master/docs/1.3_to_1.4_migration_guide.md - call `indy_set_protocol_version(2)` during app initialization

JoostZ (Fri, 15 Jun 2018 08:43:15 GMT):
Has joined the channel.

gravity (Fri, 15 Jun 2018 20:53:27 GMT):
Has joined the channel.

ricardoshuree (Sat, 16 Jun 2018 13:59:22 GMT):
Has joined the channel.

DuncanMuhoro (Sun, 17 Jun 2018 11:04:06 GMT):
Has joined the channel.

Miguel_Jimenez (Mon, 18 Jun 2018 16:10:20 GMT):
Has joined the channel.

nehalshah50 (Mon, 18 Jun 2018 16:14:03 GMT):
Has joined the channel.

InukaGunawardana (Mon, 18 Jun 2018 17:34:25 GMT):
Has joined the channel.

dungnguyen116 (Tue, 19 Jun 2018 08:16:38 GMT):
(node:11738) UnhandledPromiseRejectionWarning: IndyError: LedgerInvalidTransaction at Object.callback (/home/nbxtruong/Documents/Hyperledger-Indy-NodeJS/node_modules/indy-sdk/src/wrapIndyCallback.js:15:10)

dungnguyen116 (Tue, 19 Jun 2018 08:16:38 GMT):
``` (node:11738) UnhandledPromiseRejectionWarning: IndyError: LedgerInvalidTransaction at Object.callback (/home/nbxtruong/Documents/Hyperledger-Indy-NodeJS/node_modules/indy-sdk/src/wrapIndyCallback.js:15:10)

dungnguyen116 (Tue, 19 Jun 2018 08:18:09 GMT):
how can i fix this problem ? :(

tylerhester (Tue, 19 Jun 2018 15:37:55 GMT):
Has joined the channel.

tylerhester (Tue, 19 Jun 2018 21:36:15 GMT):
Hmmm, I am trying get up and running with indy and I'm following the "how-tos" on writing a did and querying a verkey. I seem to be having some trouble here. I'm using python, on MacOSX, and running the python file results in an error `ErrorCode: Pool Genesis Transactions are not compatible with Protocol version Libindy PROTOCOL_VERSION is 1 but Pool Genesis Transactions are of version 1.4` I try the following: `await pool.set_protocol_version(2)` and I get another error `AttributeError: module 'indy.pool' has no attribute 'set_protocol_version'

tylerhester (Tue, 19 Jun 2018 21:36:15 GMT):
Hmmm, I am trying get up and running with indy and I'm following the "how-tos" on writing a did and querying a verkey. I seem to be having some trouble here. I'm using python, on MacOSX, and running the python file results in an error `ErrorCode: Pool Genesis Transactions are not compatible with Protocol version Libindy PROTOCOL_VERSION is 1 but Pool Genesis Transactions are of version 1.4` I try the following: `await pool.set_protocol_version(2)` and I get another error `AttributeError: module 'indy.pool' has no attribute 'set_protocol_version'`

PhillipGibb (Wed, 20 Jun 2018 12:40:01 GMT):
Hi, is there a place we can talk #VSX ? In particular provisioning an agent.

PhillipGibb (Wed, 20 Jun 2018 12:40:01 GMT):
Hi, is there a place we can talk #VCX ? In particular provisioning an agent.

tomislav (Wed, 20 Jun 2018 12:41:17 GMT):
There's agent related channel at #indy-agent

PhillipGibb (Wed, 20 Jun 2018 12:41:33 GMT):
tks

CarlosAvim (Wed, 20 Jun 2018 12:50:41 GMT):
Has joined the channel.

CarlosAvim (Wed, 20 Jun 2018 13:34:03 GMT):
Hello Indy ppl, I'm planning to do a thesis work on decentralized identities and wondering if Indy is stable enough to build a full application around the ledger. Or should I go for HL Fabric and add SSI with the business model ? Can someone advice me by giving a feedback on his experience, any poc ? Thanks!

tomislav (Wed, 20 Jun 2018 13:40:15 GMT):
@CarlosAvim Stable channel of Indy is stable and you can build full application platform on top of it. Thesis on DIDs should use a ledger that's built for SSI. Verifiable credentials is unique feature of Indy SDK. I'm currently doing work across different application platforms and use the stable channel, things run pretty well across the board. A bit of a learning curve, but nothing too terrible.

CarlosAvim (Wed, 20 Jun 2018 13:50:25 GMT):
@tomislav I am not afraid of the learning curve :) Thanks for the feedback! I guess I should get started and see this by myself.

asardo (Wed, 20 Jun 2018 14:23:25 GMT):
Has joined the channel.

kdenhartog (Wed, 20 Jun 2018 16:47:56 GMT):
@tylerhester i believe changes are being made in master right now that are affecting this part of the sdk. Try working from the stable version instead of master.

expire (Wed, 20 Jun 2018 16:48:49 GMT):
Has joined the channel.

Sean_Bohan (Wed, 20 Jun 2018 16:51:24 GMT):
INDY WG Call Agenda - Tomorrow Thurs 6/21/2018 AGENDA: Housekeeping (Sean) Evolution of the Current Indy Crypto codebase (MikeL) Shared Crypto Library at Hyperledger, kickoff and next steps at AMS Hackfest (NathanG) Open Discussion (All) When: TOMORROW - Thurs June 21, 2018 8amPT, 9amMT, 11amET, 3pm GMT Where: Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/232861185 Or iPhone one-tap (US Toll): +16465588656,232861185# or +14086380968,232861185# Or Telephone: Dial: +1 646 558 8656 (US Toll) or +1 408 638 0968 (US Toll) Meeting ID: 232 861 185 International numbers available: https://zoom.us/zoomconference?m=a0jD_rTMnh0ZYGQDOKPCNrK_0dP7WPfp1 Hyperledger Indy Community Indy Rocketchat: https://chat.hyperledger.org/channel/indyhttps://chat.hyperledger.org/channel/indy-agenthttps://chat.hyperledger.org/channel/indy-node Indy Wiki: https://wiki.hyperledger.org/projects/indy Indy WG Agendas: https://drive.google.com/open?id=1wNnp1pPS6-1Y4B2oygfI6tOU4eHGmpRH Indy Videos: https://drive.google.com/open?id=1Z8-jR7hnXb57fufE0OXxIfeUn05zNmRt Indy WG Meeting Recordings: https://drive.google.com/open?id=0B_NJV6eJXAA1UlJZMXd3cm1zNDQ

tylerhester (Wed, 20 Jun 2018 16:57:20 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=57685b6a-e855-4069-8dde-f1971425525e) @kdenhartog Thanks for the heads up :) I was able to get past that error by simply upgrading my python wrapper to the right version.

tylerhester (Wed, 20 Jun 2018 16:57:20 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=57685b6a-e855-4069-8dde-f1971425525e) @kdenhartog Thanks for the heads up :) I was able to get past that error by simply upgrading my python wrapper to the right version. (.4.0-dev-591 )

tylerhester (Wed, 20 Jun 2018 16:57:20 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=57685b6a-e855-4069-8dde-f1971425525e) @kdenhartog Thanks for the heads up :) I was able to get past that error by simply upgrading my python wrapper to the right version. (1.4.0-dev-591 )

kdenhartog (Wed, 20 Jun 2018 17:00:28 GMT):
Glad to hear! Feel free to @ me if you have more questions as well.

Katie_Wei (Wed, 20 Jun 2018 18:41:39 GMT):
Has joined the channel.

tylerhester (Wed, 20 Jun 2018 18:41:58 GMT):
Can anyone help me understand the difference between evernym's "indy-sdk" and hyperledger's "indy-sdk" I know that sovrin was re-branded as Indy AFAIK, but it seems like both repositories are being developed on. Where is the line drawn between "sovrin" and "indy"

voddan (Thu, 21 Jun 2018 09:18:51 GMT):
Has joined the channel.

dbluhm (Thu, 21 Jun 2018 12:30:19 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=4YMePcHLeEt2fkL6M) @tylerhester Sovrin still exists as a separate entity from both evernym and the Hyperledger Indy project. Sovrin builds on top of and contributes actively to Indy. There should be only one "indy-sdk" but evernym did recently open source a repository called "SDK" which contains an additional API on top of indy-sdk that they use to create their products. That sdk, as I understand it, will eventually be merged with indy-sdk.

dbluhm (Thu, 21 Jun 2018 12:32:08 GMT):
You should be using hyperledger's indy-sdk repository as evernym's repository will eventually merge with it.

Sean_Bohan (Thu, 21 Jun 2018 14:41:07 GMT):
Some recent videos: 6/14 Indy WG Call: https://drive.google.com/open?id=1yL8_BvWXE0S38gBsXBdos6FfO9cO3CoL 6/15 Agent WG Call: https://drive.google.com/file/d/17ahwKndVORXM2NKCSFLYrlbs1VaD1XXn/view?usp=sharing Wallet Architecture HIPE Discussion with DanielH https://drive.google.com/open?id=1gGkjpu1ZTJiI0RtSIroB58QRzCXBkoWG BC.gov Verifiable Organizations Network Demo: https://drive.google.com/open?id=1Ms2vKqTe-jVDpz5eWtQIqIz8Jl6TLbSE Decentralized Identity Distilled (Hyperledger Webinar w/ Daniel Hardman): https://gateway.on24.com/wcc/gateway/linux/1101876/1678323/decentralized-identity-distilled

tylerhester (Thu, 21 Jun 2018 15:06:49 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=24ad3807-af2c-47c0-ad86-87c12f6f961a) @dbluhm Great! Thanks for the information :)

arunwij (Fri, 22 Jun 2018 09:17:44 GMT):
Has joined the channel.

arunwij (Fri, 22 Jun 2018 09:24:47 GMT):
hello guys. Beginner here. I just started to learn indy project and created test network according to this guide https://github.com/hyperledger/indy-node/blob/stable/environment/vagrant/training/vb-multi-vm/TestIndyClusterSetup.md. Since there are many repositories i am confused where to start learn indy. Can anyone give me suggestions on that? Where to start.

ashcherbakov (Fri, 22 Jun 2018 11:04:04 GMT):
Hi @arunwij I think the best way to start is with a Getting Started guide: https://github.com/hyperledger/indy-sdk/blob/master/doc/getting-started/getting-started.md

hendry19901990 (Fri, 22 Jun 2018 23:34:27 GMT):
Hello everyone i need help

hendry19901990 (Fri, 22 Jun 2018 23:34:53 GMT):
there is an api like ethereum json-rpc

hendry19901990 (Fri, 22 Jun 2018 23:35:14 GMT):
for connecting ?

rudimk (Sat, 23 Jun 2018 04:15:49 GMT):
Has joined the channel.

arunwij (Mon, 25 Jun 2018 06:28:18 GMT):
@ashcherbakov Thank you. How can I setup environment to run this guide? Can I use VM based indy test network for this?

srinivasanraghavan (Mon, 25 Jun 2018 08:22:11 GMT):
Apologies as this might be naive question here. I looking into an experimentation to integrate a cloud hsm to indy and offload the crypto ops to hsm and let all keys stay in HSM . The first point I see is like implementing a custom wallet using api/wallet.rs in indy-sdk which could talk to the cloud HSM using PKCS11 interface . Is this a good to start and would I see the code wallet is passed as an handle to other ops .

srinivasanraghavan (Mon, 25 Jun 2018 08:22:11 GMT):
Apologies as this might be naive question here. I am looking into an experimentation to integrate a cloud hsm to indy and offload the crypto ops to hsm and let all keys stay in HSM . The first point I see is like implementing a custom wallet using api/wallet.rs in indy-sdk which could talk to the cloud HSM using PKCS11 interface . Is this a good to start and would I see the code wallet is passed as an handle to other ops .

srinivasanraghavan (Mon, 25 Jun 2018 08:22:11 GMT):
Apologies as this might be naive question here. I am looking into an experimentation to integrate a cloud hsm to indy and offload the crypto ops to hsm and let all keys stay in HSM . The first point I see is like implementing a custom wallet using api/wallet.rs in indy-sdk which could talk to the cloud HSM using PKCS11 interface . Is this a good to start as I see the in the code wallet is passed as an handle to other ops .

ashcherbakov (Mon, 25 Jun 2018 08:32:02 GMT):
@arunwij https://github.com/hyperledger/indy-sdk/blob/master/doc/getting-started/run-getting-started.md You can run the pool of nodes in the Docker as described there

Antonio_M (Mon, 25 Jun 2018 14:16:33 GMT):
Has joined the channel.

Antonio_M (Mon, 25 Jun 2018 14:17:19 GMT):
hi, is there a way of using indy as a fabric's msp? if so, are there any nice examples available? thank you

SherifMuhammed (Mon, 25 Jun 2018 20:16:13 GMT):
Hi everyone , I am using Indy 1.3.62 & trying to connect to sandbox it gives me "Connecting to sandbox..." I am using vagrant to set up the environment. Thank You.

alvaradojl (Tue, 26 Jun 2018 07:56:33 GMT):
Has joined the channel.

Xzccc (Tue, 26 Jun 2018 11:14:14 GMT):
Has joined the channel.

Xzccc (Tue, 26 Jun 2018 11:18:26 GMT):
Hi guys,when I run "start_indy_node" command with "start_indy_node Node1 0.0.0.0 9701 0.0.0.0 9702", it tell me "ValueError: invalid literal for int() with base 10: '0.0.0.0' ". However I can run init_indy_keys successfully. I download the indy by "pip3 install indy-node" and my os is ubuntu 16.04 x86 and python version is 3.5.2 What's wrong with me?Thanks a lot

Xzccc (Tue, 26 Jun 2018 11:18:26 GMT):
Hi guys,when I run "start_indy_node" command with "start_indy_node Node1 0.0.0.0 9701 0.0.0.0 9702", it tell me "ValueError: invalid literal for int() with base 10: '0.0.0.0' ". However I can run init_indy_keys successfully. I download the indy by "pip3 install indy-node" and my os is ubuntu 16.04 x86 and python version is 3.5.2 What's wrong with me?Thanks a lot

Xzccc (Tue, 26 Jun 2018 11:18:45 GMT):

Clipboard - June 26, 2018 7:18 PM

Xzccc (Tue, 26 Jun 2018 11:19:21 GMT):
root@cly:~# start_indy_node Node1 0.0.0.0 9701 0.0.0.0 9702 2018-06-26 19:00:26,547 | DEBUG | __init__.py ( 60) | register | Registered VCS backend: git 2018-06-26 19:00:26,600 | DEBUG | __init__.py ( 60) | register | Registered VCS backend: hg 2018-06-26 19:00:26,639 | DEBUG | __init__.py ( 60) | register | Registered VCS backend: svn 2018-06-26 19:00:26,639 | DEBUG | __init__.py ( 60) | register | Registered VCS backend: bzr Traceback (most recent call last): File "/usr/local/bin/start_indy_node", line 17, in run_node(config, self_name, int(sys.argv[2]), int(sys.argv[3])) ValueError: invalid literal for int() with base 10: '0.0.0.0'

ltzMaxwell (Tue, 26 Jun 2018 14:24:03 GMT):
Has joined the channel.

tomislav (Tue, 26 Jun 2018 14:36:11 GMT):
@Xzccc I think the best way to run a node for development and testing is to use docker images

slr (Tue, 26 Jun 2018 17:22:18 GMT):
Using indy-node, how do you actually query the ledger? any help would be appreciated

swcurran (Tue, 26 Jun 2018 17:58:22 GMT):
@slr - take a look at UI for viewing the ledger here - http://159.89.115.24/. Click the "Ledger: Domain (pretty)" link to see the transactions (not blocks) on the ledger. The repo for this code is here - https://github.com/bcgov/von-network - take a look at the server folder to see the call for displaying the data.

mjmckean (Tue, 26 Jun 2018 19:13:55 GMT):
@SherifMuhammed I would try setting it up with docker--the vagrant process seems to be deprecated

ltzMaxwell (Wed, 27 Jun 2018 06:09:53 GMT):
@tomislav @mjmckean Hi ,do we have an already built indy_pool image? cuz it always comes error in building process.

ltzMaxwell (Wed, 27 Jun 2018 06:10:13 GMT):
Seem to be something wrong with sovrin apt repo.

ltzMaxwell (Wed, 27 Jun 2018 06:38:27 GMT):
Hit:1 http://archive.ubuntu.com/ubuntu xenial InRelease Hit:2 http://security.ubuntu.com/ubuntu xenial-security InRelease Get:3 http://archive.ubuntu.com/ubuntu xenial-updates InRelease [109 kB] Get:4 http://archive.ubuntu.com/ubuntu xenial-backports InRelease [107 kB] Ign:5 https://repo.sovrin.org/deb xenial InRelease Ign:6 https://repo.sovrin.org/deb xenial Release Ign:7 https://repo.sovrin.org/deb xenial/master amd64 Packages Ign:8 https://repo.sovrin.org/deb xenial/master all Packages Ign:7 https://repo.sovrin.org/deb xenial/master amd64 Packages Ign:8 https://repo.sovrin.org/deb xenial/master all Packages Ign:7 https://repo.sovrin.org/deb xenial/master amd64 Packages Ign:8 https://repo.sovrin.org/deb xenial/master all Packages Ign:7 https://repo.sovrin.org/deb xenial/master amd64 Packages Ign:8 https://repo.sovrin.org/deb xenial/master all Packages Ign:7 https://repo.sovrin.org/deb xenial/master amd64 Packages Ign:8 https://repo.sovrin.org/deb xenial/master all Packages Err:7 https://repo.sovrin.org/deb xenial/master amd64 Packages gnutls_handshake() failed: An unexpected TLS packet was received. Ign:8 https://repo.sovrin.org/deb xenial/master all Packages Fetched 216 kB in 2min 22s (1516 B/s) Reading package lists... Done W: The repository 'https://repo.sovrin.org/deb xenial Release' does not have a Release file. N: Data from such a repository can't be authenticated and is therefore potentially dangerous to use. N: See apt-secure(8) manpage for repository creation and user configuration details. E: Failed to fetch https://repo.sovrin.org/deb/dists/xenial/master/binary-amd64/Packages gnutls_handshake() failed: An unexpected TLS packet was received. E: Some index files failed to download. They have been ignored, or old ones used instead.

Xzccc (Wed, 27 Jun 2018 06:45:03 GMT):
@tomislav Hello guy, when I use docker imagers by running ./pool_start.sh as follow the doc in github, it goes wrong .Should I use root to run this shell script? If I use root, it can run but not successfully because I cannot see the docker daemon with "docker ps" . If i don't use sudo, it cannot run and tell me permission denied...And I wonder somewhere has the detailed guidance doc about use docker to run node? Thank you!

Xzccc (Wed, 27 Jun 2018 06:48:14 GMT):
Step 9/13 : RUN useradd -ms /bin/bash -l -u $uid indy ---> Running in 6c3e54654392 useradd: UID 0 is not unique The command '/bin/sh -c useradd -ms /bin/bash -l -u $uid indy' returned a non-zero code: 4 Building node4 Sending build context to Docker daemon 34.3kB Step 1/20 : FROM indycore pull access denied for indycore, repository does not exist or may require 'docker login' Node Node4 0.0.0.0 9707 0.0.0.0 9708 created Writing pool data node1 10.0.0.2 9701 9702,node2 10.0.0.3 9703 9704,node3 10.0.0.4 9705 9706,node4 10.0.0.5 9707 9708 to file pool_data Pool created Reading pool data Pool data is node1 10.0.0.2 9701 9702,node2 10.0.0.3 9703 9704,node3 10.0.0.4 9705 9706,node4 10.0.0.5 9707 9708 Creating pool network pool-network 12e6a227fff7042f20ac27f40653899f3d4ceccaf7f6c3c459abfcf29f118a61 Starting pool Starting node node1 10.0.0.2 Unable to find image 'node1:latest' locally docker: Error response from daemon: pull access denied for node1, repository does not exist or may require 'docker login'. See 'docker run --help'. Error: No such container: node1 Node node1 started on 10.0.0.2 Starting node node2 10.0.0.3 Unable to find image 'node2:latest' locally docker: Error response from daemon: pull access denied for node2, repository does not exist or may require 'docker login'. See 'docker run --help'. Error: No such container: node2 Node node2 started on 10.0.0.3 Starting node node3 10.0.0.4 Unable to find image 'node3:latest' locally docker: Error response from daemon: pull access denied for node3, repository does not exist or may require 'docker login'. See 'docker run --help'. Error: No such container: node3 Node node3 started on 10.0.0.4 Starting node node4 10.0.0.5 Unable to find image 'node4:latest' locally docker: Error response from daemon: pull access denied for node4, repository does not exist or may require 'docker login'. See 'docker run --help'. Error: No such container: node4 Node node4 started on 10.0.0.5 Pool started lab420@lab420-PowerEdge-R730:/opt/indy-node/environment/docker/pool$ sudo docker ps CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES lab420@lab420-PowerEdge-R730:/opt/indy-node/environment/docker/pool$

tomislav (Wed, 27 Jun 2018 12:33:58 GMT):
@Xzccc I just run build and run the pool directly from the indy-sdk main directory. `docker build -f ci/indy-pool.dockerfile -t indy_pool .` to build and `docker run -itd -p 9701-9709:9701-9709 indy_pool` to run the built image

tomislav (Wed, 27 Jun 2018 12:33:58 GMT):
@Xzccc @ltzMaxwell I just run build and run the pool directly from the indy-sdk main directory. `docker build -f ci/indy-pool.dockerfile -t indy_pool .` to build and `docker run -itd -p 9701-9709:9701-9709 indy_pool` to run the built image

tomislav (Wed, 27 Jun 2018 12:35:14 GMT):
This will run 4 nodes in a container with 127.0.0.1 as their IP and genesis txn

Xzccc (Wed, 27 Jun 2018 13:11:30 GMT):
@tomislav Ok,Thanks a lot. I will try

Xzccc (Wed, 27 Jun 2018 13:22:33 GMT):
Could I use docker to setup a cluster on several servers?

adamludvik (Wed, 27 Jun 2018 21:37:45 GMT):
Has joined the channel.

sjain74 (Thu, 28 Jun 2018 02:50:46 GMT):
Has joined the channel.

sjain74 (Thu, 28 Jun 2018 02:52:10 GMT):
Hi, on mac, while doing docker-compose up following https://github.com/hyperledger/indy-sdk/blob/master/doc/getting-started/run-getting-started.md, I am seeing the following error:

sjain74 (Thu, 28 Jun 2018 02:52:24 GMT):
test ledger_demo_works ... thread 'ledger_demo_works' panicked at 'called `Result::unwrap()` on an `Err` value: Timeout', libcore/result.rs:945:5

chilipepper (Thu, 28 Jun 2018 08:16:06 GMT):
Has joined the channel.

sergey.minaev (Thu, 28 Jun 2018 08:51:30 GMT):
@sjain74 on mac docker from GSG can't be used for samples or test. Please read https://github.com/hyperledger/indy-sdk/blob/master/README.md#how-to-start-local-nodes-pool-with-docker

SherifMuhammed (Thu, 28 Jun 2018 08:55:18 GMT):
Hi. Are you planning to support integration Indy with popular enterprise blockchains like Fabric and ethereum/quorum ?

Xzccc (Thu, 28 Jun 2018 10:58:30 GMT):
@tomislav Hi,how can I use indy_cli to connect the pool which run as docker locally?I have already run the pool node by `docker run -itd -p 9701-9709:9701-9709 indy_pool` however I cannot use indy_cli to connet in the same server `root@lab:/opt/indy-sdk/cli/target/debug# ./indy-cli indy> pool list There are no pool indy> pool connect sandbox Pool "sandbox" does not exist. ` Sorry to bother you again

Xzccc (Thu, 28 Jun 2018 10:58:30 GMT):
@tomislav Hi,how can I use indy_cli to connect the pool which run as docker locally?I have already run the pool node by `docker run -itd -p 9701-9709:9701-9709 indy_pool` however I cannot use indy_cli to connet in the same server ``root@lab:/opt/indy-sdk/cli/target/debug# ./indy-cli indy> pool list There are no pool indy> pool connect sandbox Pool "sandbox" does not exist.` ` Sorry to bother you again

Xzccc (Thu, 28 Jun 2018 11:10:55 GMT):
I think I have found the doc, follow this guidance doc right? https://github.com/hyperledger/indy-sdk/blob/master/doc/how-tos/write-did-and-query-verkey/python/README.md

arunwij (Thu, 28 Jun 2018 11:14:40 GMT):
Hello everyone, is indy based sovrin network live now? if it so, can anyone create an identity there?

PhillipGibb (Thu, 28 Jun 2018 11:31:28 GMT):
There is the Sovrin Test Network where you can add your Trust Anchor DID and VerKey

PhillipGibb (Thu, 28 Jun 2018 11:33:25 GMT):
@arunwij you have had to already created a DID: https://s3.us-east-2.amazonaws.com/evernym-cs/sovrin-STNnetwork/www/trust-anchor.html

Sean_Bohan (Thu, 28 Jun 2018 14:32:32 GMT):
Folks:

Sean_Bohan (Thu, 28 Jun 2018 14:32:36 GMT):
For those at the Hackfest, we will be doing the Thursday WG call in breakout room 23 1. Housekeeping 2. Hackfest Update 3. New Release Announcement 4. Open Discussion (edited) in 30 minutes! Where: Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/232861185 Or iPhone one-tap (US Toll): +16465588656,232861185# or +14086380968,232861185# Or Telephone: Dial: +1 646 558 8656 (US Toll) or +1 408 638 0968 (US Toll) Meeting ID: 232 861 185 International numbers available: https://zoom.us/zoomconference?m=a0jD_rTMnh0ZYGQDOKPCNrK_0dP7WPfp1 Hyperledger Indy Community Indy Rocketchat: https://chat.hyperledger.org/channel/indyhttps://chat.hyperledger.org/channel/indy-agenthttps://chat.hyperledger.org/channel/indy-node Indy Wiki: https://wiki.hyperledger.org/projects/indy Indy WG Agendas: https://drive.google.com/open?id=1wNnp1pPS6-1Y4B2oygfI6tOU4eHGmpRH Indy Videos: https://drive.google.com/open?id=1Z8-jR7hnXb57fufE0OXxIfeUn05zNmRt Indy WG Meeting Recordings: https://drive.google.com/open?id=0B_NJV6eJXAA1UlJZMXd3cm1zNDQ

Sean_Bohan (Thu, 28 Jun 2018 15:37:51 GMT):
1.3 to 1.4 Migration Guide for Indy Node: https://github.com/hyperledger/indy-node/blob/master/docs/1.3_to_1.4_migration_guide.md

Sean_Bohan (Thu, 28 Jun 2018 15:38:28 GMT):
migration guide for Indy SDK https://github.com/hyperledger/indy-sdk/blob/master/CHANGELOG.md https://github.com/hyperledger/indy-sdk/blob/master/doc/migration-guide-1.4.0-1.5.0.md

Xzccc (Fri, 29 Jun 2018 02:11:00 GMT):
Hi guys, I amand I'm following the "how-tos" on writing a did and querying a verkey.

Xzccc (Fri, 29 Jun 2018 02:16:49 GMT):
Hi guys, I'm following the "how-tos" on writing a did and querying a verkey and running the pool successfully by docker run -itd -p 9701-9708:9701-9708 indy_pool However when I run the write_did.py followed by: https://github.com/hyperledger/indy-sdk/blob/master/doc/how-tos/write-did-and-query-verkey/python/README.md I seem to be having some trouble here, I use python3 on ubuntu16.04, and use docker to run pool on locallly. The error as follow: `# python3 /opt/python_indy/write_did.py 1. Create new pool ledger configuration to connect to ledger. 2. Open ledger and get handle _indy_loop_callback: Function returned error 112 Error occurred: ErrorCode.CommonInvalidState`

Xzccc (Fri, 29 Jun 2018 02:16:49 GMT):
Hi guys, I'm following the "how-tos" on writing a did and querying a verkey and running the pool successfully by docker run -itd -p 9701-9708:9701-9708 indy_pool However when I run the write_did.py followed by: https://github.com/hyperledger/indy-sdk/blob/master/doc/how-tos/write-did-and-query-verkey/python/README.md I seem to be having some trouble here, I use python3 on ubuntu16.04, and use docker to run pool on locallly. The error as follow: `# python3 /opt/python_indy/write_did.py 1. Create new pool ledger configuration to connect to ledger. 2. Open ledger and get handle _indy_loop_callback: Function returned error 112 Error occurred: ErrorCode.CommonInvalidState` However I can run the sample of python in `indy-sdk/samples/python` successfully by `python3 -m src.main` How can I sovle this promble? Thanks a lot!

Xzccc (Fri, 29 Jun 2018 02:16:49 GMT):
Hi guys, I'm following the "how-tos" on writing a did and querying a verkey and running the pool successfully by docker run -itd -p 9701-9708:9701-9708 indy_pool However when I run the write_did.py followed by: https://github.com/hyperledger/indy-sdk/blob/master/doc/how-tos/write-did-and-query-verkey/python/README.md I seem to be having some trouble here, I use python3 on ubuntu16.04, and use docker to run pool on locallly. The error as follow: `# python3 /opt/python_indy/write_did.py 1. Create new pool ledger configuration to connect to ledger. 2. Open ledger and get handle _indy_loop_callback: Function returned error 112 Error occurred: ErrorCode.CommonInvalidState` However I can run the sample of python in `indy-sdk/samples/python` successfully by `python3 -m src.main` How can I sovle this problem? Thanks a lot!

KanupriyaPandey (Fri, 29 Jun 2018 04:39:30 GMT):
Has joined the channel.

KanupriyaPandey (Fri, 29 Jun 2018 05:22:45 GMT):
Hey can anyone tell me where one can find the storage location of the pools and nodes being created on ubuntu vm. I am using indy-sdk. Would really appreciate the help

ricktobacco (Fri, 29 Jun 2018 06:38:10 GMT):
Has joined the channel.

ricktobacco (Fri, 29 Jun 2018 06:38:13 GMT):
So is it unheard of to run two hyperledger projects together. So if I am interested in Indy, but I want smart contract functionality, can I run them together, or at least have them coordinate in some way if they run on separate networks?

Sean_Bohan (Fri, 29 Jun 2018 08:34:38 GMT):
Not unheard of @ricktobacco - ChrisF from IBM mentioned at the hackfest on Weds that "membership as a plugin" (membership = identity a la Indy or others) was something they are/would be working on in Fabric and there is a lot of interest/effort in how Sawtooth and Indy can work together. @nage - any thoughts on running Indy as the identity layer with another framework for smart contracts?

nage (Fri, 29 Jun 2018 08:38:49 GMT):
We really like the idea of using Indy as an Identity layer for other Hyperledger chains. This isn't really possible out of the box with those platforms. It is also possible to use Plenum's extensions to add smart contracts into Indy's chain, and no one has really tried that yet out either, so you have a lot of options. @ricktobacco, if you are at the Hackfest going on now, find me and we can whiteboard through some ideas.

nage (Fri, 29 Jun 2018 08:38:49 GMT):
We really like the idea of using Indy as an Identity layer for other Hyperledger chains. This isn't really possible out of the box with those platforms (yet). It is also possible to use Plenum's extensions to add smart contracts into Indy's chain, and no one has really tried that yet out either, so you have a lot of options. @ricktobacco, if you are at the Hackfest going on now, find me and we can whiteboard through some ideas.

nage (Fri, 29 Jun 2018 08:39:41 GMT):
(I'm in room 24, breakout #3 for the next while)

jaromil (Fri, 29 Jun 2018 11:56:52 GMT):
Has joined the channel.

Xzccc (Fri, 29 Jun 2018 13:22:17 GMT):
Hello everyone. I wonder how to do when meet the error 112 `indy.error.IndyError: ErrorCode.CommonInvalidStateby` with running python sample? I see the doc of https://github.com/hyperledger/indy-sdk/blob/master/wrappers/python/indy/error.py # Invalid library state was detected in runtime. It signals library bug CommonInvalidState = 112, So can I reset the library state? Would really appreciate the help!

LuigiRiva (Fri, 29 Jun 2018 13:23:29 GMT):
Has joined the channel.

LuigiRiva (Fri, 29 Jun 2018 14:36:13 GMT):
Hello everyone, just new to Indy, internally in my company we would like to set up a test env and start playing with it. Could everyone point me out at the right resource in order to do that or, even better, provide me a comprehensive lists of tasks that should be done for having an indy env up & running?

dbluhm (Fri, 29 Jun 2018 17:19:44 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=g5ihsmien8RcHmYxx) @LuigiRiva Welcome to the Indy Community! What are your plans for using Indy? If you're going for creating applications on top of Indy, a good place to start is with the [indy-sdk README](https://github.com/hyperledger/indy-sdk) and then [running](https://github.com/hyperledger/indy-sdk/blob/master/doc/getting-started/run-getting-started.md) and [reading](https://github.com/hyperledger/indy-sdk/blob/master/doc/getting-started/getting-started.md) the getting started guide. The Getting Started Guide can be a little dense with information so if you have questions, feel free to ask around here or in #indy-sdk . After you work through some of that, you will probably be ready to set up a local development environment by following the steps on the Indy-SDK README. If you run into troubles there, again, please ask in #indy-sdk ! :slight_smile:

jpscatterday (Mon, 02 Jul 2018 00:09:12 GMT):
Has joined the channel.

UCDirk (Mon, 02 Jul 2018 07:30:10 GMT):
Has joined the channel.

janl (Mon, 02 Jul 2018 07:35:59 GMT):
Has joined the channel.

janl (Mon, 02 Jul 2018 07:36:14 GMT):
indy-cli

LuigiRiva (Mon, 02 Jul 2018 07:57:37 GMT):
@dbluhm thanks you for welcoming me and for the information. Yes we would like to create applications on top of indy and as well we are looking to see how it can be connected or integrated with Hyperledger Fabric: we have an HF network up and running and we would like to manage authentication through Indy. Any best practice around about that?

AbidiBassem (Mon, 02 Jul 2018 08:32:49 GMT):
Has joined the channel.

slr (Mon, 02 Jul 2018 17:53:27 GMT):
Hi! I was going through the demo again and was wondering what exactly it means when it says "No key present in wallet for making request on Indy, so adding one". This DID for the key doesn't seem to be used anywhere else? I'm a little confused as to what it is. Any help would be appreciated!

dbluhm (Mon, 02 Jul 2018 18:19:36 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=tyHH7dqAE76YJhCpb) @slr Which demo are you referring to?

slr (Mon, 02 Jul 2018 18:22:37 GMT):
getting started demo with indy node

dbluhm (Mon, 02 Jul 2018 18:23:12 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=5CABtjMhQ6QQ9oj7t) @LuigiRiva I'm not aware of any established best practices for integrating Indy into Fabric projects. That doesn't necessarily mean that it doesn't exist though. Conversations about using Indy for authentication and other purposes are on-going and very active.

dbluhm (Mon, 02 Jul 2018 18:24:33 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=pdjs5Y4ZwYAzPydPv) @slr What are you planning to build with Indy? Generally, the getting started guide in Indy-SDK is the best place to start for those new to the Indy Project. There is a pretty detailed descriptions of each interaction there as well

dbluhm (Mon, 02 Jul 2018 18:26:40 GMT):
@slr This is dependent on your use case but usually I'd highly recommend starting there (Indy-SDK Getting Started Guide) and then go from there. Any more questions you might have are more than welcome here and in Indy-SDK :slight_smile:

dbluhm (Mon, 02 Jul 2018 18:26:40 GMT):
@slr This is dependent on your use case but usually I'd highly recommend starting there (Indy-SDK Getting Started Guide) and then go from there. Any more questions you might have are more than welcome here and in #indy-sdk :slight_smile:

ThomasKrech (Tue, 03 Jul 2018 09:47:00 GMT):
Has joined the channel.

VitalijReicherdt (Tue, 03 Jul 2018 09:47:33 GMT):
Has joined the channel.

tripleOrange (Tue, 03 Jul 2018 17:33:51 GMT):
Has joined the channel.

tripleOrange (Tue, 03 Jul 2018 17:51:55 GMT):
Just joining - and we are only now considering Indy in our application - also based on fabric. Am I correct in understanding that Indy potentially helps us address gdpr compliance? - ie. no central authority or entity owns the individuals data so a user could potentially burn their own data account?

sejalsoftware (Wed, 04 Jul 2018 01:28:36 GMT):
Has joined the channel.

Marcus1 (Wed, 04 Jul 2018 06:46:31 GMT):
Has joined the channel.

pradeeppadmarajaiah (Wed, 04 Jul 2018 06:59:55 GMT):
Has joined the channel.

pradeeppadmarajaiah (Wed, 04 Jul 2018 07:02:57 GMT):
Hi Team, I have started exploring an Indy . Need your help, to get hello-world type (create a sample identifier in this case) example using Java which includes setup of Indy nodes to running java sample. I am confused, where to start now

arunwij (Wed, 04 Jul 2018 08:58:43 GMT):
Hello everyone, I will highly appreciate if anyone can clarify this for me, Think about a scenario a person(entity A) needs to connect to an existing self-sovereign identity network. What should be that person do first? Should they contact Trust anchor(entity B)? How they(B) add that person(A) into the network?

arunwij (Wed, 04 Jul 2018 08:58:43 GMT):
Hello everyone, I will highly appreciate if anyone can clarify this for me, Think about a scenario a person(entity A) needs to connect to an existing self-sovereign identity network. What should be that person do first? Should they contact Trust anchor(entity B )? How they(B) add that person(A) into the network?

arunwij (Wed, 04 Jul 2018 08:58:43 GMT):
Hello everyone, I will highly appreciate if anyone can clarify this for me, Think about a scenario a person(entity A) needs to connect to an existing self-sovereign identity network. What should be that person do first? Should that person contact Trust anchor(entity B )? How they(B) add that person(A) into the network?

arunwij (Thu, 05 Jul 2018 06:12:06 GMT):
Is *Public key* and *Verkey*, Same or different things?

arunwij (Thu, 05 Jul 2018 06:12:06 GMT):
Is *Public key* and *Verkey*, Same or different things in indy?

vijays817 (Thu, 05 Jul 2018 07:44:07 GMT):
Has joined the channel.

LuigiRiva (Thu, 05 Jul 2018 10:52:36 GMT):
Hi! The Sovrin whitepaper speaks a lot about the incentive model of the network based on a Sovrin token. Could anyone provide me more technical information about that? Is the possibility to use a token in the Sovrin network already available?

nage (Thu, 05 Jul 2018 14:19:30 GMT):
@LuigiRiva any Sovrin Token discussion is better suited for Sovrin slack, as that is an deployment concern for the Sovrin network specifically, not Indy in general

nage (Thu, 05 Jul 2018 14:20:32 GMT):
The new Indy-Node and Indy-SDK releases include some interface support for allowing payment as part of interactions with an Indy ledger, but no specific implementation. Again networks like Sovrin are figuring out how to implement those concepts in their own plugins and configurations

nage (Thu, 05 Jul 2018 14:21:00 GMT):
The Indy WG call is coming up in 40 minutes. The Agenda can be found here https://docs.google.com/presentation/d/1cNaIZz-iBKH9PrKSn3yVgfHXIT881BeRQePgqi_O1VE

nage (Thu, 05 Jul 2018 14:46:22 GMT):
https://zoom.us/j/232861185

herveblanc (Thu, 05 Jul 2018 15:57:17 GMT):
Has joined the channel.

dbluhm (Thu, 05 Jul 2018 19:08:28 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=SQLarsLkmQdYEKn5K) @arunwij I'm not certain but from what I understand a "verkey" or verification key is used basically the same way that a Public Key is used but the `verkey` has two parts to it: a key used to encrypt messages and a key used to verify signed messages. @nage or others, feel free to correct me if I'm wrong.

drummondreed (Fri, 06 Jul 2018 04:53:31 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=nPhtYq4tZgbCWbGqP) @tripleOrange The short answer is Yes. I highly recommend this series of Medium posts on that exact question: https://medium.com/evernym/is-self-sovereign-identity-ssi-the-ultimate-gdpr-compliance-tool-9d8110752f89

axiomdev (Fri, 06 Jul 2018 18:53:30 GMT):
Has joined the channel.

jmason900 (Fri, 06 Jul 2018 19:24:08 GMT):
Has joined the channel.

vtech (Wed, 11 Jul 2018 08:32:00 GMT):
Has joined the channel.

ghouston10 (Wed, 11 Jul 2018 10:02:43 GMT):
Has joined the channel.

ghouston10 (Wed, 11 Jul 2018 10:02:54 GMT):
Hi, I am following the indy agent nodejs quick start guide, found here, https://github.com/hyperledger/indy-agent/blob/master/nodejs/quick-start-guide.md When I run ./manage cli within the von network I recieve the following error: libpbc.so.1: cannot open shared object file: No such file or directory

robotfactory (Wed, 11 Jul 2018 14:39:35 GMT):
Has joined the channel.

laurasp (Wed, 11 Jul 2018 15:29:46 GMT):
Has joined the channel.

swcurran (Wed, 11 Jul 2018 22:16:06 GMT):
@ghouston10 - looking into it now. I've got the same thing happening.

nikhil.kumar (Thu, 12 Jul 2018 05:36:54 GMT):
Has joined the channel.

laurasp (Thu, 12 Jul 2018 07:42:04 GMT):
Hi everyone. I just started to look into Indy and I was wondering if is there any documentation related "Privacy by Design" ..........In regards to what is written in the Indy project page. " …….In particular, Privacy by Design and privacy-preserving technologies are critically important for a public identity ledger where correlation can take place……..” does exist a guideline or something similar? Hi everyone. I just started to look into Indy and I was wondering if is there any documentation related "Privacy by Design" ..........In regards to what is written in the Indy project page. " …….In particular, Privacy by Design and privacy-preserving technologies are critically important for a public identity ledger where correlation can take place……..” does exist a guideline or something similar? Hi everyone. I just started to look into Indy and I was wondering if is there any documentation related "Privacy by Design" ..........In regards to what is written in the Indy project page. " …….In particular, Privacy by Design and privacy-preserving technologies are critically important for a public identity ledger where correlation can take place……..” does exist a guideline or something similar where I can find the concerns related to privacy? Many thanks! Hi everyone. I just started to look into Indy and I was wondering if is there any documentation related "Privacy by Design" ..........In regards to what is written in the Indy project page. " …….In particular, Privacy by Design and privacy-preserving technologies are critically important for a public identity ledger where correlation can take place……..” does exist a guideline or something similar where I can find the concerns related to privacy? Many thanks! Hi everyone. I just started to look into Indy and I was wondering if is there any documentation related "Privacy by Design" ..........In regards to what is written in the Indy project page. " …….In particular, Privacy by Design and privacy-preserving technologies are critically important for a public identity ledger where correlation can take place……..” does exist a guideline or something similar where I can find the concerns related to privacy? Many thanks!Hi everyone. I just started to look into Indy and I was wondering if is there any documentation related "Privacy by Design" ..........In regards to what is written in the Indy project page. " …….In particular, Privacy by Design and privacy-preserving technologies are critically important for a public identity ledger where correlation can take place……..” does exist a guideline or something similar where I can find the concerns related to privacy? Many thanks! Hi everyone. I just started to look into Indy and I was wondering if is there any documentation related "Privacy by Design" ..........In regards to what is written in the Indy project page. " …….In particular, Privacy by Design and privacy-preserving technologies are critically important for a public identity ledger where correlation can take place……..” does exist a guideline or something similar where I can find the concerns related to privacy? Many thanks! Hi everyone. I just started to look into Indy and I was wondering if is there any documentation related "Privacy by Design" ..........In regards to what is written in the Indy project page. " …….In particular, Privacy by Design and privacy-preserving technologies are critically important for a public identity ledger where correlation can take place……..” does exist a guideline or something similar where I can find the concerns related to privacy? Many thanks! Hi everyone. I just started to look into Indy and I was wondering if is there any documentation related "Privacy by Design" ..........In regards to what is written in the Indy project page. " …….In particular, Privacy by Design and privacy-preserving technologies are critically important for a public identity ledger where correlation can take place……..” does exist a guideline or something similar where I can find the concerns related to privacy? Many thanks!

laurasp (Thu, 12 Jul 2018 07:42:42 GMT):
Hi everyone. I just started to look into Indy and I was wondering if is there any documentation related "Privacy by Design" ..........In regards to what is written in the Indy project page. " …….In particular, Privacy by Design and privacy-preserving technologies are critically important for a public identity ledger where correlation can take place……..” does exist a guideline or something similar where I can find the concerns related to privacy? Many thanks!

ghouston10 (Thu, 12 Jul 2018 08:59:10 GMT):
@swcurran ok, let me know if you were able to resolve it as I have spent a few days on it and have got no progress.

danielhardman (Thu, 12 Jul 2018 09:39:13 GMT):
Indy HIPE Status Notes 1. The HIPE on credential revocation has finished its Final Comment Period and been accepted unanimously by Indy maintainers. It was merged and is now known as Indy HIPE 0011. You can view it at https://github.com/hyperledger/indy-hipe/tree/master/text/0011-cred-revocation. 2. Per a decision of Indy maintainers on Monday, July 9, the HIPE on concurrency improvements in Indy is now entering its Final Comment Period. We believe it is ready to be accepted. You can view the PR for this HIPE at https://github.com/hyperledger/indy-hipe/pull/16. The HIPE will be accepted and merged on Monday, July 23 unless there is significant latebreaking dissonance from community members.

MHeisoku (Thu, 12 Jul 2018 11:44:36 GMT):
Has joined the channel.

MHeisoku (Thu, 12 Jul 2018 11:48:29 GMT):
Hi All. Is there an example implementation of an Indy or Sovrin wallet yet I could look at? I've read through the DKMS architecture and it seems like there's enough there to build one, just wanted to see if anyone had actually built one yet

leolustig (Thu, 12 Jul 2018 14:16:49 GMT):
Has joined the channel.

swcurran (Thu, 12 Jul 2018 14:44:20 GMT):
@MHeisoku - the term "wallet" in Indy is used for a data store of an agent. There is a defined API between the Indy-SDK and an implementation of the wallet, and there is a reference implementation in the Indy-SDK repo. The best place to get an overview of the Indy wallet design is in an in-process HIPE ("Hyperledger Indy Improvement Enhancements") - https://github.com/dhh1128/indy-hipe/blob/616e630ee4f939d1b86abc96c2548352e80cc7e9/text/wallets/README.md With Indy there are also Agents - implementations of the use of the Indy protocols to wrap an Wallet. The best starting place to look at those is probably the indy-agent repo - https://github.com/hyperledger/indy-agent.

mjmckean (Thu, 12 Jul 2018 15:23:15 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=m83DqaHrskMpegCCr) @laurasp @laurasp have you looked through the [Sovrin White Paper](https://sovrin.org/wp-content/uploads/Sovrin-Protocol-and-Token-White-Paper.pdf)?

mjmckean (Thu, 12 Jul 2018 15:23:15 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=m83DqaHrskMpegCCr) @laurasp have you looked through the [Sovrin White Paper](https://sovrin.org/wp-content/uploads/Sovrin-Protocol-and-Token-White-Paper.pdf)?

swcurran (Thu, 12 Jul 2018 16:31:41 GMT):
@laurasp - you can start from the Wikipedia page, which links to the foundational documents on Privacy By Design - https://en.wikipedia.org/wiki/Privacy_by_design

ksista (Thu, 12 Jul 2018 16:32:18 GMT):
Has joined the channel.

stanleyc-trustscience (Thu, 12 Jul 2018 16:43:19 GMT):
Hello again! @swcurran , I am wondering how to use indy-sdk to connect to indy network that is not hosted locally. What config do I do?

iGana (Thu, 12 Jul 2018 16:54:19 GMT):
Has joined the channel.

swcurran (Thu, 12 Jul 2018 16:59:42 GMT):
We're just starting to play with that, but I think the process is the same as local. You need a genesis file from the network you want to talk to. For example, the Sovrin global networks - their Provisional and Test networks - there are genesis files that you can download from somewhere (I haven't looked where but I know others have found it) and use them in place of your local network genesis file. Your client spins up and uses the new genesis file to find the nodes. Of course - the trick might be that you need to have a network connection to those nodes. In our world, we have port restrictions at the Enterrpise Firewall level that we are having to figure out. Should be easy from home though :-)

swcurran (Thu, 12 Jul 2018 16:59:42 GMT):
We're just starting to play with that, but I think the process is the same as local. You need a genesis file from the network you want to talk to. For example, the Sovrin global networks - their Provisional and Test networks - there are genesis files that you can download from somewhere (I haven't looked where but I know others have found them) and use them in place of your local network genesis file. Your client spins up and uses the new genesis file to find the nodes. Of course - the trick might be that you need to have a network connection to those nodes. In our world, we have port restrictions at the Enterrpise Firewall level that we are having to figure out. Should be easy from home though :-)

swcurran (Thu, 12 Jul 2018 17:01:37 GMT):
For example - we (BC Gov) have our own test Indy network at http://159.89.115.24/, and if you go there, we have a web page that includes a link to download the genesis file. You might want to use that to see if you can connect your client to that network.

jacobsaur (Thu, 12 Jul 2018 19:59:08 GMT):
Has joined the channel.

jacobsaur (Thu, 12 Jul 2018 20:21:05 GMT):
Hi, I'm interested in using Indy to build a decentralized credit registry. Each client and institution can have a DID for that relationship, and each credit attestation can be a verifiable claim (eg took out a loan for $100 on Jan 1, paid back $5 on Feb 2, etc). What's different about credit claims vs other claims is when the client goes to show these claims to a verifier (eg a credit bureau or other bank) they can't just present their good claims and leave out the bad ones (eg defaults), the verifier needs proof that they are seeing all claims issued for that relationship. One way to solve this to calculate the merkle root of all the claims and store it on the blockchain. This merkle root will then get updated as each new claim is issued (a similar solution described by W3C https://web-payments.org/specs/source/pop2016/). The verifier can then build up a merkle tree from all the claims submitted and if the merkle root doesn't match the one on the blockchain then they will know the subject left some out. Is this a good approach to the problem? Theoretically this per-DID merkle root could be stored on any blockchain but it would be easiest if it was stored right on the Indy blockchain. Is there a way to do this now? If not what would I need to change to implement it? The ideal solution would be if this all happened automatically, ie if a client & issuer ever established DID relationship of type 'merkle-claims' (or something) then when any claim was issued it would automatically update Indy's blockchain key-value store (where the key would be the DID and the value the merkle root of all claims issued).

mpet (Fri, 13 Jul 2018 01:25:48 GMT):
Has joined the channel.

charlieparker (Fri, 13 Jul 2018 05:33:19 GMT):
Has joined the channel.

charlieparker (Fri, 13 Jul 2018 05:34:09 GMT):
Has anyone looked at the Kriptan Network? What do people think of it?

sammitchell (Fri, 13 Jul 2018 07:43:40 GMT):
Has joined the channel.

ehsanhajian (Fri, 13 Jul 2018 09:29:13 GMT):
Has joined the channel.

VuiLenDi (Fri, 13 Jul 2018 10:33:28 GMT):
Has joined the channel.

GarryKelly (Fri, 13 Jul 2018 12:56:21 GMT):
Has joined the channel.

tomislav (Fri, 13 Jul 2018 13:25:58 GMT):
I read about Kriptan sometime ago. I wasn't able to find more technical details of the platform behind it, might be still early. Would love to get learn more about it. Any github repos you know where they publish work?

kendra.bittner (Fri, 13 Jul 2018 16:37:40 GMT):
Has joined the channel.

chandrakanthMamillapalli (Fri, 13 Jul 2018 18:29:03 GMT):
Has joined the channel.

charlieparker (Fri, 13 Jul 2018 20:23:02 GMT):
I'm not aware of any Github repos yet. The whitepaper is interesting though. I liked the architecture for anonymizing transactions and also thought their approach to ZKP had potential.

kdenhartog (Fri, 13 Jul 2018 20:25:00 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=GDZ6dDbeRHj2oqutG) @jacobsaur @jacobsaur There's a few things in this that make me think it's not going to work in the way you are thinking. First, We don't have a notion of a hashed linked list that maintains a global order of all transactions like a blockchain typically would. Rather, we're more like a cryptographically verified distributed database that uses a consensus protocol to decide what is added to the db. This is a very high level view of Indy though, so I'd suggest taking a look at the documentation to get a better understanding of how the ledger works. That can be found in the docs directory of hyperledger/indy-node and hyperledger/indy-plenum. Second, because we do not store any data on the ledger, I don't think your idea of storing a merkle roots on the ledger would work well in this case. Rather, claims would be stored in agents and if anyone wanted a record they would need to ask the payer for a copy of the claims and they could generate a proof to show that they've made the payments based upon the claims that the institution issued to them as a receipt of payment. Hopefully that helps a little in understanding better what you had in mind and feel free to ask more questions if you need. Please @ me though so I can get back to you with a response.

kdenhartog (Fri, 13 Jul 2018 20:25:00 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=GDZ6dDbeRHj2oqutG) @jacobsaur There's a few things in this that make me think it's not going to work in the way you are thinking. First, We don't have a notion of a hashed linked list that maintains a global order of all transactions like a blockchain typically would. Rather, we're more like a cryptographically verified distributed database that uses a consensus protocol to decide what is added to the db. This is a very high level view of Indy though, so I'd suggest taking a look at the documentation to get a better understanding of how the ledger works. That can be found in the docs directory of hyperledger/indy-node and hyperledger/indy-plenum. Second, because we do not store any data on the ledger, I don't think your idea of storing a merkle roots on the ledger would work well in this case. Rather, claims would be stored in agents and if anyone wanted a record they would need to ask the payer for a copy of the claims and they could generate a proof to show that they've made the payments based upon the claims that the institution issued to them as a receipt of payment. Hopefully that helps a little in understanding better what you had in mind and feel free to ask more questions if you need. Please @ me though so I can get back to you with a response.

baconsandwich (Fri, 13 Jul 2018 21:01:12 GMT):
Has joined the channel.

jacobsaur (Sat, 14 Jul 2018 00:32:33 GMT):
@kdenhartog thanks for your response! Just to check my understanding, I thought that Indy did store some things on a public ledger, such as DIDs and credential schemas. For my use case I would just need a public consensus driven key-value store which it seems like plenum supports at least under the hood (https://github.com/hyperledger/indy-plenum/blob/master/storage/kv_store_leveldb.py). One other thought I had was that the value for the merkle root could be stored as DID metadata (using sdk function setDidMetadata) since there should only ever be one merkle root for all claims associated with a DID. But I can't tell whether this would only update DID metadata in the users wallet or whether it would update it on the public ledger.

MHeisoku (Sat, 14 Jul 2018 02:11:04 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=wxm4cNWojkQn723tw) @swcurran Thank you. Those links were helpful

kdenhartog (Sat, 14 Jul 2018 05:58:57 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=ooWbeeTrzGmc5i7td) @jacobsaur That is correct, DIDs (and associated DID-Doc data stored as attribs), Schemas, Cred_defs, credential revocation registries, and I believe AuthZ device policies will all exist on ledger. One of the main reasons. Take note, no actual data that is signed in claims, or even hashes end up on the ledger. This is important to the data model because this means if an encryption scheme or a hash scheme we're ever able to be broken, all of this data could be cracked later on. So, when I read what you describe, I see it as data that shouldn't go on ledger, but rather should be maintained between the client and the institution. There's additional benefits to this as well beyond security. In this model where data is stored off ledger and signed with keys on ledger the scalability of the system becomes much easier. The core question I ask myself when considering what should go on ledger is "Does everyone need access to this data by default, or can I get by having only the involved parties and those who they choose to share that data with storing the data?"

kdenhartog (Sat, 14 Jul 2018 05:58:57 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=ooWbeeTrzGmc5i7td) @jacobsaur That is correct, DIDs (and associated DID-Doc data stored as attribs), Schemas, Cred_defs, credential revocation registries, and I believe AuthZ device policies will all exist on ledger. Take note, no actual data that is signed in claims, or even hashes end up on the ledger. This is important to the data model because this means if an encryption scheme or a hash scheme we're ever able to be broken, and this data was stored on ledger, all of this data could be cracked later on. So, when I read what you describe, I see it as data that shouldn't go on ledger, but rather should be maintained between the client and the institution. There's additional benefits to this as well beyond security. In this model where data is stored off ledger and signed with keys on ledger the scalability of the system becomes much easier. The core question I ask myself when considering what should go on ledger is "Does everyone need access to this data by default, or can I get by having only the involved parties and those who they choose to share that data with storing the data?". However, I do believe you could make a workaround to make your use case fit using Indy, but I wouldn't suggest it as you're introducing assumptions that we haven't made and I think it will become very messy quickly. If having the data exist on a ledger is absolutely necessary, I'd suggest looking at finding a way to use Indy as an identity layer with a Fabric ledger rather than trying to make Indy work in a way it wasn't intended.

stanleyc-trustscience (Sat, 14 Jul 2018 07:20:27 GMT):
I have a general question about the Alice example. Hope anyone who's interested can chime in ... First, we have government creating the transcript schema ``` await anoncreds.issuer_create_schema(government_did, 'Transcript', '1.2', json.dumps(['first_name', 'last_name', 'degree', 'status', 'year', 'average', 'ssn'])) await send_schema(pool_handle, government_wallet, government_did, transcript_schema) ``` Then, we have Fabre college sending credential definition to the ledger ``` (_, transcript_schema) = await get_schema(pool_handle, faber_did, transcript_schema_id) (faber_transcript_cred_def_id, faber_transcript_cred_def_json) = \ await anoncreds.issuer_create_and_store_credential_def(faber_wallet, faber_did, transcript_schema, 'TAG1', 'CL', '{"support_revocation": false}') await send_cred_def(pool_handle, faber_wallet, faber_did, faber_transcript_cred_def_json) ``` Then, we have Acme requesting proof of transcript that is issued by faber ``` job_application_proof_request_json = json.dumps({ 'nonce': '1432422343242122312411212', 'name': 'Job-Application', ... 'attr3_referent': { 'name': 'degree', 'restrictions': [{'cred_def_id': faber_transcript_cred_def_id}] }, .... The question here is what if Alice didn't go to faber college, but instead went to Harvard? In the real-world UI where Alice interacts with Acme college, it seems like there would be a dropdown of supported colleges, and the credential definition id on the proof request json would set to whichever college that Alice picked ```

stanleyc-trustscience (Sat, 14 Jul 2018 07:20:27 GMT):
I have a general question about the Alice example. Hope anyone who's interested can chime in ... First, we have government creating the transcript schema ``` await anoncreds.issuer_create_schema(government_did, 'Transcript', '1.2', json.dumps(['first_name', 'last_name', 'degree', 'status', 'year', 'average', 'ssn'])) await send_schema(pool_handle, government_wallet, government_did, transcript_schema) ``` Then, we have Fabre college sending credential definition to the ledger ``` (_, transcript_schema) = await get_schema(pool_handle, faber_did, transcript_schema_id) (faber_transcript_cred_def_id, faber_transcript_cred_def_json) = \ await anoncreds.issuer_create_and_store_credential_def(faber_wallet, faber_did, transcript_schema, 'TAG1', 'CL', '{"support_revocation": false}') await send_cred_def(pool_handle, faber_wallet, faber_did, faber_transcript_cred_def_json) ``` Then, we have Acme requesting proof of transcript that is issued by faber ``` job_application_proof_request_json = json.dumps({ 'nonce': '1432422343242122312411212', 'name': 'Job-Application', ... 'attr3_referent': { 'name': 'degree', 'restrictions': [{'cred_def_id': faber_transcript_cred_def_id}] }, ``` The question here is what if Alice didn't go to faber college, but instead went to Harvard? In the real-world UI where Alice interacts with Acme Corp, it seems like there would be a dropdown of supported colleges, and the credential definition id on the proof request json would set to whichever college that Alice picked

stanleyc-trustscience (Sat, 14 Jul 2018 07:20:27 GMT):
I have a general question about the Alice example. Hope anyone who's interested can chime in ... First, we have government creating the transcript schema ``` await anoncreds.issuer_create_schema(government_did, 'Transcript', '1.2', json.dumps(['first_name', 'last_name', 'degree', 'status', 'year', 'average', 'ssn'])) await send_schema(pool_handle, government_wallet, government_did, transcript_schema) ``` Then, we have Fabre college sending credential definition to the ledger ``` (_, transcript_schema) = await get_schema(pool_handle, faber_did, transcript_schema_id) (faber_transcript_cred_def_id, faber_transcript_cred_def_json) = \ await anoncreds.issuer_create_and_store_credential_def(faber_wallet, faber_did, transcript_schema, 'TAG1', 'CL', '{"support_revocation": false}') await send_cred_def(pool_handle, faber_wallet, faber_did, faber_transcript_cred_def_json) ``` Then, we have Acme requesting proof of transcript that is issued by faber ``` job_application_proof_request_json = json.dumps({ 'nonce': '1432422343242122312411212', 'name': 'Job-Application', ... 'attr3_referent': { 'name': 'degree', 'restrictions': [{'cred_def_id': faber_transcript_cred_def_id}] }, .... The question here is what if Alice didn't go to faber college, but instead went to Harvard? In the real-world UI where Alice interacts with Acme college, it seems like there would be a dropdown of supported colleges, and the credential definition id on the proof request json would set to whichever college that Alice picked

stanleyc-trustscience (Sat, 14 Jul 2018 07:20:27 GMT):
I have a general question about the Alice example. Hope anyone who's interested can chime in ... First, we have government creating the transcript schema ``` await anoncreds.issuer_create_schema(government_did, 'Transcript', '1.2', json.dumps(['first_name', 'last_name', 'degree', 'status', 'year', 'average', 'ssn'])) await send_schema(pool_handle, government_wallet, government_did, transcript_schema) ``` Then, we have Fabre college sending credential definition to the ledger ``` (_, transcript_schema) = await get_schema(pool_handle, faber_did, transcript_schema_id) (faber_transcript_cred_def_id, faber_transcript_cred_def_json) = \ await anoncreds.issuer_create_and_store_credential_def(faber_wallet, faber_did, transcript_schema, 'TAG1', 'CL', '{"support_revocation": false}') await send_cred_def(pool_handle, faber_wallet, faber_did, faber_transcript_cred_def_json) ``` Then, we have Acme requesting proof of transcript that is issued by faber ``` job_application_proof_request_json = json.dumps({ 'nonce': '1432422343242122312411212', 'name': 'Job-Application', ... 'attr3_referent': { 'name': 'degree', 'restrictions': [{'cred_def_id': faber_transcript_cred_def_id}] }, ... The question here is what if Alice didn't go to faber college, but instead went to Harvard? In the real-world UI where Alice interacts with Acme college, it seems like there would be a dropdown of supported colleges, and the credential definition id on the proof request json would set to whichever college that Alice picked

stanleyc-trustscience (Sat, 14 Jul 2018 07:20:27 GMT):
I have a general question about the Alice example. Hope anyone who's interested can chime in ... First, we have government creating the transcript schema ``` await anoncreds.issuer_create_schema(government_did, 'Transcript', '1.2', json.dumps(['first_name', 'last_name', 'degree', 'status', 'year', 'average', 'ssn'])) await send_schema(pool_handle, government_wallet, government_did, transcript_schema) ``` Then, we have Fabre college sending credential definition to the ledger ``` (_, transcript_schema) = await get_schema(pool_handle, faber_did, transcript_schema_id) (faber_transcript_cred_def_id, faber_transcript_cred_def_json) = \ await anoncreds.issuer_create_and_store_credential_def(faber_wallet, faber_did, transcript_schema, 'TAG1', 'CL', '{"support_revocation": false}') await send_cred_def(pool_handle, faber_wallet, faber_did, faber_transcript_cred_def_json) ``` Then, we have Acme requesting proof of transcript that is issued by faber ``` job_application_proof_request_json = json.dumps({ 'nonce': '1432422343242122312411212', 'name': 'Job-Application', ... 'attr3_referent': { 'name': 'degree', 'restrictions': [{'cred_def_id': faber_transcript_cred_def_id}] }, ``` The question here is what if Alice didn't go to faber college, but instead went to Harvard? In the real-world UI where Alice interacts with Acme college, it seems like there would be a dropdown of supported colleges, and the credential definition id on the proof request json would set to whichever college that Alice picked

stanleyc-trustscience (Sat, 14 Jul 2018 07:23:36 GMT):
This dropdown seems like a problem if the list is huge

stanleyc-trustscience (Sat, 14 Jul 2018 07:28:11 GMT):
Like Acme Corp, would need to know about all the colleges that it supports ... or the Thirft loan needs to know about all the credential def ids from all the companies that it supports

stanleyc-trustscience (Sat, 14 Jul 2018 07:28:11 GMT):
Like Acme Corp, would need to know about all the credential def if from the colleges that it supports ... or Thirft loan needs to know about all the credential def ids from all the companies that it supports

stanleyc-trustscience (Sat, 14 Jul 2018 07:28:11 GMT):
Like Acme Corp, would need to know about all the credential def if from the colleges that it supports ... or the Thirft loan needs to know about all the credential def ids from all the companies that it supports

stanleyc-trustscience (Sat, 14 Jul 2018 07:28:11 GMT):
Like Acme Corp, would need to know all the colleges that it supports ... or the Thirft loan needs to know about all the credential def ids from all the companies that it supports

swcurran (Sat, 14 Jul 2018 14:04:09 GMT):
@stanleyc-trustscience - good questions and the answers will evolve. First, Acme Corp might process any credential from a well-known schema, or set of schema, so that can reduce the variability of the data being received. schema.org can be a source for schema, and the approach that Sovrin/Indy supports is that the market will drive the schema used for purposes. Someone will post a good one, others will use it, it will become a defacto (or actual) standard. On the question of many orgs - you might want to look at a paper we did about how a business might build up a set of trusted sources of data. At worst, you'll get the concepts. Paper is here: Traversing the Web of Trust https://docs.google.com/document/d/1nYq0iakgtyC21oUGWa5hLuJUoKeJFpURtGz6HcLIltY/edit#heading=h.b8ud3xi74tdl

swcurran (Sat, 14 Jul 2018 14:04:09 GMT):
@stanleyc-trustscience - good questions and the answers will evolve. First, Acme Corp might process any credential from a well-known schema, or set of schema, so that can reduce the variability of the data being received. schema.org can be a source for schema, and the approach that Sovrin/Indy supports is that the market will drive the schema used for purposes. Someone will post a good one, others will use it, it will become a defacto (or actual) standard. On the question of many orgs - you might want to look at a paper we did about how a business might build up a set of trusted sources of data. At worst, you'll get the concepts. Paper is here: https://docs.google.com/document/d/1nYq0iakgtyC21oUGWa5hLuJUoKeJFpURtGz6HcLIltY/edit#heading=h.b8ud3xi74tdl

stanleyc-trustscience (Sat, 14 Jul 2018 18:07:21 GMT):
@swcurran The thing is on proof_request_json, it accepts a cred_def_id, not a schema_id. And cred definition is published by individual orgs and its cred can only be issued by the org which published the cred definition.

stanleyc-trustscience (Sat, 14 Jul 2018 18:35:53 GMT):
How is possible to configure such that `Acme Corp process any credential from a well-known schema, or set of schema, ` without Acme Corp knowing the exhaustive list of cred_def_id's?

stanleyc-trustscience (Sat, 14 Jul 2018 18:35:53 GMT):
How is possible to configure such that `Acme Corp process any credential from a well-known schema, or set of schema, ` without Acme Corp knowing the exhaustive list of cred_def_id's from a well-known schema?

stanleyc-trustscience (Sat, 14 Jul 2018 18:35:53 GMT):
How is possible to configure such that `Acme Corp process any credential from a well-known schema, or set of schema, ` without Acme Corp knowing the exhaustive list of cred_def_id's from the well-known schema/set of schema?

stanleyc-trustscience (Sat, 14 Jul 2018 18:57:53 GMT):
Seems like according to your paper, the process is reversed. Instead of verifying producing the proof_request_json first, the end user would first submit proof, and the verifier then determines whether to trust the issuer of the proof. This is currently not supported in indy?

stanleyc-trustscience (Sat, 14 Jul 2018 18:57:53 GMT):
Seems like according to your paper, the process is reversed. Instead of verifier specifying the issuer first (via `'restrictions': [{'cred_def_id': faber_transcript_cred_def_id}]`), the end user would first submit proof, and the verifier then determines whether to trust the issuer of the proof. This is currently not supported in indy?

stanleyc-trustscience (Sat, 14 Jul 2018 18:57:53 GMT):
Seems like according to your paper, the process is reversed. Instead of verifier producing the proof_request_json first, the end user would first submit proof, and the verifier then determines whether to trust the issuer of the proof. This is currently not supported in indy?

swcurran (Sat, 14 Jul 2018 20:45:00 GMT):
The proof request JSON definitely supports schema_id(s) as either the ID from the ledger or the tuple {issuer, schema_name, version) - you should check that again.

swcurran (Sat, 14 Jul 2018 20:45:00 GMT):
The proof request JSON definitely supports schema_id(s) as either the ID from the ledger or the tuple {issuer, schema_name, version) - you should check that again. Might be you are just looking at one example, vs. the general format.

swcurran (Sat, 14 Jul 2018 20:51:12 GMT):
On the latter - when you receive a proof, you receive (via the CredDef) a DID for the Issuer of the credential from which the claim is from. It is "an exercise left to the user" to figure out how to chase down who that Issuer is, and whether or not you trust them. That is really the core of Verifiable Credentials - when you get a proof, you know for each claim: who issued it, that it was issued to the Prover, that it was not tampered with, and that it was not revoked. What you need to decide is - do I trust the issuer? That is not so much a technical question as a social one. When I ask a person to "proof you live at the address you say" and they give you a Hydro Bill - you have to decide - do I trust that Hydro Bill? In the paper case, it's usually - do I trust it's not a faked bill. With VerfCreds, you know it's not forged, but you have to be sure it's a really issued by the Hydro company.

swcurran (Sat, 14 Jul 2018 20:51:12 GMT):
On the latter - when you receive a proof, you receive (via the CredDef) a DID for the Issuer of the credential from which the claim is from. It is "an exercise left to the user" to figure out how to chase down who that Issuer is, and whether or not you trust them. That is really the core of Verifiable Credentials - when you get a proof, you know for each claim: who issued it, that it was issued to the Prover, that it was not tampered with, and that it was not revoked. What you need to decide is - do I trust the issuer? That is not so much a technical question as a social one. When I ask a person to "proof you live at the address you say" and they give me a Hydro Bill - I have to decide - do I trust that Hydro Bill? In the paper case, it's usually - do I trust it's not a faked bill? With VerfCreds, you know it's not forged and that it was issued to person providing it, but you have to be sure it's a really issued by the Hydro company.

stanleyc-trustscience (Sat, 14 Jul 2018 21:23:03 GMT):
@swcurran Where can I find the general format of proof request json?

tomislav (Sat, 14 Jul 2018 22:57:11 GMT):
@stanleyc-trustscience https://github.com/hyperledger/indy-sdk/blob/4259a6d938f5a606f699ef7bcc84c1860d3611c3/libindy/src/api/anoncreds.rs#L857

stanleyc-trustscience (Sat, 14 Jul 2018 23:51:08 GMT):
@tomislav thanks ``` /// "restrictions": Optional<[]> // see below, /// // if specified, credential must satisfy to one of the given restriction. ``` Looks like we can have multiple restrictions. This would work? `'restrictions': [{'cred_def_id': faber_transcript_cred_def_id}, {'cred_def_id': harvard_transcript_cred_def_id}]

PeterX (Sun, 15 Jul 2018 07:45:53 GMT):
Has joined the channel.

DoctorX (Sun, 15 Jul 2018 08:37:17 GMT):
Has joined the channel.

tomislav (Sun, 15 Jul 2018 19:58:30 GMT):
Sounds about right, judging by the wording on the comment

RuWander (Mon, 16 Jul 2018 09:30:14 GMT):
Has joined the channel.

dmitry.anansky (Mon, 16 Jul 2018 11:05:20 GMT):
Has joined the channel.

dmitry.anansky (Mon, 16 Jul 2018 11:05:59 GMT):
Hi all, I was follow steps from https://github.com/hyperledger/indy-sdk/blob/master/doc/mac-build.md And got an issue with building the library: --- stdout cargo:rerun-if-env-changed=SODIUM_LIB_DIR cargo:rerun-if-env-changed=SODIUM_STATIC --- stderr thread 'main' panicked at 'called `Result::unwrap()` on an `Err` value: "`\"pkg-config\" \"--libs\" \"--cflags\" \"libsodium\"` did not exit successfully: signal: 6\n--- stderr\ndyld: Symbol not found: _cg_png_create_info_struct\n Referenced from: /System/Library/Frameworks/ImageIO.framework/Versions/A/ImageIO\n Expected in: /usr/local/lib/libPng.dylib\n in /System/Library/Frameworks/ImageIO.framework/Versions/A/ImageIO\n"', libcore/result.rs:945:5 Please let me know if someone had faced and overcome that issue

dmitry.anansky (Mon, 16 Jul 2018 11:05:59 GMT):
Hi all, I was following steps from https://github.com/hyperledger/indy-sdk/blob/master/doc/mac-build.md And got an issue with building the library: --- stdout cargo:rerun-if-env-changed=SODIUM_LIB_DIR cargo:rerun-if-env-changed=SODIUM_STATIC --- stderr thread 'main' panicked at 'called `Result::unwrap()` on an `Err` value: "`\"pkg-config\" \"--libs\" \"--cflags\" \"libsodium\"` did not exit successfully: signal: 6\n--- stderr\ndyld: Symbol not found: _cg_png_create_info_struct\n Referenced from: /System/Library/Frameworks/ImageIO.framework/Versions/A/ImageIO\n Expected in: /usr/local/lib/libPng.dylib\n in /System/Library/Frameworks/ImageIO.framework/Versions/A/ImageIO\n"', libcore/result.rs:945:5 Please let me know if someone had faced and overcome that issue

jacobsaur (Mon, 16 Jul 2018 17:35:34 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=tCRiEryfNR45jF2eL) @kdenhartog Thanks @kdenhartog . I think you're right that the best way forward might be to use Indy/Sovrin for identity, and Fabric or Sawtooth for the credit registry.

stefanopulzeic (Tue, 17 Jul 2018 07:47:48 GMT):
Has left the channel.

stanleyc-trustscience (Tue, 17 Jul 2018 17:17:02 GMT):
Hi guys, for the alice example, I have a few questions When Alice present her transcript credential to Acme corp, how does Acme corp know that the claim was issued to Alice? I notice while Alice was creating her proof request to Faber college, both `alice_faber_did` and `alice_master_secret_id` were included ``` # Alice agent (transcript_cred_request_json, transcript_cred_request_metadata_json) = \ await anoncreds.prover_create_credential_req(alice_wallet, alice_faber_did, transcript_cred_offer_json, faber_transcript_cred_def, alice_master_secret_id) ``` How were `alice_faber_did` and `alice_master_secret_id` used when Faber college creates her transcript crendential? ``` # Faber agent transcript_cred_json, _, _ = \ await anoncreds.issuer_create_credential(faber_wallet, transcript_cred_offer_json, transcript_cred_request_json, transcript_cred_values, None, None) ``` How does Acme corp verify that this transcript's credential belong to Alice? ``` # Acme Agent assert await anoncreds.verifier_verify_proof(job_application_proof_request_json, apply_job_proof_json, schemas_json, cred_defs_json, revoc_ref_defs_json, revoc_regs_json) ```

stanleyc-trustscience (Tue, 17 Jul 2018 17:17:02 GMT):
Hi guys, for the alice example, I have a few questions When Alice present her transcript credential to Acme corp, how does Acme corp know that the claim was issued to Alice? I notice while Alice was creating her proof request to Faber college, both `alice_faber_did` and `alice_master_secret_id` were included ``` (transcript_cred_request_json, transcript_cred_request_metadata_json) = \ await anoncreds.prover_create_credential_req(alice_wallet, alice_faber_did, transcript_cred_offer_json, faber_transcript_cred_def, alice_master_secret_id) ``` How were `alice_faber_did` and `alice_master_secret_id` used when Faber college creates her transcript crendential? ``` transcript_cred_json, _, _ = \ await anoncreds.issuer_create_credential(faber_wallet, transcript_cred_offer_json, transcript_cred_request_json, transcript_cred_values, None, None) ``` How does Acme corp verify that this transcript's credential belong to Alice?

stanleyc-trustscience (Tue, 17 Jul 2018 17:17:02 GMT):
Hi guys, for the alice example, I have a few questions When Alice present her transcript credential to Acme corp, how does Acme corp know that the claim was issued to Alice? I notice while Alice was creating her proof request to Faber college, both `alice_faber_did` and `alice_master_secret_id` were included ``` #Alice agent (transcript_cred_request_json, transcript_cred_request_metadata_json) = \ await anoncreds.prover_create_credential_req(alice_wallet, alice_faber_did, transcript_cred_offer_json, faber_transcript_cred_def, alice_master_secret_id) ``` How were `alice_faber_did` and `alice_master_secret_id` used when Faber college creates her transcript crendential? ``` #Faber agent transcript_cred_json, _, _ = \ await anoncreds.issuer_create_credential(faber_wallet, transcript_cred_offer_json, transcript_cred_request_json, transcript_cred_values, None, None) ``` How does Acme corp verify that this transcript's credential belong to Alice? ``` # Acme Agent assert await anoncreds.verifier_verify_proof(job_application_proof_request_json, apply_job_proof_json, schemas_json, cred_defs_json, revoc_ref_defs_json, revoc_regs_json) ```

stanleyc-trustscience (Tue, 17 Jul 2018 17:17:02 GMT):
Hi guys, for the alice example, I have a few questions When Alice present her transcript credential to Acme corp, how does Acme corp know that the claim was issued to Alice? I notice while Alice was creating her proof request to Faber college, both `alice_faber_did` and `alice_master_secret_id` were included ``` # Alice agent (transcript_cred_request_json, transcript_cred_request_metadata_json) = \ await anoncreds.prover_create_credential_req(alice_wallet, alice_faber_did, transcript_cred_offer_json, faber_transcript_cred_def, alice_master_secret_id) ``` How were `alice_faber_did` and `alice_master_secret_id` used when Faber college creates her transcript crendential? ``` # Faber agent transcript_cred_json, _, _ = \ await anoncreds.issuer_create_credential(faber_wallet, transcript_cred_offer_json, transcript_cred_request_json, transcript_cred_values, None, None) ``` How does Acme corp verify that this transcript's credential belong to Alice? ``` # Acme Agent assert await anoncreds.verifier_verify_proof(job_application_proof_request_json, apply_job_proof_json, schemas_json, cred_defs_json, revoc_ref_defs_json, revoc_regs_json) ```

ChrisMatichuk (Tue, 17 Jul 2018 17:25:46 GMT):
Has joined the channel.

CJP (Tue, 17 Jul 2018 17:43:34 GMT):
Has joined the channel.

swcurran (Tue, 17 Jul 2018 18:03:25 GMT):
@stanleyc-trustscience - the Link Secret (formerly known as Master Secret) is how Alice proves she was issued the credential. Basically, before issue the credential, Alice gives Faber some credential associated with her Link Secret - some crypto. That's put in the credential and in a proof, Alice's uses the Link Secret so to show she controls that crypto embedded in the credential. If you look up Pederson's Commitment. Not that you really need to know that :-)

sjain74 (Tue, 17 Jul 2018 19:10:53 GMT):
related question: looking at the examples in https://www.w3.org/TR/verifiable-claims-data-model/, is it true that a claim is also tied to a DID (or any other subject identifier)? And, if one can prove the ownership to the DID, can therefore prove the ownership to the claim too?

alanz (Tue, 17 Jul 2018 19:13:31 GMT):
Has joined the channel.

voddan (Tue, 17 Jul 2018 19:39:25 GMT):
Hi! Version 1.4 of the sdk started using something called `schemaId`, which is a structured string (see `Anoncreds.issuerCreateSchema` return value). Can you tell me how stable `schemaId` format is? Do you plan to use those ids to publicly identify schemes, or do you reserve a right to change their format in the future? What are the compatibility and migration guaranties from Indy? Can I share a `schemaId` with a user as-is, or are those internal ids and subgect to change? Thanks!

voddan (Tue, 17 Jul 2018 19:39:25 GMT):
Hi! Need help! Version 1.4 of the sdk started using something called `schemaId`, which is a structured string (see `Anoncreds.issuerCreateSchema` return value). Can you tell me how stable `schemaId` format is? Do you plan to use those ids to publicly identify schemes, or do you reserve a right to change their format in the future? What are the compatibility and migration guaranties from Indy? Can I share a `schemaId` with a user as-is, or are those internal ids and subgect to change? Thanks!

tomislav (Tue, 17 Jul 2018 19:53:04 GMT):
@voddan SchemaId format doesn't change over the course of credential issuance scenario. They seem to be pretty stable now, but format _may_ change in future. If that were to happen, it would be a breaking change, so systems will need to adopt. Currently, schemaId can be shared with a user as-is, so can credential definition ids

Raghav_rt (Tue, 17 Jul 2018 20:49:53 GMT):
Has joined the channel.

stanleyc-trustscience (Tue, 17 Jul 2018 23:17:57 GMT):
@swcurran Thanks :) but why is `alice_faber_did` included in ``` await anoncreds.prover_create_credential_req(alice_wallet, alice_faber_did, transcript_cred_offer_json, faber_transcript_cred_def, alice_master_secret_id ``` if all it needs is alice's master_secret_id

stanleyc-trustscience (Tue, 17 Jul 2018 23:17:57 GMT):
@swcurran Thanks but why is `alice_faber_did` included in `await anoncreds.prover_create_credential_req(alice_wallet, alice_faber_did, transcript_cred_offer_json, faber_transcript_cred_def, alice_master_secret_id)`, if all it needs is alice's master_secret_id

stanleyc-trustscience (Tue, 17 Jul 2018 23:17:57 GMT):
@swcurran Thanks :) but why is `alice_faber_did` included in `await anoncreds.prover_create_credential_req(alice_wallet, alice_faber_did, transcript_cred_offer_json, faber_transcript_cred_def, alice_master_secret_id)` if all it needs is alice's master_secret_id

stanleyc-trustscience (Tue, 17 Jul 2018 23:17:57 GMT):
@swcurran Thanks :) but why is `alice_faber_did` included in ``` await anoncreds.prover_create_credential_req(alice_wallet, alice_faber_did, transcript_cred_offer_json, faber_transcript_cred_def, alice_master_secret_id ``)` if all it needs is alice's master_secret_id

swcurran (Wed, 18 Jul 2018 03:34:27 GMT):
@stanleyc-trustscience - I'm not sure. Notice I didn't answer that one :-). It isn't used in the credential. Others might know (@nage?) As I think it through, I'm betting that the cred_req that Alice is building gets signed using the DID by which Faber knows Alice. That way, Faber can be sure that it is Alice requesting the Credential. The high level sequence is: * Alice <--> Faber connect - create pairwise DIDs * Faber gets proofs/data from Alice to link Alice to backend records they have about her - e.g. her transcripts, degrees, etc. * Faber offers Alice a Credential * Alice sends Faber a credential request, signing it so that Faber knows it's the same person that proved who they were. * Faber sends Alice a Credential By signing the actual request, Faber knows it's Alice, even if the Credential Request came in some other manner than a message. It's a theory... :-)

nabheetmadan (Wed, 18 Jul 2018 06:30:54 GMT):
Has joined the channel.

FlavioSS (Wed, 18 Jul 2018 10:51:00 GMT):
Has joined the channel.

Ribas (Wed, 18 Jul 2018 13:05:10 GMT):
Has joined the channel.

Arvindraja (Wed, 18 Jul 2018 14:51:42 GMT):
Has joined the channel.

dklesev (Wed, 18 Jul 2018 14:57:50 GMT):
Has joined the channel.

dklesev (Wed, 18 Jul 2018 15:03:33 GMT):
hi all is there something like explorer for indy?

shazzle (Wed, 18 Jul 2018 17:05:23 GMT):
Has joined the channel.

dileban (Thu, 19 Jul 2018 02:39:28 GMT):
Has joined the channel.

thPart (Thu, 19 Jul 2018 07:34:53 GMT):
Has joined the channel.

iwinoto (Thu, 19 Jul 2018 13:12:03 GMT):
Has joined the channel.

Sergey.Kupryushin (Thu, 19 Jul 2018 14:00:53 GMT):
Has joined the channel.

pandagopal (Fri, 20 Jul 2018 05:15:02 GMT):
Has joined the channel.

ajmeraharsh (Fri, 20 Jul 2018 11:48:48 GMT):
Has joined the channel.

rogermartins (Fri, 20 Jul 2018 12:44:45 GMT):
Has joined the channel.

canadaduane (Fri, 20 Jul 2018 17:03:59 GMT):
Has joined the channel.

stanleyc-trustscience (Fri, 20 Jul 2018 19:18:00 GMT):
@swcurran Thanks! ;)

stanleyc-trustscience (Fri, 20 Jul 2018 19:18:24 GMT):
I am setting up a server that runs a indy docker

stanleyc-trustscience (Fri, 20 Jul 2018 19:18:55 GMT):
How do I know what version libindy it is? I installed using ``` channel="master" # options: master, rc, stable sudo apt-key adv --keyserver keyserver.ubuntu.com --recv-keys 68DB5E88 sudo add-apt-repository "deb https://repo.sovrin.org/sdk/deb xenial $channel" sudo apt-get update sudo apt-get install -y libindy ```

stanleyc-trustscience (Fri, 20 Jul 2018 19:18:55 GMT):
How do I know what server libindy it is using? I installed using ``` channel="master" # options: master, rc, stable sudo apt-key adv --keyserver keyserver.ubuntu.com --recv-keys 68DB5E88 sudo add-apt-repository "deb https://repo.sovrin.org/sdk/deb xenial $channel" sudo apt-get update sudo apt-get install -y libindy ```

stanleyc-trustscience (Fri, 20 Jul 2018 19:21:49 GMT):
I'm trying to run `getting_started.py` And am getting this ``` INFO:__main__:------------------------------ INFO:__main__:"Sovrin Steward" -> Create wallet WARNING:indy.libindy:_do_call: Function indy_create_wallet returned error 103 WARNING:indy.libindy:_do_call: Function indy_open_wallet returned error 102 Traceback (most recent call last): File "getting_started.py", line 871, in run_coroutine(run) File "/home/dev/git/github.com/hyperledger/indy-sdk/samples/python/src/utils.py", line 47, in run_coroutine loop.run_until_complete(coroutine()) File "/usr/lib/python3.6/asyncio/base_events.py", line 468, in run_until_complete return future.result() File "getting_started.py", line 48, in run steward_wallet = await wallet.open_wallet(steward_wallet_name, None, steward_wallet_credentials) File "/usr/local/lib/python3.6/dist-packages/indy/wallet.py", line 97, in open_wallet open_wallet.cb) indy.error.IndyError: ErrorCode.CommonInvalidParam3 FATAL: exception not rethrown Aborted (core dumped) ```

stanleyc-trustscience (Fri, 20 Jul 2018 19:22:04 GMT):
Wondering if I need a specific version of libindy

stanleyc-trustscience (Fri, 20 Jul 2018 22:59:29 GMT):
@swcurran I see your geneis txn file here (http://159.89.115.24/) has different node ips. I thought you had docker running on a single node, how come they have different ips?

swcurran (Fri, 20 Jul 2018 23:05:30 GMT):
That Ledger is actually at Digital Ocean on 4 nodes - 4 VMs. In our local dev environments, we run it on one node with different ports.

swcurran (Fri, 20 Jul 2018 23:05:50 GMT):
As long as the genesis file reflects the config, all good.

sarveshgupta89 (Sat, 21 Jul 2018 00:46:01 GMT):
Has joined the channel.

sarveshgupta89 (Sat, 21 Jul 2018 00:47:32 GMT):
Hi Team, I am trying to get indy setup running on MacOS using this link :

sarveshgupta89 (Sat, 21 Jul 2018 00:47:33 GMT):
https://github.com/hyperledger/indy-sdk#how-to-start-local-nodes-pool-with-docker

sarveshgupta89 (Sat, 21 Jul 2018 00:48:32 GMT):
However, seems like we do not have a good cli setup for Mac OS. Would be great if you can guide on the right link to follow for running the INDY pool and client on MacOS. Want to run all the commands in the ALICE tutorial

sarveshgupta89 (Sat, 21 Jul 2018 00:49:01 GMT):
was using Docker env so far

sarveshgupta89 (Sat, 21 Jul 2018 00:49:09 GMT):
Please advise. Thanks

cattamanchi (Sat, 21 Jul 2018 01:02:37 GMT):
Has joined the channel.

cattamanchi (Sun, 22 Jul 2018 03:58:47 GMT):
Are there any Verify Organizations Network for US?

cattamanchi (Sun, 22 Jul 2018 04:04:17 GMT):
I ran through the examples of Indy. But now how I get to live connection, how do go about testing for identity and job verification in real world.

JaswanthReddyV (Sun, 22 Jul 2018 10:45:01 GMT):
Has joined the channel.

DonClaude (Sun, 22 Jul 2018 19:53:30 GMT):
Has joined the channel.

rogermartins (Mon, 23 Jul 2018 09:36:09 GMT):
Hi guys

rogermartins (Mon, 23 Jul 2018 09:36:32 GMT):
I'm trying to run the indy java samples, but i'm having an error while setting the protocol version

rogermartins (Mon, 23 Jul 2018 09:36:35 GMT):
an npe

rogermartins (Mon, 23 Jul 2018 09:36:49 GMT):
@ Anoncreds.java:24

rogermartins (Mon, 23 Jul 2018 09:37:08 GMT):
i'm probably missing something here, anyone knows what might be the issue?

pimotte (Mon, 23 Jul 2018 09:47:59 GMT):
@rogermartins You need to have libindy installed and libindy.so in your path. (Also, the samples are from the sdk, so #indy-sdk is the recommended channel for questions like these)

rogermartins (Mon, 23 Jul 2018 09:48:54 GMT):
@pimotte Thank you, moving channels now

xnopre (Mon, 23 Jul 2018 13:39:18 GMT):
Has joined the channel.

matthewdavie (Mon, 23 Jul 2018 17:48:16 GMT):
Has joined the channel.

tbenzies (Mon, 23 Jul 2018 20:20:53 GMT):
Has joined the channel.

tbenzies (Mon, 23 Jul 2018 20:20:55 GMT):
Reminder to register for the final Hackfest of the year in Montreal (October 3-4) http://hyperledger.org/event/hyperledger-hackfest-october-2018

Gokulraja (Tue, 24 Jul 2018 05:29:50 GMT):
Has joined the channel.

miranda (Tue, 24 Jul 2018 14:06:30 GMT):
Has joined the channel.

marrowdev (Tue, 24 Jul 2018 16:00:50 GMT):
Has joined the channel.

marksta (Tue, 24 Jul 2018 16:48:54 GMT):
Has joined the channel.

Alexi (Wed, 25 Jul 2018 19:47:04 GMT):
Has joined the channel.

Alexi (Wed, 25 Jul 2018 19:49:01 GMT):
Hey all, trying to get started with indy sdk. Was working on the 'Write a DID and Query Its Verkey' tutorial. at step 2 "open ledger and get handle" I get _indy_loop_callback: Function returned error 308 Error occurred: ErrorCode.PoolIncompatibleProtocolVersion I followed all the steps to set up the environment so I do not know what could be causing this. Does anyone have any suggestions on how I could fix this?

tomislav (Wed, 25 Jul 2018 23:12:33 GMT):
@Alexi You may need to invoke pool.set_protocol_version(2) to use the latest indy node version

DuckLover (Thu, 26 Jul 2018 11:45:53 GMT):
Has joined the channel.

DuckLover (Thu, 26 Jul 2018 11:46:26 GMT):
Greeting!

DuckLover (Thu, 26 Jul 2018 11:46:50 GMT):
Could someone pls help me out with some setup Problems?

DuckLover (Thu, 26 Jul 2018 11:46:53 GMT):
https://stackoverflow.com/questions/51536879/proper-indy-setup-with-docker

LeHieu (Thu, 26 Jul 2018 12:03:56 GMT):
Has joined the channel.

Sean_Bohan (Thu, 26 Jul 2018 13:19:50 GMT):
Hyperledger Indy WG Call is happening today (7/26)! Agenda: Housekeeping Debugging Tools in Hyperledger Indy with DevinF Digital Inclusion through Trust and Agency: India - DanielH recaps his experience and learnings Open Discussion When: 8am PT 9am MT 10am CT 11amET 4pmBST Where: Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/232861185 Or iPhone one-tap (US Toll): +16465588656,232861185# or +14086380968,232861185# Or Telephone: Dial: +1 646 558 8656 (US Toll) or +1 408 638 0968 (US Toll) Meeting ID: 232 861 185 International numbers available: https://zoom.us/zoomconference?m=a0jD_rTMnh0ZYGQDOKPCNrK_0dP7WPfp1

Alexi (Thu, 26 Jul 2018 15:14:25 GMT):
@tomislav thanks for the suggestion! so I put it in the script and that fixed the incompatibleVersion error, however now it struggles to continue. Still in step 2 when executing "pool_handle = await pool.open_pool_ledger(config_name=pool_name, config=None)" there is a long pause and then I get: _indy_loop_callback: Function returned error 307 Error occurred: ErrorCode.PoolLedgerTimeout

Alexi (Thu, 26 Jul 2018 15:14:34 GMT):
Any ideas on what may be causing this?

vladan.divac (Thu, 26 Jul 2018 15:44:00 GMT):
Has joined the channel.

yocal (Thu, 26 Jul 2018 16:58:20 GMT):
Has joined the channel.

tomislav (Thu, 26 Jul 2018 19:15:10 GMT):
Your script can't connect to the nodes. Check your genesis file and make sure the node is accessible with those IPs.

Alexi (Thu, 26 Jul 2018 19:28:08 GMT):
That did it! by default the genesis file is looking for the nodes on ip 10.0.0.2, but per the first set of instructions I am running them on local 127.0.0.1. Thanks @tomislav !

eidand (Fri, 27 Jul 2018 10:14:03 GMT):
Has joined the channel.

eidand (Fri, 27 Jul 2018 10:15:01 GMT):
hello, is there a tutorial I could use, showing all the steps from zero to hero? I'd like to get an indy setup running on ubuntu 16.04 from absolute nothing

Alexi (Fri, 27 Jul 2018 15:32:02 GMT):
Hey @eidand I also run ubuntu 16.04 this is what I did, follow the install part of this page https://github.com/hyperledger/indy-sdk then follow this guid to build form source https://github.com/hyperledger/indy-sdk/blob/master/doc/ubuntu-build.md

Alexi (Fri, 27 Jul 2018 15:32:46 GMT):
I then did the Write a DID and Query Its Verkey tutorial to get familiar with it (I ran into a couple of problems in the tutorial, the solution of whiuch you will find just above these messages in this chat)

eidand (Fri, 27 Jul 2018 20:26:54 GMT):
thank you

leondcastle (Sat, 28 Jul 2018 08:52:26 GMT):
Has joined the channel.

leondcastle (Sat, 28 Jul 2018 12:34:14 GMT):
hello, what's the minimum spec for a indy-node to work? I plan to create a test network in AWS. Also, is there a publicly available test network? thanks

tomislav (Sat, 28 Jul 2018 23:56:15 GMT):
@leondcastle You can find the sandbox and live network information here https://github.com/sovrin-foundation/sovrin/tree/master/sovrin

leondcastle (Sun, 29 Jul 2018 03:44:23 GMT):
thanks @tomislav

x1990 (Sun, 29 Jul 2018 15:49:30 GMT):
Has joined the channel.

x1990 (Sun, 29 Jul 2018 15:56:43 GMT):
hello, I have a question please: in order for an issuer to create a credential ("anoncreds.issuer_create_credential"), the Issuer needs to create an instantiated credentials json -- and within the values of the credentials json, there is a "raw" value and an "encoded" value for each attribute. May I know how is the "encoded" value calculated from the "raw" value please? for eg: claim_json = json.dumps({ "sex": { "raw": "male", "encoded": "5944657099558967239210949258394887428692050081607692519917050011144233115103"}, "name": {"raw": "Alex", "encoded": "1139481716457488690172217916278103335"}, "height": {"raw": "175", "encoded": "175"}, "age": {"raw": "28", "encoded": "28"} })

tomislav (Sun, 29 Jul 2018 23:33:54 GMT):
@x1990 It's application specific, meaning up to you to encode the values as you see fit. I think the team is working on standardizing this.

x1990 (Mon, 30 Jul 2018 10:32:21 GMT):
@tomislav Ah ok thanks very much for the clarification.

DuckLover (Mon, 30 Jul 2018 10:48:17 GMT):
Is there any clarification what indy-node and indy-sdk are capable to do? I followed @Alexi advice and installed indy-sdk and the ubuntu build. The indy-cli differs from the other alice tutorials so im not really sure what i need to do to be able to follow the tutorials. Do i need indy-node to use its as a pool? Or ist there any not deprecated tutorial i can follow? thanks

the_identity_guy (Mon, 30 Jul 2018 14:22:07 GMT):
Are there any documentation comparing the Indy DLT and other blockchains such as Ethereum, in terms of performance, and for the purpose of identity management?

dredMonkey (Mon, 30 Jul 2018 18:59:44 GMT):
Has joined the channel.

dredMonkey (Mon, 30 Jul 2018 19:00:20 GMT):
Hi every One .... I have been trying a couple of times to get a Indy to run by following the docs from github and I have been unsuccessfull so far .... is there somewhere a "oneclick" script that grabs and sets up everything with sample applications that work out of the box

geethanisp (Tue, 31 Jul 2018 01:55:30 GMT):
Has joined the channel.

geethanisp (Tue, 31 Jul 2018 02:06:06 GMT):
Hi I'm new to indy and a absolute beginner. What is the best guide to follow on indy? I want to set up a indy network so that I can try the example "Alice get Transcript"

geethanisp (Tue, 31 Jul 2018 02:23:35 GMT):
I'm trying with https://github.com/hyperledger/indy-sdk. It gives me following error. Can someone help to resolve this pls

geethanisp (Tue, 31 Jul 2018 02:25:16 GMT):

Clipboard - July 31, 2018 7:56 AM

prmdmshra (Tue, 31 Jul 2018 11:45:35 GMT):
Hi Team, I am trying to setup the following https://github.com/IBM-Blockchain-Identity/indy-ssivc-tutorial/ I am getting the following error on running : $ cd TheOrgBook/docker $ ./manage.sh build Error: Building django image ... /bin/sh: 1: /usr/libexec/s2i/assemble: Permission denied Build failed ERROR: An error occurred: non-zero (13) exit code from python-libindy Please suggest

Alexi (Tue, 31 Jul 2018 12:05:43 GMT):
@prmdmshra "/bin/sh: 1: /usr/libexec/s2i/assemble: Permission denied" try running the build command with --user option if not, while not best practice then try running it with prefix sudo

Alexi (Tue, 31 Jul 2018 12:08:43 GMT):
@geethanisp when you did the install instrctions: sudo apt-key adv --keyserver keyserver.ubuntu.com --recv-keys 68DB5E88 sudo add-apt-repository "deb https://repo.sovrin.org/sdk/deb xenial {release channel}" sudo apt-get update sudo apt-get install -y libindy did you replace {release channel} with which version you wanted? (i.e stable) so: sudo add-apt-repository "deb https://repo.sovrin.org/sdk/deb xenial stable"

geethanisp (Tue, 31 Jul 2018 14:34:31 GMT):
@Alexi thanks! did that and it worked

geethanisp (Tue, 31 Jul 2018 14:50:45 GMT):
I now have indy_pool_network running (as per https://github.com/hyperledger/indy-sdk). Also has setup the indy-cli as per the instruction given in https://github.com/hyperledger/indy-sdk/blob/master/cli/README.md

geethanisp (Tue, 31 Jul 2018 14:53:06 GMT):

Clipboard - July 31, 2018 8:24 PM

geethanisp (Tue, 31 Jul 2018 14:57:23 GMT):

Clipboard - July 31, 2018 8:28 PM

geethanisp (Tue, 31 Jul 2018 14:57:32 GMT):
unsure how I can connect indy-cli with the indy_pool running on the docker. tried using "pool connect indy_pool" from the indy-cli but it says pool doesn't exists.

geethanisp (Tue, 31 Jul 2018 14:57:32 GMT):
unsure how I can connect indy-cli with the indy_pool running on the docker. tried using "pool connect indy_pool" from the indy-cli but it says pool doesn't exists. check the image above

geethanisp (Tue, 31 Jul 2018 14:57:32 GMT):
unsure how I can connect indy-cli with the indy_pool running on the docker. tried using "pool connect indy_pool" from the indy-cli but it says pool doesn't exists. check the image below.

geethanisp (Tue, 31 Jul 2018 14:57:32 GMT):
unsure how I can connect indy-cli with the indy_pool running on the docker. tried using "pool connect indy_pool" from the indy-cli but it says pool doesn't exists. check the image below. can some one point how to connect the indy -cli to the pool. I want to try out the getting started tutorial given at https://github.com/hyperledger/indy-node/blob/master/getting-started.md

geethanisp (Tue, 31 Jul 2018 14:59:19 GMT):

Clipboard - July 31, 2018 8:30 PM

Alexi (Tue, 31 Jul 2018 15:29:55 GMT):
so the tutorial you are using is deprecated, and links to this one https://github.com/hyperledger/indy-sdk/blob/master/doc/getting-started/getting-started.md

Alexi (Tue, 31 Jul 2018 15:31:16 GMT):
personally I have not used the CLI, wehre I set up the sdk and have been playing around with scripts. where this is a good basic tutorial to start off with: https://github.com/hyperledger/indy-sdk/blob/master/doc/how-tos/write-did-and-query-verkey/README.md that being said I had two problems with it which if you scroll up about 10 messages you will see the solutions to

swcurran (Tue, 31 Jul 2018 17:25:01 GMT):
@prmdmshra - I think you just have not made /usr/libexec/s2i/assemble executable. I think it is manually installed. I think that if you do: *sudo chmod ug+x /usr/libexec/s2i/assemble*

swcurran (Tue, 31 Jul 2018 17:25:01 GMT):
@prmdmshra - I think you just have not made /usr/libexec/s2i/assemble executable. I think it is manually installed. I think that if you do: `sudo chmod ug+x /usr/libexec/s2i/assemble` you should be able to run the manage.sh script.

angieology (Tue, 31 Jul 2018 17:51:23 GMT):
Has joined the channel.

hariomgoyal (Wed, 01 Aug 2018 05:29:52 GMT):
Has joined the channel.

prmdmshra (Wed, 01 Aug 2018 10:29:42 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=aQXnuaNNn7nDp8uGq) @Alexi I am still receiving the same error on running with sudo: $ cd TheOrgBook/docker $ sudo ./manage.sh build *Error:* Building django image ... /bin/sh: 1: /usr/libexec/s2i/assemble: Permission denied Build failed ERROR: An error occurred: non-zero (13) exit code from python-libindy

musquash (Wed, 01 Aug 2018 12:08:31 GMT):
I haved looked at the indy roadmap here https://drive.google.com/drive/folders/0ByaxIFopqmUVOThlbEtsUk1OSDg and was wondering if the plan still holds. Where are we right now and what has to developed? The raodmap is from late january so quite old.

Alexi (Wed, 01 Aug 2018 15:44:22 GMT):
@prmdmshra did you try what @swcurran suggested? its probably not marked as executable

prmdmshra (Wed, 01 Aug 2018 16:21:35 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=ctnuC8orfcbGk83jk) @Alexi I am receiving the error 'No such file or directory' on running 'sudo chmod ug+x /usr/libexec/s2i/assemble` I am running on Ubuntu 16.04 and the S2i verion: s2i v1.1.10

Alexi (Wed, 01 Aug 2018 16:56:06 GMT):
@prmdmshra this means that '/usr/libexec/s2i/assemble' is located in the docker image not on your computer

swcurran (Wed, 01 Aug 2018 20:55:38 GMT):
@Alexi - thanks - you are right. @WadeBarnes - can you take a look at this and offer a suggestion?

swcurran (Wed, 01 Aug 2018 21:02:19 GMT):
@Alexi - sorry, rethinking this. The executable should be on the system building the docker image. So I think the at @prmdmshra needs to install the executable on his machine.

swcurran (Wed, 01 Aug 2018 21:02:32 GMT):
Instruction are here: https://github.com/openshift/source-to-image#for-windows

swcurran (Wed, 01 Aug 2018 21:03:34 GMT):
Instructions for Linux are here: https://github.com/openshift/source-to-image#for-linux Other platforms on the same page.

pradeeppadmarajaiah (Thu, 02 Aug 2018 10:06:05 GMT):
Hi All, I am interested to know, how Indy ledger works. Means, what exactly Indy ledger stores when the wallet data is sent to ledger?. Also, interested to know, the components of Indy Ledger. Could you guide me the link to get the overview

Sean_Bohan (Thu, 02 Aug 2018 13:21:01 GMT):
Folks: TODAY, Thursday 8/2 on the Indy WG Call: 1. Quick overview of the 2-Day agents even this week 2. Ryan M demos LibVCX 3. Paul Knowles discusses Hyperledger Indy + Overlays When: 8am PT 9am MT 10am CT 11amET 4pmBST Where: Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/232861185

uhaider (Thu, 02 Aug 2018 15:53:48 GMT):
Has joined the channel.

uhaider (Thu, 02 Aug 2018 15:54:25 GMT):
Any one from academia/industry who could propose some topic for MS thesis on blockhain? Any links to open programs?

intelligentblock (Thu, 02 Aug 2018 19:16:27 GMT):
Has joined the channel.

RubenLassau-Strauven (Fri, 03 Aug 2018 07:20:52 GMT):
Has joined the channel.

Aschi (Fri, 03 Aug 2018 07:22:51 GMT):
Has joined the channel.

zickau (Fri, 03 Aug 2018 09:03:35 GMT):
Has joined the channel.

cedelasen (Fri, 03 Aug 2018 10:41:49 GMT):
Has joined the channel.

dbluhm (Fri, 03 Aug 2018 19:00:15 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=rXJsBPoe5dGLp6KHR) @Sean_Bohan I was hoping to view the recording for the meeting but it hasn't been posted yet. Would you happen to know when that will be available?

dredMonkey (Sun, 05 Aug 2018 08:59:50 GMT):
Good morning .... i am strugling with a await ledger.build_nym_request call File "/usr/local/lib/python3.6/dist-packages/indy/ledger.py", line 233, in build_nym_request c_submitter_did = c_char_p(submitter_did.encode('utf-8')) AttributeError: 'NoneType' object has no attribute 'encode' can anyone perhaps point me into the right direction?

geethanisp (Sun, 05 Aug 2018 18:11:52 GMT):
Hi I'm trying with Indy World demo provided by IBM https://github.com/IBM-Blockchain-Identity/indy-ssivc-tutorial. Managed to get all the components (1) von-network, 2)TheOrgBook and the 3) permitify up and running. But the links inside the permitfy applications points to "localhost" I have the applications hosted on cloud. How to change the permitify applications to point to the actual URL rather than localhost

Gokulraja (Mon, 06 Aug 2018 05:19:34 GMT):
Has left the channel.

gravity (Mon, 06 Aug 2018 08:42:27 GMT):
Has left the channel.

bsuichies (Mon, 06 Aug 2018 09:54:01 GMT):
Has joined the channel.

Christian (Mon, 06 Aug 2018 10:53:44 GMT):
Has joined the channel.

Christian (Mon, 06 Aug 2018 10:57:35 GMT):
Hi all

Christian (Mon, 06 Aug 2018 11:02:46 GMT):
I have been searching but cant find a resolution to this error.. what am i missing? any advice?

Christian (Mon, 06 Aug 2018 11:05:24 GMT):
Hi all I have been installing Indy on Ubuntu 16.04 based on the tutorial below. https://github.com/hyperledger/indy-node/blob/master/docs/start-nodes.md When I get to this section I get the error below init_indy_node Alpha 0.0.0.0 9701 0.0.0.0 9702 [--seed 111111111111111111111111111Alpha] Generating keys for provided seed [--seed Init local keys for client-stack Invalid Seed How can I fix this error? Please j=help

Christian (Mon, 06 Aug 2018 11:05:47 GMT):
seed

ashcherbakov (Mon, 06 Aug 2018 11:31:16 GMT):
@cr

ashcherbakov (Mon, 06 Aug 2018 11:31:16 GMT):
@Christian The recommended way to run Indy Nodes is to use Docker: https://github.com/hyperledger/indy-node/tree/master/environment/docker/pool

LucW (Mon, 06 Aug 2018 12:25:07 GMT):
Hi guys. Sorry for the question that must be quite asked: is there any testnet implementing indy ? To try real-life situations without having to setup ourselves all the nodes and stuff.

prmdmshra (Mon, 06 Aug 2018 12:33:24 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=vkXLGesBD9ymuc69W) @swcurran As suggested, I have reinstalled S2i. The installation seems to be fine for S2i since the build passed in the ./manage.sh It is always failing at the below line of code in .manage/sh echo -e "\nBuilding django image ..." ${S2I_EXE} build \ '../tob-api' \ 'python-libindy' \ 'django' } The error displayed is: Building django image ... /bin/sh: 1: /usr/libexec/s2i/assemble: Permission denied Build failed ERROR: An error occurred: non-zero (13) exit code from python-libindy I found a similar post in github : https://github.com/openshift/source-to-image/issues/465 Kindly suggest

Sean_Bohan (Mon, 06 Aug 2018 16:27:19 GMT):
Video of last week's Indy WG Call: https://drive.google.com/open?id=1xdIUvqmZ7qsCuJvv5UtY0tz3m8YUtlVa Notes from our call: Find out more about PaulK's presentation on Overlays: #overlays-services on the Sovrin Slack LibVCX Demo with RyanM https://github.com/evernym/sdk Indy HIPEs https://github.com/hyperledger/indy-hipe Daniel's Thoughts on the Agency Summit https://docs.google.com/document/d/1TP_7MKfuIrlY3rz4cz_tuuCFi7hdUWifeKwr5h-QTYM/edit Hyperledger Meetups https://www.meetup.com/pro/Hyperledger/ Hyperledger Hackfest Montreal Signup: http://hyperledger.org/event/hyperledger-hackfest-october-2018 Where Indy folks will be gathering in the next few months: Crypto 2018 8/19-8/23 Santa Barbara, CA MyData 2018 8/29-9/1 Helsinki Finland Rebooting Web of Trust 9/24-9/26 Montreal Canada Hyperledger Member Summit 10/1-2 Montreal Canada Hyperledger Hackfest 10/3-5 Montreal Canada Blockchain Seattle 10/10-10/11 Seattle, WA W3C TPAC 10/22-10/26 Lyons France Internet Identity Workshop 10/23-10/26 Mountainview, CA Indy Folder (Agendas, videos, etc.) https://drive.google.com/drive/u/1/folders/0B_NJV6eJXAA1UWtBSmRNLTR3UjQ

swcurran (Mon, 06 Aug 2018 18:59:31 GMT):
@prmdmshra - I think I might have a handle on the issue. What platform are you running on - e.g. what is the OS into which you are typing the ./manage.sh? Linux, perhaps? Can you check the line-endings of the file "TheOrgBook/tob-api/openshift/python-libindy/s2i/bin/assemble"? I suspect that they have Windows CRLF line endings vs. just LF that you need. You can use instructions here to look at that: https://stackoverflow.com/questions/3569997/how-to-find-out-line-endings-in-a-text-file. You might want to check the other files in the same directory. If so, I suspect the s2i is trying to run "/bin/bash" instead of "/bin/bash" - which makes sense. Not 100% sure as the "Permission denied" thing is weird - vs. Not found. There is a reference in TheOrgBook readme.md to an error like this being seen.

Cleverence (Tue, 07 Aug 2018 07:17:30 GMT):
Has joined the channel.

microinclude (Wed, 08 Aug 2018 09:45:32 GMT):
Has joined the channel.

microinclude (Wed, 08 Aug 2018 09:45:38 GMT):
Hi everyone, good day, I am running indy-sdk on python but keeps getting errors when I try to open pool ledger. I am using the genesis sandbox file from this link `https://github.com/sovrin-foundation/sovrin/blob/master/sovrin/pool_transactions_sandbox_genesis`, stack backtrace is below: ```thread '' panicked at 'called `Option::unwrap()` on a `None` value', libcore/option.rs:335:21 stack backtrace: 0: 0x1068b7c8b - std::sys::unix::backtrace::tracing::imp::unwind_backtrace::h971d95e9e3db308e 1: 0x1068a728b - std::sys_common::backtrace::print::h4029c802e44d6cdd 2: 0x1068b25ed - std::panicking::default_hook::{{closure}}::h1ecea4a975914b6f 3: 0x1068b234a - std::panicking::default_hook::h7cf0c6fc3b540dfe 4: 0x1068b2a36 - std::panicking::rust_panic_with_hook::h66177ef9f133b67f 5: 0x1068b286a - std::panicking::begin_panic_fmt::h01fcfca126e50287 6: 0x1068b2762 - rust_begin_unwind 7: 0x1068cd505 - core::panicking::panic_fmt::h0146da1a7e276266 8: 0x1068cd406 - core::panicking::panic::h28c5890fb1451720 9: 0x1061135d9 - >::unwrap::h1e36f2938c2fb447 10: 0x105d1bd33 - indy::services::ledger::merkletree::::consistency_proof::h0cfceea6e11df377 11: 0x1060c036d - indy::services::pool::catchup::CatchupHandler::check_cons_proofs::h7eae9966dfe1b538 12: 0x1060bf49d - indy::services::pool::catchup::CatchupHandler::catchup_step::h4a4c54e39d0a571e 13: 0x1060be356 - indy::services::pool::catchup::CatchupHandler::process_catchup_rep::h8d359d2ed19cf2a8 14: 0x1060bb3a0 - indy::services::pool::catchup::CatchupHandler::process_msg::h73faf88b363f8289 15: 0x105b93e41 - indy::services::pool::PoolWorkerHandler::process_msg::h744b1e0d971539bb 16: 0x105b975cc - indy::services::pool::PoolWorker::process_actions::h3ecf6dcf45b512a5 17: 0x105b9730f - indy::services::pool::PoolWorker::_run::hea1801bb3f88dc96 18: 0x105b96ac6 - indy::services::pool::PoolWorker::run::hfaacf6e93d303f83 19: 0x105ecc91f - indy::services::pool::Pool::new::{{closure}}::h3747611cf24869ef 20: 0x1061ae044 - std::sys_common::backtrace::__rust_begin_short_backtrace::hfc9218aeea3dc5dc 21: 0x105b617ab - std::thread::Builder::spawn::{{closure}}::{{closure}}::hec090bb7571a5b30 22: 0x105fafc34 - as core::ops::function::FnOnce<()>>::call_once::h640ddec952e32834 23: 0x105e25614 - std::panicking::try::do_call::he1501eb0abe31c91 24: 0x1068c684e - __rust_maybe_catch_panic 25: 0x105e25489 - std::panicking::try::h831c2fa4620c42fe 26: 0x105fafc84 - std::panic::catch_unwind::h10b0bff00e600a98 27: 0x105b612f8 - std::thread::Builder::spawn::{{closure}}::h086eb89fe156a6c2 28: 0x105b61965 - >::call_box::hb10f46eea60cd339 29: 0x1068ba3f7 - std::sys_common::thread::start_thread::hffc3810a6e03b74d 30: 0x1068a6188 - std::sys::unix::thread::Thread::new::thread_start::hbd2eb72915a2a86e 31: 0x7fff70ba8660 - _pthread_body 32: 0x7fff70ba850c - _pthread_start ```

microinclude (Wed, 08 Aug 2018 09:45:38 GMT):
Hi everyone, good day, I am running indy-sdk on python but keeps getting errors when I try to open pool ledger. I am using the genesis sandbox file from this link `https://github.com/sovrin-foundation/sovrin/blob/master/sovrin/pool_transactions_sandbox_genesis`, stack backtrace is below: ```thread '' panicked at 'called `Option::unwrap()` on a `None` value', libcore/option.rs:335:21 stack backtrace: 0: 0x1068b7c8b - std::sys::unix::backtrace::tracing::imp::unwind_backtrace::h971d95e9e3db308e 1: 0x1068a728b - std::sys_common::backtrace::print::h4029c802e44d6cdd 2: 0x1068b25ed - std::panicking::default_hook::{{closure}}::h1ecea4a975914b6f 3: 0x1068b234a - std::panicking::default_hook::h7cf0c6fc3b540dfe 4: 0x1068b2a36 - std::panicking::rust_panic_with_hook::h66177ef9f133b67f 5: 0x1068b286a - std::panicking::begin_panic_fmt::h01fcfca126e50287 6: 0x1068b2762 - rust_begin_unwind 7: 0x1068cd505 - core::panicking::panic_fmt::h0146da1a7e276266 8: 0x1068cd406 - core::panicking::panic::h28c5890fb1451720 9: 0x1061135d9 - >::unwrap::h1e36f2938c2fb447 10: 0x105d1bd33 - indy::services::ledger::merkletree::::consistency_proof::h0cfceea6e11df377 11: 0x1060c036d - indy::services::pool::catchup::CatchupHandler::check_cons_proofs::h7eae9966dfe1b538 12: 0x1060bf49d - indy::services::pool::catchup::CatchupHandler::catchup_step::h4a4c54e39d0a571e 13: 0x1060be356 - indy::services::pool::catchup::CatchupHandler::process_catchup_rep::h8d359d2ed19cf2a8 14: 0x1060bb3a0 - indy::services::pool::catchup::CatchupHandler::process_msg::h73faf88b363f8289 15: 0x105b93e41 - indy::services::pool::PoolWorkerHandler::process_msg::h744b1e0d971539bb 16: 0x105b975cc - indy::services::pool::PoolWorker::process_actions::h3ecf6dcf45b512a5 17: 0x105b9730f - indy::services::pool::PoolWorker::_run::hea1801bb3f88dc96 18: 0x105b96ac6 - indy::services::pool::PoolWorker::run::hfaacf6e93d303f83 19: 0x105ecc91f - indy::services::pool::Pool::new::{{closure}}::h3747611cf24869ef 20: 0x1061ae044 - std::sys_common::backtrace::__rust_begin_short_backtrace::hfc9218aeea3dc5dc 21: 0x105b617ab - std::thread::Builder::spawn::{{closure}}::{{closure}}::hec090bb7571a5b30 22: 0x105fafc34 - as core::ops::function::FnOnce<()>>::call_once::h640ddec952e32834 23: 0x105e25614 - std::panicking::try::do_call::he1501eb0abe31c91 24: 0x1068c684e - __rust_maybe_catch_panic 25: 0x105e25489 - std::panicking::try::h831c2fa4620c42fe 26: 0x105fafc84 - std::panic::catch_unwind::h10b0bff00e600a98 27: 0x105b612f8 - std::thread::Builder::spawn::{{closure}}::h086eb89fe156a6c2 28: 0x105b61965 - >::call_box::hb10f46eea60cd339 29: 0x1068ba3f7 - std::sys_common::thread::start_thread::hffc3810a6e03b74d 30: 0x1068a6188 - std::sys::unix::thread::Thread::new::thread_start::hbd2eb72915a2a86e 31: 0x7fff70ba8660 - _pthread_body 32: 0x7fff70ba850c - _pthread_start ``` What causes this issue? Did I miss something wrong?

mero (Wed, 08 Aug 2018 10:32:19 GMT):
Has joined the channel.

prmdmshra (Wed, 08 Aug 2018 11:41:03 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=EQN6t9YqaPkvBP8qy) @swcurran The files seems to be /bin/bash. Please find the results of 'file --keep-going -- *' below: assemble: Bourne-Again shell script text executable\012- a /bin/bash script, ASCII text executable\012- data run: Bourne-Again shell script text executable\012- a /bin/bash script, ASCII text executable\012- data usage: Bourne-Again shell script text executable\012- a /bin/bash script, ASCII text executable\012- data Still receiving the same error on running './manage.sh build' in 'TheOrgBook/docker' Building django image ... /bin/sh: 1: /usr/libexec/s2i/assemble: Permission denied Build failed ERROR: An error occurred: non-zero (13) exit code from python-libindy

vtech (Wed, 08 Aug 2018 11:58:26 GMT):

Sean_Bohan (Wed, 08 Aug 2018 14:19:21 GMT):
This Thursday (TOMORROW!) August 9 is the Hyperledger Indy WG Call. Times and call info below. Agenda: Housekeeping Hyperledger & Indy Events: Aug-Sept-Oct DID Auth and Indy with Kyle DH Open Discussion When: 8am PT 9am MT 10am CT 11amET 4pmBST Where: Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/232861185 Or iPhone one-tap (US Toll): +16465588656,232861185# or +14086380968,232861185# Or Telephone: Dial: +1 646 558 8656 (US Toll) or +1 408 638 0968 (US Toll) Meeting ID: 232 861 185 International numbers available: https://zoom.us/zoomconference?m=a0jD_rTMnh0ZYGQDOKPCNrK_0dP7WPfp1

prmdmshra (Wed, 08 Aug 2018 16:01:48 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=bibWvgpa9hBsscb6f) It would be great if we can resolve the error by today. I wanted to present the capabilities of Indy tomorrow

swcurran (Wed, 08 Aug 2018 16:26:37 GMT):
Sorry @prmdmshra - I don't have the tools to determine why your seeing the error. I have gone through the full process since we started this discussion and was able to make it work - Windows 10 machine with docker, docker-compose, git-bash and s2i. I've googled the various reasons for such an error and can't see what's happening. What machine are you running on?

prmdmshra (Wed, 08 Aug 2018 17:11:06 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=urfrHzaDQWuShPYsK) @swcurran I am on Ubuntu 16.04 OS

swcurran (Wed, 08 Aug 2018 17:11:57 GMT):
OK - I'll try to run a VM with Ubuntu on my machine to recreate, but I can't promise to do it immediately. Sorry about that.

prmdmshra (Wed, 08 Aug 2018 17:13:51 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=ft5JiMkEtc4Qw5grK) @swcurran No issues.......btw thanks a lot for your suggestions and support. I am also trying to resolve it.....Thanks :-)

NataliaDracheva (Thu, 09 Aug 2018 12:53:55 GMT):
Has joined the channel.

disrupt101 (Thu, 09 Aug 2018 14:30:19 GMT):
Has joined the channel.

nage (Thu, 09 Aug 2018 15:16:35 GMT):
Welcome @hvandurme !!

nage (Thu, 09 Aug 2018 15:19:42 GMT):
Welcome Reddy and Jan (rocket.chat handles aren't readily tab-completing)!

hvandurme (Thu, 09 Aug 2018 15:21:46 GMT):
Thanks! @nage

kdenhartog (Thu, 09 Aug 2018 15:25:27 GMT):
https://docs.google.com/presentation/d/1LiVi3wmUNFxUc5AXQyZVW2Gn1X7hcS9obgQ6uuoNZTI/edit?usp=sharing

kdenhartog (Thu, 09 Aug 2018 15:25:27 GMT):
https://docs.google.com/presentation/d/1LiVi3wmUNFxUc5AXQyZVW2Gn1X7hcS9obgQ6uuoNZTI/edit?usp=sharing Slides have been updated now. If you find other things that I missed let me know and I'll update it

swcurran (Thu, 09 Aug 2018 15:25:55 GMT):
DID Auth presentation from today's Developer's Call ^^^

nage (Thu, 09 Aug 2018 15:35:04 GMT):
To see progress on the agent specifications take a look at the PRs proposed against https://github.com/hyperledger/indy-hipe repository

nage (Thu, 09 Aug 2018 15:36:20 GMT):
(this is Indy's RFC process for capturing development and architecture thinking and ratify standard practices and conventions for Indy, the Sovrin Foundation extends these in this repo https://github.com/sovrin-foundation/sovrin-sip)

Vlad (Thu, 09 Aug 2018 15:58:27 GMT):
Has joined the channel.

x1990 (Thu, 09 Aug 2018 20:33:21 GMT):
Hello, I have a question regarding the Onboarding process described in the example case study "https://github.com/hyperledger/indy-sdk/blob/master/doc/getting-started/getting-started.md", under point number 11 please. For convenience, the relevant excerpt is: "Faber anonymously encrypts the connection response by calling crypto.anon_crypt with the Steward verkey. The Anonymous-encryption schema is designed for the sending of messages to a Recipient which has been given its public key. Only the Recipient can decrypt these messages, using its private key. While the Recipient can verify the integrity of the message, it cannot verify the identity of the Sender." My question is: what is the protection against someone (for eg, a MITM) who pretends to be Faber and submits its own DID,VerKey pair (instead of Faber's) as connection response to Steward?

stanleyc-trustscience (Thu, 09 Aug 2018 21:17:17 GMT):
With the new indy-sdk, how do we not encrypt the wallet? What value should I set `credentials` to? `wallet.create_wallet(pool_name, wallet_name, None, "{}", credentials) I tried None (or null) and empty string, `{"key":""}` and `{}`. And none of them worked

stanleyc-trustscience (Thu, 09 Aug 2018 21:17:17 GMT):
With the new indy-sdk, how do we not encrypt the wallet? What value should I set `credentials` to? `wallet.create_wallet(pool_name, wallet_name, None, "{}", credentials)` I tried None (or null) and empty string, `{"key":""}` and `{}`. And none of them worked

stanleyc-trustscience (Thu, 09 Aug 2018 21:17:17 GMT):
With the new indy-sdk, how do we not encrypt the wallet? What value should I set `credentials` to? `wallet.create_wallet(pool_name, wallet_name, None, "{}", credentials) I tried None (or null) and empty string, `{"key":""}` and `{}`. And none of them worked

Penh (Fri, 10 Aug 2018 10:59:44 GMT):
Has joined the channel.

marrowdev (Fri, 10 Aug 2018 14:15:26 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=56AXvDLb5AovFK939) @x1990 I am also wondering this, is it assumed that the connection request is sent via some other secure means?

marrowdev (Fri, 10 Aug 2018 14:26:40 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=56AXvDLb5AovFK939) @x1990 Quoting a response I got from #indy-sdk channel "@marrowdev The connection establishment process encrypts all messages agent-to-agent, intercepting wouldn't compromise it. The initial invitiation/offer should be delivered via secure way in already established trusted environment. These protocols and concepts are discussed in great detail in #indy-agent channel."

swcurran (Fri, 10 Aug 2018 16:43:45 GMT):
@marrowdev @x1990 - I agree with what was said above about agent communications and the methods to prevent compromise. However, there is another key thing to keep in mind in this ecosystem. A connection alone should not be trusted - it merely establishing a communication channel between two identities. The use of Verifiable Credentials should be used after the connection is established to determine if you should trust the Identity for the purposes of the connection. Thus, if a MITM attack could occur, the connection might be established, but the attacker could not prove, for example, that the Identity is John Smith as issued by the DMV (e.g. on their Driver's Licence) to them and that their TaxID is XXXXXX as issued by the Federal Government to them (if that information is needed). Just knowing that data is no longer enough - you have to be able to prove who issued it to you. VCs are the key for establishing the necessary level of trust for continuing with the relationship. Note as well that the verification can (and should) be both ways - Alice must prove with VCs who she is to Faber, and Faber must prove with VCs to Alice who it is. It is a peer-to-peer relationship.

stanleyc-trustscience (Fri, 10 Aug 2018 17:11:31 GMT):
@swcurran Very good post. I'd like brainstorm a bit about the process: Establishing a secure connection with pairwise DID: 1) Alice send Faber a connection request 2) Faber responds with a faber_alice_did and nonce 3) Alice sends Faber a connection response (alice_faber_did, alice_faber_verkey, nonce) encrypted with faber_alice_verkey 4) Faber decrypts the connection response with faber_alice_verkey and make sure nonce match 5) Connection is now established Note at this point, they don't know each other's identity within the society, or haven't prove their identity to each other yet Vetting each other's Identity 6) Alice gets a Social Security Number proof request from Faber 7) Alice searches within her wallet for the credential issued by the government, and sends the credentials to Faber 8) Fabers confirms the proof by verifying it against indy ledger 9) Faber gets a Education Instituion proof request from Alice 10) Faber searches within its wallet for the credential issued by the government, and send the credentials to Alice 11) Alice confirms the proof by verifying it against indy ledger 12) Both Alice and Faber now confirm each other's identity within the society The process of vetting identity doesn't necessary need to be performed using the steps above. It could be real life interaction, or phone , email, etc. But if we were to do it in the VC way, this is what it'd look like, I'd imagine Also the protocol of vetting identity using the VC way, doesn't seem to have a standard yet? So perhaps a standard will emerge for exchanging VC for the purpose of vetting social identity

stanleyc-trustscience (Fri, 10 Aug 2018 17:11:31 GMT):
@swcurran Very good post. I'd like brainstorm a bit about the process: Establishing a secure connection with pairwise DID: 1) Alice send Faber a connection request 2) Faber responds with a faber_alice_did and nonce 3) Alice sends Faber a connection response (alice_faber_did, alice_faber_verkey, nonce) encrypted with faber_alice_verkey 4) Faber decrypts the connection response with faber_alice_verkey and make sure nonce match 5) Connection is now established Vetting each other's Identity 6) Alice gets a Social Security Number proof request from Faber 7) Alice searches within her wallet for the credential issued by the government, and sends the credentials to Faber 8) Fabers confirms the proof by verifying it against indy ledger 9) Faber gets a Education Instituion proof request from Alice 10) Faber searches within its wallet for the credential issued by the government, and send the credentials to Alice 11) Alice confirms the proof by verifying it against indy ledger 12) Both Alice and Faber now confirm the identity of each other within the society The process of vetting identity doesn't necessary need to be performed using the steps above. It could be real life interaction, or phone , email, etc. But if we were to do it in the VC way, this is what it'd look up

stanleyc-trustscience (Fri, 10 Aug 2018 17:11:31 GMT):
@swcurran Very good post. I'd like brainstorm a bit about the process: Establishing a secure connection with pairwise DID: 1) Alice send Faber a connection request 2) Faber responds with a faber_alice_did and nonce 3) Alice sends Faber a connection response (alice_faber_did, alice_faber_verkey, nonce) encrypted with faber_alice_verkey 4) Faber decrypts the connection response with faber_alice_verkey and make sure nonce match 5) Connection is now established Note at this point, they don't know each other's identity within the society, or haven't prove their identity to each other yet Vetting each other's Identity 6) Alice gets a Social Security Number proof request from Faber 7) Alice searches within her wallet for the credential issued by the government, and sends the credentials to Faber 8) Fabers confirms the proof by verifying it against indy ledger 9) Faber gets a Education Instituion proof request from Alice 10) Faber searches within its wallet for the credential issued by the government, and send the credentials to Alice 11) Alice confirms the proof by verifying it against indy ledger 12) Both Alice and Faber now confirm the identity of each other within the society The process of vetting identity doesn't necessary need to be performed using the steps above. It could be real life interaction, or phone , email, etc. But if we were to do it in the VC way, this is what it'd look up

stanleyc-trustscience (Fri, 10 Aug 2018 17:11:31 GMT):
@swcurran Very good post. I'd like brainstorm a bit about the process: Establishing a secure connection with pairwise DID: 1) Alice send Faber a connection request 2) Faber responds with a faber_alice_did and nonce 3) Alice sends Faber a connection response (alice_faber_did, alice_faber_verkey, nonce) encrypted with faber_alice_verkey 4) Faber decrypts the connection response with faber_alice_verkey and make sure nonce match 5) Connection is now established Note at this point, they don't know each other's identity within the society, or haven't prove their identity to each other yet Vetting each other's Identity 6) Alice gets a Social Security Number proof request from Faber 7) Alice searches within her wallet for the credential issued by the government, and sends the credentials to Faber 8) Fabers confirms the proof by verifying it against indy ledger 9) Faber gets a Education Instituion proof request from Alice 10) Faber searches within its wallet for the credential issued by the government, and send the credentials to Alice 11) Alice confirms the proof by verifying it against indy ledger 12) Both Alice and Faber now confirm each other's identity within the society The process of vetting identity doesn't necessary need to be performed using the steps above. It could be real life interaction, or phone , email, etc. But if we were to do it in the VC way, this is what it'd look up

stanleyc-trustscience (Fri, 10 Aug 2018 17:11:31 GMT):
@swcurran Very good post. I'd like brainstorm a bit about the process: Establishing a secure connection with pairwise DID: 1) Alice send Faber a connection request 2) Faber responds with a faber_alice_did and nonce 3) Alice sends Faber a connection response (alice_faber_did, alice_faber_verkey, nonce) encrypted with faber_alice_verkey 4) Faber decrypts the connection response with faber_alice_verkey and make sure nonce match 5) Connection is now established Note at this point, they don't know each other's identity within the society, or haven't prove their identity to each other yet Vetting each other's Identity 6) Alice gets a Social Security Number proof request from Faber 7) Alice searches within her wallet for the credential issued by the government, and sends the credentials to Faber 8) Fabers confirms the proof by verifying it against indy ledger 9) Faber gets a Education Instituion proof request from Alice 10) Faber searches within its wallet for the credential issued by the government, and send the credentials to Alice 11) Alice confirms the proof by verifying it against indy ledger 12) Both Alice and Faber now confirm each other's identity within the society The process of vetting identity doesn't necessary need to be performed using the steps above. It could be real life interaction, or phone , email, etc. But if we were to do it in the VC way, this is what it'd look like, I'd imagine

stanleyc-trustscience (Fri, 10 Aug 2018 17:11:31 GMT):
@swcurran Very good post. I'd like brainstorm a bit about the process: Establishing a secure connection with pairwise DID: 1) Alice send Faber a connection request 2) Faber responds with a faber_alice_did and nonce 3) Alice sends Faber a connection response (alice_faber_did, alice_faber_verkey, nonce) encrypted with faber_alice_verkey 4) Faber decrypts the connection response with faber_alice_verkey and make sure nonce match 5) Connection is now established Note at this point, they don't know each other's identity within the society, or haven't prove their identity to each other yet Vetting each other's Identity 6) Alice gets a Social Security Number proof request from Faber 7) Alice searches within her wallet for the credential issued by the government, and sends the credentials to Faber 8) Fabers confirms the proof by verifying it against indy ledger 9) Faber gets a Education Instituion proof request from Alice 10) Faber searches within its wallet for the credential issued by the government, and send the credentials to Alice 11) Alice confirms the proof by verifying it against indy ledger 12) Both Alice and Faber now confirm each other's identity within the society The process of vetting identity doesn't necessary need to be performed using the steps above. It could be real life interaction, or phone , email, etc. But if we were to do it in the VC way, this is what it'd look like, I'd imagine Also the protocol of vetting identity using the VC way, doesn't seem to have a standard yet? So perhaps a standard will emerge for exchanging VC

geethanisp (Sun, 12 Aug 2018 00:57:13 GMT):
Usage of the command should be as below

geethanisp (Sun, 12 Aug 2018 00:58:57 GMT):
@Christian try it like below. Worked for me.

geethanisp (Sun, 12 Aug 2018 00:58:57 GMT):
@Christian try it like below. Worked for me. init_indy_node Node2 0.0.0.0 9701 0.0.0.0 9702 b'Gca0b1Cce09E72cBafb1f967E7BEGdb' where b'Gca0b1Cce09E72cBafb1f967E7BEGdb' is the seed value

geethanisp (Sun, 12 Aug 2018 00:58:57 GMT):
@Christian try it like below. Worked for me. init_indy_node Node2 0.0.0.0 9701 0.0.0.0 b'Gca0b1Cce09E72cBafb1f967E7BEGdb'

geethanisp (Sun, 12 Aug 2018 00:58:57 GMT):
@Christian try it like below. Worked for me. init_indy_node Node2 0.0.0.0 9701 0.0.0.0 b'Gca0b1Cce09E72cBafb1f967E7BEGdb' where b'Gca0b1Cce09E72cBafb1f967E7BEGdb' is the seed value

swcurran (Sun, 12 Aug 2018 18:23:11 GMT):
@stanleyc-trustscience - That's about how I envision it happening at a high level. And absolutely, the details are to be worked out and an issuer ecosystem created. Our project (Verifiable Organizations Network) might be the path to Faber (or some other organization) proving who they are, as we are working with Registries - the gov't entities that track organizations - corps, professional organizations, societies, etc. A comparable, standardized credential issuance for a Driver's Licence organization should be relatively fast to happen - such organizations are in close communication and they face a pretty significant problem with forgeries. They are motivated to correct the problem.

swcurran (Sun, 12 Aug 2018 18:23:11 GMT):
@stanleyc-trustscience - That's about how I envision it happening at a high level. And absolutely, the details are to be worked out and an issuer ecosystem created. Our project (Verifiable Organizations Network) might be the path to Faber (and other organizations) proving who they are, as we are working with Registries - the gov't entities that track organizations - corps, professional organizations, societies, etc. A comparable, standardized credential issuance for a Driver's Licence organization should be relatively fast to happen - such organizations are in close communication and they face a pretty significant problem with forgeries. They are motivated to correct the problem.

paulransfield (Mon, 13 Aug 2018 08:41:08 GMT):
Has joined the channel.

hvandurme (Mon, 13 Aug 2018 11:43:42 GMT):
When running the commands: docker build -f ci/indy-pool.dockerfile -t indy_pool . docker run -itd -p 9701-9708:9701-9708 indy_pool I get the following output in docker: 2018-08-13 11:41:14,793 CRIT Set uid to user 1000 2018-08-13 11:41:14,795 INFO supervisord started with pid 1 2018-08-13 11:41:15,804 INFO spawned: 'node1' with pid 11 2018-08-13 11:41:15,806 INFO spawned: 'node3' with pid 12 2018-08-13 11:41:15,808 INFO spawned: 'node2' with pid 13 2018-08-13 11:41:15,810 INFO spawned: 'node4' with pid 14 2018-08-13 11:41:16,060 DEBG 'node1' stderr output: Traceback (most recent call last): File "/usr/local/bin/start_indy_node", line 18, in run_node(config, self_name, int(sys.argv[2]), int(sys.argv[3])) File "/usr/local/lib/python3.5/dist-packages/indy_node/utils/node_runner.py", line 14, in run_node Logger().enableFileLogging(logFileName) File "/usr/local/lib/python3.5/dist-packages/stp_core/common/log.py", line 100, in enableFileLogging os.makedirs(d) File "/usr/lib/python3.5/os.py", line 241, in makedirs mkdir(name, mode) FileExistsError: [Errno 17] File exists: '/var/log/indy/sandbox' 2018-08-13 11:41:16,078 DEBG fd 6 closed, stopped monitoring (stdout)> 2018-08-13 11:41:16,078 DEBG fd 8 closed, stopped monitoring (stderr)> 2018-08-13 11:41:16,078 INFO exited: node1 (exit status 1; not expected) 2018-08-13 11:41:16,079 DEBG received SIGCLD indicating a child quit 2018-08-13 11:41:17,045 INFO success: node3 entered RUNNING state, process has stayed up for > than 1 seconds (startsecs) 2018-08-13 11:41:17,045 INFO success: node2 entered RUNNING state, process has stayed up for > than 1 seconds (startsecs) 2018-08-13 11:41:17,046 INFO success: node4 entered RUNNING state, process has stayed up for > than 1 seconds (startsecs) 2018-08-13 11:41:17,315 DEBG 'node4' stderr output: Traceback (most recent call last): File "/usr/local/bin/start_indy_node", line 18, in run_node(config, self_name, int(sys.argv[2]), int(sys.argv[3])) File "/usr/local/lib/python3.5/dist-packages/indy_node/utils/node_runner.py", line 27, in run_node from indy_node.server.node import Node File "/usr/local/lib/python3.5/dist-packages/indy_node/server/node.py", line 18, in from plenum.server.node import Node as PlenumNode File "/usr/local/lib/python3.5/dist-packages/plenum/server/node.py", line 65, in from plenum.server.monitor import Monitor File "/usr/local/lib/python3.5/dist-packages/plenum/server/monitor.py", line 26, in pluginManager = PluginManager() File "/usr/local/lib/python3.5/dist-packages/plenum/server/notifier_plugin_manager.py", line 35, in __init__ self.importPlugins() File "/usr/local/lib/python3.5/dist-packages/plenum/server/notifier_plugin_manager.py", line 102, in importPlugins plugins = self._findPlugins() File "/usr/local/lib/python3.5/dist-packages/plenum/server/notifier_plugin_manager.py", line 131, in _findPlugins for pkg in pip.utils.get_installed_distributions() AttributeError: module 'pip' has no attribute 'utils' 2018-08-13 11:41:17,317 INFO spawned: 'node1' with pid 19

hvandurme (Mon, 13 Aug 2018 11:44:13 GMT):
Is this supposed to be the case? It just keeps spewing the same for the rest of the nodes.

hvandurme (Mon, 13 Aug 2018 11:45:08 GMT):
Using the indy-sdk v1.3.0 on MacOS FYI

Alexi (Mon, 13 Aug 2018 17:26:26 GMT):
Hey all, for some reason when creating wallets on the indy pool I am getting the following error: `_do_call: Function indy_create_wallet returned error 103 Error occurred: ErrorCode.CommonInvalidParam4 `

Alexi (Mon, 13 Aug 2018 17:27:34 GMT):
this came out of no where, where it was not happening before. it even happens when I run the write_did script from the documentation. which I could run perfectly fine before. SO I do not know what is causing this to happen all of a sudden.

Alexi (Mon, 13 Aug 2018 17:28:58 GMT):
This call will cuase the error in the write_did script: print_log('\n3. Create new identity wallet\n') await wallet.create_wallet(pool_name, wallet_name, None, None, wallet_credentials) so it says the 4th parameter in this case 'None' is invalid. But why is this is the case when its the boilerplate given by the github documentation

Alexi (Mon, 13 Aug 2018 18:09:37 GMT):
ive been trying to figure this out but looking into it makes it somewhat more confusing to why this is happening: trying to remove the fourth parameter 'None' results in a missing parameter error meaning we need all five parameters to create the wallet. So I tried looking into what the fourth parameter requires, I went to the indy-sdk python library at wallet https://github.com/hyperledger/indy-sdk/blob/master/wrappers/python/indy/wallet.py but found that the create_wallet command in the library is actually only calling for two parameters: async def create_wallet(config: str, credentials: str) -> None: so I am somewhat more lost than before, any ideas?

coderintherye (Mon, 13 Aug 2018 18:46:56 GMT):
Has joined the channel.

microinclude (Tue, 14 Aug 2018 01:57:59 GMT):
@Alexi same problem here

microinclude (Tue, 14 Aug 2018 01:59:34 GMT):
The problem is on this code: ```res = await do_call('indy_open_wallet', c_config, c_credentials, open_wallet.cb)```

microinclude (Tue, 14 Aug 2018 02:01:47 GMT):
open_wallet.cb is the invalid parameter, somewhat the callback is different from the expected?

voddan (Tue, 14 Aug 2018 11:19:14 GMT):
Hi! Please tell me what's `NYM`? What does it stand for?

tomislav (Tue, 14 Aug 2018 13:13:30 GMT):
"Words ending in –nym are often used to describe different classes of words, and the relationships between words. The –nym literally means name, from the Greek onoma meaning name or word." @voddan

tomislav (Tue, 14 Aug 2018 13:16:12 GMT):
In indy, a NYM is a public information about an identity defined by a Did + Verkey.

Alexi (Tue, 14 Aug 2018 16:20:29 GMT):
@microinclude that's wierd. Is anyone else having the same invalidParam4 problem? or have suggestions on what to do? I tried re building the docker indy pool just in case, and I have not changed any of the indy sdk files yet it still is happening. so I have no idea what could be causing this problem to have suddenly started. Also on a side note, the write_did script uses the following command to create wallet await wallet.create_wallet(pool_name, wallet_name, None, None, wallet_credentials) I am wondering why the create_wallet call takes in 5 parameters and what each of them does / what type they have to be. Maybe this will give some insight into why this error is occurring

tomislav (Tue, 14 Aug 2018 17:53:48 GMT):
@Alexi The python wrapper you are using is out of date. The wallet API has changes since and has a new method signature for open and create. It is no longer dependent on pool name.

Alexi (Tue, 14 Aug 2018 20:56:44 GMT):
@tomislav the wrapper I have seems to match with the one on the sdk github repo. Ill re-download and set up everything to be sure, but since the api has since changed this means the tutorials are deprecated so what would be the correct way to call create_wallet at this point? Same parameters as before but without the pool name?

asmallfurrywombat (Wed, 15 Aug 2018 02:39:46 GMT):
Has joined the channel.

asmallfurrywombat (Wed, 15 Aug 2018 02:47:33 GMT):
I'm running the test on the nodejs wrapper using the current indy-sdk source. The first 4 tests pass but the next 8 fail due to PoolLedgerTimeout. Am I missing something? I'm running on the latest Ubuntu.

asmallfurrywombat (Wed, 15 Aug 2018 03:02:40 GMT):
I've also tried running the nodejs sample. Everything builds ok and the ledger appears to launch fine, but I get a DidAlreadyExistsError in response to the createAndStoreMyDid call on line 51 of gettingStarted.js . The indy_pool logs just show a 'Set uid to user ...'. The sdk innards appear to be rust. Out of idle curiosity, is there a recommended way to debug this issue?

microinclude (Wed, 15 Aug 2018 10:32:23 GMT):
hi @tomislav, the sdk i am using is from the master branch. I just tried this one on my code: ```config = json.dumps({"id": self.name}) credentials = json.dumps({"key": self.key}) await wallet.create_wallet(config, credentials)```

microinclude (Wed, 15 Aug 2018 10:33:33 GMT):
am i doing something wrong? i still have the CommonInvalid4 error

tomislav (Wed, 15 Aug 2018 13:33:15 GMT):
@asmallfurrywombat PoolLedgerTimeout most likely means your script can't access the IP of the pool containers. Check that both the pool and your script have the same genesis file (the IP inside it) and they are accessible. DidAlreadyExistsError means exactly that, did is already present in the wallet. Clean up any wallets by removing ~/.indy_client directory.

tomislav (Wed, 15 Aug 2018 13:34:58 GMT):
@microinclude That's about right. Can you check if your installed libindy library is also of the latest version? One way to debug things is to add environment variable RUST_LOG=trace to your process

asmallfurrywombat (Wed, 15 Aug 2018 14:33:41 GMT):
thanks @tomislav .. removing ~/.indy_client got past line 51 :). Now it is dumping core on line 695 during a call to capi.cryptoAnonCrypt(). The core shows a SIGSEGV in libzmq.. will create an issue with the stack details.

PatrikStas (Thu, 16 Aug 2018 04:39:44 GMT):
Has joined the channel.

maydar (Thu, 16 Aug 2018 12:00:33 GMT):
Has joined the channel.

maydar (Thu, 16 Aug 2018 12:23:35 GMT):
Hello. I am new to Indy community. I would likt to install Indy for self-sovereign identity. I would like to create a proof of concept. How do I start? the installation documentation was not clear to me.

Sean_Bohan (Thu, 16 Aug 2018 14:47:26 GMT):
TODAY is the Hyperledger Indy WG Call (times and call info below). Agenda: Housekeeping Hyperledger Indy Project Report (Nathan, Richard, MichaelM) AGENTS! (MatthewH, StephenC, JohnJ) Architecture Working Group & Whitepaper (Nathan) Identity Working Group & Whitepaper (Vipin) Open Discussion When: 8am PT 9am MT 10am CT 11amET 4pmBST Where: Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/232861185 Or iPhone one-tap (US Toll): +16465588656,232861185# or +14086380968,232861185# Or Telephone: Dial: +1 646 558 8656 (US Toll) or +1 408 638 0968 (US Toll) Meeting ID: 232 861 185 International numbers available: https://zoom.us/zoomconference?m=a0jD_rTMnh0ZYGQDOKPCNrK_0dP7WPfp1

Antonio_M (Thu, 16 Aug 2018 14:55:33 GMT):
@maydar If you like to learn by writing code then this is a place to start https://github.com/hyperledger/indy-sdk/tree/master/doc/how-tos

Antonio_M (Thu, 16 Aug 2018 14:55:33 GMT):
@maydar If you like to learn by writing code then this is a place to start https://github.com/hyperledger/indy-sdk/tree/master/doc/how-tos Some examples for certain languages are still missing, but it's possible to follow the Python tutorials and figure out how to make it work in any other of available languages

VipinB (Thu, 16 Aug 2018 15:39:32 GMT):
Identity WG paper is: https://docs.google.com/document/d/1ExFNRx-yYoS8FnDIUX1_0UBMha9TvQkfts2kVnDc4KE/edit#

Sean_Bohan (Thu, 16 Aug 2018 15:55:47 GMT):
awesome - thank you

coderintherye (Thu, 16 Aug 2018 18:29:20 GMT):
What's the best way to help clean up outdated documentation in the hylerledger-indy repo, should it be done via PRs?

coderintherye (Thu, 16 Aug 2018 18:30:04 GMT):
for example https://github.com/hyperledger/indy-node/blob/stable/getting-started.md references a command that doesn't exist, as well as purports to link to a doc about starting a pool, but actually links to something that has no such info

Alexi (Thu, 16 Aug 2018 19:28:32 GMT):
@coderintherye yep documentation and README changes can be done through PR's and is always helpful as those kind of things are generally in a lower priority for the team when they are continually working on the project

microinclude (Thu, 16 Aug 2018 23:00:05 GMT):
@tomislav just missed this step: Note: After building libindy, add the path containing the library the LD_LIBRARY_PATH andDYLD_LIBRARY_PATH environment variables. This is necessary for dynamically linking your application with libindy. The dynamic linker will first check for the library in LD_LIBRARY_PATH if the library in your application doesn't include directory names. If the library in your application does include any directory name, then dynamic linker will search for the library in DYLD_LIBRARY_PATH (not LD_LIBRARY_PATH) so we recommend you set both variables to be safe. Mybad... I'll update you once everything's working

FCode (Fri, 17 Aug 2018 01:37:19 GMT):
Has joined the channel.

FCode (Fri, 17 Aug 2018 01:45:44 GMT):
>plenum Error: Traceback (most recent call last): File "/home/user/projects/indy-plenum/scripts/plenum", line 58, in run_cli() File "/home/user/projects/indy-plenum/scripts/plenum", line 41, in run_cli nodeReg = config.nodeReg AttributeError: module 'indy_config.py' has no attribute 'nodeReg'

baconsandwich (Fri, 17 Aug 2018 06:31:33 GMT):
This [table](https://docs.google.com/spreadsheets/d/1TWXF7NtBjSOaUIBeIH77SyZnawfo91cJ_ns4TR-wsq4/edit#gid=0) uses the abbreviations MGL, TGB and TF. What do they stand for?

simonmullaney (Fri, 17 Aug 2018 10:00:31 GMT):
Has joined the channel.

vtech (Fri, 17 Aug 2018 10:54:27 GMT):
Hi, is there any way to use Indy as login management for fabric ?

vtech (Fri, 17 Aug 2018 10:54:27 GMT):
Is communication between indy & fabric is possible ?

simonmullaney (Fri, 17 Aug 2018 11:07:39 GMT):
Hey guys just wondering if anyone here has any suggestions on what is the best option for setting up decentralised login functionality to my blockchain business network (business network setup on hyperledger composer which interacts with Hyperledger Fabric. My business network is then exposed as a REST server which my angular application interacts with.) I know of Uport & Hyperledger Indy but was just wondering about pros and cons of both platforms and how easy it is to integrate a solution into your stack. Any help is greatly appreciated

tomislav (Fri, 17 Aug 2018 12:28:00 GMT):
@simonmullaney Indy is a great fit for decentralized auth, biggest advantage over uport is the verifiable claims support provided by the SDK. However, project is still in incubation and there aren't many (if any) production ready solutions for this, such as uport. I am involved in one project that will tackle this, I'm happy to share some details with you if you're interested.

tomislav (Fri, 17 Aug 2018 12:28:52 GMT):
@vtech You can do DID Auth using Indy, essentially it comes down to proving an ownership over certain DID, which is super easy to implement with Indy

simonmullaney (Fri, 17 Aug 2018 12:32:45 GMT):
@tomislav absolutely if you could that would be great!

baconsandwich (Fri, 17 Aug 2018 13:08:30 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=XNL2N883XiHCYbzui) I found TGB=Technical Governance Board but the other two are still a mystery.

swcurran (Fri, 17 Aug 2018 13:17:11 GMT):
TF is trust framework.

Antonio_M (Fri, 17 Aug 2018 13:47:07 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=YhPr73Qt7MQeE6hco) @tomislav I would also greatly appreciate those details, as I am working on a Composer/Fabric POC app and would love to use Indy, so I don't mind if it's not production ready

baconsandwich (Fri, 17 Aug 2018 13:50:04 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=dcHeNyDYkN2wid74Y) @swcurran Thanks!

Sean_Bohan (Fri, 17 Aug 2018 16:50:21 GMT):
Awesome: @peacekeeper is profiled in the Hyperledger Developer Showcase! https://www.hyperledger.org/blog/2018/08/15/developer-showcase-series-markus-sabadello-danube-tech

stanleyc-trustscience (Fri, 17 Aug 2018 19:24:16 GMT):
Just want to make sure one thing, when using predicate in proof request, this allow the prover not sending over the actual value of the credential correct? Prover is just sending over the proof that he has a credential that satisfies the predicate, and sending over not the actual credential

stanleyc-trustscience (Fri, 17 Aug 2018 19:24:16 GMT):
Just want to make sure one thing, when using predicate in proof request, this allow the prover not sending over the actual value of the credential correct? Prover is just sending over the proof that he has a credential that satisfies the predicate, and not sending over the actual credential

swcurran (Fri, 17 Aug 2018 19:57:17 GMT):
@stanleyc-trustscience - correct. You should probably use the term "claim" in that phrase. You receive claims in a credential, you prove claims in a proof. If you use a predicate on a claim, you will get a proof of the claim meeting the predicate, but not the value.

stanleyc-trustscience (Fri, 17 Aug 2018 20:00:56 GMT):
@swcurran Nice, thank you! this is exciting :)

stanleyc-trustscience (Fri, 17 Aug 2018 20:04:02 GMT):
On a side note, I do see proof, credential, and claim being thrown around a bit .... now that you've clarify credential and claim, not sure about proof now lol

swcurran (Fri, 17 Aug 2018 20:31:14 GMT):
Early on claim and credential were used interchangeably (such that the W3C team call their group "Verifiable Claims" and couldn't change it later), but the terms are at different levels - a Credential contain Claims. On the Proof side, you prove claims - at least in Indy. Some other systems only support proving Credentials, which means you lose selective disclosure (only disclosing a subset of claims in a Credential) and ZKPs - e.g. the use of predicates. So in Indy, a Proof involves proving a set of claims from one or more Credentials.

waleed (Sat, 18 Aug 2018 14:45:53 GMT):
Has joined the channel.

waleed (Sat, 18 Aug 2018 15:02:45 GMT):
Hi Everyone :) I am new to Indy. I have Fabric/composer background. I would like to know how to start with Indy, etc...

Alexi (Mon, 20 Aug 2018 16:58:55 GMT):
hey all, I have bee struggling since a recent update in the wallet API, the wrappers have reflected the change in that when creating a wallet it only takes two arguments. However when I try to create a wallet as such I keep getting type error stating I am missing 3 arguments (this pointing to before when the create wallet method took in 5 arguments) however adding the missing arguments leads to and invalid parameter error. according to me I have everything up to date, so I am unsure why I am getting this error. Anyone have any suggestions? here is the full error after running this code: `await wallet.create_wallet(wallet_config, wallet_credentials)` ``` Traceback (most recent call last): File "write_did.py", line 173, in main() File "write_did.py", line 168, in main loop.run_until_complete(write_nym_and_query_verkey()) File "/usr/lib/python3.5/asyncio/base_events.py", line 387, in run_until_complete return future.result() File "/usr/lib/python3.5/asyncio/futures.py", line 274, in result raise self._exception File "/usr/lib/python3.5/asyncio/tasks.py", line 239, in _step result = coro.send(None) File "write_did.py", line 57, in write_nym_and_query_verkey await wallet.create_wallet(wallet_config, wallet_credentials) TypeError: create_wallet() missing 3 required positional arguments: 'xtype', 'config', and 'credentials' ```

baconsandwich (Mon, 20 Aug 2018 19:46:20 GMT):
did you update indy-sdk and libindy?

Alexi (Mon, 20 Aug 2018 19:52:41 GMT):
@baconsandwich I tried uninstalling and rebuilding everything from scratch and got the same error

baconsandwich (Mon, 20 Aug 2018 20:00:20 GMT):
the python code seems still to be the old one, i.e. indy-sdk. if you are using an IDE try to jump to the function definition to see which file it is

Alexi (Mon, 20 Aug 2018 20:30:45 GMT):
the python wrapper in my indy-sdk project is up to date with the repo, in it the method only takes in two parameters: ``` async def create_wallet(config: str, credentials: str) -> None: ```

Alexi (Mon, 20 Aug 2018 20:46:46 GMT):
ok so I figured out what the problem was, I updated everything but `.../.local/lib/python3.5/site-packages/indy` still had the old wrappers saved so I just copied the updated ones from indy-sdk into my sitepackages and this solved my problem. @baconsandwich thanks, your message made me stop and think that python was probably not looking at the correct libraries.

baconsandwich (Mon, 20 Aug 2018 21:35:05 GMT):
@Alexi Glad I could help

josh.hill (Tue, 21 Aug 2018 05:15:18 GMT):
Has joined the channel.

Smit95shah (Tue, 21 Aug 2018 05:26:26 GMT):
Has joined the channel.

Smit95shah (Tue, 21 Aug 2018 05:44:48 GMT):
HI, What is this group all about?

Sean_Bohan (Tue, 21 Aug 2018 15:19:16 GMT):
Hi @Smit95shah this is the main rocketchat channel for Hyperledger Indy - one of the projects in Hyperledger. Indy is a blockchain purpose-built for identity. You can find out more here or at hyperledger.org https://wiki.hyperledger.org/projects/indy

Alexi (Tue, 21 Aug 2018 19:49:14 GMT):
hey all, does anyone know the format of the json which is returned by: ``` async def list_pools() -> None: """ Lists names of created pool ledgers :return: Error code """ logger = logging.getLogger(__name__) logger.debug("list_pools: >>> ") if not hasattr(list_pools, "cb"): logger.debug("list_pools: Creating callback") list_pools.cb = create_cb(CFUNCTYPE(None, c_int32, c_int32, c_char_p)) res = await do_call('indy_list_pools', list_pools.cb) res = json.loads(res.decode()) logger.debug("list_pools: <<< res: %r", res) return res ``` in other words, how can I iterate through the list of pools?

Alexi (Tue, 21 Aug 2018 19:49:14 GMT):
hey all, does anyone know the format of the json which is returned by: ``` async def list_pools() -> None: """ Lists names of created pool ledgers :return: Error code """ logger = logging.getLogger(__name__) logger.debug("list_pools: >>> ") if not hasattr(list_pools, "cb"): logger.debug("list_pools: Creating callback") list_pools.cb = create_cb(CFUNCTYPE(None, c_int32, c_int32, c_char_p)) res = await do_call('indy_list_pools', list_pools.cb) res = json.loads(res.decode()) logger.debug("list_pools: <<< res: %r", res) return res ```

Alexi (Tue, 21 Aug 2018 19:49:14 GMT):
hey all, does anyone know the format of the json which is returned by: ``` async def list_pools() -> None: """ Lists names of created pool ledgers :return: Error code """ logger = logging.getLogger(__name__) logger.debug("list_pools: >>> ") if not hasattr(list_pools, "cb"): logger.debug("list_pools: Creating callback") list_pools.cb = create_cb(CFUNCTYPE(None, c_int32, c_int32, c_char_p)) res = await do_call('indy_list_pools', list_pools.cb) res = json.loads(res.decode()) logger.debug("list_pools: <<< res: %r", res) return res ```

tomislav (Tue, 21 Aug 2018 20:12:28 GMT):
`[{\"pool\":\"pool_name1\"},{\"pool\":\"pool_name2\"},{\"pool\":\"pool_name3\"}]`

Alexi (Tue, 21 Aug 2018 20:20:18 GMT):
@tomislav thanks!

claudiobizzotto (Tue, 21 Aug 2018 20:25:50 GMT):
Has joined the channel.

Alexi (Tue, 21 Aug 2018 22:07:03 GMT):
hey all, had some more questions out of curiosity to see how I can work with indy for the solution I am building -now that wallet methods do not require pool name to work, will the wallet methods simply work in the context of the last 'opened' pool? How would you go about manipulating wallets in different pools? -is there a way to list out the wallets in a pool? -will closing/deleting a pool, by default close/delete the wallets associated with it?

Alexi (Tue, 21 Aug 2018 22:07:03 GMT):
hey all, had some more questions out of curiosity to see how I can work with indy for the solution I am building -now that wallet methods do not require pool name to work, will the wallet methods simply work in the context of the last 'opened' pool? How would you go about manipulating wallets in different pools? -is there a way to list out the wallets in a pool? -will closing/deleting a pool, by default close/delete the wallets associated with it?

bdjidi (Tue, 21 Aug 2018 22:51:08 GMT):
Has joined the channel.

gaurisankarj (Thu, 23 Aug 2018 07:38:42 GMT):
Has joined the channel.

gaurisankarj (Thu, 23 Aug 2018 07:38:52 GMT):
$ vagrant ssh agent01 vagrant@agent01:~$ python /usr/local/lib/python3.5/dist-packages/indy_client/test/agent/faber.py --port 5555 --network

gaurisankarj (Thu, 23 Aug 2018 07:39:24 GMT):
Is it your network name that has to be added?

gaurisankarj (Thu, 23 Aug 2018 07:39:54 GMT):
Or is it the Indy Network that we are setting up?

gaurisankarj (Thu, 23 Aug 2018 08:49:31 GMT):
Can somebody explain how to do this : Set Up Cluster of Indy Validator Nodes?

bricg (Thu, 23 Aug 2018 13:22:05 GMT):
Has joined the channel.

bricg (Thu, 23 Aug 2018 13:25:16 GMT):
I am following the getting started tutorial on a mac and I have got as far as running the local pool. How can I check that is is running? Also, can someone explain exactly how I do this ```add the path containing the library the LD_LIBRARY_PATH and DYLD_LIBRARY_PATH environment variables. ```

bricg (Thu, 23 Aug 2018 13:25:16 GMT):
I am following the getting started tutorial on a mac and I have got as far as running the local pool. How can I check that is is running? Also, can someone tell me what path i should use for this ```add the path containing the library the LD_LIBRARY_PATH and DYLD_LIBRARY_PATH environment variables. ```

Hasse (Thu, 23 Aug 2018 15:23:55 GMT):
Has joined the channel.

Alexi (Thu, 23 Aug 2018 20:16:17 GMT):
are DID and Verkeys always the same lenght\

Alexi (Thu, 23 Aug 2018 20:16:17 GMT):
are DID and Verkeys always the same length?

canadaduane (Thu, 23 Aug 2018 21:22:35 GMT):
@bricg I downloaded `indy-sdk` to a directory on my machine `/Users/duane/Mainframe/indy-sdk` so after compiling the debug target I set LD_LIBRARY_PATH to `/Users/duane/Mainframe/indy-sdk/libindy/target/debug/`

canadaduane (Thu, 23 Aug 2018 21:23:11 GMT):
That seems to have worked, as I was then able to compile the `cli`

canadaduane (Thu, 23 Aug 2018 21:23:32 GMT):
I don't know how to check if the local pool is running

microinclude (Thu, 23 Aug 2018 21:31:10 GMT):
@bricg im using DYLD_LIBRARY_PATH env

microinclude (Thu, 23 Aug 2018 21:31:50 GMT):
If you are running the local pool on docker, you can see the docker logs

microinclude (Thu, 23 Aug 2018 21:32:53 GMT):
also, don't forget to run indy node start and as well as mount the genesis files you generated.

bricg (Fri, 24 Aug 2018 09:19:09 GMT):
@microinclude @canadaduane thanks for your suggestions. I've now discovered the getting-started demo so I have decided to try it instead. I'm following the steps indicated here - https://github.com/hyperledger/indy-sdk/blob/master/doc/getting-started/Trouble_shoot_GSG.md ; but I am getting the following error ```Step 4/22 : RUN pip3 install -U pip==9.0.3 setuptools ---> Running in a5564edbb3fc Collecting pip==9.0.3 Exception: Traceback (most recent call last): File "/usr/share/python-wheels/urllib3-1.13.1-py2.py3-none-any.whl/urllib3/connectionpool.py", line 346, in _make_request self._validate_conn(conn) File "/usr/share/python-wheels/urllib3-1.13.1-py2.py3-none-any.whl/urllib3/connectionpool.py", line 787, in _validate_conn conn.connect() File "/usr/share/python-wheels/urllib3-1.13.1-py2.py3-none-any.whl/urllib3/connection.py", line 252, in connect ssl_version=resolved_ssl_version) File "/usr/share/python-wheels/urllib3-1.13.1-py2.py3-none-any.whl/urllib3/util/ssl_.py", line 305, in ssl_wrap_socket return context.wrap_socket(sock, server_hostname=server_hostname) File "/usr/lib/python3.5/ssl.py", line 377, in wrap_socket _context=self) File "/usr/lib/python3.5/ssl.py", line 752, in __init__ self.do_handshake() File "/usr/lib/python3.5/ssl.py", line 988, in do_handshake self._sslobj.do_handshake() File "/usr/lib/python3.5/ssl.py", line 633, in do_handshake self._sslobj.do_handshake() socket.timeout: _ssl.c:629: The handshake operation timed out```. Has anyone any suggestions?

bricg (Fri, 24 Aug 2018 09:19:09 GMT):
@microinclude @canadaduane thanks for your suggestions. I've now discovered the getting-started demo so I have decided to try it instead.

bricg (Fri, 24 Aug 2018 14:09:46 GMT):
I have been going through the getting started demo and I think its a great example use case to demo what SSI is all about. The main thing that is not clear to me however is where the data is actually stored during each stage. Are there any documents out there that might help me?

Sean_Bohan (Fri, 24 Aug 2018 18:55:34 GMT):
Hyperledger Indy WG Call for Thurs 8/23/2018 Recap Agenda: https://drive.google.com/open?id=1NmFASN0kNmbgK9_yOC0F5I8Onb_hKoj5rTHZXryJfNE Video: https://drive.google.com/open?id=1dOKajuJWdUepvh-c8-K4P-eDVlNWCG5X

InfoMiner (Fri, 24 Aug 2018 19:37:08 GMT):
Has joined the channel.

xum7331 (Sat, 25 Aug 2018 02:38:28 GMT):
Has joined the channel.

baconsandwich (Mon, 27 Aug 2018 09:11:27 GMT):
What is meant by "walleted" entity? This term is used here for example: https://github.com/hyperledger/indy-sdk/blob/677a0439487a1b7ce64c2e62671ed3e0079cc11f/doc/design/002-anoncreds/anoncreds-workflow.svg The diagram shows a normal issuer and a walleted issuer, a normal prover and a walleted prover. Is a walleted X the wallet of X?

ashcherbakov (Tue, 28 Aug 2018 10:54:54 GMT):
@baconsandwich It means an Entity having access to the Wallet. For example, one can consider a use case where Issuer consist of two parts: internal (secure) one, which has access to the wallet, but doesn't have access to the network, and external one which doesn't have access to secure part (wallet, private keys, etc.) but has access to the network. So, the current API supports separation of dealing with secrets and sending/exchanging data.

tomislav (Tue, 28 Aug 2018 14:09:54 GMT):
Yes

swcurran (Tue, 28 Aug 2018 14:52:24 GMT):
@ashcherbakov - is that supported today with the current ledger and libindy? Since all transactions to the ledger have to be signed (I think, as a last step of constructing the transaction), doesn't that mean the entity that talks to the ledger must have access to a private key? Are there any examples in the test cases of this being used?

ashcherbakov (Tue, 28 Aug 2018 14:54:20 GMT):
I believe signing and sending are separate operations, there is a call `indy_sign` and a call `indy_submit`, so one party can sign it, and another can submit/send it. @gudkov?

brycebudd (Tue, 28 Aug 2018 15:00:06 GMT):
Has joined the channel.

gudkov (Tue, 28 Aug 2018 16:32:50 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=y4ddurxPcm75KzgxR) @ashcherbakov Yes sign and submit are present as separate actions in indy sdk

swcurran (Tue, 28 Aug 2018 18:55:04 GMT):
@ashcherbakov @gudkov - we're trying to come up with a way to launch our app without waiting for the next SLN upgrade, but we need to be on a 1.6 SDK. Do you think we could take an Agent running on the 1.6 SDK (python), and when sending to the ledger, reorganize the message to match the current Ledger before signing and sending the message (and vice versa for reads)? We would need to do that for a handful of DID, schema and CredDef creation. Doable, impossible or worth trying because it might work?

szewong (Tue, 28 Aug 2018 20:44:03 GMT):
Has joined the channel.

gudkov (Wed, 29 Aug 2018 08:28:22 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=SCJgTLFhfuYTRkR3f) @swcurran Indy SDK 1.6 support both Node protocols. You can use it to send requests to Node with protocol version 1 and to Node with protocol version 2. There is a call set_protocol_version. The only problem for you is that your agent app should support both protocol versions, but i am not sure that it is complex as you need to use only few transactions.

gudkov (Wed, 29 Aug 2018 08:36:12 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=DPRYcmnyy2a7QHTnY) Indy SDK already provides simple abstraction layer over node protocol. For your request there is build method and corresponded parse method that will provide the same results independent on node protocol version. Most probably if your app uses methods like indy_build_get_schema_request/indy_parse_get_schema_response you are already ok and just need to write few lines of code to switch between protocol versions.

gudkov (Wed, 29 Aug 2018 08:36:12 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=DPRYcmnyy2a7QHTnY) Indy SDK already provides simple abstraction layer over node protocol. For each request there is build method and corresponded parse method that will provide the same results independent on node protocol version. Most probably if your app uses methods like indy_build_get_schema_request/indy_parse_get_schema_response you are already ok and just need to write few lines of code to switch between protocol versions.

gut (Wed, 29 Aug 2018 10:08:33 GMT):
Has joined the channel.

RubenLassau-Strauven (Wed, 29 Aug 2018 12:34:55 GMT):
Hey all

RubenLassau-Strauven (Wed, 29 Aug 2018 12:42:27 GMT):
Hey all I'm currently experimenting with Indy and some things aren't clear. I'm trying to build an app with Indy in Node.js and i'm trying to create a credential offer. The problem I have is how do I get the credential definition ID for the credential definition I want to use for this offer? In most tutorials these ID's are available from previous actions, but how am I supposed to retrieve them when they are not locally saved? The only way I see how to do this is to send the buildSchemaRequest again, and then use the ID from the result to recreate the credential definition. But this seems to be way too much work to do every time a credential offer has to be made for the same credential definition. Am I misunderstanding the use of schemas and credential definitions? Can someone provide some clarity?

RubenLassau-Strauven (Wed, 29 Aug 2018 12:42:27 GMT):
Hey all I'm currently experimenting with Indy and some things aren't clear. I'm trying to build an app with Indy in Node.js and i'm trying to create a credential offer. The problem I have is how do I get the credential definition ID for the credential definition I want to use for this offer? In most tutorials these ID's are available from previous actions, but how am I supposed to retrieve them when they are not locally saved? The only way I see how to do this is to send the buildSchemaRequest again, and then use the ID from the result to recreate the credential definition. But this seems to be way too much work to do every time a credential offer has to be made for the same credential definition. Am I misunderstanding the use of schemas and credential definitions? Can someone provide some clarity? Thanks in advance!

RubenLassau-Strauven (Wed, 29 Aug 2018 12:42:27 GMT):
Hey all I'm currently experimenting with Indy and some things aren't clear. I'm trying to build an app with Indy in Node.js and i'm trying to create a credential offer. The problem I have is how do I get the credential definition ID for the credential definition I want to use for this offer? In most tutorials these ID's are available from previous actions, but how am I supposed to retrieve them when they are not locally saved? The only way I see how to do this is to send the buildSchemaRequest again, and then use the ID from the result to recreate the credential definition. But this seems to be way too much work to do every time a credential offer has to be made for the same credential definition. Am I misunderstanding the use of schemas and credential definitions? Can someone provide some clarity? Thanks in advance! Edit: I'm using indy-sdk v1.6.2

tomislav (Wed, 29 Aug 2018 14:10:21 GMT):
@RubenLassau-Strauven You will need to persist the credential definition id you have created and use it later. You can use the non_secrets API to do this, or your own storage.

RubenLassau-Strauven (Wed, 29 Aug 2018 14:14:11 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=Z8mouCgFBy76z76qS) @tomislav Alright, thanks!

gskerry (Wed, 29 Aug 2018 15:54:47 GMT):
Has joined the channel.

RubenLassau-Strauven (Thu, 30 Aug 2018 07:51:03 GMT):
Can someone also EiZDAJ3M4HnQRYovwkamgx:2:firstName:2.09 EiZDAJ3M4HnQRYWvwkamgx:3:CL:5459

RubenLassau-Strauven (Thu, 30 Aug 2018 07:54:45 GMT):
Schema ID's and Credential definition ID's are made up of several parts. Can someone give more explanation on what these parts are and what they are used for? Thanks. Schema ID: EiZ9AJiM4HnxRYWvwkamgx:2:firstName:2.09 Cred def ID: EiZ9AJiM4HnxRYWvwkamgx:3:CL:5459

huxiangdong (Thu, 30 Aug 2018 09:02:48 GMT):
Has joined the channel.

st (Thu, 30 Aug 2018 09:16:13 GMT):
Has joined the channel.

ashcherbakov (Thu, 30 Aug 2018 10:47:59 GMT):
@RubenLassau-Strauven This is a unique key of the entity in State Trie (Patricia State Trie)

RubenLassau-Strauven (Thu, 30 Aug 2018 11:20:47 GMT):
@ashcherbakov Yes, but it is comprised of multiple parts separated by colons, right?

ashcherbakov (Thu, 30 Aug 2018 11:22:57 GMT):
Yes, because the key is organized this way. You can have a look at https://github.com/hyperledger/indy-node/blob/master/indy_common/state/domain.py for more details

xxjoshuaxx (Thu, 30 Aug 2018 13:32:52 GMT):
Has joined the channel.

xxjoshuaxx (Thu, 30 Aug 2018 13:33:19 GMT):
I am trying out Plenum, I keep getting "Not connected: Beta, Alpha, Gamma" and unable to create clients.. Why would this be?

trthhrtz (Fri, 31 Aug 2018 04:25:17 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=LC4M2wxMfwR7NqkmQ) @Sean_Bohan Hi! Is there a recap for 8/30/2018?

sebastian (Fri, 31 Aug 2018 12:33:07 GMT):
Has joined the channel.

sureshtedla (Fri, 31 Aug 2018 13:43:31 GMT):
Has joined the channel.

Tulika8 (Sat, 01 Sep 2018 10:12:40 GMT):
Has joined the channel.

Tulika8 (Sat, 01 Sep 2018 10:16:23 GMT):
Hi Everyone I just joined this chat! I was trying out hyperledger indy.. read and got https://github.com/hyperledger/indy-sdk/blob/master/doc/getting-started/getting-started.md this example up on my local indy test network.. However the code of this example (https://github.com/hyperledger/indy-sdk/blob/master/samples/python/src/getting_started.py) does not have any NYM transaction's code (ledger.sign_and_submit_request), then how are did and other transactions getting added in the indy blockchain? How can I see the current state of my indy blockchian? Any cli commands to see the list of trasaction etc?

Tulika8 (Sat, 01 Sep 2018 10:16:23 GMT):
Can someone please help me understand the basic concept of Indy. In the getting started example is it like ..Faber college and Acme maintain there own identity stores where they maintain the details of their users? Alice, Acme and Faber have DID's in the blockchain and what ever trasactions are happening between them are recorded in the blockchain that can be later used for proof?They also have public keys published in the blockchain that they can use later for proof? When Alice makes a connection request with Faber they register there DID in the blockchain, but the transcript is passed on to Alice through backchannel by encypting it with Alice public key ?

Tulika8 (Sat, 01 Sep 2018 10:16:23 GMT):
I ran sample java program Annoncreds.java and have put a sleep of 5 mins in the end of this class before it deletes all walets etc. Before it deletes these wallets , I run wallet list command on indy-cli and I get a message that no open wallet present. PLeas elet me know why its so? This command should list the wallets opened or dids in created in the AnonCreds,java?

Tulika8 (Sat, 01 Sep 2018 10:35:37 GMT):
Can someone please help me understand the basic concept of Indy. In the getting started exmaple is it like ..Faber college and Acme maintain there own identity stores where they maintain the details of their users? Alice, Acme and Faber have DID's in the blockchain and what ever trasactions are happening between them are recorded in the blockchain that can be later used for proof?They also have public keys published in the blockchain that they can use later for proof? When Alice makes a connection request with Faber they register there DID in the blockchain, but the transcript is passed on to Alice through backchannel by encypting it with Alice public key ?

Tulika8 (Sat, 01 Sep 2018 11:49:41 GMT):
I ran sample java program Annoncreds.java and have put a sleep of 5 mins in the end of this class before it deletes all walets etc. Before it deletes these wallets , I run wallet list command on indy-cli and I get a message that no open wallet present. PLeas elet me know why its so? This command should list the wallets opened or dids in created in the AnonCreds,java?

SherifMuhammed (Sun, 02 Sep 2018 15:07:38 GMT):
Hi , Is it possible to use Indy as an Authenticator ?

PatrickDLT (Sun, 02 Sep 2018 21:01:38 GMT):
Has joined the channel.

PatrickDLT (Sun, 02 Sep 2018 21:06:41 GMT):
Hi :) I'm new here and my job is to create a projekt for university which is based on Hyperledger Indy but I'm struggel with setting up HL Indy. Why does some Links not work on GitHub reffering to Indy SDK (https://github.com/hyperledger/indy-sdk)?

abraham (Mon, 03 Sep 2018 05:17:45 GMT):
Hi Daniel Hardman , I am planning to talk about Indy project during technical conference organized by csdn, which is biggest developers community in China. Can I use your presentation materials which you talked about "decentralized indentiy, distilled" by webinair on June,2018 ? thanks a lot

AbhinayB (Mon, 03 Sep 2018 06:27:03 GMT):
Has joined the channel.

AbhinayB (Mon, 03 Sep 2018 10:07:42 GMT):
Hi everyone! I am a beginner to Indy and had a look at its principle at https://github.com/hyperledger/indy-sdk/blob/rc/doc/getting-started/getting-started.md I have a doubt. Suppose in the Alice example, if Alice had to store her picture as part of her credential, is it possible in Indy? Can we store an image as a credential in the identity record?

AbhinayB (Mon, 03 Sep 2018 10:07:42 GMT):
Hi everyone! I am a beginner to Indy and had a look at its principle at https://github.com/hyperledger/indy-sdk/blob/rc/doc/getting-started/getting-started.md I have a doubt. Suppose in the Alice example, if Alice had to store her picture as part of her credential, is it possible in Indy? Can we store an image as a credential in the identity record?

AbhinayB (Mon, 03 Sep 2018 10:07:42 GMT):
Hi everyone! I am a beginner to Indy and had a look at its principle at https://github.com/hyperledger/indy-sdk/blob/rc/doc/getting-started/getting-started.md I have a doubt. Suppose in the Alice example, if Alice had to store her picture as part of her credential, is it possible in Indy? Can we store an image as a credential in the identity record? Also, Is it possible to have a cross-ledger identity management between hyperledger fabric and indy?

nakul (Mon, 03 Sep 2018 16:27:48 GMT):
Has joined the channel.

InfoMiner (Tue, 04 Sep 2018 01:03:18 GMT):
to whomever it may concern: on this page https://wiki.hyperledger.org/projects/indy/documentation I find a listing for the sovrin technical yellow paper https://drive.google.com/file/d/1zZv9-jd5skno3rUKwAW2Z621B6rztAjy/view but this is private.. if it's supposed to be private, ok, but why put it on the wiki?

tnr@womeninlinux.com (Tue, 04 Sep 2018 06:42:26 GMT):
Has joined the channel.

zerppen (Tue, 04 Sep 2018 08:32:52 GMT):
Has joined the channel.

tkdp (Tue, 04 Sep 2018 09:13:33 GMT):
Has joined the channel.

voddan (Tue, 04 Sep 2018 13:10:26 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=Xws858SmQDcnh3MBH) @tomislav That makes zero sense to me

RyanWest (Tue, 04 Sep 2018 16:01:34 GMT):
Has joined the channel.

SRibeiro (Wed, 05 Sep 2018 12:55:16 GMT):
Has joined the channel.

canadaduane (Wed, 05 Sep 2018 14:39:08 GMT):
How is a verkey created? Is there somewhere I can go to learn about this?

canadaduane (Wed, 05 Sep 2018 14:39:47 GMT):
I'm interested in understanding what it is and how a DID + something else is used to generate it.

Dubh3124 (Thu, 06 Sep 2018 01:00:46 GMT):
Has joined the channel.

trthhrtz (Thu, 06 Sep 2018 03:26:59 GMT):
@canadaduane take a look at the INDY-SDK getting started guide (https://github.com/hyperledger/indy-sdk/blob/master/doc/getting-started/getting-started.md) It describes briefly DID/verkey and Python usage of them

fhmarino (Thu, 06 Sep 2018 13:35:08 GMT):
Has joined the channel.

canadaduane (Thu, 06 Sep 2018 13:58:36 GMT):
@trthhrtz That's actually what I've been reading, but it isn't clear to me how a verkey is created. It talks about verkeys in "abbreviated form" in one section, and it introduces verkeys as "The most important fields of this transaction are dest (target DID), role (role of a user NYM record being created for) and the verkey (target verification key)." but doesn't explain what exactly a verkey is. By analogy, I've seen verkey described as the "public key" in pub/priv key pair. But I'd like to see more detail: like, is it the DID concatenated with a random string?

canadaduane (Thu, 06 Sep 2018 13:58:36 GMT):
@trthhrtz That's actually what I've been reading, but it isn't clear to me how a verkey is created. It talks about verkeys in "abbreviated form" in one section, and it introduces verkeys when explaining "The most important fields of this transaction are dest (target DID), role (role of a user NYM record being created for) and the verkey (target verification key)." but doesn't explain what exactly a verkey is. By analogy, I've seen verkey described as the "public key" in pub/priv key pair. But I'd like to see more detail: like, is it the DID concatenated with a random string?

canadaduane (Thu, 06 Sep 2018 13:58:36 GMT):
@trthhrtz That's actually what I've been reading, but it isn't clear to me how a verkey is created. It talks about verkeys in "abbreviated form" in one section, and it introduces verkeys when explaining "The most important fields of this transaction are dest (target DID), role (role of a user NYM record being created for) and the verkey (target verification key)." but doesn't explain what exactly a verkey is. I've seen verkey described by analogy as the "public key" in a public/private keypair. But I'd like to see more detail: like, is it the DID concatenated with a random string?

canadaduane (Thu, 06 Sep 2018 14:15:22 GMT):
Another question: I've been listening to Drummond Reed present about DPKS on ssimeetup.org and in the section about social recovery of private keys, he talks about "trustees" as the preferred name for friends or family who are willing to provide this service for you. Is this "trustee" different from the role of "trustee" in Indy? e.g. "Roles.TRUSTEE" here: https://github.com/hyperledger/indy-node/blob/master/indy_common/roles.py

canadaduane (Thu, 06 Sep 2018 14:15:22 GMT):
Another question: I've been listening to Drummond Reed present about DKMS on ssimeetup.org and in the section about social recovery of private keys, he talks about "trustees" as the preferred name for friends or family who are willing to provide this service for you. Is this "trustee" different from the role of "trustee" in Indy? e.g. "Roles.TRUSTEE" here: https://github.com/hyperledger/indy-node/blob/master/indy_common/roles.py

Sean_Bohan (Thu, 06 Sep 2018 14:56:24 GMT):
This week's Indy WG Call agenda: Housekeeping AGENTS! Upcoming Thurs WG Call Agendas (SeanB) Enterprise Wallet Use Cases (RichardE) DIF Hubs and Indy Agents (All) Open Discussion Happening in 5 minutes: https://zoom.us/j/232861185

mighty-pirate (Thu, 06 Sep 2018 17:55:27 GMT):
Has joined the channel.

swcurran (Thu, 06 Sep 2018 18:11:56 GMT):
I was wondering who is in charge of the Indy Project page on hyperledger.org? https://www.hyperledger.org/projects/hyperledger-indy I had some thoughts on this page - at least one bug to report, and some suggestions.

InfoMiner (Thu, 06 Sep 2018 22:26:23 GMT):
indy's documentation doesn't mention the word "modular" as much as fabric and sawtooth. is that still an important focus of its design?

shrimanwar92 (Fri, 07 Sep 2018 07:00:05 GMT):
Has joined the channel.

mawi (Fri, 07 Sep 2018 09:23:24 GMT):
Has left the channel.

Antonio_M (Fri, 07 Sep 2018 11:39:33 GMT):
hi guys, anyone have an example of how to encode strings to int as seen here https://github.com/hyperledger/indy-sdk/blob/aa43653319d8ec4d2ffd83c2499f78e82a9897ea/wrappers/java/src/test/java/org/hyperledger/indy/sdk/anoncreds/AnoncredsIntegrationTest.java#L51

canadaduane (Fri, 07 Sep 2018 14:49:18 GMT):
Once a connection is established between two identities, where are those connections stored?

canadaduane (Fri, 07 Sep 2018 14:49:33 GMT):
on-chain? locally on device?

MALodder (Fri, 07 Sep 2018 17:49:40 GMT):
locally on device

esplinr (Fri, 07 Sep 2018 17:55:04 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=epzZ4s6eMGkPSPHTN) @swcurran I believe that both @tkuhrt and @rjones can make improvements to that page, but the Hyperledger marketing team owns it.

rjones (Fri, 07 Sep 2018 17:55:06 GMT):
Has joined the channel.

esplinr (Fri, 07 Sep 2018 17:56:12 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=WGPipTXBJuGZojW5g) @InfoMiner Indy Node is not very modular today. We have discussed adopting a more service oriented architecture in the future, but I personally still question the benefits of that approach over what we are doing now.

esplinr (Fri, 07 Sep 2018 17:57:29 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=PGxQXwj8BAXLjyNN5) @canadaduane Your trustees for key recover are different from the trustees of an Indy Network. The overlap in terminology has many of us using the term "recovery buddy" for social key recovery.

esplinr (Fri, 07 Sep 2018 18:00:23 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=69gwvyS5Po8tX4Sgb) @abraham @abraham In the future you should reference someone's handle if you want them to respond individually. It can be hard to keep up with all the traffic in a busy channel like this. I work with @danielhardman . Most of our public presentation materials are under a CC-BY license. I would be surprised if Daniel has a concern with you reusing them, but he'll have to respond if there isn't already a license specified in the deck.

esplinr (Fri, 07 Sep 2018 18:00:23 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=69gwvyS5Po8tX4Sgb) @abraham In the future you should reference someone's handle if you want them to respond individually. It can be hard to keep up with all the traffic in a busy channel like this. I work with @danielhardman . Most of our public presentation materials are under a CC-BY license. I would be surprised if Daniel has a concern with you reusing them, but he'll have to respond if there isn't already a license specified in the deck.

canadaduane (Fri, 07 Sep 2018 19:30:25 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=gtHBQfyokcET2HdpE) @MALodder Thanks

canadaduane (Fri, 07 Sep 2018 20:16:39 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=uQYeJdRymJgHrmG82) To answer my own question, I *think* the way a verkey is generated is it's just a randomly generated public key (32 bytes long?). The first few bytes (16?) are the DID.

Antonio_M (Sat, 08 Sep 2018 15:35:07 GMT):
does anyone have any idea how to debug this, I'm getting an error with this: ``` var proofRequest = ProofRequest( "proof_req_1", "0.1", "123432421212", mapOf(Pair("attr1_referent", AttributeReferent( "name", arrayListOf( Filter( schemaName = "gwt", schemaVersion = "1.0" ))))), mapOf(Pair("predicate1_referent", PredicateReferent("age", ">=", 18, arrayListOf()) )) ) val proverWallet = walletManager.getWallet(proverWalletId) var getCredentialsForProofResponse = Anoncreds.proverGetCredentialsForProofReq(walletHandle, gson.toJson(proofRequest)).get() println("getCredentialsForProofResponse: " + getCredentialsForProofResponse) ``` It's written in Kotlin so it uses the Java SDK 1.6.1

Antonio_M (Sat, 08 Sep 2018 15:35:07 GMT):
does anyone have any idea how to debug this, I'm getting an WalletStorageException with this: ``` var proofRequest = ProofRequest( "proof_req_1", "0.1", "123432421212", mapOf(Pair("attr1_referent", AttributeReferent( "name", arrayListOf( Filter( schemaName = "gwt", schemaVersion = "1.0" ))))), mapOf(Pair("predicate1_referent", PredicateReferent("age", ">=", 18, arrayListOf()) )) ) val proverWallet = walletManager.getWallet(proverWalletId) var getCredentialsForProofResponse = Anoncreds.proverGetCredentialsForProofReq(walletHandle, gson.toJson(proofRequest)).get() println("getCredentialsForProofResponse: " + getCredentialsForProofResponse) ``` It's written in Kotlin so it uses the Java SDK 1.6.1

Antonio_M (Sat, 08 Sep 2018 15:35:07 GMT):
does anyone have any idea how to debug this, I'm getting an WalletStorageException with this: ``` var proofRequest = ProofRequest( "proof_req_1", "0.1", "123432421212", mapOf(Pair("attr1_referent", AttributeReferent( "name", arrayListOf( Filter( schemaName = "gwt", schemaVersion = "1.0" ))))), mapOf(Pair("predicate1_referent", PredicateReferent("age", ">=", 18, arrayListOf()) )) ) val proverWallet = walletManager.getWallet(proverWalletId) var getCredentialsForProofResponse = Anoncreds.proverGetCredentialsForProofReq(walletHandle, gson.toJson(proofRequest)).get() println("getCredentialsForProofResponse: " + getCredentialsForProofResponse) ``` It's written in Kotlin so it uses the Java SDK 1.6.1 wallet manager.getManager this ``` fun createWallet(id: String, key: String): Wallet { // 3. Creates a new identity wallet val walletConfig = gson.toJson(WalletConfig(id)) val walletCreds = gson.toJson(WalletCredentials(key)) try { Wallet.createWallet(walletConfig, walletCreds).get() } catch (plcee: ExecutionException) { // 3.a. Wallet already exists } // 4. Open identity wallet and get the wallet handle from libindy return Wallet.openWallet(walletConfig, walletCreds) .get() .also { wallets[WalletInfo(id, key)] = it } } ```

Antonio_M (Sat, 08 Sep 2018 15:35:07 GMT):
does anyone have any idea how to debug this, I'm getting an WalletStorageException with this: ``` var proofRequest = ProofRequest( "proof_req_1", "0.1", "123432421212", mapOf(Pair("attr1_referent", AttributeReferent( "name", arrayListOf( Filter( schemaName = "gwt", schemaVersion = "1.0" ))))), mapOf(Pair("predicate1_referent", PredicateReferent("age", ">=", 18, arrayListOf()) )) ) val proverWallet = walletManager.getWallet(proverWalletId) var getCredentialsForProofResponse = Anoncreds.proverGetCredentialsForProofReq(walletHandle, gson.toJson(proofRequest)).get() println("getCredentialsForProofResponse: " + getCredentialsForProofResponse) ``` It's written in Kotlin so it uses the Java SDK 1.6.1

Antonio_M (Sat, 08 Sep 2018 15:35:07 GMT):
does anyone have any idea how to debug this, I'm getting an WalletStorageException (210) with this: ``` var proofRequest = ProofRequest( "proof_req_1", "0.1", "123432421212", mapOf(Pair("attr1_referent", AttributeReferent( "name", arrayListOf( Filter( schemaName = "gwt", schemaVersion = "1.0" ))))), mapOf(Pair("predicate1_referent", PredicateReferent("age", ">=", 18, arrayListOf()) )) ) val proverWallet = walletManager.getWallet(proverWalletId) var getCredentialsForProofResponse = Anoncreds.proverGetCredentialsForProofReq(walletHandle, gson.toJson(proofRequest)).get() println("getCredentialsForProofResponse: " + getCredentialsForProofResponse) ``` It's written in Kotlin so it uses the Java SDK 1.6.1

PhillipGibb (Mon, 10 Sep 2018 07:33:07 GMT):
Please check out the stackoverflow proposal for questions concerning Indy (that would include sovrin, vcx, etc) https://area51.stackexchange.com/proposals/120021/hyperledger-indy?referrer=2Kckofvf6W-eWWrqyLlr9A2 I think that if there are some well defined questions in a contextual and better searchable place we can better develop SSI applications based on indy

phinpope (Mon, 10 Sep 2018 10:26:56 GMT):
Has joined the channel.

phinpope (Mon, 10 Sep 2018 10:28:02 GMT):
Hi, quick q... I installed Indy on a Linode box and it is not being used but it is churning CPU at 20% per node (5 nodes) without any network traffic, is this normal?

Antonio_M (Mon, 10 Sep 2018 18:29:25 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=nqzF8fKzre3qjRSqk) @PhillipGibb Hi, these proposals questions can only be 150 or so characters, which is way to low for a decent question, or I'm doing something wrong?

canadaduane (Mon, 10 Sep 2018 20:07:15 GMT):
How is a new identity created in Sovrin/Indy? For instance, in ethereum's ERC-725, a new smart contract must be deployed (expensive); in uPort, a new ethereum address is created (free). What is the minimum set of things needed in Indy (trust anchor, money, connection to internet?)

canadaduane (Mon, 10 Sep 2018 20:07:15 GMT):
How is a new identity created in Sovrin/Indy? For instance, in ethereum's ERC-725, a new smart contract must be deployed (expensive); in uPort, a new ethereum address is created (free). What is the minimum set of things needed in Indy? (trust anchor, money, connection to internet?)

canadaduane (Mon, 10 Sep 2018 20:07:15 GMT):
How is a new identity created in Sovrin/Indy? For instance, in ethereum's ERC-725, a new smart contract must be deployed (expensive); in uPort, a new ethereum address is created (free, offline). What is the minimum set of things needed in Indy? (trust anchor, money, connection to internet?)

Dubh3124 (Mon, 10 Sep 2018 20:46:58 GMT):
Hi, I've been trying to make an client using the python indy-sdk. I've been following on the How-Tos https://github.com/hyperledger/indy-sdk/tree/master/doc/how-tos. I am able to save the trust anchor DID and steward DID to a wallet, however, I do not see how to retrieve those DIDs from the user's wallet to make transactions (such as rotating keys). Does the DIDs need to be persisted outside of a wallet?

lynn.bendixsen (Mon, 10 Sep 2018 22:46:40 GMT):
Has joined the channel.

PhillipGibb (Tue, 11 Sep 2018 06:19:20 GMT):
I agree, but I think that these are just kick starter question

canadaduane (Tue, 11 Sep 2018 14:00:31 GMT):
Does this comparison chart I'm working on look right?

canadaduane (Tue, 11 Sep 2018 14:00:33 GMT):
https://discourse-cdn-sjc1.com/business4/uploads/mainframe/original/1X/868ecbaced866ba804e2141f69e7542b0cb26871.png

canadaduane (Tue, 11 Sep 2018 14:00:49 GMT):
(from https://community.mainframe.com/t/decentralized-identity-projects-we-should-keep-an-eye-on/17/10)

abraham (Tue, 11 Sep 2018 14:53:57 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=zb6x5SoMPr3RpbuLf) @esplinr thanks for your reply. i did not find his contact here before. Finally we already agreed with using it.

baoyangc (Tue, 11 Sep 2018 17:02:30 GMT):
Has joined the channel.

andrewtan (Wed, 12 Sep 2018 01:35:06 GMT):
As I get to know more about this project, it seems that this project is based on a combination of the following features: 1. DID [Open Standards] 2. JSON Document 3. Zero Knowledge Proof [Though I have yet figured out which Zero Knowledge Proof Lib it is using] 4. Selective Disclosure of Information 5. Underlying Ledger using BFT Am I right to say that excluding point 5, point 1-4 can be replicated and write to other types of Ledger or am I missing out something here. What I can see is that the designated ledger resolves scalability issues (at the point of this project development a few years back) and the Trust Framework that governs any information placed on the network. With Hedera HashGraph launching soon, does it make sense to leverage on that ledger as the underlying ledger given that Hashgraph claims to have solve the throughput and scaling issues?

Dubh3124 (Wed, 12 Sep 2018 07:00:02 GMT):
I am having trouble understanding how to get either the DID or Verkey of a steward after wallet creation. I am using the Indy python sdk

Dubh3124 (Wed, 12 Sep 2018 09:14:13 GMT):
I figured out a way to get the DID from created wallet. When creating a wallet I used set_did_metadata to set some kind of identifier then wrote a function to basically run list_my_dids_with_meta which returns a list of did/metadata json object in which you can parse for the metadata identifier and return the did in the json object. Create wallet function ``` async def create_wallet(self, wallet_config, wallet_credentials): print_log('\n3. Creating new secure wallet\n') try: await wallet.create_wallet(wallet_config, wallet_credentials) wallet_handle = await wallet.open_wallet(wallet_config, wallet_credentials) pool_handle = await pool.open_pool_ledger(config_name=self.pool_name, config=None) steward_did, steward_verkey = await self._generate_steward_DID(wallet_handle) await did.set_did_metadata(wallet_handle, steward_did, json.dumps({'Name': 'Steward'})) ```

Dubh3124 (Wed, 12 Sep 2018 09:14:13 GMT):
I figured out a way to get the DID from created wallet. When creating a wallet I used set_did_metadata to set some kind of identifier then wrote a function to basically run list_my_dids_with_meta which returns a list of did/metadata json object in which you can parse for the metadata identifier and return the did in the json object. Create wallet function ``` async def create_wallet(self, wallet_config, wallet_credentials): print_log('\n3. Creating new secure wallet\n') try: await wallet.create_wallet(wallet_config, wallet_credentials) wallet_handle = await wallet.open_wallet(wallet_config, wallet_credentials) pool_handle = await pool.open_pool_ledger(config_name=self.pool_name, config=None) steward_did, steward_verkey = await self._generate_steward_DID(wallet_handle) await did.set_did_metadata(wallet_handle, steward_did, json.dumps({'Name': 'Steward'})) ``` ``` async def _didfromseed(self, wallet_handle): didlist = json.loads(await did.list_my_dids_with_meta(wallet_handle)) if didlist: for didobject in didlist: if 'metadata' in didobject: try: meta = json.loads(didobject['metadata']) if isinstance(meta, dict) and meta.get('Name', None) == "Steward": return didobject.get('did') except Exception: raise Exception ```

maydar (Wed, 12 Sep 2018 13:30:01 GMT):
Built von-network using: ./manage build . Getting the following error while starting it, ./manage start von-web_1 | ERROR|indy::errors::indy | src/errors/indy.rs:68 | Casting error to ErrorCode: Invalid library state: Unexpected SQLite error: execute returned results - did you mean to call query? von-web_1 | _indy_loop_callback: Function returned error 112 von-web_1 | Traceback (most recent call last): von-web_1 | File "server.py", line 295, in von-web_1 | loop.run_until_complete(boot()) von-web_1 | File "/usr/lib/python3.5/asyncio/base_events.py", line 387, in run_until_complete von-web_1 | return future.result() von-web_1 | File "/usr/lib/python3.5/asyncio/futures.py", line 274, in result von-web_1 | raise self._exception von-web_1 | File "/usr/lib/python3.5/asyncio/tasks.py", line 241, in _step von-web_1 | result = coro.throw(exc) von-web_1 | File "server.py", line 121, in boot von-web_1 | await trust_anchor.open() von-web_1 | File "/home/indy/.local/share/virtualenvs/server-8XoupS0v/lib/python3.5/site-packages/von_agent/agents.py", line 126, in open von-web_1 | await self.wallet.open() von-web_1 | File "/home/indy/.local/share/virtualenvs/server-8XoupS0v/lib/python3.5/site-packages/von_agent/wallet.py", line 186, in open von-web_1 | credentials=json.dumps(self.creds) if self.creds else None) von-web_1 | File "/home/indy/.local/share/virtualenvs/server-8XoupS0v/lib/python3.5/site-packages/indy/wallet.py", line 52, in create_wallet von-web_1 | create_wallet.cb) von-web_1 | File "/usr/lib/python3.5/asyncio/futures.py", line 361, in iter von-web_1 | yield self # This tells Task to wait for completion. von-web_1 | File "/usr/lib/python3.5/asyncio/tasks.py", line 296, in _wakeup von-web_1 | future.result() von-web_1 | File "/usr/lib/python3.5/asyncio/futures.py", line 274, in result von-web_1 | raise self._exception von-web_1 | indy.error.IndyError: ErrorCode.CommonInvalidState

maydar (Wed, 12 Sep 2018 13:32:20 GMT):
I installed and ran IBM's Indy tutorial example using https://github.com/IBM-Blockchain-Identity/indy-ssivc-tutorial. I am receiving the error "indy.error.IndyError: ErrorCode.CommonInvalidState" error while starting von-network. I built von-network using: `./manage build .` Getting the following error while starting it,` ./manage start` von-web_1 | ERROR|indy::errors::indy | src/errors/indy.rs:68 | Casting error to ErrorCode: Invalid library state: Unexpected SQLite error: execute returned results - did you mean to call query? von-web_1 | _indy_loop_callback: Function returned error 112 von-web_1 | Traceback (most recent call last): von-web_1 | File "server.py", line 295, in von-web_1 | loop.run_until_complete(boot()) von-web_1 | File "/usr/lib/python3.5/asyncio/base_events.py", line 387, in run_until_complete von-web_1 | return future.result() von-web_1 | File "/usr/lib/python3.5/asyncio/futures.py", line 274, in result von-web_1 | raise self._exception von-web_1 | File "/usr/lib/python3.5/asyncio/tasks.py", line 241, in _step von-web_1 | result = coro.throw(exc) von-web_1 | File "server.py", line 121, in boot von-web_1 | await trust_anchor.open() von-web_1 | File "/home/indy/.local/share/virtualenvs/server-8XoupS0v/lib/python3.5/site-packages/von_agent/agents.py", line 126, in open von-web_1 | await self.wallet.open() von-web_1 | File "/home/indy/.local/share/virtualenvs/server-8XoupS0v/lib/python3.5/site-packages/von_agent/wallet.py", line 186, in open von-web_1 | credentials=json.dumps(self.creds) if self.creds else None) von-web_1 | File "/home/indy/.local/share/virtualenvs/server-8XoupS0v/lib/python3.5/site-packages/indy/wallet.py", line 52, in create_wallet von-web_1 | create_wallet.cb) von-web_1 | File "/usr/lib/python3.5/asyncio/futures.py", line 361, in iter von-web_1 | yield self # This tells Task to wait for completion. von-web_1 | File "/usr/lib/python3.5/asyncio/tasks.py", line 296, in _wakeup von-web_1 | future.result() von-web_1 | File "/usr/lib/python3.5/asyncio/futures.py", line 274, in result von-web_1 | raise self._exception von-web_1 | indy.error.IndyError: ErrorCode.CommonInvalidState http://ip:9000/ and http://ip:9000/ are not coming up.

swcurran (Wed, 12 Sep 2018 15:26:26 GMT):
Just reran this and it worked fine. What is your environment (Mac, Windows, Linux)?

kenebert (Wed, 12 Sep 2018 16:53:50 GMT):
Has joined the channel.

Sean_Bohan (Wed, 12 Sep 2018 22:05:37 GMT):
Here is the agenda for tomorrow's call: 1. Housekeeping 2. AGENTS! 3. Upcoming Thurs WG Call Agendas (SeanB) 4. Finding patterns in users' traffic - a discussion on correlation (AmineK) 5. Plug-in version of a wallet storage for Postgre (IanC) Open Discussion Agenda: Housekeeping Open Discussion When: 8am PT 9am MT 10am CT 11amET 4pmBST Where: Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/232861185 Or iPhone one-tap (US Toll): +16465588656,232861185# or +14086380968,232861185# Or Telephone: Dial: +1 646 558 8656 (US Toll) or +1 408 638 0968 (US Toll) Meeting ID: 232 861 185 International numbers available: https://zoom.us/zoomconference?m=a0jD_rTMnh0ZYGQDOKPCNrK_0dP7WPfp1

andrewtan (Wed, 12 Sep 2018 22:10:21 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=bGCrEcBt7Wfejwtif) Any comments or thoughts?

gzkoala (Thu, 13 Sep 2018 04:03:47 GMT):
Has joined the channel.

maydar (Thu, 13 Sep 2018 05:43:15 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=qBpb7JjkXNgJ5EYXz) @swcurran It is Ubuntu 16.04.2 64 bit. lsb_release -a No LSB modules are available. Distributor ID: Ubuntu Description: Ubuntu 16.04.2 LTS Release: 16.04 Codename: xenial I work behind a proxy. Could that be a problem?

waleed (Thu, 13 Sep 2018 09:31:48 GMT):
hi every one :) Is it possible to use Indy with fabric?

szewong (Thu, 13 Sep 2018 12:59:30 GMT):
Hi, I am in the Hyperledger Identity working group. We are putting together a white-paper about the Digital Identity on blockchains. The target audience of this white paper are people who are just getting into the Identity world. I would like to ask the following question with regard to Indy. Any insight will be great. Our goal is to help people understand more about the fast changing landscape. 1. What is the key differentiator of Indy? 2. How is privacy handled? Where does PII reside in your solution and how is it protected? From correlation attacks? 3. How do you address the “Username/Password” issue? Meaning people have to maintain too many passwords and the fact that passwords are not secure. 4. How do you address interoperability with existing identity systems (LDAP and Active Directory)? 5. What are the special features that make your solution fit well within a blockchain context?

Rantwijk (Thu, 13 Sep 2018 13:15:04 GMT):
Has joined the channel.

wip-abramson (Thu, 13 Sep 2018 16:48:21 GMT):
Has joined the channel.

QinghuaLu (Fri, 14 Sep 2018 01:40:45 GMT):
Has joined the channel.

QinghuaLu (Fri, 14 Sep 2018 01:42:10 GMT):
Hi, I got a 114 error when i tried to run write_did.py example. The error message is

QinghuaLu (Fri, 14 Sep 2018 01:42:17 GMT):
_indy_loop_callback: Function returned error 114

QinghuaLu (Fri, 14 Sep 2018 01:42:36 GMT):
Error occurred: ErroCode.CommonIOError

QinghuaLu (Fri, 14 Sep 2018 01:42:50 GMT):
Anyone knows how to address this issue? Thanks

swcurran (Fri, 14 Sep 2018 02:19:23 GMT):
@maydar - I think the proxy could be a problem. It's probably limiting what ports are open and indy needs some non-standard ports open for this conversation. I run this on a Linux VM and it worked fine. Can you try it on another environment without the proxy?

jljordan_bcgov (Fri, 14 Sep 2018 04:52:28 GMT):
You will find a lot of those answers in https://sovrin.org/wp-content/uploads/Sovrin-Protocol-and-Token-White-Paper.pdf which is a network that uses Indy as its core tech @szewong [ ](https://chat.hyperledger.org/channel/indy?msg=7dBaZn4sZZC22gLH7)

Tnigam (Fri, 14 Sep 2018 06:44:50 GMT):
Has joined the channel.

Tnigam (Fri, 14 Sep 2018 06:45:56 GMT):
Hi Guys, I am new here does anybody know where is the Documentation for the .NET wrapper for indy? the link in broken on https://github.com/hyperledger/indy-sdk/blob/master/wrappers/dotnet/README.md

MaheshSharma (Fri, 14 Sep 2018 07:33:17 GMT):
Has joined the channel.

MaheshSharma (Fri, 14 Sep 2018 08:00:19 GMT):
I'm using Indy-cli: We have run command pool(pool1):wallet(alice_wallet):did(Av6...4c3):indy> ledger nym did= adddid verkey= add verkey but we have get from this command:Transaction has been rejected: Invalid format of command params. Please check format of posted JSONs, Keys, DIDs and etc...

Sean_Bohan (Fri, 14 Sep 2018 14:16:25 GMT):
Hey everyone. Video from yesterday's WG Call: https://drive.google.com/open?id=12J1bWBkueCiOdrZ94jIxcVbUYVD585TJ Notes from yesterdays INDY WG Call: Read the docs HIPE: https://github.com/hyperledger/indy-hipe/pull/24 NPM INdy SDK: https://www.npmjs.com/package/indy-sdk Hyperledger Identity WG Call Next call: 9/19 Noon ET https://zoom.us/my/hyperledger.community https://wiki.hyperledger.org/groups/identity/identity-wg Hyperledger Hackfest Fall 2018: Oct 3-4 in Montreal, Canada https://events.linuxfoundation.org/events/hyperledger-hackfest-montreal-2018/ CryptoLib meeting (Indy Crypto Library destined to merge) Internet Identity Workshop: Oct 23-25 in Mountain View, CA Website: https://www.internetidentityworkshop.com/ Hyperledger Community Discount (20% off): https://iiw27.eventbrite.com?discount=HLI_XXVII_20

PhillipGibb (Sun, 16 Sep 2018 15:17:15 GMT):
out of interest; Is there a reason my Rust was chosen as the primary language for Indy?

Shyam_Pratap_Singh (Mon, 17 Sep 2018 07:10:07 GMT):
Has joined the channel.

canadaduane (Mon, 17 Sep 2018 14:23:52 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=nagRq9RCXyYDGKtaY) @PhillipGibb Do you mean just "Rust" generally (or is there something special about your Rust?)

PhillipGibb (Mon, 17 Sep 2018 14:26:44 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=JSuW7JTBZ5NwgrZpf) @canadaduane well, specifically for Indy-sdk and indy-node. I am interested in the motivation because 1. I am developing in the SSI space 2. it is always fun to learn a new language

danielhardman (Mon, 17 Sep 2018 16:55:30 GMT):
@PhillipGibb : I was the one who wrote the original spec for indy-sdk, and stipulated that Rust would be the implementation language. We knew we needed something that was c-callable, so we could wrap it easily in many different languages. Our main choices were C/C++, Go, Rust, and maybe D. We discounted D as not having the momentum we wanted. We seriously considered Go but ended up choosing Rust because we didn't believe Go's natural concurrency model was right for us, and because we liked Rust's correctness guarantees, LLVM backend, and so forth. An extra plus was that some crypto libraries were available in a Rust flavor.

PhillipGibb (Mon, 17 Sep 2018 18:06:28 GMT):
@danielhardman excellent, thanks

VarunMathur (Tue, 18 Sep 2018 08:20:44 GMT):
Has joined the channel.

adamludvik (Tue, 18 Sep 2018 15:10:00 GMT):
Has left the channel.

RBorst (Tue, 18 Sep 2018 15:44:42 GMT):
Has joined the channel.

rgunn (Thu, 20 Sep 2018 03:59:15 GMT):
Has joined the channel.

anand.fast (Thu, 20 Sep 2018 06:23:48 GMT):
Has joined the channel.

anand.fast (Thu, 20 Sep 2018 06:28:27 GMT):
Could you please assist me to setup a hyperledger indy network (from scratch on ubuntu 16.04) so as to be able to follow the indy tutorial (Alice use-case)? The current documentation is confusing as I am not able to follow the path.

arati_baliga (Thu, 20 Sep 2018 07:45:26 GMT):
Has joined the channel.

dklesev (Thu, 20 Sep 2018 11:13:34 GMT):
can anyone explain to me where all the issues are gone? (https://github.com/hyperledger/indy-sdk) Seems to be disabled... Whats the reason for this?

dklesev (Thu, 20 Sep 2018 11:15:50 GMT):
ah, I see everything is in jira now?

Sean_Bohan (Thu, 20 Sep 2018 14:13:42 GMT):
Hyperledger Indy Working Group Here is the agenda for today's call: Housekeeping AGENTS! Upcoming Thurs WG Call Agendas (SeanB) Rebooting Web of Trust Planning Open Discussion When (TODAY): 8am PT 9am MT 10am CT 11amET 4pmBST Where: Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/232861185

Sean_Bohan (Thu, 20 Sep 2018 14:14:26 GMT):
this call is recorded, previous calls can be found here: https://drive.google.com/drive/u/1/folders/111FWCb8cIsEzN-5yhTACyXtKwCnenLDu

esplinr (Thu, 20 Sep 2018 16:15:27 GMT):
The Evernym Indy teams have posted our sprint demos to YouTube. You can see them here: https://www.youtube.com/playlist?list=PLRp0viTDxBWGLdZk0aamtahB9cpJGV7ZF Everything is there since May, except for one recording from early July that I lost.

kdenhartog (Thu, 20 Sep 2018 20:22:31 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=DxenBs6ZM2kCnuLue) @nage @esplinr where's the best place to redirect people to jira so we don't get this question repetitively.

nage (Thu, 20 Sep 2018 20:33:01 GMT):
SDK https://jira.hyperledger.org/browse/IS Ledger/Node https://jira.hyperledger.org/browse/INDY Agents https://jira.hyperledger.org/browse/IA

nage (Thu, 20 Sep 2018 20:36:01 GMT):
We don't use github issues because they don't allow for responsible disclosure of security issues. We will like GitHub issues, and will reevaluate if they add that functionality in the future.

nage (Thu, 20 Sep 2018 20:36:01 GMT):
We don't use github issues because they don't allow for responsible disclosure of security issues. We like GitHub issues, and will reevaluate if they add that functionality in the future.

rjones (Thu, 20 Sep 2018 21:13:13 GMT):
the downside of turning off issues is the open ones that were open are now hidden. I think this is a design flaw in github

rjones (Thu, 20 Sep 2018 21:13:26 GMT):
same for the wiki. you can't put it in read-only mode

burdettadam (Thu, 20 Sep 2018 22:27:36 GMT):
The issues were imported to Jira before GitHub issue tracking was turned off. If you do an advanced search in the indy-sdk Jira project with this ``` project='IS' AND issuekey>=IS-958 AND issuekey<=IS-1000 ``` you will get the bulk of them.

sithara99 (Fri, 21 Sep 2018 03:09:09 GMT):
Has joined the channel.

freeman (Fri, 21 Sep 2018 12:28:46 GMT):
Has joined the channel.

ppirrip (Fri, 21 Sep 2018 16:33:47 GMT):
Has joined the channel.

canadaduane (Fri, 21 Sep 2018 18:44:41 GMT):
Are there any examples of claims schemas out in the wild?

MHeisoku (Sat, 22 Sep 2018 10:09:35 GMT):
Can anyone help explain why indy is being built with a blockchain behind it? It seems strange that the blockchain needs to be setup and maintained by one entity (so there's really a central system admin) and that all changes require consultation and updates across the network (otherwise you risk forks). Plenum surely can't scale to handle large numbers of nodes and high volumes of writes per second (no consensus algorithm fares convincingly well for business needs), and given that all bllckchains are centrally administered anyway why can't there just be a central database instead of a blockchain for Indy?

arjunkhera (Sun, 23 Sep 2018 17:26:36 GMT):
Has joined the channel.

ClearFoundation (Mon, 24 Sep 2018 06:39:39 GMT):
Has joined the channel.

Rod (Mon, 24 Sep 2018 18:48:59 GMT):
Has joined the channel.

ShivVenkatraman (Mon, 24 Sep 2018 20:53:28 GMT):
Has joined the channel.

ShivVenkatraman (Mon, 24 Sep 2018 20:53:57 GMT):
From the white paper, https://www.evernym.com/wp-content/uploads/2017/07/What-Goes-On-The-Ledger.pdf, the revocation registries are published by the issuer and stored in Sovrin. How fast (and big) does the revocation registration grow? If 1000s of claims are revoked, what are the implications on the run time performance during validation/verification?

VaTcHEr (Mon, 24 Sep 2018 21:01:54 GMT):
Has joined the channel.

Taaanos (Tue, 25 Sep 2018 16:17:51 GMT):
Has joined the channel.

fhmarino (Tue, 25 Sep 2018 19:55:26 GMT):
Hi there! I have been studying the Indy Platform and some doubts about architecture were raised: where is the identity wallet and which data is write in this wallet? where is the directory and ledger file (in the peer)? the kind of data and transactions its been written to the ledger?

fhmarino (Tue, 25 Sep 2018 19:55:26 GMT):
Hi there! I have been studying the Indy Platform and some doubts about architecture were raised: 1) where is the identity wallet and which data is write in this wallet? 2) where is the directory and ledger file (in the peer)? 3) the kind of data and transactions its been written to the ledger?

alainN (Tue, 25 Sep 2018 23:06:31 GMT):
Has joined the channel.

hinhrt (Wed, 26 Sep 2018 00:21:25 GMT):
Has joined the channel.

swcurran (Wed, 26 Sep 2018 01:12:02 GMT):
@ShivVenkatraman - the bigger performance hit for revocation occurs at Credential Definition Creation time, so it's a one time cost. The more credentials you plan to issue, the longer it will take. The next longest time, which will not be long, is creating the proof. The verifier just has to review the Proof and verify that the "Proof of non-Revocation" is sufficiently recent. Note that the number of credentials that can be issued is set at Credential Definition creation time - a new Cred Def may have to be created if you run out. FYI - we found the Cred Def with revocation took a REALLY long time to create when the code has DEBUG turned off. Without that, it's relatively fast. Still takes awhile - lots of crypto-math - but not intolerable for a one time event.

ShivVenkatraman (Wed, 26 Sep 2018 01:41:27 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=p6MrBK8HW2aNxYFwL) @swcurran Thanks @swcurran. Can revocation be done at attribute level? Suppose my credential schema has attributes name, age, height. Can revocation be done just for age or is it done for the entire credential?

swcurran (Wed, 26 Sep 2018 01:59:59 GMT):
No - revocation is on a credential. However, our expectation is that the revocation feature will be used a lot more the changed data, not necessarily to take something away. Thus, if a single claim changed, you would revoke the current credential, and issue a new one. For example - A Driver's License - when you move to a new address, you revoke the old one, and issue a new one with the new address. Beats the sticker that they currently give out when you move. I had the sticker on a my licence for 3 years once!

ravneetkaur (Wed, 26 Sep 2018 05:43:36 GMT):
Has joined the channel.

ravneetkaur (Wed, 26 Sep 2018 05:43:49 GMT):
Hi, I am a student working on setting up hyperledger indy on windows. And went through lot of documentation. But I am not really sure how to setup the basic testbed. Can someone help me with this?

rmorbach (Wed, 26 Sep 2018 10:32:40 GMT):
Has joined the channel.

nage (Wed, 26 Sep 2018 18:00:17 GMT):
slides I'm sharing at RWoT https://docs.google.com/presentation/d/1980Jta3pt_Qxa9JY1KSMiOWkgxeMcMGFkJL4A9EWIhc

Sean_Bohan (Wed, 26 Sep 2018 22:48:49 GMT):
Indy WG Call 9/27Here is the agenda for tomorrow's call: Housekeeping AGENTS! Kuzma’s Internship @ Hyperledger Indy (Kuzma) Hyperledger Hackfest Planning (all) Open Discussion When: 8am PT 9am MT 10am CT 11amET 4pmBST Where: Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/232861185 Or iPhone one-tap (US Toll): +16465588656,232861185# or +14086380968,232861185# Or Telephone: Dial: +1 646 558 8656 (US Toll) or +1 408 638 0968 (US Toll) Meeting ID: 232 861 185 International numbers available: https://zoom.us/zoomconference?m=a0jD_rTMnh0ZYGQDOKPCNrK_0dP7WPfp1

Sean_Bohan (Wed, 26 Sep 2018 22:48:49 GMT):
Indy WG Call 9/27 Here is the agenda for tomorrow's call: Housekeeping AGENTS! Kuzma’s Internship @ Hyperledger Indy (Kuzma) Hyperledger Hackfest Planning (all) Open Discussion When: 8am PT 9am MT 10am CT 11amET 4pmBST Where: Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/232861185 Or iPhone one-tap (US Toll): +16465588656,232861185# or +14086380968,232861185# Or Telephone: Dial: +1 646 558 8656 (US Toll) or +1 408 638 0968 (US Toll) Meeting ID: 232 861 185 International numbers available: https://zoom.us/zoomconference?m=a0jD_rTMnh0ZYGQDOKPCNrK_0dP7WPfp1

kingpasan (Thu, 27 Sep 2018 08:49:02 GMT):
Has joined the channel.

karthikpanicker (Thu, 27 Sep 2018 10:23:42 GMT):
Has joined the channel.

sklump (Thu, 27 Sep 2018 11:02:07 GMT):
Has joined the channel.

sklump (Thu, 27 Sep 2018 11:05:38 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=p6MrBK8HW2aNxYFwL) @swcurran Correction, the issuer must issue a new Revocation Registry (not Cred Def) when the existing one runs out. A cred def can father many revocation registries.

sklump (Thu, 27 Sep 2018 11:05:38 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=p6MrBK8HW2aNxYFwL) @swcurran Correction, the issuer must issue a new Revocation Registry (not Cred Def) when the existing one runs out. A cred def can father many revocation registries. Also, revocation registry creation takes show-stopper long (0.25 sec per credential) when DEBUG is ON, not OFF. I.e., build libindy with `cargo build --release` for reasonable results.

sunburntcat (Thu, 27 Sep 2018 18:31:13 GMT):
Has joined the channel.

sunburntcat (Thu, 27 Sep 2018 18:35:58 GMT):
Hey #indy I'm working with IBM's ssivc tutorial, and Bryce Cutiss (one of the maintainers of the demo) gave us some suggestions on how to issue a new schema. I've updated the config.toml, proof_request.json, and schemas.json files. How do we generate a new issuer (on our own network) so that we can deploy a new schema. I apologize if these are IBM-tutorial specific questions. It's hard to see where IBM stops and indy-SDK begins

sunburntcat (Thu, 27 Sep 2018 18:35:58 GMT):
Hey #indy I'm working with IBM's ssivc tutorial, and Bryce Cutiss (one of the maintainers of the demo) gave us some suggestions on how to issue a new schema. I've updated the config.toml, proof_request.json, and schemas.json files. How do we generate a new issuer (on our own network) so that we can deploy a new schema? I apologize if these are IBM-tutorial specific questions. It's hard to see where IBM stops and indy-SDK begins

tarun.sharma (Thu, 27 Sep 2018 23:07:04 GMT):
Has joined the channel.

VipulDabhi (Fri, 28 Sep 2018 05:23:10 GMT):
Has joined the channel.

VipulDabhi (Fri, 28 Sep 2018 05:23:13 GMT):
hi

VipulDabhi (Fri, 28 Sep 2018 05:23:15 GMT):
IndyError: PoolLedgerTimeout

karthick15v (Fri, 28 Sep 2018 06:18:04 GMT):
Has joined the channel.

Tiruyasha (Fri, 28 Sep 2018 11:21:02 GMT):
Has joined the channel.

sklump (Fri, 28 Sep 2018 11:23:08 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=4AYqY96XH8rtopvTe) @sunburntcat 1. Create and open a wallet `wallet.create_wallet()`, `wallet.open_wallet()` 2. Generate verification key `did.create_and_store_my_did()` 3. Send NYM request to Steward `ledger.build_nym_request(), ledger.sign_and_submit_request()`

Tiruyasha (Fri, 28 Sep 2018 12:03:24 GMT):
Hello i have a problem. When i run the getting started project in python i get a commoninvalidstate error. I already tried the recovery steps: https://github.com/hyperledger/indy-sdk/blob/master/doc/getting-started/Trouble_shoot_GSG.md

Tiruyasha (Fri, 28 Sep 2018 12:26:18 GMT):
I also get WARNING:indy.libindy:_indy_loop_callback: Function returned error 306/112

baconsandwich (Fri, 28 Sep 2018 13:10:19 GMT):
Hey, maybe this is a dumb question because I forgot something...but if I create new "pairwises" for each DID I connect to, is there a reason to create more than one DID for a wallet? What would be the use case to create more than one DID in a wallet?

Tiruyasha (Fri, 28 Sep 2018 13:13:17 GMT):
I fixed it... I set the wrong ip address :')

forrunner (Sat, 29 Sep 2018 04:34:30 GMT):
Has joined the channel.

vishal21pandita (Sat, 29 Sep 2018 07:08:31 GMT):
Has joined the channel.

tomislav (Sat, 29 Sep 2018 20:49:00 GMT):
@sklump How does one decide and track the size of the revocation registry?

berserkr (Sun, 30 Sep 2018 08:23:55 GMT):
Has joined the channel.

berserkr (Sun, 30 Sep 2018 08:28:35 GMT):
Hi All, quick question. I am working on a PoC and have about a week to get the blockchain piece done. I want to use indy as the project deals with identity. I know the HLF inside and out, but I don't want to use Fabric any more unless I really have to. Before jumping to Ethereum, I want to know, from your experience, how involved is the process in getting a simple indy-based service up and running with some restful APIs to issue identities?

the_identity_guy (Sun, 30 Sep 2018 14:58:47 GMT):
Hello, is there an enterprise Indy subscription? if so can I have the link to it.

sklump (Mon, 01 Oct 2018 13:38:48 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=n7DiH3oEX473KMwNb) @tomislav Hi, There is a natural design tension. The code itself can't know _a priori_ how big the registry is going to need to be, and it costs to make it bigger. So what I've done is to have a rev reg size parameter at cred def creation for an override of default behaviour. The default behaviour for `von_anchor.Issuer` is to start with a smaller registry of size 256, and double it every time it fills, up to size 4096. I use a numeric value, incrementing with rev reg cardinality, to track what the next size ought to be by default. It's a heuristic; any implementation can choose its own way forward here. You don't have to track the size of the revocation registry: indy-sdk will raise `ErrorCode.AnoncredsRevocationRegistryFullError` when you try to create a new credential on a full revocation registry.

sklump (Mon, 01 Oct 2018 13:38:48 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=n7DiH3oEX473KMwNb) @tomislav Hi, There is a natural design tension. The code itself can't know _a priori_ how big the registry is going to need to be, and it costs to make it bigger. So what I've done is to have a rev reg size parameter at cred def creation for an override of default behaviour. The default behaviour for `von_anchor.Issuer` is to start with a smaller registry of size 256, and double it every time it fills, up to size 4096. I use a numeric value for the rev reg tag, incrementing with rev reg cardinality, to track what the next size ought to be by default. It's a heuristic; any implementation can choose its own way forward here. You don't have to track the size of the revocation registry: indy-sdk will raise `ErrorCode.AnoncredsRevocationRegistryFullError` when you try to create a new credential on a full revocation registry.

sklump (Mon, 01 Oct 2018 13:38:48 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=n7DiH3oEX473KMwNb) @tomislav Hi, There is a natural design tension. The code itself can't know _a priori_ how big the registry is going to need to be, and it costs to make it bigger. So what I've done is to have a rev reg size parameter at cred def creation for an override of default behaviour. The default behaviour for `von_anchor.Issuer` is to start with a smaller registry of size 256, and double it every time it fills, up to size 4096. I use a numeric value for the rev reg tag, incrementing with rev reg cardinality, to generate a suggestion for the next one's size. It's a heuristic; any implementation can choose its own way forward here. You don't have to track the size of the revocation registry: indy-sdk will raise `ErrorCode.AnoncredsRevocationRegistryFullError` when you try to create a new credential on a full revocation registry.

sklump (Mon, 01 Oct 2018 13:38:48 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=n7DiH3oEX473KMwNb) @tomislav Hi, There is a natural design tension. The code itself can't know _a priori_ how big the registry is going to need to be, and it costs to make it bigger. So what I've done is to have a rev reg size parameter at cred def creation for an override of default behaviour. The default behaviour for `von_anchor.Issuer` is to start with a smaller registry of size 256, and double it every time it fills, up to size 4096. I use a numeric value for the rev reg tag, incrementing with rev reg cardinality, to generate a suggestion for the next one's size _(it's OK to put this duty on the tag; the only indy-sdk requirement is uniqueness and, presumably, printability)_. It's a heuristic; any implementation can choose its own way forward here. You don't have to track the size of the revocation registry: indy-sdk will raise `ErrorCode.AnoncredsRevocationRegistryFullError` when you try to create a new credential on a full revocation registry.

sklump (Mon, 01 Oct 2018 13:38:48 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=n7DiH3oEX473KMwNb) @tomislav Hi, There is a natural design tension. The code itself can't know _a priori_ how big the registry is going to need to be, and it costs to make it bigger. So what I've done is to have a rev reg size parameter at cred def creation for an override of default behaviour. The default behaviour for `von_anchor.Issuer` is to start with a smaller registry of size 256, and double it every time it fills, up to size 4096 (any OS can handle flat directories of thousands of tails files scalably, so I cap it there and the operation scales to the millions). I use a numeric value for the rev reg tag, incrementing with rev reg cardinality, to generate a suggestion for the next one's size _(it's OK to put this duty on the tag; the only indy-sdk requirement is uniqueness and, presumably, printability)_. It's a heuristic; any implementation can choose its own way forward here. You don't have to track the size of the revocation registry: indy-sdk will raise `ErrorCode.AnoncredsRevocationRegistryFullError` when you try to create a new credential on a full revocation registry.

tomislav (Mon, 01 Oct 2018 14:20:48 GMT):
@sklump Fantastic, exactly what I was asking for. Thank you

Taaanos (Mon, 01 Oct 2018 15:04:52 GMT):
Hey everyone, I am working on my thesis at IBM on Indy + IoT.

Taaanos (Mon, 01 Oct 2018 15:05:21 GMT):
Is there a proper way to give Identity to a thing?A thing is a sovrin Entity (as stated in the Sovrin Provisional Trust Framework Whitepaper*), but how can you distinguish a thing from an Identity owner? Should that be implemented in the DDO? I am currently trying to find out what is defined/implemented for IoT use cases. *https://docs.google.com/document/d/18V1c0rOQYxNMleuV_2z7yQny0KdBnuDkWlN8DNUrioM/edit (3.2.1)

Taaanos (Mon, 01 Oct 2018 15:05:21 GMT):
Is there a proper way to give Identity to a thing?A thing is a sovrin Entity (as stated in the Sovrin Provisional Trust Framework Whitepaper), but how can you distinguish a thing from an Identity owner? Should that be implemented in the DDO? I am currently trying to find out what is defined/implemented for IoT use cases.

phaniac (Mon, 01 Oct 2018 18:04:40 GMT):
Has joined the channel.

handaanmol (Tue, 02 Oct 2018 05:56:25 GMT):
Has joined the channel.

peacekeeper (Tue, 02 Oct 2018 10:12:34 GMT):
Upcoming Hyperledger meetup in Vienna on 16th October focusing on Indy: https://www.meetup.com/Hyperledger-Vienna/events/255191300/

dono (Tue, 02 Oct 2018 21:46:01 GMT):
Has joined the channel.

aniv (Wed, 03 Oct 2018 09:56:10 GMT):
Has joined the channel.

starkriedesel (Wed, 03 Oct 2018 17:30:54 GMT):
Has joined the channel.

adamludvik (Wed, 03 Oct 2018 18:22:01 GMT):
Has joined the channel.

adamludvik (Wed, 03 Oct 2018 18:47:39 GMT):
Hello! Is this the right forum to ask a question about plenum?

dbluhm (Wed, 03 Oct 2018 19:56:49 GMT):
@adamludvik here is a good place to start :slight_smile:

adamludvik (Wed, 03 Oct 2018 20:14:48 GMT):
I am researching how different consensus algorithms handle "fair ordering" of transactions. Specifically, I am interested in if and how protocols handle nodes ordering transactions in a way that unfairly benefits them, or even generating and inserting new transactions where it benefits them. In lottery-style algorithms, I believe this is just ignored because the probability of winning when a node wants to is low. With leader-style algorithms that maintain a single leader for an extended time, such as PBFT and rBFT, it seems like this is more important. Does anyone know if rBFT or Plenum monitor that the leader is including transactions within a reasonable amount of time or that it isn't unfairly ordering the transactions to benefit itself?

dbluhm (Wed, 03 Oct 2018 20:30:07 GMT):
That is a good question for @lovesh

agunde (Wed, 03 Oct 2018 20:46:58 GMT):
Has joined the channel.

srunpengsreang (Thu, 04 Oct 2018 04:53:59 GMT):
Has joined the channel.

srunpengsreang (Thu, 04 Oct 2018 04:54:26 GMT):
Hello guys here, nice to meet you here

srunpengsreang (Thu, 04 Oct 2018 04:54:49 GMT):
Right now I am researching Hyperledger Indy with iOS

srunpengsreang (Thu, 04 Oct 2018 04:56:20 GMT):
I don't understand much about the flow of creating app with Indy

Dpkkmr (Thu, 04 Oct 2018 06:37:24 GMT):
Hi I am new to Indy, trying to get hands-on on Indy project. I am following the tutorial provided in https://github.com/IBM-Blockchain-Identity/indy-tutorial-sandbox When I am creating the Nodes cluster, all nodes in the cluster are exiting with this exception: c211c7037b82 indy-base "/bin/bash -c 'creat…" 4 minutes ago Exited (1) 4 minutes ago Node4 f29e84e71826 indy-base "/bin/bash -c 'creat…" 4 minutes ago Exited (1) 4 minutes ago Node3 a7ba9d57712a indy-base "/bin/bash -c 'creat…" 4 minutes ago Exited (1) 4 minutes ago Node2 45008dd837f7 indy-base "/bin/bash -c 'creat…" 4 minutes ago Exited (1) 4 minutes ago Node1

Dpkkmr (Thu, 04 Oct 2018 06:37:24 GMT):
Hi I am new to Indy, trying to get hands-on on Indy project. I am following the tutorial provided in https://github.com/IBM-Blockchain-Identity/indy-tutorial-sandbox When I am creating the Nodes cluster, all nodes in the cluster are exiting with this exception in each node container logs: Proof of possession for BLS key is RPLagxaR5xdimFzwmzYnz4ZhWtYQEj8iR5ZU53T2gitPCyCHQneUn2Huc4oeLd2B2HzkGnjAff4hWTJT6C7qHYB1Mv2wU5iHHGFWkhnTX9WsEAbunJCV2qcaXScKj4tTfvdDKfLiVuU2av6hbsMztirRze7LvYBkRHV3tGwyCptsrP Traceback (most recent call last): File "/usr/local/bin/start_indy_node", line 12, in raise Exception("Provide name and two pairs of IP/port for running the node " Exception: Provide name and two pairs of IP/port for running the node and client stacks in form 'node_name node_ip node_port client_ip client_port' c211c7037b82 indy-base "/bin/bash -c 'creat…" 4 minutes ago Exited (1) 4 minutes ago Node4 f29e84e71826 indy-base "/bin/bash -c 'creat…" 4 minutes ago Exited (1) 4 minutes ago Node3 a7ba9d57712a indy-base "/bin/bash -c 'creat…" 4 minutes ago Exited (1) 4 minutes ago Node2 45008dd837f7 indy-base "/bin/bash -c 'creat…" 4 minutes ago Exited (1) 4 minutes ago Node1 Tried changing the IP address to local system IP address, still no luck. Any suggestions on resolving this will be helpful.

greg2git (Thu, 04 Oct 2018 13:23:13 GMT):
Has joined the channel.

phaniac (Thu, 04 Oct 2018 15:57:15 GMT):
Couldn't clap earlier - so congrats to the TheOrgBook team! That's exciting stuff.

lovesh (Thu, 04 Oct 2018 17:22:18 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=FsHbKKoSsvqtEvdmc) @adamludvik RBFT runs several instances of the protocol and there is only one instance (known beforehand) called the master instance whose execution affects the ledger. Each instance has its own leader and the leader of the master instance can be changed if it performs significantly worse than other instances. Now if this leader was ordering txns in the batch in an unfair manner, it will lose some performance and might get replaced. There are other protocols like HoneyBadger where users send encrypted txns so nodes cannot do favouritism. Another interesting one is Helix by http://www.orbs.com. Here users send encrypted txns and the nodes use threshold decryption to decrypt the txns, this way it ensures that nodes can't see the txn content unless the block is finalised.

adamludvik (Thu, 04 Oct 2018 18:52:04 GMT):
@lovesh thanks for the answer! I will take a loook at those other examples too.

nage (Thu, 04 Oct 2018 19:04:28 GMT):
At the hackfest we are having a planning meeting to talk about problems people are seeing, ideas for changes to the platform, what the team is thinking about and what people here want to work on. The link the the doc we are using can be found here https://docs.google.com/document/d/1DWtAuGys8tfzzz013ZgR_1dHXfUX_DVUMp-BQEc7FEU/edit?usp=sharing

nage (Thu, 04 Oct 2018 19:04:55 GMT):
(I expect we will have these sessions at Hackfests going forward -- hopefully it is an incentive to see more of you here! :sunglasses: )

Sean_Bohan (Thu, 04 Oct 2018 19:09:32 GMT):
go here for the zoom: https://zoom.us/j/6466466446

rikgig (Thu, 04 Oct 2018 19:19:03 GMT):
Has joined the channel.

manuvarghese (Thu, 04 Oct 2018 19:19:46 GMT):
Has joined the channel.

ikinique (Thu, 04 Oct 2018 19:39:47 GMT):
Has joined the channel.

adamludvik (Thu, 04 Oct 2018 20:51:27 GMT):
@lovesh Do replicas check that the leader is not intentionally ignoring a transaction?

srunpengsreang (Fri, 05 Oct 2018 03:51:36 GMT):
Hello all

srunpengsreang (Fri, 05 Oct 2018 03:52:05 GMT):
Right now I have written a sample using Indy with iOS platform

srunpengsreang (Fri, 05 Oct 2018 03:52:56 GMT):
But the framework is built old version of Xcode

srunpengsreang (Fri, 05 Oct 2018 03:53:03 GMT):
I am using Xcode 10

srunpengsreang (Fri, 05 Oct 2018 03:53:40 GMT):
So Could the team of iOS update the framework to support Xcode 10?

VimalJose (Fri, 05 Oct 2018 08:07:43 GMT):
Has joined the channel.

Elakkiyaselvan (Fri, 05 Oct 2018 08:29:19 GMT):
Has joined the channel.

jlin (Fri, 05 Oct 2018 15:07:40 GMT):
Has joined the channel.

lovesh (Fri, 05 Oct 2018 15:47:30 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=WKM6e4NzXKtrJ6cMX) @adamludvik By replicas, i assume you mean replicas on other nodes, then no. It is possible but would be expensive to ensure

adamludvik (Fri, 05 Oct 2018 15:53:15 GMT):
I think withholding transactions could be checked by either the secondaries in a single instance of the protocol or when comparing the orderings generated by the different instances. Are neither of these checking for this behavior? Does a client have any recourse if it sees that a leader is ignoring its transactions?

lovesh (Fri, 05 Oct 2018 16:01:38 GMT):
Yes, secondaries can check but its expensive to make that check, each secondary sends the txn to the primary and get a signed acknowledgement from the primary, the primary if does not include the txn in a batch under a timeout, then the secondariness share the signed acknowledgement with each other. "orderings generated by the different instances" is not enough since the primary can choose not to participate in any instance for that txn. As of now the client has no recourse

lovesh (Fri, 05 Oct 2018 16:04:16 GMT):
But if we have signed acknowledgements from the node to the client, the client can share that with other nodes and complain

lovesh (Fri, 05 Oct 2018 16:08:36 GMT):
Currently, we don't separate consensus from execution but if we separate consensus from execution then having encrypted txns makes more sense to ensure fairness. If we had simpler txns then we can use snarks to prove that encrypted txns are actually correct and will execute correctly, that will let us keep consensus and execution combined

adamludvik (Fri, 05 Oct 2018 16:47:34 GMT):
Thanks for the great answers!

lovesh (Fri, 05 Oct 2018 16:52:11 GMT):
No probem

darkblue-b (Fri, 05 Oct 2018 17:27:11 GMT):
Has joined the channel.

darkblue-b (Fri, 05 Oct 2018 17:28:20 GMT):
hi - my first day here.. I am about to jump into an Ethereum event, a big one.. I have been building and running Ethereum toolchain on Linux for the last week or so.. I just noticed this today

darkblue-b (Fri, 05 Oct 2018 17:28:47 GMT):
I could try to build this one, if that is an option.. in "discovery" mode

sjrichter (Fri, 05 Oct 2018 20:07:42 GMT):
Has joined the channel.

marksta (Sat, 06 Oct 2018 09:50:39 GMT):
I want to support the sovrin project by hosting a observer node. Is this possible without being a steward? (Is this even the right place to ask this?)

bootstrapsp (Sun, 07 Oct 2018 14:11:12 GMT):
Has joined the channel.

supun 1 (Sun, 07 Oct 2018 16:18:53 GMT):
Has joined the channel.

supun 1 (Sun, 07 Oct 2018 16:29:29 GMT):
I am using windows, and new to hypeldger

supun 1 (Sun, 07 Oct 2018 16:30:06 GMT):
I want to start with indy-sdk with java

Nareshtej (Mon, 08 Oct 2018 05:22:33 GMT):
Has joined the channel.

lijiachuan (Mon, 08 Oct 2018 05:59:51 GMT):
Has joined the channel.

lijiachuan (Mon, 08 Oct 2018 06:01:02 GMT):
Hi All, am a new one for Indy, I would like to confirm one quite basic question, what does 'NYM' mean, what is the full description of this abbreviation? Thanks.

lijiachuan (Mon, 08 Oct 2018 07:44:02 GMT):
And follow the setup dev document, there was an error in the last step "pip install -e .[tests]", two errors as below `rocksdb/_rocksdb.cpp:549:27: fatal error: rocksdb/slice.h: No such file or directory compilation terminated. error: command 'x86_64-linux-gnu-gcc' failed with exit status 1 _______ Failed building wheel for python-rocksdb _______ Command "/home/lijiachuan/Desktop/Indy/indy-env/bin/python3.5 -u -c "import setuptools, tokenize;__file__='/tmp/pip-install-p38d8pma/python-rocksdb/setup.py';f=getattr(tokenize, 'open', open)(__file__);code=f.read().replace('\r\n', '\n');f.close();exec(compile(code, __file__, 'exec'))" install --record /tmp/pip-record-w6j78jky/install-record.txt --single-version-externally-managed --compile --install-headers /home/lijiachuan/Desktop/Indy/indy-env/include/site/python3.5/python-rocksdb" failed with error code 1 in /tmp/pip-install-p38d8pma/python-rocksdb/`

lijiachuan (Mon, 08 Oct 2018 07:44:02 GMT):
And follow the setup dev document, there was an error in the last step "pip install -e .[tests]", two errors as below ``` rocksdb/_rocksdb.cpp:549:27: fatal error: rocksdb/slice.h: No such file or directory compilation terminated. error: command 'x86_64-linux-gnu-gcc' failed with exit status 1 _______ Failed building wheel for python-rocksdb _______ Command "/home/lijiachuan/Desktop/Indy/indy-env/bin/python3.5 -u -c "import setuptools, tokenize;__file__='/tmp/pip-install-p38d8pma/python-rocksdb/setup.py';f=getattr(tokenize, 'open', open)(__file__);code=f.read().replace('\r\n', '\n');f.close();exec(compile(code, __file__, 'exec'))" install --record /tmp/pip-record-w6j78jky/install-record.txt --single-version-externally-managed --compile --install-headers /home/lijiachuan/Desktop/Indy/indy-env/include/site/python3.5/python-rocksdb" failed with error code 1 in /tmp/pip-install-p38d8pma/python-rocksdb/ ```

lijiachuan (Mon, 08 Oct 2018 09:42:38 GMT):
I am not sure whether this line of code is correct or not in https://github.com/hyperledger/indy-sdk/blob/master/samples/python/src/getting_started.py line 828 `assert sender_verkey == await did.key_for_did(pool_handle, from_wallet, to_from_did)` From the setup dev document, this line of code should looks like ``` # Steward Agent faber_verkey = await did.key_for_did(pool_handle, from_wallet, faber_did_info['did']) ``` So whether to_from_did is the same as faber_did_info['did]? I think this DID should be faber's Verinym DID but not the connection DID.

lijiachuan (Mon, 08 Oct 2018 10:46:45 GMT):
Hello, may I have anyone's help on my questions? Thanks a lot.

micahsoft (Mon, 08 Oct 2018 12:16:37 GMT):
Has joined the channel.

lovesh (Mon, 08 Oct 2018 14:14:13 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=B38bdv4i6tnJ3FQGG) @lijiachuan There is no full description, its last part of the word Cryptonym, short for a cryptographic name.

lovesh (Mon, 08 Oct 2018 14:15:50 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=sBrBRxxecZEaj7ZXa) @lijiachuan You need to install rocksdb. You can follow instructions from their website or follow instructions here https://github.com/hyperledger/indy-node/blob/master/dev-setup/ubuntu/setup-dev-depend-ubuntu16.sh

greg2git (Mon, 08 Oct 2018 14:30:20 GMT):
what's up with rocksdb? is that a persistent storage dependency?

agunde (Mon, 08 Oct 2018 14:30:39 GMT):
Has left the channel.

lijiachuan (Mon, 08 Oct 2018 16:29:41 GMT):
hi @lovesh, when I execute "sudo apt-key adv --keyserver keyserver.ubuntu.com --recv-keys 68DB5E88" this command, it returned below timeout error: ``` gpg: requesting key 68DB5E88 from hkp server keyserver.ubutntu.com gpg: keyserver timed out gpg: keyserver receive failed: keyserver error ``` Whether this is my local network issue or something wrong for keyserver.ubuntu.com?

lijiachuan (Mon, 08 Oct 2018 16:30:42 GMT):
I am using a VirtualBox with Ubuntu 16 for this

lijiachuan (Mon, 08 Oct 2018 17:04:58 GMT):
hi @lovesh , I have some more questions after reading the get start document, could you please provide your comments for these questions? Thanks. 1. Whether the credential verifier will store the credential in their wallet? For example, Acme will request some credentials from Alice, such as first name, last name, ssn, etc. Once Alice provide the proof for this(proof also includes Alice's first name, last name, ssn, etc), whether the verifier Acme will store this information into its wallet? If yes, what is the difference between this new personal information storing with traditional database storing, which means Acme stored this information in their own company's database 2. Alice found there are two credentials can meet Loan-Application-KYC Proof Request(one was provided from Faber, the other was provided from Acme), how does she choose which credential she will use as the proof? This was not mentioned in the documentation

lovesh (Mon, 08 Oct 2018 20:11:39 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=dkPeDRXBZWw9RE7cX) @greg2git We need a key value store for persistence. Rocksdb is not a hard dependency, you can change the config file to work with leveldb but that will be suboptimal

lovesh (Mon, 08 Oct 2018 20:17:15 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=qkPTEgky9HR7ACBZh) @lijiachuan Hi. 1. Alice does not transfer the credential to the verifier Acme but only a proof computed from the credential. Acme might store this information but they cannot use this info to create another proof to give to another verifier. 2. Alice will look at the proof request and then decide which credential needs to be used. This decision will be influenced by conditions in the proof request, so the proof request can specify that it needs proof of age only from a certain credential or certain issuer or proof of age from a credential that specifies the social security number or something similar

only1chanakya (Tue, 09 Oct 2018 12:09:35 GMT):
Has joined the channel.

only1chanakya (Tue, 09 Oct 2018 12:10:04 GMT):
Has anyone setup Hyperledger-Indy on a Windows machine? Any pointers would be appreciated.

dono (Tue, 09 Oct 2018 15:19:48 GMT):
Anyone here willing to help out a beginner with the indy-sdk python wrapper

kdenhartog (Tue, 09 Oct 2018 18:09:08 GMT):
@dono list out your questions and I'll try to get back ASAP

kdenhartog (Tue, 09 Oct 2018 18:09:08 GMT):
@dono could you add your questions here and I'll try to get back ASAP (@ me for faster responses)

dono (Tue, 09 Oct 2018 18:14:01 GMT):
@kdenhartog Thanks, I was trying to run pytest on the SDK Python tests and was receiving a bunch of errors, most of them being function not found errors. I'm running Windows and set my Path environment variable to point to libindy_1.6.7/lib, python3-indy is properly installed with pip. Am I missing something? For now I'm trying it out on a Ubuntu 16.04 vm

haggs (Wed, 10 Oct 2018 01:57:09 GMT):
Have a stack trace by chance Dono? Did you try Pip3?

haggs (Wed, 10 Oct 2018 01:58:00 GMT):
@dono When you go to `python -v` in your unix terminal (not command prompt, something like git bash or the like) what happens? Same with pip

haggs (Wed, 10 Oct 2018 01:58:17 GMT):
I'd post the errors regardless!

dono (Wed, 10 Oct 2018 02:26:16 GMT):
hey @haggs I get a lot of import statements and at the end Python 3.7.1rc1 (v3.7.1rc1:2064bcf6ce, Sep 26 2018, 15:15:36) [MSC v.1914 64 bit (AMD64)] on win32 Type "help", "copyright", "credits" or "license" for more information.

dono (Wed, 10 Oct 2018 02:29:19 GMT):
when I type pip -v i get the standard help output

dono (Wed, 10 Oct 2018 02:29:34 GMT):
I've been using git bash so no need to worry about that

haggs (Wed, 10 Oct 2018 02:43:44 GMT):
@dono, I doubt it matters, but the current version of the python wrapper requires 3.6 vs 3.7. 3.7 Is quite new (a month or two) so I would recommending downgrading from your release candidate (3.7.*1rc1*) to 3.6

haggs (Wed, 10 Oct 2018 02:44:30 GMT):
Mind posting the errors for functions?

haggs (Wed, 10 Oct 2018 02:44:50 GMT):
Also maybe try `pip3`?

dono (Wed, 10 Oct 2018 02:45:22 GMT):
well its a lot of output, I can take a screenshot of what fits on one page length

dono (Wed, 10 Oct 2018 02:45:34 GMT):
pip3 -v does the same thing

haggs (Wed, 10 Oct 2018 02:46:18 GMT):
Do you have Docker or just a VirtualBox?

haggs (Wed, 10 Oct 2018 02:47:15 GMT):
\[FROM ubuntu:16.04 ARG uid=1000 RUN apt-get update && \ apt-get install -y \ pkg-config \ libssl-dev \ libgmp3-dev \ curl \ build-essential \ libsqlite3-dev \ cmake \ git \ python3.5 \ python3-pip \ python-setuptools \ apt-transport-https \ ca-certificates \ debhelper \ wget \ devscripts \ libncursesw5-dev \ libzmq3-dev \ zip \ unzip \ jq # install nodejs and npm RUN curl -sL https://deb.nodesource.com/setup_8.x | bash - RUN apt-get install -y nodejs RUN pip3 install -U \ pip \ setuptools \ virtualenv \ twine \ plumbum \ deb-pkg-tools RUN cd /tmp && \ curl https://download.libsodium.org/libsodium/releases/libsodium-1.0.14.tar.gz | tar -xz && \ cd /tmp/libsodium-1.0.14 && \ ./configure --disable-shared && \ make && \ make install && \ rm -rf /tmp/libsodium-1.0.14 RUN apt-get update && apt-get install openjdk-8-jdk -y ENV JAVA_HOME /usr/lib/jvm/java-8-openjdk-amd64 RUN apt-get update && apt-get install -y maven RUN apt-get install -y zip RUN useradd -ms /bin/bash -u $uid indy USER indy RUN curl https://sh.rustup.rs -sSf | sh -s -- -y --default-toolchain 1.26.0 ENV PATH /home/indy/.cargo/bin:$PATH WORKDIR /home/indy\]

haggs (Wed, 10 Oct 2018 02:47:20 GMT):
FROM ubuntu:16.04 ARG uid=1000 RUN apt-get update && \ apt-get install -y \ pkg-config \ libssl-dev \ libgmp3-dev \ curl \ build-essential \ libsqlite3-dev \ cmake \ git \ python3.5 \ python3-pip \ python-setuptools \ apt-transport-https \ ca-certificates \ debhelper \ wget \ devscripts \ libncursesw5-dev \ libzmq3-dev \ zip \ unzip \ jq # install nodejs and npm RUN curl -sL https://deb.nodesource.com/setup_8.x | bash - RUN apt-get install -y nodejs RUN pip3 install -U \ pip \ setuptools \ virtualenv \ twine \ plumbum \ deb-pkg-tools RUN cd /tmp && \ curl https://download.libsodium.org/libsodium/releases/libsodium-1.0.14.tar.gz | tar -xz && \ cd /tmp/libsodium-1.0.14 && \ ./configure --disable-shared && \ make && \ make install && \ rm -rf /tmp/libsodium-1.0.14 RUN apt-get update && apt-get install openjdk-8-jdk -y ENV JAVA_HOME /usr/lib/jvm/java-8-openjdk-amd64 RUN apt-get update && apt-get install -y maven RUN apt-get install -y zip RUN useradd -ms /bin/bash -u $uid indy USER indy RUN curl https://sh.rustup.rs -sSf | sh -s -- -y --default-toolchain 1.26.0 ENV PATH /home/indy/.cargo/bin:$PATH WORKDIR /home/indy

dono (Wed, 10 Oct 2018 02:47:21 GMT):
I was using Docker when this issue happened, but I don't think it matters because it doesn't seem to be a problem with connecting to the local pool i ran, but gimmie a sec i'm getting the screenshot now

haggs (Wed, 10 Oct 2018 02:47:33 GMT):
Tru that script

haggs (Wed, 10 Oct 2018 02:47:45 GMT):
'To run on your machine, make sure you have docker, cd into this directory, and then build the image: docker build -t ubuntu-1604 . -t is added for tag for the image After the docker image has been built and tagged: docker run -it $NAMEOFTAG -i flag is added for interactive to enter the entrypoint of the shell Please make any questions or PR's if this doesn't work for you'

dono (Wed, 10 Oct 2018 02:48:07 GMT):
That is the indy-pool dockerfile right?

dono (Wed, 10 Oct 2018 02:48:15 GMT):
I had no problems running the indy_pool image

dono (Wed, 10 Oct 2018 02:49:00 GMT):

pytest_issue.PNG

dono (Wed, 10 Oct 2018 02:49:23 GMT):
@haggs ^ thats a snippet of the output from pytest, the last of it specifically

dono (Wed, 10 Oct 2018 02:49:57 GMT):
shows the same thing whether indy_pool is running or not

dono (Wed, 10 Oct 2018 02:50:06 GMT):
I can try downgrading to python 3.6

dono (Wed, 10 Oct 2018 02:51:27 GMT):
should 3.6.7rc1 work?

haggs (Wed, 10 Oct 2018 02:53:17 GMT):
I'd grab a non release0candidate

dono (Wed, 10 Oct 2018 02:53:40 GMT):
kk

dono (Wed, 10 Oct 2018 02:55:54 GMT):
Does that snippet mean anything to you? I've never used pytest before

haggs (Wed, 10 Oct 2018 02:57:30 GMT):
Did your run `pip install -U pytest1

haggs (Wed, 10 Oct 2018 02:57:30 GMT):
Did your run `pip install -U pytest`

dono (Wed, 10 Oct 2018 02:58:05 GMT):
yes

haggs (Wed, 10 Oct 2018 02:58:18 GMT):
Then `piytest --version`

haggs (Wed, 10 Oct 2018 02:58:18 GMT):
Then `pytest --version`

dono (Wed, 10 Oct 2018 02:58:31 GMT):
Right now I'm completely uninstalling Python and reinstalling with version 3.6.6

haggs (Wed, 10 Oct 2018 02:58:48 GMT):
A new terminal window might help

dono (Wed, 10 Oct 2018 02:58:55 GMT):
k

dono (Wed, 10 Oct 2018 02:59:29 GMT):
okay with fresh install what are requirements

dono (Wed, 10 Oct 2018 02:59:45 GMT):
upgraded pip, python3-indy, pytest?

haggs (Wed, 10 Oct 2018 03:01:16 GMT):
yep, don't be afraid to overdo the 3's as you might have an env variable pointing the other way

dono (Wed, 10 Oct 2018 03:01:28 GMT):
kk

haggs (Wed, 10 Oct 2018 03:02:49 GMT):
asyncio is brand new to a brand new change in major language, that's my vues

haggs (Wed, 10 Oct 2018 03:02:49 GMT):
asyncio is brand new to a brand new change in major language, that's my guess

dono (Wed, 10 Oct 2018 03:03:26 GMT):
So I have this:

dono (Wed, 10 Oct 2018 03:03:27 GMT):
$ py -m pip --version pip 18.1 from C:\Users\Brandon\AppData\Local\Programs\Python\Python36\lib\site-packages\pip (python 3.6)

dono (Wed, 10 Oct 2018 03:03:57 GMT):
wont let me run pip without py -m and git bash wont recognize python, python3, or py3 as a command only py

haggs (Wed, 10 Oct 2018 03:04:28 GMT):
Try adding it to your shell

haggs (Wed, 10 Oct 2018 03:04:33 GMT):
Or restarting your shell

dono (Wed, 10 Oct 2018 03:04:41 GMT):
I did select install py launcher during install which is probably why

dono (Wed, 10 Oct 2018 03:05:00 GMT):
should I remove and add python to Path just incase

haggs (Wed, 10 Oct 2018 03:05:18 GMT):
If it's pointing to a different library version, then yes

dono (Wed, 10 Oct 2018 03:05:50 GMT):
well it can only be pointing to 3.6.6 because thats all i have installed right now

haggs (Wed, 10 Oct 2018 03:06:30 GMT):
Well assuming your env vars updated, a restart might help but a `--python --version` should be for sure

haggs (Wed, 10 Oct 2018 03:06:30 GMT):
Well assuming your env vars updated, a restart might help but a `python --version` should be for sure

dono (Wed, 10 Oct 2018 03:06:45 GMT):
right, I should restart. Gimmie a second

haggs (Wed, 10 Oct 2018 03:07:03 GMT):
before you do

haggs (Wed, 10 Oct 2018 03:07:32 GMT):
try `whereis python`

dono (Wed, 10 Oct 2018 03:08:19 GMT):
too late

dono (Wed, 10 Oct 2018 03:08:21 GMT):
lol

dono (Wed, 10 Oct 2018 03:08:31 GMT):
whereis command not found

haggs (Wed, 10 Oct 2018 03:08:54 GMT):
Unix command it must be entirely

dono (Wed, 10 Oct 2018 03:09:04 GMT):
python and python3 not recognized as commands still, let me check my Path

haggs (Wed, 10 Oct 2018 03:09:22 GMT):
`where` is the Windows equivalent

dono (Wed, 10 Oct 2018 03:09:33 GMT):
Seems like the installer didnt even add it to the Path variable because py launcher was installed

haggs (Wed, 10 Oct 2018 03:09:41 GMT):
hmm

dono (Wed, 10 Oct 2018 03:09:45 GMT):
I

haggs (Wed, 10 Oct 2018 03:09:47 GMT):
restart shell

haggs (Wed, 10 Oct 2018 03:10:20 GMT):
I don't have a windows machine close by to test this, but I have ran into issues with 3.7

dono (Wed, 10 Oct 2018 03:10:47 GMT):
well worse comes to worse I'll just use a ubuntu virtual machine

haggs (Wed, 10 Oct 2018 03:11:10 GMT):
That or a docker image are great, just watch out for the port mapping!

dono (Wed, 10 Oct 2018 03:11:27 GMT):
tomorrow I'm getting help from Thomas from the message list I think you saw\

dono (Wed, 10 Oct 2018 03:11:39 GMT):
Yeah I could try that too

haggs (Wed, 10 Oct 2018 03:12:13 GMT):
Still, I wish the message were more verbose. Mind writing down all you did and when you get it sending me what happepened? I'll replicated in the morning

dono (Wed, 10 Oct 2018 03:13:22 GMT):
``` ```

dono (Wed, 10 Oct 2018 03:21:44 GMT):
First I installed latest python 3.7.1rc1 using Windows executable installer, using recommended settings and selecting add Python to Path. installed the required pip packages, python3-indy and asyncio I can't remember the exact command but I'm pretty sure I only had to enter *pip install*, not *py -m pip install*. Then I downloaded latest stable release of libindy for Windows and unzipped into a directory in Program Files, pointed Path to it - C:\Program Files\libindy_1.6.7\lib Then I git cloned indy-sdk, built the indy-pool.dockerfile image, and ran indy_pool image on loopback exactly as they show to do. Finally I cd over to indy-sdk/wrappers/python ran the command *pytest* nothing else and I got the output shown.

dono (Wed, 10 Oct 2018 03:22:20 GMT):
Sorry I can't give you better step by step instructions it was earlier in the day and I go through trial and error so fast I usually don't keep track of exact steps I did

dono (Wed, 10 Oct 2018 03:22:28 GMT):
@haggs ^

dono (Wed, 10 Oct 2018 03:25:20 GMT):
its 64bit windows and versions of python btw

haggs (Wed, 10 Oct 2018 03:43:23 GMT):
My only guess is that you installed release channel and another release channel https://github.com/hyperledger/indy-sdk#windows. @nhelmy May be able to help, but in general running a docker version of `libindy` . Tough to say as I don't own a windows machine. I'll look again tomorrow. I'll leave this for reference: `pip3 install -U pip`, https://github.com/hyperledger/indy-sdk/blob/master/doc/windows-build.md is your best bet, I'll check back through tomorrrow and possibly do a screenshare as I'm stumped myself. A VM or cloud ubuntu box might make it easier, still a correct Docker setup should mitigate most of this besides network mapping. Best of luck!

haggs (Wed, 10 Oct 2018 03:43:23 GMT):
My only guess is that you installed release channel and another release channel https://github.com/hyperledger/indy-sdk#windows. @nhelmy May be able to help, but in general running a docker version of `libindy` in ubuntu 16.04 is ideal . Tough to say as I don't own a windows machine. I'll look again tomorrow and I'll leave this for reference: `pip3 install -U pip`, https://github.com/hyperledger/indy-sdk/blob/master/doc/windows-build.md is your best bet. A VM or cloud ubuntu box might make it easier, still a correct Docker setup should mitigate most of this besides network mapping. Best of luck, seems like you're close and followed all the steps, so I'm happy to share a screen tomorrow to get to the bottom of this.

sklump (Wed, 10 Oct 2018 10:50:33 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=LftfhyeTbMPiYebaq) @dono You could try pip 9.0.x: pip 10+ broke indy pytests.

lijiachuan (Wed, 10 Oct 2018 11:40:39 GMT):
hi, I have one question about where to host the wallet for individual user, normally when we talk about a digital wallet, that will be a mobile application, but from the get started example, sometimes the initial connection might be triggered from a website, such as Alice click "Get Transcript" button on Faber's website, this makes the communication between website and mobile app difficult, if Alice is using a web browser in her laptop. So I would like to get some best practice what is the best way to build the wallet for individual user and organization

haggs (Wed, 10 Oct 2018 13:30:17 GMT):
Ah yes @sklump forgot about that. I downgraded to `pip 9.0.3`

haggs (Wed, 10 Oct 2018 13:31:59 GMT):
`sudo pip install --force-reinstall pip==9.0.3` @dono

dono (Wed, 10 Oct 2018 19:38:18 GMT):
@haggs @sklump I've tried your suggestions, it didn't change anything. @twshelton and I have been trying to find the root cause and we think it is an issue with libindy not recognizing/finding dependencies on my system

dono (Wed, 10 Oct 2018 19:47:28 GMT):
For now I'm going to run through the Build from source process and then try again

haggs (Wed, 10 Oct 2018 20:54:55 GMT):
Alright sounds good

haggs (Wed, 10 Oct 2018 20:54:55 GMT):
Alright sounds good and good luck!

sandeepsethia62 (Thu, 11 Oct 2018 01:48:03 GMT):
Has joined the channel.

sandeepsethia62 (Thu, 11 Oct 2018 01:52:36 GMT):
This is my first time building Indy locally and running the test. as of now, when I run the test using the guide on this location (https://github.com/hyperledger/indy-sdk/blob/master/doc/ubuntu-build.md), I have a failing test case (test ledger_demo_works ...FAILED ). I am building the master branch as of now. Does anybody has a hint?

sandeepsethia62 (Thu, 11 Oct 2018 06:07:02 GMT):
another question: can somebody point me to a document or link which list "how to store the Indy data on a regular relational database instead of sqlite" ? my requirement is I want to store the data in either mysql or postgreSQL

srunpengsreang (Thu, 11 Oct 2018 07:03:35 GMT):
#indy-sdk Hello guys here

lijiachuan (Thu, 11 Oct 2018 08:55:49 GMT):
hi, I followed this start node guideline step 1, https://github.com/hyperledger/indy-node/blob/master/docs/start-nodes.md "Install Indy Node", I complete running that 4 lines of command in Ubuntu, and when I try to run the next step "Initialize Node to be included into the Network", it has one description about updating the network name in config file "NETWORK_NAME={network_name} where {network_name} matches the one in genesis transaction files above", but I didn't generate any genesis transaction files before this, anyone can let me know where is the step guide to generate it?

dbluhm (Thu, 11 Oct 2018 13:16:16 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=2PEM45s3Q9k2zmr4Q) @sandeepsethia62 That error is likely because the tests were unable to connect to a local pool network. This is done using docker typically. Instructions can be found in the indy-sdk main README.

osesov (Thu, 11 Oct 2018 15:37:32 GMT):
Has joined the channel.

dhuseby (Thu, 11 Oct 2018 16:08:55 GMT):
https://medium.com/@charltoons/tutorial-compiling-rust-to-wasm-b70d4a7d428f

dhuseby (Thu, 11 Oct 2018 16:09:17 GMT):
For those who are curious about rust and wasm.

dhuseby (Thu, 11 Oct 2018 16:13:36 GMT):
How much web can the rust-web rust if the rust-web could rust web? It can rust as much web as a rust-web can rust since the rust-web can rust web.

esplinr (Thu, 11 Oct 2018 19:39:36 GMT):
A recording of the sprint demo for the Evernym Indy teams: https://www.youtube.com/watch?v=jt9VLhs885o&list=PLRp0viTDxBWGLdZk0aamtahB9cpJGV7ZF&index=2&t=0s

swcurran (Thu, 11 Oct 2018 22:18:06 GMT):
@sandeepsethia62 - we (BC Gov) are running Indy using Postgres for wallet storage. @ianco is currently working on a PR to indy-sdk to add a pluggable wallet implementation based on that work. Stay tuned - should be a PR in about a week.

xnopre (Fri, 12 Oct 2018 09:43:11 GMT):
Hi all. Based on this script (https://github.com/hyperledger/indy-sdk/blob/master/doc/getting-started/getting-started.ipynb), I have rewritten the story until "Alice -> Store 'Job-Certificate' Credential", with NodeJs SDK wrapper, and separating each actors (Steward, Alice, Faber and Acme) in different processes, especially to separate operations for each actor, and to implement (and show) communications and data exchanged between actors. All is packaged to run in different docker containers, with "make" commands to simply run the script. I think it can be interesting to share it, as example (demo). And I can share it :-) . But the question is who or where ? Perhaps it could be integrated in https://github.com/hyperledger/indy-sdk with a pull request ? Or elsewhere ? What do you think about ? Thanks :-) [posted on #indy and #indy-sdk ]

kdenhartog (Fri, 12 Oct 2018 09:54:14 GMT):
I just put together a simple development environment to run IndySDK code in a docker environment. If anyone is having trouble getting IndySDK setup or wants an easy way to develop with the SDK on any OS, check out this repo. https://github.com/kdenhartog/Indy-dev

kdenhartog (Fri, 12 Oct 2018 09:54:14 GMT):
I just put together a simple development environment to run IndySDK code in a docker environment. If anyone is having trouble getting IndySDK setup or wants an easy way to develop with the SDK on any OS, check out this repo. https://github.com/kdenhartog/Indy-dev Right now it only supports Python development, but I'm planning on getting all the wrappers supported.

xnopre (Fri, 12 Oct 2018 12:42:58 GMT):
@kdenhartog Effectively, there are different subjects (interests) in my project. 1/ I can share information to run NodeJs script in indy environment, or contribute in your project 2/ My scripts (1 for each actor of the story) are samples for how using NodeJs SDK Indy wrapper 3/ And the entire project show communication and data between each actors (no present in https://github.com/hyperledger/indy-sdk/blob/master/doc/getting-started/getting-started.ipynb) My question is how/where to share this to be visible and foundable by a maximum of indy SDK users. For example, I didn't known your repo. So where is the best place to share this code to be visible and foundable by future indy SDK users ? Perhaps in "official" indy SDK repo ? What is the opinion of others ? Thanks :-)

kdenhartog (Fri, 12 Oct 2018 16:05:17 GMT):
This sounds like it might be a good getting started guide for the Indy-agents repository specifically the NodeJS reference agent. @mhailstone any thoughts on pulling something like this in to have an example of agents communicating?

kdenhartog (Fri, 12 Oct 2018 16:05:54 GMT):
@xnopre at the very least, adding a reference to the Indy wiki would be useful, so others can find it.

Sean_Bohan (Fri, 12 Oct 2018 19:07:54 GMT):
folks: Indy WG call from yesterday - the video is posted here; https://drive.google.com/drive/u/0/folders/1AwHWN95KmSEi5fijraID0tFFMzYHoMwt?ogsrc=32 Agenda: - Recap of the Hyperledger Hackfest Montreal (last week) with all weighing in - Shared Crypto Library, aka CryptoLib, with DaveH (Linux Foundation), NathanG (Sovrin Foundation), MikeL (Sovrin Foundation), all - Building bridges between Indy and Fabric communities with TroyR (SecureKey) and NathanG (Sovrin Foundation), all

Sean_Bohan (Fri, 12 Oct 2018 19:07:54 GMT):
folks: Indy WG call from yesterday - the video is posted here; https://drive.google.com/drive/u/0/folders/1AwHWN95KmSEi5fijraID0tFFMzYHoMwt?ogsrc=32 Agenda: - Recap of the Hyperledger Hackfest Montreal (last week) with all weighing in - Shared Crypto Library, aka CryptoLib, with DaveH (Linux foundation), NathanG (Sovrin), MikeL (Sovrin), all - Building bridges between Indy and Fabric communities with TroyR (SecureKey) and NathanG (Sovrin), all

mhailstone (Fri, 12 Oct 2018 19:54:17 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=be5bb468-ecc6-4f52-99ee-13b733625c9a) @kdenhartog Yes. I think we need to consider a better getting started guide for the agent community complimentary to the getting started content for indy-sdk pages in GitHub. I'm hoping next week's call will be a springboard for some of that.

ahern077 (Fri, 12 Oct 2018 20:54:40 GMT):
Has joined the channel.

dono (Fri, 12 Oct 2018 22:27:53 GMT):
@kdenhartog Running the indy-dev image you provided, when I get into the shell the home directory is empty. Can't cd python, did you mean to git clone indy-sdk and then run the getting started?

dono (Fri, 12 Oct 2018 23:09:59 GMT):
@kdenhartog I created a virtual machine so that I have more control and followed the sequence of actions from your indy-dev dockerfile. Ran getting started and everything finished without error. But running pytest inside the indy-sdk/wrappers/python directory shows many errors with all test modules. Is it just cause they haven't been updated and I should ignore them, or is there something I'm missing?

kdenhartog (Fri, 12 Oct 2018 23:37:35 GMT):
@dono Likely I haven't added the dependencies to get the how to's working yet. I'll get that fixed, but if you run into any other problems, please file an issue or a PR. This is still very early stages so I expect more problems m

dono (Fri, 12 Oct 2018 23:38:06 GMT):
Okay, thank you

SumanPapanaboina (Sun, 14 Oct 2018 15:36:28 GMT):
Has joined the channel.

calvingerling (Sun, 14 Oct 2018 22:19:09 GMT):
Has joined the channel.

calvingerling (Sun, 14 Oct 2018 22:24:28 GMT):
Hi everyone, my name is Calvin Gerling and I'm doing some research work in the self-sovereign identity topic. I intend to create a mobile client to the Hyperledger Indy network but I'm having some problems to get a network up and running inside a Digital Ocean Droplet, it's an Ubuntu 16.04 x64. Can someone help me?

lijiachuan (Mon, 15 Oct 2018 15:09:46 GMT):
hi everyone, is there any documentation about how to build a wallet application for Indy project?

baconsandwich (Mon, 15 Oct 2018 15:40:49 GMT):
Hey, this might be a dumb question, but after working with indy for a few months I just wondered: **Why do we even need a DID?** It's the first 16 bytes of the verification key. Why not just use that? Yeah, it's longer but its the "source" of the identity and isn't this breaking DRY? The only advantage I can see is that the DID is shorter and therefore more human readable.

baconsandwich (Mon, 15 Oct 2018 15:42:58 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=EhmAHYS4xmqnRhGdd) @lijiachuan This might be a starting point. Follow the other steps to get a feeling for what you need to do.

jljordan_bcgov (Mon, 15 Oct 2018 15:46:01 GMT):
@baconsandwich in short .. key rotation

jljordan_bcgov (Mon, 15 Oct 2018 15:46:01 GMT):
@baconsandwich in short .. I believe an important piece is key rotation

baconsandwich (Mon, 15 Oct 2018 15:52:35 GMT):
ok, that makes sense. so DID is derived "arbitrarily" from the original verkey, but looses this relation as soon as the verkey is changed.

baconsandwich (Mon, 15 Oct 2018 15:52:35 GMT):
ok, this makes sense. so the DID is derived "arbitrarily" from the original verkey, but looses this relation as soon as the verkey is changed and then only serves as "some string" to identify that NYM

baconsandwich (Mon, 15 Oct 2018 15:53:59 GMT):
would this be a sensible description

baconsandwich (Mon, 15 Oct 2018 15:53:59 GMT):
would this be a sensible description?

arjunkhera (Mon, 15 Oct 2018 16:00:57 GMT):
Hi everyone, i was trying out the getting started example of Indy SDK and was able to run the demo successfully a few times. However, a few days ago when I tried running the sample again, it failed with a few errors, the first being error 306 which in error.py reads as # Attempt to create pool ledger config with name used for another existing pool PoolLedgerConfigAlreadyExistsError = 306,

arjunkhera (Mon, 15 Oct 2018 16:01:24 GMT):
Can somebody help find out the reason for why this is happening ?

arjunkhera (Mon, 15 Oct 2018 16:01:24 GMT):
Can somebody help me find out the reason for why this is happening ?

arjunkhera (Mon, 15 Oct 2018 16:03:31 GMT):
Just fyi, I clean up the containers running the local pool of nodes after I am finished running the sample

haggs (Mon, 15 Oct 2018 16:10:25 GMT):
@arjunkhera Mind doing a ```ls -als``` in your working directory?

haggs (Mon, 15 Oct 2018 16:13:07 GMT):
Not sure if you have VCS highlighting or not but if there are any hidden files or directories you might want to delete those if they weren't there before

haggs (Mon, 15 Oct 2018 16:40:47 GMT):
@baconsandwich The DID can be arbitrarily set from the original verkey first 16 bits OR explicitly provided. https://github.com/hyperledger/indy-sdk/blob/4e8d9a47e2a7d0368c26c26c0209caa6965c7b60/libindy/src/api/did.rs#L21, but at its heart it's essentially trustless.

haggs (Mon, 15 Oct 2018 16:40:47 GMT):
@baconsandwich The DID can be arbitrarily set from the original verkey first 16 bits OR explicitly provided. https://github.com/hyperledger/indy-sdk/blob/4e8d9a47e2a7d0368c26c26c0209caa6965c7b60/libindy/src/api/did.rs#L21, but at its heart it's essentially trustless..at first, until verifiable claims are made by the issuer. After key rotation a DID pairwise connection in a secured wallet can keep track of those changes and be decentrally owned and signed. At least...that's my understanding of it! Hope that helps :D

arjunkhera (Mon, 15 Oct 2018 18:14:04 GMT):
@haggs , I installed the sdk and node repositories again, yet the issue persists. I am quite bamboozled as to why this is the case, as not only are all the containers removed, but also I have reinstalled the respective folders.

arjunkhera (Mon, 15 Oct 2018 18:16:57 GMT):
ubuntu 16.04

calvingerling (Mon, 15 Oct 2018 22:35:40 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=9KRfevHsYkFHiN4xm) @arjunkhera Wich guide are you following to run the demo?

kdenhartog (Tue, 16 Oct 2018 02:45:07 GMT):
@arjunkhera have you removed the hidden directory in your home directory? I think its either .indy or .indy_client. alternatively, check out https://github.com/kdenhartog/indy-dev

kdenhartog (Tue, 16 Oct 2018 02:46:02 GMT):
@baconsandwich thats a sensible description

kdenhartog (Tue, 16 Oct 2018 03:08:08 GMT):
@lijiachuan that is very nearly what agents are intended to do in a general way. I'd suggest looking at the indy-HIPE repository and it's pull requests.

MohitJuneja (Tue, 16 Oct 2018 03:25:51 GMT):
Has joined the channel.

lijiachuan (Tue, 16 Oct 2018 04:07:23 GMT):
thanks @baconsandwich and @jljordan_bcgov for your comments.

lijiachuan (Tue, 16 Oct 2018 04:10:00 GMT):
I have another question about how to self-issue some identities. For example, for such as user's first name, last name or birth of date this type of information, based on current reality, we can not request any organization to use the new technology and issue this information to user, so in my mind, user should have one something like a "self-issued" identities and can use it in later, is there such as concept in Indy?

lijiachuan (Tue, 16 Oct 2018 04:23:26 GMT):
I think the get start document is based on Faber already has basic information of Alice, so they can issue the credential directly to Alice, but just in case if there is no such organization/user/thing can know this initially, how should we handle the initialization

knagware9 (Tue, 16 Oct 2018 07:04:18 GMT):
Has joined the channel.

alexeidebono (Tue, 16 Oct 2018 11:41:40 GMT):
Has joined the channel.

alexeidebono (Tue, 16 Oct 2018 11:42:45 GMT):
Hi all - I'm try to do the first tutorial in the Indy Github on python and I keep getting this error when I just run the template file. Any idea what I may be doing wrong ?

alexeidebono (Tue, 16 Oct 2018 11:42:52 GMT):

Clipboard - October 16, 2018 1:42 PM

sklump (Tue, 16 Oct 2018 12:10:23 GMT):
Looks like you need to `export LD_LIBRARY_PATH` or `sudo cp target/release/libindy.so /usr/lib`

baconsandwich (Tue, 16 Oct 2018 14:37:46 GMT):
What purpose does the alias field in a NYM transaction serve? Wouln't any semantic information like "John Smith" or "NYM used for user agent X" compromise the anonymity of that DID? [The documentation](https://github.com/hyperledger/indy-node/blob/master/docs/transactions.md#nym) only says `NYM's alias.`

baconsandwich (Tue, 16 Oct 2018 14:37:46 GMT):
What purpose does the alias field in a NYM transaction serve? Wouldn't any semantic information like "John Smith" or "NYM used for user agent X" compromise the anonymity of that DID? [The documentation](https://github.com/hyperledger/indy-node/blob/master/docs/transactions.md#nym) only says `NYM's alias.`

sklump (Tue, 16 Oct 2018 15:02:20 GMT):
Friendly name for display

alexeidebono (Tue, 16 Oct 2018 16:38:35 GMT):
any reason why i keep getting a time out error when opening a ledger ? I'm running on mac osx and tried this with a local docker (127.0.0.1) and method 3 docker with network (10.0.0.2) - i also tried to change the genesis file to use 127.0.0.1 on one try but still got the time out error. any ideas what can work here ?

jedrazb (Tue, 16 Oct 2018 17:03:41 GMT):
Has joined the channel.

baconsandwich (Tue, 16 Oct 2018 19:15:10 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=XdBZDtAWYKHRHyhZX) @sklump But its written to the ledger, isn't it? Why would you add a public "friendly name"?

msalmansaeedch786 (Tue, 16 Oct 2018 21:06:17 GMT):
Has joined the channel.

grapebaba (Wed, 17 Oct 2018 01:50:59 GMT):
Has joined the channel.

alexeidebono (Wed, 17 Oct 2018 09:11:43 GMT):
Hey all - I'm still trying to fix the timeout error i'm getting but no luck. Any idea what I can do to find out where I'm going wrong ?

alexeidebono (Wed, 17 Oct 2018 09:12:07 GMT):

current ports i have with method 3 (10.0.0.2 ip )

aronvanammers (Wed, 17 Oct 2018 09:25:51 GMT):
Has joined the channel.

sklump (Wed, 17 Oct 2018 10:20:10 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=dYcsNt8Fudew6RmQ2) @baconsandwich Yes, it's written to the ledger, as it's part of the identity of the owner. At one point it was a desirable feature. You don't have to specify anything.

ClearFoundation (Wed, 17 Oct 2018 14:57:43 GMT):
Hello, Now the concern is that if we move to STN then what will we have some already issuer and verifier, those will verify our data or should we move to directly on the live server?

Sean_Bohan (Wed, 17 Oct 2018 16:57:08 GMT):
Hyperledger Indy WG Call for tomorrow, 10/18 Agenda: - Housekeeping - Agents WG Friday Call: Getting Started/Approaches with Indy Agent Development - Tieto Demo/Discussion on their work with CORDA/INDY (Antti @ Tieto) - IIW Planning /Prep (All) - W3C TPAC Planning / Prep (All) When: 8am PT 9am MT 10am CT 11amET 4pmBST Where: Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/232861185

dwaite (Wed, 17 Oct 2018 17:51:58 GMT):
Has joined the channel.

twshelton (Wed, 17 Oct 2018 22:08:18 GMT):
@alexeidebono is this an environment you recently built? Can you provide a bit more detail about your setup?

asmallfurrywombat (Wed, 17 Oct 2018 22:26:22 GMT):
I'm curious about bootstrapping. Say someone starts a ledger which provides a proof for the entire US then, later, a more reliable ledger is created to provide the proof for Texas. Is there a planned mechanism for migration or federation?

karthikmohan91 (Thu, 18 Oct 2018 04:47:50 GMT):
Has joined the channel.

ShubhamSingh18 (Thu, 18 Oct 2018 10:37:26 GMT):
Has joined the channel.

ShubhamSingh18 (Thu, 18 Oct 2018 10:41:21 GMT):
can any ony tell me the steps to develop client application by setting indy-sdk

ShubhamSingh18 (Thu, 18 Oct 2018 10:50:56 GMT):
I have already setup indy node and indy sdk

lijiachuan (Thu, 18 Oct 2018 11:14:20 GMT):
Hi, can anyone provide a guidance that how one user initially store his/her claims information in his/her wallet but not let others issue it, and later he/she can consent to share which claims will be shared to others?

lijiachuan (Thu, 18 Oct 2018 11:17:22 GMT):
The scenario should be like this, there is one open market, it has one pre-defined schema/definition, anyone can register in this market and only basic claims will be shared in this market, only when one user wants to connect, then the registered user will decide whether share details or not.

ShubhamSingh18 (Thu, 18 Oct 2018 13:11:15 GMT):
Getting error while setting indy sdk

ShubhamSingh18 (Thu, 18 Oct 2018 13:11:15 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=tTLCDACeHSiSMR6hn) @twshelton Getting error while setting indy sdk

ShubhamSingh18 (Thu, 18 Oct 2018 13:13:32 GMT):
error: aborting due to previous error error: Could not compile `indy-cli`. To learn more, run the command again with --verbose. error: aborting due to previous error error: Could not compile `indy-cli`. To learn more, run the command again with --verbose.

ShubhamSingh18 (Thu, 18 Oct 2018 13:13:55 GMT):
while executing RUSTFLAGS=" -L ../libindy/target/debug" cargo build

ShubhamSingh18 (Thu, 18 Oct 2018 13:14:05 GMT):
Please help

arjunkhera (Thu, 18 Oct 2018 13:33:31 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=4366b663-d559-466b-a795-a21a20c757c2) @calvingerling [this one](https://github.com/hyperledger/indy-sdk/blob/master/doc/getting-started/getting-started.md)

arjunkhera (Thu, 18 Oct 2018 13:34:44 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=abaf3fca-2b77-43af-b00f-c3b66e98c916) @kdenhartog You are right, I found the .indy_client folder in the home directory and removed it. However, when i try to run the demo, this time it shows a different error [ 112 ].

arjunkhera (Thu, 18 Oct 2018 13:35:17 GMT):
What is more interesting is that following your repo to install the containers and run the demo, I get the same error

arjunkhera (Thu, 18 Oct 2018 13:35:47 GMT):

why2.png

arjunkhera (Thu, 18 Oct 2018 14:07:18 GMT):
I was finally able to run the demo. The thing is I had made my indy_pool image from the dockerfile of the latest commit of indy-sdk and it was not working. But using the indy_pool build from @kdenhartog dockerfile, it seems to work perfectly. Can't figure out why this is the case

twshelton (Thu, 18 Oct 2018 14:28:03 GMT):
@ShubhamSingh18 can you provide some more detail on your setup? Looks like you might be missing some dependencies

theoturner (Thu, 18 Oct 2018 15:01:08 GMT):
Has joined the channel.

theoturner (Thu, 18 Oct 2018 15:11:52 GMT):
I have a question regarding pairwise DIDs, if this is a good place to ask.

theoturner (Thu, 18 Oct 2018 15:12:51 GMT):
Is there potential for network spam? If I'm not mistaken, received requests are recorded at endpoints even if the request is not accepted. As such, could a malicious actor not make continued connection requests to a target, forcing each one to be recorded? Additionally, if an adversary were to set up both sides of a connection (i.e. both sides of the pairwise DID), is there potential for network spam by continually changing the pairwise DID? Perhaps this is DKMS open issue #8? https://github.com/hyperledger/indy-sdk/blob/master/doc/design/005-dkms/DKMS%20Design%20and%20Architecture%20V3.md#11-open-issues Cheers!

theoturner (Thu, 18 Oct 2018 15:13:02 GMT):
@aronvanammers

kdenhartog (Thu, 18 Oct 2018 15:28:06 GMT):
@arjunkhera it has to do with docker network config stuff. When using the docker compose stuff a network is generated on docker, where as witg mine I'm relying on the images using the host network and just changing ports.

kdenhartog (Thu, 18 Oct 2018 15:34:07 GMT):
@arjunkhera it has to do with docker network config stuff. When using the docker compose stuff a network is generated with docker-compose while mine is relying on the images using the host network and just changing ports. As far as the 112 error it happens because the SDK and the pool ledger have to be in sync for data being passed back and forth and how the data and APIs are structured.

twshelton (Thu, 18 Oct 2018 15:52:36 GMT):
@kdenhartog someone was having problems with your new docker solution ... was getting tumeout when trying to run the getting_started_py file. Running on a mac. Any thoughts?

jlin (Thu, 18 Oct 2018 23:41:17 GMT):
I realize there's a big time shift, but is there a copy of the Corda slides from the WG call today?

kdenhartog (Fri, 19 Oct 2018 06:28:02 GMT):
@twshelton was it a 307 or a 306 error. I haven't seen a 307 error yet, but I'd be curious to figure out why it's not working. If its the 306 error, its because the .indy_client repo thats created needs to be deleted. I need to change the guides so thats done automatically at the beginning instead.

bartolomeu (Fri, 19 Oct 2018 09:18:30 GMT):
Has joined the channel.

bartolomeu (Fri, 19 Oct 2018 09:23:33 GMT):
Can someone justify why it is not advisable send the verkey together with the DID in a connection request. I'm referring to the Step 4, point 5 in the Getting Started Guide (https://github.com/hyperledger/indy-sdk/blob/master/doc/getting-started/getting-started.md#explore-the-code). Instead, the DID and the verkey are first stored in the ledger to be later searched by the Faber, something that could be done directly. What am I missing?

zixian5 (Sun, 21 Oct 2018 02:33:02 GMT):
Has joined the channel.

zixian5 (Sun, 21 Oct 2018 02:52:29 GMT):
I have a question. When i run this code(https://github.com/hyperledger/indy-sdk/blob/master/samples/python/src/anoncreds.py), i find that Prover and issuer don't connect with each other. Prover's funtcion use the issuer's information directly. So if prover and issuer use two clients , are there any api i can use to communite in the jdk ? Or i should reach it by myself?

yusra (Sun, 21 Oct 2018 06:40:32 GMT):
Has joined the channel.

Abdul_khaliq (Sun, 21 Oct 2018 10:43:03 GMT):
Has joined the channel.

Abdul_khaliq (Sun, 21 Oct 2018 10:43:47 GMT):
hi everyone i need some help i m new to Hyperledger can anyone suggest me the best framework for python language

zixian5 (Sun, 21 Oct 2018 11:25:56 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=M7XZM3cyifg7KNCgR) @Abdul_khaliq what you use depends on what you want to do

Abdul_khaliq (Sun, 21 Oct 2018 12:03:23 GMT):
@zixian5 i want to built an application related to healthcare for medical record storing of the patient using hyperledger i have no idea which tool is best and which framework is good. i just know that smart contract data exchange system used in it. Even i dont know how to start this is my semester project. Can you guide me or help me in this regard.

zixian5 (Sun, 21 Oct 2018 12:12:18 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=Mxf8JCTBs4Z9P8jFi) @Abdul_khaliq fabric may be your first choice. U can't write the smart contract in Indy. It just for decentralize identity. Your can read the docs about fabric(https://hyperledger-fabric.readthedocs.io/en/release-1.3/) .Python is supported by fabric

rbole (Sun, 21 Oct 2018 13:53:48 GMT):
Has joined the channel.

ShubhamSingh18 (Mon, 22 Oct 2018 04:46:46 GMT):
Install Rust and rustup (https://www.rust-lang.org/install.html). Install required native libraries and utilities: apt-get update && \ apt-get install -y \ build-essential \ pkg-config \ cmake \ libssl-dev \ libsqlite3-dev \ libzmq3-dev \ libncursesw5-dev libindy requires the modern 1.0.14 version of libsodium but Ubuntu 16.04 does not support installation it's from apt repository. Because of this, it requires to build and install libsodium from source: cd /tmp && \ curl https://download.libsodium.org/libsodium/releases/libsodium-1.0.14.tar.gz | tar -xz && \ cd /tmp/libsodium-1.0.14 && \ ./configure --disable-shared && \ make && \ make install && \ rm -rf /tmp/libsodium-1.0.14 Build libindy git clone https://github.com/hyperledger/indy-sdk.git cd ./indy-sdk/libindy cargo build cd ..

ShubhamSingh18 (Mon, 22 Oct 2018 04:48:15 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=ogwH84Nmu6Z3caFfc) Start the pool of local nodes on 127.0.0.1:9701-9708 with Docker by running: docker build -f ci/indy-pool.dockerfile -t indy_pool . docker run -itd -p 9701-9708:9701-9708 indy_pool then

ShubhamSingh18 (Mon, 22 Oct 2018 04:48:57 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=ySikfiZriz3KNvch9) when i run this Run tests cd libindy RUST_TEST_THREADS=1 cargo test then i get erreor

ShubhamSingh18 (Mon, 22 Oct 2018 06:52:22 GMT):
I run this docker file https://github.com/hyperledger/indy-sdk/blob/master/libindy/ci/ubuntu.dockerfile and already setup the indy network I am getting error while executing the command Run integration tests: Start local nodes pool with Docker If you use this method then you have to specify the TEST_POOL_IP as specified below when running the tests. It can be useful if we want to launch integration tests inside another container attached to the same docker network. Run tests cd libindy RUST_TEST_THREADS=1 cargo test It is possible to change ip of test pool by providing of TEST_POOL_IP environment variable: RUST_TEST_THREADS=1 TEST_POOL_IP=10.0.0.2 cargo test Build indy-cli (Optional) indy-cli is dependent on libindy and should be built after it. cd cli/ RUSTFLAGS=" -L ../libindy/target/debug" cargo build If you have followed the instructions to build libindy above, the default build type will be debug Make sure to add the libindy to the path. Replace /path/to with the actual path to the libindy directory. Using bash: echo "export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/path/to/libindy/target/{BUILD TYPE}" >> ~/.bashrc sudo ldconfig source ~/.bashrc To run indy-cli, navigate to cli/target/debug and run ./indy-cli

ShubhamSingh18 (Mon, 22 Oct 2018 06:53:46 GMT):

indy error.png

sandeepsethia62 (Mon, 22 Oct 2018 07:43:25 GMT):
@ShubhamSingh18 I do not know the exact reason for this, but I also faced similar issue.. precisely this: `error: linking with `cc` failed: exit code: 1 = note: /home/sandy/opensource/indy-sdk-postgres/indy-sdk/cli/target/debug/deps/indy_cli-37ef1ab258dc7f72.1jy2bey351zfhh1v.rcgu.o: In function `indy_cli::libindy::logger::set_indy_logger': /home/sandy/opensource/indy-sdk-postgres/indy-sdk/cli/src/libindy/logger.rs:25: undefined reference to `indy_set_logger' collect2: error: ld returned 1 exit status`

sandeepsethia62 (Mon, 22 Oct 2018 07:43:25 GMT):
@ShubhamSingh18 I do not know the exact reason for this, but I also faced similar issue.. precisely this: `error: linking with `cc` failed: exit code: 1 = note: /home/sandy/opensource/indy-sdk-postgres/indy-sdk/cli/target/debug/deps/indy_cli-37ef1ab258dc7f72.1jy2bey351zfhh1v.rcgu.o: In function `indy_cli::libindy::logger::set_indy_logger': /home/sandy/opensource/indy-sdk-postgres/indy-sdk/cli/src/libindy/logger.rs:25: undefined reference to `indy_set_logger' collect2: error: ld returned 1 exit status`

sandeepsethia62 (Mon, 22 Oct 2018 07:43:25 GMT):
@ShubhamSingh18 I do not know the exact reason for this, but I also faced similar issue.. precisely this: `error: linking with `cc` failed: exit code: 1 = note: /home/sandy/opensource/indy-sdk-postgres/indy-sdk/cli/target/debug/deps/indy_cli-37ef1ab258dc7f72.1jy2bey351zfhh1v.rcgu.o: In function `indy_cli::libindy::logger::set_indy_logger': /home/sandy/opensource/indy-sdk-postgres/indy-sdk/cli/src/libindy/logger.rs:25: undefined reference to `indy_set_logger' collect2: error: ld returned 1 exit status`

cguerin (Mon, 22 Oct 2018 09:20:01 GMT):
Has joined the channel.

louisk (Mon, 22 Oct 2018 13:33:30 GMT):
Has joined the channel.

louisk (Mon, 22 Oct 2018 13:34:43 GMT):
hi good morning all

louisk (Mon, 22 Oct 2018 13:36:05 GMT):
playing with Indy a bit and working on contributing some docs & (maybe) an easier to drop in SDK for node.js

louisk (Mon, 22 Oct 2018 13:37:45 GMT):
encountering an issue when trying to create a new pairwise DID though and submit the signed NYM request

louisk (Mon, 22 Oct 2018 13:38:33 GMT):
I'm receiving a 212 / WalletItemNotFound error when i definitely have a created & opened wallet, as well as a DID from the `createAndStoreMyDId` method

louisk (Mon, 22 Oct 2018 13:38:33 GMT):
I'm receiving a 212 / WalletItemNotFound error when i definitely have a created & opened wallet, as well as a DID and verkey from the `createAndStoreMyDId` method

louisk (Mon, 22 Oct 2018 13:44:05 GMT):
with a request that looks like this: ```{"reqId":1540188203034500000,"identifier":"Th7MpTaRZVRYnPiabds81Y","operation":{"dest":"BHXxQCC7uLPPEbMLodeo7c","role":null,"type":"1","verkey":"someVerkeyHere"},"protocolVersion":2}```

Paul-TAB (Mon, 22 Oct 2018 14:30:58 GMT):
Has joined the channel.

rjones (Mon, 22 Oct 2018 14:33:24 GMT):
@louisk awesome!

louisk (Mon, 22 Oct 2018 14:34:41 GMT):
it seems a little bit anti-pattern though to be building a server for this since it's supposed to be p2p

louisk (Mon, 22 Oct 2018 14:34:49 GMT):
though i don't see how else an agent would function?

Paul-TAB (Mon, 22 Oct 2018 14:35:09 GMT):
Hi all - looking for the best way to clear all locally stored indy stuff: wallet, dids, pool data etc created while trying to get an app working. Where is the cache stored? Could we just delete that?

sklump (Mon, 22 Oct 2018 15:07:55 GMT):
rm -rf ~/.indy_client

sklump (Mon, 22 Oct 2018 15:07:55 GMT):
@Paul-TAB ``` $ rm -rf ~/.indy_client ```

Paul-TAB (Mon, 22 Oct 2018 15:16:43 GMT):
Thanks! If this is deleted will it recreate as needed?

sklump (Mon, 22 Oct 2018 15:28:21 GMT):
If using indy-sdk, this is its default top dir for indy artifacts.

Paul-TAB (Mon, 22 Oct 2018 15:31:59 GMT):
Thanks sklump - amazingly helpful!

Paul-TAB (Mon, 22 Oct 2018 15:38:20 GMT):
I have another question relating to a current demo we're trying to create based on the Getting Started walkthrough. This involves a user purchasing an item from a retailer. The retailer will initiate the onboarding via a QR code. We are stumped as to how the user then accesses the retailers verkey in order to encrypt their response. In our implementation, using verkey = keyForDid(poolHandle, userWalletHandle, retailerDid) fails with walletItemNotFound 212. Whereas in the Getting Started example, using the toWallet (ie userWallet) succeeds. Any thoughts?

Paul-TAB (Mon, 22 Oct 2018 15:41:29 GMT):
I should add that we suppose the retailers verkey can be retrieved on the user side because previously the retailer wrote it to the ledger using *their* wallet

sklump (Mon, 22 Oct 2018 16:12:02 GMT):
GET_NYM returns a structure like ``` { "dest": "V4SGRU86Z58d6TV7PBUZ6f", "identifier": null, "role": "0", "seqNo": 123, "txnTime": null, "verkey": "~KoRVR13DVYsWPtK8uAzNbx" } ``` Just peel out the verkey from the JSON.

sklump (Mon, 22 Oct 2018 16:12:31 GMT):
You do have to keep track of DIDs for parties of interest.

louisk (Mon, 22 Oct 2018 16:52:39 GMT):
why GET_NYM instead of KEY_FOR_DID ? (if you already have the DID for retailer in this case)

sklump (Mon, 22 Oct 2018 17:09:07 GMT):
Sure, KEY_FOR_DID is probably better

sklump (Mon, 22 Oct 2018 17:09:07 GMT):
Sure, KEY_FOR_DID is better

manishcm (Tue, 23 Oct 2018 03:27:34 GMT):
Has joined the channel.

ShubhamSingh18 (Tue, 23 Oct 2018 07:00:32 GMT):
Getting this error while running python sampleindy@Shubham:~/samples/python/src$ python3.5 getting_started.py Traceback (most recent call last): File "getting_started.py", line 3, in from indy import anoncreds, crypto, did, ledger, pool, wallet ImportError: No module named 'indy'

ShubhamSingh18 (Tue, 23 Oct 2018 07:00:32 GMT):
Getting this error while running python sample indy@Shubham:~/samples/python/src$ python3.5 getting_started.py Traceback (most recent call last): File "getting_started.py", line 3, in from indy import anoncreds, crypto, did, ledger, pool, wallet ImportError: No module named 'indy'

Paul-TAB (Tue, 23 Oct 2018 08:14:56 GMT):
@sklump: The result of signAndSubmitRequest(buildGetNymRequest(submitterDid, targetDid)) ```{ result: { txnTime: null, seqNo: null, dest: 'Sa2MZfSD6ycRtxaxh1J2ux', reqId: 1540282035769940000, data: null, identifier: 'JC8y9KJ1qbbqJHm3dh9DeM', type: '105' }, op: 'REPLY' }``` Not seeing a verkey at all. keyForDid does what we want, but in our demo the party using it doesn't have the necessary wallet handle.

NikolaosTsoniotis (Tue, 23 Oct 2018 08:44:05 GMT):
Has joined the channel.

ShubhamSingh18 (Tue, 23 Oct 2018 11:55:21 GMT):
How to solve this error please help while executing the python sample indy@Shubham:~/samples/python$ python3 -m src.main INFO:src.getting_started:Getting started -> started INFO:src.getting_started:Open Pool Ledger: pool1 ERROR:indy.libindy:_load_cdll: Can't load libindy: libindy.so: cannot open shared object file: No such file or directory Traceback (most recent call last): File "/usr/lib/python3.5/runpy.py", line 184, in _run_module_as_main "__main__", mod_spec) File "/usr/lib/python3.5/runpy.py", line 85, in _run_code exec(code, run_globals) File "/home/indy/samples/python/src/main.py", line 16, in run_coroutine(main) File "/home/indy/samples/python/src/utils.py", line 47, in run_coroutine loop.run_until_complete(coroutine()) File "/usr/lib/python3.5/asyncio/base_events.py", line 387, in run_until_complete return future.result() File "/usr/lib/python3.5/asyncio/futures.py", line 274, in result raise self._exception File "/usr/lib/python3.5/asyncio/tasks.py", line 239, in _step result = coro.send(None) File "/home/indy/samples/python/src/main.py", line 8, in main await getting_started.run() File "/home/indy/samples/python/src/getting_started.py", line 26, in run await pool.set_protocol_version(PROTOCOL_VERSION) File "/home/indy/.local/lib/python3.5/site-packages/indy/pool.py", line 208, in set_protocol_version delete_pool_ledger_config.cb) File "/home/indy/.local/lib/python3.5/site-packages/indy/libindy.py", line 24, in do_call err = getattr(_cdll(), name)(command_handle, File "/home/indy/.local/lib/python3.5/site-packages/indy/libindy.py", line 90, in _cdll _cdll.cdll = _load_cdll() File "/home/indy/.local/lib/python3.5/site-packages/indy/libindy.py", line 121, in _load_cdll raise e File "/home/indy/.local/lib/python3.5/site-packages/indy/libindy.py", line 116, in _load_cdll res = CDLL(libindy_name) File "/usr/lib/python3.5/ctypes/__init__.py", line 347, in _init_ self._handle = _dlopen(self._name, mode) OSError: libindy.so: cannot open shared object file: No such file or directory indy@Shubham:~/samples/python$

sklump (Tue, 23 Oct 2018 12:08:34 GMT):
You need to `export LD_LIBRARY_PATH` or `sudo cp .../indy-sdk/libindy/target/*/libindy.so /usr/lib`

ShubhamSingh18 (Tue, 23 Oct 2018 13:51:35 GMT):
@sklump already done

jlin (Tue, 23 Oct 2018 15:37:48 GMT):
Hi, was there a link to the Corda slides from the Indy WG meeting last week? I scrolled up but couldn't find it

Sean_Bohan (Tue, 23 Oct 2018 16:40:42 GMT):
JLin - we are late getting the video posted. i will get it up and running tomorrow

jlin (Tue, 23 Oct 2018 16:44:44 GMT):
Ok thanks Sean

zixian5 (Wed, 24 Oct 2018 03:52:47 GMT):
Do any one know how the client or sdk connect to the node ? Do the client connect to the one node in the genesis transactions OR connect to the all nodes in the genesis transactions?

zixian5 (Wed, 24 Oct 2018 03:52:47 GMT):
Does any one know how the client or sdk connect to the node ? Does the client connect to the one node in the genesis transactions OR connect to the all nodes in the genesis transactions?

zixian5 (Wed, 24 Oct 2018 03:52:47 GMT):
Please help me . It suffers me a lot. Does any one know how the client or sdk connect to the node ? Does the client connect to one node in the genesis transactions OR connect to the all nodes in the genesis transactions?

ShubhamSingh18 (Wed, 24 Oct 2018 04:54:40 GMT):
Do anyone have made demo in python or java

Abdul_khaliq (Wed, 24 Oct 2018 16:55:18 GMT):
@zixian5 thank you for guiding me

swcurran (Wed, 24 Oct 2018 18:44:35 GMT):
@ShubhamSingh18 - the https://github.com/hyperledger/indy-agent repo contains examples of python and nodejs agents. As well, this README - https://github.com/swcurran/education/blob/master/LFS171x/indy-material/nodejs/README.md - provides an easy to use example of a point in time version of the nodejs agent, including how to build, run and walk through the demo that you can run in a browser or on your laptop.

swcurran (Wed, 24 Oct 2018 18:46:29 GMT):
@MALodder - question. When a verifier gets a proof that contains claims from multiple credentials, can the verifier tell if the same linked secret was used to prove that different claims?

arunwij (Thu, 25 Oct 2018 00:45:53 GMT):
Hello Everyone, I highly appreciate if anyone can give me guide one this. Think if a person without having any verifiable credential(VC) initially. How that person going to get first VC from a issuer. How identity owner going to verify it is him/her. Do we have to have physical contact with the issuing entity? What are the possible solutions for this matter? (Does indy have a solution for this?) Thank you.

arunwij (Thu, 25 Oct 2018 00:45:53 GMT):
Hello Everyone, can someone give me guide one this? Think if a person without having any verifiable credential(VC) initially. How that person going to get first VC from a issuer. How identity owner going to verify it is him/her. Do we have to have physical contact with the issuing entity? What are the possible solutions for this matter? (Does indy have a solution for this?) Thank you.

arunwij (Thu, 25 Oct 2018 00:45:53 GMT):
Hi, can someone give me guide one this? Think if a person without having any verifiable credential(VC) initially. How that person going to get first VC from an issuer. How identity owner going to verify it is him/her. Do we have to have physical contact with the issuing entity? What are the possible solutions for this matter? (Does Indy have a solution for this?) Thank you.

Shyam_Pratap_Singh (Thu, 25 Oct 2018 04:22:54 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=6Xfc6cA889Aq2AApX) @arunwij As per my understanding, Alice use case has mentioned such scenerio, so what happen exactly is, Alice ( Person) establish trust with the issuing organisation though either physical contact or using email or any medium of contact. Once Entity verify Alice, then it start onboarding process of Alice over Indy network and hence issue verifiable claims to her.

arunwij (Thu, 25 Oct 2018 04:34:07 GMT):
@shyam 40 Thank you for the reply. Don't you think if we contact a person with email or other digital mechanism, there is a chance to impersonate the identity owner by another party? Physically meeting a issuer is may not feasible, other wise we still have to depend on our traditional identity methods like identity cards. What are your thoughts?

Shyam_Pratap_Singh (Thu, 25 Oct 2018 04:46:01 GMT):
Initially some how , you have to establish a relationship with any of the TrustAnchor for eg. In India you might visit to nearby Aadhar center and show your current available identity , then Aadhar center based on initial proof, start process of onboarding to Sovrin Network.

arunwij (Thu, 25 Oct 2018 07:07:12 GMT):
Thank you. Is there anyone who have other suggestions on how we should handle this on practical situation?

arunwij (Thu, 25 Oct 2018 07:07:39 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=jSqaFsCzjCj5Kh3DL) For the reference.

cagdast (Thu, 25 Oct 2018 08:11:39 GMT):
Has joined the channel.

richzhao (Thu, 25 Oct 2018 08:33:12 GMT):
Has joined the channel.

ShubhamSingh18 (Thu, 25 Oct 2018 11:02:05 GMT):
What can be developed by using indy by using indy sdk code By current status

xnopre (Thu, 25 Oct 2018 12:48:22 GMT):
Hi all. I'm setup our new Steward node on Sovrin :-) . My application is using the SDK, and connects the ledger loading a "pool transactions genesis" file, containing information about some nodes on the ledger. OK. But I don't really understand why we need to have a Steward node to connect the ledger knowing that our node (IP and port) are not in the genesis file. Where can I found information or schemas to understand this point ? Thanks

xnopre (Thu, 25 Oct 2018 12:48:22 GMT):
Hi all. I'm setup our new Steward node on Sovrin :-) . My application is using the Indy SDK, and connects the ledger loading a "pool transactions genesis" file, containing information about some nodes on the ledger. OK. But I don't really understand why we need to have a Steward node to connect the ledger knowing that our node (IP and port) are not in the genesis file. Where can I found information or schemas to understand this point ? Thanks

ShubhamSingh18 (Thu, 25 Oct 2018 13:32:15 GMT):
can you share your code

ShubhamSingh18 (Thu, 25 Oct 2018 13:32:23 GMT):
@xnopre

xnopre (Thu, 25 Oct 2018 14:07:07 GMT):
@ShubhamSingh18 No interest to share my code. My question is very general. Is it needed to have a "Sovrin validator node" to be able to link my application to the "live" Sovrin ledger ? I thought that my application will communicate to my Sovrin node, but apparently not, the IP address of my node is note in the genesis file loaded by my application to communicate with Sovrin ledger.... ??

Sean_Bohan (Thu, 25 Oct 2018 16:07:55 GMT):
FYI: Friday Indy Agent WG Design/Discussion Date: Friday October 26, 2018 Time: 7amPT, 8amMT, 9amCT, 10amET, 3pmBST Agenda IIW Review Reference Agent Review Open Discussion Have a question, issue, use case you want to share, etc.? Bring it to the call. https://byu.zoom.us/j/2627891784

twshelton (Thu, 25 Oct 2018 19:08:01 GMT):
@xnopre ... you can change the genesis file to point to your local node and not use the Sovrin network for development purposes. Can you provide some more information on how you configured your environment and the environment itself (OSX, Linux, Windows)? Then we can help you get everything connected.

ShivVenkatraman (Thu, 25 Oct 2018 22:03:21 GMT):
Basic question: what is the role of the consensus in Hyperledger Indy; particularly with regard to claims/identity? What do the nodes agree/disagree upon?

ShivVenkatraman (Thu, 25 Oct 2018 22:03:21 GMT):
Basic question: what is the role of the consensus in Hyperledger Indy; particularly with regard to claims/identity? What do the nodes agree/disagree upon? Also, is the consensus module pluggable like with chaincodes in Fabric or ABCI implementation in Tendermint?

haniavis (Thu, 25 Oct 2018 22:45:50 GMT):
Has joined the channel.

xnopre (Fri, 26 Oct 2018 07:57:20 GMT):
@twshelton Sorry, it seems I'm not so clear in my question. I have no problem to connect to ledger (no yet tested). I'm doing the setup of my Sovrin Steward node, and I try to understand the link (if there is one) between connecting the Sovrin ledger, and the need to have a Steward node ? Why is it needed (or not ?) to have a Steward node to be able to use the ledger ? cc @ShubhamSingh18

ShubhamSingh18 (Fri, 26 Oct 2018 08:12:58 GMT):
@xnopre Refer this document it might help you https://github.com/hyperledger/indy-sdk/blob/master/doc/getting-started/getting-started.md

nage (Fri, 26 Oct 2018 12:42:54 GMT):
Just a note to be careful about Working Group times through the next few weeks. With the differences in time zone changes between the US and other parts of the world it is important to check your calendaring applications to see if you're coming to the meetings at the right times.

mhailstone (Fri, 26 Oct 2018 13:52:20 GMT):
The agent call will begin in 10 minutes: https://byu.zoom.us/j/2627891784

lightcap (Fri, 26 Oct 2018 18:05:56 GMT):
Has joined the channel.

xnopre (Sat, 27 Oct 2018 09:35:09 GMT):
@ShubhamSingh18 Thanks :-)

ShubhamSingh18 (Sat, 27 Oct 2018 09:37:14 GMT):
My web socket is closed after sending the connection request from AlicetoBobConnection

ShubhamSingh18 (Sat, 27 Oct 2018 09:37:32 GMT):
shubhamsingh@Shubham:~/Desktop/indy-agent/python$ docker run -p 8095:8095 -e PORT=8095 indy-agent ===== Starting Server on: http://localhost:8095 ===== Your UI Token is: 5a0c7b936ab54668b36fccf1d4bf6580 Received "{"type":"did:sov:BzCbsNYhMrjHiqZDTUASHg;spec/sovrin.org/ui/1.0/state_request","id":"5a0c7b936ab54668b36fccf1d4bf6580","content":null}" Sending "{"type": "did:sov:BzCbsNYhMrjHiqZDTUASHg;spec/sovrin.org/ui/1.0/state", "content": {"initialized": false, "agent_name": null}}" Received "{"type":"did:sov:BzCbsNYhMrjHiqZDTUASHg;spec/sovrin.org/ui/1.0/initialize","id":"5a0c7b936ab54668b36fccf1d4bf6580","content":{"name":"Bob","passphrase":"sdfghj"}}" Sending "{"type": "did:sov:BzCbsNYhMrjHiqZDTUASHg;spec/sovrin.org/ui/1.0/state", "content": {"initialized": true, "agent_name": "Bob"}}" websocket connection closed Received "{"type":"did:sov:BzCbsNYhMrjHiqZDTUASHg;spec/sovrin.org/ui/1.0/state_request","id":"5a0c7b936ab54668b36fccf1d4bf6580","content":null}" Sending "{"type": "did:sov:BzCbsNYhMrjHiqZDTUASHg;spec/sovrin.org/ui/1.0/state", "content": {"initialized": true, "agent_name": "Bob"}}" websocket connection closed Received "{"type":"did:sov:BzCbsNYhMrjHiqZDTUASHg;spec/sovrin.org/ui/1.0/state_request","id":"5a0c7b936ab54668b36fccf1d4bf6580","content":null}" Sending "{"type": "did:sov:BzCbsNYhMrjHiqZDTUASHg;spec/sovrin.org/ui/1.0/state", "content": {"initialized": true, "agent_name": "Bob"}}" websocket connection closed Received "{"type":"did:sov:BzCbsNYhMrjHiqZDTUASHg;spec/sovrin.org/ui/1.0/state_request","id":"5a0c7b936ab54668b36fccf1d4bf6580","content":null}" Sending "{"type": "did:sov:BzCbsNYhMrjHiqZDTUASHg;spec/sovrin.org/ui/1.0/state", "content": {"initialized": true, "agent_name": "Bob"}}" websocket connection closed Received "{"type":"did:sov:BzCbsNYhMrjHiqZDTUASHg;spec/sovrin.org/ui/1.0/state_request","id":"5a0c7b936ab54668b36fccf1d4bf6580","content":null}" Sending "{"type": "did:sov:BzCbsNYhMrjHiqZDTUASHg;spec/sovrin.org/ui/1.0/state", "content": {"initialized": true, "agent_name": "Bob"}}" websocket connection closed Received "{"type":"did:sov:BzCbsNYhMrjHiqZDTUASHg;spec/sovrin.org/ui/1.0/state_request","id":"5a0c7b936ab54668b36fccf1d4bf6580","content":null}" Sending "{"type": "did:sov:BzCbsNYhMrjHiqZDTUASHg;spec/sovrin.org/ui/1.0/state", "content": {"initialized": true, "agent_name": "Bob"}}" websocket connection closed Received "{"type":"did:sov:BzCbsNYhMrjHiqZDTUASHg;spec/sovrin.org/ui/1.0/state_request","id":"5a0c7b936ab54668b36fccf1d4bf6580","content":null}" Sending "{"type": "did:sov:BzCbsNYhMrjHiqZDTUASHg;spec/sovrin.org/ui/1.0/state", "content": {"initialized": true, "agent_name": "Bob"}}" websocket connection closed Received "{"type":"did:sov:BzCbsNYhMrjHiqZDTUASHg;spec/sovrin.org/ui/1.0/state_request","id":"5a0c7b936ab54668b36fccf1d4bf6580","content":null}" Sending "{"type": "did:sov:BzCbsNYhMrjHiqZDTUASHg;spec/sovrin.org/ui/1.0/state", "content": {"initialized": true, "agent_name": "Bob"}}" websocket connection closed Received "{"type":"did:sov:BzCbsNYhMrjHiqZDTUASHg;spec/sovrin.org/ui/1.0/state_request","id":"5a0c7b936ab54668b36fccf1d4bf6580","content":null}" Sending "{"type": "did:sov:BzCbsNYhMrjHiqZDTUASHg;spec/sovrin.org/ui/1.0/state", "content": {"initialized": true, "agent_name": "Bob"}}" websocket connection closed Received "{"type":"did:sov:BzCbsNYhMrjHiqZDTUASHg;spec/sovrin.org/ui/1.0/state_request","id":"5a0c7b936ab54668b36fccf1d4bf6580","content":null}" Sending "{"type": "did:sov:BzCbsNYhMrjHiqZDTUASHg;spec/sovrin.org/ui/1.0/state", "content": {"initialized": true, "agent_name": "Bob"}}" websocket connection closed

ShubhamSingh18 (Sat, 27 Oct 2018 11:57:08 GMT):
Can any one tell me the architecture of indy

danielhardman (Sun, 28 Oct 2018 00:53:12 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=mrHTEXqEo3MAuuoMd) @ShubhamSingh18 Indy is a python-based daemon that runs on linux and implements a byzantine fault tolerant consensus algorithm to protect the integrity of writes to the ledger. The ledger storage uses RocksDB. Does that answer your question? Or could you be more specific?

andrewtan (Mon, 29 Oct 2018 02:14:24 GMT):
Do have a question and hope this community can help to share some insights. With MyCUID, I believe there will be credentials issued to the client's MyCUID Wallet. If the same client is also having the wallet for BC Gov's VON's Network for example or other SSI Wallets, how does he or she used a mixture of credentials across different wallets for a specific purpose? If the different credentials from different issuers are all stored in the same wallet, I can see that happening easily but across multiple wallets, how will that works? Any document or links that explains more on that will be helpful.

swcurran (Mon, 29 Oct 2018 04:13:01 GMT):
Not sure about MyCUID wallet, but we aren't planning a wallet for VON. We're assuming that vendors will create wallets that work with VON Issuer/Verifier Services. This points to the importance of interoperability in the community. That said, we do think that a Wallet that supports extensions - eg an app store-type ecosystem - would enable done pretty amazing capabilities. As you note, we don't think each Service Provider will have their own Agent/Wallet.

swcurran (Mon, 29 Oct 2018 04:13:01 GMT):
Not sure about MyCUID wallet, but we aren't planning a personal agent/wallet just for VON. We're assuming that vendors will create wallets that work with VON Issuer/Verifier Services. This points to the importance of interoperability in the community. That said, we do think that a Wallet that supports extensions - eg an app store-type ecosystem - would enable some pretty amazing capabilities. As you note, we don't think each Service Provider will have their own Agent/Wallet.

ShubhamSingh18 (Mon, 29 Oct 2018 04:49:08 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=aQw6dPEEhfYx996Hb) @danielhardman Thanks for the information but can you give this all information in detail how the what are the component of architecture,which database for wallet,network,backup

ShubhamSingh18 (Mon, 29 Oct 2018 07:12:27 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=9kgHkEZ2DiTepYf6Q) how the components are connected to each other

ShubhamSingh18 (Mon, 29 Oct 2018 07:13:40 GMT):
Can anyone tell me the architecture of hyperledger indy and how the components are connected to each other

ShubhamSingh18 (Mon, 29 Oct 2018 07:18:34 GMT):
if any diagram please give

theoturner (Mon, 29 Oct 2018 09:29:12 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=HiyPTrwLQcxQhQhNE) Would like to reiterate this if anyone has any info

ShubhamSingh18 (Mon, 29 Oct 2018 11:43:17 GMT):
Please answer the question

danielhardman (Mon, 29 Oct 2018 16:46:20 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=reDmk5DZrhxWQd9Ah) @theoturner Pairwise DIDs will mostly be written to microledgers, so they impose little or no burden on the global, public ledger. Further, even if there is no microledger (every DID must be written to the global, public ledger), a connection request does not automatically result in a connection acceptance that creates a DID. You only end up creating a DID when the connection is accepted. Now, if every client decides to accept all connection requests and then evaluate whether they're desirable, after the fact, then the potential for spam exists as you have imagined it. The disincentive for doing this is the cost of writing a DID. DIDs are not rotated, so I don't see any potential for abuse related to "continually changing the pairwise DID." Unless you mean, continually rotating the public key associated with a DID? As with other ledger writes, this write operation will have a cost, which provides a built-in disincentive to such spam.

danielhardman (Mon, 29 Oct 2018 16:49:29 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=oLSZPbsWnt2JXSLhQ) @xnopre You don't need a validator node to use Indy, any more than you need to be a bank operating an ATM to be able to use ATM machines. Your application communicates with *all* validator nodes, not just one that you operate. If you're not operating a node, then your application communicates with other nodes; it doesn't need anything local.

danielhardman (Mon, 29 Oct 2018 16:52:42 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=4H74Y2KNzfYgdDCaz) @ShivVenkatraman Consensus is used to order transactions, impose timestamps, and harmonize the view of what is on the ledger. It has no direct relationship to claims/credentials. A credential is not agreed upon by validator nodes; rather, a credential is an independent JSON-based doc that validator nodes don't care about--but it is digitally signed using a private key for which the corresponding public key, credential definition, and DID are stored on the ledger. The consensus algorithm is somewhat decoupled from the rest of the system, but this decoupling is not crisp enough that another consensus algorithm like Tendermint could be plugged in out-of-the-box. However, there are active discussions about plugging in other consensus algorithms, so this decoupling is likely to get crisper in coming months.

danielhardman (Mon, 29 Oct 2018 16:54:16 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=FeEMcCh3Fjz9A4Low) @ShubhamSingh18 See http://bit.ly/2PpBdHz

danielhardman (Mon, 29 Oct 2018 17:26:24 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=2u9E9Nd5GR84oJ3tQ) @arunwij Arun: the assumption is that issuers don't give credentials unless they have some reason to do so--and that this reason includes some basis for trust. If the credential is issued by a bank, then the person receiving the credential is likely a bank customer, and the bank has policies and procedures in place to identify their customer and decide what their customer's attributes are. For example, the bank probably has a secure website where its customers can login and receive a credential. Similarly, if the issuer is an educational institution, it knows its students and has some sort of secure communication mechanism with them. Or if the issuer is a government, it knows its citizens and has some mechanism for determining whether it should issue credentials to them. The scenario you seem to be focusing on is one that seems extremely unlikely to me: two parties who have never interacted in any way want to begin their association by having one give the other a credential. Why would that ever be useful? What seems far more likely to me is that two parties who have never interacted create a relationship and then, AFTER trust accumulates in that relationship, one of the parties gives the other a credential on the basis of the trust that has accumulated. For example, two parties that do not know one another conduct some business on eBay--and the resulting credential states that party A was a good customer for party B. It cannot state that party A has a known name or bank balance, since the relationship never involved evidence of those things--but it can state that A paid promptly and was pleasant to work with, and used a particular eBay username.

danielhardman (Mon, 29 Oct 2018 17:33:15 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=8QXe4RayAPgzFm5jp) @zixian5 The SDK uses the genesis transactions file to contact some of the original nodes in the validator pool. It then asks those nodes if any new nodes have been added. If the answer is yes, then the original nodes from the validator pool prove that the new nodes were added according to correct rules of consensus, and supplies the contact information for the additional nodes. This is a process called "catchup", and it results in client code having an accurate idea of which validator nodes are currently canonical for the pool. After this is done, the client code then sends a read or write request to multiple validator nodes. Which nodes it picks is not under client control. This is a security design choice. We assume that one or more of the validator nodes could be malicious (attempt to lie to a client). Therefore the client randomly contacts several of them (enough so that at least one of them is not malicious). The nodes process the transaction request, and provide a reply. The reply has to be signed by multiple nodes, such that any single node cannot forge a response--only a response backed by genuine consensus is legitimate. Communication between clients and nodes currently uses ZMQ technology. Did this answer your question?

haggs (Mon, 29 Oct 2018 18:24:15 GMT):
@danielhardman Sorry to berate you with questions as you seem to be getting a lot today. Thanks for your extremely thoughtful and detailed responses! But just curious on your thoughts about this comment and an ongoing theme of microledgers. "Pairwise DIDs will mostly be written to microledgers, so they impose little or no burden on the global, public ledger." That, as well as the recently updated (last Monday I believe) Evernym updated paper on "What Goes on the Ledger", which also points to microledgers as helping with a lot of this friction on the ledger.

haggs (Mon, 29 Oct 2018 18:25:01 GMT):
Anyways, my question is, are DAD and derived DID's going to be the answer/solution? I'm referring to Sam Smith's work from IIW: https://github.com/SmithSamuelM/Papers/blob/master/presentations/DID_Everything_IIW_20181023.pdf

haggs (Mon, 29 Oct 2018 18:25:01 GMT):
Anyways, my question is, are DADi 's and derived DID's going to be the answer/solution? I'm referring to Sam Smith's work from IIW: https://github.com/SmithSamuelM/Papers/blob/master/presentations/DID_Everything_IIW_20181023.pdf. I know you probably have no definite answer but that presentation crossed a lot of mental checkboxes for me as far as microledgers go. Thanks for any input Daniel!

danielhardman (Mon, 29 Oct 2018 18:35:17 GMT):
I haven't seen Sam's preso, but I'm generally aware of the concepts after getting a brain dump from some people who attended. I will need to sit down and digest it before I'm qualified to have an informed opinion. But my initial reaction is enthusiastic; it seems like a nice addition to the thinking.

haggs (Mon, 29 Oct 2018 18:46:43 GMT):
Cool, thanks @danielhardman for your immediate input! I'm still trying to digest it as well :D

MichaelLitchev (Mon, 29 Oct 2018 21:50:45 GMT):
Has joined the channel.

MichaelLitchev (Mon, 29 Oct 2018 21:50:56 GMT):
Hey everybody!

MichaelLitchev (Mon, 29 Oct 2018 21:51:16 GMT):
Anyone have any insight into a Indy 309 error when trying parseGetSchemaResponse with Node.js?

MichaelLitchev (Mon, 29 Oct 2018 21:51:39 GMT):
```Credential Definition Creation Failed { IndyError: 309 at Object.callback (/Users/Mikhael/Workspace/Persisto/indy-server/node_modules/indy-sdk/src/wrapIndyCallback.js:15:10) name: 'IndyError', message: '309' }```

MichaelLitchev (Mon, 29 Oct 2018 21:51:39 GMT):
```Credential Definition Creation Failed { IndyError: 309 at Object.callback (/Users/User/Workspace/path/node_modules/indy-sdk/src/wrapIndyCallback.js:15:10) name: 'IndyError', message: '309' }```

MichaelLitchev (Mon, 29 Oct 2018 21:54:07 GMT):
I am able to buildGetSchemaRequest and submitRequest before trying parseGetSchemaResponse

MichaelLitchev (Mon, 29 Oct 2018 22:15:57 GMT):
The schema response here: ```{ op: 'REPLY', result: { identifier: 'Th7MpTaRZVRYnPiabds81Y', type: '107', seqNo: null, dest: 'Th7MpTaRZVRYnPiabds81Y', data: { name: 'Job-Certificate', version: '0.4' }, state_proof: { multi_signature: [Object], root_hash: '2hm6BKvUN7kWB3L35MmVtA7JVMHis3mkQkPJJkRaMmXr', proof_nodes: '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' }, reqId: 1540850921557444000, txnTime: null } } Credential Definition Creation Failed { IndyError: 309 at Object.callback (/Users/Mikhael/Workspace/Persisto/indy-server/node_modules/indy-sdk/src/wrapIndyCallback.js:15:10) name: 'IndyError', message: '309' }```

MichaelLitchev (Mon, 29 Oct 2018 22:15:57 GMT):
The schema response here: ```{ op: 'REPLY', result: { identifier: 'Th7MpTaRZVRYnPiabds81Y', type: '107', seqNo: null, dest: 'Th7MpTaRZVRYnPiabds81Y', data: { name: 'Job-Certificate', version: '0.4' }, state_proof: { multi_signature: [Object], root_hash: '2hm6BKvUN7kWB3L35MmVtA7JVMHis3mkQkPJJkRaMmXr', proof_nodes: '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' }, reqId: 1540850921557444000, txnTime: null }```

MichaelLitchev (Mon, 29 Oct 2018 22:17:59 GMT):
This is my first post on this chat - is this channel the best place for such questions?

haniavis (Mon, 29 Oct 2018 22:30:13 GMT):
Hi, I am designing a distributed system and I would like to use indy for the identity management of the nodes. I see that there are different roles in indy (trustees, stewards, trust anchors) but my system will be t

haniavis (Mon, 29 Oct 2018 22:30:13 GMT):
Hi, I am designing a distributed system and I would like to use indy for the identity management of the nodes. I see that there are different roles in indy (trustees, stewards, trust anchors) but my system will be arbitrary and nodes will join and leave on demand

haniavis (Mon, 29 Oct 2018 22:32:26 GMT):
can I somehow bypass these roles and have equal nodes in terms of role?

swcurran (Mon, 29 Oct 2018 22:35:40 GMT):
Nodes do not have roles as you outline. DIDs that created in the ledger have the roles you mention.

swcurran (Mon, 29 Oct 2018 22:35:40 GMT):
Nodes do not have roles as you outline. DIDs that are created on the ledger have the roles you mention.

ShivVenkatraman (Mon, 29 Oct 2018 23:26:28 GMT):
What is indy written in? Rust or Python?

ShivVenkatraman (Tue, 30 Oct 2018 00:00:40 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=yFuivKKXJtpNgKkzo) @danielhardman Thanks @danielhardman. When changes are made to the ledger; can the clients listen to it so that they can take appropriate action if required? Tendermint has this feature, I believe.

danielhardman (Tue, 30 Oct 2018 00:26:29 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=jHeWcbxwrTm478AE9) @ShivVenkatraman No. However, most of the kinds of data that you might be interested in tracking are not on The Ledger anyway. So I'm not sure whether that really matters. What specific types of information do you want to track?

ShivVenkatraman (Tue, 30 Oct 2018 00:31:02 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=42604d71-2331-4bbd-8cf1-ac938e069b7d) @danielhardman @danielhardman In "what goes on ledger" (https://www.evernym.com/wp-content/uploads/2017/07/What-Goes-On-The-Ledger.pdf), there is a reference to storing public claims on the ledger. One scenario was to store the issuer info (with his consent) on the ledger; and so that verifiers can get more info on the issuer. So if the issuer has updated his phone number (for example), the verifiers and provers across different node instances know by querying the ledger instead of making cross-node calls

danielhardman (Tue, 30 Oct 2018 03:53:19 GMT):
I think you are misunderstanding that doc. The version of it that I am reading at that URL does not mention storing claims on the registry. It does talk about storing schemas (which tell which sorts of claims will be in credentials of that type, when they are finally issued), and credential definitions (which tell who will be signing credentials that use a given schema, and how revocation data will be managed). But claims themselves are never stored on the ledger.

danielhardman (Tue, 30 Oct 2018 03:54:33 GMT):
If someone wants to tell the world their definitive phone number, they don't do it by publishing the number to the ledger. Rather, they publish a DID Doc that tells the world how to talk to their agent at a chosen endpoint. The world can then call that endpoint and interrogate the agent to find out their phone number, mailing address, etc. The endpoint doesn't change, but all the other info about the company can vary in realtime.

wl_Carl (Tue, 30 Oct 2018 05:40:53 GMT):
Has joined the channel.

NikitaVolkov (Tue, 30 Oct 2018 11:51:01 GMT):
Has joined the channel.

wip-abramson (Tue, 30 Oct 2018 12:45:07 GMT):
Hi, I am trying to work through the docs on how to get started with indy. Currently I am here: https://github.com/hyperledger/indy-node/blob/master/environment/docker/pool/README.md I have started a pool but cannot start indy client using ./client_for_pool_start.sh I get - No such container: indyclient - any ideas?

wip-abramson (Tue, 30 Oct 2018 12:47:01 GMT):
I would also like to help write a tutorial/documentation for how an developer can actually get started with indy. Currently I am finding it really confusing.

lijiachuan (Tue, 30 Oct 2018 14:57:19 GMT):
Hi everyone, I still need someone can clarify this question "the identity owner will not share the actual data to the requester", I know this can be explained if the requester wants the identity owner to prove you are more than 18 years old, but if the requester needs to know your exactly SSN, does Indy still can support not share the actual SSN to the requester but instead will send something else? Kindly advise.

swcurran (Tue, 30 Oct 2018 15:14:48 GMT):
@lijiachuan - the identity owner could share that they have a credential from the Issuing Authority that they have an SSN, but do not share the SSN itself.

swcurran (Tue, 30 Oct 2018 15:17:56 GMT):
But if the requester needs to know the SSN (as you say) the identity owner would need to share it to meet the requirements - assuming they want the service offered.

haniavis (Tue, 30 Oct 2018 15:33:29 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=2YS8tCtXkQb8CB9yo) @swcurran So can I have a node that doesn't have any of these roles? and provide roles to some of my DIDs? e.g. having a network with no steward or trustee

swcurran (Tue, 30 Oct 2018 15:39:40 GMT):
Yes. We think of the nodes kind of like a database. It's the DIDs that provide the roles for performing network operations. Unless you are planning on conributing to indy-node, you should try to get a network running and then ignore it. :-)

ShivVenkatraman (Tue, 30 Oct 2018 16:54:18 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=Kug4cy6ZizoaG4wS3) @danielhardman You are right. I somehow had a 2017 version of the doc. It also had a whole paragraph on Public Claims (which I pasted to the sdk thread). The 2018 version of the doc does not refer to Public Claims. Here is the bottom of the 2017 doc: "Goes on Sovrin 􀂌 Decentralised identifiers and associated DDOs with verification keys and endpoints. 􀂌 Schemas and claim definitions 􀂌 Proof of consent for data sharing 􀂌 Public claims 􀂌 Revocation registries Does not go on Sovrin 􀂞 Private data of any kind (including hashed personal data) 􀂞 Private proof of existence".

twshelton (Tue, 30 Oct 2018 20:59:02 GMT):
@wip-abramson start here instead https://github.com/hyperledger/indy-sdk ... this should allow you to get the pool up and running in a docker container and then you can look through the wrappers to dind you language of choice to investigate further. Ping me directly if you need further guidance and I'll walk you through it

twshelton (Tue, 30 Oct 2018 20:59:35 GMT):
@wip-abramson we're also working on something a little more user friendly that I can walk you through just to get a feel of how it all fits together

ShivVenkatraman (Wed, 31 Oct 2018 01:11:59 GMT):
In https://github.com/hyperledger/indy-node/blob/master/docs/start-nodes.md, it mentions about domain_transactions_genesis file that has initial NYM transactions for trustee, steward etc. How are the entries for verkey and dest (did) generated in that file? Is there a way to control what gets into that file? Also, when new trust_anchors are created explicitly via buildNymRequest, where do the entries get populated? I don't see them in domain_transactions_genesis

Sasadara (Wed, 31 Oct 2018 04:16:51 GMT):
Has joined the channel.

ShubhamSingh18 (Wed, 31 Oct 2018 07:44:48 GMT):
How indy is different from the fabric as fabric can also customized for identity management system

sergey.khoroshavin (Wed, 31 Oct 2018 08:49:26 GMT):
@ShubhamSingh18 Fabric is oriented for private use (installations are inaccessible for general public), but Indy is public permissioned ledger

xnopre (Wed, 31 Oct 2018 09:05:33 GMT):
Where can I find (and follow) the (last) versions of `indy` (node and SDK) and release notes ?

arunwij (Wed, 31 Oct 2018 10:09:39 GMT):
@xnopre Is this are you looking for? https://github.com/hyperledger/indy-sdk/releases

fcabo (Wed, 31 Oct 2018 10:49:07 GMT):
Has joined the channel.

dnnn (Wed, 31 Oct 2018 10:59:13 GMT):
Has joined the channel.

dnnn (Wed, 31 Oct 2018 12:39:53 GMT):
Hi everyone! I am trying to run demo applications of iOS wrapper from indy-sdk repo and am facing an issue with missing "platform.hpp" file while the dependency pod `libzmq` is built. Could you please share some advice on how to fix it? I am using xcode 10

dnnn (Wed, 31 Oct 2018 12:39:53 GMT):
Hi everyone! I am trying to run demo application of iOS wrapper from indy-sdk repo and am facing an issue with missing "platform.hpp" file while the dependency pod `libzmq` is built. Could you please share some advice on how to fix it? I am using xcode 10

xnopre (Wed, 31 Oct 2018 14:03:29 GMT):
@arunwij Yes, thanks ! :-)

bhagadorn (Wed, 31 Oct 2018 18:57:42 GMT):
Has joined the channel.

NateThelen (Wed, 31 Oct 2018 21:10:21 GMT):
Has joined the channel.

ShivVenkatraman (Thu, 01 Nov 2018 00:01:07 GMT):
If the verkey of the issuer (who had earlier issued claims to a prover) is replaced, should the subsequent (after key replacement) proof verification (by another verifier) of the proof shown by prover fail? It's not in my test case; but I am not sure what is the expected behavior

ShivVenkatraman (Thu, 01 Nov 2018 00:23:26 GMT):
The Sovrin whitepaper https://sovrin.org/wp-content/uploads/2018/03/Sovrin-Protocol-and-Token-White-Paper.pdf refers to observer and validator nodes. Does Indy support such a model? Currently all nodes seem to be validators

haggs (Thu, 01 Nov 2018 01:04:29 GMT):
@ShivVenkatraman The node consensus algorithm is very performant and favors diversity, high throughput, and other things as far as my understanding. From what I can gather some validator nodes may be dual purpose depending on that algorithm and the current state of the node.

haggs (Thu, 01 Nov 2018 01:04:29 GMT):
@ShivVenkatraman The node consensus algorithm is very performant and favors diversity, high throughput, and other things as far as my outside understanding. Like other consensus algorithms, some nodes may become validators or observers depending on their current state and the throughput of validation and observation needed. Hope that helps!

haggs (Thu, 01 Nov 2018 01:04:29 GMT):
@ShivVenkatraman The node consensus algorithm is very performant and favors regional diversity, high throughput, and other things as far as my outside understanding. Like other consensus algorithms, some nodes may become validators or observers depending on their current state and the throughput of validation and observation needed (reads/writes). Hope that helps!

haggs (Thu, 01 Nov 2018 01:04:29 GMT):
@ShivVenkatraman The node consensus algorithm is very performant and favors regional diversity, high throughput, and other things as far as my outside understanding. Like other consensus algorithms, some nodes may become validators or observers depending on their current state and the throughput of validation and observation needed (reads/writes). Hope that helps! For an expert opinion I would probably stream your thoughts towards indy-node where someone more knowledgeable might help answer your question and the context behind it.

haggs (Thu, 01 Nov 2018 01:04:29 GMT):
@ShivVenkatraman The node consensus algorithm is very performant and favors regional diversity, high throughput, and other things as far as my limited understanding. Like other consensus algorithms, some nodes may become validators or observers depending on their current state and the throughput of validation and observation needed (reads/writes). Hope that helps! For an expert opinion I would probably stream your thoughts towards indy-node where someone more knowledgeable might help answer your question and the context behind it.

jljordan_bcgov (Thu, 01 Nov 2018 04:41:45 GMT):
Observer nodes are not done yet .. on the roadmap https://wiki.hyperledger.org/projects/indy/roadmap

Sean_Bohan (Thu, 01 Nov 2018 13:10:09 GMT):
Today's Indy WG Call Agenda and Info: AGENDA: - Housekeeping - AGENTS! - Recap IIW (All) - Recap W3C TPAC (All) - Luxoft Discusses Indy/Corda Integration (Luxoft team) - Open Discussion When: 8am PT, 9am MT, 10am CT, 11amET, 3pmGMT Where: Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/232861185

MALodder (Thu, 01 Nov 2018 14:26:26 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=bYGHrGbjd5hNHFSQ3) @ShivVenkatraman Verkey is different than issuance key. Verkey is to verify the connection to the issuer. Issuance key is used to verify credential signatures. Verkey is meant to be pairwise so rotating it only changes the connection to you not to everyone else. If the issuer did indeed rotate the issuance key in the credential definition, then that doesn't necessarily invalidate existing credentials. It just means newer credentials are not using the older key. If the issuer wanted to revoke all issued credentials, they would zero out the revocation registry.

dhuseby (Thu, 01 Nov 2018 15:04:54 GMT):
Dramatic reading ++

ShivVenkatraman (Thu, 01 Nov 2018 17:34:57 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=59bMwETwHsWyBM8mT) @MALodder Does the SDK (Java) have examples of issuance key vs verkey? I will ask this in sdk channel as well

MALodder (Thu, 01 Nov 2018 17:41:25 GMT):
Verkeys are associated with DIDs

MALodder (Thu, 01 Nov 2018 17:41:36 GMT):
issuance keys are associated with credential definitions

MALodder (Thu, 01 Nov 2018 17:42:04 GMT):
I don't know if the Java SDK has examples of that but I hope this leads you to the right functions

arunwij (Fri, 02 Nov 2018 05:09:21 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=pnJ7ThQekRxn7fu9X) @danielhardman Thank you. This actually makes sense.

TammyPlatero (Fri, 02 Nov 2018 18:35:33 GMT):
Has joined the channel.

TammyPlatero (Fri, 02 Nov 2018 18:36:13 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=xwfmnLFz3tWxWKQA5) @Tnigam @Tnigam Did you get a response? I'm new too and the link is broken

TammyPlatero (Fri, 02 Nov 2018 18:37:58 GMT):
I downloaded the indy-sdk-dotnet and copied the libindy dlls but when I run the Samples solution I get a 307 error. I tried running as an administrator but then got a 306 error. Anyone have any tips or reference material to install the .net sdk?

dbluhm (Fri, 02 Nov 2018 19:41:37 GMT):
@TammyPlatero you might be able to find more help by directing your question to the #indy-sdk channel

TammyPlatero (Fri, 02 Nov 2018 19:50:34 GMT):
Good deal - thanks :D

drummondreed (Sat, 03 Nov 2018 01:05:48 GMT):
Although Sovrin is just one deployment of Hyperledger Indy, it may still be of interest to the Indy community that, after a year’s worth of work by the Sovrin Governance Framework Working Group, the Sovrin Foundation has announced the beginning of the public review period for the Sovrin Governance Framework V2 (formerly called the Sovrin Trust Framework). As the “constitution” of the Sovrin Network, this set of documents specifies the purpose, principles, and policies that govern Sovrin as a global public utility for self-sovereign identity. Read the full story on the Sovrin blog: https://sovrin.org/announcing-public-review-of-the-sovrin-governance-framework-v2/

calvingerling (Sat, 03 Nov 2018 06:52:04 GMT):
Hi, I'm following the Getting Started guide with the indy-cli into a test network and I didn't understood very well the concept of a Credential Definition and how it's related with a Credential Schema?

calvingerling (Sat, 03 Nov 2018 07:02:17 GMT):
In the guide it's needed to create Credential Definitions to Faber (Transcription) and Acme (Job-Certificate). I'm using the `ledger cred-def` command in the cli it has been a little troublesome with the `primary` and `revocation` arguments, the `primary` have a really specific structure (I've found in `indy-crypto` code the struct of it but there's no documentation about what each thing represents), can someone explain me the purpose and usage of the `primary` argument?

Isaiah_Kim (Sat, 03 Nov 2018 14:18:16 GMT):
Has joined the channel.

MinuJawahar (Mon, 05 Nov 2018 12:58:59 GMT):
Has joined the channel.

MinuJawahar (Mon, 05 Nov 2018 13:04:39 GMT):
Hello everyone! I am currently working with a biometrics project. I want it to be implement in block chain technology. I have chosen INDY for my project. Could some one please suggest whether INDY will be the correct choice or not? and how to create a blockchain using INDY? Hello everyone! I am currently working with a biometrics project. I want it to be implement in block chain technology. I have chosen INDY for my project. Could some one please suggest whether INDY will be the correct choice or not? and how to create a blockchain using INDY? Hello everyone! I am currently working with a biometrics project. I want it to be implement in block chain technology. I have chosen INDY for my project. Could some one please suggest whethe r INDY will be the correct choice or not? and how to create a blockchain using INDY? Hello everyone! I am currently working with a biometrics project. I want it to be implement in block chain technology. I have chosen INDY for my project. Could some one please suggest whethe r INDY will be the correct choice or not? and how to create a blockchain using INDY? Hello everyone! I am currently working with a biometrics project. I want it to be implement in block chain technology. I have chosen INDY for my project. Could some one please suggest whethe r INDY will be the correct choice or not? and how to create a blockchain using INDY? Hello Every One. Currently I am working with a biometrics project. I want it to be implement using blockchain technology. for this I have chosen INDY. Could some one please suggest whether INDY would be a correct choice? And How to create blockchain using indy?

MinuJawahar (Mon, 05 Nov 2018 13:05:10 GMT):
Hello everyone! I am currently working with a biometrics project. I want it to be implement in block chain technology. I have chosen INDY for my project. Could some one please suggest whethe r INDY will be the correct choice or not? and how to create a blockchain using INDY?

MinuJawahar (Mon, 05 Nov 2018 13:05:10 GMT):
Hello everyone! Currently I am working with a Biometrics project. I want it to be implement in block chain technology. I have chosen INDY for my project. Could some one please suggest whether INDY will be the correct choice or not? and how to create a blockchain using INDY?

lightcap (Tue, 06 Nov 2018 00:33:57 GMT):
I seem to be able to change the Common Name for a CA without a problem using the `cryptogen` tool. I can validate it with `openssl x509`. But, I can't find *any place* to do the same for the TLS cert. It always uses `tlsca` as the hostname in the CN. Does anyone know to change either the CN for the TLS cert -or- the FQDN so it matches the default CN?

NV1988 (Tue, 06 Nov 2018 03:40:25 GMT):
Has joined the channel.

kannancet (Tue, 06 Nov 2018 22:27:43 GMT):
Has joined the channel.

kannancet (Tue, 06 Nov 2018 22:33:04 GMT):
Hello everyone. I am getting started with indy using docs here https://github.com/hyperledger/indy-sdk/blob/master/doc/mac-build.md. My installation fails at `cargo build` with `error: failed to run custom build command for `openssl-sys v0.9.36`. Can anyone help out ?

kannancet (Tue, 06 Nov 2018 22:33:04 GMT):
Hello everyone. I am getting started with indy using docs here https://github.com/hyperledger/indy-sdk/blob/master/doc/mac-build.md. My installation fails at `cargo build` with `error: failed to run custom build command for openssl-sys v0.9.36`. Can anyone help out ?

saeed_taurus (Wed, 07 Nov 2018 04:14:22 GMT):
Has joined the channel.

saeed_taurus (Wed, 07 Nov 2018 04:15:23 GMT):
Hi there is there any complete guide on how to build a hyperledger Indy test network

saeed_taurus (Wed, 07 Nov 2018 04:16:01 GMT):
I'm facing some difficulties in making pools and connecting clients

saeed_taurus (Wed, 07 Nov 2018 05:01:37 GMT):
I have made a pool

saeed_taurus (Wed, 07 Nov 2018 05:02:01 GMT):
But I don't know how to connect indy-cli to my pool

haggs (Wed, 07 Nov 2018 09:18:54 GMT):
@saeed_taurus This might help or guide you along the way

haggs (Wed, 07 Nov 2018 09:19:09 GMT):
https://github.com/blhagadorn/Indy-dev , simple setup that conencts to the pool

haggs (Wed, 07 Nov 2018 09:19:28 GMT):
Run the makefile commands for docker!

marchon (Wed, 07 Nov 2018 17:04:31 GMT):
Has joined the channel.

Sean_Bohan (Wed, 07 Nov 2018 19:37:02 GMT):
Gang: This Thursday's Indy WG call Agenda: - AGENTS! - Upcoming Indy Calls - LibVCX Update (RichardE) - Message Trust Contexts (DanielH) When: 8am PT, 9am MT, 10am CT, 11amET, 3pmGMT Where: Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/232861185 Or iPhone one-tap (US Toll): +16465588656,232861185# or +14086380968,232861185# Or Telephone: Dial: +1 646 558 8656 (US Toll) or +1 408 638 0968 (US Toll) Meeting ID: 232 861 185 International numbers available: https://zoom.us/zoomconference?m=a0jD_rTMnh0ZYGQDOKPCNrK_0dP7WPfp1

kasimmm (Thu, 08 Nov 2018 04:22:24 GMT):
Has joined the channel.

Taaanos (Thu, 08 Nov 2018 08:17:14 GMT):
Has left the channel.

HiranKumar (Thu, 08 Nov 2018 10:05:48 GMT):
Has joined the channel.

HiranKumar (Thu, 08 Nov 2018 10:06:33 GMT):
indy.createAndStoreMyDid always returns undefined.Please share your thoughts on this

mxs1491 (Thu, 08 Nov 2018 10:38:25 GMT):
Has joined the channel.

TelegramSam (Thu, 08 Nov 2018 15:32:16 GMT):
Project update delivered to the TSC on their call today: https://wiki.hyperledger.org/groups/tsc/project-updates/indy-2018-nov

lynn.bendixsen (Thu, 08 Nov 2018 15:47:10 GMT):
Thanks Sam! how did it go? @TelegramSam

sklump (Thu, 08 Nov 2018 15:51:57 GMT):
Has left the channel.

twshelton (Thu, 08 Nov 2018 15:53:24 GMT):
@kannancet It looks like your oppenssl installation might differ from the README doc. I can hop on a zoom chat with you and attempt to walk you through it if you still need assistance.

HiranKumar (Thu, 08 Nov 2018 15:53:47 GMT):
Hi,I always got error Commoninvalidstate, while connecting to the network using Indy-cli,what would be the common reason

HiranKumar (Thu, 08 Nov 2018 15:54:05 GMT):
Please help

TelegramSam (Thu, 08 Nov 2018 16:01:57 GMT):
Comment after presentation: "Indy is setting a high bar for project reports!"

TelegramSam (Thu, 08 Nov 2018 16:02:09 GMT):
@lynn.bendixsen ^

haggs (Thu, 08 Nov 2018 16:40:46 GMT):
Agent Taxonomy: https://docs.google.com/presentation/d/1ExQM_suu9MISrPanpK9sBGqVZFxf8pp4GvytcsKbkVM/edit#slide=id.p

haggs (Thu, 08 Nov 2018 17:09:14 GMT):
Continuing on with the conversation that was in the meeting: here's a transcript blurb: ``` Should the unpack method then be statically configured to determined the trust context and security updates over time change the static configuration?
 From Kyle Den Hartog to Everyone: (11:57 AM) 
Time check 3 minutes
 From Sam Curren (TelegramSam) to Everyone: (11:57 AM) 
Matthew, not sure what you mean.
 From Matthew Hailstone to Everyone: (11:58 AM) 
Unpack basically determines the security context from what I heard, correct?
 From Stephen Curran to Everyone: (11:59 AM) 
Unpack tells you what the security context is - the attributes of the message. Then you decide if you are OK with that.
 From Me to Everyone: (11:59 AM) 
I think I heard that, I guess my question would be how do you dynamically change the security context you’d like over time?
 From Matthew Hailstone to Everyone: (11:59 AM) 
If that’s the case, then the implementation of unpack needs to determine how safe the message is or based on current cryptography schemes and their current state determine whether the message is trusted (or other context states).
 From Matthew Hailstone to Everyone: (12:01 PM) 
I’m recommending that the determination of the trust context be something that is statically built into the version of the indy-sdk/unpack/pack functions and that updates to the indy-sdk would update how unpack determines the trust context.
 From Me to Everyone: (12:01 PM) 
Matthew, semi-related - Message trust entirely based on crypto and security is good, but seems not like the silver bullet - I think message trust along with reputation of those connecting (and over time) is the combination that solves this issue ```

haggs (Thu, 08 Nov 2018 17:10:31 GMT):
Anyone know how to snippet?

kannancet (Thu, 08 Nov 2018 20:47:20 GMT):
@twshelton - Resolved that issue and I was able to setup the pool. I am trying to run a boilerplate DID creation script as described on SDK guide and got stuck here now `OSError: dlopen(libindy.dylib, 6): image not found`

kannancet (Thu, 08 Nov 2018 20:47:20 GMT):
@twshelton - Resolved that issue and I was able to setup the pool. I am trying to run a boilerplate DID creation script as described on SDK guide and got stuck here now `OSError: dlopen(libindy.dylib, 6): image not found` . Any help would be appreciated :)

kannancet (Thu, 08 Nov 2018 20:47:20 GMT):
@twshelton - Resolved that issue and I was able to setup the pool. I am trying to run a boilerplate DID creation script using python as described on SDK guide and got stuck here now `OSError: dlopen(libindy.dylib, 6): image not found` . Any help would be appreciated :)

twshelton (Thu, 08 Nov 2018 21:18:23 GMT):
did you set LD_LIBRARY_PATH or DYLIB_LIBRARY_PATH? Looks like Python can't find libindy

twshelton (Thu, 08 Nov 2018 21:19:57 GMT):
@kannancet ^

kannancet (Thu, 08 Nov 2018 22:26:16 GMT):
@twshelton - Resolved, Stuck on another one now ``` _indy_loop_callback: Function returned error 113 Error occurred: ErrorCode.CommonInvalidStructure ```

mwherman2000 (Thu, 08 Nov 2018 22:34:13 GMT):
Has joined the channel.

mwherman2000 (Thu, 08 Nov 2018 22:36:01 GMT):
Suggestions for how to get started using Indy for "things"? I

mwherman2000 (Thu, 08 Nov 2018 22:36:01 GMT):
Suggestions for how to get started using Indy for "things"? I'm specifically interested in representing Non-Fungible Entities like purchase orders, invoices, waybills, time sheets, expense reports, etc. on Indy/Sovrin.

haggs (Fri, 09 Nov 2018 03:20:17 GMT):
@mwherman2000 I would tune into the next Overlays WG as there are some folks working in that realm specifically

haggs (Fri, 09 Nov 2018 03:21:20 GMT):
What's your slack or rocketchat sovrin Username? I'll invite you to the room

mwherman2000 (Fri, 09 Nov 2018 04:12:06 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=e3BYApw53agrgSMgF) @haggs mwherman2000

haggs (Fri, 09 Nov 2018 04:18:02 GMT):
schemas-and-overlays

haggs (Fri, 09 Nov 2018 04:18:11 GMT):
@mwherman2000

haggs (Fri, 09 Nov 2018 04:18:56 GMT):
couldn't add you but I'd join that room and attend next week's meeting

haggs (Fri, 09 Nov 2018 04:19:46 GMT):
as well as indy-agent, as the agents will be dealing with those.

mwherman2000 (Fri, 09 Nov 2018 04:20:54 GMT):
Which room? ....do you have a link?

haggs (Fri, 09 Nov 2018 04:23:19 GMT):
https://sovrin.slack.com/messages/CA9NZGG3Z

HiranKumar (Fri, 09 Nov 2018 10:34:58 GMT):
while using indy-cli,pool is created,but could not able to connect to the network using pool.I got the below error while connecting indy> pool connect pool45 Indy SDK error occurred CommonInvalidState

haggs (Fri, 09 Nov 2018 12:18:24 GMT):
@HiranKumar if using docker, try throwing ```--no-cache```

HiranKumar (Fri, 09 Nov 2018 14:10:16 GMT):
yes, I am using docker container

HiranKumar (Fri, 09 Nov 2018 14:10:53 GMT):
@haggs yes.I am using docker container

HiranKumar (Fri, 09 Nov 2018 14:17:48 GMT):
@haggs also got the same error while using Nodejs sdk

haggs (Fri, 09 Nov 2018 14:18:35 GMT):
Try rebuilding it with that flag, I've found it cleans up a lot of things docker likes to keep.

HiranKumar (Fri, 09 Nov 2018 14:36:52 GMT):
@haggs Same error

haggs (Fri, 09 Nov 2018 14:39:25 GMT):
Their might be an ip address you need to change

haggs (Fri, 09 Nov 2018 14:39:25 GMT):
There might be an ip address you need to change

mwherman2000 (Fri, 09 Nov 2018 14:51:00 GMT):
If a DID/DID-document referred to/was, say, a purchase order or an invoice, who would the trusted Issuer be? ...ideally as well as in practice? Would it be the Company or Company Department or Company Person who authorized/approved/issued the specific document? ...or would the previous be expected to submit the document to some sort of issuing authority (outside the Company)? ...who would then become the trusted Issuer for that specific document?

HiranKumar (Fri, 09 Nov 2018 15:20:54 GMT):
@haggs address in the genesis file?

HiranKumar (Fri, 09 Nov 2018 15:20:54 GMT):
@haggs ,Is there any agent call now?

HiranKumar (Fri, 09 Nov 2018 15:20:54 GMT):
@haggs I mean any meeting regarding agent call i sovrin

haggs (Fri, 09 Nov 2018 15:23:35 GMT):
Yes hop over to indy-agent :D

haggs (Fri, 09 Nov 2018 15:24:01 GMT):
Details in there

twshelton (Fri, 09 Nov 2018 16:46:30 GMT):
@kannancet Thats a new one ... we can jump on a zoom chat and try to figure it out.

haggs (Fri, 09 Nov 2018 16:56:46 GMT):
@twshelton Care if I join

haggs (Fri, 09 Nov 2018 16:59:39 GMT):
Or get the outcome, also curious. I'd like to see a set of versions and environment commands we can use to help identify problems

danielhardman (Fri, 09 Nov 2018 16:59:51 GMT):
@JanL (I think I have the right handle) wanted to socialize a HIPE proposal that was built around the concept of consent receipts, but was having trouble with Rocket Chat--so I said I'd drop a link here. https://github.com/hyperledger/indy-hipe/pull/55

pknowles (Fri, 09 Nov 2018 17:01:12 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=AkhrZtdQrBZyG73Hk) @danielhardman @JanL 5 appears to be the right handle.

janl (Fri, 09 Nov 2018 17:05:03 GMT):
Thanks @danielhardman. Now I managed to connect. The initial goal with the HIPE was to document what has been informally been discussed surrounding how consent receipt shall be used. Hope the community is in support of the initial proposal. Thanx

HiranKumar (Fri, 09 Nov 2018 17:12:39 GMT):
@haggs I created the pool using the docker container which is in side a vm.docker build -f ci/indy-pool.dockerfile -t indy_pool . docker run -itd -p 9701-9708:9701-9708 indy_pool Then I tried to connect the network using the indy-cli.Pool created using the below genesis file

HiranKumar (Fri, 09 Nov 2018 17:12:41 GMT):
{"reqSignature": {}, "txn": {"data": {"data": {"alias": "Node2", "blskey": "37rAPpXVoxzKhz7d9gkUe52XuXryuLXoM6P6LbWDB7LSbG62Lsb33sfG7zqS8TK1MXwuCHj1FKNzVpsnafmqLG1vXN88rt38mNFs9TENzm4QHdBzsvCuoBnPH7rpYYDo9DZNJePaDvRvqJKByCabubJz3XXKbEeshzpz4Ma5QYpJqjk", "blskey_pop": "Qr658mWZ2YC8JXGXwMDQTzuZCWF7NK9EwxphGmcBvCh6ybUuLxbG65nsX4JvD4SPNtkJ2w9ug1yLTj6fgmuDg41TgECXjLCij3RMsV8CwewBVgVN67wsA45DFWvqvLtu4rjNnE9JbdFTc1Z4WCPA3Xan44K1HoHAq9EVeaRYs8zoF5", "client_ip": "104.211.210.10", "client_port": 9704, "node_ip": "104.211.210.10", "node_port": 9703, "services": ["VALIDATOR"]}, "dest": "8ECVSk179mjsjKRLWiQtssMLgp6EPhWXtaYyStWPSGAb"}, "metadata": {"from": "EbP4aYNeTHL6q385GuVpRV"}, "type": "0"}, "txnMetadata": {"seqNo": 2, "txnId": "1ac8aece2a18ced660fef8694b61aac3af08ba875ce3026a160acbc3a3af35fc"}, "ver": "1"} {"reqSignature": {}, "txn": {"data": {"data": {"alias": "Node3", "blskey": "3WFpdbg7C5cnLYZwFZevJqhubkFALBfCBBok15GdrKMUhUjGsk3jV6QKj6MZgEubF7oqCafxNdkm7eswgA4sdKTRc82tLGzZBd6vNqU8dupzup6uYUf32KTHTPQbuUM8Yk4QFXjEf2Usu2TJcNkdgpyeUSX42u5LqdDDpNSWUK5deC5", "blskey_pop": "QwDeb2CkNSx6r8QC8vGQK3GRv7Yndn84TGNijX8YXHPiagXajyfTjoR87rXUu4G4QLk2cF8NNyqWiYMus1623dELWwx57rLCFqGh7N4ZRbGDRP4fnVcaKg1BcUxQ866Ven4gw8y4N56S5HzxXNBZtLYmhGHvDtk6PFkFwCvxYrNYjh", "client_ip": "104.211.210.10", "client_port": 9706, "node_ip": "104.211.210.10", "node_port": 9705, "services": ["VALIDATOR"]}, "dest": "DKVxG2fXXTU8yT5N7hGEbXB3dfdAnYv1JczDUHpmDxya"}, "metadata": {"from": "4cU41vWW82ArfxJxHkzXPG"}, "type": "0"}, "txnMetadata": {"seqNo": 3, "txnId": "7e9f355dffa78ed24668f0e0e369fd8c224076571c51e2ea8be5f26479edebe4"}, "ver": "1"} {"reqSignature": {}, "txn": {"data": {"data": {"alias": "Node1", "blskey": "4N8aUNHSgjQVgkpm8nhNEfDf6txHznoYREg9kirmJrkivgL4oSEimFF6nsQ6M41QvhM2Z33nves5vfSn9n1UwNFJBYtWVnHYMATn76vLuL3zU88KyeAYcHfsih3He6UHcXDxcaecHVz6jhCYz1P2UZn2bDVruL5wXpehgBfBaLKm3Ba", "blskey_pop": "RahHYiCvoNCtPTrVtP7nMC5eTYrsUA8WjXbdhNc8debh1agE9bGiJxWBXYNFbnJXoXhWFMvyqhqhRoq737YQemH5ik9oL7R4NTTCz2LEZhkgLJzB3QRQqJyBNyv7acbdHrAT8nQ9UkLbaVL9NBpnWXBTw4LEMePaSHEw66RzPNdAX1", "client_ip": "104.211.210.10", "client_port": 9702, "node_ip": "104.211.210.10", "node_port": 9701, "services": ["VALIDATOR"]}, "dest": "Gw6pDLhcBcoQesN72qfotTgFa7cbuqZpkX3Xo6pLhPhv"}, "metadata": {"from": "Th7MpTaRZVRYnPiabds81Y"}, "type": "0"}, "txnMetadata": {"seqNo": 1, "txnId": "fea82e10e894419fe2bea7d96296a6d46f50f93f9eeda954ec461b2ed2950b62"}, "ver": "1"} {"reqSignature": {}, "txn": {"data": {"data": {"alias": "Node4", "blskey": "2zN3bHM1m4rLz54MJHYSwvqzPchYp8jkHswveCLAEJVcX6Mm1wHQD1SkPYMzUDTZvWvhuE6VNAkK3KxVeEmsanSmvjVkReDeBEMxeDaayjcZjFGPydyey1qxBHmTvAnBKoPydvuTAqx5f7YNNRAdeLmUi99gERUU7TD8KfAa6MpQ9bw", "blskey_pop": "RPLagxaR5xdimFzwmzYnz4ZhWtYQEj8iR5ZU53T2gitPCyCHQneUn2Huc4oeLd2B2HzkGnjAff4hWTJT6C7qHYB1Mv2wU5iHHGFWkhnTX9WsEAbunJCV2qcaXScKj4tTfvdDKfLiVuU2av6hbsMztirRze7LvYBkRHV3tGwyCptsrP", "client_ip": "104.211.210.10", "client_port": 9708, "node_ip": "104.211.210.10", "node_port": 9707, "services": ["VALIDATOR"]}, "dest": "4PS3EDQ3dW1tci1Bp6543CfuuebjFrg36kLAUcskGfaA"}, "metadata": {"from": "TWwCRQRZ2ZHMJFn9TzLp7W"}, "type": "0"}, "txnMetadata": {"seqNo": 4, "txnId": "aa5e817d7cc626170eca175822029339a444eb0ee8f0bd20d3b0b76e566fb008"}, "ver": "1"}

HiranKumar (Fri, 09 Nov 2018 17:14:26 GMT):
@haggs Here ip address I mentioned in the genesis file is ip address of the vm

haggs (Fri, 09 Nov 2018 17:21:56 GMT):
@kannancet and @twshelton https://github.com/hyperledger/indy-sdk/blob/fcf7cbb0bf2aa410e0f8ae2ff32f46271be5d5df/libindy/src/errors/common.rs#L43 https://github.com/hyperledger/indy-sdk/blob/55aaec34e023980cfc6dc8bf912006cf3b8e8c88/wrappers/nodejs/test/index.js#L32

haggs (Fri, 09 Nov 2018 17:22:00 GMT):
There's a start

haggs (Fri, 09 Nov 2018 17:23:49 GMT):
Looks like the key is abbreviated

haggs (Fri, 09 Nov 2018 17:24:53 GMT):
`err = await t.throws(indy.abbreviateVerkey('?', verkey)) t.is(err + '', 'IndyError: CommonInvalidStructure') t.is(err.indyCode, 113) t.is(err.indyName, 'CommonInvalidStructure')`

haggs (Fri, 09 Nov 2018 17:24:53 GMT):
```err = await t.throws(indy.abbreviateVerkey('?', verkey)) t.is(err + '', 'IndyError: CommonInvalidStructure') t.is(err.indyCode, 113) t.is(err.indyName, 'CommonInvalidStructure')```

haggs (Fri, 09 Nov 2018 17:26:09 GMT):
Not not abbreviated

haggs (Fri, 09 Nov 2018 17:30:48 GMT):
@HiranKumar Which of these methods are you using? Looks like you have your client IP address in there (I would hide that)

haggs (Fri, 09 Nov 2018 17:30:51 GMT):
https://github.com/hyperledger/indy-sdk#how-to-start-local-nodes-pool-with-docker

haggs (Fri, 09 Nov 2018 17:32:10 GMT):
Some custom docker port mapping you might have to do

haggs (Fri, 09 Nov 2018 17:32:16 GMT):
ah ok ip of vm

haggs (Fri, 09 Nov 2018 17:33:03 GMT):
What OS are you using? Docker inside of a VM is somewhat unchartered

HiranKumar (Fri, 09 Nov 2018 17:33:18 GMT):
ubantu 16.04

HiranKumar (Fri, 09 Nov 2018 17:33:55 GMT):
@haggs ubuntu 16.04

haggs (Fri, 09 Nov 2018 17:34:23 GMT):
As the VM OS?

HiranKumar (Fri, 09 Nov 2018 17:35:29 GMT):
both are ubuntu

twshelton (Fri, 09 Nov 2018 19:48:00 GMT):
@haggs @kannancet Bryant ... will keep you posted if we set up a zoom time

haggs (Fri, 09 Nov 2018 21:16:20 GMT):
Right on!

gy8409i (Sun, 11 Nov 2018 03:05:43 GMT):
Has joined the channel.

kannancet (Sun, 11 Nov 2018 04:17:03 GMT):
@twshelton - Sure. Thanks. Looking forward to it :)

mwherman2000 (Sun, 11 Nov 2018 10:43:26 GMT):
Hi, I'm working through section 3.7 Definitions and Terminology in the Provisional Trust Framework. Looking at just the first 7 definitions on page 11, they don't hang/connect together as good as I expect them to. For example, I'm trying to draw a conceptual/logical architecture diagram/data model relating these 7 concepts and it's not that easy. Are these definitions frozen? Are people interested in feedback/improvements? Does anyone know of an existing (detailed) conceptual/logical architecture diagram/data model for Sovrin?

swcurran (Sun, 11 Nov 2018 20:07:15 GMT):
I've posted a HIPE that proposes we move away from the current PR-centric approach to HIPEs, where a PR for a HIPE is only merged after it is accepted, to one that is based on merging HIPE PRs more or less immediately, and tracking the status of a HIPE by it's location in a set of repo folders - propsed, final-review, accepted, archived and rejected. The PR for the HIPE can be found here: https://github.com/hyperledger/indy-hipe/pull/56 The current text of the HIPE can be found here: https://github.com/hyperledger/indy-hipe/blob/31446b154f2ac8eb991b6e1d4725623f7ba587a8/text/new-hipe-process/new-hipe-process.md

gy8409i (Mon, 12 Nov 2018 02:40:02 GMT):
What does HIPE stand for?

azur3s0ng (Mon, 12 Nov 2018 03:12:51 GMT):
Has left the channel.

swcurran (Mon, 12 Nov 2018 03:43:05 GMT):
HIPE is Hyperledger Indy Product Enhancements

gy8409i (Mon, 12 Nov 2018 07:36:19 GMT):
Thanks. Btw, I got several errors when I build libindy-crypto.dylib on Mac. Could you give me some help on it?

gy8409i (Mon, 12 Nov 2018 07:37:19 GMT):
error: Could not compile `version_check`, `proc-macro2`, `cfg-if`, `typenum`,`libc`....

uhef (Mon, 12 Nov 2018 08:56:45 GMT):
Has joined the channel.

uhef (Mon, 12 Nov 2018 09:03:36 GMT):
Hi All! I'm newbie with Indy and Sovrin so excuse for potentially stupid questions but I would be interested in knowing if there is project to provide a Sovrin mobile wallet app going on anywhere. Functionality that I see this could provide are for instance: - A service (as in an Android service) that runs on a background and that maintains the user wallet on the device. User interface components can integrate with this service to provide user with information about the wallet or to make changes to the wallet. - User interface to see the state of the wallet (perhaps some functions around updating the wallet as well, even though those might be provided by separate, case-specific user interface applications) - Way to backup and restore the contents of a wallet on users explicit approval (to a cloud provider, probably)

wangdong (Mon, 12 Nov 2018 09:51:22 GMT):
is there a limit to the number of steward nodes?

wangdong (Mon, 12 Nov 2018 09:52:16 GMT):
I think that the all the service provider can be as a validator node, but from the document I found 'For the General Availability Network, once there is a sufficient number of Stewards, the Sovrin Trust Framework SHOULD incorporate policies for ensuring System Diversity and Diffuse Trust that can be enforced algorithmically and dynamically by the Sovrin Open Source Code operating on all Nodes to determine which Nodes should be operating as Validator Nodes at any one point in time.`

wangdong (Mon, 12 Nov 2018 09:52:16 GMT):
I think that the all the service provider can be as a validator node, but from the document I found `For the General Availability Network, once there is a sufficient number of Stewards, the Sovrin Trust Framework SHOULD incorporate policies for ensuring System Diversity and Diffuse Trust that can be enforced algorithmically and dynamically by the Sovrin Open Source Code operating on all Nodes to determine which Nodes should be operating as Validator Nodes at any one point in time.`

wangdong (Mon, 12 Nov 2018 09:52:16 GMT):
I think that the all the service provider can be as a validator node, but from the document I found ```For the General Availability Network, once there is a sufficient number of Stewards, the Sovrin Trust Framework SHOULD incorporate policies for ensuring System Diversity and Diffuse Trust that can be enforced algorithmically and dynamically by the Sovrin Open Source Code operating on all Nodes to determine which Nodes should be operating as Validator Nodes at any one point in time.```

wangdong (Mon, 12 Nov 2018 09:52:53 GMT):
So the role of node is changing from time to time.

wangdong (Mon, 12 Nov 2018 09:53:44 GMT):
but how to guareet that the service is consistent for a service provider?

wangdong (Mon, 12 Nov 2018 09:53:44 GMT):
but how to guarantee that the service is consistent for a service provider?

wangdong (Mon, 12 Nov 2018 14:25:12 GMT):
```Could not find a version that satisfies the requirement sovtoken==0.9.5 (from sovrin) (from versions: ) No matching distribution found for sovtoken==0.9.5 (from sovrin)```

wangdong (Mon, 12 Nov 2018 14:25:32 GMT):
when I try to install sovrin from cmd line.

twshelton (Mon, 12 Nov 2018 16:11:39 GMT):
@kannancet let me know what works for you this week. I generally available with a little notice.

umamaheswarv (Mon, 12 Nov 2018 16:48:31 GMT):
Has joined the channel.

trthhrtz (Mon, 12 Nov 2018 18:32:55 GMT):
Hi everyone! I started a list of awesome resources to create application with Hyperledger Indy (https://github.com/trthhrtz/hyperledger-indy-awesome). Feel free to create PR's!

trthhrtz (Mon, 12 Nov 2018 18:32:55 GMT):
Hi everyone! I started a list of awesome resources to create applications with Hyperledger Indy (https://github.com/trthhrtz/hyperledger-indy-awesome). Feel free to create PR's!

TelegramSam (Mon, 12 Nov 2018 19:19:37 GMT):
New HIPE Approved: DID Doc Conventions, by @swcurran https://github.com/hyperledger/indy-hipe/tree/master/text/0023-diddoc-conventions

TelegramSam (Mon, 12 Nov 2018 19:26:29 GMT):
New HIPE Approved: Adding forward secrecy to Agent to Agent communications, by @MALodder: https://github.com/hyperledger/indy-hipe/tree/master/text/0024-a2a-forward-secrecy

LucW (Tue, 13 Nov 2018 10:19:02 GMT):
I don't know if this was mentioned already or not, but it would be really cool to develop an app for the Ledger Nano https://ledger.readthedocs.io/en/latest/additional/publishing_an_app.html So people could use it to authenticate with it. Though I don't know how exactly it would work and if they'd accept such early stage application

Sean_Bohan (Tue, 13 Nov 2018 20:42:55 GMT):
This Week's Hyperledger Indy WG Call: Date: Thurs Nov 15 When: 8am PT, 9am MT, 10am CT, 11amET, 4pmGMT Where: https://zoom.us/j/232861185 Agenda: - Overlays 101 - presentation/demo (P.Knowles/R.Mitwicki) - IXO Team Presentation (IXO Team) Previous Recordings and Agendas: https://drive.google.com/drive/u/0/folders/1AwHWN95KmSEi5fijraID0tFFMzYHoMwt?ogsrc=32 (edited)

Sean_Bohan (Tue, 13 Nov 2018 20:42:55 GMT):
Date: Thurs Nov 15 When: 8am PT, 9am MT, 10am CT, 11amET, 4pmGMT Where: https://zoom.us/j/232861185 Agenda: - Overlays 101 - presentation/demo (P.Knowles/R.Mitwicki) - IXO Team Presentation (IXO Team) Previous Recordings and Agendas: https://drive.google.com/drive/u/0/folders/1AwHWN95KmSEi5fijraID0tFFMzYHoMwt?ogsrc=32 (edited)

mwherman2000 (Wed, 14 Nov 2018 19:25:35 GMT):
My feedback added to the the following Medium.com article: https://hackernoon.com/the-one-thing-missing-from-erc-721-standard-for-digital-collectibles-on-the-blockchain-9ee26e4a918c "Once a Non-Fungible Entity (e.g. [Person], [Organization], [Thing], Cryptokitty, Purchase Order, Invoice, Shipping Instructions, Delivery Confirmation, etc.) have been fully realized and encased in concrete plus 100 layers of titanium and digitally signed so that it can never be altered, why can't it be accessed using a trusted service within a smart contract VM?…why can't there be a set of simple standard VM-level interop methods for querying any NFE found anywhere on the planet, selecting an entity or entities from the result set, and then calling on a standard set of accessors to deserialize the properties desired by a smart contract invocation which, in turn, is being called by an off-chain app or another on-chain smart contract?" Is anyone aware of a blockchain/VM project that is already undertaking this approach for direct NFE access from a smart contract?

Silona (Wed, 14 Nov 2018 21:38:34 GMT):
Has joined the channel.

haow85 (Thu, 15 Nov 2018 08:56:09 GMT):
Has joined the channel.

haow85 (Thu, 15 Nov 2018 08:57:09 GMT):
We are working on a survey on HyperLedger Indy. We would like to learn about Indy usage globally. Anybody can give some good ideas on this ?

mwherman2000 (Thu, 15 Nov 2018 13:43:21 GMT):
...a use case focused survey?

mwherman2000 (Thu, 15 Nov 2018 13:43:47 GMT):
...size/scalability metrics?

mwherman2000 (Thu, 15 Nov 2018 13:44:15 GMT):
...scopes? ...in terms of Persons, Orgs, Things

HiranKumar (Thu, 15 Nov 2018 13:45:26 GMT):
I am trying to create a network and start nodes.

HiranKumar (Thu, 15 Nov 2018 13:46:00 GMT):
Docker or manual step,which one is better?

HiranKumar (Thu, 15 Nov 2018 13:46:36 GMT):
I am mentioning about Indy -hyper ledger

swcurran (Thu, 15 Nov 2018 17:01:04 GMT):
@HiranKumar - if you just want to get something running to see it in action and look a the code, the easiest I know of is this - can be tried in just a browser: https://github.com/hyperledger/education/blob/master/LFS171x/indy-material/nodejs/README.md

swcurran (Thu, 15 Nov 2018 17:01:04 GMT):
@HiranKumar - if you just want to get something running to see it in action and look some code, the easiest I know of is this - can be tried in just a browser: https://github.com/hyperledger/education/blob/master/LFS171x/indy-material/nodejs/README.md

kannancet (Thu, 15 Nov 2018 19:26:37 GMT):
@twshelton - Shall we zoom around 9PM CET tomorrow ? Or you can let me know your convenient time.

kannancet (Thu, 15 Nov 2018 19:28:04 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=ze25aWtuh9WYfd5pL) @twshelton @twshelton - Shall we zoom around 9PM CET tomorrow ? Or you can let me know your convenient time.

twshelton (Thu, 15 Nov 2018 19:49:12 GMT):
@kannancet that should work. Are you in Berlin? @haggs does that work for you?

haggs (Thu, 15 Nov 2018 19:51:18 GMT):
2pm Central for me tomorrow?

haggs (Thu, 15 Nov 2018 19:51:20 GMT):
Sure

haggs (Thu, 15 Nov 2018 19:51:54 GMT):
I could do today as well, before we go - where are you in this journey so far?

kannancet (Thu, 15 Nov 2018 19:58:43 GMT):
@twshelton @haggs - I am based in Berlin. I will explain my product use-case on call. Currently I am just getting started using https://github.com/blhagadorn/Indy-dev and trying to run basic tutorial for storing and retrieving DIDs

haggs (Thu, 15 Nov 2018 20:01:42 GMT):
How about this

haggs (Thu, 15 Nov 2018 20:02:13 GMT):
https://github.com/kdenhartog/Indy-dev

haggs (Thu, 15 Nov 2018 20:02:18 GMT):
The original

haggs (Thu, 15 Nov 2018 20:02:25 GMT):
My fork is messed up

twshelton (Thu, 15 Nov 2018 20:02:57 GMT):
this is a really slick Docker based setup ... try this one. I believe this is what we will be moving to very soon.

twshelton (Thu, 15 Nov 2018 20:02:58 GMT):
https://github.com/hyperledger/education/blob/master/LFS171x/indy-material/nodejs/README.md

twshelton (Thu, 15 Nov 2018 20:03:54 GMT):
the one @haggs just referenced is a step in that direction but the one I posted can be completely setup in a cloud docker-based session thats pretty quick and easy to manage

kannancet (Thu, 15 Nov 2018 20:03:57 GMT):
I was looking into it. Looks awesome. Will be trying it out.

haggs (Thu, 15 Nov 2018 20:19:10 GMT):
I think that's preferrable

haggs (Thu, 15 Nov 2018 20:20:42 GMT):
@twshelton Docker compose is the way to go. Currently working on a rendition similar, https://github.com/hyperledger/indy-agent/pull/33

haggs (Thu, 15 Nov 2018 20:21:10 GMT):
that PR is still burning and is the same stuff beautiful work by @swcurran and his team

haow85 (Fri, 16 Nov 2018 01:31:38 GMT):
There is a built-in zero-knowledge-proof functionality in HyperLedger Indy called idenxkeygen or something similar but I forgot the name . Can someone help me with the name ?

haow85 (Fri, 16 Nov 2018 04:12:36 GMT):
a use case focued survey

swcurran (Fri, 16 Nov 2018 04:36:47 GMT):
Identity Mixer from IBM - idemix - https://www.zurich.ibm.com/identity_mixer/

haow85 (Fri, 16 Nov 2018 04:57:09 GMT):
Is identity mixer for Fabric only or it's also working for Indy ?

haow85 (Fri, 16 Nov 2018 05:36:43 GMT):
@mwherman2000 a use case focused survey.

Ankita1 (Fri, 16 Nov 2018 06:51:41 GMT):
Has joined the channel.

Ankita1 (Fri, 16 Nov 2018 06:56:28 GMT):
Hello Everyone, I am setting up the indy-world demo using https://github.com/IBM-Blockchain-Identity/indy-ssivc-tutorial. While I try to run ./manage.sh start, it is giving issues while pulling schema-spy. Pulling schema-spy (schema-spy:)... ERROR: The image for the service you're trying to recreate has been removed. If you continue, volume data could be lost. Consider backing up your data before continuing. Continue with the new image? [yN]y Pulling schema-spy (schema-spy:)... ERROR: pull access denied for schema-spy, repository does not exist or may require 'docker login' Does anyone have an idea what issue it could be? in the ./manage.sh build, the schema-spy image was built, but when I check docker image ls, I do not see the schema-spy image present. Not sure how it gets deleted. Please let me know if you have any suggestions. Thanks

DCSnail (Fri, 16 Nov 2018 09:59:56 GMT):
Has joined the channel.

swcurran (Fri, 16 Nov 2018 15:05:03 GMT):
You can ignore the schema spy piece. It is just used to look at the data schema, and really should be removed from that. If that's the only issue, you can proceed.

kannancet (Fri, 16 Nov 2018 19:47:51 GMT):
@twshelton - Hi extremely sorry. Got stuck with a personal exigency today. Can we reschedule our call ?

twshelton (Fri, 16 Nov 2018 19:59:55 GMT):
@kannancet @haggs absolutely ... was starting to get things in order so no worries. Let me know a good time next week and we'll get it scheduled.

kannancet (Fri, 16 Nov 2018 20:05:31 GMT):
@haggs @twshelton - Definitely. Tuesday 8pm CET would work for you ?

haggs (Fri, 16 Nov 2018 20:28:58 GMT):
Possibly

haggs (Fri, 16 Nov 2018 20:29:06 GMT):
@kannancet I'

haggs (Fri, 16 Nov 2018 20:29:11 GMT):
ll let you know Monday

stanleyc-trustscience (Fri, 16 Nov 2018 21:21:02 GMT):
Hi guys, I was wondering what the latest development and discussion for agent to agent protocol?

dbluhm (Fri, 16 Nov 2018 21:21:39 GMT):
@stanleyc-trustscience join us in #indy-agent!

stanleyc-trustscience (Fri, 16 Nov 2018 21:22:16 GMT):
@dbluhm Thank you! :)

InfoMiner (Sat, 17 Nov 2018 01:07:10 GMT):
https://github.com/infominer33/awesome-decentralized-id

InfoMiner (Sat, 17 Nov 2018 01:07:27 GMT):
would love some help filling in the gaps

InfoMiner (Sat, 17 Nov 2018 01:07:27 GMT):
would love some help filling in the gaps re: indy

scli (Sun, 18 Nov 2018 11:25:01 GMT):
Has joined the channel.

baconsandwich (Sun, 18 Nov 2018 14:48:32 GMT):
What would be the rational on why Trustees should not add validator nodes and this is reserved to Stewards? Is this more like a separation of concerns? I mean by proxy of creating a steward they can circumvent this...althogh this action would be public of course. But why was this decided on purpose?

baconsandwich (Sun, 18 Nov 2018 14:48:32 GMT):
What would be the rationale on why Trustees should not add validator nodes and this is reserved to Stewards? Is this more like a separation of concerns? I mean by proxy of creating a steward they can circumvent this...althogh this action would be public of course. But why was this decided on purpose?

mwherman2000 (Sun, 18 Nov 2018 17:21:13 GMT):
What is an example for the most detailed Indy/Sovrin software architecture/data model diagrams available?

drummondreed (Mon, 19 Nov 2018 02:37:38 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=dpCLxHcEb29rDfF5h) @baconsandwich Can you give some context on your question? As far as I understand it, with the current permissioning model, both Trustees and Stewards could add validator nodes.

baconsandwich (Mon, 19 Nov 2018 02:44:34 GMT):
My source is this: https://docs.google.com/spreadsheets/d/1TWXF7NtBjSOaUIBeIH77SyZnawfo91cJ_ns4TR-wsq4/edit#gid=0 See row 12

baconsandwich (Mon, 19 Nov 2018 02:44:34 GMT):
@drummondreed My source is this: https://docs.google.com/spreadsheets/d/1TWXF7NtBjSOaUIBeIH77SyZnawfo91cJ_ns4TR-wsq4/edit#gid=0 See row 12

pknowles (Mon, 19 Nov 2018 08:13:24 GMT):
*SSIMeetup* webinar today at 1pm MT / 9pm CET - Title: *Overlays 1O1: Establishing Schema Definitions within the Self-Sovereign Identity (SSI) Ecosystem* by @pknowles + live demo from Robert Mitwicki - http://ssimeetup.org/overlays-1o1-establishing-schema-definitions-self-sovereign-identity-ssi-ecosystem-paul-knowles-webinar-17/

mwherman2000 (Mon, 19 Nov 2018 13:06:37 GMT):
Naive question: Is there an existing Indy/Sovrin wallet that I can download and start using?

mwherman2000 (Mon, 19 Nov 2018 13:06:37 GMT):
Naive question: Is there an existing Indy/Sovrin wallet implementation that I can download/access and start using?

mwherman2000 (Mon, 19 Nov 2018 13:06:37 GMT):
Naive question: Is there an existing Indy/Sovrin wallet implementation that I can download/access and start using? ...any specific recommendations if there are move than one?

mgbailey (Mon, 19 Nov 2018 14:49:43 GMT):
@baconsandwich @drummondreed The trustee role in indy cannot add node credentials to the ledger, and belongs to an individual rather than an organization. However, a trustee can be the employee of an organization that is a steward. There are a few instances of this now.

Sean_Bohan (Mon, 19 Nov 2018 15:33:38 GMT):
Last week's Indy WG Call featuring Overlays video has been posted and can be found here: https://drive.google.com/open?id=1a4ydpu6RDlyrqWX7eLomElR_CTj8hUug

nage (Mon, 19 Nov 2018 15:40:33 GMT):
@tkuhrt or @rjones may we get #indy-semantics created as a channel with @pknowles @kenebert and I as owners? We are moving the effort that started in Sovrin Slack over here. 😇

vishumanvi (Mon, 19 Nov 2018 19:07:15 GMT):
Has joined the channel.

pknowles (Mon, 19 Nov 2018 21:14:09 GMT):
:bangbang: Following the progressive work being undertaken by the Schemas and Overlays WG and the W3C efforts around JSON-LD that Ken Ebert has been spearheading, a new *#indy-semantics* channel has been created in *chat.hyperledger.org* . In order for all data capture and semantics discussion threads to be housed under one roof, all sub-categorical channels that fall into that remit will be archived over the next few weeks. In the meantime, head over to *chat.hyperledger.org* [ *#indy-semantics* ] to engage in all discussions related to data capture and semantics including schemas and overlays. @kenebert , @nage and I are the owners of the new channel. If you have any queries about the migration process, please feel free to message me directly. :bangbang:

pknowles (Mon, 19 Nov 2018 21:14:09 GMT):
:bangbang: Following the progressive work being undertaken by the Schemas and Overlays WG and the W3C efforts around JSON-LD that Ken Ebert has been spearheading, a new *#indy-semantics* channel has been created in *chat.hyperledger.org* . In order for all data capture and semantics discussion threads to be housed under one roof, all associated channels that fall into that remit will be archived over the next few weeks. In the meantime, head over to *chat.hyperledger.org* [ *#indy-semantics* ] to engage in all discussions related to data capture and semantics including schemas and overlays. @kenebert , @nage and I are the owners of the new channel. If you have any queries about the migration process, please feel free to message me directly. :bangbang:

pknowles (Mon, 19 Nov 2018 21:14:09 GMT):
:bangbang: Following the progressive work being undertaken by the Schemas and Overlays WG and the W3C efforts around JSON-LD that Ken Ebert has been spearheading, a new *#indy-semantics* channel has been created. In order for all data capture and semantics discussion threads to be housed under one roof, all associated channels that fall into that remit will be archived over the next few weeks. In the meantime, head over to *#indy-semantics* to engage in all discussions related to data capture and semantics including schemas and overlays. @kenebert, @nage and I are the owners of the new channel. If you have any queries about the migration process, please feel free to message me directly. :bangbang:

pknowles (Mon, 19 Nov 2018 21:14:09 GMT):
:bangbang: Following the progressive work being undertaken by the Schemas and Overlays WG and the W3C efforts around JSON-LD that Ken Ebert has been spearheading, a new *#indy-semantics* channel has been created. In order for all data capture and semantics discussion threads to be housed under one roof, all associated channels that fall under that remit will be archived over the next few weeks. In the meantime, head over to *#indy-semantics* to engage in all discussions related to data capture and semantics including schemas and overlays. @kenebert, @nage and I are the owners of the new channel. If you have any queries about the migration process, please feel free to message me directly. :bangbang:

drummondreed (Mon, 19 Nov 2018 23:19:53 GMT):
Great stuff, @pknowles. It just highlights that, as we get the rest of SSI infrastructure into place (no small job), interoperable semantics will become the whole focus. Great to see this happening.

pradeepjain (Tue, 20 Nov 2018 05:09:40 GMT):
Has joined the channel.

pradeepjain (Tue, 20 Nov 2018 05:13:24 GMT):
Hi All I am trying to get the credentials which are issued from trust anchor to the user, But I am getting following error: TypeError: cryptoAuthDecrypt expects Uint8Array for encryptedMsgRaw     at Object.cryptoAuthDecrypt (/root/indy-sdk/wrappers/nodejs/src/index.js:243:8)     at Object.authDecrypt (/root/api/functions.js:120:63)     at credIssue (/root/api/routes/anchor.js:211:67) POST /anchor/credential 400 2465.311 ms – 11

pknowles (Tue, 20 Nov 2018 12:44:23 GMT):
The video recording and slideshare presentation from yesterday’s *SSIMeetup* are now available for viewing. The title of the presentation: " *Overlays 1O1: Establishing Schema Definitions within the Self-Sovereign Identity (SSI) Ecosystem* ". https://chat.hyperledger.org/channel/indy-semantics?msg=66Pr6ptZMw5aDkGE3

pknowles (Tue, 20 Nov 2018 12:44:23 GMT):
https://chat.hyperledger.org/channel/indy-semantics?msg=66Pr6ptZMw5aDkGE3

kannancet (Tue, 20 Nov 2018 14:23:04 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=XLKZvFKN8sTdZ3oBX) @haggs @haggs @twshelton - Shall we do it today at 8pm CET ?

twshelton (Tue, 20 Nov 2018 19:57:22 GMT):
@kannancet @haggs ... I've been stuck on another project and traveling for the holidays. I might be able to make tomorrow work around the same time.

kannancet (Tue, 20 Nov 2018 20:01:52 GMT):
@twshelton @haggs - Works for me.We can meet here https://meet.google.com/ekn-zuux-xrh?authuser=1 at 8PM CET tommorow.

aaronr (Tue, 20 Nov 2018 21:57:34 GMT):
Sorry to bug you guys, but I'm getting conflicting information. It looks like, to my untrained eye, that revocation currently exists in Indy and as far as I can tell, is completely implemented. But I've also heard that Indy doesn't have revocation, yet. Am I missing something?

aaronr (Tue, 20 Nov 2018 21:58:31 GMT):
completely implmeneted = tests exist for current level of functionality and that succeed

esplinr (Tue, 20 Nov 2018 23:24:00 GMT):
@aaronr it depends on the layer.

esplinr (Tue, 20 Nov 2018 23:24:30 GMT):
Revocation is implemented in Indy Node. Within Indy SDK, LibIndy implements it, and the pull request into LibVCX is almost done. I'm not aware of an agent that has revocation implemented yet.

aaronr (Wed, 21 Nov 2018 04:52:41 GMT):
@esplinr exactly what I was looking for, thanks for the detail!

HiranKumar (Wed, 21 Nov 2018 06:34:58 GMT):
I tried to install node network using indy-node installation.But failed got an error while installing indy-plenum Command "/usr/bin/python3 -u -c "import setuptools, tokenize;__file__='/tmp/pip-build-uv4s4kwz/python-rocksdb/setup.py';f=getattr(tokenize, 'open', open)(__file__);code=f.read().replace('\r\n', '\n');f.close();exec(compile(code, _file_, 'exec'))" install --record /tmp/pip-wxw2rg93-record/install-record.txt --single-version-externally-managed --compile" failed with error code 1 in /tmp/pip-build-uv4s4kwz/python-rocksdb/

HiranKumar (Wed, 21 Nov 2018 07:32:20 GMT):
The following packages have unmet dependencies: indy-node : Depends: indy-plenum (= 1.6.53) but it is not going to be installed Depends: libsodium18 but it is not installable

Ankita1 (Wed, 21 Nov 2018 07:56:55 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=KPAC4fTsc4Lu6yyfj) @swcurran Sure, thank you. I had tried rebuilding the image post which it worked.

pknowles (Wed, 21 Nov 2018 08:40:29 GMT):
I'll be attending *Hyperledger Global Forum* (Basel, Switzerland) in December with my #indy-semantics hat on. If any other community members are planning to attend, let me know.

pknowles (Wed, 21 Nov 2018 08:40:29 GMT):
I'll be attending *Hyperledger Global Forum* (Basel, Switzerland) in December with my #indy-semantics hat on. If any other community members are planning to attend, let me know. https://events.linuxfoundation.org/events/hyperledger-global-forum-2018/

pknowles (Wed, 21 Nov 2018 08:40:29 GMT):
I'll be attending *Hyperledger Global Forum* (December 12th-15th - Basel, Switzerland) with my #indy-semantics hat on. If any other community members are planning to attend, let me know. https://events.linuxfoundation.org/events/hyperledger-global-forum-2018/

mwherman2000 (Wed, 21 Nov 2018 13:25:55 GMT):
The type of architecture model/diagram I'm trying to create for Sovrin/Indy is sometimes called an Architecture Reference Model (ARM). Here's an example of an ARM I created for the Microsoft SharePoint platform: https://hyperonomy.com/2016/06/15/arms-for-metadata-driven-lob-apps-sharepoint-2013sharepoint-2016/

mwherman2000 (Wed, 21 Nov 2018 13:28:21 GMT):
Here's another example ARM for Microsoft Azure Stack ...on-premise version of Microsoft's cloud platform: https://hyperonomy.com/2016/04/29/microsoft-azure-stack-poc-architecture-archimate-model-version-1-0-1-april-29-2016/

mwherman2000 (Wed, 21 Nov 2018 13:28:58 GMT):
Anyone interested in collaborating on this? I will be in Basel from Dec. 6-16.

baconsandwich (Wed, 21 Nov 2018 13:32:20 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=hBs2a9HGS5oZY2BxY) @mgbailey Isn't this a circular argument? My question was why a Trustee can not add nodes, and you answer that he can not answer because he is a Trustee. My main quesiton is why this separation was made. The Trustee has "higher" privileges than a Steward, so why was it decided that he ca not add nodes himself. My assumption that this has to do with separation of duties to avoid concentrating too much power in one entity. But I would like to know what the "official" reason is.

pknowles (Wed, 21 Nov 2018 13:58:21 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=4rN74Qw2FeSNurA3Q) @mwherman2000 Is the Architecture Reference Model that you're proposing specific for Agent interaction, data architecture/semantics or both? The data capture architecture that we are building is platform agnostic whereas the Agent interaction piece is more deeply rooted on the ledger. I'm involved in the data architecture/semantics side of things. I'd suggest that @danielhardman might be a good point of contact for Agent interaction. In any case, I'll be in Basel at the same time as you and would be happy to work on this with you if you think that that would be a worthwhile exercise.

pknowles (Wed, 21 Nov 2018 13:58:21 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=4rN74Qw2FeSNurA3Q) @mwherman2000 Is the Architecture Reference Model that you're proposing specific to Agent interaction, data architecture/semantics or both? The data capture architecture that we are building is platform agnostic whereas the Agent interaction piece is more deeply rooted on the ledger. I'm involved in the data architecture/semantics side of things. I'd suggest that @danielhardman might be a good point of contact for Agent interaction. In any case, I'll be in Basel at the same time as you and would be happy to work on this with you if you think that that would be a worthwhile exercise.

pknowles (Wed, 21 Nov 2018 13:58:21 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=4rN74Qw2FeSNurA3Q) @mwherman2000 Is the Architecture Reference Model that you're proposing specific to Agent interaction, data architecture/semantics or both? The data capture architecture that we are building is platform agnostic whereas the Agent interaction piece is more deeply rooted on the ledger. I'm involved in the data architecture/semantics side of things. I'd suggest that @danielhardman might be a good point of contact for Agent interaction. In any case, I'll be in Basel on those dates and would be happy to work on this with you if you think that that would be a worthwhile exercise.

pknowles (Wed, 21 Nov 2018 13:58:21 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=4rN74Qw2FeSNurA3Q) @mwherman2000 Is the Architecture Reference Model that you're proposing specific to Agent interaction, data architecture/semantics or both? The data capture architecture that we are building is platform agnostic whereas the Agent interaction piece is more deeply rooted on the ledger. I'm involved in the data architecture/semantics side of things. I'd suggest that @danielhardman might be a good point of contact for Agent interaction. In any case, I'll be in Basel on those dates and would be happy to work with you on this if deemed appropriate.

pknowles (Wed, 21 Nov 2018 13:58:21 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=4rN74Qw2FeSNurA3Q) @mwherman2000 Is the Architecture Reference Model that you're proposing specific to Agent interaction, data architecture/semantics or both? The data capture architecture that we are building is platform agnostic whereas the Agent interaction piece is specific to the master branch for reference implementations which is more deeply rooted on the ledger. I'm involved in the data architecture/semantics side of things. I'd suggest that @danielhardman might be a good point of contact for Agent interaction. In any case, I'll be in Basel on those dates and would be happy to work with you on this if deemed appropriate.

pknowles (Wed, 21 Nov 2018 13:58:21 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=4rN74Qw2FeSNurA3Q) @mwherman2000 Is the Architecture Reference Model that you're proposing specific to Agent interaction, data architecture/semantics or both? The data capture architecture that we are building is platform agnostic whereas the Agent interaction piece is specific to the master branch for reference implementations which is more ledger focussed. I'm involved in the data architecture/semantics side of things. I'd suggest that @danielhardman might be a good point of contact for Agent interaction. In any case, I'll be in Basel on those dates and would be happy to work with you on this if deemed appropriate.

pknowles (Wed, 21 Nov 2018 13:58:21 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=4rN74Qw2FeSNurA3Q) @mwherman2000 Is the Architecture Reference Model that you're proposing specific to Agent interaction, data architecture/semantics or both? The data capture architecture that we are building is platform agnostic whereas the Agent interaction piece is specific to the master branch for reference implementations which is more ledger focussed. I'm involved in the data architecture/semantics side of things. I'd suggest that @danielhardman might be a good point of contact for Agent interaction. In any case, I'll be in Basel on those dates and, if deemed appropriate, would be more than happy to work with you on this.

pknowles (Wed, 21 Nov 2018 13:58:21 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=4rN74Qw2FeSNurA3Q) @mwherman2000 Is the Architecture Reference Model that you're proposing specific to Agent interaction, data architecture/semantics or both? The data capture architecture that we are building is platform agnostic whereas the Agent interaction piece is specific to the master branch for reference implementations which is more ledger focussed. I'm involved in the data architecture/semantics side of things. I'd suggest that @danielhardman might be a good point of contact for Agent interaction. In any case, I'll be in Basel on those dates and, if deemed appropriate, would be more than happy to work with you on this from a semantics perspective.

pknowles (Wed, 21 Nov 2018 14:10:14 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=4rN74Qw2FeSNurA3Q) @mwherman2000 Is the Architecture Reference Model that you're proposing specific to Agent interaction, data architecture/semantics or both? The data capture architecture that we are building is platform agnostic whereas the Agent interaction piece is more ledger specific. I'm involved in the data architecture/semantics side of things. I'd suggest that @danielhardman might be a good point of contact for Agent interaction. In any case, I'll be in Basel on those dates and would be happy to work with you on this if deemed appropriate.

mwherman2000 (Wed, 21 Nov 2018 15:15:30 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=3SAMWg2Wui4XyMdfc) @pknowles Both @pknowles ...as comprehensive as possible. I think these types of assets are incredibly important - particularly accelerating adoption by large enterprise organizations (which I assume must be a primary audience for Indy/Sovrin).

mwherman2000 (Wed, 21 Nov 2018 15:15:30 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=3SAMWg2Wui4XyMdfc) @pknowles Both ...as comprehensive as possible. I think these types of assets are incredibly important - particularly accelerating adoption by large enterprise organizations (which I assume must be a primary audience for Indy/Sovrin).

mwherman2000 (Wed, 21 Nov 2018 15:15:30 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=3SAMWg2Wui4XyMdfc) @pknowles Both ...as comprehensive as possible. I think these types of assets are incredibly important - particularly wrt accelerating adoption by large enterprise organizations (which I assume must be a primary audience for Indy/Sovrin).

mwherman2000 (Wed, 21 Nov 2018 15:15:30 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=3SAMWg2Wui4XyMdfc) @pknowles Both ...as comprehensive as possible. I think these types of assets are incredibly important - particularly wrt accelerating adoption by large enterprise organizations (which I assume must be a primary audience for Indy/Sovrin). In the SharePoint ARM example, I captured the logical data model side-by-side against the software architecture.

mwherman2000 (Wed, 21 Nov 2018 15:15:30 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=3SAMWg2Wui4XyMdfc) @pknowles Both ...as comprehensive as possible. I think these types of assets are incredibly important - particularly wrt accelerating adoption by large enterprise organizations (which I assume must be a primary audience for Indy/Sovrin). In the SharePoint ARM example, I captured the high-level data model side-by-side against the software architecture. Let's book some time for when we're in Basel.

mwherman2000 (Wed, 21 Nov 2018 15:15:30 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=3SAMWg2Wui4XyMdfc) @pknowles Both ...as comprehensive as possible. I think these types of assets are incredibly important - particularly wrt accelerating adoption by large enterprise organizations (which I assume must be a primary audience for Indy/Sovrin). In the SharePoint ARM example, I captured the high-level data model side-by-side against the software architecture. Let's book some time for when we're in Basel along with @danielhardman and anyone else that is interested ...an impromptu post-con workshop.

mgbailey (Wed, 21 Nov 2018 15:22:47 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=rocZA6ASLwfh3TJGz) @baconsandwich Think of it this way. A Trustee is by nature an individual, and exercises his role directly. A steward is an organization, and has a team dedicated to maintaining a validator. The governance framework requires this organization / team orientation for maintaining a validator, for good reason.

mwherman2000 (Wed, 21 Nov 2018 15:23:27 GMT):
Here

mwherman2000 (Wed, 21 Nov 2018 15:23:27 GMT):
Here's a link to the Archi EA modeling tool I use (free and open source - Phil loves to receive donations): https://www.archimatetool.com/download/

mwherman2000 (Wed, 21 Nov 2018 15:27:23 GMT):
FYHumor (from the Enterprise Architecture space): "A fool with a tool is still a fool" ;-)

mwherman2000 (Wed, 21 Nov 2018 15:29:41 GMT):
[We can also capture at the Business Architecture level some of the high-level business processes and roles ...e.g. the Trustee discussion that is going on simultaneously with this thread.]

baconsandwich (Wed, 21 Nov 2018 15:31:54 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=3Rv5wZ2YtFtyzF5ZQ) @mgbailey Hm, I don't see a fundamental difference. I mean, I understand what you are saying, but I don't see *why* this has to be this way. Why can't the Trustee not be an organization too? And the Steward can add other Trust Anchors, so this is something that is a concrete action and will be performed by a Human probably.....So I guess there was some Idea behind how these roles were created. And this is still uncreal to me. :confused:

mwherman2000 (Wed, 21 Nov 2018 17:33:13 GMT):

57219119-read-the-fine-print-3d-rendering-a-red-wax-seal.jpg

mwherman2000 (Wed, 21 Nov 2018 17:33:13 GMT):

57219119-read-the-fine-print-3d-rendering-a-red-wax-seal.jpg

mwherman2000 (Wed, 21 Nov 2018 17:33:13 GMT):

57219119-read-the-fine-print-3d-rendering-a-red-wax-seal.jpg

pknowles (Wed, 21 Nov 2018 18:01:34 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=9oNfwXfPgLvxtw2ef) @mwherman2000 There are obviously separate pieces in this space with specialised experts for each part. It'll be a case of building the model piece by piece and then dovetailing it all together to complete the tapestry. It'll take some time to complete but certainly sounds like a useful exercise.

mwherman2000 (Wed, 21 Nov 2018 20:04:35 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=ZmBmusxaGP4BGHpuZ) @pknowles Here's a strawman ...2 diagrams initially ...one ARM specifically for Indy/Sovrin...

mwherman2000 (Wed, 21 Nov 2018 20:05:23 GMT):

HyperLedger Indy-Sovrin ARM - Strawman v0.1 - 2018-11-21-Annotated.png

mwherman2000 (Wed, 21 Nov 2018 20:06:04 GMT):
This can be expanded upon/detailed - part-by-part as you mentioned in your msg.

mwherman2000 (Wed, 21 Nov 2018 20:06:04 GMT):
This can be expanded upon/detailed - part-by-part as you mentioned in your msg. It's an important starting point for me as I will attempt to detail it with each new Indy/Sovrin concept I encounter. e.g. reading through the Glossary.

mwherman2000 (Wed, 21 Nov 2018 20:09:24 GMT):
Here's a second ARM example where I'm working to visualize how an SSI platform (Indy/Sovrin) can be tightly integrated with a more general-purpose blockchain platform (e.g. the Stratis Platform that underpins my Hyperonomy Business Blockchain project)...

mwherman2000 (Wed, 21 Nov 2018 20:09:24 GMT):
Here's a second conceptual architecture example where I'm working to visualize how an SSI platform (Indy/Sovrin) can be tightly integrated with a more general-purpose blockchain platform (e.g. the Stratis Platform that underpins my Hyperonomy Business Blockchain project)...coince

mwherman2000 (Wed, 21 Nov 2018 20:09:24 GMT):
Here's a second conceptual architecture example where I'm working to visualize how an SSI platform (Indy/Sovrin) can be tightly integrated with a more general-purpose blockchain platform (e.g. the Stratis Platform that underpins my Hyperonomy Business Blockchain project)...

mwherman2000 (Wed, 21 Nov 2018 20:10:06 GMT):

HBB Conceptual Architecture v0.3-Annotated.png

darrell.odonnell (Wed, 21 Nov 2018 20:41:49 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=mtLjpv9AP8BWoKda7) @mwherman2000 @nage I strongly recommend you and @mwherman2000 touch base both at Basel and in advance. He and I have some history and he can help shape a solid future here...

rjones (Wed, 21 Nov 2018 20:42:11 GMT):
Has left the channel.

peacekeeper (Thu, 22 Nov 2018 06:55:44 GMT):
TLS certificate for https://repo.sovrin.org/ expired a few hours ago.

peacekeeper (Thu, 22 Nov 2018 07:58:51 GMT):
Okay looks like it just got renewed.

peacekeeper (Thu, 22 Nov 2018 07:58:51 GMT):
Okay looks like it just got renewed. Thanks.

KindleBitSoln (Thu, 22 Nov 2018 09:57:29 GMT):
Has joined the channel.

KindleBitSoln (Thu, 22 Nov 2018 10:00:58 GMT):
Hi Everyone Can you please tell me how to publish an identity into sovrin ledger containing a DNA data. I am new in this platform so you help would be appreciable for me?

KindleBitSoln (Thu, 22 Nov 2018 10:08:37 GMT):
Is there official tutorial for INDY ?

aristide (Thu, 22 Nov 2018 11:47:40 GMT):
Has joined the channel.

aristide (Thu, 22 Nov 2018 11:58:16 GMT):
Hello guys, is there some kind of "public" or "general purpose" indy network that one can attach his node to? Or the way is deploying a new indy network for every use case? Thanks

carsten (Fri, 23 Nov 2018 03:58:57 GMT):
Has joined the channel.

manner (Fri, 23 Nov 2018 09:57:55 GMT):
Has joined the channel.

Derashe (Fri, 23 Nov 2018 09:59:51 GMT):
Hello everybody. We have finally removed old client code, which was merged with our plenum & node codebases. For now the only way to connect to ledger and send transactions is Indy SDK (https://github.com/hyperledger/indy-sdk). Old getting started guide is deprecated too. You can find new one at https://github.com/hyperledger/indy-sdk/blob/master/doc/getting-started/getting-started.md

kdenhartog (Sun, 25 Nov 2018 22:52:18 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=uuzTNSjHAgsYKWMaG) @aristide Since Indy is a public-permissioned network, you would need permission to join a network. Once observer nodes are completed (lower priority on roadmap for core dev team), then it may be possible depending on if observer nodes are given permissionless or permissioned access to the network. My best guess is that they will be permissioned as well to prevent DDoS attacks on the validator pool of nodes, but until someone writes the code it's just a design.

kdenhartog (Sun, 25 Nov 2018 22:59:01 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=dzXqrmKEKWFfGtJbo) @KindleBitSoln This is possible, but it requires advanced knowledge in the architecture and the IndySDK codebase. Have you completed the getting started guide yet and gotten a better than good understanding of the reference agent implementions yet?

KindleBitSoln (Mon, 26 Nov 2018 05:42:09 GMT):
@kdenhartog : I am new in this. I need to know is it possible so that i can proceed with indy

louisk (Mon, 26 Nov 2018 12:44:00 GMT):
@KindleBitSoln - have a read through the getting started guide, paying close attention to schemas and credentials. you can find a lot more detail about these concepts in the W3C community guide on verifiable claims - https://github.com/w3c/verifiable-claims

louisk (Mon, 26 Nov 2018 12:44:52 GMT):
can't say exactly what's possible and not for you, only you can decide if these constructs can be applied to your idea

janl (Mon, 26 Nov 2018 15:28:18 GMT):
I have done a major update to my Consent Receipt HIPE pull request (https://github.com/hyperledger/indy-hipe/pull/55). Now it includes a more detailed sequence diagram and sample schema/overlays to help explain the higher level diagrams. (@danielhardman, @swcurran, @sergey.minaev)

lwan2000 (Mon, 26 Nov 2018 17:11:19 GMT):
Has joined the channel.

infominer33 (Tue, 27 Nov 2018 06:02:07 GMT):
Has joined the channel.

JoanArcas (Tue, 27 Nov 2018 11:08:43 GMT):
Has joined the channel.

mwherman2000 (Tue, 27 Nov 2018 17:02:19 GMT):
In an online commerce scenario, if Alice working for Acme Org acting as a Customer sends an electronic Purchase Order (a Sovrin entity) that is received by Bob working for Bobbles Org acting as a Supplier... who are the most likely actors responsible for the Holder, Inspector, Issuer, and Verifier roles? ...feel free to speculate if this is a new(er) use case. Thx

mwherman2000 (Tue, 27 Nov 2018 17:06:31 GMT):

TODPycw.png

infominer33 (Wed, 28 Nov 2018 01:54:41 GMT):
to be clear, the indy software runs the sovrin ledger... but theoretically it could be used to create an independent network... basically trying to understand how to talk \write about it

infominer33 (Wed, 28 Nov 2018 01:55:54 GMT):
do I say "indy creates a public ledger" or can I write: "since indy is a public ledger" or is it just as well to refer to the code as indy and the ledger\network as sovrin?

kdenhartog (Wed, 28 Nov 2018 02:05:13 GMT):
it's fine to refer to the code as indy, and the network+governance as sovrin

infominer33 (Wed, 28 Nov 2018 03:00:26 GMT):
thx. one more question:

infominer33 (Wed, 28 Nov 2018 03:00:44 GMT):
did document for a public DID is stored on or off ledger?

infominer33 (Wed, 28 Nov 2018 03:01:11 GMT):
assuming it's off... but assuming is rough work

kdenhartog (Wed, 28 Nov 2018 03:07:50 GMT):
globally public DIDs (usually used for issuing credentials) are stored on ledger. private DIDs (usually used for encrypting communications). Here's a link to slides that outline this well. https://docs.google.com/presentation/d/1UnC_nfOUK40WS5TD_EhyDuFe5cStX-u0Z7wjoae_PqQ/

infominer33 (Wed, 28 Nov 2018 03:30:16 GMT):
that is a useful document, @kdenhartog What I'm trying to distinguish is the DID from the DID Doc\JSON File

infominer33 (Wed, 28 Nov 2018 03:30:24 GMT):
where does the did doc live?

infominer33 (Wed, 28 Nov 2018 03:30:52 GMT):
for private, this is with some agent I'm supposing...

infominer33 (Wed, 28 Nov 2018 05:19:53 GMT):
ok, moving on.. cause it doesn't really matter to me where public DID docs are stored I'm just a nit-picker

infominer33 (Wed, 28 Nov 2018 05:20:33 GMT):
aaand I just answered my own next question before I finished typing it

drummondreed (Wed, 28 Nov 2018 05:46:05 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=3Wt8XookBnzKms2Kt) Michael, first, see Figure 1 of https://w3c.github.io/vc-data-model/ for the current terms the W3C Verifiable Claims Working Group is using for these roles: issuer, holder, verifier, and identifier registry (the final term which is still being debated as its role is broader than just a registry for identifiers).

drummondreed (Wed, 28 Nov 2018 05:54:39 GMT):
Second, in your scenario, you are introducing the idea that a verifiable credential (a Purchase Order) can itself be a Sovrin Entity. That's a cool concept that I haven't seen modeled much yet. However I don't think it changes the basic roles. Acme Org is the Issuer because the P.O. is from Acme Org, not from Alice. Alice, acting as a Delegate of the Issuer, triggers the issuance of the P.O. as a VC. Since the P.O. is to Bobbles Org, I believe Bobbles.Org is both the Holder and the Verifier. Bob is acting as a Delegate of Bobbles Org just like Alice is acting as a Delegate of Acme.Org.

drummondreed (Wed, 28 Nov 2018 06:00:17 GMT):
When you map out the agent-to-agent (A2A) flows, they will involve four agents: two enterprise agents (Acme Org and Bobbles Org) and two personal agents (Alice and Bob). Alice's agent has a connection with the Acme Org agent and has been issued a Delegate VC. Bob's agent has a connection with Bobbles Org and has been issued a Delegate VC. Alice uses her Delegate VC to triggers the issuance of the P.O. as a VC sent from the Acme Org agent to the Bobbles Org agent who stores it as the Holder. The Bobbles Org agent then verifies the VC as the Verifier and notifies Bob's agent of the result.

swcurran (Wed, 28 Nov 2018 06:04:37 GMT):
@drummondreed - very cool to see you write that out so clearly. The interplay between delegates and organizations was one of our first conundrums in this space (way back in 2017 - ancient history). That description is basically where we got as well. Nice!

drummondreed (Wed, 28 Nov 2018 06:22:37 GMT):
@swcurran Thanks. Yes, I remember it well. I've become a massive fan of delegates (and delegate VCs) as I think they are the ones that do the vast majority of authorizing work and making actual decisions in the A2A world. Very elegant.

infominer33 (Wed, 28 Nov 2018 06:23:11 GMT):
do pairwise DIDs have any data whatsoever on the public chain? such as a public key? or.. what I think I'm understanding is that only public parties store data on the ledger that is used to verify and support off-chain relationships with private dids?

drummondreed (Wed, 28 Nov 2018 06:29:10 GMT):
@infominer33 Your understanding is correct if you are talking about a *Sovrin* A2A network architecture. I say that because other deployments of Hyperledger Indy agents might choose to have a different set of dependencies and services from an Indy ledger network. But with Sovrin Privacy by Design architecture, a private pairwise pseudonymous DID-to-DID connection has no direct dependency on the Sovrin public ledger. Thus, as you say, the main purpose of the Sovrin public ledger is for public DIDs that issuers use for discovery and to sign VCs that need to be publicly verifiable.

drummondreed (Wed, 28 Nov 2018 06:31:56 GMT):
Note, however, that there are two designed-but-not-yet implemented-and-deployed features of off-ledger pairwise pseudonymous DID-to-DID connections that *do* use the Sovrin public ledger. The first one is called dead drops—used to reconnect two agents who happen to lose their connection (e.g., both moved agencies at the same time and could not update each other with new service endpoints). The second one is called Agent Authorization Policies, and that's a security feature of DKMS that's explained further in the DKMS Design and Architecture doc: http://bit.ly/dkmsv3

infominer33 (Wed, 28 Nov 2018 06:32:55 GMT):
thank you for sharing the finer points!

infominer33 (Wed, 28 Nov 2018 06:33:49 GMT):
tbh, it took me quite a few reads of "what goes on the ledger" to finally come to the conclusion that the pairwise did probably doesn't even have a signature or anything on the chain

drummondreed (Wed, 28 Nov 2018 06:34:23 GMT):
Exactly. We should strive to make that clearer in the next version of the paper.

infominer33 (Wed, 28 Nov 2018 06:35:40 GMT):
cause the wp is about what goes on the ledger.. but spends a bunch of time about DIDs while not being super explicit about which part if any is on the chain for private

infominer33 (Wed, 28 Nov 2018 06:35:52 GMT):
anyway.. good stuff, I'll be back for more :)

mwherman2000 (Wed, 28 Nov 2018 13:58:14 GMT):
Thank you @drummondreed ... extending the use case ...Alice works in the Central Purchasing department of Acme.org. ...Bob works in the Order Desk department of Bobbles.org. The Purchase Order (aka PO VC) is an aggravated NFE (Non-Fungibe Entity) - a collection of Facts that are references to other externally separate entities: Purchase Order Number, Customer Contact (e.g. Alice), Customer Address, Supplier Contact (e.g. Bob), Supplier Address, Ship To Contact (Charlie who works in the Receiving department also at Acme.org), Ship To Address (e

mwherman2000 (Wed, 28 Nov 2018 13:58:14 GMT):
Thank you @drummondreed ... extending the use case ...Alice works in the Central Purchasing department of Acme.org. ...Bob works in the Order Desk department of Bobbles.org. The Purchase Order (aka PO VC) is an aggravated NFE (Non-Fungibe Entity) - a collection of Facts that are references to other externally separate entities: Purchase Order Number, Customer Contact (e.g. Alice), Customer Address, Supplier Contact (e.g. Bob), Supplier Address, Ship To Contact (Charlie who works in the Receiving department also at Acme.org), Ship To Address (e.g. an Acme warehouse location), a sequence of Line Items detailing Part Number, Description, Quantity, Unit Price, Discount Amount, Extended Price for each Line Item. Attached to the Purchase Order but separate from it are a Shipping Instructions document. Call the two of these (PO and Shipping Instructions) a Purchase Order Bundle.

mwherman2000 (Wed, 28 Nov 2018 13:58:14 GMT):
Thank you @drummondreed ... extending the use case ...Alice works in the Central Purchasing department of Acme.org. ...Bob works in the Order Desk department of Bobbles.org. The Purchase Order (aka PO VC) is an aggravated NFE (Non-Fungibe Entity) - a collection of Facts that are references to other externally separate entities: Purchase Order Number, Customer Contact (e.g. Alice), Customer Address, Supplier Contact (e.g. Bob), Supplier Address, Ship To Contact (Charlie who works in the Receiving department also at Acme.org), Ship To Address (e.g. an Acme warehouse location), a sequence of Line Items detailing Part Number, Description, Quantity, Unit Price, Discount Amount, Extended Price for each Line Item. Attached to the Purchase Order but separate from it is a Shipping Instructions document. Call the two of these (PO and Shipping Instructions) a Purchase Order Bundle.

mwherman2000 (Wed, 28 Nov 2018 13:58:14 GMT):
Thank you @drummondreed ... extending the use case ...Alice works in the Central Purchasing department of Acme.org. ...Bob works in the Order Desk department of Bobbles.org. The Purchase Order (aka PO VC) is an aggravated NFE (Non-Fungibe Entity) - a collection of Facts that are references to other externally separate entities: Purchase Order Number, Customer Contact (e.g. Alice), Customer Address, Supplier Contact (e.g. Bob), Supplier Address, Ship To Contact (Charlie who works in the Receiving department also at Acme.org), Ship To Address (e.g. an Acme.org warehouse location), a sequence of Line Items detailing Part Number, Description, Quantity, Unit Price, Discount Amount, Extended Price for each Line Item. Attached to the Purchase Order but separate from it is a Shipping Instructions document. Call the two of these (PO and Shipping Instructions) a Purchase Order Bundle.

mwherman2000 (Wed, 28 Nov 2018 13:58:14 GMT):
Thank you @drummondreed Purchase Order Extended Use Case: Alice works in the Central Purchasing department of Acme.org. ...Bob works in the Order Desk department of Bobbles.org. The Purchase Order (aka PO VC) is an aggravated NFE (Non-Fungibe Entity) - a collection of Facts that are references to other externally separate entities: Purchase Order Number, Customer Contact (e.g. Alice), Customer Address, Supplier Contact (e.g. Bob), Supplier Address, Ship To Contact (Charlie who works in the Receiving department also at Acme.org), Ship To Address (e.g. an Acme.org warehouse location), a sequence of Line Items detailing Part Number, Description, Quantity, Unit Price, Discount Amount, Extended Price for each Line Item. Attached to the Purchase Order but separate from it is a Shipping Instructions document. Call the two of these (PO and Shipping Instructions) a Purchase Order Bundle.

mwherman2000 (Wed, 28 Nov 2018 13:58:14 GMT):
Thank you @drummondreed Purchase Order Extended Use Case: Alice works in the Central Purchasing department of Acme.org. ...Bob works in the Order Desk department of Bobbles.org. The Purchase Order (aka PO VC) is an aggregated NFE (Non-Fungibe Entity) - a collection of Facts that are references to other externally separate entities: Purchase Order Number, Customer Contact (e.g. Alice), Customer Address, Supplier Contact (e.g. Bob), Supplier Address, Ship To Contact (Charlie who works in the Receiving department also at Acme.org), Ship To Address (e.g. an Acme.org warehouse location), a sequence of Line Items detailing Part Number, Description, Quantity, Unit Price, Discount Amount, Extended Price for each Line Item. Attached to the Purchase Order but separate from it is a Shipping Instructions document. Call the two of these (PO and Shipping Instructions) a Purchase Order Bundle.

mwherman2000 (Wed, 28 Nov 2018 14:08:58 GMT):
I presume each of the Facts in my use case are also Verfiable Claims? ...I acknowledge I've presented a lengthy list but I pose the same question for each Fact: who is the likely Issuer, Verifier, Holder, Inspector, etc. for each of the listed Facts?

pknowles (Wed, 28 Nov 2018 14:53:51 GMT):
@mwherman2000 In your example, at the time of initial transaction, the person issuing the purchase order would be the "Issuer" (that purchase order would be issued in the form of an overlaid schema), the customer contact would be the "Holder" (their personal data might be held in a data wallet on their mobile phone) and the supplier would be the "Verifier".

pknowles (Wed, 28 Nov 2018 14:53:51 GMT):
@mwherman2000 In your example, at the time of initial transaction, the person issuing the purchase order would be the "Issuer" (that purchase order would be issued in the form of an overlaid schema), the customer would be the "Holder" (their personal data might be held in a data wallet on their mobile phone) and the supplier would be the "Verifier".

pknowles (Wed, 28 Nov 2018 14:53:51 GMT):
@mwherman2000 In your example, at the time of initial transaction, the actor issuing the purchase order would be the "Issuer" (that purchase order would be issued in the form of an overlaid schema), the customer would be the "Holder" (their personal data might be held in a data wallet on their mobile phone) and the supplier would be the "Verifier".

pknowles (Wed, 28 Nov 2018 15:03:19 GMT):
@mwherman2000 At the time of shipping, the actors would switch roles. The supplier would be the "Issuer" (that shipping order would be issued in the form of an overlaid schema), the person who issued the original purchase order would be the "Holder" and the customer contact would be the "Verifier".

pknowles (Wed, 28 Nov 2018 15:03:19 GMT):
@mwherman2000 At the time of shipping, the actors would switch roles. The supplier would be the "Issuer" (that shipping order would be issued in the form of an overlaid schema), the person who issued the original purchase order would become the "Holder" and the customer contact would be the "Verifier".

pknowles (Wed, 28 Nov 2018 15:03:19 GMT):
@mwherman2000 At the time of shipping, the actors would switch roles. The supplier would become the "Issuer" (that shipping order would be issued in the form of an overlaid schema), the person who issued the original purchase order would become the "Holder" and the customer contact would be the "Verifier".

pknowles (Wed, 28 Nov 2018 15:03:19 GMT):
@mwherman2000 At the time of shipping, the actors would switch roles. The supplier would become the "Issuer" (that shipping order would be issued in the form of an overlaid schema), the person who issued the original purchase order would become the "Holder" and the customer contact would become the "Verifier".

pknowles (Wed, 28 Nov 2018 15:03:19 GMT):
@mwherman2000 At the time of shipping, the actors would switch roles. The supplier would become the "Issuer" (that shipping order would be issued in the form of an overlaid schema), the person who issued the original purchase order would become the "Holder" and the customer would become the "Verifier".

pknowles (Wed, 28 Nov 2018 15:03:19 GMT):
@mwherman2000 At the time of shipping, the actors would switch roles. The supplier would become the "Issuer" (that shipping order would be issued in the form of an overlaid schema), the actor who issued the original purchase order would become the "Holder" and the customer would become the "Verifier".

changeagent (Wed, 28 Nov 2018 19:39:04 GMT):
Has joined the channel.

changeagent (Wed, 28 Nov 2018 19:43:35 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=eiYx3YSAFb5Le2N44) @infominer33 @infominer33 have you looked at how DIDs relate to "On Boarding" as described in the getting started guide? I am trying to understand more about how the ledger helps with verification

mwherman2000 (Wed, 28 Nov 2018 21:50:45 GMT):
@pknowles Is Alice the Issuer of the PO? ...or is Acme.org the Issuer of the PO (per @drummondreed's comments above)?

mwherman2000 (Wed, 28 Nov 2018 21:51:11 GMT):

HBB-Gumball Protocol-Indy-Sovrin-Mapping v0.4.png

mwherman2000 (Wed, 28 Nov 2018 21:52:02 GMT):
Here's a sketch to help illustrate the Purchase Order Extended Use Case (see above)

mwherman2000 (Wed, 28 Nov 2018 21:52:02 GMT):
Here's a sketch to help illustrate the Purchase Order Extended Use Case (see below)

mwherman2000 (Wed, 28 Nov 2018 22:07:22 GMT):

HBB-Gumball Protocol-Indy-Sovrin-Mapping v0.4.png

mwherman2000 (Wed, 28 Nov 2018 22:32:50 GMT):

HBB-Gumball Protocol-Indy-Sovrin-Mapping v0.4.png

pknowles (Wed, 28 Nov 2018 22:51:07 GMT):
@mwherman2000 Acme.org would be the "Issuer" of the purchase order. I'll take a more detailed at your schematic diagram in more detail tomorrow.

pknowles (Wed, 28 Nov 2018 22:51:07 GMT):
@mwherman2000 Acme.org would be the "Issuer" of the purchase order. I'll take a more detailed look at your schematic tomorrow.

pknowles (Wed, 28 Nov 2018 22:51:07 GMT):
@mwherman2000 The "Issuer" is the actor that creates and issues the data capture request. In this case, Acme.org would appear to be the "Issuer" of the purchase order. (I'll study your diagram in more detail in the morning with fresh senses!)

pknowles (Wed, 28 Nov 2018 22:51:07 GMT):
@mwherman2000 The "Issuer" is the actor that creates and issues the data capture request. In this case, Acme.org would appear to be the "Issuer" of the purchase order. (I'll study your diagram in more detail in the morning with fresher senses!)

pknowles (Wed, 28 Nov 2018 22:51:07 GMT):
@mwherman2000 The "Issuer" is the actor that creates and issues the data capture request. In this case, Acme.org would appear to be the "Issuer" of the purchase order. (I'll study your diagram in more detail in the morning with fresher senses to make sure I've understood the data flow!)

pknowles (Wed, 28 Nov 2018 22:51:07 GMT):
@mwherman2000 The "Issuer" is the actor that creates and issues the data capture request. In this case, Acme.org would appear to be the "Issuer" of the purchase order. (I'll study your diagram in more detail in the morning with fresher senses to make sure I've understood the data flow properly!)

pknowles (Wed, 28 Nov 2018 22:51:07 GMT):
@mwherman2000 The "Issuer" is the actor that creates and issues the data capture request. In this case, Acme.org would appear to be the "Issuer" of the purchase order. (I'll study your diagram in more detail in the morning with fresher senses to make sure I've understood it properly!)

mwherman2000 (Thu, 29 Nov 2018 03:11:29 GMT):
@pknowles FYI: The diagram depicts aggregation, composition, and realization relationship (primarily) - not data flows.

mwherman2000 (Thu, 29 Nov 2018 03:17:25 GMT):
@pknowles When you look at the aggregation relationships in middle blue Logical Data Model layers (the open diamond shapes), st this level of detail, I think there will/may be different Issuers for each Fact e.g. the Issuer of the next (sequential) PO No. will not be the same as the Issuer of the Customer Address or the Shipping Instructions or the Line Items ...potentially for (sub) Issuers in addition to the Issue of the aggregrated Purchase Order or Purchase Order Bundle NFE.

mwherman2000 (Thu, 29 Nov 2018 03:17:25 GMT):
@pknowles When you look at the aggregation relationships in the middle blue Logical Data Model layer (the open diamond shapes), at this level of detail, I think there will/may be different Issuers for each Fact - for example, the Issuer of the next sequential PO No. will not be the same as the Issuer of the Customer Address or Issuer of the Shipping Instructions or Issuer of the Line Items ...in addition to these (sub) Issuers, there is the Issuer of the aggregated Purchase Order or Purchase Order Bundle NFE.

mwherman2000 (Thu, 29 Nov 2018 03:23:34 GMT):
@pknowles Some of these sub Issuers may be specific departments within Acme Org (e.g. Central Purchasing, Receiving, etc.). Issuers of the Address NFEs may be the National Post Service or similar government service or a contractor operating on their behalf. The latter may be Issuers and/or Verifiers?

mwherman2000 (Thu, 29 Nov 2018 03:23:34 GMT):
@pknowles Some of these sub Issuers may be specific departments within Acme Org (e.g. Central Purchasing, Receiving, etc.). Issuers of the various Address NFEs may be the National Post Service or similar government service or a contractor operating on their behalf. The latter may be Issuers and/or Verifiers?

mwherman2000 (Thu, 29 Nov 2018 04:08:14 GMT):

HBB-Gumball Protocol-Indy-Sovrin-Mapping v0.4-Annotated.png

infominer33 (Thu, 29 Nov 2018 04:38:40 GMT):
@changeagent basically, as I understand it. The ledger is used to validate organizations, who then validate user credentials(off-ledger)... and those user credentials can be verified as 'legit' by the ledger. So the user, or organizations the user wants to transact with can verify the credential by checking the ledger. does that answer your question?

infominer33 (Thu, 29 Nov 2018 04:38:40 GMT):
@changeagent basically, as I understand it. The ledger is used to validate organizations, who then validate user credentials(off-ledger)... and those user credentials can be verified as 'legit' by the ledger. So the user, or organizations the user wants to transact with can verify eachothers credentials by checking the ledger. does that answer your question?

infominer33 (Thu, 29 Nov 2018 04:38:40 GMT):
@changeagent basically, as I understand it. The ledger is used to validate organizations, who then validate user credentials(off-ledger)... and those user credentials can be verified as 'legit' by the ledger. So the user, or organizations the user wants to transact with can verify eachothers credentials by checking the ledger. does that answer your question? (i'm still somewhat of a newb)

infominer33 (Thu, 29 Nov 2018 04:45:10 GMT):
and now I have a question: Its difficult to discern exactly the line between sovrin and indy. When describing Indy, it's easier to default to describing sovrin... I'm just confused, and now re-reading the edx course to figure out how to describe it... as I'm writing about Indy, atm... but how to do that without mostly talking about it as used in sovrin

infominer33 (Thu, 29 Nov 2018 04:48:58 GMT):

Clipboard - November 28, 2018 11:48 PM

infominer33 (Thu, 29 Nov 2018 04:49:03 GMT):
@changeagent maybe this picture helps

infominer33 (Thu, 29 Nov 2018 04:51:09 GMT):
but back to my question... it would be superduper if there was a document that clearly communicated the distinctions of how indy could be used differently than the way sovrin uses it

pknowles (Thu, 29 Nov 2018 06:46:12 GMT):
@mwherman2000 Schema attributes (Address, etc.) can obviously be aggregated from various sources. In the case of aggregated data capture, the creator and subsequent issuer of the aggregated schema would be the "Issuer". Not sure if that helps or not. Perhaps we go through the diagram together in person at the Hyperledger Global Forum in Basel. That'll be a nice way to introduce me to ARM design. Shall we earmark December 14th for a deep dive on this?

pknowles (Thu, 29 Nov 2018 06:46:12 GMT):
@mwherman2000 Schema attribute names (Address, etc.) can obviously be aggregated from various sources. In the case of aggregated data capture, the creator and subsequent issuer of the aggregated schema would be the "Issuer". Not sure if that helps or not. Perhaps we go through the diagram together in person at the Hyperledger Global Forum in Basel. That'll be a nice way to introduce me to ARM design. Shall we earmark December 14th for a deep dive on this?

pknowles (Thu, 29 Nov 2018 06:46:12 GMT):
@mwherman2000 Schema attribute names (Address, etc.) can obviously be aggregated from various sources. In the case of aggregated data capture, the creator and subsequent issuer of the aggregated schema would be the "Issuer". Not sure if that helps or not. Perhaps we can go through the diagram together in person at the Hyperledger Global Forum in Basel. That'll be a nice way to introduce me to ARM design. Shall we earmark December 14th for a deep dive on this?

pknowles (Thu, 29 Nov 2018 06:46:12 GMT):
@mwherman2000 Schema attribute names (Customer Address, Supplier Address, etc.) can obviously be aggregated from various sources. In the case of aggregated data capture, the creator and subsequent issuer of the aggregated schema would be the "Issuer". Not sure if that helps or not. Perhaps we can go through the diagram together in person at the Hyperledger Global Forum in Basel. That'll be a nice way to introduce me to ARM design. Shall we earmark December 14th for a deep dive on this?

pknowles (Thu, 29 Nov 2018 06:46:12 GMT):
@mwherman2000 Schema attribute names (Customer Address, Supplier Address, etc.) can obviously be aggregated from various sources. In the case of aggregated data capture, the creator and subsequent issuer of the aggregated schema would be the "Issuer". Not sure if that helps or not. I've tweaked some of my previous responses to add clarity. Perhaps we can go through the diagram together in person at the Hyperledger Global Forum in Basel. That'll be a nice way to introduce me to ARM design. Shall we earmark December 14th for a deep dive on this?

pknowles (Thu, 29 Nov 2018 06:46:12 GMT):
@mwherman2000 Schema attribute names (Customer Address, Supplier Address, etc.) can obviously be aggregated from various sources. In the case of aggregated data capture, the creator and subsequent issuer of the aggregated schema would be the "Issuer". Not sure if that helps or not. I've tweaked some of my previous responses to add clarity. Perhaps you, @swcurran and I can go through the diagram together in person at the Hyperledger Global Forum in Basel. That'll be a nice way to introduce me to ARM design. Shall we earmark December 14th for a deep dive on this?

pknowles (Thu, 29 Nov 2018 06:46:12 GMT):
@mwherman2000 Schema attribute names and/or credentials (Customer Address, Supplier Address, etc.) can obviously be aggregated from various sources. In the case of aggregated data capture, the creator and subsequent issuer of the aggregated schema would be the "Issuer". Not sure if that helps or not. I've tweaked some of my previous responses to add clarity. Perhaps you, @swcurran and I can go through the diagram together in person at the Hyperledger Global Forum in Basel. That'll be a nice way to introduce me to ARM design. Shall we earmark December 14th for a deep dive on this?

pknowles (Thu, 29 Nov 2018 06:46:12 GMT):
@mwherman2000 Schema attributes and/or Credentials (Customer Address, Supplier Address, etc.) can obviously be aggregated from various sources. In the case of aggregated data capture, the creator and subsequent issuer of the aggregated schema would be the "Issuer". Not sure if that helps or not. I've tweaked some of my previous responses to add clarity. Perhaps you, @swcurran and I can go through the diagram together in person at the Hyperledger Global Forum in Basel. That'll be a nice way to introduce me to ARM design. Shall we earmark December 14th for a deep dive on this?

pknowles (Thu, 29 Nov 2018 06:46:12 GMT):
@mwherman2000 Attributes and/or Credentials (Customer Address, Supplier Address, etc.) can obviously be aggregated from various sources. In the case of aggregated data capture, the creator and subsequent issuer of the aggregated schema would be the "Issuer". Not sure if that helps or not. I've tweaked some of my previous responses to add clarity. Perhaps you, @swcurran and I can go through the diagram together in person at the Hyperledger Global Forum in Basel. That'll be a nice way to introduce me to ARM design. Shall we earmark December 14th for a deep dive on this?

pknowles (Thu, 29 Nov 2018 06:46:12 GMT):
@mwherman2000 Schema attributes and Credentials (Customer Address, Supplier Address, etc.) can obviously be aggregated from various sources. In the case of aggregated data capture, the creator and subsequent issuer of the aggregated schema would be the "Issuer". Not sure if that helps or not. I've tweaked some of my previous responses to add clarity. Perhaps you, @swcurran and I can go through the diagram together in person at the Hyperledger Global Forum in Basel. That'll be a nice way to introduce me to ARM design. Shall we earmark December 14th for a deep dive on this?

pknowles (Thu, 29 Nov 2018 06:46:12 GMT):
@mwherman2000 Schema attributes and credentials (Customer Address, Supplier Address, etc.) can obviously be aggregated from various sources. In the case of aggregated data capture, the creator and subsequent issuer of the aggregated schema would be the "Issuer". Not sure if that helps or not. I've tweaked some of my previous responses to add clarity. Perhaps you, @swcurran and I can go through the diagram together in person at the Hyperledger Global Forum in Basel. That'll be a nice way to introduce me to ARM design. Shall we earmark December 14th for a deep dive on this?

swcurran (Thu, 29 Nov 2018 07:26:32 GMT):
@mwherman2000 - an interesting use case. I'd be interested in participating in a discussion in Basel about it. Keep me in the loop. One way we've been thinking about the design of Credentials is determining how the holder of the Verifiable Credential would have to use it - what would they need to prove and to whom? In thinking about it that way, are all of the Facts in the use case VCs? Perhaps, but I'm not certain at the level I've gone through it. Something to consider. On the Issuer/Holder question - I think that in general, the role of Alice and Bob are indeed as delegates of Acme and Bobbles, respectively, and so Acme/Bobbles are the likely Issuers/Holders of the VCs. The delegate relationship is (current implementation at least) still informally defined but as these more complex use cases get implemented, it will crystalize. It will be extremely common.

infominer33 (Thu, 29 Nov 2018 09:03:28 GMT):
ok, @drummondreed now I'm watching a video of you on ssimeetup where you say that the DID is the address to a public key on the ledger

infominer33 (Thu, 29 Nov 2018 09:03:28 GMT):
ok, @drummondreed now I'm watching a video of you on ssimeetup where you say that the DID is the address to a public key on the ledger... so i'm a little confused :D

infominer33 (Thu, 29 Nov 2018 09:03:28 GMT):
ok, @drummondreed now I'm watching a video of you on ssimeetup where you say that the DID is the address to a public key on the ledger... so i'm a little confused :)

infominer33 (Thu, 29 Nov 2018 09:05:47 GMT):
video being "The Story of Open SSI Standards"

infominer33 (Thu, 29 Nov 2018 09:05:47 GMT):
video being "The Story of Open SSI Standards" @ 17:50

infominer33 (Thu, 29 Nov 2018 09:09:20 GMT):
but what I thought we cleared up is that a pairwise did public hash is stored on a microledger w my agent, and only communicated p2p with whoever I'm transacting

swcurran (Thu, 29 Nov 2018 10:59:20 GMT):
Think of "public" as published to those that need to access it. If only a single party needs to know it, share it directly (eg microledger), if arbitrary parties need access to it, put it on the Public Ledger (eg. Sovrin).

mwherman2000 (Thu, 29 Nov 2018 10:59:56 GMT):
@pknowles Dec. 14th certainly works for me to meet and discussion how to mostly effectively represent Business Documents as Sovrin Entities (subjec to some additional requirements which I will document beforehand). @swcurran ? @drummondreed ? @danielhardman ? @darrell.odonnell Who else?

swcurran (Thu, 29 Nov 2018 11:00:02 GMT):
Think of "public" as published to those that need to access it. If only a single party needs to know it, share it directly (eg microledger), if arbitrary parties need access to it, put it on the Public Ledger (eg. Sovrin).

swcurran (Thu, 29 Nov 2018 11:00:02 GMT):
Think of "public" as published to those that need to access it. If only a single party needs to know it, share it directly (eg microledger), if arbitrary parties need access to it, put it on a Public Ledger (eg. Sovrin).

infominer33 (Thu, 29 Nov 2018 11:56:36 GMT):
thx @swcurran I have a question also about the edX course... does it teach indy generally as deployed by sovrin? trying to clarify as Drummond pointed out that the code could be used differently than as in sovrin

infominer33 (Thu, 29 Nov 2018 11:58:37 GMT):
based on the "what goes on the blockchain" section of edx, it seems to be taught as implemented in sovrin.. but it would be nice to know if there are any differences I should look out for

infominer33 (Thu, 29 Nov 2018 11:59:30 GMT):
because I'm working on some educational material also, and I don't want to produce conflicting information

swcurran (Thu, 29 Nov 2018 12:23:23 GMT):
I focused on what was implemented in Indy at the time of writing. Since Sovrin is runs Indy software, it's much the same, but you do have to be carefully in taking about what is there now and what is intended.

swcurran (Thu, 29 Nov 2018 12:23:23 GMT):
I focused on what was implemented in Indy at the time of writing. Since Sovrin is runs Indy software, it's much the same, but you do have to be carefully in talking about what is there now and what is intended.

infominer33 (Thu, 29 Nov 2018 12:31:41 GMT):
btw, is there some procedure for wiki edits? I replaced a broken link earlier... but I didn't expect my change to automatically go thru. I never wiki'd before.. but I expect I will be doing a lot of it into the future, generally speaking

kdenhartog (Thu, 29 Nov 2018 12:35:24 GMT):
nah, any updates are appreciated. I believe changes are tracked so if it's defaced it's easy to change back. Let the wiki wars happen :)

infominer33 (Thu, 29 Nov 2018 12:37:34 GMT):
yes, I see there is a revision history.. the very first link was: https://github.com/hyperledger/indy-rfc/blob/master/text/ssi-notation.md (which doesn't exist) and I changed it to the only thing I could find (but maybe not the perfectly correct link): https://github.com/hyperledger/indy-hipe/tree/master/text/0014-ssi-notation

Sean_Bohan (Thu, 29 Nov 2018 15:19:31 GMT):
At the top of the hour is the Weekly INDY WG Call where we will be joined by Silas from the BURROW team and talk about building bridges between the two teams. Join us! *Thursday INDY WG Call:* Date: Thurs Nov 29 When: 8am PT, 9am MT, 10am CT, 11amET, 4pmGMT Where: https://zoom.us/j/232861185 Agenda: Building bridges with Burrow team (SilasD to join us) Previous Recordings and Agendas: https://drive.google.com/drive/u/0/folders/1AwHWN95KmSEi5fijraID0tFFMzYHoMwt?ogsrc=32 (edited) (edited)

mhailstone (Thu, 29 Nov 2018 16:04:07 GMT):
Meeting is back on.

silasdavis (Thu, 29 Nov 2018 17:22:38 GMT):
Has joined the channel.

silasdavis (Thu, 29 Nov 2018 17:23:16 GMT):
Could someone link me to something on the foundation of linked secrets?

changeagent (Thu, 29 Nov 2018 20:40:33 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=jnteRP9Gcy8LHhyov) @infominer33 Thanks! Sort of... After further research, my understanding is that DIDs and the identity records associated with them can either be published to the chain, or, as in Sovrin, they can be used to set up an encrypted communication channel between two parties (the pseudonymous option, where you get the benefit of public-private key cryptography but it does not use the ledger).

changeagent (Thu, 29 Nov 2018 20:40:33 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=jnteRP9Gcy8LHhyov) @infominer33 Thanks! Sort of... After further research, my understanding is that DIDs and the identity records associated with them can either be published to the chain, or, as in Sovrin, they can be used to set up an encrypted communication channel between two parties (the pseudonymous option, where you get the benefit of public-private key cryptography but it does not use the public ledger).

Sean_Bohan (Thu, 29 Nov 2018 21:18:12 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=rmzgSqBC7Z9ZS5qZb) @silasdavis From Mike @ Sovrin: https://github.com/hyperledger/indy-crypto/blob/master/libindy-crypto/docs/AnonCred.pdf

infominer33 (Thu, 29 Nov 2018 21:58:28 GMT):
@changeagent yes, that's the point ,it uses the public ledger for public information only... and there is no need for the entire world to have your public key, only the private parties you are transacting with.

infominer33 (Thu, 29 Nov 2018 22:10:52 GMT):
we basically are saying the same thing back to eachother

danielhardman (Thu, 29 Nov 2018 23:36:26 GMT):
@mwherman2000 I will be in Basel on the 14th. Not sure of my exact schedule that day--but assuming we don't pick a time when I'm otherwise committed, by all means let's meet then.

jljordan_bcgov (Fri, 30 Nov 2018 00:22:04 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=cdtnFxaSZDRabJ6G4) @Sean_Bohan https://github.com/hyperledger/indy-sdk/blob/master/doc/design/005-dkms/DKMS%20Design%20and%20Architecture%20V3.md @silasdavis This is the design document for decentralized key management ... See section 5.1

jljordan_bcgov (Fri, 30 Nov 2018 00:22:04 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=cdtnFxaSZDRabJ6G4) @Sean_Bohan https://github.com/hyperledger/indy-sdk/blob/master/doc/design/005-dkms/DKMS%20Design%20and%20Architecture%20V3.md This is the design document for decentralized key management ... See section 5.1

jljordan_bcgov (Fri, 30 Nov 2018 00:28:14 GMT):
^^ @silasdavis

mtfk (Fri, 30 Nov 2018 13:11:46 GMT):
Has joined the channel.

donqui (Fri, 30 Nov 2018 15:34:03 GMT):
Has joined the channel.

changeagent (Fri, 30 Nov 2018 16:37:16 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=AJZSXxkqCgMae9Msr) @infominer33 tomato... tomato... :)

TramAnh (Fri, 30 Nov 2018 17:30:46 GMT):
Has joined the channel.

TramAnh (Fri, 30 Nov 2018 17:31:51 GMT):
hello guys

mboyd (Fri, 30 Nov 2018 20:23:29 GMT):
Hi all, I've created a consolidated Indy Documentation website hosted on ReadtheDocs at http://indy.readthedocs.io. *It's still in beta*. What do you think of it? How would you improve it? Please add your feedback to my proposal here: https://github.com/hyperledger/indy-hipe/pull/63 Thanks!

mboyd (Fri, 30 Nov 2018 20:23:29 GMT):
Hi all, I've created a consolidated Indy Documentation website hosted on ReadtheDocs at http://indy.readthedocs.io. *It's still in beta*. What do you think of it? How would you improve it? Please add your feedback to my Hyperledger Improvement Proposal here: https://github.com/hyperledger/indy-hipe/pull/63 Thanks!

infominer33 (Sat, 01 Dec 2018 03:08:28 GMT):
is there any special process or requirements for getting access to Jira?

dbluhm (Sat, 01 Dec 2018 04:43:57 GMT):
You log into Jira using a Linux foundation account

dbluhm (Sat, 01 Dec 2018 04:44:36 GMT):
https://identity.linuxfoundation.org/user

infominer33 (Sat, 01 Dec 2018 05:19:06 GMT):
oh thx! I tried, but maybe not hard enough

infominer33 (Sat, 01 Dec 2018 05:19:54 GMT):
very helpful link !

tuckerg (Sat, 01 Dec 2018 09:01:37 GMT):
Has joined the channel.

NursultanMakhanov (Sat, 01 Dec 2018 15:25:44 GMT):
Has joined the channel.

mwherman2000 (Sat, 01 Dec 2018 15:30:55 GMT):
I'm working on a fairly detailed architecture model of the Getting Started user scenario ...from a which a series of views (simple diagrams) can created using drag&drop. Here's what I call an "all-in" construction view from page 2 (only) of the Getting Started document (What We'll Cover and "About Alice")...

mwherman2000 (Sat, 01 Dec 2018 15:30:55 GMT):
FYI: I'm working on a fairly detailed architecture model of the Getting Started user scenario ...from a which a series of views (simple diagrams) can created using drag&drop. Here's what I call an "all-in" construction view from page 2 (only) of the Getting Started document (What We'll Cover and "About Alice")...

mwherman2000 (Sat, 01 Dec 2018 15:30:55 GMT):
FYI: I'm working on a fairly detailed architecture model of the #indy-sdk Getting Started user scenario ...from a which a series of views (simple diagrams) can created using drag&drop. Here's what I call an "all-in" construction view from page 2 (only) of the Getting Started document (What We'll Cover and "About Alice")...

mwherman2000 (Sat, 01 Dec 2018 15:30:55 GMT):
FYI: I'm working on a fairly detailed architecture model of the #indy-sdk Getting Started user scenario ...from which a series of views (simple diagrams) can created using drag&drop. Here's what I call an "all-in" construction view from page 2 (only) of the Getting Started document (What We'll Cover and "About Alice")...

mwherman2000 (Sat, 01 Dec 2018 15:30:55 GMT):
FYI: I'm working on a fairly detailed architecture model of the #indy-sdk Getting Started user scenario ...from which a series of views (simpler diagrams) can created using drag&drop. Here's what I call an "all-in" construction view from page 2 (only) of the Getting Started document (What We'll Cover and "About Alice")...

mwherman2000 (Sat, 01 Dec 2018 15:30:55 GMT):
FYI: I'm working on a fairly detailed architecture model of the #indy-sdk Getting Started user scenario ...from which a series of views (simpler diagrams) can created using drag&drop. Here's what I call an "all-in" construction view based only on page 2 (so far) of the Getting Started document (What We'll Cover and "About Alice")...bsed

mwherman2000 (Sat, 01 Dec 2018 15:30:55 GMT):
FYI: I'm working on a fairly detailed architecture model of the #indy-sdk Getting Started user scenario ...from which a series of views (simpler diagrams) can created using drag&drop. Here's what I call an "all-in" construction view based only on page 2 (so far) of the Getting Started document (What We'll Cover and "About Alice")...

mwherman2000 (Sat, 01 Dec 2018 15:31:25 GMT):

Indy-SDK-Getting-Started-Step 0 v0.3.png

mwherman2000 (Sat, 01 Dec 2018 15:32:45 GMT):
The intent is for some of these views to be added to the Getting Started document at some point.

Shyam_Pratap_Singh (Sat, 01 Dec 2018 16:10:08 GMT):
Hi Everyone, I have created a post on Hyperledger Indy and installation with Ubuntu. Kindly provide your valuable feedback to take this forward. Link : https://medium.com/@spsingh559/hyperledger-indy-installation-with-ubuntu-and-node-js-4c1bff81ebc7

mwherman2000 (Sat, 01 Dec 2018 19:12:56 GMT):
@shyam 40 Hi, I know you have great intentions but I have only read the first 2 paragraphs and found myself forming some rather interesting opinions/reactions wrt the title of the article and these 2 paragraphs ...checkout https://medium.com/@mwherman2000/i-have-only-read-the-first-2-paragraphs-and-found-myself-forming-some-rather-interesting-9e7fd91f67d1 ...continuing with the rest of the article.

Shyam_Pratap_Singh (Sat, 01 Dec 2018 19:44:40 GMT):
Focus of article was about Indy set up and instead of directly going to set up, I just added content so that people can appreciate real power of Indy. Thanks for your response.

mwherman2000 (Sun, 02 Dec 2018 03:12:34 GMT):
@kdenhartog In your recent youtube video (https://youtu.be/llwfb5Ut5sg?t=386), you mention DIDs are based on RFC2141. My reading of RFC2141 indicates that the scheme portion of a URN needs to be "urn:" ...not anything else (e.g. "did:"). Should your slide reference some other, more general, specification that permits alternative scheme identifiers? Reference: https://www.ietf.org/rfc/rfc2141.txt

mwherman2000 (Sun, 02 Dec 2018 03:12:34 GMT):
@kdenhartog Q1. In your recent youtube video (https://youtu.be/llwfb5Ut5sg?t=386), you mention DIDs are based on RFC2141. My reading of RFC2141 indicates that the scheme portion of a URN needs to be "urn:" ...not anything else (e.g. "did:"). Should your slide reference some other, more general, specification that permits alternative scheme identifiers? Reference: https://www.ietf.org/rfc/rfc2141.txt

kdenhartog (Sun, 02 Dec 2018 03:35:42 GMT):
It may need to this is a slide that I borrowed from @drummondreed he may be able to provide more insight into this.

kdenhartog (Sun, 02 Dec 2018 03:35:42 GMT):
It may need to this is a slide that I borrowed from @drummondreed he may be able to provide more insight into this. As I go and reread it I think it's supposed to reference RFC 3986 which is the URI rfc.

mwherman2000 (Sun, 02 Dec 2018 05:58:55 GMT):
@kdenhartog Follow-up question: Neither the Indy nor Sovrin Glossaries use the word credential e.g. Credential Definition - they seem to prefer the word Claim instead e.g. Claim Definition. Which is the correct/current terminology? References: https://docs.google.com/document/d/1giOzpTFXypJ6bAUp_6g93kYOEiNa5eWI1KeIg6wb598/edit#heading=h.h07siytb69z and https://wiki.hyperledger.org/projects/indy/glossary

mwherman2000 (Sun, 02 Dec 2018 05:58:55 GMT):
@kdenhartog Q2. Follow-up question: Neither the Indy nor Sovrin Glossaries use the word credential e.g. Credential Definition - they seem to prefer the word Claim instead e.g. Claim Definition. Which is the correct/current terminology? References: https://docs.google.com/document/d/1giOzpTFXypJ6bAUp_6g93kYOEiNa5eWI1KeIg6wb598/edit#heading=h.h07siytb69z and https://wiki.hyperledger.org/projects/indy/glossary

mwherman2000 (Sun, 02 Dec 2018 06:30:17 GMT):
@kdenhartog Q3. Follow-up re: your draft Global Forum presentation. In your presentation of the "3-axis" model, I think you're missing a 4th axis: Time ...because the organizational role I'm associated with at any given instant is like to change over time. I don't always have the same role (aka relationship) with my university/school, my back, whoever is my current employer, etc. Time is an important element. More importantly, does the Time dimension affect the requirements for the Microledger? Here's a visual example of all of Alice's roles changing over time...

mwherman2000 (Sun, 02 Dec 2018 06:30:53 GMT):

Indy-SDK-Getting-Started-Step 0 v0.4.png

swcurran (Sun, 02 Dec 2018 06:43:37 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=Cpka9Tg7ciMieSa4h) @mwherman2000 This has been a bit of a hot topic in the past - e.g. several years ago. It's mostly settled now AFAIK. The word "Claim" was used in previous works and then in the name of the W3C Working Group (names of which are not easy to change). However, the term Credential is now considered to mean a group of Claims. So a Verifiable Credential is issued to a Holder and that Verifiable Credential contains one or more Verifiable Claims. The Holder can then prove (caveat here) the individual Claims within the Credential to a Verifier. The caveat is that some Verifiable Credential systems handle proofs only of the Credential, not of the Claims within a Credential. That approach means that selective disclosure, ZKPs and Proofs referencing Claims from multiple Credentials are not supported. Indy (and hence Sovrin) support all of those capabilities.

swcurran (Sun, 02 Dec 2018 06:43:37 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=Cpka9Tg7ciMieSa4h) @mwherman2000 This has been a bit of a hot topic in the past - e.g. several years ago. It's mostly settled now AFAIK. "Claim" was used in previous works and in the name of the W3C Verifiable Claims Working Group (names of which are not easy to change). However, the term Credential is now considered to mean a group of Claims. So a Verifiable Credential is issued to a Holder and that Verifiable Credential contains one or more Verifiable Claims. The Holder can then prove (caveat here) the individual Claims within the Credential to a Verifier. The caveat is that some Verifiable Credential systems handle proofs only of the Credential, not of the Claims within a Credential. That approach means that selective disclosure, ZKPs and Proofs referencing Claims from multiple Credentials are not supported. Indy (and hence Sovrin) support all of those capabilities.

mwherman2000 (Sun, 02 Dec 2018 07:54:09 GMT):
@swcurran Is there a plan to update the Glossaries? ....it's confusing when core content doesn't reflect reality.

swcurran (Sun, 02 Dec 2018 08:28:50 GMT):
@mwherman2000 - yes. You are looking at the V1 version that is currently linked on the Sovrin site. However, the V2 version is in Draft along with the rest of the Sovrin Governance Framework. Here is a link to Claim in the document that talks about the change from V1 to V2. https://docs.google.com/document/d/1gfIz5TT0cNp2kxGMLFXr19x1uoZsruUe_0glHst2fZ8/edit#heading=h.r4mpdswkxgk2

swcurran (Sun, 02 Dec 2018 08:28:50 GMT):
@mwherman2000 - yes. You are looking at the V1 version that is currently linked on the Sovrin site. However, the V2 version is in Draft along with the rest of the Sovrin Governance Framework V2. Here is a link to Claim in the document that talks about the change from V1 to V2. https://docs.google.com/document/d/1gfIz5TT0cNp2kxGMLFXr19x1uoZsruUe_0glHst2fZ8/edit#heading=h.r4mpdswkxgk2

swcurran (Sun, 02 Dec 2018 12:42:38 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=wyMCzReRRgPHXvXum) @danielhardman @mwherman2000 @pknowles @darrell.odonnell - on the 14th, I have an all morning workshop that we are giving, so having this meeting in the afternoon would be appreciated. Thanks

darrell.odonnell (Sun, 02 Dec 2018 12:52:16 GMT):
I should be able to adjust to whenever. Kick some butt there dudes!

pknowles (Sun, 02 Dec 2018 13:25:26 GMT):
Perhaps we can convene right after lunch on the 14th. I'll keep the afternoon clear.

mwherman2000 (Sun, 02 Dec 2018 13:41:27 GMT):
Sounds like we have some "consensus" for the afternoon of Friday Dec. 14. #imin

mwherman2000 (Sun, 02 Dec 2018 13:45:36 GMT):
Confirming the theme for the Dec. 14 meeting is: _Generalizable approaches for representing NFEs (non-fungible entities) in Hyperledger Indy_?

mwherman2000 (Sun, 02 Dec 2018 13:46:55 GMT):
...with the UBL Purchase Order as a use case: http://docs.oasis-open.org/ubl/os-UBL-2.2/UBL-2.2.html#S-INVOICE-SCHEMA

mwherman2000 (Sun, 02 Dec 2018 13:46:55 GMT):
...using the UBL Purchase Order schema as a use case: http://docs.oasis-open.org/ubl/os-UBL-2.2/UBL-2.2.html#S-INVOICE-SCHEMA

mwherman2000 (Sun, 02 Dec 2018 13:46:55 GMT):
...using the UBL *Invoice* schema as a use case: http://docs.oasis-open.org/ubl/os-UBL-2.2/UBL-2.2.html#S-INVOICE-SCHEMA

mwherman2000 (Sun, 02 Dec 2018 16:07:57 GMT):
Who is the presenter/voiceover in this video?

pknowles (Sun, 02 Dec 2018 16:56:45 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=YS5ALd5kWbs59cTFC) @mwherman2000 Presenter: @danielhardman

mwherman2000 (Sun, 02 Dec 2018 17:34:36 GMT):
@danielhardman @mwherman2000 @pknowles @darrell.odonnell @swcurran If you DM me your email address, I'll add it to the Dec.14 meeting invite I sent out a minute or so ago to myself, Paul and Darrell.

mwherman2000 (Sun, 02 Dec 2018 17:35:39 GMT):
If anyone else has an interest in this topic (NFEs on the Indy Ledger), DM me your email address as well.

mwherman2000 (Sun, 02 Dec 2018 18:14:52 GMT):
@danielhardman I'm watching - the part where you're discussing crypto vs. identity wallets and was wondering... What is a modern name for an (Indy-Sovin) place/app that acts as a repository for (UBL) business documents? First, the documents need to be properly grouped/organized in to subledgers (subsidiary ledgers of a general ledger …ref: https://www.cliffsnotes.com/study-guides/accounting/accounting-principles-i/subsidiary-ledgers-and-special-journals/subsidiary-ledgers). If crypto has crypto currency wallets, and digital identities have identity wallets, what do we call this thing that stores general ledgers, subledgers, accounts, and business documents ...in addition to identities and crypto assets? …a portfolio, a register, a journal, a quickbooks (oops, that one is already taken), …? I kind of like Business Register ...Organization Business Register and Personal Business Register ...the OBR and the PBR. Your thoughts?

mwherman2000 (Sun, 02 Dec 2018 18:14:52 GMT):
@danielhardman I'm watching - the part where you're discussing crypto vs. identity wallets and was wondering... What is a modern name for an (Indy-Sovin) place/app that acts as a repository for (UBL) business documents? First, the documents need to be properly grouped/organized in to subledgers (subsidiary ledgers of a general ledger …ref: https://www.cliffsnotes.com/study-guides/accounting/accounting-principles-i/subsidiary-ledgers-and-special-journals/subsidiary-ledgers). If crypto has crypto currency wallets, and digital identities have identity wallets, what do we call this thing that stores general ledgers, subledgers, accounts, and business documents ...in addition to identities and crypto assets? …a portfolio, a register, a journal, a quickbooks (oops, that one is taken already), …? I kind of like Business Register ...Organization Business Register and Personal Business Register ...the OBR and the PBR. Your thoughts?

mwherman2000 (Sun, 02 Dec 2018 18:18:00 GMT):

20890ngr002.jpg

mwherman2000 (Sun, 02 Dec 2018 18:18:00 GMT):

20890ngr002.jpg

cfzhang (Sun, 02 Dec 2018 19:30:06 GMT):
Has joined the channel.

cfzhang (Sun, 02 Dec 2018 19:34:51 GMT):
hello, all, I am setting up Indy with docker and play the Indy CLI as https://indy.readthedocs.io/projects/sdk/en/latest/design/001-cli/README.html#commands. I could create DID and create a new schema with command 'ledger schema'. But what exactly is the command 'ledger cred-def'? what is the 'schema-id' and 'primary' in the command 'ledger cred-def schema_id= signature_type= [tag=] primary='?

HaiTranThanh (Mon, 03 Dec 2018 08:27:45 GMT):
Has joined the channel.

bootstrapsp (Mon, 03 Dec 2018 17:14:52 GMT):
Looking for some help on understanding what should be the network name and is the indy_config.py is the right file to set this parameter in? Installed the IndyNode using the deb pacakge for ubuntu so its basically a non docker installation

HiranKumar (Tue, 04 Dec 2018 06:39:26 GMT):
I got an error like poolledgernotcreatederror

HiranKumar (Tue, 04 Dec 2018 06:39:49 GMT):
Would you please tell me the reason for the above error

Cliff.Huang (Tue, 04 Dec 2018 09:34:57 GMT):
Has joined the channel.

Francis-Guttridge (Tue, 04 Dec 2018 17:19:04 GMT):
Has joined the channel.

Francis-Guttridge (Tue, 04 Dec 2018 17:23:23 GMT):
Hello all. I've been working in the blockchain space for a while now, and in particular I've had a keen interest in identity. I've heard a lot about this project but I'm only just now finding time to start digging into it. I'm wondering if there is any high level documentation & process flows to help get me up to speed with some of the terminology before I dig into deploying/contributing. Would Sovrin be the right place to look?

MiguelFJimenezSola (Tue, 04 Dec 2018 19:25:46 GMT):
Has joined the channel.

kdenhartog (Tue, 04 Dec 2018 20:28:33 GMT):
check out https://indy.readthedocs.io https://wiki.hyperledger.org/projects/indy

kdenhartog (Tue, 04 Dec 2018 20:29:55 GMT):
@Francis-Guttridge check out https://indy.readthedocs.io https://wiki.hyperledger.org/projects/indy and https://vonx.io Sovrin also has a lot of great documentation on this stuff as well.

kdenhartog (Tue, 04 Dec 2018 20:34:01 GMT):
Another great option from @infominer33 is https://github.com/infominer33/awesome-decentralized-id

infominer33 (Tue, 04 Dec 2018 20:38:52 GMT):
this is a good intro: https://github.com/infominer33/awesome-decentralized-id/blob/master/awesome-sovrin/README.md

infominer33 (Tue, 04 Dec 2018 20:39:44 GMT):
https://medium.com/axiom-tech/hyperledger-indy-the-future-of-decentralized-identity-9de5c2459e4e

Francis-Guttridge (Tue, 04 Dec 2018 20:41:59 GMT):
Thanks !

danielhardman (Tue, 04 Dec 2018 22:07:17 GMT):
I have been exploring the idea of writing a did:peer method spec. This would document how to manage a relationship and its associated DIDs and keys purely through peer-to-peer communication. It would not be specific to Indy (or, indeed, to any particular blockchain), though it would be implemented using Indy messaging concepts because A2A addresses many of the mechanisms it needs. Anybody want to collaborate?

haow85 (Wed, 05 Dec 2018 01:14:25 GMT):
Does anyone know how zero-knowledge-proof works in HyperLedger Indy ?

infominer33 (Wed, 05 Dec 2018 03:29:52 GMT):
@haow85 https://github.com/hyperledger/indy-anoncreds

infominer33 (Wed, 05 Dec 2018 03:37:06 GMT):
https://github.com/infominer33/awesome-decentralized-id/blob/master/README.md#idemix-

Shyam_Pratap_Singh (Wed, 05 Dec 2018 09:06:46 GMT):
Hi Team, can you please help me in solving this issue " (node:11797) UnhandledPromiseRejectionWarning: IndyError: WalletInvalidHandle". I am not getting why this error is coming?

kdenhartog (Wed, 05 Dec 2018 09:47:18 GMT):
@shyam 40 This is an error that occurs when the wallet_handle that's being passed isn't able to be found. Typically, this occurs because you haven't created/opened a wallet yet which returns the wallet_handle.

kdenhartog (Wed, 05 Dec 2018 09:47:18 GMT):
@shyam\ 40 This is an error that occurs when the wallet_handle that's being passed isn't able to be found. Typically, this occurs because you haven't created/opened a wallet yet which returns the wallet_handle.

kdenhartog (Wed, 05 Dec 2018 09:47:18 GMT):
@shyam 40 This is an error that occurs when the wallet_handle that's being passed isn't able to be found. Typically, this occurs because you haven't created/opened a wallet yet which returns the wallet_handle.

mtfk (Wed, 05 Dec 2018 09:49:38 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=Fi8CFes3W4p55hcBM) @danielhardman Sounds very interesting, I don't know much about A2A from sovrin yet but I was looking for solution for identification in completely offline world base only on mobile devices and the networks/connections which you would be able to get there (Wifi direct, BT LE) in flight mode. If there will be a way to contribute from my side would love to.

PatrikStas (Wed, 05 Dec 2018 10:47:35 GMT):
Hi, I am trying to add Trust Anchor on Sovrin Testnet via the https://s3.us-east-2.amazonaws.com/evernym-cs/sovrin-STNnetwork/www/trust-anchor.html page but I am getting back `502 internal server error` Anyone knows if something is going on with that application, or if I might be doing something wrong? But I am pretty sure I am putting in valid DID and valid abbreviated verkey.

Shyam_Pratap_Singh (Wed, 05 Dec 2018 11:08:53 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=BNfh3Fe9cySb3LntL) @kdenhartog Thank you so much, now issue has been resolved!!

Shyam_Pratap_Singh (Wed, 05 Dec 2018 11:10:44 GMT):
I can see two version of proverStoreCredential function in node js wrapper. In typical Alice use case, for transcript await indy.proverStoreCredential(aliceWallet, null, transcriptCredRequestMetadataJson, authdecryptedTranscriptCredJson, faberTranscriptCredDef, null);

JonathanGiglio (Wed, 05 Dec 2018 17:52:13 GMT):
Has joined the channel.

Francis-Guttridge (Wed, 05 Dec 2018 19:48:31 GMT):
Is there an updated version of the faber college sample that works with the new cli? I'm also curious as to exactly what they .indy files are doing in these samples. Are they being processed as custom transactions? As an aside to that question, how exactly are the relationships between entities managed. I read the document explaining the various relationships and I understood that, but within indy itself, is each new pairwise relationship represented by a new DID(issued by the trusted org), or is it possible to have a single DID with multiple attestations to the credibility of the holder? For example on a single DID I may have multiple TRUST_ANCHORS signatures.

Francis-Guttridge (Wed, 05 Dec 2018 19:48:31 GMT):
Is there an updated version of the faber college sample that works with the new cli? I'm also curious as to exactly what they .indy files are doing in these samples. Are they being processed as custom transactions? As an aside to that question, how exactly are the relationships between entities managed. I read the document explaining the various relationships and I understood that, but within indy itself, is each new pairwise relationship represented by a new DID(issued by the trusted org), or is it possible to have a single DID with multiple attestations to the credibility of the holder? For example on a single DID I may have multiple TRUST_ANCHORS signatures. I believe something along the lines was mentioned A+B+C:D

esplinr (Wed, 05 Dec 2018 21:09:03 GMT):
An architecture overview of Indy Plenum: https://www.youtube.com/watch?v=WZin717AT_A&list=PLRp0viTDxBWGLdZk0aamtahB9cpJGV7ZF

srottem (Thu, 06 Dec 2018 01:03:20 GMT):
I've asked this multiple times before on the indy-sdk channel but have had no response: Who is responsible for the Hyperledger Indy Github repo? The current .NET wrapper's README.md points to the Hyperledger repo for the .NET docs but it's not working because the repo needs to be configured with "Github Pages" with the source as the Master branch. I'd like to get this sorted out, otherwise we'll have to find another source for hosting the docs since they're inaccessible. See here for details on the change that needs to be made: https://help.github.com/articles/configuring-a-publishing-source-for-github-pages/

dbluhm (Thu, 06 Dec 2018 01:36:22 GMT):
@srottem Sorry for the frustration. Part of what may be causing issues is that the Hyperledger help desk has to make these sort of changes. Just to be sure, @nage does that seem accurate to you? With the recent work on a read the docs site for Indy by @mboyd it might be the case that github-pages isn't the ideal place for this documentation anymore anyways. Thoughts?

dbluhm (Thu, 06 Dec 2018 01:37:06 GMT):
Though it clearly is a problem that we have broken links in the README

SanjeevKumarn (Thu, 06 Dec 2018 06:05:51 GMT):
Has joined the channel.

AvikHazra (Thu, 06 Dec 2018 10:53:17 GMT):
when i`m trying to build cargo i`m getting failed to run custom build command for `openssl v0.9.24` error. can anyone help me ?

twshelton (Thu, 06 Dec 2018 16:09:03 GMT):
@AvikHazra ... can you provide sme details on the build environment?

nage (Thu, 06 Dec 2018 16:45:14 GMT):
Please take a look at https://indy.readthedocs.io as a new documentation portal and provide feedback to @mboyd

nage (Thu, 06 Dec 2018 16:46:54 GMT):
@gudkov was the main contact on the Indy-ask repo. @MattRaffel and @sergey.minaev should be able to help on repo issues there as well @srottem [ ](https://chat.hyperledger.org/channel/indy?msg=4EStbaHMiyiy6fpSp)

MattRaffel (Thu, 06 Dec 2018 16:46:55 GMT):
Has joined the channel.

Francis-Guttridge (Thu, 06 Dec 2018 18:07:42 GMT):
is cli version 1.6.8 still WIP?

nage (Thu, 06 Dec 2018 19:20:19 GMT):
I show the 1.6.8 as the latest sdk release, so I would expect that CLI to work properly. There is still a lot of functionality it lacks that we would like to add (specifically around credentials exchange and wallet management)

nage (Thu, 06 Dec 2018 19:20:36 GMT):
what are you seeing?

Francis-Guttridge (Thu, 06 Dec 2018 19:34:43 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=7Ne6xPjmYrwcm7JNc) @nage It works great and it's a lot more clean than the previous versions, but it does seem to be limited in it's capabilities, specifically with regards to it's ability to follow the Alice & faber college getting started sample. can't seem to load any of the .indy files and what not

nage (Thu, 06 Dec 2018 19:53:20 GMT):
The new sdk based cli made the deliberate choice to focus on node transactions and interacting with the ledger over the credentials protocol. It is better to use the sdk api directly in a programming language for most of that work. Specifically the python examples in iPython or similar systems is more flexible and will get to a working POC without requiring a rewrite like deploying on the old CLI might have

esplinr (Thu, 06 Dec 2018 21:33:05 GMT):
Today's sprint demo from the Evernym team: https://www.youtube.com/watch?v=D3HrL4tvpP8

coderintherye (Fri, 07 Dec 2018 00:09:35 GMT):
Which actors in the Indy terminology can be issuers of verifiable claims? For instance, a trust anchor can create identities, presumably they can also issue claims. May someone with one of those newly created identities who is *not* a trust anchor also be able to issue verifiable claims?

kdenhartog (Fri, 07 Dec 2018 00:13:57 GMT):
@coderintherye any identity can be an issuer

Sean_Bohan (Fri, 07 Dec 2018 15:47:38 GMT):
GREAT Indy WG call yesterday: @jljordan_bcgov and @ianco from the VON team (BC.gov) shared a status update for VON and gave us a sneak preview of what they will be presenting next week at the Hyperledger Global Forum @ashcherbakov gave an overview of documentation in Plenum, which was great. LINKS: https://theorgbook.pathfinder.gov.bc.ca/api/ https://indy.readthedocs.io https://indy.readthedocs.io/projects/plenum/en/latest/index.html VIDEO: https://drive.google.com/open?id=1166XpTM8WgZVMN2ca53CRCJapZlAeUhM NEXT THURSDAY: The IXO team will share the work they have been doing with INDY.

Sean_Bohan (Fri, 07 Dec 2018 15:47:38 GMT):
GREAT Indy WG call yesterday: @jljordan_bcgov and @ianco from the VON team (BC.gov) shared a status update for VON and gave us a sneak preview of what they will be presenting next week at the Hyperledger Global Form @ashcherbakov gave an overview of documentation in Plenum, which was great. Links: https://theorgbook.pathfinder.gov.bc.ca/api/ https://indy.readthedocs.io https://indy.readthedocs.io/projects/plenum/en/latest/index.html https://drive.google.com/open?id=1166XpTM8WgZVMN2ca53CRCJapZlAeUhM NEXT THURSDAY: The IXO team will share the work they have been doing with INDY.

Sean_Bohan (Fri, 07 Dec 2018 15:47:38 GMT):
GREAT Indy WG call yesterday: @jljordan_bcgov and @ianco from the VON team (BC.gov) shared a status update for VON and gave us a sneak preview of what they will be presenting next week at the Hyperledger Global Form @ashcherbakov gave an overview of documentation in Plenum, which was great. LINKS: https://theorgbook.pathfinder.gov.bc.ca/api/ https://indy.readthedocs.io https://indy.readthedocs.io/projects/plenum/en/latest/index.html VIDEO: https://drive.google.com/open?id=1166XpTM8WgZVMN2ca53CRCJapZlAeUhM NEXT THURSDAY: The IXO team will share the work they have been doing with INDY.

axel (Fri, 07 Dec 2018 16:44:29 GMT):
yes, interesting stuff indeed! thanks!

the_identity_guy (Sun, 09 Dec 2018 01:36:23 GMT):
Question about the trust model with Indy/Sovrin: Given that the trust placed by service providers / identity consumers on given claims, really depends on the level of trust the service provider have for the claim 'issuer', How does the SSI concept or Indy solve the decentralized trust model. It seems the services provides still have to be able to trust the identity providers somehow, which maybe done offline and outside of the Indy network.. perhaps through traditional PKI or WebofTrust (reputation based) approaches. Does Indy/Sovrin offer any novel solution in how different entities can trust each others public keys, or is this requirement outside of the Indy ecosystem. I am aware of the trust framework that Sovrin has, but I think that is also a form of weboftrust/pki no?

the_identity_guy (Sun, 09 Dec 2018 01:36:23 GMT):
Question about the trust model with Indy/Sovrin: Given that the trust placed by service providers / identity consumers on given claims, really depends on the level of trust the service provider have for the claim 'issuer', How does the SSI concept or Indy solve the decentralized trust model. It seems the services provides still have to be able to trust the identity providers somehow, which maybe done offline and outside of the Indy network.. perhaps through traditional PKI or WebofTrust (reputation based) approaches. How does Indy/Sovrin facilitate the trust on the public keys, or is this requirement outside of the Indy ecosystem. I have limited understanding of the trust framework that Sovrin provides, perhaps the key piece is the trust framework agreed by all parties?

the_identity_guy (Sun, 09 Dec 2018 01:36:23 GMT):
Question about the trust model within Indy/Sovrin: Given that the trust placed by service providers / identity consumers on given claims, really depends on the level of trust the service provider have for the claim 'issuer', How does the SSI concept or Indy solve the decentralized trust model. It seems the services provides still have to be able to trust the identity providers somehow, which maybe done offline and outside of the Indy network.. perhaps through traditional PKI or WebofTrust (reputation based) approaches. How does Indy/Sovrin facilitate the trust on the public keys, or is this requirement outside of the Indy ecosystem. I have limited understanding of the trust framework that Sovrin provides, perhaps the key piece is the trust framework agreed by all parties?

drummondreed (Sun, 09 Dec 2018 05:27:43 GMT):
@the_identity_guy Fantastic question. Short answer: while Indy doesn't address this directly, Sovrin does. The Sovrin Governance Framework Working Group has been active at the Sovrin Foundation for over two years and is just about done with the Sovrin Governance Framework V2. It's been in public review since Nov 1 and should be approved by the Sovrin Board of Trustees by the end of the year. You can see all the main documents at https://sovrin.org/governance-framework/

drummondreed (Sun, 09 Dec 2018 05:45:56 GMT):
The SGF V2 in turn lays the groundwork for what we call the Sovrin Web of Trust Model. This is a globally scalable formal web of trust model based on DIDS and verifiable credentials. It defines four main classes of issuers in terms of escalating strength-of-trust: Peer, Trust Anchor, Trust Hub, Governance Authority. All the roles are optional, i.e., any trust community can assemble the roles it needs to serve its purposes. For any Proof received by a Verifier, the Verifier can walk the trust chain up as far as that particular chain goes, i.e., if the Proof is from a Peer, does the Peer have a Proof from a Trust Anchor? Does the Trust Anchor have a Proof from a Trust Hub? Does the Trust Hub have a Proof from a Governance Authority? Governance Authorities publish what we call domain-specific governance frameworks, so they are the real backbones of trust, and under this model there can be tens of thousands of them and they can cross-certify to form a true global web of trust.

drummondreed (Sun, 09 Dec 2018 05:47:33 GMT):
Once the Sovrin Governance Framework V2 is passed, the next major job of the SGFWG is to formalize the Sovrin Web of Trust Governance Framework. We welcome anyone who wants to join us in this work; the Sovrin Foundation is an open governance community just like Hyperledger is an open source community. See the https://sovrin.org/governance-framework/ page for details of how to engage.

jeff_g (Sun, 09 Dec 2018 12:37:05 GMT):
Has joined the channel.

DJ_HC (Sun, 09 Dec 2018 14:55:58 GMT):
Has joined the channel.

SanjayJain (Sun, 09 Dec 2018 19:14:35 GMT):
Has joined the channel.

the_identity_guy (Sun, 09 Dec 2018 21:27:18 GMT):
Thank you very much for explaining this @drummondreed !. I will refer to https://sovrin.org/governance-framework/ to understand the governance framework of sovrin.

srottem (Sun, 09 Dec 2018 21:58:27 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=B3KJP1DCGItryfKvob) @nage Thanks @nage. The issue is that the .NET docs are API docs generated from the source as a collection of HTML pages so ReadTheDocs isn't an appropriate source for them unless they can host a collection of static HTML pages.

srottem (Sun, 09 Dec 2018 22:02:29 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=B3KJP1DCGItryfKvob) @nage The .NET wrapper API docs are generated from source using Docfx to produce a fully indexed collection of static HTML pages that match the way Microsoft lay out their MSDN documentation. Not sure this is compatible with ReadTheDocs.

nage (Mon, 10 Dec 2018 08:51:30 GMT):
@srottem Pinging @mboyd so that he will take a look at the SDK .NET docs to see what is going on there (I'm sure we can figure something out)

JeromeK13 (Mon, 10 Dec 2018 09:10:55 GMT):
Has joined the channel.

mwherman2000 (Mon, 10 Dec 2018 10:05:11 GMT):
As an aside, it's a bright sunny day here in Basel :-)

mwherman2000 (Mon, 10 Dec 2018 10:05:11 GMT):
As an aside, it's a bright sunny day here in Basel :-) ...a bit of cloud

mwherman2000 (Mon, 10 Dec 2018 10:05:11 GMT):
As an aside, it's a bright sunny day here in Basel :-) ...a bit of cloud ...interesting metaphor ;-)

manishcm (Tue, 11 Dec 2018 05:33:04 GMT):
@drummondreed This is regarding the Economic Policies of Sovrin Framework. Some countries like India, China, Russia to name a few, have banned cyrpto currencies. If Sovrin comes with it's own cyrptocurrency, then it can't be used in such countries that contribute to major population. Any thoughts on this?

HPlus (Tue, 11 Dec 2018 06:38:22 GMT):
Has joined the channel.

HPlus (Tue, 11 Dec 2018 06:38:45 GMT):
Hi there I am looking to put my farm produce on the blockchain to allow users to verify the source and authenticity. Can someone from hyperledger assist? How do I go about utilising hyperledger to get me a solution for users to scan the produce QR code and retrieve information about the source and other useful information?

HPlus (Tue, 11 Dec 2018 06:38:54 GMT):
Am in in the correct chat?

HPlus (Tue, 11 Dec 2018 06:38:54 GMT):
Am I in the correct chat?

swcurran (Tue, 11 Dec 2018 07:18:23 GMT):
@HPlus - probably not in the right channel. Supply-chain solutions have tended in the past to be built on private blockchains such as Hyperledger Fabric or Sawtooth, so you should look for those channels.

HPlus (Tue, 11 Dec 2018 08:31:19 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=WvAGG7kTyXBWSqJkT) @swcurran Okay thanks

Sean_Bohan (Tue, 11 Dec 2018 14:38:40 GMT):
*Thursday INDY WG Call:* Date: Thurs Dec 13 When: 8am PT, 9am MT, 10am CT, 11amET, 4pmGMT Where: https://zoom.us/j/232861185 Agenda: IXO Team shares their work and what they have been doing with INDY Previous Recordings and Agendas: https://drive.google.com/drive/u/0/folders/1AwHWN95KmSEi5fijraID0tFFMzYHoMwt?ogsrc=32 (edited)

vikpande (Tue, 11 Dec 2018 16:48:15 GMT):
Has joined the channel.

dbmath (Tue, 11 Dec 2018 20:51:40 GMT):
Has joined the channel.

nanmin (Tue, 11 Dec 2018 21:41:37 GMT):
Has joined the channel.

nanmin (Tue, 11 Dec 2018 21:46:15 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=YhPr73Qt7MQeE6hco) @tomislav @tomislav I also would greatly appreciate any details on integrating Indy and Fabric.

sam-kaw (Tue, 11 Dec 2018 21:59:38 GMT):
Has joined the channel.

yisheng (Wed, 12 Dec 2018 03:40:00 GMT):
Has joined the channel.

skaree (Wed, 12 Dec 2018 05:22:52 GMT):
Has joined the channel.

skaree (Wed, 12 Dec 2018 05:26:13 GMT):
hi...can someone please explain me the two types of DIDs Verinym and Pseudonym in simple terms?

mwherman2000 (Wed, 12 Dec 2018 06:59:31 GMT):
In Basel this week, I'm wearing a bright blue "Malta" cap. Looking forward to meeting everyone.

DuckLover (Wed, 12 Dec 2018 12:01:12 GMT):
Hello, from the Sovrin Paper i read different information on what public Keys/ DiDs are stored on the Ledger. Can anyone clarify if only the first public key and did is stored on the ledger or every public key /did that is used for direct interaction?

jadhavajay (Wed, 12 Dec 2018 14:19:08 GMT):
Has joined the channel.

vikpande (Wed, 12 Dec 2018 14:48:08 GMT):
@skaree ; this is gonna be a long explanation. Like explaining all simple things , its difficult to explain but i try my best. In DID as you know you can have two paths - Annonymity ( keeping the identity of the user annonymous - not known to anyone, more for one time use or pseudonymous - using zero knowledge proof for more than one time use) to verinimity ( able to verify the identity of the user using Trust Anchor - organizations issuing identity). So; in short, they are blinded of which Pseudonym is a part( you are not able to see the details of the user) and verifiable - read verinym ( you can verify the details fo the users). Do let me know if that helps .

vikpande (Wed, 12 Dec 2018 14:48:37 GMT):
@kdenhartog , can you please give your valuable inputs here . thanks.

vikpande (Wed, 12 Dec 2018 14:53:25 GMT):

DID_Sovrin.png

sbtmentor (Wed, 12 Dec 2018 14:59:32 GMT):
Has joined the channel.

xnopre (Wed, 12 Dec 2018 15:16:52 GMT):
Is it possible for an issuer to provide a credential with a date of end of validity (or a period of validity) ? Same question for a prover to provide a proof with a date of end of validity (or a period of validity) ?

DuckLover (Wed, 12 Dec 2018 15:44:09 GMT):
@vikpande can you link the whitepaper? Cant find that image in the sovrin library for papers

vikpande (Wed, 12 Dec 2018 16:01:57 GMT):
@DuckLover , pls see here https://sovrin.org/wp-content/uploads/2018/03/Sovrin-Provisional-Trust-Framework-2017-06-28.pdf

vikpande (Wed, 12 Dec 2018 16:02:09 GMT):
@ page 9 . Happy reading.

vikpande (Wed, 12 Dec 2018 16:20:27 GMT):
@DuckLover , maybe this paper is more relevant to your use case. https://www.evernym.com/wp-content/uploads/2017/07/What-Goes-On-The-Ledger.pdf . BTW, i like your username :-)

vikpande (Wed, 12 Dec 2018 21:29:37 GMT):
@DuckLover , thanks for asking this question. a good read/information for me too. so here is the answer to your question from the paper written 3 months ago- refer page 10 .

vikpande (Wed, 12 Dec 2018 21:30:17 GMT):

On Sovrin.png

vikpande (Wed, 12 Dec 2018 21:31:06 GMT):

Off Sovrin.png

vikpande (Wed, 12 Dec 2018 21:32:18 GMT):
Answer to your question - every public key /did that is used for direct interaction is stored. :-)

DuckLover (Wed, 12 Dec 2018 21:32:18 GMT):
im not sure about this. Refering to this every DID is on the ledger

DuckLover (Wed, 12 Dec 2018 21:32:19 GMT):
https://forum.sovrin.org/t/public-facing-dids-vs-private-pairwise-dids/243/5?u=eduarda

DuckLover (Wed, 12 Dec 2018 21:33:14 GMT):
https://forum.sovrin.org/t/public-facing-dids-vs-private-pairwise-dids/243/5

DuckLover (Wed, 12 Dec 2018 21:34:39 GMT):
everything else would make any sense or am i wrong? How do the other party check for the public key if the DID is not on the sovrin ledger?

vikpande (Wed, 12 Dec 2018 21:37:14 GMT):
so, if you want someone to check for your public key through verifiable credential then you keep it on ledger; else you keep it off ledger ( a choice you make based on your needs/use case)

DuckLover (Wed, 12 Dec 2018 21:38:30 GMT):
isnt the public key used to enable a secure communication? so in nearly every case the other party should check for it?

mgbailey (Wed, 12 Dec 2018 21:41:00 GMT):
Public meaning that the DID is an identifier for an entity that wants to issue credentials that are able to be verified against the ledger. Typically corporations, NGOs, governments, etc. Individuals will typically not be issuing credentials to others, so no need of a (possibly permanent) DID on a ledger for them. In fact, such a DID would allow cross correlation, which is undesirable for individuals.

DuckLover (Wed, 12 Dec 2018 21:42:04 GMT):
Also refering to https://sovrin.org/wp-content/uploads/2018/03/How-Sovrin-Works.pdf the public Key of both parties ist stored on the ledger, even Alice Key should be off-ledger because its is created for a direct communication between every party

mgbailey (Wed, 12 Dec 2018 21:42:23 GMT):
or if an individual does need such a DID, it would be used only for issuing credentials. Other private DIDs would be used for all other interactions

DuckLover (Wed, 12 Dec 2018 21:46:45 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=DuRzzjqy9NePzzeTb) @mgbailey As a regular user i dont have any DID on the Blockchain at all? With regular i mean, that i only interact with corperations

mgbailey (Wed, 12 Dec 2018 21:48:05 GMT):
No. When you connect to the corp, you will generate a new DID that you will share with the corp, and them only. They will keep track of it internally.

mgbailey (Wed, 12 Dec 2018 21:48:30 GMT):
There is no need for individuals to leave any trace on the ledger

DuckLover (Wed, 12 Dec 2018 21:50:48 GMT):
So is this answer https://forum.sovrin.org/t/public-facing-dids-vs-private-pairwise-dids/243/5 not correct, because this DID is only stored local on my device and at the corp?

mgbailey (Wed, 12 Dec 2018 21:51:43 GMT):
This DID generation is hidden from the individual by the software. It puts the DID and its private key in the individual's local wallet, and sends the DID and public verkey to the corp. It is then pulled from the local wallet when needed for comms with the corp.

mgbailey (Wed, 12 Dec 2018 21:54:21 GMT):
This is kinda old, and the thinking on this has been evolving. Let me look through it for a minute

mgbailey (Wed, 12 Dec 2018 21:55:00 GMT):
(old in a new-technology, start-up kind of way) :)

DuckLover (Wed, 12 Dec 2018 21:56:12 GMT):
At this point thank you very much for your help. That was my opinion that i had before reading the link above today. I am writing my bachelor thesis about sovrin and its sometimes really hard to check if 1 year old paper is still up-to-date

mgbailey (Wed, 12 Dec 2018 21:59:01 GMT):
The people in this thread are all authoritative, but Andy's comment is very outdated. No longer do we write all DIDs to the ledger. Most are not.

DuckLover (Wed, 12 Dec 2018 22:02:15 GMT):
OK. Thank you very much, i keep reading through the forum. Maybe i need your help sometimes again :sweat_smile:

mgbailey (Wed, 12 Dec 2018 22:02:52 GMT):
NP. I added a comment on that thread to help future wanderers.

mwherman2000 (Wed, 12 Dec 2018 22:11:20 GMT):
And all through the Congress Center, not a creature was stirring ...not even mouse ;-)

skaree (Thu, 13 Dec 2018 06:39:37 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=PszmwA8x4AWk2p7ci) @vikpande @vikpande thank you for that.by the way who generates the DIDs?is it by the Indy SDks?if so,then how these different kinds of DIDs are generated ?

vtech (Thu, 13 Dec 2018 07:11:22 GMT):
Hi Team, As currently Hyperledger Indy is in incubation . Are there any possible dates to for a stable release ? Thanks !

sairak (Thu, 13 Dec 2018 09:22:35 GMT):
Has joined the channel.

vikpande (Thu, 13 Dec 2018 09:53:58 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=QzbZYCYzgXaEASmpJ) @mgbailey :thumbsup:

vikpande (Thu, 13 Dec 2018 13:00:20 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=WXc4ZQDy2nLm4u9d4) @skaree Yes. @skaree, you are right. the DID's are generated by indy SDK. Indy uses "base58Check encoding" to geenrate these DID numbers.

vikpande (Thu, 13 Dec 2018 13:00:20 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=WXc4ZQDy2nLm4u9d4) @skaree Yes. @skaree, you are right. the DID's are generated by indy SDK. Indy uses "base58Check encoding" to geenrate these DID numbers.

vikpande (Thu, 13 Dec 2018 13:00:20 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=WXc4ZQDy2nLm4u9d4) @skaree Yes. @skaree, you are right. the DID's are generated by indy SDK. Indy uses "base58Check encoding" to generate these DID numbers. there is a lot happening under the hood which you can see in the indy github description like wallet creation, did creation, base58 and checksum using SHA256 algorithm.

abilashs (Thu, 13 Dec 2018 13:57:03 GMT):
Has joined the channel.

CHempel (Thu, 13 Dec 2018 16:19:53 GMT):
Has joined the channel.

markoprljic (Thu, 13 Dec 2018 16:43:49 GMT):
Has joined the channel.

ingo_so (Thu, 13 Dec 2018 17:13:04 GMT):
Has joined the channel.

esplinr (Thu, 13 Dec 2018 21:40:53 GMT):
@swcurran @sklump Does Hyperledger Indy Catalyst have GitHub repos yet, or are you still working in the BCGov OrgBook / VON repos?

sklump (Thu, 13 Dec 2018 21:40:53 GMT):
Has joined the channel.

swcurran (Fri, 14 Dec 2018 04:52:18 GMT):
@esplinr - still in bcgov and pspc repos. That's an exercise to come.

drummondreed (Fri, 14 Dec 2018 06:46:28 GMT):
@DuckLover @vikpande et al: I'm at the Hyperledger Global Forum in Basel and just caught up with this thread. The Sovrin Glossary, which originally was at the start of the Sovrin Provisional Trust Framework, is now a standalone document that is MUCH more comprehensive that the first one. We're going to maintain it as a Google doc so that every term is independently linkable. Here's the link: https://docs.google.com/document/d/1gfIz5TT0cNp2kxGMLFXr19x1uoZsruUe_0glHst2fZ8/edit?usp=sharing

drummondreed (Fri, 14 Dec 2018 06:47:48 GMT):
It's one of the documents that is slated to be approved by the Sovrin Board of Trustees at their year-end meeting on Dec 18th. You can see the full set of docs at https://sovrin.org/governance-framework/

vikpande (Fri, 14 Dec 2018 07:03:08 GMT):
thats good to hear @drummondreed . thanks for your inputs. Appreciate it.

Dan (Fri, 14 Dec 2018 09:54:23 GMT):
@ianco https://github.com/bcgov/von-agent-template/pull/22 is this good for :beers: ?

mwherman2000 (Fri, 14 Dec 2018 10:55:05 GMT):
If you're in Basel, the Indy@Basel ARM and NFE discussion from 1pm to 3pm has been moved to **BOSTON 3**. See you in about 1 hour.

Dan (Fri, 14 Dec 2018 11:39:15 GMT):
@ianco here's a hostname instead of an IP address https://github.com/bcgov/von-agent-template/pull/23/files

ianco (Fri, 14 Dec 2018 13:56:38 GMT):
Awesome thanks @Dan are you around to claim your bounty?

ianco (Fri, 14 Dec 2018 13:56:49 GMT):
Tested it out and everything works as advertised :-)

Dan (Fri, 14 Dec 2018 13:57:54 GMT):
Cool .. thanks for cleaning that up. I would like you to donate my bounty to needy orphans. :D

ianco (Fri, 14 Dec 2018 13:59:57 GMT):
OK I will, thanks!

abdfaye (Fri, 14 Dec 2018 14:31:55 GMT):
Has joined the channel.

axel (Fri, 14 Dec 2018 14:46:10 GMT):
@kdenhartog I see you are using AdBlock Plus on firefox, may I recommend switching to uBlock origin? ;)

axel (Fri, 14 Dec 2018 14:54:01 GMT):
just a thought as we just watched a video demo (the IBM loan demo) at the Hyperledger Global Forum workshop, in the demos, the convenience aspect is often put forth (ie: look at how easy it is to fill forms with valid data!), but the empowering aspect is not showcased or really not shown at all (ie: you can't refuse to share some of your information)

axel (Fri, 14 Dec 2018 14:54:45 GMT):
I think that aspect is going to be very important to demonstrate to show people how empowering SSI can be to them

esplinr (Fri, 14 Dec 2018 15:10:31 GMT):
@swcurran @sklump What are the pspc repos? I'm not familiar with them. I'm specifically looking for your implementation of credential exchange (vonanchor?) so that we can compare it with libvcx in the SDK. Can you provide a link?

swcurran (Fri, 14 Dec 2018 15:21:02 GMT):
Here is the von_anchor repo: https://github.com/PSPC-SPAC-buyandsell/von_anchor

swcurran (Fri, 14 Dec 2018 15:21:43 GMT):
There are a few others under that organization - Public Service and Procurement Canada. SPAC is the French acronym.

swcurran (Fri, 14 Dec 2018 15:30:40 GMT):
Use the ReadTheDocs link in there to get a summary of von_anchor - https://von-anchor.readthedocs.io/en/latest/

swcurran (Fri, 14 Dec 2018 15:34:34 GMT):
VON-X wraps von_anchor to provide a Credential Exchange REST interface for Issuer/Verifier Services.

swcurran (Fri, 14 Dec 2018 15:35:16 GMT):
VON-X docs - https://von-x.readthedocs.io/en/latest/ VONX repo - https://github.com/PSPC-SPAC-buyandsell/von-x

swcurran (Fri, 14 Dec 2018 15:36:57 GMT):
The docs for von-x aren't as complete.

swcurran (Fri, 14 Dec 2018 15:37:05 GMT):
Working on them.

swcurran (Fri, 14 Dec 2018 15:37:05 GMT):
Working on them. :-)

DuckLover (Fri, 14 Dec 2018 18:26:12 GMT):
Hello guys, i often read that Sovrin is a ledger and sometimes that it is a blockchain. I know that Blockchains are a subset of ledgers but is Sovrin explicit a blockchain or do people start to use it synonymous?

DuckLover (Fri, 14 Dec 2018 18:45:51 GMT):
Second question: I read that private Keys never leave the local wallet. How do you communicate with an corp. where you already created pairwise DID? Just creating a new pair of DID and proof trust all over again via transmitting Credentials?

TammyPlatero (Fri, 14 Dec 2018 19:56:36 GMT):
Is there a way to see the/a DID schema in the indy-sdk? I can see the generated DID from a wallet and json (a string) but is there a schema of some kind?

TammyPlatero (Fri, 14 Dec 2018 20:13:34 GMT):
Er, how do we add a Did document or create your own Did schema?

dbluhm (Sat, 15 Dec 2018 00:59:18 GMT):
I think Sovrin is explicitly a ledger but people are using blockchain and ledger interchangeably

dbluhm (Sat, 15 Dec 2018 00:59:18 GMT):
I think Sovrin is explicitly a ledger but people are using the terms blockchain and ledger interchangeably

dbluhm (Sat, 15 Dec 2018 00:59:27 GMT):
@DuckLover ^

dbluhm (Sat, 15 Dec 2018 01:00:22 GMT):
@TammyPlatero what are you referring to when you say "DID schema"? The term schema is kind of overloaded

dbluhm (Sat, 15 Dec 2018 01:04:35 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=hn9hD6L2PRfmXheNL) @DuckLover Pairwise connections allow you to continue communication as long as both parties want to. The secret keys never leaving the wallet means that the Indy SDK takes as a parameter a wallet handle for any operations requiring the secret key, pulling it up and using it but then properly wiping memory after use and making sure the secret key isn't used in inappropriate ways.

infominer33 (Sat, 15 Dec 2018 01:09:09 GMT):
ok, I got a question what is DKMS vs DPKI? they must be pretty closely related, but... itd be nice to know their precise dividing lines

infominer33 (Sat, 15 Dec 2018 01:10:26 GMT):
@TammyPlatero idk if this helps: https://github.com/sovrin-foundation/protocol/blob/master/themis/schema.md

infominer33 (Sat, 15 Dec 2018 01:10:26 GMT):
@TammyPlatero idk if this helps or is what you meant: https://github.com/sovrin-foundation/protocol/blob/master/themis/schema.md

infominer33 (Sat, 15 Dec 2018 01:37:29 GMT):
https://github.com/hyperledger/indy-sdk/blob/master/doc/getting-started/getting-started.md#step-5-credential-schemas-setup

sairak (Sat, 15 Dec 2018 06:37:58 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=R98c479F89eAhGHkB) @vikpande @vikpande...thank you for the reply...and what i intend to ask is as we have two kinds of Dids Verinym and Pseudonym are generated on what basis? suppose in the case of Alice use case,where she wants to get a Digital Transcript,she registers herself in the Faber College network and obtains a DID..in that case what kind of DID she gets?i am sorry if i that doubt is too simple::disappointed_relieved:

sairak (Sat, 15 Dec 2018 08:28:15 GMT):
also in order to get DID, Alice has to provide her personal details to the ledger...in that case ,how we can ensure security and privacy to data?

drummondreed (Sat, 15 Dec 2018 08:37:05 GMT):
It’s a very broad question but here’s an answer: DPKI is an architectural philosophy about the advantages of decentralized PKI. The term was created at the first Rebooting Web of Trust workshop in San Francisco about 4 years ago. Here’s a link to the paper: https://github.com/WebOfTrustInfo/rebooting-the-web-of-trust/blob/master/draft-documents/Decentralized-Public-Key-Infrastructure-CURRENT.md [ ](https://chat.hyperledger.org/channel/indy?msg=nFeZ5ByqJBSQqS6Sh)

drummondreed (Sat, 15 Dec 2018 08:40:21 GMT):
DPKI is not a spec about how to actually implement decentralized PKI. So the next step was to start to create the specs to actually implement it. The first one was DIDs (Decentralized Identifiers). https://w3c-ccg.github.io/did-spec/

drummondreed (Sat, 15 Dec 2018 08:42:37 GMT):
I also recommend the DID Primer for an easy overview of DIDs: https://w3c-ccg.github.io/did-primer/

drummondreed (Sat, 15 Dec 2018 08:46:22 GMT):
Now, once you have DIDs, the biggest remaining problem is now to do the decentralized key management that you need for decentralized identifiers. Like with DIDs, it demands an open standard spec that any digital wallet vendor can implement. That spec came to be called DKMS (Decentralized Key Management System). The first step towards creating that spec was a year-long project that Evernym carried out under a research contract from the U.S. Dept of Homeland Security. The resulting design and architecture was published last March in the Hyperledger Indy repo: https://github.com/hyperledger/indy-sdk/blob/master/doc/dkms/DKMS%20Design%20and%20Architecture%20V3.md

drummondreed (Sat, 15 Dec 2018 08:47:12 GMT):
The next step is to formally submit DKMS to a standards body (OASIS is the leading choice). DM me if you’re interested in being part of that.

chipleinen (Sat, 15 Dec 2018 08:51:55 GMT):
Has joined the channel.

vikpande (Sat, 15 Dec 2018 09:02:28 GMT):
Yes please @drummondreed. Would like to be engaged in indy as much as I could. Pls include me. Thank you for asking.

vikpande (Sat, 15 Dec 2018 09:02:28 GMT):
Yes please @drummondreed . Would like to be engaged in indy as much as I could. Pls include me. Thank you for asking.

mwherman2000 (Sat, 15 Dec 2018 13:20:49 GMT):
Floating the idea of DIRL (Digital Identity Resource Locators) being used to retrieve a DID Document ...which internally contain a DID Document Identifier. The nickname "DID" is only used as a reference to a DID Document. @drummondreed

mwherman2000 (Sat, 15 Dec 2018 13:20:49 GMT):
Floating the idea of DIRL (Digital Identity Resource Locators) being used to retrieve a DID Document ...which internally contain a DID Document Identifier. The nickname "DID" is only used as a reference to a DID Document. @drummondreed @nage

mwherman2000 (Sat, 15 Dec 2018 13:20:49 GMT):
Floating the idea of a DIRL (Digital Identity Resource Locators) being used to retrieve a DID Document ...which internally contain a DID Document Identifier. The nickname "DID" is only used as a reference to a DID Document. @drummondreed @nage

mwherman2000 (Sat, 15 Dec 2018 13:20:49 GMT):
Floating the idea of a DIRL (Digital Identity Resource Locator) being used to retrieve a DID Document ...which internally contain a DID Document Identifier. The nickname "DID" is only used as a reference to a DID Document. @drummondreed @nage

mwherman2000 (Sat, 15 Dec 2018 13:20:49 GMT):
Floating the idea of a DIRL (Digital Identity Resource Locator) being used to retrieve a DID Document ...which, internally, contain a DID Document Identifier. The nickname "DID" is only used as a reference to a DID Document. @drummondreed @nage

mwherman2000 (Sat, 15 Dec 2018 13:20:49 GMT):
Floating the idea of a DIRL (Digital Identity Resource Locator) being used to retrieve a DID Document ...which, internally, contain a DID Document Identifier. The nickname "DID" is only used as a reference to a DID Document. @drummondreed @nage @danielhardman

mwherman2000 (Sat, 15 Dec 2018 13:20:49 GMT):
Floating the idea of a DIRL (Digital Identity Resource Locator) being used to retrieve a DID Document ...which, internally, contains a DID Document Identifier. The nickname "DID" is only used as a reference to a DID Document. @drummondreed @nage @danielhardman

mwherman2000 (Sat, 15 Dec 2018 13:20:49 GMT):
Floating the idea of a DIRL (Digital Identity Resource Locator) being used to retrieve a DID Document ...which, internally, contains a DID Document Identifier. The nickname "DID" is only used as a reference to a DID Document. @drummondreed @nage @danielhardman ...your thoughts?

mwherman2000 (Sat, 15 Dec 2018 13:20:49 GMT):
Floating the idea of a DIRL (Decentralized Identity Resource Locator) being used to retrieve a DID Document ...which, internally, contains a DID Document Identifier. The nickname "DID" is only used as a reference to a DID Document. @drummondreed @nage @danielhardman ...your thoughts?

mwherman2000 (Sat, 15 Dec 2018 13:29:15 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=WrAKmTEhDkNZafJJN) Question: Is it required/desirable/not desirable for a DIRL for a DID be required to include the DID Document Identifier (aka "DID Identifier) for the DID?

mwherman2000 (Sat, 15 Dec 2018 13:29:15 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=WrAKmTEhDkNZafJJN) Question: Is it required/desirable/not desirable for a DIRL for a DID be required to include the DID Document Identifier (aka "DID Identifier) for the DID? [serious question]

mwherman2000 (Sat, 15 Dec 2018 13:29:15 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=WrAKmTEhDkNZafJJN) Question: Is it required/desirable/not desirable for a DIRL for a DID be required to include the DID Document Identifier (aka "DID Identifier) for the DID? [serious question] ...or can the DID be retrieved using some other (unique) identifier string?

mwherman2000 (Sat, 15 Dec 2018 13:29:15 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=WrAKmTEhDkNZafJJN) Question: Is it required/desirable/not desirable for a DIRL for a DID be required to include the DID Document Identifier (aka "DID Identifier") for the DID? [serious question] ...or can the DID be retrieved using some other (unique) identifier string?

mwherman2000 (Sat, 15 Dec 2018 13:29:15 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=WrAKmTEhDkNZafJJN) Question: Is it required/desirable/not desirable for a DIRL for a DID be required to include the DID Document Identifier (aka "DID Identifier") for the DID? [serious question] ...or can the DID be retrieved using some other (unique) identifier string in the DIRL?

mwherman2000 (Sat, 15 Dec 2018 16:07:13 GMT):
Floating the idea of a "Digital Trust Assistant" ...a category name for mobile apps that include a mobile UX, mobile agent, and mobile wallet. Your thoughts?

mwherman2000 (Sat, 15 Dec 2018 16:07:13 GMT):
Floating the idea of a "Trusted Digital Assistant" ...a category name for mobile apps that include a mobile UX, mobile agent, and mobile wallet. Your thoughts?

mwherman2000 (Sat, 15 Dec 2018 16:07:13 GMT):
Floating the idea of a "Trusted Digital Assistant" ...a category name for mobile apps that include a mobile UX, mobile AI, mobile SSI agent, and mobile SSI wallet. Your thoughts?

infominer33 (Sat, 15 Dec 2018 20:47:11 GMT):
so, if I understand correctly: DPKI broadly speaks of DID\AUTH\Credentials\DKMS which are all specific methods that will work together in the quest for DPKI?

infominer33 (Sat, 15 Dec 2018 20:47:11 GMT):
so, if I understand correctly: DPKI broadly speaks of DID\AUTH\Credentials\DKMS which are all specific methods that will work together in the quest for DPKI

infominer33 (Sat, 15 Dec 2018 20:47:11 GMT):
so, if I understand correctly: DPKI broadly speaks of DID\AUTH\Credentials\DKMS which are all specific methods that will work together in the quest for DPKI DKMS relates to hubs and agents

infominer33 (Sat, 15 Dec 2018 20:47:45 GMT):
thank you for the thorough response, btw

infominer33 (Sat, 15 Dec 2018 20:57:53 GMT):
also, while I'm asking questions are did auth and universal resolver related? I keep feeling like they are, but haven't found documentation confirm or deny my suspicion

infominer33 (Sat, 15 Dec 2018 21:46:48 GMT):
and is there a significant difference between verifiable claims vs verifiable credentials?

mwherman2000 (Sun, 16 Dec 2018 07:39:38 GMT):
Floating the idea of #TheTrustedDigitalWeb... #leadership #innovation #blockchain #technology #banking #foodforthought #cool #switzerland #sharepoint #reputation #wordsofwisdom #digitalidentity #trustworthy #indy #fabric-gossip

mwherman2000 (Sun, 16 Dec 2018 07:39:53 GMT):

TheTrustedDigitalWeb.png

mwherman2000 (Sun, 16 Dec 2018 07:47:01 GMT):
...the above is inspired by some of the most diversely knowledgeable and experienced minds that I had the privilege to meet in Basel this past week. Thank you!

mwherman2000 (Sun, 16 Dec 2018 08:21:13 GMT):
@mtfk Hi

mwherman2000 (Sun, 16 Dec 2018 08:28:07 GMT):
@mtkf and I just had our own mini-summit this morning here at the Basel airport and he easily convinced me that DIRL is not the right concept because it is refers to a Resource Location ...not appropriate in a decentralized world ...but we were easily able to arrive a different (and,I think better sounding) concept: a DRI ...Decentralized Resource Identifier. The scheme remains the same (aka "did") but the string things we now call DRIs and they resolve to DID Documents. Your thoughts?

mwherman2000 (Sun, 16 Dec 2018 08:28:07 GMT):
@mtkf and I just had our own mini-summit this morning here at the Basel airport and he easily convinced me that DIRL is not the right concept because it is refers to a Resource Location ...not appropriate in a decentralized world ...but we were easily able to arrive a different (and,I think better sounding) concept: a DRI ...Decentralized Resource Identifier. The scheme remains the same (aka "did") but the string things we are called DRIs and they resolve to DID Documents. Your thoughts?

mwherman2000 (Sun, 16 Dec 2018 08:28:07 GMT):
@mtfk and I just had our own mini-summit this morning here at the Basel airport and he easily convinced me that DIRL is not the right concept because it is refers to a Resource Location ...not appropriate in a decentralized world ...but we were easily able to arrive a different (and,I think better sounding) concept: a DRI ...Decentralized Resource Identifier. The scheme remains the same (aka "did") but the string things we are called DRIs and they resolve to DID Documents. Your thoughts?

mwherman2000 (Sun, 16 Dec 2018 08:28:07 GMT):
@mtfk and I just had our own mini-summit this morning here at the Basel airport and he easily convinced me that DIRL is not the right concept because it is refers/implies a Resource Location ...not appropriate in a decentralized world ...but we were easily able to arrive a different (and,I think better sounding) concept: a DRI ...Decentralized Resource Identifier. The scheme remains the same (aka "did") but the string things we are called DRIs and they resolve to DID Documents. Your thoughts?

mwherman2000 (Sun, 16 Dec 2018 08:28:07 GMT):
@mtfk and I just had our own mini-summit this morning here at the Basel airport and he easily convinced me that DIRL is not the right concept because it is refers to/implies a Resource Location ...not appropriate in a decentralized world ...but we were easily able to arrive a different (and,I think better sounding) concept: a DRI ...Decentralized Resource Identifier. The scheme remains the same (aka "did") but the string things we are called DRIs and they resolve to DID Documents. Your thoughts?

mwherman2000 (Sun, 16 Dec 2018 08:28:07 GMT):
@mtfk and I just had our own mini-summit this morning here at the Basel airport and he easily convinced me that DIRL is not the right concept because it is refers to/implies a Resource Location ...not appropriate in a decentralized world ...but we were easily able to arrive a different (and,I think better sounding) concept: a DRI ...Decentralized Resource Identifier. The scheme remains the same (aka "did") but we are now calling the string things DRIs and they resolve to DID Documents. Your thoughts?

mwherman2000 (Sun, 16 Dec 2018 08:28:07 GMT):
@mtfk and I just had our own mini-summit this morning here at the Basel airport and he easily convinced me that DIRL is not the right concept because it is refers to/implies a Resource Location ...not appropriate in a decentralized world ...but we were easily able to arrive a different (and,I think better sounding) concept: a DRI ...Decentralized Resource Identifier. The scheme remains the same (aka "did") but we are now calling the string things DRIs and they, by default, resolve to DID Documents. Your thoughts?

mwherman2000 (Sun, 16 Dec 2018 08:30:04 GMT):

TheTrustedDigitalWeb v0.2.png

mwherman2000 (Sun, 16 Dec 2018 08:30:04 GMT):

TheTrustedDigitalWeb v0.2.png

mwherman2000 (Sun, 16 Dec 2018 08:32:10 GMT):
...the above is inspired by some of the most diversely knowledgeable and experienced minds that I had the privilege to meet in Basel this past week. Thank you!

mwherman2000 (Sun, 16 Dec 2018 08:46:10 GMT):
@mtfk Maybe we need a "didp" scheme in additional to the "did" schema for DRIs? ...and we shouldn't be overloaded one scheme (i.e. did) for two different purposes?

mwherman2000 (Sun, 16 Dec 2018 08:46:10 GMT):
@mtfk Maybe we need a "didp" scheme in additional to the "did" scheme (for DRIs)? ...and we shouldn't be overloaded one scheme (i.e. did) for two different purposes

mwherman2000 (Sun, 16 Dec 2018 08:46:10 GMT):
@mtfk Maybe we need a "didp" scheme in addition to the "did" scheme (for DRIs)? ...and we shouldn't be overloaded one scheme (i.e. did) for two different purposes

mwherman2000 (Sun, 16 Dec 2018 08:46:10 GMT):
@mtfk Maybe we need a "didp" scheme in addition to the "did" scheme (for DRIs)? ...thst is, we shouldn't be overloaded one scheme (i.e. did) for two different purpos

mwherman2000 (Sun, 16 Dec 2018 08:46:10 GMT):
@mtfk Maybe we need a "didp" scheme in addition to the "did" scheme (for DRIs)? ...thst is, we shouldn't be overloading one scheme (i.e. did) for two different purposes.

mwherman2000 (Sun, 16 Dec 2018 08:46:10 GMT):
@mtfk Maybe we need a "didp" scheme in addition to the "did" scheme (for DRIs)? ...that is, we shouldn't be overloading one scheme (i.e. did) for two different purposes.

mwherman2000 (Sun, 16 Dec 2018 08:46:43 GMT):
did:sov:1234abcd cand be the DRI (decentralized resource identifier) ...for a DID Document

mwherman2000 (Sun, 16 Dec 2018 08:46:43 GMT):
did:sov:1234abcd can be the DRI (decentralized resource identifier) ...for a DID Document

mwherman2000 (Sun, 16 Dec 2018 08:47:48 GMT):
didp:///did:sov:1234abcd is an example for the locationless protocol for retrieving the DID Document for did:sov:1234abcd

mwherman2000 (Sun, 16 Dec 2018 08:47:48 GMT):
didp:///did:sov:1234abcd is an example of the locationless protocol for retrieving the DID Document for did:sov:1234abcd

mwherman2000 (Sun, 16 Dec 2018 08:49:58 GMT):
didp:///did:sov:1234abcd/schema or, alternatively, didp:///did:sov:1234abcd?schema or ...something else could be the protocol for retrieving the Schema Document for DID Document did:sov:1234abcd. Your thoughts?

mwherman2000 (Sun, 16 Dec 2018 08:49:58 GMT):
didp:///did:sov:1234abcd/schema or, alternatively, didp:///did:sov:1234abcd?schema or ...something else ...could be the protocol for retrieving the Schema Document for DID Document did:sov:1234abcd. Your thoughts?

mwherman2000 (Sun, 16 Dec 2018 08:49:58 GMT):
didp:///did:sov:1234abcd/schema ...or, alternatively, didp:///did:sov:1234abcd?schema or ...something else ...could be the protocol for retrieving the Schema Document for DID Document did:sov:1234abcd. Your thoughts?

pknowles (Sun, 16 Dec 2018 11:28:43 GMT):
@mwherman2000 @mtfk - Rest easy, gents! If you need to tweak anything substantial, just keep me posted so that I can keep an eye on the semantics modelling.

pknowles (Sun, 16 Dec 2018 11:28:43 GMT):
@mwherman2000 @mtfk - Rest easy, gents! If you need to tweak anything substantial, just keep me posted so that I can keep an eye on things re the capture model. You'll probably just need an entry and a source overlay somewhere in the high-level cake. Over to the tech gurus.

pknowles (Sun, 16 Dec 2018 11:28:43 GMT):
@mwherman2000 @mtfk - You'll probably just need an entry and a source overlay somewhere in the high-level cake. Over to the tech gurus for their input.

pknowles (Sun, 16 Dec 2018 11:28:43 GMT):
@mwherman2000 @mtfk - You'll probably just need an *entry overlay* and a *source overlay* somewhere in the high-level cake. Over to the tech gurus for their input. From my end, no issue.

pknowles (Sun, 16 Dec 2018 11:28:43 GMT):
@mwherman2000 @mtfk - You'll probably just need an *entry overlay* and a *source overlay* somewhere in the high-level cake. From my side, I don't envisage an issue. Over to the tech gurus for their input.

pknowles (Sun, 16 Dec 2018 11:28:43 GMT):
@mwherman2000 @mtfk - You'll probably just need an *entry overlay* and a *source overlay* somewhere in the high-level cake. From my side, I it should be pretty straightforward. Over to the tech gurus for their input.

pknowles (Sun, 16 Dec 2018 11:28:43 GMT):
@mwherman2000 @mtfk - You'll probably just need an *entry overlay* and a *source overlay* somewhere in the high-level cake. From a capture perspective, I it should be pretty straightforward. Over to the tech gurus for their input.

pknowles (Sun, 16 Dec 2018 11:28:43 GMT):
@mwherman2000 @mtfk - You'll probably just need an *entry overlay* and a *source overlay* somewhere in the high-level cake. From a capture perspective, it should be pretty straightforward. Over to the tech gurus for their input.

drummondreed (Sun, 16 Dec 2018 12:40:40 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=FmjTyCiC2b7priJWP) I disagree—not because @mtfk does not have a point, but because there's a different way to look at it. A DID is *by definition* decentralized. But a DIRL (see note below RE naming) actually DOES define the location of a resource. Using a simplified pseudocode: DIRL = DID + ServiceID + ResourceID (the actual syntax is did:method:specificid;serviceid/path?query#frag where every component except did:method:specificid is optional). So what happens is that while the DID is decentralized (by design), the ServiceID component maps the DIRL to a specific service endpoint URL—which *is* a resource locator—and then adds the optional path?query#fragment components which at that point *are* also resource locators.

drummondreed (Sun, 16 Dec 2018 12:42:46 GMT):
Net net: what I like about the term "DIRL" is that it's a simple, catchy term for the combination of a DID that is a true decentralized identifier and a URL that is a true uniform resource locator.

drummondreed (Sun, 16 Dec 2018 12:43:39 GMT):
Which finally brings me to my point that DID + URL seems like it would more logically combine into "DURL" (Decentralized URL).

drummondreed (Sun, 16 Dec 2018 12:44:54 GMT):
As someone who is constantly having to refer to "the way you can turn a DID into a URL" (which currently in the spec is called a "DID reference"), I really like the idea of a simple, catchy acronym for it: DURL.

drummondreed (Sun, 16 Dec 2018 12:45:51 GMT):
Major kudos to @mwherman2000 for the light bulb for that term.

swcurran (Sun, 16 Dec 2018 15:02:19 GMT):
I agree @drummondreed - we don't need another standard - the DID standard covers it. dURL might be useful to mean "the concrete URL formed by resolving the DID and applying the DID path fragment to the DID endpoint". If nothing else, it's a reminder that the capability is there.

swcurran (Sun, 16 Dec 2018 15:02:19 GMT):
I agree @drummondreed - we don't need another standard - the DID standard covers it. dURL might be useful to mean "the concrete URL formed by resolving the DID and applying the DID path fragment to the DID endpoint". If nothing else, it's a reminder that the capability is there already, in the standard.

swcurran (Sun, 16 Dec 2018 15:05:19 GMT):
FYI - an example of the use of this is in the Message Type indy-hipe - notably in this section (but you might want to read the whole HIPE for context) - https://github.com/hyperledger/indy-hipe/tree/master/text/0021-message-types#service-selection-by-id

drummondreed (Sun, 16 Dec 2018 15:06:13 GMT):
@swcurran Completely agreed that it's not anything new in terms of functionality—just an catchy new name for the functionality that was already there. I personally believe this capability to root conventional URLs on DIDs in order to make them domain-independent and persistent for very long periods will become very popular for certain types of resource references.

jljordan_bcgov (Sun, 16 Dec 2018 16:18:50 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=yuuwtZSCAHmfQWa77) @infominer33 We have a reasonable overview of SSI, DID, Verifiable Credentials so forth here https://vonx.io/getting_started/vons-blockchain-basis/ .. some folks my find it helpful ( @sairak @vikpande )

infominer33 (Sun, 16 Dec 2018 16:21:07 GMT):
hey! that's a pretty good page @jljordan_bcgov

infominer33 (Sun, 16 Dec 2018 16:35:11 GMT):
@jljordan_bcgov VON now has its own section on the /awesome-indy page. will you check it out and let me know if there's anything you think is missing? https://github.com/infominer33/awesome-decentralized-id/tree/master/awesome-indy#bcgovs-verifiable-organizations-network-von

infominer33 (Sun, 16 Dec 2018 16:37:20 GMT):
that entire page got much more orderly and useful for indy folk since yesterday

vikpande (Sun, 16 Dec 2018 17:34:55 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=b4H4oYM7jhbhYXrue) @jljordan_bcgov :thumbsup:

mtfk (Sun, 16 Dec 2018 22:00:23 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=doZFCeKSnpTyum2xy) @drummondreed The problem which we are trying to address with "renaming" stuff is to figure out how we can solve problem with DID as a unit of control. Decentralize identifier is something which should serve in any use case is very generic and fits very well in multiple cases (for example by simply changing main context of DID Doc). The problem is that in the current implementation/spec is tied to the public key. If we would like to use DID for something else then unit of control like schema (unit of language) we entering very dangers space as we are mixing control with language. We had multiple discussions during HL Global Forum about that problem end even that I am not fully understand all consequences yet I kind of got the point. I think is very important to distinguishes between DID as a unit of control from the rest use cases where not necessary we want to tied to someone key. Language (schema + overlays) are good example were nobody should have control over that what we can or can't speak about. But still we would like to use DID as a generic decentralize identifier to resolve content behind it. Does that have any sens?

mtfk (Sun, 16 Dec 2018 22:02:16 GMT):
To avoid confusion with naming I started experimenting with the schema for the URN like: ``` did-schema:sov:1231415124151 did-overlay:sov:13215512312 did-schema-element:sov:5814108410924 ```` But I think everyone would agree that this is not the way to go. Personally I like the concept of reusing main @context in DID document to tell resolver what type of the content it is. Similar to the idea with MIME Type in the document resolved via HTTP/S

mtfk (Sun, 16 Dec 2018 22:06:02 GMT):
So as soon as the I will try to resolve `did:sov:8291239123` I check the @context of the document to figure out with what I have to deal and for example it could be: ``` @context: ["https://domain.com/language/v1"] @context: ["https://domain.com/control/v1"] @context: ["https://domain.com/..../v1"] ``` I can't remember what other type of units we were talking about during the conf but those two are the most needed right now control is relation between did and public key, where language is did and data schema/language - something which describe how to talk not who controls it.

mtfk (Sun, 16 Dec 2018 22:10:08 GMT):
Maybe @nage could comment on that as he had very nice arguments about whole topic. As well @nage if you could remind me what other "units" you mentioned about, for sure unit of control and unit of language that stuck in my head but I know that there was more. Would help to look on that problem from a bigger perspective.

dznz (Mon, 17 Dec 2018 01:25:21 GMT):
Has joined the channel.

pknowles (Mon, 17 Dec 2018 01:57:34 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=37gmiNAnqEXCpMZvN) @mtfk `did-overlay:sov:13215512312` is like me and @drummondreed shaking hands! Probably not the solution for scalability though ... the ID, not the friendship!

pknowles (Mon, 17 Dec 2018 01:57:34 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=37gmiNAnqEXCpMZvN) @mtfk `did-overlay:sov:13215512312` is like me and @drummondreed shaking hands! Probably not the best solution for scalability though ... the ID, not the friendship!

pknowles (Mon, 17 Dec 2018 01:57:34 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=37gmiNAnqEXCpMZvN) @mtfk `did-overlay:sov:13215512312` is like me and @drummondreed shaking hands! Probably not the best solution for scalability though ... the ID character length, not the friendship!

DuckLover (Mon, 17 Dec 2018 08:36:13 GMT):
Hello, can someone explain or provide some recent information about the usage of microledgers? Regards

donqui (Mon, 17 Dec 2018 10:26:23 GMT):
@DuckLover have you read this HIPE and the documents it refers to https://github.com/hyperledger/indy-hipe/pull/31 ?

glauserr (Mon, 17 Dec 2018 11:19:27 GMT):
Has joined the channel.

glauserr (Mon, 17 Dec 2018 12:31:43 GMT):
hello, question to the roadmap of the indy project. When will it be ready for production? And at which point will the DID/DIO standards and W3C standards for verifiable credentials be implemented? all best

vikpande (Mon, 17 Dec 2018 13:54:33 GMT):
@glauserr , maybe @drummondreed could help us with this question

mxs1491 (Mon, 17 Dec 2018 14:58:19 GMT):
@infominer33 Found a couple more resources that may be helpful for your page: 1. From the New America foundation, a pretty good, non-technical review of SSI https://www.newamerica.org/future-property-rights/reports/nail-finds-hammer/introduction 2. A recent report from World Economic Forum on Digital Identity http://www3.weforum.org/docs/WEF_INSIGHT_REPORT_Digital%20Identity.pdf

Sean_Bohan (Mon, 17 Dec 2018 15:02:50 GMT):
This week: *Thursday INDY WG Call:* Date: Thurs Dec 20 When: 8am PT, 9am MT, 10am CT, 11amET, 4pmGMT Where: https://zoom.us/j/232861185 Agenda: Sami presents Sovrin Connector App demo Hyperledger Global Forum recap (all) Previous Recordings and Agendas: https://drive.google.com/drive/u/0/folders/1AwHWN95KmSEi5fijraID0tFFMzYHoMwt?ogsrc=32 (edited) NOTE: No call next week 12/24

DuckLover (Mon, 17 Dec 2018 15:14:15 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=r8Cfz2BXPTPu5Yygc) @donqui Thank you. I will check it out. Im getting a little bit more in the Code and working through the Getting Started (https://github.com/hyperledger/indy-sdk/blob/master/doc/getting-started/getting-started.md). Maybe someone can help me understand how the Steward knows the Endpoint of Faber when establishing the connection and writing Fabers DID and VerKey to the ledger. "At this point Faber is connected to the Steward and can interact in a secure peer-to-peer way. Faber can trust the response is from Steward because: it connects to the current endpoint ..."

dnnn (Mon, 17 Dec 2018 16:25:31 GMT):
Hi everyone, I have a question regarding `plugins`. As far as I've seen in `indy-plenum` documentation there is a possibility to extend the set of ledgers and transactions with the help of plugins. I have 2 questions: 1) Does the mechanism described for `indy-plenum` works for `indy-node`. Is there any differences in implementation of plugins? 2) If the plugin is used to add some additional transaction, how I can send such a transaction from `indy-sdk`? Is there a way to write plugin for `indy-sdk` or the only way would be to fork and expand the functionality in the code? Any replies or hints would be highly appreciated

infominer33 (Mon, 17 Dec 2018 17:14:06 GMT):
good picks @mxs1491

mwherman2000 (Mon, 17 Dec 2018 18:33:15 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=doZFCeKSnpTyum2xy) @drummondreed When you refer to DID what are you referring to?? ...the string identifier? ...or the document it links to?

jljordan_bcgov (Mon, 17 Dec 2018 18:35:23 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=5ycduwoTmBY3q2x3y) @mwherman2000 It is assumed to be both

jljordan_bcgov (Mon, 17 Dec 2018 18:35:39 GMT):
A DID without the associated document is not of any value

jljordan_bcgov (Mon, 17 Dec 2018 18:38:32 GMT):
There is a discussion on that here in a w3c mailing list

jljordan_bcgov (Mon, 17 Dec 2018 18:42:08 GMT):
https://lists.w3.org/Archives/Public/public-credentials/2018Dec/0055.html

mwherman2000 (Mon, 17 Dec 2018 18:46:07 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=smNTyDqCH3MKHZutp) @swcurran I believe we need to separate 2 concepts (and not overload them on to a single idea): 1. There's the need for an "identifier spec" ...the string thing, and 2. logically related but separate is the "protocol spec" for universally interacting with DID Documents ... attributes, schema, keys, etc. etc. Continuing to overload these onto one idea will be forever confuding to people new to Indy (I was trying to type confusing but like the new word better 🙂).

mwherman2000 (Mon, 17 Dec 2018 18:46:07 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=smNTyDqCH3MKHZutp) @swcurran I believe we need to separate 2 concepts (and not overload them on to a single idea): 1. There's the need for an "identifier spec" ...the string thing, and 2. logically related but separate is the "protocol spec" for universally interacting with decentralized resources (DID Documents being one physical manifestation of a DR) ... attributes, schema, keys, etc. etc. Continuing to overload these onto one idea will be forever confuding to people new to Indy (I was trying to type confusing but like the new word better 🙂).

mwherman2000 (Mon, 17 Dec 2018 18:48:58 GMT):
Secondarily, I'm proposing DRI (decentralized resource identifier) and did: as the name and protocol, respectively, for the string identifier, and...

mwherman2000 (Mon, 17 Dec 2018 18:52:59 GMT):
...and, (perhaps) DRAP (decentralized resource access protocol) and drap: (or didp:) as the name and protocol for interacting with DID Documents ...or something like that

mwherman2000 (Mon, 17 Dec 2018 18:58:40 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=ecMujEBHJhuJvxvCK) @jljordan_bcgov That's a part of what I and I think Robert are trying to address. It can't be both. It's confuding ...to use a new word

jljordan_bcgov (Mon, 17 Dec 2018 19:00:09 GMT):
water under the bridge methinks

mwherman2000 (Mon, 17 Dec 2018 19:01:18 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=5oZjeGazyFGKws5fH) @jljordan_bcgov The "DID" specification isn't final

jljordan_bcgov (Mon, 17 Dec 2018 19:02:59 GMT):
i suppose .. I'm not a standards definition guy so I'll just bow out of the discussion ...

jljordan_bcgov (Mon, 17 Dec 2018 19:03:11 GMT):
i prefer clicky things :)

mwherman2000 (Mon, 17 Dec 2018 19:04:31 GMT):
Confuding: synonyms confusing. Spreading fear, uncertainty and doubt in the minds of new project participants. "The specification was confuding to most new developers."

infominer33 (Mon, 17 Dec 2018 19:04:59 GMT):
confounding is also an enjoyable word

infominer33 (Mon, 17 Dec 2018 19:05:21 GMT):
< word nerd ;)

infominer33 (Mon, 17 Dec 2018 19:06:09 GMT):
tbh, I'm still trying to DRAP my head around this

mwherman2000 (Mon, 17 Dec 2018 19:06:41 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=BJrWuCiuxfJDn9caq) @jljordan_bcgov ...with one large meta button? ...or a reasonable number of buttons necessary to address the task at hand? 😉🙂🙃

mwherman2000 (Mon, 17 Dec 2018 19:10:11 GMT):
Now to let this perculote through the 27 world timezones 🙃

mwherman2000 (Mon, 17 Dec 2018 19:22:23 GMT):
If DRIs and DRAP can be defined in a single document that would be awesome ...but that's a decision best delegated to the W3C governance processes.

dbluhm (Mon, 17 Dec 2018 20:48:33 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=t2ySbHTBDeuBJaa5G) @DuckLover This is bumping into the agent protocol which is in active discussion in the form of HIPEs (Hyperledger Indy Project Enhancements, basically an RFC) at https://github.com/hyperledger/indy-hipe , in weekly calls, and in #indy-agent

mogamboizer (Mon, 17 Dec 2018 20:49:23 GMT):
Has joined the channel.

drummondreed (Tue, 18 Dec 2018 13:47:19 GMT):
Guys, sorry, I've been tied up in Madrid and have fallen behind on this thread. One thing I can easily clarify, since it's straight out of the spec: *DID* always refers to the *decentralized identifier*, period. If you want to talk about the JSON-LD document that the DID resolution process returns, that's called the *DID document*. The DID document contains a copy of the DID itself, but the *DID document is not the DID* and the *DID is not the DID document*.

greg2git (Tue, 18 Dec 2018 13:49:47 GMT):
Has left the channel.

drummondreed (Tue, 18 Dec 2018 13:50:10 GMT):
Now, the whole question that was raised in the last conversations in Basel about using DIDs to address OTHER types of objects on a ledger besides a DID document is *hugely important*. Excellent discussions so far. I'm cooking on it but have to prepare for my final meetings in Madrid now. I have a 9 hour flight from Madrid to Miami tomorrow so I plan to do a writeup on it then.

esplinr (Tue, 18 Dec 2018 14:24:20 GMT):
@Sean_Bohan I believe that you are canceling the Indy WG call on Thursday the 27th. Is that correct?

Sean_Bohan (Tue, 18 Dec 2018 15:06:47 GMT):
there is no Indy WG call on 12/27 - we return to our regularly scheduled indy WG call on 1/3/19

mwherman2000 (Tue, 18 Dec 2018 15:37:11 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=EG4bSzomLLoNokbt4) @drummondreed So Drummond, then you'll agree that it doesn't make sense for anyone to talk about a DID having keys, service endpoints, etc. ...if a DID is just a string identifier? In addition, your important clarification suggests the spec perhaps needs to call out the concept of a DID Entity ...the concept that aggregates the DID and the DID Document ...the latter being a serialization of a DID Entity. The former being the identifier or locator for a DID Entity (and it's serialization). Your thoughts?

mwherman2000 (Tue, 18 Dec 2018 15:37:11 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=EG4bSzomLLoNokbt4) @drummondreed So Drummond, then you'll agree that it doesn't make sense for anyone to talk about a DID having keys, service endpoints, etc. ...if a DID is just a string identifier? In addition, your important clarification suggests the spec perhaps needs to call out the concept of a DID Entity (or ...perhaps Decentralized Resource is a better term) ...the concept that aggregates the DID and the DID Document ...the latter being a serialization of a DID Entity. The former being the identifier or locator for a DID Entity (and it's serialization). Your thoughts?

jljordan_bcgov (Tue, 18 Dec 2018 15:53:37 GMT):
the place for this discussion is the W3C mailing list I think

jljordan_bcgov (Tue, 18 Dec 2018 15:53:57 GMT):
where the maintainers of the DID spec live

mwherman2000 (Tue, 18 Dec 2018 15:54:12 GMT):
link

mwherman2000 (Tue, 18 Dec 2018 15:54:12 GMT):
link?

jljordan_bcgov (Tue, 18 Dec 2018 15:55:50 GMT):
here is where a discussion about these kinds of topics happened recently https://lists.w3.org/Archives/Public/public-credentials/2018Dec/0055.html

jljordan_bcgov (Tue, 18 Dec 2018 15:58:22 GMT):
for example .. here is a summary of this very topic ... https://lists.w3.org/Archives/Public/public-credentials/2018Dec/0045.html

DuckLover (Tue, 18 Dec 2018 16:15:40 GMT):
Short question: Is the DID itself stored on the ledger?

jljordan_bcgov (Tue, 18 Dec 2018 16:16:31 GMT):
yes

jljordan_bcgov (Tue, 18 Dec 2018 16:17:55 GMT):
but not alone .. in indy it is with a key and looks like this ```{ "auditPath": [ "Eo4yYVy6r3tfGW7UfVtr9aLRXdocHThN5e3JGicBUrgW", "F86xX3HPjdhFYoLZQN6BTW84Zkbx9xKfHtv2ZLqKzZpe", "7g1hz1NJQyB4rcA7zRRQiL5h4UxDgSvUgFbUem1Dun9M" ], "reqSignature": { "type": "ED25519", "values": [ { "from": "V4SGRU86Z58d6TV7PBUe6f", "value": "5B98GuE7R7TFEcnC3d4wLjRufMKnoh9MfC2MzUzz6QtarKfndrYsAB5431ntCnVw3GxqNZr7J6Ux2DF9vSuCFFTG" } ] }, "rootHash": "7pKG5hiTVvbKgZfJJPFy3nz6R4NaTRuw1rzY9dY2AsmD", "txn": { "data": { "alias": "jj-test-dec14-1", "dest": "VqZUX1bWiR1tMDLpeuhZC1", "role": "101", "verkey": "Gibv7rcr2xQ63LQmtwQCtS8vY2JwXHpZHZyv1s4W9dWD" }, "metadata": { "digest": "a13d25243839dd5de4fd68240e5dfefbae21c0b95237052a4e57a3ae82092aaf", "from": "V4SGRU86Z58d6TV7PBUe6f", "reqId": 1544803317096239000 }, "protocolVersion": 2, "type": "1" }, "txnMetadata": { "seqNo": 38, "txnId": "cc7ff937c6cea79ff2328aafa9a4fee679b9475eb852837bf2c5de624d53ab40", "txnTime": 1544803317 }, "ver": "1" }```

jljordan_bcgov (Tue, 18 Dec 2018 16:17:55 GMT):
but not alone .. in indy it is with a key and looks like this from one of our test ledgers ... you can browse the ledger with our ledger browser http://138.197.138.255 (this is a dev service ... unsupported_ ```{ "auditPath": [ "Eo4yYVy6r3tfGW7UfVtr9aLRXdocHThN5e3JGicBUrgW", "F86xX3HPjdhFYoLZQN6BTW84Zkbx9xKfHtv2ZLqKzZpe", "7g1hz1NJQyB4rcA7zRRQiL5h4UxDgSvUgFbUem1Dun9M" ], "reqSignature": { "type": "ED25519", "values": [ { "from": "V4SGRU86Z58d6TV7PBUe6f", "value": "5B98GuE7R7TFEcnC3d4wLjRufMKnoh9MfC2MzUzz6QtarKfndrYsAB5431ntCnVw3GxqNZr7J6Ux2DF9vSuCFFTG" } ] }, "rootHash": "7pKG5hiTVvbKgZfJJPFy3nz6R4NaTRuw1rzY9dY2AsmD", "txn": { "data": { "alias": "jj-test-dec14-1", "dest": "VqZUX1bWiR1tMDLpeuhZC1", "role": "101", "verkey": "Gibv7rcr2xQ63LQmtwQCtS8vY2JwXHpZHZyv1s4W9dWD" }, "metadata": { "digest": "a13d25243839dd5de4fd68240e5dfefbae21c0b95237052a4e57a3ae82092aaf", "from": "V4SGRU86Z58d6TV7PBUe6f", "reqId": 1544803317096239000 }, "protocolVersion": 2, "type": "1" }, "txnMetadata": { "seqNo": 38, "txnId": "cc7ff937c6cea79ff2328aafa9a4fee679b9475eb852837bf2c5de624d53ab40", "txnTime": 1544803317 }, "ver": "1" }```

jljordan_bcgov (Tue, 18 Dec 2018 16:17:55 GMT):
but not alone .. in indy it is with a key and looks like this from one of our test ledgers ... you can browse the ledger with our ledger browser http://138.197.138.255 (this is a dev service ... unsupported) ```{ "auditPath": [ "Eo4yYVy6r3tfGW7UfVtr9aLRXdocHThN5e3JGicBUrgW", "F86xX3HPjdhFYoLZQN6BTW84Zkbx9xKfHtv2ZLqKzZpe", "7g1hz1NJQyB4rcA7zRRQiL5h4UxDgSvUgFbUem1Dun9M" ], "reqSignature": { "type": "ED25519", "values": [ { "from": "V4SGRU86Z58d6TV7PBUe6f", "value": "5B98GuE7R7TFEcnC3d4wLjRufMKnoh9MfC2MzUzz6QtarKfndrYsAB5431ntCnVw3GxqNZr7J6Ux2DF9vSuCFFTG" } ] }, "rootHash": "7pKG5hiTVvbKgZfJJPFy3nz6R4NaTRuw1rzY9dY2AsmD", "txn": { "data": { "alias": "jj-test-dec14-1", "dest": "VqZUX1bWiR1tMDLpeuhZC1", "role": "101", "verkey": "Gibv7rcr2xQ63LQmtwQCtS8vY2JwXHpZHZyv1s4W9dWD" }, "metadata": { "digest": "a13d25243839dd5de4fd68240e5dfefbae21c0b95237052a4e57a3ae82092aaf", "from": "V4SGRU86Z58d6TV7PBUe6f", "reqId": 1544803317096239000 }, "protocolVersion": 2, "type": "1" }, "txnMetadata": { "seqNo": 38, "txnId": "cc7ff937c6cea79ff2328aafa9a4fee679b9475eb852837bf2c5de624d53ab40", "txnTime": 1544803317 }, "ver": "1" }```

jljordan_bcgov (Tue, 18 Dec 2018 16:18:08 GMT):
Indy is not yet aligned with the DID spec

jljordan_bcgov (Tue, 18 Dec 2018 16:19:08 GMT):
so if you would like to see a real DID and what the DID doc look like here is the BC Registries Issuer DID https://uniresolver.io/#did=did:sov:HR6vs6GEZ8rHaVgjg2WodM

jljordan_bcgov (Tue, 18 Dec 2018 16:19:50 GMT):
Click on the DID Document tab and this is assembled from various transactions on the Sovrin ledger and transformed into the DID Doc spec format

jljordan_bcgov (Tue, 18 Dec 2018 16:20:06 GMT):
```{ "id": "did:sov:HR6vs6GEZ8rHaVgjg2WodM", "service": [], "authentication": { "type": "Ed25519SignatureAuthentication2018", "publicKey": [ "did:sov:HR6vs6GEZ8rHaVgjg2WodM#key-1" ] }, "publicKey": [ { "id": "did:sov:HR6vs6GEZ8rHaVgjg2WodM#key-1", "type": "Ed25519VerificationKey2018", "publicKeyBase58": "~VMZDS6JkbUvxibhqr6cbT2" } ], "@context": "https://w3id.org/did/v1" }```

brentzundel (Tue, 18 Dec 2018 16:26:31 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=pMgjyzHqEhMGzKaG6) @jljordan_bcgov It is important to remember that there is no "DID Spec" There is only a draft community group report that may some day become a W3C recommendation (or a specification from some other standards body).

jljordan_bcgov (Tue, 18 Dec 2018 16:32:07 GMT):
I am aware of that

mwherman2000 (Tue, 18 Dec 2018 16:39:53 GMT):
@jljordan_bcgov Your thoughts on the following....

mwherman2000 (Tue, 18 Dec 2018 16:40:13 GMT):

DID Logical Architecture.png

DuckLover (Tue, 18 Dec 2018 16:40:59 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=gLovCpLPFMrqCB7Mm) @jljordan_bcgov So with a DID i can resolve to a specific DID Doc in the Ledger but there is the DID also written in?

jljordan_bcgov (Tue, 18 Dec 2018 16:43:21 GMT):
Yes that is my understanding .. I think @AndrewHughes3000 offering in this message thread is generating good discussion around these kinds of contemplations ... https://lists.w3.org/Archives/Public/public-credentials/2018Dec/0045.html

mwherman2000 (Tue, 18 Dec 2018 16:44:12 GMT):

DID Logical Architecture.png

mwherman2000 (Tue, 18 Dec 2018 16:45:14 GMT):

DID Logical Architecture.png

swcurran (Tue, 18 Dec 2018 17:03:49 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=S9sngboGi9Q5HECK2) @brentzundel @brentzundel - a nit about this. There is a DID Spec, at the link Drumond mentioned above. It's a spec that has been defined through considerable effort, discussion and thought. However, it is not a *finalized *spec, it will undoubtably evolve somewhat, and most importantly, it has not yet been accepted by any Standards body. My point in adding this is that any discussion about DIDs needs to start from the spec as it stands and participants should be well versed in the current spec.

jljordan_bcgov (Tue, 18 Dec 2018 17:14:44 GMT):
I am listeing in on the W3C credential community call right now (its weekly) ... https://docs.google.com/document/d/1JIWWs8YTWP83Hao5UXyrgpddYu9F0v8lGDUo0Usor10/edit is a document being worked on by the community .. a DID Explainer

jljordan_bcgov (Tue, 18 Dec 2018 17:14:44 GMT):
I am listening in on the W3C credential community call right now (its weekly) ... https://docs.google.com/document/d/1JIWWs8YTWP83Hao5UXyrgpddYu9F0v8lGDUo0Usor10/edit is a document being worked on by the community .. a DID Explainer

brentzundel (Tue, 18 Dec 2018 17:16:12 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=xSQZC923BYZzHN3fa) @swcurran No disagreement from me on your points. I think it is important to remember that the existing document will change and is not final.

mwherman2000 (Tue, 18 Dec 2018 17:37:14 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=xSQZC923BYZzHN3fa) @swcurran A big part of the confusion (confudsion) arose last week in a number of presentations and conversations where the term DID was used in many different and differing ways. It was hard to separate the wheat (proper usage) from the chaff (overly casual usage of the term DID). @drummondreed clarified the expected usage in the last 12 hours (see above) ...but I guess if one (e.g. myself) feels the need for more precise terminology, it needs to be referenced against the specification? ...that's a bit weighty (given the lack clarity/precision), then that's what needs to be done. Correct?

mwherman2000 (Tue, 18 Dec 2018 17:37:14 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=xSQZC923BYZzHN3fa) @swcurran A big part of the confusion (confudsion) arose last week in a number of presentations and conversations where the term DID was used in many different and differing ways. It was hard to separate the wheat (proper usage) from the chaff (overly casual usage of the term DID). @drummondreed clarified the expected usage in the last 12 hours (see above) ...but I guess if one (e.g. myself) feels the need for more precise terminology, it needs to be referenced against the specification? ...that's a bit weighty (given the lack clarity/precision), but if that is what needs to be done, that's what needs to done. Correct?

mwherman2000 (Tue, 18 Dec 2018 17:37:14 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=xSQZC923BYZzHN3fa) @swcurran Specification or no specification, a big part of the confusion (confudsion) arose last week in a number of presentations and conversations where the term DID was used in many different and differing ways. It was hard to separate the wheat (proper usage) from the chaff (overly casual usage of the term DID). @drummondreed clarified the expected usage in the last 12 hours (see above) ...but I guess if one (e.g. myself) feels the need for more precise terminology, it needs to be referenced against the specification? ...that's a bit weighty (given the lack clarity/precision), but if that is what needs to be done, that's what needs to done. Correct?

mwherman2000 (Tue, 18 Dec 2018 17:37:14 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=xSQZC923BYZzHN3fa) @swcurran Specification or no specification, a big part of the confusion (confudsion) arose last week in a number of presentations and conversations where the term DID was used in many different and differing ways. It was hard to separate the wheat (proper/precise usage) from the chaff (overly casual usage of the term DID). @drummondreed clarified the expected usage in the last 12 hours (see above) ...but I guess if one (e.g. myself) feels the need for more precise terminology, it needs to be referenced against the specification? ...that's a bit weighty (given the lack clarity/precision), but if that is what needs to be done, that's what needs to done. Correct?

mwherman2000 (Tue, 18 Dec 2018 17:49:53 GMT):
RE: Link to current draft DID specification. I scrolled by quite a ways (e.g. a couple days) and couldn't find it. I don't want to be working from an "old one". What is the current/best URL to the draft DID specification?

mwherman2000 (Tue, 18 Dec 2018 17:49:53 GMT):
RE: Link to current draft DID specification. I scrolled back quite a ways (e.g. a couple days) and couldn't find it. I don't want to be working from an "old one". What is the current/best URL to the draft DID specification?

swcurran (Tue, 18 Dec 2018 17:50:04 GMT):
@mwherman2000 - it's an interesting question. I agree with you that if a newcomer to a situation has difficulty, that needs to be addressed. It's not the newcomers fault - whether it is software or specs. That said, the first step is making sure that all of the information available resolves the issue. As a newcomer, the starting point should have been the spec and if it wasn't, that's the first thing that needs to be fixed within the community. If you still have confudsion after that, then yes, fixing the spec is the right place, and as John said, that conversation needs to happen in that context - W3C Community Group.

swcurran (Tue, 18 Dec 2018 17:50:54 GMT):
The URL is always here, and it shows the date of the latest update - Dec. 11, 2018 as of now. https://w3c-ccg.github.io/did-spec/

swcurran (Tue, 18 Dec 2018 17:50:54 GMT):
The URL is always the following, and it shows the date of the latest update - Dec. 11, 2018 as of now. https://w3c-ccg.github.io/did-spec/

swcurran (Tue, 18 Dec 2018 17:51:53 GMT):
The spec is in a repo that is open to all and change is driven by github issues and pull requests.

swcurran (Tue, 18 Dec 2018 17:52:23 GMT):
W3C has a pretty formal process for evolving specs and the groups are diligent at sticking to the process.

mwherman2000 (Tue, 18 Dec 2018 17:52:45 GMT):
...and this is the correct place to create an Issue if needed? https://github.com/w3c-ccg/did-spec/issues

swcurran (Tue, 18 Dec 2018 17:53:28 GMT):
The status of the group is that they are working hard to transition from a Community Group (those can be created without appoval) to a Working Group (approved by W3C).

mwherman2000 (Tue, 18 Dec 2018 17:56:43 GMT):
Is the correct place to create an Issue if needed? https://github.com/w3c-ccg/did-spec/issues

swcurran (Tue, 18 Dec 2018 17:59:36 GMT):
Yup, you got it. There is a "File a Bug" link that goes there from the did-spec URL.

mwherman2000 (Tue, 18 Dec 2018 18:28:56 GMT):
#supertedius 3 issues opened to far and I'm only part way through the Overview. https://github.com/w3c-ccg/did-spec/issues

kdenhartog (Tue, 18 Dec 2018 18:46:47 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=ZPQJbJKJiQeaMDv7k) @vikpande This looks correct to me. I'd have to double check with the Sovrin glossary as I typically refer to the lump sum as DIDs. @drummondreed does this align with your thinking?

mwherman2000 (Tue, 18 Dec 2018 19:04:30 GMT):
@kdenhartog I think this diagram might have the same faults as the DID draft specification. DIDs are just character strings (confirmed by @drummondreed). Hence, I don't think they can have "types". I think what is depicted are actually DID Entities ...not DIDs. Scan the first 5 issues here: https://github.com/w3c-ccg/did-spec/issues (issues 115-119) ...119 refers to this diagram...

mwherman2000 (Tue, 18 Dec 2018 19:04:30 GMT):
@kdenhartog I think this diagram might have the same faults as the DID draft specification. DIDs are just character strings (confirmed by @drummondreed). Hence, I don't think they can have "types". I think what is depicted are actually different types of DID Entities ...not DIDs. Scan the first 5 issues here: https://github.com/w3c-ccg/did-spec/issues (issues 115-119) ...119 refers to this diagram...

mwherman2000 (Tue, 18 Dec 2018 19:05:12 GMT):

DID Logical Architecture.png

mwherman2000 (Tue, 18 Dec 2018 19:08:29 GMT):

DID Logical Architecture.png

TammyPlatero (Tue, 18 Dec 2018 19:32:27 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=076c2a0b-a4e2-4170-919b-0a5f8fa556ad) @dbluhm Here is what I was looking for :) https://w3c-ccg.github.io/did-spec/#the-generic-did-scheme

TammyPlatero (Tue, 18 Dec 2018 19:33:09 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=yBp5QejdX7ust56vx) @infominer33 Thank you! That's a handy document. I'll keep that bookmarked. I was looking for this document: https://w3c-ccg.github.io/did-spec/#the-generic-did-scheme

brenzi (Tue, 18 Dec 2018 21:55:40 GMT):
Has joined the channel.

mwherman2000 (Wed, 19 Dec 2018 01:57:08 GMT):

DID Logical Architecture v0.3.png

mwherman2000 (Wed, 19 Dec 2018 01:57:08 GMT):

DID Logical Architecture v0.3.png

AladdinMH (Wed, 19 Dec 2018 11:52:36 GMT):
Has joined the channel.

drummondreed (Wed, 19 Dec 2018 14:05:03 GMT):
@mwherman2000 I think your "logical architecture" is good. One nit: you say "the DID is never written to the ledger" but some DID methods will do just that. In the did:sov: method, the DID is written to the Sovrin ledger using a NYM transaction. The properties of the associated DID document are written using ATTRIB transactions. You can't currently write a DID document directly to the Sovrin ledger, but that feature is on the roadmap (according to a recent post by @danielhardman). For details see https://github.com/hyperledger/indy-node/blob/master/docs/requests.md.

drummondreed (Wed, 19 Dec 2018 14:06:46 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=1d82f936-d013-40ee-ba87-48998269ba3d) Just to note: this diagram was from the glossary section of the original Sovrin Provisional Trust Framework and is now completely outdated. The new Sovrin Glossary is at https://docs.google.com/document/d/1gfIz5TT0cNp2kxGMLFXr19x1uoZsruUe_0glHst2fZ8/edit?usp=sharing

drummondreed (Wed, 19 Dec 2018 14:06:46 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=1d82f936-d013-40ee-ba87-48998269ba3d) *Answering another question from vikpande above* This diagram was from the glossary section of the original Sovrin Provisional Trust Framework and is now completely outdated. The new Sovrin Glossary V2 is at https://docs.google.com/document/d/1gfIz5TT0cNp2kxGMLFXr19x1uoZsruUe_0glHst2fZ8/edit?usp=sharing

drummondreed (Wed, 19 Dec 2018 14:09:40 GMT):
Also note that we are still writing new "explainers" of certain sets of glossary terms in the Appendicies of Sovrin Glossary V2. Anyone who would like to help in that process is welcome.

mwherman2000 (Wed, 19 Dec 2018 14:34:53 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=WKCwH9BLTkkaAA3oM) @drummondreed My "DID" comments/diagrams from now on are now entirely focused on migrating from the current state of the draft DID specification to some future state of the same; regardless of the current state of any actual implementation. It's understandable that particular implementations are going to lag behind until some time after the spec is finalized. I'll make corrections re: NYMs and repost.

mwherman2000 (Wed, 19 Dec 2018 14:38:00 GMT):
Is there a URL for a Ledger Browser for the Indy/Sovrin distributed ledger?

mwherman2000 (Wed, 19 Dec 2018 14:38:00 GMT):
@drummondreed Is there a URL for a Ledger Browser for the public Indy/Sovrin distributed ledger?

mwherman2000 (Wed, 19 Dec 2018 14:38:00 GMT):
@drummondreed Is there a URL for a Ledger Browser for the public Indy/Sovrin distributed ledger? Found it http://138.197.138.255/browse/domain?page=1&query=mwherman2002&txn_type=

DuckLover (Wed, 19 Dec 2018 14:47:44 GMT):
Maybe something like this? it was posted above

DuckLover (Wed, 19 Dec 2018 14:47:45 GMT):
138.197.138.255

DuckLover (Wed, 19 Dec 2018 14:58:03 GMT):
Is someone used with the ibm implementation for the Alice-Faber Example?

DuckLover (Wed, 19 Dec 2018 15:00:38 GMT):
I dont really get it why i should fill out the form. For example the step where i already got a Gov ID and want to get a faber transcript. It uses the GOV Creds but let me type in the degree and year. Cant Faber just send me all the data based on my Gov Creds?

swcurran (Wed, 19 Dec 2018 16:35:42 GMT):
@DuckLover - the IBM demo was based on the BC Gov VON work. In that demo, the Credentials (Permits and Licences) were issued by a series of Gov't Services, and the information entered in the form was from the Organization wanting the Permit as an application. The IBM folks adopted the code and made it work for the Alice/Faber/Acme example, but the flow (as you note) is a bit odd. Bottom line - it was just a demo and should be thought of as that.

swcurran (Wed, 19 Dec 2018 16:35:42 GMT):
@DuckLover - the IBM demo was based on the BC Gov VON work. In the BC Gov demo, the Credentials (Permits and Licences) were issued by a series of Gov't Services, and the information entered in the form was from the Organization wanting the Permit as an application. The IBM folks adopted the code and made it work for the Alice/Faber/Acme example, but the flow (as you note) is a bit odd. Bottom line - it was just a demo and should be thought of as that.

Sean_Bohan (Wed, 19 Dec 2018 17:12:05 GMT):
*Tomorrow's Thursday INDY WG Call:* Note - change in agenda (Sami has a conflict - will share the connector app after the New Year) Date: Thurs Dec 20 When: 8am PT, 9am MT, 10am CT, 11amET, 4pmGMT Where: https://zoom.us/j/232861185 Agenda: Documentation In GitHub Repos (Richard) Hyperledger Global Forum recap (all) Previous Recordings and Agendas: https://drive.google.com/drive/u/0/folders/1AwHWN95KmSEi5fijraID0tFFMzYHoMwt?ogsrc=32 (edited) NOTE: No call next week 12/27

mwherman2000 (Wed, 19 Dec 2018 17:44:57 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=WKCwH9BLTkkaAA3oM) @drummondreed I'm having second thoughts about "some DID methods will do just that. In the did:sov: method, the DID is written to the Sovrin ledger using a NYM transaction." ...that is, "is a NYM a DID" and "is a DID a NYM"? ...I don't know (but don't think so) ...waiting for a resolution to a series of questions in #indy-node.

mwherman2000 (Wed, 19 Dec 2018 17:44:57 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=WKCwH9BLTkkaAA3oM) @drummondreed I'm having second thoughts about "some DID methods will do just that. In the did:sov: method, the DID is written to the Sovrin ledger using a NYM transaction." ...that is, "is a NYM a DID?" and "is a DID a NYM?" ...I don't know (but don't think so) ...waiting for a resolution to a series of questions in #indy-node.

mwherman2000 (Wed, 19 Dec 2018 17:44:57 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=WKCwH9BLTkkaAA3oM) @drummondreed I'm having second thoughts about "some DID methods will do just that. In the did:sov: method, the DID is written to the Sovrin ledger using a NYM transaction." ...that is, "is a NYM a DID?" and "is a DID a NYM?" ...I don't know (but don't think so) ...waiting for a resolution to a series of questions in #indy-node. DIDs being an attrribute of a NYM Tx really isn't any different that a DID being part of a DID Document. @drummondreed

vikpande (Wed, 19 Dec 2018 17:52:11 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=2vB7SwkhJPseyDidK) @drummondreed :thumbsup:

infominer33 (Wed, 19 Dec 2018 18:54:26 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=PLM4LebF7H6kYc5gZ) @mwherman2000 is bcgov part of the sovrin ledger? or its own instance?

jakubkoci (Wed, 19 Dec 2018 18:59:41 GMT):
Has joined the channel.

swcurran (Wed, 19 Dec 2018 19:10:45 GMT):
@infominer33 - the production instance of BC Gov's OrgBook and related Issuer/Verifier Agents are using the Sovrin Live (Main?) Net as it's root of trust for DIDs, Schema and Cred Defs. The non-production instances of the applications/agents are using the BCovrin (http://138.197.138.255) ledger to anchor DIDs, etc. Ontario Gov is doing the same for it's production and test instances (using Sovrin and BCovrin respectively).

swcurran (Wed, 19 Dec 2018 19:10:45 GMT):
@infominer33 - the production instance of BC Gov's OrgBook and related Issuer/Verifier Agents are using the Sovrin Live (Main?) Net as their root of trust for DIDs, Schema and Cred Defs. The non-production instances of the applications/agents are using the BCovrin (http://138.197.138.255) ledger to anchor DIDs, etc. Ontario Gov is doing the same for it's production and test instances (using Sovrin and BCovrin respectively).

infominer33 (Wed, 19 Dec 2018 19:15:38 GMT):
so that link I quoted is for the test net, basically?

infominer33 (Wed, 19 Dec 2018 19:15:38 GMT):
so that link I quoted is for the test net, basically.. I see I have a lot to learn :)

DuckLover (Wed, 19 Dec 2018 19:30:20 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=as3pGRimTfoXRKrNY) @swcurran Thanks for that! Can someone recommend an easy source for a simple setup. I just need to share creds between 3 or 4 comp. Setup from https://github.com/IBM-Blockchain-Identity/indy-tutorial-sandbox seems not to work

swcurran (Wed, 19 Dec 2018 19:33:36 GMT):
@DuckLover - the VON stuff in BC Gov may work for you, but I'd need a bit more context to know for sure. It certainly is super easy to set up Issuer/Verifiers with VON, but it is intended is an SSI bootstrapping mechanism for creating a supply of Credential issuers - but not for individuals. DM we in rocketchat and we can see if it's a fit for you.

infominer33 (Wed, 19 Dec 2018 20:16:55 GMT):
any chance there is (or will be) video available from the hyperledger forum? I've been quietly hoping...

swcurran (Wed, 19 Dec 2018 21:56:24 GMT):
@infominer33 - what topic? They have released a number of them.

swcurran (Wed, 19 Dec 2018 21:56:24 GMT):
@infominer33 - what topic? They have released a number of them (I think...).

infominer33 (Wed, 19 Dec 2018 22:01:32 GMT):
I was thinking specifically of the one you were part of for Verifiable Organizations Network

infominer33 (Wed, 19 Dec 2018 22:02:51 GMT):
looks like you had a few sessions on the subject

swcurran (Wed, 19 Dec 2018 22:06:57 GMT):
Yes - we had fun over there. We did a workshop and plan is to do a recording of the workshop to walk through the background and the steps of the workshop. A recording from the actual workshop would have been difficult.

swcurran (Wed, 19 Dec 2018 22:07:07 GMT):
We'll post here when available.

infominer33 (Wed, 19 Dec 2018 22:09:02 GMT):
excellent! I figured useful material was coming down the pipe, but I had to ask.

drummondreed (Thu, 20 Dec 2018 02:05:58 GMT):
I highly recommend whatever the BC Gov team puts together to capture the essence of their workshop. It was the cats pajamas.

allisongs (Thu, 20 Dec 2018 03:56:15 GMT):
Has joined the channel.

vikpande (Thu, 20 Dec 2018 08:26:35 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=ZyvHAzgwMw9e4qrmZ) @swcurran :thumbsup:

knagware9 (Thu, 20 Dec 2018 13:58:51 GMT):
Hi , I have one confusion. HOw I can use sovrin network ? How i can use to store my identities ? and even how I can connect to sovrin network

DuckLover (Thu, 20 Dec 2018 15:08:51 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=gz9uMfFg7MhAo2HAd) @knagware9 Correct me when im wrong. 1) You need a client(+Agent) that interacts with the sovrin ledger. For this the client need the genesis record of sovrin that defines the first pool nodes of sovrin. Storage can be done locally on your Client (Mobile Phone, Laptop...), Agent or somewhere else but the clinet/agent have to be able to deliver them to you.

DuckLover (Thu, 20 Dec 2018 15:11:49 GMT):
I got a question on my own. How is the authentification handled when someone wants to be a trust anchor? Are the use cases always limited to KYC situations?

DuckLover (Thu, 20 Dec 2018 15:12:41 GMT):
I mean if i want to be an trust anchor, i need another trust anchor write me as that in the ledger, but how does he know who i am, what my endpoints are and if im trustworthy?

knagware9 (Thu, 20 Dec 2018 18:07:16 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=m3vxXxkvJMnQ4BpMp) @DuckLover Right.. thanks for the details..let me try

swcurran (Thu, 20 Dec 2018 18:18:14 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=takPXnoQMqyoj8JF9) @DuckLover This is less a techical question as a Business/Legal question and that's where the Sovrin Governance Framework comes into play. Technically, it's easy for a Trust Anchor to execute a transaction to create another Trust Anchor. However, getting a TA to do that for requires "proving" in the social sense who you are and what you are planning to do. This will become solidified over time based on the Sovrin Gov Framework. Right now I believe (at least in my direct experience) on direct interactions with a Trust Anchor.

MayukhGhosh (Fri, 21 Dec 2018 09:57:56 GMT):
Has joined the channel.

xnopre (Fri, 21 Dec 2018 10:58:19 GMT):
Hi all :-) . I have this use case : the user get a "earning credential" with its salary from its employer. He provide to the bank, a proof including this data (earning). The bank want to know when the user earning has changed. What is the best way to update the proof for the bank ? Is it possible ? Can the issuer update a credential ? Does the bank need to request regularly a new proof ? ..... Thanks for you solution or idea :-)

jljordan_bcgov (Fri, 21 Dec 2018 11:42:52 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=SBHW9mns3QHJfmxMp) @xnopre So, in answering your question I will quote from a very interesting paper that the German Blockchain Association published 23 October 2018 which I am just making my way through today ... https://jolocom.io/wp-content/uploads/2018/10/Self-sovereign-Identity-_-Blockchain-Bundesverband-2018.pdf ... so far I would recommend it as a great overview of SSI. ``` It is important to say that the exchange of claims happens in a peer-to-peer manner and the Holder is always one of the peers to the interaction. No information is exchanged outside the Holder’s control, a principle that is ensured by storing the Claims under the Holder’s control and requiring cryptographic signatures for each interaction, based on keys only the Holder can access and control.``` So there are a few different aspects to your questions I will address individually. If the bank wants to know if the earnings have changed .. they must ask the holder of the credential (employee/customer) via a proof request. As per the paragraph quoted above, this is a fundamental design intent of an SSI system. The holder MUST be involved. Now, this could be perhaps automated, perhaps when the employee/bank customer agreed to receive a loan from the bank they also consented to answering earnings update requests from the bank every X period of time. This request may be manually handled by the employee/bank customer or perhaps the tell their persistent agent to automatically respond to these proof requests from the bank. This is a business/design decision. Can the issuer (employer) update a credential. Technically they can not "update" a credential. What they can do is issue a new credential and revoke the previous one. Credentials could also include expiry dates that would offer verifiers a way to ask for "current" credentials. As well, credentials likely will have an issuance date (I think that is a good idea in general) so verifiers can request that claim in a proof request to see how "fresh" the claim is. Now, all that being said I can promote/test the waters on an upcoming feature of the coming Hyperledger Indy Catalyst's Credential Registry component (aka OrgBook https://orgbook.gov.bc.ca is an example) ... This feature will allow verifiers to subscribe (via a webhook) to an organization page or a credential page so they can be notified of changes such as a legal name change, the issuance of a new permit, the change in legal status of an organization, etc. Now this is only possible when it is ok to have a multi-subject holder (subjects in this case being legal entities). Since we are using public open data that is ok .. we would NOT, at least in our jurisdiction, see this as a pattern for personal data. The use cases for Hyperledger Indy Catalyst's Credential Registry component are for these kinds of public collections of accreditations. For example, OrgBook BC provides a pillar of trust around legal entities in British Columbia, we hope other jurisdictions might adopt the OrgBook model. However there are many more we think - College of Physicians and Surgeons, Engineering Associations, Law Societies, Universities and Colleges, Government Ministries/Departments, membership organizations (Hyperledger, Sovrin, Industry Associations, etc) and more I am sure. All these have the pattern of having public statements of accreditation of their members. A very long response .. hope that was helpful.

joaquimpedrooliveira (Fri, 21 Dec 2018 20:39:15 GMT):
Has joined the channel.

joaquimpedrooliveira (Fri, 21 Dec 2018 20:40:29 GMT):
Hi, everyone. I'm starting with Indy and following this tutorial: https://github.com/hyperledger/indy-node/blob/master/environment/docker/pool/StartIndyAgents.md

joaquimpedrooliveira (Fri, 21 Dec 2018 20:40:29 GMT):
Hi, everyone. I'm starting with Indy and following this tutorial: https://github.com/hyperledger/indy-node/blob/master/getting-started.md#using-the-indy-cli

joaquimpedrooliveira (Fri, 21 Dec 2018 20:44:00 GMT):
I've noticed that the indy client executable changed from `indy' to `indy-cli` and that some commands also do not correspond. For instance, command for creating a wallet is described in tutorial as `new wallet Alice` but this does not exist. It appears to be `wallet create key=`

joaquimpedrooliveira (Fri, 21 Dec 2018 20:44:24 GMT):
Can someone confirm that these docs are really outdated? Am I missing something here?

esplinr (Fri, 21 Dec 2018 22:25:11 GMT):
That answer by @jljordan_bcgov is fantastic. It makes me think we should move more of these discussions into the forums where they are persisted and Google searchable.

esplinr (Fri, 21 Dec 2018 22:25:11 GMT):
That answer by @jljordan_bcgov is fantastic. It makes me think we should move more of these discussions into someplace like the Sovrin forums where they are persisted and Google searchable.

esplinr (Fri, 21 Dec 2018 22:25:33 GMT):
Something to discuss next year.

esplinr (Fri, 21 Dec 2018 22:25:49 GMT):
I'm leaving on vacation now. Happy holidays everyone and I'll talk to you next year!

smaldeniya (Sat, 22 Dec 2018 17:56:26 GMT):
Has joined the channel.

smaldeniya (Sat, 22 Dec 2018 17:56:33 GMT):
i am new to this and currently having local pool of hyperledger indy nodes. this connect to a node app using indy sdk. i have been able to create dids and add them to ledger using sdk. but when i use indy-cli to get those added nyms it says no did found. why can't i get the nyms using indy-cli which i added using indy sdk?

smaldeniya (Sat, 22 Dec 2018 17:56:33 GMT):
hi all, i am new to blockchains and currently trying with a local pool of hyperledger indy nodes. this pool connect to a node app using indy sdk. i have been able to create dids and add them to ledger using sdk. but when i use indy-cli to get those added nyms it says no did found. why can't i get the nyms using indy-cli which i added using indy sdk?

sergey.minaev (Mon, 24 Dec 2018 15:07:36 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=uE3vtMB5oGHNHZsdE) @joaquimpedrooliveira getting started inside `indy-node` repo is really outdated as it claimed by first paragraph of this document https://github.com/hyperledger/indy-node/blob/master/getting-started.md#getting-started-with-indy

infominer33 (Mon, 24 Dec 2018 20:16:31 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=BJES5EoGwZ9icJCMK) @esplinr I was recently thinking similarl, that an indy forum would be a good resource

infominer33 (Mon, 24 Dec 2018 20:16:31 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=BJES5EoGwZ9icJCMK) @esplinr I was recently thinking similar, that an indy forum would be a good resource

infominer33 (Mon, 24 Dec 2018 20:37:54 GMT):
perhaps it wouldn't be so much trouble to archive channel history, and put it somewhere searchable? https://rocket.chat/docs/developer-guides/rest-api/channels/history/

DuckLover (Tue, 25 Dec 2018 00:39:07 GMT):
Does someone have a paper or any other source that can clarify why the sovrin ledger is still immutable even if it doesnt use chained blocks as a structure?

infominer33 (Tue, 25 Dec 2018 03:21:45 GMT):
I would start with the indy plenum wiki, @DuckLover https://github.com/hyperledger/indy-plenum/wiki

infominer33 (Tue, 25 Dec 2018 03:26:40 GMT):
from a browse of the Redundant Byzantine Fault Tolerance paper, they took the traditional BFT algorithm and improved its performance by running multiple instances

infominer33 (Tue, 25 Dec 2018 03:26:40 GMT):
from a browse of the Redundant Byzantine Fault Tolerance paper, they took the traditional BFT algorithm and improved its performance by running multiple instances:

infominer33 (Tue, 25 Dec 2018 03:29:47 GMT):
>all existing BFT protocols targeting high throughput use a special replica, called the primary, which indicates to other replicas the order in which requests should be processed. This primary can be smartly malicious and degrade the performance of the system without being detected by correct replicas. > >In this paper, we propose a new approach, called RBFT for Redundant-BFT: we execute multiple instances of the same BFT protocol, each with a primary replica executing on a different machine. All the instances order the requests, but only the requests ordered by one of the instances, called the master instance, are actually executed. > >The performance of the different instances is closely monitored, in order to check that the master instance provides adequate performance. If that is not the case, the primary replica of the master instance is considered malicious and replaced. We implemented RBFT and compared its performance to that of other existing robust protocols. > >Our evaluation shows that RBFT achieves similar performance as the most robust protocols when there is no failure and that, under faults, its maximum performance degradation is about 3%, whereas it is at least equal to 78% for existing protocols.

infominer33 (Tue, 25 Dec 2018 03:29:47 GMT):
>all existing BFT protocols targeting high throughput use a special replica, called the primary, which indicates to other replicas the order in which requests should be processed. This primary can be smartly malicious and degrade the performance of the system without being detected by correct replicas. > >In this paper, we propose a new approach, called RBFT for Redundant-BFT: we execute multiple instances of the same BFT protocol, each with a primary replica executing on a different machine. All the instances order the requests, but only the requests ordered by one of the instances, called the master instance, are actually executed. > >The performance of the different instances is closely monitored, in order to check that the master instance provides adequate performance. If that is not the case, the primary replica of the master instance is considered malicious and replaced. We implemented RBFT and compared its performance to that of other existing robust protocols. > >Our evaluation shows that RBFT achieves similar performance as the most robust protocols when there is no failure and that, under faults, its maximum performance degradation is about 3%, whereas it is at least equal to 78% for existing protocols.

infominer33 (Tue, 25 Dec 2018 03:29:47 GMT):
>all existing BFT protocols targeting high throughput use a special replica, called the primary, which indicates to other replicas the order in which requests should be processed. This primary can be smartly malicious and degrade the performance of the system without being detected by correct replicas. > >In this paper, we propose a new approach, called RBFT for Redundant-BFT: we execute multiple instances of the same BFT protocol, each with a primary replica executing on a different machine. All the instances order the requests, but only the requests ordered by one of the instances, called the master instance, are actually executed. > >The performance of the different instances is closely monitored, in order to check that the master instance provides adequate performance. If that is not the case, the primary replica of the master instance is considered malicious and replaced. We implemented RBFT and compared its performance to that of other existing robust protocols. > >Our evaluation shows that RBFT achieves similar performance as the most robust protocols when there is no failure and that, under faults, its maximum performance degradation is about 3%, whereas it is at least equal to 78% for existing protocols.

infominer33 (Tue, 25 Dec 2018 03:30:02 GMT):
idk if that helps

infominer33 (Tue, 25 Dec 2018 03:37:03 GMT):
next I would read: http://www.windley.com/archives/2016/09/self-sovereign_identity_and_the_legitimacy_of_permissioned_ledgers.shtml

infominer33 (Tue, 25 Dec 2018 03:37:36 GMT):
starting at the section: "The Legitimacy of Distributed Ledger Identity Systems

infominer33 (Tue, 25 Dec 2018 03:37:36 GMT):
starting at the section: "The Legitimacy of Distributed Ledger Identity Systems"

infominer33 (Tue, 25 Dec 2018 03:38:20 GMT):
where Phil compares the tradeoffs between public and private ledgers

DuckLover (Tue, 25 Dec 2018 12:11:48 GMT):
thanks for your reply. Im not sure thats answers my question but i will definitly check out the plenum paper more. My question was more about the security aspects. In Bitcoin a block contains the hash of the previous block so that if you want to replace a block of data you need to change all the data that follows. I try to understand how Sovrin solves that security problem. But i guess i will try to dig deeper in the plenum papers

infominer33 (Tue, 25 Dec 2018 12:14:07 GMT):
thx for the clarification, lemme see what else I can dig up

infominer33 (Tue, 25 Dec 2018 12:19:09 GMT):
@DuckLover the Sovrin White Paper says: 1. Each transaction in the blockchain is digitally signed by the originator. 2. Each transaction—singly or in blocks—is chained to the prior via a digital hash.5 3. Validated transactions are replicated across all machines using a consensus algorithm https://sovrin.org/wp-content/uploads/Sovrin-Protocol-and-Token-White-Paper.pdf

infominer33 (Tue, 25 Dec 2018 12:20:13 GMT):
sorry, that's not super helpful

DuckLover (Tue, 25 Dec 2018 12:20:47 GMT):
yeah i think thats correct for blockchains but sovrin uses a ledger as i know

DuckLover (Tue, 25 Dec 2018 12:26:38 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=wgHCiypS9anSKvPAN) @infominer33 i took a quick look in the paper and its mostly deal with performance and malicious nodes that want to change the current data and not data that was written in the past. I will try to look around more about that

infominer33 (Tue, 25 Dec 2018 12:27:14 GMT):
I could be reading this wrong, but it appears that each transaction is hashed to the previous transaction? http://138.197.138.255/browse/domain?page=1&query=&txn_type=

infominer33 (Tue, 25 Dec 2018 12:27:14 GMT):
I could be reading this wrong, but it appears that each transaction is refrenced to the previous transaction? http://138.197.138.255/browse/domain?page=1&query=&txn_type=

infominer33 (Tue, 25 Dec 2018 12:28:58 GMT):
I'm thinking the auditpath is the key to your answer.. but I'm not 100%

DuckLover (Tue, 25 Dec 2018 12:34:48 GMT):
well it really look like they are using hash values... Not sure how to understand it with my previous information that it doesnt use chained blocks.

infominer33 (Tue, 25 Dec 2018 12:49:50 GMT):
yeah, but I think it's just one transaction after the other, not stored up in blocks before confirmation

infominer33 (Tue, 25 Dec 2018 12:50:41 GMT):
hashed to previous transaction(s?) already chained... but not in blocks... because this ledger isn't fast moving like bitcoin

infominer33 (Tue, 25 Dec 2018 12:51:31 GMT):
I feel like I can remember now reading about that in a forum somewhere

infominer33 (Tue, 25 Dec 2018 12:52:54 GMT):
since it's basically just a distributed registry of public keys, that enable all kinda other interactions, off-chain

infominer33 (Tue, 25 Dec 2018 12:52:54 GMT):
since it's basically just a distributed registry of public keys of public organizations, enabling all kinda other interactions, off-chain

infominer33 (Tue, 25 Dec 2018 12:54:22 GMT):
@DuckLover

infominer33 (Tue, 25 Dec 2018 12:55:43 GMT):
no need to tag you I guess, I was just excited to remember

jljordan_bcgov (Tue, 25 Dec 2018 17:20:56 GMT):
https://github.com/hyperledger/indy-plenum/blob/master/docs/storage.md Provides info on the design of the ledger in Indy

haow85 (Wed, 26 Dec 2018 01:17:46 GMT):
Does anyone has a tutorial or example on how to tak advantage of HyperLedger Indy to develop Dapps ?

sanjeevkkn (Wed, 26 Dec 2018 06:18:22 GMT):
Has joined the channel.

joaquimpedrooliveira (Wed, 26 Dec 2018 14:26:13 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=SzGFH8Kr9a9prBECd) @sergey.minaev Thank you for your answer. I've really missed the warning. So, the official getting started is the one from `indy-sdk` repo? https://github.com/hyperledger/indy-sdk/blob/master/doc/getting-started/getting-started.md

sergey.minaev (Wed, 26 Dec 2018 15:02:36 GMT):
@joaquimpedrooliveira yes, actual official GSG is in indy-sdk now

Cato (Thu, 27 Dec 2018 17:23:08 GMT):
Has joined the channel.

Cato (Thu, 27 Dec 2018 17:26:55 GMT):
Hi, all. Has zero-knowledge-proof been integrated in Sovrin network? If so, anyone can explain that how ZKP works in Sovrin.

infominer33 (Thu, 27 Dec 2018 20:04:31 GMT):
yes, @Cato zkp are a core component to the function of Sovrin. https://github.com/hyperledger/indy-anoncreds is where the code lives https://sovrin.org/the-sovrin-network-and-zero-knowledge-proofs/ gives a high level walk through of how it works

Cato (Fri, 28 Dec 2018 01:53:33 GMT):
@infominer33 Thank you for you reply, it helps me a lot.

Cato (Fri, 28 Dec 2018 01:53:33 GMT):
@infominer33 Thank you for your reply, it helps me a lot.

ashokkj (Fri, 28 Dec 2018 09:34:51 GMT):
Has joined the channel.

ashokkj (Fri, 28 Dec 2018 09:36:06 GMT):
I want to know when NYM transaction is registered, what are its content, I mean what are the parameters that are stored at Ledger? Can anyone point the doc?

jljordan_bcgov (Fri, 28 Dec 2018 11:55:27 GMT):
If you look here http://138.197.138.255/browse/domain you can see the formatted and raw contents of an Indy ledger. This is one of our small test ledgers but it is a current version of Indy. @ashokkj [ ](https://chat.hyperledger.org/channel/indy?msg=D79uxJhxbDoje6ZNY)

ashokkj (Fri, 28 Dec 2018 14:40:58 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=pSapEkGzWiKE1EQZD2) @jljordan_bcgov Thank you for the reply @jljordan_bcgov

DuckLover (Fri, 28 Dec 2018 15:59:18 GMT):
It is possible to change the private key for a DID. Why is it possible? Are there no concerns that a second DID could be created with the same new Keypair?

DuckLover (Fri, 28 Dec 2018 16:01:50 GMT):
or that severals DID can be controlled by one keypair? like one password for all accounts today?

anant706 (Fri, 28 Dec 2018 17:22:55 GMT):
Has joined the channel.

swcurran (Fri, 28 Dec 2018 18:30:10 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=rE3kK7bKGQZDofeZS) @DuckLover Answers - yes, it's possible (and expected) to change the private key for a DID - key rotation - and you must have sign the request with the current private key for the DID. A second DID cannot be created with the same key pair - at most, the DID can be updated because the request is signed by the private key of the DID. However, one of the tenents of these systems is that you don't let others know your private key or (if you keep it) the seed used to create the keypair. That's the only way a second person could update the DID. Yes, in theory several DIDs could be controlled by one keypair, but you would have to go out of the way to do it. When you create a DID in Indy, the DID string is based on the initial public key of the pair. So to have two keys with one keypair, you would have to create each DID and then rotate the keys to match. Not sure why you would want to do that, but sure, fill your boots. Since the public keys would be the same, people would know they had the same private key. So possible but dumb (same as one password) and inconvenient (different from one password).

pknowles (Sat, 29 Dec 2018 13:59:02 GMT):
In order to receive *Indy Semantics WG call calendar invites*, please add your contact details to the following distribution list. https://docs.google.com/document/d/1NL36ZIksk4DmquRNvxpyZugWyjqCYa6n20FMzUnf-fY/edit?usp=sharing

pknowles (Sat, 29 Dec 2018 13:59:02 GMT):
In order to receive *Indy Semantics WG calendar invites*, please add your contact details to the following distribution list. https://docs.google.com/document/d/1NL36ZIksk4DmquRNvxpyZugWyjqCYa6n20FMzUnf-fY/edit?usp=sharing

mwherman2000 (Sun, 30 Dec 2018 22:54:22 GMT):
[CROSSPOSTING] I don't know how many of you are also on the public-credentials@w3.org distribution list ...so I'm crossing this here,,,

mwherman2000 (Sun, 30 Dec 2018 22:56:19 GMT):
From: Michael Herman (Parallelspace) Sent: December 30, 2018 2:27 PM To: daniel.hardman@evernym.com Cc: public-credentials@w3.org Subject: Draft DID spec: DID ARM (Architecture Reference Model) diagram version 0.11 Daniel and list, I’ve updated the DID ARM (Architecture Reference Model) diagram to version 0.11. I believe if we can get an agreeable “picture” of what we’re talking about, the correct words will come a lot easier …at least, that’s my experience in the past. The ARM is posted here: https://hyperonomy.com/2018/12/21/decentralized-identifiers-dids-architecture-reference-model-arm/. Scroll down the page to the heading Version 0.11 – December 30, 2018 and start reading from there. If you need/want more background, start reading from the top. Here’s a copy of what I’ve written there (minus the ARM diagram)… Version 0.11 – December 30, 2018 The updates in version 0.11 of the Decentralized Identifiers (DIDs) ARM (architecture reference model) are limited to the Business Architecture Layer; more specifically, clarifying to roles and relationships between: - Actors (Persons and Organizations), and - Things (Pets, Cars, Houses, Business Documents, Products, Assemblies and Parts). The role of Actor in the model is a key learning that came out of the Basel workshop. I’ve taken the idea a step further to find a proper “place and set of relationships” for the Sovrin concept of “Things”. I believe I’ve (successfully?) done that but am looking for feedback/consensus about these ideas. They are presented visually in the diagram below. The new Business Architecture basic principles are: 1. An Actor is either a Person or an Organization. 2. Each Person or Organization has one or more SS Digital Identities associated with it. 3. Actors (Persons and Organizations) participate in Processes. 4. A Process acts on/accesses Things (e.g. Pet, Car, House, Business Document, Product, Assembly, Part) to perform work. 5. Business Documents, Products, Assemblies and Parts are different from the traditional or generic Sovrin concept of a Thing (e.g. Pet, Car, House). 6. Each Thing has one or more SS Digital Identities associated with it. To bridge from previous versions of the ARM (see below), the following Application Architecture basic principles also apply: 1. A DID Document is a JSON-LD serialization of a DID Entity. 2. A DID Entity is the in-memory, application-specific object that represents a de-serialized DID Document. 3. A DID Entity is what application developers work with (program against) at the Application Architecture level of an app. 4. DID Entities have a set of attributes such as the following: - id (DID) - service (endpoints) - authentication - publicKey - @context - etc. 5. “id (DID)” exists as an attribute of a DID Entity (and by implication, as an attribute of DID Document, the JSON-LD serialization of the corresponding DID Entity). 6. The “id (DID)” attribute is given the nickname “DID” (aka Decentralized Identifier) for convenience; but more importantly, to clarify what a DID specifically refers to (as well as to clarify what the term DID specifically does not refer to). “DID” should only be used to refer to the “id (DID)” attribute of a DID Entity (or DID Document). 7. DIDs are used to index, find, and retrieve DID Documents from the Technology/Infrastructure Architecture layer. 8. DID Documents represent the primary object type that is exchanged between the Application Architecture layer and the Technology/Infrastructure layer. Please reply with your thoughts and feedback. Best regards (and Happy New Year :-)), Michael Herman (Toronto/Calgary/Seattle)

mwherman2000 (Sun, 30 Dec 2018 22:56:19 GMT):
From: Michael Herman (Parallelspace) Sent: December 30, 2018 2:27 PM To: daniel.hardman@evernym.com Cc: public-credentials@w3.org Subject: Draft DID spec: DID ARM (Architecture Reference Model) diagram version 0.11 Daniel and list, I’ve updated the DID ARM (Architecture Reference Model) diagram to version 0.11. I believe if we can get an agreeable “picture” of what we’re talking about, the correct words will come a lot easier …at least, that’s my experience in the past. The ARM is posted here: https://hyperonomy.com/2018/12/21/decentralized-identifiers-dids-architecture-reference-model-arm/. Scroll down the page to the heading *Version 0.11 – December 30, 2018* and start reading from there. If you need/want more background, start reading from the top. Here’s a copy of what I’ve written there (minus the ARM diagram)… Version 0.11 – December 30, 2018 The updates in version 0.11 of the Decentralized Identifiers (DIDs) ARM (architecture reference model) are limited to the Business Architecture Layer; more specifically, clarifying to roles and relationships between: - Actors (Persons and Organizations), and - Things (Pets, Cars, Houses, Business Documents, Products, Assemblies and Parts). The role of Actor in the model is a key learning that came out of the Basel workshop. I’ve taken the idea a step further to find a proper “place and set of relationships” for the Sovrin concept of “Things”. I believe I’ve (successfully?) done that but am looking for feedback/consensus about these ideas. They are presented visually in the diagram below. The new Business Architecture basic principles are: 1. An Actor is either a Person or an Organization. 2. Each Person or Organization has one or more SS Digital Identities associated with it. 3. Actors (Persons and Organizations) participate in Processes. 4. A Process acts on/accesses Things (e.g. Pet, Car, House, Business Document, Product, Assembly, Part) to perform work. 5. Business Documents, Products, Assemblies and Parts are different from the traditional or generic Sovrin concept of a Thing (e.g. Pet, Car, House). 6. Each Thing has one or more SS Digital Identities associated with it. To bridge from previous versions of the ARM (see below), the following Application Architecture basic principles also apply: 1. A DID Document is a JSON-LD serialization of a DID Entity. 2. A DID Entity is the in-memory, application-specific object that represents a de-serialized DID Document. 3. A DID Entity is what application developers work with (program against) at the Application Architecture level of an app. 4. DID Entities have a set of attributes such as the following: - id (DID) - service (endpoints) - authentication - publicKey - @context - etc. 5. “id (DID)” exists as an attribute of a DID Entity (and by implication, as an attribute of DID Document, the JSON-LD serialization of the corresponding DID Entity). 6. The “id (DID)” attribute is given the nickname “DID” (aka Decentralized Identifier) for convenience; but more importantly, to clarify what a DID specifically refers to (as well as to clarify what the term DID specifically does not refer to). “DID” should only be used to refer to the “id (DID)” attribute of a DID Entity (or DID Document). 7. DIDs are used to index, find, and retrieve DID Documents from the Technology/Infrastructure Architecture layer. 8. DID Documents represent the primary object type that is exchanged between the Application Architecture layer and the Technology/Infrastructure layer. Please reply with your thoughts and feedback. Best regards (and Happy New Year :-)), Michael Herman (Toronto/Calgary/Seattle)

mwherman2000 (Sun, 30 Dec 2018 22:56:19 GMT):
From: Michael Herman (Parallelspace) Sent: December 30, 2018 2:27 PM To: daniel.hardman@evernym.com Cc: public-credentials@w3.org Subject: Draft DID spec: DID ARM (Architecture Reference Model) diagram version 0.11 Daniel and list, I’ve updated the DID ARM (Architecture Reference Model) diagram to version 0.11. I believe if we can get an agreeable “picture” of what we’re talking about, the correct words will come a lot easier …at least, that’s my experience in the past. The ARM is posted here: https://hyperonomy.com/2018/12/21/decentralized-identifiers-dids-architecture-reference-model-arm/. Scroll down the page to the heading *Version 0.11 – December 30, 2018* and start reading from there. If you need/want more background, start reading from the top. Here’s a copy of what I’ve written there (minus the ARM diagram)… Version 0.11 – December 30, 2018 The updates in version 0.11 of the Decentralized Identifiers (DIDs) ARM (architecture reference model) are limited to the Business Architecture Layer; more specifically, clarifying to roles and relationships between: - Actors (Persons and Organizations), and - Things (Pets, Cars, Houses, Business Documents, Products, Assemblies and Parts). The role of Actor in the model is a key learning that came out of the Basel workshop. I’ve taken the idea a step further to find a proper “place and set of relationships” for the Sovrin concept of “Things”. I believe I’ve (successfully?) done that but am looking for feedback/consensus about these ideas. They are presented visually in the diagram below. The new Business Architecture basic principles are: 1. An Actor is either a Person or an Organization. 2. Each Person or Organization has one or more SS Digital Identities associated with it. 3. Actors (Persons and Organizations) participate in Processes. 4. A Process acts on/accesses Things (e.g. Pet, Car, House, Business Document, Product, Assembly, Part) to perform work. 5. Business Documents, Products, Assemblies and Parts are different from the traditional or generic Sovrin concept of a Thing (e.g. Pet, Car, House). 6. Each Thing has one or more SS Digital Identities associated with it. To bridge from previous versions of the ARM (see below), the following Application Architecture basic principles also apply: 1. A DID Document is a JSON-LD serialization of a DID Entity. 2. A DID Entity is the in-memory, application-specific object that represents a de-serialized DID Document. 3. A DID Entity is what application developers work with (program against) at the Application Architecture level of an app. 4. DID Entities have a set of attributes such as the following: - id (DID) - service (endpoints) - authentication - publicKey - @context - etc. 5. “id (DID)” exists as an attribute of a DID Entity (and by implication, as an attribute of DID Document, the JSON-LD serialization of the corresponding DID Entity). 6. The “id (DID)” attribute is given the nickname “DID” (aka Decentralized Identifier) for convenience; but more importantly, to clarify what a DID specifically refers to (as well as to clarify what the term DID specifically does not refer to). “DID” should only be used to refer to the “id (DID)” attribute of a DID Entity (or DID Document). 7. DIDs are used to index, find, and retrieve DID Documents from the Technology/Infrastructure Architecture layer. 8. DID Documents represent the primary object type that is exchanged between the Application Architecture layer and the Technology/Infrastructure layer. Please reply with your thoughts and feedback. Best regards (and Happy New Year :-)), Michael Herman (Toronto/Calgary/Seattle)

mwherman2000 (Sun, 30 Dec 2018 22:56:19 GMT):
From: Michael Herman (Parallelspace) Sent: December 30, 2018 2:27 PM To: daniel.hardman@evernym.com Cc: public-credentials@w3.org Subject: Draft DID spec: DID ARM (Architecture Reference Model) diagram version 0.11 Daniel and list, I’ve updated the DID ARM (Architecture Reference Model) diagram to version 0.11. I believe if we can get an agreeable “picture” of what we’re talking about, the correct words will come a lot easier …at least, that’s my experience in the past. The ARM is posted here: https://hyperonomy.com/2018/12/21/decentralized-identifiers-dids-architecture-reference-model-arm/. Scroll down the page to the heading *Version 0.11 – December 30, 2018* and start reading from there. If you need/want more background, start reading from the top. Here’s a copy of what I’ve written there (minus the ARM diagram)… Version 0.11 – December 30, 2018 The updates in version 0.11 of the Decentralized Identifiers (DIDs) ARM (architecture reference model) are limited to the Business Architecture Layer; more specifically, clarifying to roles and relationships between: - Actors (Persons and Organizations), and - Things (Pets, Cars, Houses, Business Documents, Products, Assemblies and Parts). The role of Actor in the model is a key learning that came out of the Basel workshop. I’ve taken the idea a step further to find a proper “place and set of relationships” for the Sovrin concept of “Things”. I believe I’ve (successfully?) done that but am looking for feedback/consensus about these ideas. They are presented visually in the diagram below. The new Business Architecture basic principles are: 1. An Actor is either a Person or an Organization. 2. Each Person or Organization has one or more SS Digital Identities associated with it. 3. Actors (Persons and Organizations) participate in Processes. 4. A Process acts on/accesses Things (e.g. Pet, Car, House, Business Document, Product, Assembly, Part) to perform work. 5. Business Documents, Products, Assemblies and Parts are different from the traditional or generic Sovrin concept of a Thing (e.g. Pet, Car, House). 6. Each Thing has one or more SS Digital Identities associated with it. To bridge from previous versions of the ARM (see below), the following Application Architecture basic principles also apply: 1. A DID Document is a JSON-LD serialization of a DID Entity. 2. A DID Entity is the in-memory, application-specific object that represents a de-serialized DID Document. 3. A DID Entity is what application developers work with (program against) at the Application Architecture level of an app. 4. DID Entities have a set of attributes such as the following: - id (DID) - service (endpoints) - authentication - publicKey - @context - etc. 5. “id (DID)” exists as an attribute of a DID Entity (and by implication, as an attribute of DID Document, the JSON-LD serialization of the corresponding DID Entity). 6. The “id (DID)” attribute is given the nickname “DID” (aka Decentralized Identifier) for convenience; but more importantly, to clarify what a DID specifically refers to (as well as to clarify what the term DID specifically does not refer to). “DID” should only be used to refer to the “id (DID)” attribute of a DID Entity (or DID Document). 7. DIDs are used to index, find, and retrieve DID Documents from the Technology/Infrastructure Architecture layer. 8. DID Documents represent the primary object type that is exchanged between the Application Architecture layer and the Technology/Infrastructure layer. Please reply with your thoughts and feedback. Best regards (and Happy New Year :-)), Michael Herman (Toronto/Calgary/Seattle)

mwherman2000 (Sun, 30 Dec 2018 22:56:19 GMT):
From: Michael Herman (Parallelspace) Sent: December 30, 2018 2:27 PM To: daniel.hardman@evernym.com Cc: public-credentials@w3.org Subject: Draft DID spec: DID ARM (Architecture Reference Model) diagram version 0.11 Daniel and list, I’ve updated the DID ARM (Architecture Reference Model) diagram to version 0.11. I believe if we can get an agreeable “picture” of what we’re talking about, the correct words will come a lot easier …at least, that’s my experience in the past. The ARM is posted here: https://hyperonomy.com/2018/12/21/decentralized-identifiers-dids-architecture-reference-model-arm/. Scroll down the page to the heading *Version 0.11 – December 30, 2018* and start reading from there. If you need/want more background, start reading from the top. Here’s a copy of what I’ve written there (minus the ARM diagram)… Version 0.11 – December 30, 2018 The updates in version 0.11 of the Decentralized Identifiers (DIDs) ARM (architecture reference model) are limited to the Business Architecture Layer; more specifically, clarifying to roles and relationships between: - Actors (Persons and Organizations), and - Things (Pets, Cars, Houses, Business Documents, Products, Assemblies and Parts). The role of Actor in the model is a key learning that came out of the Basel workshop. I’ve taken the idea a step further to find a proper “place and set of relationships” for the Sovrin concept of “Things”. I believe I’ve (successfully?) done that but am looking for feedback/consensus about these ideas. They are presented visually in the diagram below. The new Business Architecture basic principles are: 1. An Actor is either a Person or an Organization. 2. Each Person or Organization has one or more SS Digital Identities associated with it. 3. Actors (Persons and Organizations) participate in Processes. 4. A Process acts on/accesses Things (e.g. Pet, Car, House, Business Document, Product, Assembly, Part) to perform work. 5. Business Documents, Products, Assemblies and Parts are different from the traditional or generic Sovrin concept of a Thing (e.g. Pet, Car, House). 6. Each Thing has one or more SS Digital Identities associated with it. To bridge from previous versions of the ARM (see below), the following Application Architecture basic principles also apply: 1. A DID Document is a JSON-LD serialization of a DID Entity. 2. A DID Entity is the in-memory, application-specific object that represents a de-serialized DID Document. 3. A DID Entity is what application developers work with (program against) at the Application Architecture level of an app. 4. DID Entities have a set of attributes such as the following: - id (DID) - service (endpoints) - authentication - publicKey - @context - etc. 5. “id (DID)” exists as an attribute of a DID Entity (and by implication, as an attribute of DID Document, the JSON-LD serialization of the corresponding DID Entity). 6. The “id (DID)” attribute is given the nickname “DID” (aka Decentralized Identifier) for convenience; but more importantly, to clarify what a DID specifically refers to (as well as to clarify what the term DID specifically does not refer to). “DID” should only be used to refer to the “id (DID)” attribute of a DID Entity (or DID Document). 7. DIDs are used to index, find, and retrieve DID Documents from the Technology/Infrastructure Architecture layer. 8. DID Documents represent the primary object type that is exchanged between the Application Architecture layer and the Technology/Infrastructure layer. Please reply with your thoughts and feedback. Best regards (and Happy New Year :-)), Michael Herman (Toronto/Calgary/Seattle)

ashokkj (Tue, 01 Jan 2019 09:00:38 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=GbdWKB6EBomMS5x4W) Is it possible to see wallet (Trust anchors, users) contents as well ?

RaysUncle (Wed, 02 Jan 2019 09:36:03 GMT):
Has joined the channel.

seddoni (Wed, 02 Jan 2019 10:03:52 GMT):
Has joined the channel.

NedSnark (Wed, 02 Jan 2019 19:31:08 GMT):
Has joined the channel.

aappddeevv (Wed, 02 Jan 2019 20:52:12 GMT):
Has joined the channel.

mwherman2000 (Wed, 02 Jan 2019 21:20:34 GMT):
Is anyone familiar with this announcement? Is it important? https://cointelegraph.com/news/visa-set-to-launch-blockchain-based-digital-identity-system-with-ibm-in-q1-2019

pknowles (Wed, 02 Jan 2019 21:34:50 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=5DTvCdo7NCKBX9mki) @mwherman2000 ... not to mention Mastercard and Microsoft announcing a similar partnership! https://www.linkedin.com/feed/update/urn:li:activity:6475974366244012032

pknowles (Wed, 02 Jan 2019 21:34:50 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=5DTvCdo7NCKBX9mki) @mwherman2000 ... with Mastercard and Microsoft announcing a similar partnership! https://www.linkedin.com/feed/update/urn:li:activity:6475974366244012032

mwherman2000 (Wed, 02 Jan 2019 23:35:24 GMT):
@pknowles Any idea what MS-MC are planning to use for a platform? ...I think the IBM-Visa platform choice might be obvious.

pknowles (Thu, 03 Jan 2019 03:20:36 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=4q8FuoDtJ8AY3YuSB) @mwherman2000 Pass. "Intelligent Edge" is ultimately where MS are heading with this though. See https://www.forbes.com/sites/janakirammsv/2018/05/12/microsoft-starts-to-make-serious-progress-on-the-intelligent-edge-vision/#1eacc94c2608 . Kaliya ("Identity Woman") recently published the following article re the MS-MC announcement ... https://identitywoman.net/exciting-ssi-announcement-was-not-well-received-by-some/

pknowles (Thu, 03 Jan 2019 03:20:36 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=4q8FuoDtJ8AY3YuSB) @mwherman2000 Pass. "Intelligent Edge" is ultimately where MS are heading with this though. See https://www.linkedin.com/feed/update/urn:li:activity:6401052455131447296 . Kaliya ("Identity Woman") recently published the following article re the MS-MC announcement ... https://identitywoman.net/exciting-ssi-announcement-was-not-well-received-by-some/

pknowles (Thu, 03 Jan 2019 03:20:36 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=4q8FuoDtJ8AY3YuSB) @mwherman2000 Pass. "Intelligent Edge" is ultimately where MS are heading with this though. See https://www.linkedin.com/feed/update/urn:li:activity:6401052455131447296 . Kaliya ("Identity Woman") also recently published the following article re the MS-MC announcement ... https://identitywoman.net/exciting-ssi-announcement-was-not-well-received-by-some/

pknowles (Thu, 03 Jan 2019 03:20:36 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=4q8FuoDtJ8AY3YuSB) @mwherman2000 Pass. "Intelligent Edge" is ultimately where MS are heading with this though ... https://www.linkedin.com/feed/update/urn:li:activity:6401052455131447296 . Kaliya ("Identity Woman") also recently published the following article re the MS-MC announcement ... https://identitywoman.net/exciting-ssi-announcement-was-not-well-received-by-some/ . Also check out the conversation thread at the top of the following page re the MS-MC collab ... https://www.reddit.com/user/csuwildcat

pknowles (Thu, 03 Jan 2019 03:20:36 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=4q8FuoDtJ8AY3YuSB) @mwherman2000 Pass. "Intelligent Edge" is ultimately where MS are heading with this though ... https://www.linkedin.com/feed/update/urn:li:activity:6401052455131447296 . Kaliya ("Identity Woman") also recently published the following article re the MS-MC announcement ... https://identitywoman.net/exciting-ssi-announcement-was-not-well-received-by-some/ . Also, check out the conversation thread at the top of the following page re the MS-MC collab ... https://www.reddit.com/user/csuwildcat

pknowles (Thu, 03 Jan 2019 03:20:36 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=4q8FuoDtJ8AY3YuSB) @mwherman2000 Pass. "Intelligent Edge" is ultimately where MS are heading with this though ... https://www.linkedin.com/feed/update/urn:li:activity:6401052455131447296 . Kaliya ("Identity Woman") also recently published the following article re the MS-MC announcement ... https://identitywoman.net/exciting-ssi-announcement-was-not-well-received-by-some/ . Finally, check out the conversation thread at the top of the following page re the MS-MC collab ... https://www.reddit.com/user/csuwildcat

pknowles (Thu, 03 Jan 2019 03:20:36 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=4q8FuoDtJ8AY3YuSB) @mwherman2000 Pass. Their "Intelligent Cloud / Intelligent Edge" model is ultimately where MS are heading with this though ... https://www.linkedin.com/feed/update/urn:li:activity:6401052455131447296 . Kaliya ("Identity Woman") also recently published the following article re the MS-MC announcement ... https://identitywoman.net/exciting-ssi-announcement-was-not-well-received-by-some/ . Finally, check out the conversation thread at the top of the following page re the MS-MC collab ... https://www.reddit.com/user/csuwildca

pknowles (Thu, 03 Jan 2019 03:20:36 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=4q8FuoDtJ8AY3YuSB) @mwherman2000 Pass. Their "Intelligent Cloud / Intelligent Edge" model is ultimately where MS are heading with this though ... https://www.linkedin.com/feed/update/urn:li:activity:6401052455131447296 . Kaliya ("Identity Woman") also recently published the following article re the MS-MC announcement ... https://identitywoman.net/exciting-ssi-announcement-was-not-well-received-by-some/ . Finally, check out the conversation thread at the top of the following page re the MS-MC collab ... https://www.reddit.com/user/csuwildcat

pknowles (Thu, 03 Jan 2019 03:20:36 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=4q8FuoDtJ8AY3YuSB) @mwherman2000 Pass. MS's "Intelligent Cloud / Intelligent Edge" model is ultimately where they're heading with this though ... https://www.linkedin.com/feed/update/urn:li:activity:6401052455131447296 . Kaliya ("Identity Woman") also recently published the following article re the MS-MC announcement ... https://identitywoman.net/exciting-ssi-announcement-was-not-well-received-by-some/ . Finally, check out the conversation thread at the top of the following page re the MS-MC collab ... https://www.reddit.com/user/csuwildcat

Gilang (Thu, 03 Jan 2019 03:20:41 GMT):
Has joined the channel.

xadhoom76 (Thu, 03 Jan 2019 11:18:19 GMT):
Has joined the channel.

devin-fisher (Thu, 03 Jan 2019 17:59:26 GMT):
I noticed that a new repo was created called `indy-docs`. Can whoever is championing this, submit a PR to indy-hipe to update the HIPE-0018 to document the purpose of this new repo. https://github.com/hyperledger/indy-hipe/tree/master/text/0018-indy-git-repos

mboyd (Thu, 03 Jan 2019 18:05:12 GMT):
That was me, as has been documented in https://github.com/hyperledger/indy-hipe/blob/master/text/0025-indy-docs-framework/README.md#relevant-repositories I'll update the other HIPE accordingly

mboyd (Thu, 03 Jan 2019 18:05:12 GMT):
That was me, as has been documented in https://github.com/hyperledger/indy-hipe/blob/master/text/0025-indy-docs-framework/README.md#relevant-repositories I'll update the

devin-fisher (Thu, 03 Jan 2019 18:24:16 GMT):
Thanks

SaraInadam (Fri, 04 Jan 2019 11:32:01 GMT):
Has joined the channel.

mxs1491 (Fri, 04 Jan 2019 21:13:06 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=twWC4v84oLgbvKii6) @pknowles @pknowles Kaliya does make some very good points in her piece. Could be a very prescient long strategy to "re-level" the playing field since in markets they have never had much success in. LinkedIn had recently changed their rules around communications to groups as well as the ability to 'extract' emails of members of groups.

mxs1491 (Fri, 04 Jan 2019 21:13:06 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=twWC4v84oLgbvKii6) @pknowles @pknowles Kaliya does make some very good points in her piece. Could be a very prescient long strategy to "re-level" the playing field since in markets they have never had much success in. LinkedIn had recently changed their rules around communications to groups as well as the ability to 'extract' emails of members of groups. An interesting announcement to watch...

mwherman2000 (Sat, 05 Jan 2019 05:40:41 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=twWC4v84oLgbvKii6) @pknowles Microsoft "Bali" ...sounds like a wallet? https://amp.businessinsider.com/microsoft-working-on-project-bali-to-give-people-control-over-data-2019-1

tuckerg (Sat, 05 Jan 2019 06:25:20 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=jWHSN5W6XJ3w37cce) @mwherman2000 In the LlinkedIn article @pknowles claims "Microsoft's "Intelligent Cloud / Intelligent Edge" model described in the following Forbes article is very much aligned with a Sovrin Foundation "Global Data Store" concept..." Still in learning mode here, so can someone hold my hand through the conceptual pieces I am missing? Indy is a distributed mechanism for validating claims. How does this help us manage IoT assets, mobile edge computing assets in "the fog", replicate, distribute, or discover data in the fog, etc? I guess I am still trying to make the conceptual leap from global distributed claim validation to global data store.

tuckerg (Sat, 05 Jan 2019 06:29:29 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=jWHSN5W6XJ3w37cce) @mwherman2000 Precisely what I was thinking, Bali sounds like a data wallet controlled by MS. I am trying hard to understand why I would want this.

pknowles (Sat, 05 Jan 2019 07:36:34 GMT):
@tuckerg @mwherman2000 I think a number of concepts are getting "confuded" here: 1.) Sovrin Foundation has never mentioned a "Global Data Store". That is a concept that I've had running in my mind for the past few years. It's very much a personal concept rather than a Sovrin one. I don't want to confuse that fact at the risk of getting my wrist's slapped by the Technical Ambassadors! 2.) MS's "Intelligent Cloud / Intelligent Edge" model looks like a separate thing to "Bali". The former looks and feels similar to the early ramblings of my "Global Data Store" concept. The latter looks more like a wallet.

pknowles (Sat, 05 Jan 2019 07:36:34 GMT):
@tuckerg @mwherman2000 I think a number of concepts are getting "confuded" here: 1.) Sovrin Foundation has never mentioned a "Global Data Store". That is a concept that I've had running in my mind for the past few years. It's very much a personal concept rather than a Sovrin one. I don't want to confuse that fact at the risk of getting my wrist's slapped by the Technical Ambassadors! 2.) MS's "Intelligent Cloud / Intelligent Edge" model looks like a separate thing to "Bali". The former looks and feels similar to the early ramblings of my "Global Data Store" concept. The latter looks more like a wallet. I envisage that "Bali" will sit on the "Intelligent Edge" part of their future "Intelligent Cloud / Intelligent Edge" model. I hope that "de-confuds" things!

pknowles (Sat, 05 Jan 2019 07:36:34 GMT):
@tuckerg @mwherman2000 I think a number of concepts are getting "confuded" here: 1.) Sovrin Foundation has never mentioned a "Global Data Store". That is a concept that I've had running in my mind for the past few years. It's very much a personal concept rather than a Sovrin one. I don't want to confuse that fact at the risk of getting my wrist's slapped by the Technical Ambassadors! 2.) MS's "Intelligent Cloud / Intelligent Edge" model looks like a separate thing to "Bali". The former looks and feels similar to the early ramblings of my "Global Data Store" concept. The latter looks more like a wallet. I envisage that "Bali" will sit on the "Intelligent Edge" layer of their future "Intelligent Cloud / Intelligent Edge" model. I hope that "de-confuds" things!

pknowles (Sat, 05 Jan 2019 07:36:34 GMT):
@tuckerg @mwherman2000 I think a number of concepts are getting "confuded" here: 1.) Sovrin Foundation has never mentioned a "Global Data Store". That is a concept that I've had running in my mind for the past few years. It's very much a personal concept rather than a Sovrin one. I don't want to confuse that fact at the risk of getting my wrist's slapped by the Technical Ambassadors! 2.) MS's "Intelligent Cloud / Intelligent Edge" model looks like a separate thing to "Bali". The former looks and feels similar to my early "Global Data Store" ramblings. The latter looks more like a wallet. I envisage that "Bali" will sit on the "Intelligent Edge" layer of MS's future "Intelligent Cloud / Intelligent Edge" model. I hope that "de-confuds" things!

pknowles (Sat, 05 Jan 2019 07:36:34 GMT):
@tuckerg @mwherman2000 I think a number of concepts are getting "confuded" here: 1.) Sovrin Foundation has never mentioned a "Global Data Store". That is a concept that I've had running in my mind for the past few years. It's very much a personal concept rather than a Sovrin one. I don't want to confuse that fact at the risk of getting my wrist's slapped by the Technical Ambassadors! 2.) MS's "Intelligent Cloud / Intelligent Edge" model looks like a separate thing to "Bali". The former looks and feels similar to my early "Global Data Store" architecture ramblings. The latter looks more like a wallet. I envisage that "Bali" will sit on the "Intelligent Edge" layer of MS's future "Intelligent Cloud / Intelligent Edge" model. I hope that "de-confuds" things!

pknowles (Sat, 05 Jan 2019 07:36:34 GMT):
@tuckerg @mwherman2000 I think a number of concepts are getting "confuded" here: 1.) Sovrin Foundation has never mentioned a "Global Data Store". That is a concept that I've had running in my mind for the past few years. It's very much a personal concept rather than a Sovrin one. I don't want to confuse that fact at the risk of getting my wrist's slapped by the Technical Ambassadors! 2.) MS's "Intelligent Cloud / Intelligent Edge" model looks like a separate thing to "Bali". The former looks and feels similar to my early "Global Data Store" architecture ramblings. The latter looks more like a wallet. I envisage that "Bali" will sit on the "Intelligent Edge" layer of MS's future "Intelligent Cloud / Intelligent Edge" model. I hope that "de-confuds" things somewhat!

pknowles (Sat, 05 Jan 2019 07:36:34 GMT):
@tuckerg @mwherman2000 I think a number of concepts are getting "confuded" here: 1.) Sovrin Foundation has never touted a "Global Data Store". That is a concept that I've had running in my mind for the past few years. It's very much a personal concept rather than a Sovrin one. I don't want to confuse that fact at the risk of getting my wrist's slapped by the Technical Ambassadors! 2.) MS's "Intelligent Cloud / Intelligent Edge" model looks like a separate thing to "Bali". The former looks and feels similar to my early "Global Data Store" architecture ramblings. The latter looks more like a wallet. I envisage that "Bali" will sit on the "Intelligent Edge" layer of MS's future "Intelligent Cloud / Intelligent Edge" model. I hope that "de-confuds" things somewhat!

pknowles (Sat, 05 Jan 2019 07:36:34 GMT):
@tuckerg @mwherman2000 I think a number of concepts are getting "confuded" here: 1.) Sovrin Foundation has never touted a "Global Data Store" angle in their literature. That is a concept that I've had running in my mind for the past few years. It's very much a personal concept rather than a Sovrin one. I don't want to confuse that fact at the risk of getting my wrist's slapped by the Technical Ambassadors! 2.) MS's "Intelligent Cloud / Intelligent Edge" model looks like a separate thing to "Bali". The former looks and feels similar to my early "Global Data Store" architecture ramblings. The latter looks more like a wallet. I envisage that "Bali" will sit on the "Intelligent Edge" layer of MS's future "Intelligent Cloud / Intelligent Edge" model. I hope that "de-confuds" things somewhat!

pknowles (Sat, 05 Jan 2019 07:36:34 GMT):
@tuckerg @mwherman2000 I think a number of concepts are getting "confuded" here: 1.) Sovrin Foundation has never touted a "Global Data Store" in their literature. That is a concept that I've had running in my mind for the past few years. It's very much a personal concept rather than a Sovrin one. I don't want to confuse that fact at the risk of getting my wrist's slapped by the Technical Ambassadors! 2.) MS's "Intelligent Cloud / Intelligent Edge" model looks like a separate thing to "Bali". The former looks and feels similar to my early "Global Data Store" architecture ramblings. The latter looks more like a wallet. I envisage that "Bali" will sit on the "Intelligent Edge" layer of MS's future "Intelligent Cloud / Intelligent Edge" model. I hope that "de-confuds" things somewhat!

pknowles (Sat, 05 Jan 2019 07:36:34 GMT):
@tuckerg @mwherman2000 I think a number of concepts are getting "confuded" here: 1.) Sovrin Foundation has never touted a "Global Data Store" in their literature. That is a concept that I've had running in my mind for the past few years. It's very much a personal concept rather than a Sovrin one. I don't want to confuse that fact at the risk of getting my wrist's slapped by the Technical Ambassadors! 2.) MS's "Intelligent Cloud / Intelligent Edge" model looks like a separate thing to "Bali". The former looks and feels similar to my early "Global Data Store" architecture ramblings. The latter looks like a wallet. I envisage that "Bali" will sit on the "Intelligent Edge" layer of MS's future "Intelligent Cloud / Intelligent Edge" model. I hope that "de-confuds" things somewhat!

mwherman2000 (Sat, 05 Jan 2019 14:02:34 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=j2nxJPZHe2rJb9XCi) @pknowles I'm not familiar with the term Global Data Store but if you expanded the term to be Global Decentralized Trusted Store for Actors (people and organizations) as well as Things (pets, cars, houses, products, business documents, etc. owned by Actors), that's exactly that my work-in-progress vision for an SSI based solution matches up with.

mwherman2000 (Sat, 05 Jan 2019 14:02:34 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=j2nxJPZHe2rJb9XCi) @pknowles I'm not familiar with the term Global Data Store but if you expanded the term to be Global Decentralized Trusted Store for Actors (people and organizations and software agents) as well as Things (pets, cars, houses, products, business documents, etc. owned by Actors), that's exactly that my work-in-progress vision for an SSI based solution matches up with.

mwherman2000 (Sat, 05 Jan 2019 14:02:34 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=j2nxJPZHe2rJb9XCi) @pknowles I'm not familiar with the term Global Data Store but if you expanded the term to be Global Decentralized Trusted Store for Actors (people and organizations and software agents) as well as Things (pets, cars, houses, products, business documents, etc. owned by Actors), that's exactly that my work-in-progress vision for an SSI based solution matches up with the Global Data Store idea.

mwherman2000 (Sat, 05 Jan 2019 14:02:59 GMT):

path-id-DID-real-life-somethings v0.1.png

mwherman2000 (Sat, 05 Jan 2019 14:02:59 GMT):

path-id-DID-real-life-somethings v0.1.png

mwherman2000 (Sat, 05 Jan 2019 14:02:59 GMT):

path-id-DID-real-life-somethings v0.1.png

mwherman2000 (Sat, 05 Jan 2019 14:06:34 GMT):
https://hyperonomy.com/2019/01/04/the-path-from-a-id-did-to-a-real-life-something/

mwherman2000 (Sat, 05 Jan 2019 14:06:34 GMT):
Here's a new graphic that illustrates this. To dereference the numbers, read . Let me know what you like or don't like. What needs to be added, removed, or changed?

mwherman2000 (Sat, 05 Jan 2019 14:17:37 GMT):

path-id-DID-real-life-somethings v0.1.png

mwherman2000 (Sat, 05 Jan 2019 14:33:58 GMT):

path-id-DID-real-life-somethings v0.2.png

aappddeevv (Sat, 05 Jan 2019 17:31:05 GMT):
I was working through some of the demo scripts/software and videos on youtube. It appears that we should consider that the DID model should be considered gen3 where gen2 was oauth2+openId. I've not seen any material that covers how to frame the did model for application IAM. Should I just be considering did+verifiable credentials to be the new oauth2+openId for IAM?

aappddeevv (Sat, 05 Jan 2019 17:31:05 GMT):
I was working through some of the demo scripts/software and videos on youtube. It appears that the DID model should be considered gen3 where gen2 was oauth2+openId. I've not seen any material that covers how to frame the DID model for application IAM actions. Should I just be considering did+verifiable credentials to be the new oauth2+openId for IAM?

pknowles (Sun, 06 Jan 2019 04:59:28 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=qyERwNpEcpgP8hJDG) @mwherman2000 As we continue to build and implement innovative pieces within the Indy stack, some of these GDS concepts will naturally start to align. Just make sure that the pieces that you need are being aired in the appropriate working groups. BTW, I just came across Finema's Verify dApp and thought that it might be of interest to you. https://verify.finema.co

jljordan_bcgov (Sun, 06 Jan 2019 05:04:47 GMT):
This is a nice paper that should address your question -> https://www.bundesblock.de/2018/10/23/position-paper-self-sovereign-identity/ As well the Sovrin Whitepaper is informative -> https://sovrin.org/wp-content/uploads/2018/03/Sovrin-Protocol-and-Token-White-Paper.pdf

jljordan_bcgov (Sun, 06 Jan 2019 05:04:47 GMT):
@aappddeevv This is a nice paper that should address your question -> https://www.bundesblock.de/2018/10/23/position-paper-self-sovereign-identity/ As well the Sovrin Whitepaper is informative -> https://sovrin.org/wp-content/uploads/2018/03/Sovrin-Protocol-and-Token-White-Paper.pdf

anant706 (Sun, 06 Jan 2019 07:25:01 GMT):
Is Indy Digital ID and Wallet available to end users operating OFF-LINE? If no, then is there any future possibility and solution. Thanks and Regards

anant706 (Sun, 06 Jan 2019 07:26:16 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=7mk6JcqRGwsq37dfd) @pknowles I think https://verify.finema.co is a dummy site. Correct me if I am wrong?

anant706 (Sun, 06 Jan 2019 07:27:08 GMT):
Is Indy Digital ID and Wallet available to end users operating OFF-LINE? If no, then is there any future possibility and solution. Thanks and Regards

anant706 (Sun, 06 Jan 2019 07:27:58 GMT):
Is Indy Digital ID and Wallet available to end users operating OFF-LINE? If no, then is there any future possibility and solution. Thanks and Regards

pknowles (Sun, 06 Jan 2019 08:16:25 GMT):
nage :top: "Is Indy Digital ID and Wallet available to end users operating OFF-LINE?"

pknowles (Sun, 06 Jan 2019 08:16:25 GMT):
@nage :top: "Is Indy Digital ID and Wallet available to end users operating OFF-LINE?"

pknowles (Sun, 06 Jan 2019 08:16:25 GMT):
@nage :top: "Is Indy Digital ID and Wallet available to end users operating OFF-LINE?" - @anant706

pknowles (Sun, 06 Jan 2019 08:57:50 GMT):
@anant706 Finema's Verify dApp is definitely a dummy site. However, long-time contributor and friend to this community, @ewelton , is working closely with Finema on Blockchain based voting. They are in good hands. https://www.ethnews.com/thailand-to-introduce-blockchain-based-voting

pknowles (Sun, 06 Jan 2019 08:57:50 GMT):
@anant706 Finema's Verify dApp is definitely a dummy site. However, long-time contributor and friend to this community, @ewelton , is working closely with Finema on Blockchain based voting. They are being well advised. https://www.ethnews.com/thailand-to-introduce-blockchain-based-voting

pknowles (Sun, 06 Jan 2019 08:57:50 GMT):
@anant706 Finema's Verify dApp is definitely a dummy site. However, long-time contributor and friend to this community, @ewelton , is working closely with Finema re NECTEC and Blockchain based voting. They are being well advised. https://www.ethnews.com/thailand-to-introduce-blockchain-based-voting

frankz (Sun, 06 Jan 2019 13:58:26 GMT):
Has joined the channel.

anant706 (Sun, 06 Jan 2019 19:30:40 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=QHu6GAeY6Wy9E8L5K) @pknowles Thanks.. Actually I am looking for any concrete work going on for Document management System based on Blockchain. Any suggestion?

mwherman2000 (Sun, 06 Jan 2019 20:16:19 GMT):
@pknowles Content Management and Distribution on the blockchain: https://phantasma.io/home

pknowles (Sun, 06 Jan 2019 20:49:29 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=NCAMp7SBqgBHFwycr) @mwherman2000 Thanks for sharing. Following a quick recce, I'm guessing that Phantasma aren't SSI-configured and they look to be heading down the _data reclusion_ route. I'm seeing a number of startups heading down that road but, without non-PII data being made publicly available for societal benefit, I'm not convinced by the model. Due to the bad press coming out of the Facebook/Cambridge Analytica fiasco, startups have tended towards data reclusion. The *Blinding Identity Taxonomy (BIT)* was created to help developers change that mindset. In my humble opinion, any element not included in the BIT can add great value to our tech future.

pknowles (Sun, 06 Jan 2019 20:49:29 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=NCAMp7SBqgBHFwycr) @mwherman2000 Thanks for sharing. Following a quick recce, I'm guessing that Phantasma aren't SSI-configured and they look to be heading down the _data reclusion_ route. I'm seeing a number of startups heading down that road but, without non-PII data being made publicly available for societal benefit, I'm not convinced by the model. Due to the bad press coming out of the Facebook/Cambridge Analytica fiasco, startups have tended towards data reclusion. The *Blinding Identity Taxonomy (BIT)* was created to help change that mindset. In my humble opinion, any element not included in the BIT can add great value to the global tech tapestry.

pknowles (Sun, 06 Jan 2019 20:49:29 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=NCAMp7SBqgBHFwycr) @mwherman2000 Thanks for sharing. Following a quick recce, I'm guessing that Phantasma aren't SSI-configured and they look to be heading down the _data reclusion_ route. I'm seeing a number of startups heading down that road but, without non-PII data being made publicly available for societal benefit, I'm not convinced by the model. Due to the bad press coming out of the Facebook/Cambridge Analytica fiasco, startups have tended towards data reclusion. The *Blinding Identity Taxonomy (BIT)* was created to help change that mindset. In my humble opinion, any element not included in the BIT can add great value to the global data tapestry.

pknowles (Sun, 06 Jan 2019 20:49:29 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=NCAMp7SBqgBHFwycr) @mwherman2000 Thanks for sharing. Following a quick recce, I'm guessing that Phantasma aren't SSI-configured and they look to be heading down the _data reclusion_ route. I'm seeing a number of startups heading down that road but, without non-PII data being made publicly available for societal benefit, I'm not convinced by the model. Due to the bad press coming out of the Facebook/Cambridge Analytica fiasco, startups have tended towards data reclusion. The *Blinding Identity Taxonomy (BIT)* was created to help change that mindset. In my humble opinion, any element not included in the BIT will add value to the global data tapestry.

pknowles (Sun, 06 Jan 2019 20:49:29 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=NCAMp7SBqgBHFwycr) @mwherman2000 Thanks for sharing. Following a quick recce, I'm guessing that Phantasma aren't SSI-configured and they look to be heading down the _data reclusion_ route. I'm seeing a number of startups heading down that road but, without non-PII data being made publicly available for societal benefit, I'm not convinced by the model. Due to the bad press coming out of the Facebook/Cambridge Analytica fiasco, startups have tended towards data reclusion. The *Blinding Identity Taxonomy (BIT)* was created to help change that mindset. In my humble opinion, any element not included in the BIT adds value to the global data tapestry.

pknowles (Sun, 06 Jan 2019 20:49:29 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=NCAMp7SBqgBHFwycr) @mwherman2000 Thanks for sharing. Following a quick recce, I'm guessing that Phantasma aren't SSI-configured and they look to be heading down the _data reclusion_ route. I'm seeing a number of startups heading down that road but, without non-PII data being made publicly available for societal benefit, I'm not convinced by the model. Due to the bad press coming out of the Facebook/Cambridge Analytica fiasco, startups have tended towards data reclusion. The *Blinding Identity Taxonomy (BIT)* was created to help change that mindset. In my humble opinion, any element not included in the BIT adds value to the global data economy.

pknowles (Sun, 06 Jan 2019 20:49:29 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=NCAMp7SBqgBHFwycr) @mwherman2000 Thanks for sharing. Following a quick recce, I'm guessing that Phantasma aren't SSI-configured and they look to be heading down the _data reclusion_ route. I'm seeing a number of startups heading down that road but, without non-PII data being made publicly available for societal benefit, I'm not convinced by the model. Due to the bad press coming out of the Facebook/Cambridge Analytica fiasco, startups have tended towards data reclusion. The *Blinding Identity Taxonomy (BIT)* was created to help change that mindset. In my humble opinion, any element not included in the BIT adds value to the global data economy and should be put to use.

pknowles (Sun, 06 Jan 2019 20:49:29 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=NCAMp7SBqgBHFwycr) @mwherman2000 Thanks for sharing. Following a quick recce, I'm guessing that Phantasma aren't SSI-configured and they look to be heading down the _data reclusion_ route. I'm seeing a number of startups heading down that road but I'm not convinced by the model. Due to the bad press coming out of the Facebook/Cambridge Analytica fiasco, startups have tended towards data reclusion. The *Blinding Identity Taxonomy (BIT)* was created to help change that mindset. In my humble opinion, any element not included in the BIT adds value to the global data economy and should be put to use for societal benefit.

pknowles (Sun, 06 Jan 2019 20:49:29 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=NCAMp7SBqgBHFwycr) @mwherman2000 Thanks for sharing. Following a quick recce, I'm guessing that Phantasma aren't SSI-configured and they look to be heading down the _data reclusion_ route. I'm seeing a number of startups heading down that road but I'm not convinced. Due to the bad press coming out of the Facebook/Cambridge Analytica fiasco, startups have tended towards data reclusion. The *Blinding Identity Taxonomy (BIT)* was created to help change that mindset. In my humble opinion, any element not included in the BIT adds value to the global data economy and should be put to use for societal benefit.

pknowles (Sun, 06 Jan 2019 20:49:29 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=NCAMp7SBqgBHFwycr) @mwherman2000 Thanks for sharing. Following a quick recce, I'm guessing that Phantasma aren't SSI-configured and they look to be heading down the _data reclusion_ route. I'm seeing a number of startups heading down that road but I'm not convinced that is the right approach. Due to the bad press coming out of the Facebook/Cambridge Analytica fiasco, startups have tended towards data reclusion. The *Blinding Identity Taxonomy (BIT)* was created to help change that mindset. In my humble opinion, any element not included in the BIT adds value to the global data economy and should be put to use for societal benefit.

pknowles (Sun, 06 Jan 2019 20:49:29 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=NCAMp7SBqgBHFwycr) @mwherman2000 Thanks for sharing. Following a quick recce, I'm guessing that Phantasma aren't SSI-configured and they look to be heading down the _data reclusion_ route. I'm seeing a number of startups heading down that road but I'm not convinced that that is the right approach. Due to the bad press coming out of the Facebook/Cambridge Analytica fiasco, startups have tended towards data reclusion. The *Blinding Identity Taxonomy (BIT)* was created to help change that mindset. In my humble opinion, any element not included in the BIT adds value to the global data economy and should be put to use for societal benefit.

pknowles (Sun, 06 Jan 2019 20:49:29 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=NCAMp7SBqgBHFwycr) @mwherman2000 Thanks for sharing. Following a quick recce, I'm guessing that Phantasma aren't SSI-configured and they look to be heading down the _data reclusion_ route. I'm seeing a number of startups heading down that road but I'm not convinced that that is the right way to go. Due to the bad press coming out of the Facebook/Cambridge Analytica fiasco, startups have tended towards data reclusion. The *Blinding Identity Taxonomy (BIT)* was created to help change that mindset. In my humble opinion, any element not included in the BIT adds value to the global data economy and should be put to use for societal benefit.

aappddeevv (Sun, 06 Jan 2019 21:38:59 GMT):
@jljordan_bcgov Thanks for those links. I think the content suggests that it is a yes, where the verifier is some software that validates a credential that is issued to the "user". I guess to put time limits on it e.g. revocable at a certain time, the verifiable claim and revocation comes into play. I guess the specific patterns for developing a mobile or web app based on this is a developing story.

aappddeevv (Sun, 06 Jan 2019 21:38:59 GMT):
@jljordan_bcgov Thanks for those links. I think the content suggests that it is a yes, where the verifier is some software that validates a credential that is issued to the "user". I guess to put time limits on it e.g. revocable at a certain time, the verifiable claim and revocation comes into play. I guess the specific patterns for developing a mobile or web app based on IAM this is a developing story.

KukuhTw (Mon, 07 Jan 2019 09:56:04 GMT):
Has joined the channel.

aappddeevv (Mon, 07 Jan 2019 14:52:45 GMT):
I noticed that in the hyperledger videos, there was some mentions of mobile wallet apps for indy. ATB mentioned in their video they were going to release a wallet app they developed (OSS or smartphone?). Is there a reference impl of a mobile wallet released? It is my understanding that the ["education"](https://github.com/swcurran/education/blob/add-indy/LFS171x/indy-material/nodejs/AgentDemoScript.md) demo the web pages for Alice and Bob are web app wallets themselves.

aappddeevv (Mon, 07 Jan 2019 14:52:45 GMT):
I noticed that in the hyperledger videos, there was some mentions of mobile wallet apps for indy. ATB mentioned in their video they were going to release a wallet app they developed (OSS or smartphone?). Is there a reference impl of a mobile wallet released? It is my understanding that the [education](https://github.com/swcurran/education/blob/add-indy/LFS171x/indy-material/nodejs/AgentDemoScript.md) demo the web pages for Alice and Bob are web app wallets themselves.

aappddeevv (Mon, 07 Jan 2019 14:52:45 GMT):
I noticed that in the hyperledger videos, there was some mentions of mobile wallet apps for indy. ATB mentioned in their video they were going to release a wallet app they developed (OSS for smartphone?). Is there a reference impl of a mobile wallet released? It is my understanding that the [education](https://github.com/swcurran/education/blob/add-indy/LFS171x/indy-material/nodejs/AgentDemoScript.md) demo the web pages for Alice and Bob are web app wallets themselves.

sklump (Mon, 07 Jan 2019 15:15:19 GMT):
Has left the channel.

infominer33 (Mon, 07 Jan 2019 18:17:43 GMT):
i'm not 100% if this helps, @aappddeevv However, I'll share a couple links in case they might be of use, and someone who knows better than me will probably have more information: https://www.devteam.space/blog/how-to-build-a-self-sovereign-identity-wallet/ https://github.com/sovrin-foundation/connector-app https://github.com/hyperledger/indy-sdk/blob/master/doc/android-build.md

aappddeevv (Mon, 07 Jan 2019 18:47:07 GMT):
Thanks. That first link is a reply of the general web based demo to some degree. I reviewed the android link...the gist seems to be that its under construction so I'm going to guess that if libindy is the key dependency to a mobile based wallet (agent I guess with a wallet underneath), then a mobile agent/wallet based on it is still in the early stages. The connector-app, which I had not seen before, seems to be targeting the mobile agent/wallet concept square on.

infominer33 (Mon, 07 Jan 2019 18:51:11 GMT):
I recently saw mention of a wallet project proposal, over in #identity-wg

pknowles (Mon, 07 Jan 2019 20:46:23 GMT):
Are there any JSON-LD experts on this channel who can help with some code review?

pknowles (Mon, 07 Jan 2019 20:46:23 GMT):
Are there any JSON-LD experts on this channel who can help with some code review? @/bueller @/anyone

kdenhartog (Mon, 07 Jan 2019 22:38:29 GMT):
@kenebert or @brentzundel might be able to help with this. I believe they've been digging around this to explore things with schemas recently.

kdenhartog (Mon, 07 Jan 2019 22:38:29 GMT):
@kenebert or @brentzundel might be able to help with this. I believe they've been digging around this to explore things with schemas recently. ^

pknowles (Mon, 07 Jan 2019 22:40:35 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=P5iTkuShgEZSNwGic) @kdenhartog Thanks, Kyle. I'm pestering @kenebert already. (Sorry, Ken!)

anant706 (Tue, 08 Jan 2019 03:34:32 GMT):
How to manage documents on Private Blockchain? Any suggestion?

anant706 (Tue, 08 Jan 2019 03:34:32 GMT):
How to manage documents on Private Blockchain? Any suggestion? actually, i am working on Indy digit ID and wallet example but what if I have PDF format of my certificates, and I want that to be uploaded in my Wallet

infominer33 (Tue, 08 Jan 2019 03:37:14 GMT):
you might do better asking that question in #fabric @anant706

infominer33 (Tue, 08 Jan 2019 03:37:35 GMT):
but it's a super vague... so asking a more specific question might help

infominer33 (Tue, 08 Jan 2019 03:38:48 GMT):
Indy supports a public chain

infominer33 (Tue, 08 Jan 2019 03:40:49 GMT):
btw... on a completely different subject. I was thinking it would be cool to have a channel for anything not specifically related to 'indy work' maybe #indy-watercooler ?

infominer33 (Tue, 08 Jan 2019 03:47:45 GMT):
I don't see any "general discussion" channels anywhere in this server, so... idk. I might find some identity related links and see if other people are interested or, maybe even just #watercooler would be good then again, this has probably all been discussed before and decided against? :shrug_tone2:

anant706 (Tue, 08 Jan 2019 03:50:29 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=X294iMBrRDMsXf7GE) @infominer33 Thank you for your time and response

infominer33 (Tue, 08 Jan 2019 04:05:14 GMT):
In direct message, @anant706 asked: >actually, i am working on Indy digit ID and wallet example but what if I have PDF format of my certificates, and I want that to be uploaded in my Wallet.

xnopre (Tue, 08 Jan 2019 12:01:55 GMT):
Hi all. What are the consequences to set "trust anchor" role to any actors in my applications (organisations and each user) ? For more details, I try to implement the consent management by the user. For that, the user will be its own issuer of credential for "consent". He has to create his own "credential definition" and to store this in the ledger, he has to be "trust anchor".

jljordan_bcgov (Tue, 08 Jan 2019 15:38:43 GMT):
This is basically contrary to the design of Indy which is data driven. Verifiable credentials are collections of claims (attributes) that can be individually proven by disclosure, or use of a zero knowledge predicate (including proof of holding without disclosure of the value of the claim). In this design context a monolithic document does not make sense. @anant706 [ ](https://chat.hyperledger.org/channel/indy?msg=Za3wfPY5RauSrjjXt)

jljordan_bcgov (Tue, 08 Jan 2019 15:40:27 GMT):
You could technically encode the blob in some manner and store that in a credential however that is likely not the most elegant solution.

sstone1 (Tue, 08 Jan 2019 16:01:35 GMT):
Has left the channel.

pknowles (Tue, 08 Jan 2019 17:49:08 GMT):
This week's *Indy Semantics WG* call starts in 10 minutes. Zoom link: https://zoom.us/j/2157245727

infominer33 (Wed, 09 Jan 2019 06:05:46 GMT):
we were talking about wallets recently: https://github.com/streetcred-id/indy-sdk-storage

firewater (Wed, 09 Jan 2019 07:03:53 GMT):
Has joined the channel.

firewater (Wed, 09 Jan 2019 07:07:38 GMT):
i am using indy cli and with command pool create stn gen_txn_file=pool_transactions_sandbox_genesis

firewater (Wed, 09 Jan 2019 07:08:55 GMT):
i put the file under indy-sdk/cli folder but it says pool genesis file is invalid or does not exist

firewater (Wed, 09 Jan 2019 07:09:04 GMT):
where should i put it in

firewater (Wed, 09 Jan 2019 07:10:51 GMT):
same thing if i use pool create stn gen_txn_file=docker_pool_transactions_genesis which is the default file located in indy-sdk/cli

firewater (Wed, 09 Jan 2019 07:26:49 GMT):
where is the defult wallet location store when i issue wallet create commend in indy-cli ?

firewater (Wed, 09 Jan 2019 07:26:49 GMT):
where is the defult wallet location store when i issue wallet create command in indy-cli ?

firewater (Wed, 09 Jan 2019 07:26:49 GMT):
what is the path of the default wallet store when i issue wallet create command in indy-cli ?

sergey.minaev (Wed, 09 Jan 2019 13:04:33 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=LZ5AXNMr2nWMAqDtk) @firewater by default CLI use defaults of libindy for wallet create, please find libindy API notes at the next link https://github.com/hyperledger/indy-sdk/blob/master/libindy/src/api/wallet.rs#L162

firewater (Wed, 09 Jan 2019 13:36:54 GMT):
@sergey.minaev can't find $HOME/.indy_client/wallet.

firewater (Wed, 09 Jan 2019 13:36:58 GMT):
.indy_client does not exist

firewater (Wed, 09 Jan 2019 15:16:43 GMT):
hi guys is there any testnet and testnet explorer i can connect to directly in indy without setting up node?

swcurran (Wed, 09 Jan 2019 15:26:45 GMT):
There is a Sovrin Testnet (not sure of the URL right now) and BC Gov has BCovrin at http://159.89.115.24/ that can be used for sandbox purposes. We do reset the ledger once in awhile, so be aware of that. It has a Ledger Browser, a DID Creator and a URL for downloading the genesis file needed for connecting a client to the Ledger. The software behind BCovrin is available at https://github.com/bcgov/von-network for running local copies. All our docker sandboxes are based on that.

firewater (Wed, 09 Jan 2019 17:31:06 GMT):
thanks very much

firewater (Wed, 09 Jan 2019 17:42:16 GMT):
Few questions on setting up 1) Run networks on node # This command requires the publicly accesible ip address of the machine ./manage start proxy_1,proxy_2,proxy_3,proxy_4 & Question: so i just execute this command? or where do i put the publicly accessible ip address? 2) Special requirements for BC Because outbound traffic from openshift is limited, we use proxies to redirect traffic from open ports by overloading ports 80 and 9418 (git). Run 4 extra VMs as proxies (proxy_1, proxy_2, proxy_3, proxy4). Traffic should be forwarded in the following formation: proxy_1:9418 -> von-network:9701 proxy_1:80 -> von-network:9702 Question: I am using amazon aws ec2. How can i forward this?

firewater (Wed, 09 Jan 2019 17:42:16 GMT):
Few questions on setting up 1) Run networks on node # This command requires the publicly accesible ip address of the machine ./manage start proxy_1,proxy_2,proxy_3,proxy_4 & Question: so i just execute this command? or where do i put the publicly accessible ip address? 2) Special requirements for BC Because outbound traffic from openshift is limited, we use proxies to redirect traffic from open ports by overloading ports 80 and 9418 (git). Run 4 extra VMs as proxies (proxy_1, proxy_2, proxy_3, proxy4). Traffic should be forwarded in the following formation: proxy_1:9418 -> von-network:9701 proxy_1:80 -> von-network:9702 Question: I am using amazon aws ec2. How can i forward this?

firewater (Wed, 09 Jan 2019 17:42:16 GMT):
Few questions on setting up von network on aws ec2 1) Run networks on node # This command requires the publicly accesible ip address of the machine ./manage start proxy_1,proxy_2,proxy_3,proxy_4 & Question: so i just execute this command? or where do i put the publicly accessible ip address? 2) Special requirements for BC Because outbound traffic from openshift is limited, we use proxies to redirect traffic from open ports by overloading ports 80 and 9418 (git). Run 4 extra VMs as proxies (proxy_1, proxy_2, proxy_3, proxy4). Traffic should be forwarded in the following formation: proxy_1:9418 -> von-network:9701 proxy_1:80 -> von-network:9702 Question: I am using amazon aws ec2. How can i forward this?

firewater (Wed, 09 Jan 2019 17:42:16 GMT):
Few questions on setting up von network on aws ec2 1) Run networks on node # This command requires the publicly accesible ip address of the machine ./manage start proxy_1,proxy_2,proxy_3,proxy_4 & Question: so i just execute this command? or where do i need to specify the publicly accessible ip address? 2) Special requirements for BC Because outbound traffic from openshift is limited, we use proxies to redirect traffic from open ports by overloading ports 80 and 9418 (git). Run 4 extra VMs as proxies (proxy_1, proxy_2, proxy_3, proxy4). Traffic should be forwarded in the following formation: proxy_1:9418 -> von-network:9701 proxy_1:80 -> von-network:9702 Question: I am using amazon aws ec2. How can i forward this?

firewater (Wed, 09 Jan 2019 17:42:16 GMT):
Few questions on setting up von network on aws ec2 1) Run networks on node # This command requires the publicly accesible ip address of the machine ./manage start proxy_1,proxy_2,proxy_3,proxy_4 & Question: so i just execute this command? or where do i need to specify the publicly accessible ip address? 2) Special requirements for BC Because outbound traffic from openshift is limited, we use proxies to redirect traffic from open ports by overloading ports 80 and 9418 (git). Run 4 extra VMs as proxies (proxy_1, proxy_2, proxy_3, proxy4). Traffic should be forwarded in the following formation: proxy_1:9418 -> von-network:9701 proxy_1:80 -> von-network:9702 Question: - what is the meaning of BC?I am using amazon aws ec2. How can i forward this?

firewater (Wed, 09 Jan 2019 17:42:16 GMT):
Few questions on setting up von network on aws ec2 1) Run networks on node # This command requires the publicly accesible ip address of the machine ./manage start proxy_1,proxy_2,proxy_3,proxy_4 & Question: so i just execute this command? or where do i need to specify the publicly accessible ip address? 2) Special requirements for BC Because outbound traffic from openshift is limited, we use proxies to redirect traffic from open ports by overloading ports 80 and 9418 (git). Run 4 extra VMs as proxies (proxy_1, proxy_2, proxy_3, proxy4). Traffic should be forwarded in the following formation: proxy_1:9418 -> von-network:9701 proxy_1:80 -> von-network:9702 Question: - what is the meaning of BC? - I am using amazon aws ec2. How can i forward this?

firewater (Wed, 09 Jan 2019 17:42:16 GMT):
Few questions on setting up von network on aws ec2 1) Run networks on node # This command requires the publicly accesible ip address of the machine ./manage start proxy_1,proxy_2,proxy_3,proxy_4 & Question: so i just execute this command? or where do i need to specify the publicly accessible ip address? 2) Special requirements for BC Because outbound traffic from openshift is limited, we use proxies to redirect traffic from open ports by overloading ports 80 and 9418 (git). Run 4 extra VMs as proxies (proxy_1, proxy_2, proxy_3, proxy4). Traffic should be forwarded in the following formation: proxy_1:9418 -> von-network:9701 proxy_1:80 -> von-network:9702 Question: - what is the meaning of BC? - I am using ubuntu on amazon aws ec2. How can i forward this?

swcurran (Wed, 09 Jan 2019 19:56:38 GMT):
@firewater - BC is British Columbia. @WadeBarnes can help you with specific questions, I think - you should DM him (unless something generally useful comes out). Wade has set up von-network instances on Digital Ocean about a 1000 times :-). He also runs our OpenShift based instances.

WadeBarnes (Wed, 09 Jan 2019 21:09:05 GMT):
@firewater, I'm responding via DM.

cam-parra (Wed, 09 Jan 2019 21:13:03 GMT):
Has joined the channel.

WadeBarnes (Wed, 09 Jan 2019 21:42:09 GMT):
@firewater, I've also updated the documentation for the repo. I've removed section about special requirements for BC (British Columbia) OpenShift environments. We don't have the outbound port restrictions anymore. That section added a bit of confusion, but was there because for a while we needed to remap the Indy Node ports and IP addresses to route traffic through proxies.

jp1972-01-01 (Wed, 09 Jan 2019 22:51:24 GMT):
Has joined the channel.

firewater (Thu, 10 Jan 2019 02:30:55 GMT):
@WadeBarnes thanks for reply i have dm u as well

ShrinivasDeshmukh (Thu, 10 Jan 2019 13:37:12 GMT):
Has joined the channel.

xvilapueyo (Thu, 10 Jan 2019 13:48:12 GMT):
Has joined the channel.

YogiGolle (Thu, 10 Jan 2019 17:03:12 GMT):
Has joined the channel.

mwherman2000 (Thu, 10 Jan 2019 19:23:04 GMT):
Does anyone know the *davidm* that was on this morning's WG call?

firewater (Sat, 12 Jan 2019 03:29:33 GMT):
is cred definition need to be created once only and reuse or created everytime issuer need to create credential for same schema?

firewater (Sat, 12 Jan 2019 03:29:33 GMT):
is cred definition need to be created once only and reuse or created everytime before issuer create credential for same schema?

infominer33 (Sat, 12 Jan 2019 22:39:15 GMT):
my medium acct got suspended with no explanation, here's a new link to the article I wrote about indy history: https://www.axiomtech.io/blog-feed/hyperledger-indy-decentralized-identity

mwherman2000 (Sun, 13 Jan 2019 15:52:28 GMT):
*Hyperledger Indy/Sovrin Comprehensive Architecture Reference Model (INDY ARM) v0.9* is finally complete (at least the first draft). This is the completed version of what I talked about on last week's Indy WG call. It has morphed and matured a lot since Thursday (even (and most importantly) the name of the ARM has changed). The Narration section is now complete. The *INDY ARM* model contains more than 40 elements. Checkout https://hyperonomy.com/2019/01/13/hyperledger-indy-sovrin-comprehensive-architecture-reference-model-arm/ ...and yes, I continue to be "starving" for additional feedback. CC: @Sean_Bohan @drummondreed @danielhardman @kdenhartog @darrell.odonnell @swcurran @pknowles

mwherman2000 (Sun, 13 Jan 2019 15:52:28 GMT):
DID Comprehensive Architecture Reference Model (DID ARM) v0.8 is finally complete (at least the first draft). This is the completed version of what I talked about on last week's Indy WG call. It has morphed and matured a lot since Thursday. The Narration section is now complete. The DID ARM model contains more than 40 elements. Checkout https://hyperonomy.com/2019/01/10/did-comprehensive-architecture-reference-model-arm/ ...and yes, I continue to be "starving" for addition feedback. CC: @Sean_Bohan @drummondreed @danielhardman @kdenhartog @darrell.odonnell @swcurran

mwherman2000 (Sun, 13 Jan 2019 15:52:28 GMT):
DID Comprehensive Architecture Reference Model (DID ARM) v0.8 is finally complete (at least the first draft). This is the completed version of what I talked about on last week's Indy WG call. It has morphed and matured a lot since Thursday. The Narration section is now complete. The DID ARM model contains more than 40 elements. Checkout https://hyperonomy.com/2019/01/10/did-comprehensive-architecture-reference-model-arm/ ...and yes, I continue to be "starving" for addition feedback. CC: @Sean_Bohan @drummondreed @danielhardman @kdenhartog @darrell.odonnell @swcurran

mwherman2000 (Sun, 13 Jan 2019 15:52:28 GMT):
DID Comprehensive Architecture Reference Model (DID ARM) v0.8 is finally complete (at least the first draft). This is the completed version of what I talked about on last week's Indy WG call. It has morphed and matured a lot since Thursday. The Narration section is now complete. The DID ARM model contains more than 40 elements. Checkout https://hyperonomy.com/2019/01/10/did-comprehensive-architecture-reference-model-arm/ ...and yes, I continue to be "starving" for addition feedback. CC: @Sean_Bohan @drummondreed @danielhardman @kdenhartog @darrell.odonnell @swcurran @pknowles

mwherman2000 (Sun, 13 Jan 2019 15:52:28 GMT):
DID Comprehensive Architecture Reference Model (DID ARM) v0.8 is finally complete (at least the first draft). This is the completed version of what I talked about on last week's Indy WG call. It has morphed and matured a lot since Thursday. The Narration section is now complete. The DID ARM model contains more than 40 elements. Checkout https://hyperonomy.com/2019/01/10/did-comprehensive-architecture-reference-model-arm/ ...and yes, I continue to be "starving" for additional feedback. CC: @Sean_Bohan @drummondreed @danielhardman @kdenhartog @darrell.odonnell @swcurran @pknowles

mwherman2000 (Sun, 13 Jan 2019 15:52:28 GMT):
*HyperLedger Indy/Sovrin Comprehensive Architecture Reference Model (ARM) v0.9* is finally complete (at least the first draft). This is the completed version of what I talked about on last week's Indy WG call. It has morphed and matured a lot since Thursday. The Narration section is now complete. The *INDY ARM* model contains more than 40 elements. Checkout https://hyperonomy.com/2019/01/13/hyperledger-indy-sovrin-comprehensive-architecture-reference-model-arm/ ...and yes, I continue to be "starving" for additional feedback. CC: @Sean_Bohan @drummondreed @danielhardman @kdenhartog @darrell.odonnell @swcurran @pknowles

mwherman2000 (Sun, 13 Jan 2019 15:52:28 GMT):
*HyperLedger Indy/Sovrin Comprehensive Architecture Reference Model (ARM) v0.9* is finally complete (at least the first draft). This is the completed version of what I talked about on last week's Indy WG call. It has morphed and matured a lot since Thursday (even and most importantly, the name has changed). The Narration section is now complete. The *INDY ARM* model contains more than 40 elements. Checkout https://hyperonomy.com/2019/01/13/hyperledger-indy-sovrin-comprehensive-architecture-reference-model-arm/ ...and yes, I continue to be "starving" for additional feedback. CC: @Sean_Bohan @drummondreed @danielhardman @kdenhartog @darrell.odonnell @swcurran @pknowles

mwherman2000 (Sun, 13 Jan 2019 15:52:28 GMT):
*HyperLedger Indy/Sovrin Comprehensive Architecture Reference Model (ARM) v0.9* is finally complete (at least the first draft). This is the completed version of what I talked about on last week's Indy WG call. It has morphed and matured a lot since Thursday (even (and most importantly) the name of the ARM has changed). The Narration section is now complete. The *INDY ARM* model contains more than 40 elements. Checkout https://hyperonomy.com/2019/01/13/hyperledger-indy-sovrin-comprehensive-architecture-reference-model-arm/ ...and yes, I continue to be "starving" for additional feedback. CC: @Sean_Bohan @drummondreed @danielhardman @kdenhartog @darrell.odonnell @swcurran @pknowles

mwherman2000 (Sun, 13 Jan 2019 15:52:28 GMT):
*Hyperledger Indy/Sovrin Comprehensive Architecture Reference Model (ARM) v0.9* is finally complete (at least the first draft). This is the completed version of what I talked about on last week's Indy WG call. It has morphed and matured a lot since Thursday (even (and most importantly) the name of the ARM has changed). The Narration section is now complete. The *INDY ARM* model contains more than 40 elements. Checkout https://hyperonomy.com/2019/01/13/hyperledger-indy-sovrin-comprehensive-architecture-reference-model-arm/ ...and yes, I continue to be "starving" for additional feedback. CC: @Sean_Bohan @drummondreed @danielhardman @kdenhartog @darrell.odonnell @swcurran @pknowlesl

mwherman2000 (Sun, 13 Jan 2019 15:52:28 GMT):
*Hyperledger Indy/Sovrin Comprehensive Architecture Reference Model (ARM) v0.9* is finally complete (at least the first draft). This is the completed version of what I talked about on last week's Indy WG call. It has morphed and matured a lot since Thursday (even (and most importantly) the name of the ARM has changed). The Narration section is now complete. The *INDY ARM* model contains more than 40 elements. Checkout https://hyperonomy.com/2019/01/13/hyperledger-indy-sovrin-comprehensive-architecture-reference-model-arm/ ...and yes, I continue to be "starving" for additional feedback. CC: @Sean_Bohan @drummondreed @danielhardman @kdenhartog @darrell.odonnell @swcurran @pknowles

mwherman2000 (Sun, 13 Jan 2019 16:04:21 GMT):
...I'm especially looking for any changes that need to be made in terminology used in the DID ARM model.

mwherman2000 (Sun, 13 Jan 2019 16:04:21 GMT):
...I'm especially looking for any changes that need to be made in terminology used in the INDY ARM model.

pknowles (Mon, 14 Jan 2019 04:29:27 GMT):
Thanks, @mwherman2000 . Regarding the HL Indy data capture and semantics initiatives that are being implemented, do you envisage that information being documented in this ARM document or should that be held somewhere else?

mwherman2000 (Mon, 14 Jan 2019 04:38:25 GMT):
@pknowles Ideally, I would like to include where/how schema is part of the INDY ARM...I do want it to be comprehensive. I think a set of VC elements needs to be added first (not exactly sure how/where) ...then the schema elements on top of that?? Can you and/or Robert help out with this?

mwherman2000 (Mon, 14 Jan 2019 04:38:25 GMT):
@pknowles Ideally, I would like to include where/how schema is part of the INDY ARM...I do want the ARM to be comprehensive. I think a set of VC elements needs to be added first (not exactly sure how/where) ...then the schema elements on top of that?? Can you and/or Robert help out with this?

mwherman2000 (Mon, 14 Jan 2019 04:38:25 GMT):
@pknowles Ideally, I would like to include where/how schema is part of the INDY ARM...I do want the ARM to be comprehensive. I think a set of VC-related elements needs to be added first (not exactly sure how/where) ...then the schema elements on top of that?? Can you and/or Robert help out with this?

mwherman2000 (Mon, 14 Jan 2019 04:38:25 GMT):
@pknowles Ideally, I would like to include where/how schema is part of the INDY ARM...I do want the ARM to be comprehensive. I think a set of VC-related elements needs to be added first (not exactly sure how/where) ...then the schema-related elements on top of that?? Can you and/or Robert help out with this?

pknowles (Mon, 14 Jan 2019 07:03:31 GMT):
*Verifiable Credentials* overview deck is now stored on the HL Indy server as a Google Doc - https://drive.google.com/drive/u/0/folders/1UxLLugRQKuV8Mdvv_X9Y6ty4szSi5ZNU?ogsrc=32 [Cc: @kenebert (Mr. VC)]

pknowles (Mon, 14 Jan 2019 07:03:31 GMT):
The latest *Verifiable Credentials* overview deck is now stored on the HL Indy server as a Google Doc - https://drive.google.com/drive/u/0/folders/1UxLLugRQKuV8Mdvv_X9Y6ty4szSi5ZNU?ogsrc=32 [Cc: @kenebert (Mr. VC)]

pknowles (Mon, 14 Jan 2019 07:03:31 GMT):
@mwherman2000 The latest *Verifiable Credentials* overview deck is now stored on the HL Indy server as a Google Doc - https://drive.google.com/drive/u/0/folders/1UxLLugRQKuV8Mdvv_X9Y6ty4szSi5ZNU?ogsrc=32 [Cc: @kenebert (Mr. VC)]

pknowles (Mon, 14 Jan 2019 07:03:31 GMT):
@mwherman2000 The latest *Verifiable Credentials* overview deck is stored on the HL Indy server as a Google Doc - https://drive.google.com/drive/u/0/folders/1UxLLugRQKuV8Mdvv_X9Y6ty4szSi5ZNU?ogsrc=32 [Cc: @kenebert (Mr. VC)]

ashokkj (Mon, 14 Jan 2019 09:19:46 GMT):
can anybody tell me What is the significance of "Master Secret key" in user wallet ?

misaelssantos (Mon, 14 Jan 2019 14:14:57 GMT):
Has joined the channel.

swcurran (Mon, 14 Jan 2019 14:52:56 GMT):
@ashokkj - the Master Secret is used by a Holder to create a blinded secret that is given to an Issuer, who then embeds it in a Credential issued back to the Holder. The blinded secret is then proven in zero knowledge by the Holder in a proof. It is the mechanism that proves the Credential was issued to the Holder. An identity uses the same Master Secret (also known in Indy as a Link Secret) for all the Credentials they hold so that they can prove claims from different credentials in a single proof as all being issued to them.

swcurran (Mon, 14 Jan 2019 14:52:56 GMT):
@ashokkj - the Master Secret is used by a Holder to create a blinded secret that is given to an Issuer, who then embeds the blinded secret in a Credential issued back to the Holder. The blinded secret is then proven in zero knowledge by the Holder in a proof. It is the mechanism that proves the Credential was issued to the Holder. An identity uses the same Master Secret (also known in Indy as a Link Secret) for all the Credentials they hold so that they can prove claims from different credentials in a single proof as all being issued to them.

mwherman2000 (Mon, 14 Jan 2019 14:56:05 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=LtzfyBYJ5gnGex7rn) Thk you Paul and @kenebert. Couple comments/questions. 1. Can/should the title on slide 7 be changed from "Credential" to "Verified Credential" to match slide 2? 2. I'm thinking the best way to start is to capture the 3-layer VC data model as a 5th perspective: "VC Data Model" ...renaming the current "Data Model" perspective to specifically be the "DID Data Model". 3. For the elements that are stored "off the ledger", where is the most likely place that these will be stored? 4. For later, what are the additional software components needed to support VCs and Schema? I'll also need some additional information (e.g. Tx types etc. for the Technology layer, etc.). I'll get back to you when I'm ready to add these. I'm still digging deeper into the DID side.

mwherman2000 (Mon, 14 Jan 2019 14:56:05 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=LtzfyBYJ5gnGex7rn) Thk you Paul and @kenebert . Couple comments/questions. 1. Can/should the title on slide 7 be changed from "Credential" to "Verified Credential" to match slide 2? 2. I'm thinking the best way to start is to capture the 3-layer VC data model as a 5th perspective: "VC Data Model" ...renaming the current "Data Model" perspective to specifically be the "DID Data Model". 3. For the elements that are stored "off the ledger", where is the most likely place that these will be stored? 4. For later, what are the additional software components needed to support VCs and Schema? I'll also need some additional information (e.g. Tx types etc. for the Technology layer, etc.). I'll get back to you when I'm ready to add these. I'm still digging deeper into the DID side.

mwherman2000 (Mon, 14 Jan 2019 15:02:54 GMT):

Verifiable Credentials Schema-Slide 2.png

pknowles (Tue, 15 Jan 2019 00:41:43 GMT):
@brentzundel :top:

Jinheon (Tue, 15 Jan 2019 01:54:20 GMT):
Has joined the channel.

GEEKTEDDY (Tue, 15 Jan 2019 05:43:16 GMT):
Has joined the channel.

GEEKTEDDY (Tue, 15 Jan 2019 05:43:38 GMT):
Hi, is there any mature business product or solution which using indy inside? maybe sovrin is one of them.

thiloo (Tue, 15 Jan 2019 10:02:14 GMT):
Has joined the channel.

cam-parra (Tue, 15 Jan 2019 14:52:55 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=W6ia7zLKoMfwwPf8W) @GEEKTEDDY @drummondreed :arrow_up:

cam-parra (Tue, 15 Jan 2019 14:52:55 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=W6ia7zLKoMfwwPf8W) @GEEKTEDDY @drummondreed @nage :arrow_up:

amActiveHello (Wed, 16 Jan 2019 10:25:30 GMT):
Has joined the channel.

mwherman2000 (Wed, 16 Jan 2019 14:11:09 GMT):
Interesting SSIMeetup.org webcast today: http://ssimeetup.org/overview-proposed-pan-canadian-trust-framework-ssi-tim-bouma-webinar-19/

mwherman2000 (Wed, 16 Jan 2019 14:11:09 GMT):
Interesting SSIMeetup.org webcast today: http://ssimeetup.org/overview-proposed-pan-canadian-trust-framework-ssi-tim-bouma-webinar-19/ CC: @trbouma

mwherman2000 (Wed, 16 Jan 2019 14:13:17 GMT):

trbouma1.jpg

firewater (Wed, 16 Jan 2019 15:27:28 GMT):
assuming that i have two phone, phone A and phone B. Each phone has a client agent app. is that possible that i maintain the exact same identity in both phone client agent? (in simple words, how can i sync the same wallet in multiple devices).

swcurran (Wed, 16 Jan 2019 15:42:05 GMT):
@firewater - I think a lot of people are considering that issue, but there is not yet a consensus on what is the right answer - and there may not ever be. Since that question is not really an interoperability issue (interop - agents controlled by different entities being able to collaborate), it has not been a major focus of discussion. Specific agent implementations can handle that issue in different ways and still be able to communicate with other agents/other identities.

swcurran (Wed, 16 Jan 2019 15:42:05 GMT):
@firewater - I think a lot of people are considering that issue, but there is not yet a consensus on what is the right answer - and there may not ever be. Since that question is not really an interoperability issue (interop - agents controlled by different entities being able to collaborate/communicate), it has not been a major focus of discussion. Specific agent implementations can handle that issue in different ways and still be able to communicate with other agents/other identities.

firewater (Wed, 16 Jan 2019 15:44:44 GMT):
@swcurran thanks probably i think in a centralized way like on whatever device that i login to i expect to have same identity

firewater (Wed, 16 Jan 2019 15:46:08 GMT):
is a lot of things that is seems to be very convenient in centralized way seems still not yet been solve in SSI yet?

firewater (Wed, 16 Jan 2019 15:46:08 GMT):
is a lot of things that is seems to be very convenient in centralized way still not yet been solve in SSI yet?

swcurran (Wed, 16 Jan 2019 15:54:10 GMT):
I agree that the user experience should be as you describe. The issues are around the convenience/security tradeoff and how to do that.

firewater (Wed, 16 Jan 2019 17:42:59 GMT):
if somehow the wallet that use for issue credential is lost, could the identity it issued previously still be able to be verified in proof negotiation?

firewater (Wed, 16 Jan 2019 17:42:59 GMT):
if somehow the wallet that use for issue credential is lost, could the identity it issued previously still be able to be verified?

firewater (Wed, 16 Jan 2019 17:45:44 GMT):
scenario: faber college issue credential to Alice and Bob. Alice and bob need proof of faber college credential to apply for loan. But somehow now faber college wallet is lost. So Alice and Bob still can proof that they have faber college credential?

firewater (Wed, 16 Jan 2019 17:45:44 GMT):
scenario: faber college issue student credential to Alice and Bob. Alice and bob need proof of faber college credential to apply for job at ACME Corp. But somehow now faber college wallet is lost. So is that still possible for Alice and Bob to proof that they are faber college student to ACME Corp?

firewater (Wed, 16 Jan 2019 17:45:44 GMT):
scenario: faber college issue student credential to Alice and Bob. Alice and bob need proof of faber college credential to apply for job at ACME Corp. But somehow now faber college wallet is lost. So is that still possible for Alice and Bob to proof to ACME Corp that hey are student of faber college?

firewater (Wed, 16 Jan 2019 17:45:44 GMT):
scenario: faber college issue student credential to Alice and Bob. Alice and bob need proof of faber college credential to apply for job at ACME Corp. But somehow now faber college lost its wallet. So is that still possible for Alice and Bob to proof to ACME Corp that hey are student of faber college?

swcurran (Wed, 16 Jan 2019 18:21:21 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=y9PmZZ3fRAoJKeYoc) @firewater Yes - sort of. The proof could still be proven - the resources needed for that are available to the Holder (who creates the proof) and the Verifier (who verifies the proof). What cannot be done is that the Verifier cannot find out more about how issued the Credential. If they know already Faber's DID (because they have previously verified claims from Faber Credentials), that's OK. But if they have never heard of them, they won't be able to find out who they were - Faber can't respond to any messages from ACME for information about who they are and why they are authorized to issue Transcript Credentials.

sebastianrs95 (Wed, 16 Jan 2019 23:23:44 GMT):
Has joined the channel.

Sean_Bohan (Thu, 17 Jan 2019 02:37:08 GMT):
This week's Hyperledger Indy WG call is tomorrow 1/17: Housekeeping Agents WG, Semantics WG, URSA Indy Catalyst (JohnJ) Indy Roadmap (NathanG) Open Discussion When: 8am PT, 9am MT, 10am CT, 11amET, 3pmGMT Where: Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/232861185 Or iPhone one-tap (US Toll): +16465588656,232861185# or +14086380968,232861185# Or Telephone: Dial: +1 646 558 8656 (US Toll) or +1 408 638 0968 (US Toll) Meeting ID: 232 861 185 International numbers available: https://zoom.us/zoomconference?m=a0jD_rTMnh0ZYGQDOKPCNrK_0dP7WPfp1

vsadriano (Thu, 17 Jan 2019 11:19:38 GMT):
Has joined the channel.

vivekjoshi (Thu, 17 Jan 2019 11:51:33 GMT):
Has joined the channel.

vivekjoshi (Thu, 17 Jan 2019 11:55:12 GMT):
Hello Everyone, I want to learn about hyperledger indy and also want to implement a local hyperledger indy blockchain for doing this i am try to understand about it's blockchan and try to run a node into our local system but i am not able to do it. please help me in this. For how to implement it

vivekjoshi (Thu, 17 Jan 2019 11:55:12 GMT):
Hello Everyone, I want to learn about hyperledger indy and also want to implement a local hyperledger indy blockchain for doing this i am try to understand about it's blockchan and try to run a node into our local system but i am not able to do it. please help me in this. For how to implement it. Thanks in advance.

xnopre (Thu, 17 Jan 2019 14:41:23 GMT):
Hi all :-) . Is `DID Auth` just a concept, an idea ? Or is there somewhere a standardization or a spec ? And even an implementation ? Thanks

firewater (Thu, 17 Jan 2019 15:18:36 GMT):
how does one know a credential belongs to which schema in ledger?

firewater (Thu, 17 Jan 2019 15:18:36 GMT):
how does one know a CRED_DEF belongs to which SCHEMA in ledger?

firewater (Thu, 17 Jan 2019 15:18:36 GMT):
how does one know a CRED_DEF belongs to which SCHEMA in ledger? and vice versa

firewater (Thu, 17 Jan 2019 15:18:58 GMT):
and the other way also, how does one know this schema have what credential def?

firewater (Thu, 17 Jan 2019 15:18:58 GMT):
and the other way also, how does one know this schema have what credential def tie to it?

firewater (Thu, 17 Jan 2019 15:18:58 GMT):
and the other way also, how does one know a schema have what credential def tie to it?

firewater (Thu, 17 Jan 2019 15:20:29 GMT):
eg: i created a SCHEMA A and a CRED_DEF under SCHEMA A

firewater (Thu, 17 Jan 2019 15:21:02 GMT):
1) given the schema ID, how can i locate all the CRED_DEF under it

firewater (Thu, 17 Jan 2019 15:21:02 GMT):
1) given the schema ID, how can i locate all the corresponding CRED_DEF

firewater (Thu, 17 Jan 2019 15:21:02 GMT):
1) given the SCHEMA ID, how can i locate all the corresponding CRED_DEF SC

firewater (Thu, 17 Jan 2019 15:21:13 GMT):
2) Given the CRED_DEF, how can i locate the SCHEMA?

firewater (Thu, 17 Jan 2019 15:21:13 GMT):
2) Given the CRED_DEF, how can i locate the corresponding SCHEMA?

firewater (Thu, 17 Jan 2019 15:21:13 GMT):
2) Given the CRED_DEF ID, how can i locate the corresponding SCHEMA?

firewater (Thu, 17 Jan 2019 15:22:53 GMT):
looking from the data that is recorded in the ledger, i can't see any way to identify it.

firewater (Thu, 17 Jan 2019 15:22:53 GMT):
looking from the data that is recorded in the ledger, i can't see any way to identify it ( i suppose to SCHEMA id will be embedded in CRED_DEF ID but i couldn't find it)

firewater (Thu, 17 Jan 2019 15:22:53 GMT):
looking from the data that is recorded in the ledger, i can't see any way to identify it ( i suppose to SCHEMA id will be embedded in CRED_DEF ID in the ledger but i couldn't find it)

swcurran (Thu, 17 Jan 2019 17:33:05 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=BmywSSEGaYWcamPpW) @vivekjoshi A good way to get something running to play with and review is the Indy chapter of Hyperledger's Blockchain for Business edX course. The workshop for that course material can be found here, and that will get you up and running with a small Indy implementation using only a browser, or on your local machine. https://github.com/hyperledger/education/tree/master/LFS171x/indy-material/nodejs

swcurran (Thu, 17 Jan 2019 17:38:11 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=XNdTzbztkxdbkM2B7) @xnopre Hi - BC Gov funded some great DID-Auth work by @peacekeeper that was donated to the Decentralized Identity Foundation (DIF - https://identity.foundation/) and can be found in their repos (link below). BC Gov has that implemented and in use in our VON production implementation for system-to-system authentication. https://github.com/decentralized-identity?utf8=%E2%9C%93&q=AUTH&type=&language=

swcurran (Thu, 17 Jan 2019 17:44:51 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=DuzmGjPhpbg8Mp3QP) @firewater CredDef -> Schema discovery is easy - there is a reference in the CredDef to the Schema. There are not currently any indy-node operations (and hence, any indy-sdk calls) that provide Schema -> CredDef discovery. BC Gov has implemented a Ledger Browser by querying the Ledger for all transactions. Basically, it polls the ledger for the latest transactions by number and stores the results in a database. Kinda crude, but it works and gives use that kind of discovery mechanism. You can see it in use here: http://prod.bcovrin.vonx.io/ Code for that is in: https://github.com/bcgov/von-network von-network is a tool for running local or cloud-based Indy networks.

swcurran (Thu, 17 Jan 2019 17:44:51 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=DuzmGjPhpbg8Mp3QP) @firewater CredDef -> Schema discovery is easy - there is a reference in the CredDef to the Schema. There are not currently any indy-node operations (and hence, any indy-sdk calls) that provide Schema -> CredDef discovery. BC Gov has implemented a Ledger Browser by querying the Ledger for all transactions. Basically, it polls the ledger for the latest transactions by number and stores the results in a database. Kinda crude, but it works and gives us that kind of discovery mechanism. You can see it in use here: http://prod.bcovrin.vonx.io/ Code for that is in: https://github.com/bcgov/von-network von-network is a tool for running local or cloud-based Indy networks.

swcurran (Thu, 17 Jan 2019 17:44:51 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=DuzmGjPhpbg8Mp3QP) @firewater CredDef -> Schema discovery is easy - there is a reference in the CredDef to the Schema. There are not currently any indy-node operations (and hence, any indy-sdk calls) that provide Schema -> CredDef discovery. BC Gov has implemented a Ledger Browser by querying the Ledger for all transactions. Basically, it polls the ledger for the latest transactions by number and stores the results in a database. Kinda crude, but it works and gives us that kind of discovery mechanism. You can see it in use here: http://prod.bcovrin.vonx.io/ - click the "Domain" link to see transaction on the ledger. Code for that is in: https://github.com/bcgov/von-network von-network is a tool for running local or cloud-based Indy networks.

firewater (Thu, 17 Jan 2019 18:39:07 GMT):
@swcurran perhaps i should refine my question, i can get the reference number of SCHEMA by querying CredDef. However how can i return response of SCHEMA since the sdk accepts SCHEMA ID is input but not reference number

firewater (Thu, 17 Jan 2019 18:39:07 GMT):
@swcurran perhaps i should refine my question, i can get the reference number of SCHEMA by querying CredDef. However how can i return response of SCHEMA since the sdk accepts SCHEMA ID as input but not reference number

SaraInadam (Thu, 17 Jan 2019 18:39:16 GMT):
Hi every one ,can anyone please tell me if HL Indy is a blockchain framework like fabric or a software am already familiar with fabric but am a bit confused about Indy .

firewater (Thu, 17 Jan 2019 18:41:26 GMT):
await sdk.buildGetSchemaRequest(did, id) accepts schemaID but the CRED DEF response returns reference number to schema

firewater (Thu, 17 Jan 2019 18:48:29 GMT):
@swcurran the getCredDefResponse in sdk returns follow: schemaId: '28', but this is not the correct schema id format to be pass in into getSchemaRequest.

firewater (Thu, 17 Jan 2019 18:49:10 GMT):
the correct schema id foramt is something like follow: Th7MpTaRZVRYnPiabds81Y:2:Government-ID-XX:1.1

firewater (Thu, 17 Jan 2019 18:49:10 GMT):
the correct schema id foramt to be pass in is something like follow: Th7MpTaRZVRYnPiabds81Y:2:Government-ID-XX:1.1

kdenhartog (Thu, 17 Jan 2019 19:07:19 GMT):
@xnopre also worth noting that there's yet to be a spec for Indy's implementation of DID Auth yet, but it's essentially going to be credential exchange. On the other hand the W3C does want to also standardize a specification I've heard. @peacekee

kdenhartog (Thu, 17 Jan 2019 19:07:19 GMT):
@xnopre also worth noting that there's yet to be a spec for Indy's implementation of DID Auth yet, but it's essentially going to be credential exchange. On the other hand the W3C does want to also standardize a specification I've heard. @peacekeeper has been the one who's done the most on this topic.

swcurran (Thu, 17 Jan 2019 19:41:32 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=3kmJNZ8HMkXyg53yY) @firewater Ah...I didn't realize that. Sigh. Then you would be back to the same answer as I said for CredDef from Schema - tracking the transactions on the ledger. That's not good. @gudkov - are we wrong on this? CredDef returns Schema ID, but not transaction ID, and there is not an easy way to go from Schema ID to the Schema transaction?

rdbmsdata78 (Fri, 18 Jan 2019 00:29:33 GMT):
Has joined the channel.

KokoKoko (Fri, 18 Jan 2019 03:20:29 GMT):
Has joined the channel.

sahanmal (Fri, 18 Jan 2019 05:26:55 GMT):
Has joined the channel.

Yair (Fri, 18 Jan 2019 11:00:03 GMT):
Has joined the channel.

haniavis (Fri, 18 Jan 2019 21:31:29 GMT):
@swcurran is there a way to run the The VON-Network Ledger Browser and API on top of my local indy network and pool?

haniavis (Fri, 18 Jan 2019 21:31:42 GMT):
In the github readme, I can see installation instructions for deploying von-network together with a pool/network of nodes using docker.

swcurran (Fri, 18 Jan 2019 21:49:08 GMT):
@haniavis - Should be able to. Some thoughts on doing that: - The Ledger Browser code is in the `/server` folder and there is a start webserver bash script in the `/scripts` folder. That's the main part you would have to be able to run. - You can look in the dockerfile and docker-compose.yml files to see how it is built and deployed. - I suspect you could just change the place in the code (script) where it is getting the genesis file use the one from your local code. If you are still at a deadend, I could connect you to the dev that wrote it.

haniavis (Fri, 18 Jan 2019 21:50:03 GMT):
OK thank you! I will give it a try

firewater (Fri, 18 Jan 2019 21:58:21 GMT):
suppose that i have a website that require a login integration with hyperledger indy, how can the architecture/flow be done?

firewater (Fri, 18 Jan 2019 22:00:37 GMT):
eg website ACME issue a credential to user A wallet. how can user a use his/her indy agent to login into website ACME

firewater (Fri, 18 Jan 2019 22:00:37 GMT):
eg website ACME issue a credential to user A wallet. how can user a use his/her indy agent to proof its identity login into website ACME (without typing username and password)

firewater (Fri, 18 Jan 2019 22:00:37 GMT):
eg website ACME issue a credential to user A wallet. how can user a use his/her indy agent to proof its identity and login into website ACME (without typing username and password)

kdenhartog (Fri, 18 Jan 2019 23:09:51 GMT):
@firewater here's a presentation I gave on this topic awhile back. Take a look at the 3rd flow on page 4. In this case, Alice and the 3rd party issuer would be Acme.

kdenhartog (Fri, 18 Jan 2019 23:09:54 GMT):
https://docs.google.com/presentation/d/1LiVi3wmUNFxUc5AXQyZVW2Gn1X7hcS9obgQ6uuoNZTI/edit

sohanln (Sun, 20 Jan 2019 11:02:29 GMT):
Has joined the channel.

sohanln (Sun, 20 Jan 2019 11:07:14 GMT):
any suggestion for installation of libindy ; repositry has been moved so cant install it correctly

sohanln (Sun, 20 Jan 2019 11:07:18 GMT):

Clipboard - January 20, 2019 4:37 PM

Basil-3 (Sun, 20 Jan 2019 12:26:40 GMT):
Has joined the channel.

haggs (Sun, 20 Jan 2019 19:05:37 GMT):
@sohanln It appears you're dealing with node.js dependencies, have you tried blasting the `node_modules` directory apart?

haggs (Sun, 20 Jan 2019 19:06:24 GMT):
Also, are you following the nodejs wrapper guide, you might need a dynamic link to `/usr/lib/libindy.so` before using the wrapper

haggs (Sun, 20 Jan 2019 19:07:04 GMT):
I would follow these steps since you appear to be on Ubuntu 16.04 https://github.com/hyperledger/indy-sdk#ubuntu-based-distributions-ubuntu-1604 and then trying the `npm install...` commands

firewater (Mon, 21 Jan 2019 02:00:01 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=jbLYjPnyqEeBP9afS) @kdenhartog which is the 3rd flow? the DID authentication and 3rd party authorization?

kdenhartog (Mon, 21 Jan 2019 03:33:25 GMT):
yup. Those 3 boxes are supposed to be 3 separate flows that represent DID Auth.

firewater (Mon, 21 Jan 2019 05:34:45 GMT):
@kdenhartog i still didn't get it how does it use for login and maintain session is the site?

firewater (Mon, 21 Jan 2019 05:34:45 GMT):
@kdenhartog i still didn't get it how does it use for login and maintain login session in the site?

kdenhartog (Mon, 21 Jan 2019 05:41:46 GMT):
login goes through a connection protocol setup, and once the connection is setup, credentials can be passed to increase privilege of the user to account the different role levels that the website has.

kdenhartog (Mon, 21 Jan 2019 05:42:16 GMT):
So, if it's just a general user using the website, they just need to go through the connection protocol. If it's an admin of the website, they need to present a credential.

kdenhartog (Mon, 21 Jan 2019 05:43:37 GMT):
Ultimately, every form of authentication relies on the setup of a connection which exchanges keys. These keys can then be used to derive authenticated symmetrical encryption keys which is how the user is authenticated. As long as both parties can encrypt and decrypt messages, the "session" remains valid.

firewater (Mon, 21 Jan 2019 05:46:36 GMT):
so every exchange of information that requires login credential requires an exchange of keys?

firewater (Mon, 21 Jan 2019 05:46:37 GMT):
Actually i am expecting something like this: 1) In traditional login i need use username and password to login 2) Now, let say i have a credential issue from ACME. I tap on the ACME credential given and it directly login to ACME website or mobile app. 3) Every other connection then it will be done on ACME website or mobile app in a login session stored Not sure if this is the correct way to do with SSI login and how can i implement that.

firewater (Mon, 21 Jan 2019 05:46:37 GMT):
Actually i am expecting something like this: 1) In traditional login i need use username and password to login into a site 2) Now, let say i have a credential issue from ACME. I tap on the ACME credential given and it directly login to ACME website or mobile app. 3) Every other connection then it will be done on ACME website or mobile app in a login session stored Not sure if this is the correct way to do with SSI login and how can i implement that.

firewater (Mon, 21 Jan 2019 05:46:37 GMT):
Actually i am expecting something like this: 1) In traditional login i need use username and password to login into a site 2) Now, let say i have a credential issue from ACME. I tap on the ACME credential given and it directly login to ACME website or mobile app. 3) Every other action then i perform on ACME website or mobile app will then treated as the login user performing the session until the login session expired. Not sure if this is the correct way to do with SSI login and how can i implement that.

firewater (Mon, 21 Jan 2019 05:46:37 GMT):
Actually i am expecting something like this: 1) In traditional login i need use username and password to login into a site 2) Now, let say i have a credential issue from ACME. I tap on the ACME credential given in my mobile app client and it directly login to ACME website or mobile app. 3) Every other action then i perform on ACME website or mobile app will then treated as the login user performing the session until the login session expired. Not sure if this is the correct way to do with SSI login and how can i implement that.

firewater (Mon, 21 Jan 2019 05:48:01 GMT):
basically like single sign on login like dat

kdenhartog (Mon, 21 Jan 2019 05:56:25 GMT):
this is one flow that could work. The way I've imagined this is that the user would show up at the website, and click a "login with sovrin" button that would setup a connection between the user's browser agent and the website's server. Then once the connection is setup, the website could request proofs that meet the business flows. Then when the user returns the browser agent would recognize the URL (through a browser handler potentially) and submit a trust ping automatically on behalf of the user, and if correctly responded to, the user would have their data loaded and would be "logged in". If any part fails, such as the trust ping, then the website would load the original sign up page that the user originally saw.

kdenhartog (Mon, 21 Jan 2019 05:56:25 GMT):
this is one flow that could work. The way I've imagined this is that the user would show up at the website, and click a "login with sovrin" (or Indy) button that would setup a connection between the user's browser agent and the website's server. Then once the connection is setup, the website could request proofs that meet the business flows. Then when the user returns the browser agent would recognize the URL (through a browser handler potentially) and submit a trust ping automatically on behalf of the user, and if correctly responded to, the user would have their data loaded and would be "logged in". If any part fails, such as the trust ping, then the website would load the original sign up page that the user originally saw.

kdenhartog (Mon, 21 Jan 2019 06:03:17 GMT):
As I reread your message, one thing I want to point out is that sessions don't "expire", keys do. This is important because in traditional session based systems, they rely ona a caching mechanism between sessions to have good UX. However, the caching mechanism is the point most often exploited in traditional IAM.

firewater (Mon, 21 Jan 2019 06:03:26 GMT):
that would be the case for browser agent. what about a mobile app agent to ACME browser and mobile app interaction?

firewater (Mon, 21 Jan 2019 06:03:53 GMT):
especially from mobile app to mobile app interaction, i really don't see a way for that.

firewater (Mon, 21 Jan 2019 06:03:53 GMT):
especially for client mobile agent to ACME mobile app interaction, i really don't see a way for that.

kdenhartog (Mon, 21 Jan 2019 06:04:17 GMT):
Different agents but similar flows. Rather than using a URL handler you could use something like a schema handler.

kdenhartog (Mon, 21 Jan 2019 06:07:39 GMT):
Essentially what you're wondering about is connection protocol setup which is actively being developed as an Indy-HIPE PR right now. We've been investigating this and believe we have a solution where mobile OS and browsers like android, iOS, and Chromium based browsers, provide us with the tools to make this possible.

kdenhartog (Mon, 21 Jan 2019 06:08:47 GMT):
https://developer.android.com/training/app-links/

kdenhartog (Mon, 21 Jan 2019 06:09:49 GMT):
this is how it's done in Android. There's similar capabilities in iOS and chrome. For example, slack uses `slack://` to direct users to their local client on a computer.

Basil-3 (Mon, 21 Jan 2019 12:51:17 GMT):
Is there anyone who can help me a bit. I am not able to execute the last line of dev-setup of Hyperledger Indy (pip install -e .[tests]) so can anyone let me know how to execute this?

DirkKrueger (Mon, 21 Jan 2019 13:49:17 GMT):
Has joined the channel.

Basil-3 (Tue, 22 Jan 2019 04:54:04 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=x5fh3NYnQh7e3haFn) I am getting an error : Directory '.' is not installable. File 'setup.py' not found

RahulToraskar (Tue, 22 Jan 2019 08:43:05 GMT):
Has joined the channel.

amActiveHello (Tue, 22 Jan 2019 09:33:25 GMT):
_do_call: Function indy_prover_get_credentials_for_proof_req returned error 113 Error occurred: (, 'Error: Invalid structure\n Caused by: Invalid ProofRequest json has been passed\n Caused by: missing field `name` at line 1 column 341\n')

amActiveHello (Tue, 22 Jan 2019 09:33:44 GMT):
9. Prover gets Credentials for Proof Request Proof Request: {'name': 'proof_req_1', 'nonce': '123432421212', 'requested_attrs': {'attr1_referent': {'name': 'name', 'restrictions': [{'issuer_did': 'NcYxiDXkpYi6ov5FcYDi1e'}]}}, 'requested_predicates': {'predicate1_referent': {'attr_name': 'age', 'p_type': '>=', 'restrictions': [{'issuer_did': 'NcYxiDXkpYi6ov5FcYDi1e'}], 'value': 18}}, 'version': '0.1'} ('{"nonce": "123432421212", "name": "proof_req_1", "version": "0.1", ' '"requested_attrs": {"attr1_referent": {"name": "name", "restrictions": ' '[{"issuer_did": "NcYxiDXkpYi6ov5FcYDi1e"}]}}, "requested_predicates": ' '{"predicate1_referent": {"attr_name": "age", "p_type": ">=", "value": 18, ' '"restrictions": [{"issuer_did": "NcYxiDXkpYi6ov5FcYDi1e"}]}}}') _do_call: Function indy_prover_get_credentials_for_proof_req returned error 113 Error occurred: (, 'Error: Invalid structure\n Caused by: Invalid ProofRequest json has been passed\n Caused by: missing field `name` at line 1 column 341\n')

pknowles (Tue, 22 Jan 2019 17:00:33 GMT):
This week's *Indy Semantics WG* call starts in 1 hour. Zoom link: https://zoom.us/j/2157245727

jljordan_bcgov (Tue, 22 Jan 2019 18:28:31 GMT):
I'm very happy and proud to officially announce OrgBook BC ... the Minister of Citizens' Services announce it at Identity North https://www.identitynorth.ca/events/identitynorth-western-workshop-2019/ and we have a press release here https://news.gov.bc.ca/releases/2019CITZ0002-000062

jljordan_bcgov (Tue, 22 Jan 2019 18:29:00 GMT):
Many thanks to this community ... would not have been possible without open source software and in particular Hyperledger Indy

jljordan_bcgov (Tue, 22 Jan 2019 18:31:06 GMT):
If you are interested in the under the covers information ... check out https://vonx.io

coderintherye (Tue, 22 Jan 2019 18:56:18 GMT):
fyi that https://www.hyperledger.org/projects/hyperledger-indy links to https://jira.hyperledger.org/projects/INDY/summary which has the following error displayed: " This gadget cannot be displayed on your dashboard. This could be due to a licensing problem or an application error. If you need this gadget, contact your administrator for assistance. Otherwise, you can remove it from your dashboard."

drummondreed (Tue, 22 Jan 2019 19:21:12 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=mFxB8qYcLzTwrn5bt) Huge contrats to you and the whole BC Gov and OrgBook team. A very major milestone for Hyperledger Indy and Sovrin and the entire SSI community.

nemesisinhell (Wed, 23 Jan 2019 12:23:05 GMT):
Has joined the channel.

nemesisinhell (Wed, 23 Jan 2019 12:30:16 GMT):
Hi fellas, 1) in hyperledger indy, does the owner of the node role is steward ? 2) in hyperledger indy, in order to establish connection end user with trust anchor, does trust anchor had to host a api server to process the connection request ? 3) in hyperledger indy, what type of format or source can trust anchor provide in order to let end user get it's data such as DID , verkey for establish connection ?

nemesisinhell (Wed, 23 Jan 2019 12:32:04 GMT):
Question: we are thinking of scanning qr code of cloud agent from client app to establish connection with them. What info does the qr code in cloud ageny should provide and is there a standard format of how it should have?

drummondreed (Wed, 23 Jan 2019 13:19:33 GMT):
Hey, these are all excellent questions. Quickly (since I'm about to board a plane): 1) yes, the role of a node owner is called a steward. 2) I think the simple answer is "yes", but the devil is in the details. Many stewards will be trust anchors (in Sovrin all stewards are automatically trust anchors), so stewards are already hosting the necessary API. 3) QR codes are currently one of the main formats being used to bootstrap a connection. As for the details on that, I'll defer to the geniuses on this channel.

nemesisinhell (Wed, 23 Jan 2019 13:53:07 GMT):
@drummondreed , thanks for the answer. There another question : is it possible to update credential which contain data of schema ? or we need revoke and create it again ?

swcurran (Wed, 23 Jan 2019 14:40:02 GMT):
@nemesisinhell - there is a repo of HL Indy Project Enhancements (HIPE), which has lots of information on direction for agents. Some are accepted and some are in pull requests. About connections and QR codes, check out this in process HIPE - https://github.com/hyperledger/indy-hipe/blob/250d65e224b982fa56a6521f4a10b4b2c25bba81/text/connection-protocol/README.md Regards your last question - that's a revoke and reissue process. You can't change an issued credential. We expect that to the most common need for revocation.

mero (Wed, 23 Jan 2019 16:12:05 GMT):
Hi, in Indy, is it correct that revoking a credential prevents further valid proofs from being created with that credential but previous proofs will still verify as valid (if using their timestamp)?

swcurran (Wed, 23 Jan 2019 16:15:27 GMT):
A valid proof can be created by using ledger data from a point in time before the revocation - e.g. when the credential was created. That might be acceptable to the verifier - it's up to them to decide if that works for them. If they can see from the ledger that revocations have occured later, they could request a fresher proof - one from a later point in time. All of that depends on the use case. What is good enough to mitigate their risk for the transaction they are trying to accomplish?

mero (Wed, 23 Jan 2019 16:22:49 GMT):
Ah, ok, but the prover wouldn't be able to provide a proof with a timestamp later than the revocation? And if the verifier accepts the older timestamp (i.e. verifies with revocation data from the ledger up-to that timestamp) the proof would be valid?

mero (Wed, 23 Jan 2019 16:22:49 GMT):
Ah, ok, but the prover wouldn't be able to provide a (valid) proof with a timestamp later than the revocation? And if the verifier accepts the older timestamp (i.e. verifies with revocation data from the ledger up-to that timestamp) the proof would be valid?

swcurran (Wed, 23 Jan 2019 16:26:44 GMT):
Yes and Yes.

mero (Wed, 23 Jan 2019 16:28:58 GMT):
Awesome, that really helped my understanding. Thanks for the quick reply!

AlexanderVtyurin (Wed, 23 Jan 2019 17:19:03 GMT):
Has joined the channel.

ShivanshVij (Wed, 23 Jan 2019 18:49:54 GMT):
Has joined the channel.

ShivanshVij (Wed, 23 Jan 2019 18:50:17 GMT):
Hey everyone, I'm wondering if there's a way to make some sort of explorer for indy

ShivanshVij (Wed, 23 Jan 2019 18:51:16 GMT):
Or if there's a way to use the hyperledger explorer code for indy

kdenhartog (Wed, 23 Jan 2019 21:52:35 GMT):
@swcurran and the team at BC Gov have built a ledger explorer. Try connecting with them to get that working.

swcurran (Wed, 23 Jan 2019 21:53:32 GMT):
You can see an example here - http://dflow.bcovrin.vonx.io/

swcurran (Wed, 23 Jan 2019 21:54:29 GMT):
The code for it is in https://github.com/bcgov/von-network - look at the server folder and the `webserver` service in the docker-compose.yml file.

ShivanshVij (Wed, 23 Jan 2019 21:55:52 GMT):
@swcurran Thank you so much!

wangdong (Thu, 24 Jan 2019 00:17:57 GMT):
Hi, I want to understand the segments in genesis pool file. Where can I find the specs if any?

adamgering (Thu, 24 Jan 2019 02:57:03 GMT):
Has joined the channel.

kdenhartog (Thu, 24 Jan 2019 16:49:48 GMT):
Use cases document for anyone who'd like to help add different ways the Indy project can be used. https://docs.google.com/document/d/1EMiqzdd-6TB2puzamDGJSLrDTwUz_QW46xdEp2FLnbI/edit?usp=sharing

SteveGoob (Thu, 24 Jan 2019 17:14:51 GMT):
^^^ This is an important document for graduation from incubation. We'll turn this into a wiki page once we get a proper mass of examples.

mogamboizer (Thu, 24 Jan 2019 18:26:20 GMT):
I want to deploy a sample Indy instance and use try out some basic DID operations how can I get started? Thanks.

mogamboizer (Thu, 24 Jan 2019 18:26:20 GMT):
I want to deploy a sample Indy instance and use try out some basic DID operations how can I get started? Thanks. I've worked with Fabric but not sure how I can quickly get started with Indy and setup some ID based applications.

kdenhartog (Thu, 24 Jan 2019 20:44:06 GMT):
@mogamboizer checkout this repo https://github.com/hyperledger/education/tree/master/LFS171x/indy-material/nodejs and https://github.com/kdenhartog/indy-dev Additionally for things related to onboarding and getting started in the ecosystem, hop on over to #indy-outreach and they'll be happy to help there.

shader (Thu, 24 Jan 2019 22:22:54 GMT):
Has joined the channel.

ShivanshVij (Fri, 25 Jan 2019 00:11:20 GMT):
I was just wondering, when issuing credentials, what information actually gets published to the blockchain?

swcurran (Fri, 25 Jan 2019 00:36:02 GMT):
@ShivanshVij - no information gets published to the blockchain. A credential goes from the Issuer to the Holder - and into the Holder's wallet.

breemark (Fri, 25 Jan 2019 13:39:16 GMT):
Has joined the channel.

alexmayrhofer (Fri, 25 Jan 2019 15:01:21 GMT):
Has joined the channel.

MikeRichardson (Fri, 25 Jan 2019 15:16:52 GMT):
Has joined the channel.

ShivanshVij (Fri, 25 Jan 2019 15:34:13 GMT):
So during verification what does indy do to verify that a credential is legitimate? What does it compare to the blockchain?

mwherman2000 (Fri, 25 Jan 2019 16:10:46 GMT):
For Your Humor ...I've got this tune buzzing around in my head this morning (change "His Hands" to "His Wallet" :-)): https://www.youtube.com/watch?v=kDBJNQiugnM Happy Friday :-)

mwherman2000 (Fri, 25 Jan 2019 16:10:46 GMT):
For Your Humor ...I've got this tune buzzing around in my head this morning (change "His Hands" to "his wallet" :-)): https://www.youtube.com/watch?v=kDBJNQiugnM Happy Friday :-)

swcurran (Fri, 25 Jan 2019 18:31:03 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=ZabpbbEeZuG3iXyJC) @ShivanshVij A proof in Indy proves 4 things: - the issuer of the credential. This is through the Credential Definition that is in the Credential and is found on the Ledger. - that the identity providing the credential is the holder. This is through a "link secret" that is embedded in the Credential and provable only by the holder - that the Credential has not been tampered with. This is based on a the signing of the credential based on public keys that are in the Credential Def. - that the credential has not been revoked. This is based on revocation information found on the ledger and linked to the Credential Def. All of that data is either shared between the Holder and Verifier or pulled from the Ledger. The only additional thing that the Verifier has to do on their own - determine if they trust things that the Issuer says.

lynn.bendixsen (Fri, 25 Jan 2019 19:07:30 GMT):
@wangdong I found the following which I think answers your question from a few days ago. Search for "Genesis Transactions" at the link for a high level explanation and then read down from there for the syntax you seek: https://github.com/hyperledger/indy-node/blob/master/docs/transactions.md

wangdong (Fri, 25 Jan 2019 21:28:02 GMT):
@lynn.bendixsen Thanks very much.

firewater (Fri, 25 Jan 2019 21:39:01 GMT):
is a cloud agent endpoint url must be based on the an ip address or it could be a domain? and if i change my cloud agent endpoint url how would other agent who have establish connection with me before aware of the changes and make necessary changes to the endpoint url?

firewater (Fri, 25 Jan 2019 21:39:01 GMT):
is a cloud agent endpoint url must be based on the an ip address or it could be a domain name? and if i change my cloud agent endpoint url how would other agent who have establish connection with me before aware of the changes and make necessary changes to the endpoint url?

firewater (Fri, 25 Jan 2019 21:39:01 GMT):
is a cloud agent endpoint url must be based on the an ip address or it could be a domain name? and if i change my cloud agent endpoint url how would other agent who have establish connection with me before aware of the changes and make necessary changes to it?

firewater (Fri, 25 Jan 2019 21:40:48 GMT):
2 scenario here: 1) the did is register into ledger 2) the did is not register into ledger.

firewater (Fri, 25 Jan 2019 21:40:48 GMT):
2 scenario here: 1) the endpoint did is register into ledger 2) the endpoint did is not register into ledger.

firewater (Fri, 25 Jan 2019 21:40:48 GMT):
2 scenario here: 1) the public endpoint did is register into ledger 2) the public endpoint did is not register into ledger.

kdenhartog (Fri, 25 Jan 2019 23:51:46 GMT):
@firewater it could be a domain name, ip address, or another DID. Check out this HIPE for more info https://github.com/hyperledger/indy-hipe/tree/master/text/0023-diddoc-conventions

mossmanpete (Sat, 26 Jan 2019 00:44:00 GMT):
Has joined the channel.

mossmanpete (Sat, 26 Jan 2019 00:46:37 GMT):
hello. if keys are lost or stolen is there a "recovery" feature in indy?

mossmanpete (Sat, 26 Jan 2019 01:17:36 GMT):
i'm guessing something could be done with multi-sig to allow recovery with the help of a quorum of friends or a lawyer?

mossmanpete (Sat, 26 Jan 2019 01:18:10 GMT):
wonder if anyone has implemented this yet

minekaoru (Sat, 26 Jan 2019 10:28:01 GMT):
Has joined the channel.

mossmanpete (Sat, 26 Jan 2019 14:01:34 GMT):
i have heard about this feature with other crypto-accounts, i'm trying to find something to reference.

mossmanpete (Sat, 26 Jan 2019 14:16:05 GMT):
so, its based on weighted permissions, but this give the possibility to restore a hacked/lost account, and can be useful for any type of corporate identity... http://docs.bitshares.org/bitshares/user/account-permissions.html

mossmanpete (Sat, 26 Jan 2019 14:41:41 GMT):
think i found what i was looking for here: https://forum.sovrin.org/t/revocation-in-sovrin/468/10

vivekselvarajan (Sat, 26 Jan 2019 17:57:26 GMT):
Has joined the channel.

vivekselvarajan (Sat, 26 Jan 2019 17:57:32 GMT):
Hi

vivekselvarajan (Sat, 26 Jan 2019 17:57:36 GMT):
I'm new here

vivekselvarajan (Sat, 26 Jan 2019 17:57:47 GMT):
Can anyone help in setting up indy-sdk

vivekselvarajan (Sat, 26 Jan 2019 17:58:05 GMT):
seems to have many problem with setting up libindy

vivekselvarajan (Sat, 26 Jan 2019 17:58:25 GMT):
on ubuntu 18.04

vivekselvarajan (Sat, 26 Jan 2019 17:59:08 GMT):
Ran docker-compose for ubuntu-dockerfile and trying to start main,js in samples/nodes/src

jljordan_bcgov (Sat, 26 Jan 2019 19:30:57 GMT):
https://hub.docker.com/r/bcgovimages/von-image/

jljordan_bcgov (Sat, 26 Jan 2019 19:30:57 GMT):
On this page you will find a bunch of Indy resource alongside our Indy application components https://vonx.io/clickythings/#docker-hub-images https://hub.docker.com/r/bcgovimages/von-image/

jljordan_bcgov (Sat, 26 Jan 2019 19:30:57 GMT):
On this page you will find a bunch of Indy resource alongside our Indy application components https://vonx.io/clickythings/#docker-hub-images https://hub.docker.com/r/bcgovimages/von-image/ ... this image stays quite current with indy-sdk and other indy components

vijayanandsudhakar (Sun, 27 Jan 2019 16:13:43 GMT):
Has joined the channel.

mwklein (Mon, 28 Jan 2019 02:01:43 GMT):
Has joined the channel.

alexmayrhofer (Mon, 28 Jan 2019 07:41:26 GMT):
@swcurran - do i understand correctly that any revocation of Credentials *has* to go to the ledger? That means that the Ledger will probably contain mostly revocations after some time?

infominer33 (Mon, 28 Jan 2019 10:29:40 GMT):
https://ontvon-von-prod.pathfinder.gov.bc.ca/en/home

infominer33 (Mon, 28 Jan 2019 10:29:57 GMT):
the website says this is publicly accessible, but there is a password required

infominer33 (Mon, 28 Jan 2019 10:29:57 GMT):
the website says this is publicly accessible, but there is a password required Also, I can't replicate the DFlow demo from the VON presentation at Hyperledger Global..

infominer33 (Mon, 28 Jan 2019 10:32:28 GMT):
@mossmanpete https://eprint.iacr.org/2008/539.pdf - An Accumulator Based on Bilinear Maps and Efficient Revocation for Anonymous Credentials

HLFPOC (Mon, 28 Jan 2019 10:45:44 GMT):
Has joined the channel.

HLFPOC (Mon, 28 Jan 2019 10:47:18 GMT):
Hi Team, Can you please guide me if there is any notification/event service available in Indy so that any org in the network can be notified whenever there is a request from requester to issue the claims.

edisinovcic (Mon, 28 Jan 2019 13:30:55 GMT):
Has joined the channel.

bharadwajambati (Mon, 28 Jan 2019 14:34:52 GMT):
Has joined the channel.

swcurran (Mon, 28 Jan 2019 15:39:49 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=qpuLGw9n8ayYRisuJ) @alexmayrhofer It's not a one-to-one transaction on ledger per revocation. Basically, each time an Issuer wants to issue one or more revocations of credentials, they write a transaction that updates an accumulator and related entries. So a busy service, say a Driver's Licence, might be send a transaction a day. But yes, it's a lot of transactions to achieve the ZKP property.

swcurran (Mon, 28 Jan 2019 15:48:41 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=8SsW2i8Pw6dwAHGZm) @infominer33 That's an internal path to the site. Correct one is - https://www.von.gov.on.ca/ However, they have it locked down right now. The BC one is https://orgbook.gov.bc.ca and it's open.

swcurran (Mon, 28 Jan 2019 15:49:45 GMT):
You can also look at vonx.io and especially https://vonx.io/clickythings/ for links to all the VON related pages. dFlow is now at https://dev-dflow.orgbook.gov.bc.ca/

infominer33 (Mon, 28 Jan 2019 20:21:57 GMT):
thx. @swcurran @jljordan_bcgov you will appreciate this: https://github.com/infominer33/awesome-decentralized-id/blob/master/indy-sovrin-evernym/HyperledgerGlobalForum-VON-transcript.md

infominer33 (Mon, 28 Jan 2019 20:22:48 GMT):
why I was asking, is because I'm working on a user friendly version of that youtube vid, combining transcript and screencaps

swcurran (Mon, 28 Jan 2019 20:23:25 GMT):
That's awesome! Thanks! The Ontario one should be publically available Real Soon Now!

infominer33 (Mon, 28 Jan 2019 20:25:32 GMT):
nbd, thx for the update, I'll leave note of that on the page

infominer33 (Mon, 28 Jan 2019 20:25:32 GMT):
thx for the update, I'll leave note of that on the page

infominer33 (Mon, 28 Jan 2019 20:30:40 GMT):
and dflow is open, but empty. so I assume there is no way to replicate that workflow atm?

swcurran (Mon, 28 Jan 2019 20:48:25 GMT):
Should be - what link are you using?

infominer33 (Mon, 28 Jan 2019 22:39:40 GMT):
looks like this is the address used in the video. I found a working one last night, just wasn't able to bring up the diagrams like in the presentation http://devex-von-dflow-test.pathfinder.gov.bc.ca/

infominer33 (Mon, 28 Jan 2019 22:41:34 GMT):
so I just linked to the repo.. the repo doesn't have an address set up, yet

infominer33 (Mon, 28 Jan 2019 22:41:35 GMT):
https://github.com/bcgov/dFlow

infominer33 (Mon, 28 Jan 2019 22:42:44 GMT):
I just need to get more familiar with your website

swcurran (Mon, 28 Jan 2019 23:28:34 GMT):
We've changed things since the video. That's the issue - sorry about that. We're using more specific names now, and not the internal names such as that one. https://dflow.orgbook.gov.bc.ca for the dFlow link.

infominer33 (Tue, 29 Jan 2019 00:39:41 GMT):
after a while it will likely fork off to a more streamlined walk-through... but still want to make this as nice as possible, for posterity

jljordan_bcgov (Tue, 29 Jan 2019 05:08:06 GMT):
Awesome thanks! @infominer33 [ ](https://chat.hyperledger.org/channel/indy?msg=YNXaPvNg44GxTELZ3)

jljordan_bcgov (Tue, 29 Jan 2019 05:13:08 GMT):
Nice article stemming from a session at IIW describing the SSI “stack” ... from the DIF community. https://medium.com/decentralized-identity/the-self-sovereign-identity-stack-8a2cc95f2d45

infominer33 (Tue, 29 Jan 2019 06:42:19 GMT):
woah @swcurran just found this: https://github.com/hyperledger/education/blob/master/LFS171x/docs/introduction-to-hyperledger-indy.md

infominer33 (Tue, 29 Jan 2019 06:43:53 GMT):
oh is that all the edx material?

infominer33 (Tue, 29 Jan 2019 06:43:53 GMT):
oh is that all the edx material? or..

infominer33 (Tue, 29 Jan 2019 06:43:53 GMT):
oh is that all the edx material?

stefan.vogl (Tue, 29 Jan 2019 10:38:58 GMT):
Has joined the channel.

stefan.vogl (Tue, 29 Jan 2019 11:27:50 GMT):
sovrin

vtech (Tue, 29 Jan 2019 13:08:51 GMT):
Hi All, indy test network is configured to use steward seed as '000000000000000000000000Steward1'. How to register own steward rather using a pre-configured test seed data ?

TBigjohn (Tue, 29 Jan 2019 13:55:59 GMT):
Has joined the channel.

TBigjohn (Tue, 29 Jan 2019 14:07:58 GMT):
Hello, I am a total beginner with Indy. I am looking for a procedure to install it using docker containers if possible using yaml files. ( Like I did for fabric ) I am of course reading the doc but didn't found yet what I am looking for ... so if somebody has a direct link ... thanks

TBigjohn (Tue, 29 Jan 2019 14:30:42 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=RPbryTGZrPFJucnCi) @infominer33 Very good Link ! Thanks !

theoturner (Tue, 29 Jan 2019 15:22:36 GMT):
Hi all, I have a technical issue re: the python indy wrapper, is this the right place to ask?

theoturner (Tue, 29 Jan 2019 15:25:57 GMT):
Just saw #indy-sdk , not to worry :)

swcurran (Tue, 29 Jan 2019 15:51:20 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=RPbryTGZrPFJucnCi) @infominer33 That is indeed the edX material.

Hezaveh (Tue, 29 Jan 2019 18:35:54 GMT):
Has joined the channel.

Hezaveh (Tue, 29 Jan 2019 18:43:46 GMT):
Hi, could you please help me with this issue. I am following this: Inside samples/nodejs/ : npm install to install all NodeJS dependencies npm run ledger:start to start the ledger Docker container (needed for some samples). You must have docker installed. node main.js to run all scripts Or node to run a specific sample when I run: samples/nodejs/src$ node main.js I get an error: gettingStarted.js -> started Open Pool Ledger: pool1 (node:9337) UnhandledPromiseRejectionWarning: IndyError: CommonIOError at Object.callback (.../Downloads/indy-sdk/wrappers/nodejs/src/wrapIndyCallback.js:15:10) (node:9337) UnhandledPromiseRejectionWarning: Unhandled promise rejection. This error originated either by throwing inside of an async function without a catch block, or by rejecting a promise which was not handled with .catch(). (rejection id: 1) (node:9337) [DEP0018] DeprecationWarning: Unhandled promise rejections are deprecated. In the future, promise rejections that are not handled will terminate the Node.js process with a non-zero exit code.

lynn.bendixsen (Tue, 29 Jan 2019 21:15:18 GMT):
@vtech because the answer to your question was a little lengthy, I sent you a private message with my response on how to set up a new steward using indy-cli.

haniavis (Tue, 29 Jan 2019 22:29:24 GMT):
Hi @lynn.bendixsen I am also interested to this, could you send me your answer?

vtech (Wed, 30 Jan 2019 04:51:27 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=poTvChEtCMbktw9C9) @lynn.bendixsen Thanks

infominer33 (Wed, 30 Jan 2019 07:13:07 GMT):
https://www.axiomtech.io/blog-feed/history-of-permissioned-ledgers

infominer33 (Wed, 30 Jan 2019 07:13:07 GMT):
https://www.axiomtech.io/blog-feed/history-of-permissioned-ledgers :sweat_smile:

momoko8443 (Wed, 30 Jan 2019 07:15:16 GMT):
Has joined the channel.

momoko8443 (Wed, 30 Jan 2019 07:15:19 GMT):
Hello guys, does indy SDK provide an interface to set proxy address when connect pools?

Diiaablo95 (Wed, 30 Jan 2019 08:09:58 GMT):
Has joined the channel.

ShrutiHK (Wed, 30 Jan 2019 08:57:04 GMT):
Has joined the channel.

xnopre (Wed, 30 Jan 2019 10:09:08 GMT):
Hi all. With Indy, Alice and Bob can exchange crypted data with `cryptoAuthCrypt / cryptoAuthDecrypt`. I'm searching a solution so that Bob will store the crypted data, and decrypt this data when needed. But I want a solution for Alice to block this decryption. Is there a solution ? What is the good way ? Is the solution around key rotation by Alice ? .... Thanks for you help 🙂

Sandeep (Wed, 30 Jan 2019 17:55:24 GMT):
Has joined the channel.

momoko8443 (Thu, 31 Jan 2019 01:34:36 GMT):
@xnopre Don't send Alice's public key to bob.

momoko8443 (Thu, 31 Jan 2019 04:31:51 GMT):
Hi all, what is CommonInvalidState error? I try to open ledge by ssh tunnel (local), aways throw this exception.

haardikkk (Thu, 31 Jan 2019 05:26:27 GMT):
Has joined the channel.

cam-parra (Thu, 31 Jan 2019 12:36:15 GMT):
@momoko8443 Can you send you error logs ? That error type is in many areas of the code and would be hard to tell you what it exactly is going on

Sandeep (Thu, 31 Jan 2019 15:14:36 GMT):
@here. I am trying to use the alice,bob,faber demo. I was able to run the demo without any issues. I am trying to connect to the network from the host machine without using docker container to test out some code. I got the pool_transaction_genesis from one the node containers and the changed the IP addresses in the file to 127.0.0.1. When I try to connect to the network, I am getting the following error ``{ IndyError: CommonInvalidState at Object.callback (/Users/sandeeppulluru/work/indy/poc/node_modules/indy-sdk/src/wrapIndyCallback.js:15:10) name: 'IndyError', message: 'CommonInvalidState', indyCode: 112, indyName: 'CommonInvalidState' }``

Sandeep (Thu, 31 Jan 2019 15:14:36 GMT):
@here. I am trying to use the alice,bob,faber demo. I was able to run the demo without any issues. I am trying to connect to the network from the host machine without using docker container to test out some code. I got the pool_transaction_genesis from one the node containers and the changed the IP addresses in the file to 127.0.0.1. When I try to connect to the network, I am getting the following error \{ IndyError: CommonInvalidState at Object.callback (/Users/sandeeppulluru/work/indy/poc/node_modules/indy-sdk/src/wrapIndyCallback.js:15:10) name: 'IndyError', message: 'CommonInvalidState', indyCode: 112, indyName: 'CommonInvalidState' }\

Sandeep (Thu, 31 Jan 2019 15:14:36 GMT):
@here. I am trying to use the alice,bob,faber demo. I was able to run the demo without any issues. I am now trying to connect to the network from the host machine without using docker container to test out some code. I got the pool_transaction_genesis from one the node containers and the changed the IP addresses in the file to 127.0.0.1. When I try to connect to the network, I am getting the following error \{ IndyError: CommonInvalidState at Object.callback (/Users/sandeeppulluru/work/indy/poc/node_modules/indy-sdk/src/wrapIndyCallback.js:15:10) name: 'IndyError', message: 'CommonInvalidState', indyCode: 112, indyName: 'CommonInvalidState' }\

haggs (Thu, 31 Jan 2019 15:16:00 GMT):
@Sandeep I've seen this issue pop up a few times, could be a wallet handle left open or other call the demo doesn't account for. Do you have a hidden directory `.indy_client`

haggs (Thu, 31 Jan 2019 15:16:13 GMT):
If so, try deleting it and retry?

cam-parra (Thu, 31 Jan 2019 15:17:32 GMT):
@MattRaffel :arrow_up:

MattRaffel (Thu, 31 Jan 2019 15:21:25 GMT):
@haggs you can delete the .indy_client directory. it does delete all your pool and wallet information (obviously)

MattRaffel (Thu, 31 Jan 2019 15:21:25 GMT):
@haggs @Sandeep you can delete the .indy_client directory. it does delete all your pool and wallet information (obviously). To figure out if that is the problem, i'd need to see more logs

MattRaffel (Thu, 31 Jan 2019 15:22:27 GMT):
invalidState doesn't yield much info, unfortunately

cam-parra (Thu, 31 Jan 2019 15:23:20 GMT):
@haggs and @Sandeep Matt is one of the maintainer of the SDK and he will try to help you out where he can :)

haggs (Thu, 31 Jan 2019 15:23:47 GMT):
Hmm yes I should probably look deeper. Right now in the Indy-dev repo we just blast away directory but that probably shouldn't be the answer. See this: https://github.com/kdenhartog/indy-dev/blob/20eb31e03dbe7ff7d01c04241acf41b8c7ca03ce/scripts/cleanup.sh#L12

haggs (Thu, 31 Jan 2019 15:24:34 GMT):
not sure if it's `.indy_client/pool` or `.indy_client/wallet`. Thanks for your answers @MattRaffel and @cam-parra

haggs (Thu, 31 Jan 2019 15:24:34 GMT):
not sure if it's `.indy_client/pool` or `.indy_client/wallet`. Thanks for your answers @MattRaffel and @cam-parra !!

Sandeep (Thu, 31 Jan 2019 15:24:59 GMT):
@MattRaffel @cam-parra @haggs .Thanks for looking into this. I deleted the folder and restarted the network. Still seeing the same issue.

Sandeep (Thu, 31 Jan 2019 15:25:15 GMT):
Where can I get the logs for this ?

MattRaffel (Thu, 31 Jan 2019 15:25:16 GMT):
the tests usually cleanup fine, as long as everything is working. if something breaks, then its questionable as to what gets left behind

haggs (Thu, 31 Jan 2019 15:25:58 GMT):
Same `invalidState`?

MattRaffel (Thu, 31 Jan 2019 15:25:59 GMT):
look at the data you are passing in. can't tell what function is even called

cam-parra (Thu, 31 Jan 2019 15:26:02 GMT):
Is @kdenhartog repo updated to the latest version of master on the SDK?

haggs (Thu, 31 Jan 2019 15:27:14 GMT):
No, unfortunately it's stuck at 1.6.2 for dependency issues https://github.com/kdenhartog/indy-dev/blob/20eb31e03dbe7ff7d01c04241acf41b8c7ca03ce/indy-dev.dockerfile#L20

kdenhartog (Thu, 31 Jan 2019 15:28:39 GMT):
Yeah, I've version locked it at this point. If we've got a new version, I'd like to add that where the user can pass an argument which will launch different docker versions

Sandeep (Thu, 31 Jan 2019 15:29:07 GMT):

Pool3.txt

Sandeep (Thu, 31 Jan 2019 15:29:31 GMT):
@MattRaffel That is the data I am passing to `createPoolLedgerConfig`

cam-parra (Thu, 31 Jan 2019 15:30:03 GMT):
@kdenhartog is this docker only available on your fork? We should try to move as many people to the official fork :) If they want to of course

kdenhartog (Thu, 31 Jan 2019 15:31:47 GMT):
The docker image is based on the testing environment. However all the supporting files are different.

kdenhartog (Thu, 31 Jan 2019 15:32:51 GMT):
The problem is it's not at a point where I find it ready for the official repo.

kdenhartog (Thu, 31 Jan 2019 15:34:00 GMT):
It only supports python at this point and I'd like to add some of the Agents able to build in there. At that point it's making the sdk repo super heavy.

MattRaffel (Thu, 31 Jan 2019 15:37:15 GMT):
@Sandeep can you get us the data please

Sandeep (Thu, 31 Jan 2019 15:49:11 GMT):
@MattRaffel . What data do you need ?. Are you talking about the logs ?. If so, can you please let me know where I can find them.

Sandeep (Thu, 31 Jan 2019 16:03:18 GMT):
@haggs @cam-parra . Where can I find the logs to debug this issue ?

Sandeep (Thu, 31 Jan 2019 16:03:18 GMT):
@haggs @cam-parra @MattRaffel . Where can I find the logs to debug this issue ?

mreddy (Thu, 31 Jan 2019 16:15:55 GMT):
Has joined the channel.

nage (Thu, 31 Jan 2019 16:16:18 GMT):
#### INDY WG call going on here https://zoom.us/j/232861185 ####

nage (Thu, 31 Jan 2019 16:17:42 GMT):
Schema 2.0 presentation will start after WG updates W3C Verifiable Credentials WG
 https://www.w3.org/2017/vc/WG/
And for non W3C members, see the W3C Credentials CG
https://w3c-ccg.github.io/

nage (Thu, 31 Jan 2019 16:17:42 GMT):
Schema 2.0 presentation will start after WG updates W3C Verifiable Credentials WG
 https://www.w3.org/2017/vc/WG/
And for non W3C members, see the W3C Credentials CG
 https://w3c-ccg.github.io/

pknowles (Thu, 31 Jan 2019 16:25:58 GMT):
Note: In order to receive *Indy Semantics WG calendar invites*, make sure you've added your contact details to the following distribution list. https://docs.google.com/document/d/1NL36ZIksk4DmquRNvxpyZugWyjqCYa6n20FMzUnf-fY/edit?usp=sharing

pknowles (Thu, 31 Jan 2019 16:25:58 GMT):
Note: In order to receive *Indy Semantics WG calendar invites* ( #indy-semantics ), make sure you've added your contact details to the following distribution list. https://docs.google.com/document/d/1NL36ZIksk4DmquRNvxpyZugWyjqCYa6n20FMzUnf-fY/edit?usp=sharing

MattRaffel (Thu, 31 Jan 2019 16:26:31 GMT):
@Sandeep it depends on how you set it up. if its the demos, these are to the console. you will need to set environment variable RUST_LOG=error. there is some documentation on it: https://github.com/hyperledger/indy-sdk/blob/master/wrappers/nodejs/README.md

MattRaffel (Thu, 31 Jan 2019 16:26:31 GMT):
@Sandeep it depends on how you set it up. if its the demos, these are to the console. you will need to set environment variable RUST_LOG=trace. there is some documentation on it: https://github.com/hyperledger/indy-sdk/blob/master/wrappers/nodejs/README.md

Sandeep (Thu, 31 Jan 2019 16:34:53 GMT):
@MattRaffel. I am just using the network that was provided in the demo. I am trying to connect to that network from the host machine by creating a simple node js application. I created app.js file to open this connection. Do I have to do `RUST_LOG=trace node app.js`. If so, where can I find the logs ?. I am sorry to keep asking this as the documentation is a little hazy on debugging.

Alexi (Thu, 31 Jan 2019 16:39:15 GMT):
If I want to set up a test network is it better to use the indy-node docker set up or is the docker pool network described in the sdk repo pretty much the same thing?

Sandeep (Thu, 31 Jan 2019 16:50:03 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=HWzTRhnkctyCKa2dP) @MattRaffel @MattRaffel . I have enabled trace on the docker containers and I dont see anything in the logs when I try to open a connection. Since I am connecting to the docker container from the host, I changed the IP address in pool_transaction_genesis to 127.0.0.1 as well. I can see the logs being generated when I run the sample code. The problem there is nothing indicating an error.

Sandeep (Thu, 31 Jan 2019 16:50:40 GMT):

Indy logs

MattRaffel (Thu, 31 Jan 2019 16:51:43 GMT):
@Sandeep if you can redirect that to a file then find the call to create pool. we want to see what is being sent

Sandeep (Thu, 31 Jan 2019 16:52:42 GMT):
do you want the pool_transaction_genesis file ?

Sandeep (Thu, 31 Jan 2019 16:57:52 GMT):
@MattRaffel That is the problem. I dont see that call in the logs. Am I missing something connection wise ?. Do I have to anything different as I am trying to connect to the node containers from the host machine (not from another docker container) ?

MattRaffel (Thu, 31 Jan 2019 17:05:02 GMT):
the actual rust library call would be 'indy_create_pool_ledger_config'

Sandeep (Thu, 31 Jan 2019 17:12:21 GMT):
Yes. I can see the call in the logs. But that was done the web_server container that started with the network. I dont see the call being executed when I run it from the node js application.

Sandeep (Thu, 31 Jan 2019 17:12:21 GMT):
Yes. I can see the call in the logs. But that was done by the web_server container that started with the network. I dont see the call being executed when I run it from the node js application.

Sandeep (Thu, 31 Jan 2019 17:12:45 GMT):

Screen Shot 2019-01-31 at 12.12.17 PM.png

MattRaffel (Thu, 31 Jan 2019 17:13:53 GMT):
guessing the error then is something else

MattRaffel (Thu, 31 Jan 2019 17:13:53 GMT):
guessing the error then is something else. I know its tedious (we are working on that) keep looking through the logs until you see the CommonInvalidError

Sandeep (Thu, 31 Jan 2019 17:17:44 GMT):
I have checked the entire log. I dont see the `CommonInvalidState` error anywhere. I think the problem is connection related.

xaviervila (Thu, 31 Jan 2019 17:32:54 GMT):
Has joined the channel.

Sandeep (Thu, 31 Jan 2019 17:46:01 GMT):
@MattRaffel . Was able to get the logs working. I am able to see the errors. Can you please take a look and let me know what the issue is ?.

Sandeep (Thu, 31 Jan 2019 17:46:01 GMT):
@MattRaffel . Was able to get the logs working. I am able to see the error `CommonInvalidState`. Can you please take a look and let me know what the issue is ?.

Sandeep (Thu, 31 Jan 2019 17:46:30 GMT):

indy_log.txt

MattRaffel (Thu, 31 Jan 2019 17:58:10 GMT):
I have not seen this error 'Ledger merkle tree is not acceptable for current tree' before. @cam-parra could this be old node code running?

pknowles (Thu, 31 Jan 2019 19:04:48 GMT):
An updated version of the *Overlays Data Capture Architecture* deck has been uploaded to the following HL Indy shared area. Feel free to touch base on the #indy-semantics channel to discuss. https://drive.google.com/drive/u/0/folders/1-Q3CBSYXlRNEvTu7XQfGo-6W5H_yyOA3

pknowles (Thu, 31 Jan 2019 19:04:48 GMT):
An updated version of the *Overlays data capture architecture* deck has been uploaded to the following HL Indy shared area. Feel free to touch base on the #indy-semantics channel to discuss contents of the document. https://drive.google.com/drive/u/0/folders/1-Q3CBSYXlRNEvTu7XQfGo-6W5H_yyOA3

pknowles (Thu, 31 Jan 2019 19:04:48 GMT):
An updated version of the *Overlays data capture architecture* deck has been uploaded to the following HL Indy shared area. Feel free to touch base on the #indy-semantics channel to discuss document contents. https://drive.google.com/drive/u/0/folders/1-Q3CBSYXlRNEvTu7XQfGo-6W5H_yyOA3

pknowles (Thu, 31 Jan 2019 19:04:48 GMT):
An updated version of the *Overlays data capture architecture* deck has been uploaded to the following HL Indy shared area. Feel free to touch base on the #indy-semantics channel to discuss it's contents. https://drive.google.com/drive/u/0/folders/1-Q3CBSYXlRNEvTu7XQfGo-6W5H_yyOA3

MattRaffel (Thu, 31 Jan 2019 19:30:18 GMT):
@Sandeep what genesis transaction file are you using?

Sandeep (Thu, 31 Jan 2019 19:34:24 GMT):

nodepool.txt

Sandeep (Thu, 31 Jan 2019 19:34:33 GMT):
@MattRaffel . Please find above

chris.matichuk (Fri, 01 Feb 2019 00:16:04 GMT):
Has joined the channel.

vo2vo (Fri, 01 Feb 2019 04:39:58 GMT):
Has joined the channel.

Artemkaaas (Fri, 01 Feb 2019 06:46:45 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=tz2xgphnqdb6oJ6ve) @Sandeep pool_transaction_genesis in docker container and your local machine must be absolutely the same. if you changed something in your local copy you will get another hash. it is the cause of the error you receives.

phillip.gibb (Fri, 01 Feb 2019 10:52:32 GMT):
Has joined the channel.

DavidP (Fri, 01 Feb 2019 14:52:53 GMT):
Has joined the channel.

DavidP (Fri, 01 Feb 2019 15:01:48 GMT):
Hi, I was researching about the technical difference between indy & sovrin. Really, sovrin is only a running instance of indy, isn't it? are there any other difference?

Sandeep (Fri, 01 Feb 2019 15:28:41 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=a6NkCNmAgYt7LFDdv) @Artemkaaas @Artemkaaas . Without changing the IP Address in the pool_transaction_genesis, how can I connect to the docker container network from the host machine ?

Sandeep (Fri, 01 Feb 2019 15:28:41 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=a6NkCNmAgYt7LFDdv) @Artemkaaas @Artemkaaas . Without changing the IP Address to 127.0.0.1 in the pool_transaction_genesis, how can I connect to the docker container network from the host machine ?

Sandeep (Fri, 01 Feb 2019 15:28:41 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=a6NkCNmAgYt7LFDdv) @Artemkaaas @Artemkaaas . Without changing the IP Address to 127.0.0.1 in the pool_transaction_genesis, how can we connect to the docker container network from the host machine ?

Sandeep (Fri, 01 Feb 2019 19:21:41 GMT):
@Artemkaaas @MattRaffel @cam-parra @haggs . I was able to finally resolve this by adding host names to the docker compose file and not changing the pool_transaction_genesis file. I am able to connect from my local machine to the nodes and create nym transactions. Thanks for all the help !!!.

Doug-K1 (Fri, 01 Feb 2019 20:31:05 GMT):
Has joined the channel.

coderintherye (Sat, 02 Feb 2019 01:12:24 GMT):
Is there anything in the spec about "Trust Hubs" or are they not an actual role, but rather just a concept?

coderintherye (Sat, 02 Feb 2019 01:12:53 GMT):
they are discussed here https://github.com/WebOfTrustInfo/rwot6-santabarbara/blob/master/topics-and-advance-readings/sovrin-web-of-trust-3-primary-roles.md but are not mentioned in the auth rules https://github.com/hyperledger/indy-node/blob/master/docs/auth_rules.md

jljordan_bcgov (Sat, 02 Feb 2019 01:40:40 GMT):
In the document the “Holder” trust hub is a concept that is derived from our work in British Columbia that we just announced a beta production release of last week. Https://orgbook.gov.bc.ca You can read about the thinking behind that service here at https://vonx.io as well as links to all the demos, api, repos. Finally that idea has evolved into a “credential registry” which we are turning into an Indy sub project ... early repo here https://github.com/bcgov/indy-catalyst @coderintherye [ ](https://chat.hyperledger.org/channel/indy?msg=aKoi8yGY2KZMK89xJ)

coderintherye (Sat, 02 Feb 2019 01:47:53 GMT):
@drummondreed The RWoT notes link to a sovrin page ( https://sovrin.discoursehosting.net/t/joining-the-sovrin-trust-framework-working-group/502?u=drummond ) which leads to a broken Heroku page https://sovrin-slack-signup.herokuapp.com/ maybe worth updating the links to whatever is the working link, assuming there is a Slack still?

coderintherye (Sat, 02 Feb 2019 01:48:57 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=yfwgd19WIJ3bKnvTot) @jljordan_bcgov Thanks, yes I've been reading quite a bit about VonX and OrgBook and am following Catalyst now. Thank you for connecting the dots though that Trust Hub == Credential Registry or at least that is the incarnation. It can be confusing keeping up with all the terminology

TechleadzTester (Sat, 02 Feb 2019 09:19:43 GMT):
Has joined the channel.

TechleadzTester (Sat, 02 Feb 2019 09:20:28 GMT):
Hi friends, I installed von-network, and build TheOrgBook as well as Permitify files. network is running successfully, and i can use complete Alice "Indy Demo World" example on browser as "localhost:8080/". But i'm confused about somthing as modified data regarding Alice or New Person Registration, where is this data is saving in local machine. And how i can perform this verification with production site like "https://orgbook.gov.bc.cs". can anyone please tell me some working procedure about these verification or how it will work on my Live project. Thanks in advance

TechleadzTester (Sat, 02 Feb 2019 09:38:21 GMT):
[2/2] Actually, i'm also confused about i installed von-network in ubuntu and working wiith Alice example is fine. But i'm not sure how i will verify organization on production? is there any Api request i will send to https://orgbook.gov.bc.cs. Thanks in advance

mwherman2000 (Sat, 02 Feb 2019 16:51:03 GMT):
FYHumor: What is the link for the Indy use case document? I'd like to submit this example... :-)

mwherman2000 (Sat, 02 Feb 2019 16:51:03 GMT):
FYHumor: What is the link for the Indy use case document? I'd like to submit this example... :-) Happy Saturday

mwherman2000 (Sat, 02 Feb 2019 16:51:03 GMT):
FYHumor: What is the link to the Indy use case document? I'd like to submit this example... :-) Happy Saturday

mwherman2000 (Sat, 02 Feb 2019 16:51:03 GMT):
FYHumor: What is the link to the Indy use case document? I'd like to submit the following example... :-) Happy Saturday

mwherman2000 (Sat, 02 Feb 2019 16:52:17 GMT):

Clipboard - February 2, 2019 9:51 AM

jljordan_bcgov (Sat, 02 Feb 2019 22:28:49 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=mwisMjy4NJzBG7zMf) @coderintherye Sovrin Slack has been replaced by a Rocket.Chat instance ... https://chat.sovrin.org ... athough there does not seem to be an obvious place to find that on sovrin.org <-- @drummondreed @nage

mcemkilinc (Sun, 03 Feb 2019 12:00:16 GMT):
Has joined the channel.

mondraymond (Mon, 04 Feb 2019 12:18:46 GMT):
Has joined the channel.

apspdfoknd (Mon, 04 Feb 2019 15:55:32 GMT):
Has joined the channel.

drummondreed (Mon, 04 Feb 2019 18:37:36 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=mwisMjy4NJzBG7zMf) Thanks for catching this. I've gone in and added a new post on that Sovrin Forum thread with the new info about how to join the Sovrin Governance Framework Working Group (and the Sovrin Rocketchat channels).

drummondreed (Mon, 04 Feb 2019 18:37:36 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=mwisMjy4NJzBG7zMf) Thanks for catching this. I've gone in and added a new post on that Sovrin Forum thread with the new info about how to join the Sovrin Governance Framework Working Group (and the Sovrin Rocketchat channels). Please do join us; calls are Mondays at 16:00 UTC. All info is on the Sovrin Governance Framework Working Group Meeting Page: https://docs.google.com/document/d/1aJskOztz8NP8tI-9eaKaaOypF0Fm__SCLKR-U8ptSII/edit?usp=sharing

drummondreed (Mon, 04 Feb 2019 18:40:40 GMT):
As for the update to sovrin.org, I agree it should be much easier to find there. I'll ping the sovrin.org webmaster.

harmitfans (Mon, 04 Feb 2019 18:43:10 GMT):
Has joined the channel.

jcarpenter (Mon, 04 Feb 2019 19:34:43 GMT):
Has joined the channel.

DMPROBTC (Tue, 05 Feb 2019 08:46:38 GMT):
Has joined the channel.

apspdfoknd (Tue, 05 Feb 2019 08:54:47 GMT):
Hi all, I'm trying to call `ledger.build_get_cred_def_request` function (from Python wrapper) with credential definition ID equal to `5gbiALqa61XAgcoamTXRRX:3:CL:5gbiALqa61XAgcoamTXRRX:2:Transcript:0.3:TAG5` (obtained from `issuer_create_and_store_credential_def` function). After that I get an error saying that this ID must be an integer number (https://github.com/hyperledger/indy-sdk/blob/f9db04a182acdc2968e9956c1ec21be29048272e/libindy/src/services/ledger/mod.rs#L220). What is the correct way of calling `build_get_cred_def_request` function or how do I get credential definition ID value properly?

kdenhartog (Tue, 05 Feb 2019 12:04:47 GMT):
@apspdfoknd can you post a snippet of your code? I think the issue is that you're not submitting the cred_def or schema_def to the ledger before making this call.

apspdfoknd (Tue, 05 Feb 2019 12:50:21 GMT):
@kdenhartog Sure! source code: https://pastebin.com/MAqgQTtk and its output: https://pastebin.com/7AbsKvdA

kdenhartog (Tue, 05 Feb 2019 13:20:43 GMT):
I'm not certain this is what's causing the error for you, but I found a problem. On line 25 you provide the schema which doesn't have the seqNo in it. If you move line 43, 44, and 45 above line 22 and provide s in line 25 that should solve it.

kdenhartog (Tue, 05 Feb 2019 13:20:49 GMT):
@apspdfoknd

kdenhartog (Tue, 05 Feb 2019 13:23:43 GMT):
the schema returned on line 10 doesn't include the seqNo, so when the issuer calls `issuer_creates_and_stores_cred_def()` it doesn't have a reference point for the schema on the ledger and breaks later calls. It's one of the more tricky bugs to catch I spent a few days tracking that one down in the past.

SeanLarkin (Tue, 05 Feb 2019 13:28:36 GMT):
Has joined the channel.

apspdfoknd (Tue, 05 Feb 2019 14:25:22 GMT):
@kdenhartog Yay, it works! Thanks a lot! The problem seems to be with seqNo indeed

Alexi (Tue, 05 Feb 2019 17:24:17 GMT):
does someone have the link to the release channels workflow? the link in the sdk readme is broken

cam-parra (Tue, 05 Feb 2019 17:32:36 GMT):
@Alexi could you provide the broken link>

cam-parra (Tue, 05 Feb 2019 17:32:36 GMT):
@Alexi could you provide the broken link?

Alexi (Tue, 05 Feb 2019 17:38:07 GMT):
https://github.com/hyperledger/indy-sdk/blob/master/docs/release-workflow.md

Alexi (Tue, 05 Feb 2019 17:38:07 GMT):
@cam-parra https://github.com/hyperledger/indy-sdk/blob/master/docs/release-workflow.md

haggs (Tue, 05 Feb 2019 17:44:59 GMT):
@Alexi https://github.com/hyperledger/indy-sdk/blob/9d7eb13aa74ed7ee2e8ac1d7877ecb41a1cb64e3/doc/release-workflow.md

haggs (Tue, 05 Feb 2019 17:45:17 GMT):
Is that it?

haggs (Tue, 05 Feb 2019 17:47:51 GMT):
This is the most current version on master https://github.com/hyperledger/indy-sdk/blob/master/docs/contributors/release-workflow.md

pknowles (Tue, 05 Feb 2019 17:52:31 GMT):
This week's *Indy Semantics WG* call starts in 10 minutes. Zoom link: https://zoom.us/j/2157245727

cam-parra (Tue, 05 Feb 2019 17:53:02 GMT):
@Alexi sorry it looks like when we accepted the pull request for read the docs we didn't ask the contributor to update links on the README

cam-parra (Tue, 05 Feb 2019 17:53:19 GMT):
but yes you are right the correct link is https://github.com/hyperledger/indy-sdk/blob/master/docs/contributors/release-workflow.md

Alexi (Tue, 05 Feb 2019 17:58:26 GMT):
yes! thanks @haggs and @cam-parra

cam-parra (Tue, 05 Feb 2019 18:01:03 GMT):
@Alexi this PR will change that error https://github.com/hyperledger/indy-sdk/pull/1456/files

cam-parra (Tue, 05 Feb 2019 18:01:24 GMT):
@haggs this is your PR correct? Thanks!

haggs (Tue, 05 Feb 2019 18:01:52 GMT):
Yes @cam-parra Let me fix DCO

cam-parra (Tue, 05 Feb 2019 18:02:35 GMT):
Please update us on the PR link and we can prio this change since it's only documentation changes

cam-parra (Tue, 05 Feb 2019 18:02:50 GMT):
:grinning:

ShameemUSbank (Tue, 05 Feb 2019 18:04:49 GMT):
Has joined the channel.

haggs (Tue, 05 Feb 2019 18:12:00 GMT):
Will do, I know a lot of great work was done for helping the documentation so this is just link fixing.

haggs (Tue, 05 Feb 2019 18:20:58 GMT):
https://github.com/hyperledger/indy-sdk/pull/1457

Silona (Tue, 05 Feb 2019 20:45:30 GMT):
Hey y'alls update to the TSC is due https://wiki.hyperledger.org/display/HYP/TSC+Project+Updates

cam-parra (Tue, 05 Feb 2019 20:51:34 GMT):
@nage :arrow_up:

Sandeep (Tue, 05 Feb 2019 22:09:36 GMT):
@here. What is the difference between credential offer and a credential request ?

kdenhartog (Tue, 05 Feb 2019 22:14:50 GMT):
@Sandeep A credential offer represents the perspective of the issuer and what they're prepared to offer. The credential requests represents the perspective of the holder, and represents what they're prepared to receive. If there's a discreprency between the two, it means there's a negotiation between the holder and the issuer (not easily supported with the SDK currently) where as if it's the same, it means the two parties have agreed and they can now move forward with the actual issuance of the credential.

Sandeep (Tue, 05 Feb 2019 22:18:33 GMT):
Thanks @kdenhartog . So when a holder requests credentials from an issuer, an issuer would send a credential offer. Holder would then use this credential to create the appropriate credential request. Is that correct ?.

Sandeep (Tue, 05 Feb 2019 22:18:33 GMT):
Thanks @kdenhartog . So when a holder requests credentials from an issuer, an issuer would send a credential offer. Holder would then use this credential offer to create the appropriate credential request. Is that correct ?.

kdenhartog (Tue, 05 Feb 2019 22:19:12 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=75Kfrx6o84eSBPAg6) @Sandeep Yeah, that's correct

drummondreed (Wed, 06 Feb 2019 02:03:14 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=s7ciEbkzfRAhJpm2d) Nicely said, @kdenhartog ! You nailed it.

changeagent (Wed, 06 Feb 2019 14:33:06 GMT):
Hi all, my firm is seeking a senior blockchain dev, and we would love to have someone familiar with indy join since we are a sovrin steward. please check out the role, and if you are interested apply. DM me if you have any questions. the position is negotiable (location, etc.). https://www.bakerlaw.com/167497

Sean_Bohan (Wed, 06 Feb 2019 14:40:23 GMT):
TOMORROW, 2/7: Thursday Indy WG call: AGENDA: - Housekeeping - Updates: Agents, Semantics, SovCrypto, URSA - Truu - Doctors, Patients & Verifiable Credentials (MannyN) - Indy Roadmap (NathanG) - Open Discussion When: 8am PT, 9am MT, 10am CT, 11amET, 3pmGMT Where: Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/232861185

nanspro (Wed, 06 Feb 2019 15:25:46 GMT):
Has joined the channel.

Southside (Wed, 06 Feb 2019 17:26:15 GMT):
Has joined the channel.

Alexi (Wed, 06 Feb 2019 17:43:38 GMT):
Hello all, was wondering if someone knows if there is a reason why the python wrapper readme specifically asks for python 3.6? I am just curious as i ran the tests and all of them passed and i am using python 3.5, so am wondering if it is important i get 3.6 or can just continue developing on 3.5

ArnabChatterjee (Wed, 06 Feb 2019 18:28:57 GMT):
Has joined the channel.

ThiagoFontes (Wed, 06 Feb 2019 23:49:31 GMT):
Has joined the channel.

nage (Thu, 07 Feb 2019 16:02:51 GMT):
Link for the call's Roadmap discussion https://docs.google.com/presentation/d/1pxtvZE4Z-gm8eyIK3W29oetAdogOPBJlGbDBqH9kt2o/edit#slide=id.g4c6c106b55_0_7

bharadwajambati (Thu, 07 Feb 2019 16:32:56 GMT):
Can the issuer of the credential and verifier of the credential be two different parties ?. Lets say Company A issues credentials on behalf of Company B to Alice. Can Company C verify Alice's credentials by sending a request to Company B. Is such type of architecture possible in Indy ?.

nage (Thu, 07 Feb 2019 16:58:59 GMT):
The ledger allows for verification by company C without communicating with A or B. Any of the parties can provide the information from the ledger, so the verification interaction can happen online or offline.

nage (Thu, 07 Feb 2019 16:59:47 GMT):
Company C must have a way to know that the issue key (controlled by B represents company A--there are multiple ways to do that)

bharadwajambati (Thu, 07 Feb 2019 17:11:54 GMT):
Can you please elaborate on what do you mean by issue key and suggest one way of implementing this ?

haniavis (Thu, 07 Feb 2019 18:00:57 GMT):
So, I would like to ask if there are any known efforts to integrate Indy with Fabric in the same system, so to use Fabric as a general purpose ledger and at the same time call an Indy network for Identity management instead of using the native identity management mechanism of Fabric.

mwherman2000 (Thu, 07 Feb 2019 19:24:53 GMT):
INDY ARM: The current draft of the Hyperledger Indy/Sovrin Comprehensive Architecture Reference Model (INDY ARM) has been moved into github @ https://github.com/mwherman2000/indy-arm/blob/master/README.md. Please feel free to provide feedback and corrections using the Issues log.

mwherman2000 (Thu, 07 Feb 2019 19:24:53 GMT):
INDY ARM: The current draft of the Hyperledger Indy/Sovrin Comprehensive Architecture Reference Model (INDY ARM) has been moved into github @ https://github.com/mwherman2000/indy-arm/blob/master/README.md. Please feel free to provide feedback and corrections using the Issues log. #iDIDit

pknowles (Thu, 07 Feb 2019 19:42:04 GMT):
Top effort, @mwherman2000

mwherman2000 (Thu, 07 Feb 2019 19:50:13 GMT):
Thank you @pknowles on behalf of many contributors to this effort. The INDY ARM represents the sum total of several peoples' contributions of time and patience in explaining these concepts BB, AB, and SB (Before Basel, At Basel, and Since Basel). All the mistakes and wrong assumptions are mine but the INDY ARM is truly a community effort. In fact, the reason for moving it to github today, is that some of the Indy partners are asking their experts to look at it and we needed a better place for them to record their feedback. "Make it work. Make it work better. Make it work right."

mwherman2000 (Thu, 07 Feb 2019 19:52:56 GMT):
p.s. Ultimately, we have to give credit/blame to @darrell.odonnell for talking/dragging me into joining this great community. :-)

mwherman2000 (Thu, 07 Feb 2019 19:52:56 GMT):
p.s. Ultimately, we have to give credit/blame :-) to @darrell.odonnell for talking/dragging me into joining this great community. :-)

mwherman2000 (Thu, 07 Feb 2019 20:00:10 GMT):
The Inaugural DID Resolution Community Call is starting now... Thursday - Feb. 7, 2019 - 1pm MST (GST-7), 3pm EST, Noon PST https://zoom.us/j/7077077007 Related content: https://github.com/w3c-ccg/community/issues/20 https://github.com/w3c-ccg/did-resolution/

darrell.odonnell (Thu, 07 Feb 2019 20:10:50 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=GMyme5CEzmjrAySGN) @mwherman2000 oops

Alexi (Thu, 07 Feb 2019 20:46:02 GMT):
hello all, was wondering if anyone could answer a question: if i am working on a solution that is connected to a local pool of nodes, would it be best to keep the pool open at all times? or would it be better that whenever i make a call that requires the pool handle i open it, use the handle then close it again? I am asking as if i use the latter wouldn't there be a problem when two calls are made that require the pool to open since they both make an open_pool call and a pool with a specific name can only be opened once. In a sense i know i am answering my own question, but then I am wondering what the close_pool call is for then. I hope this question makes sense

Alexi (Thu, 07 Feb 2019 20:46:02 GMT):
hello all, was wondering if anyone could answer a question: if i am working on a solution that is connected to a local pool of nodes, would it be best to keep the pool open at all times? or would it be better that whenever i make a call that requires the pool handle; i open it, use the handle, then close it again? I am asking as if i use the latter wouldn't there be a problem when two calls are made that require the pool to open since they both make an open_pool call and a pool with a specific name can only be opened once. In a sense i know i am answering my own question, but then I am wondering what the close_pool call is for then. I hope this question makes sense

bharadwajambati (Thu, 07 Feb 2019 21:08:21 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=4t8yysCW3s5wzvd9UG) @nage Can you please elaborate on what do you mean by issue_key and suggest one way of implementing this?

nage (Thu, 07 Feb 2019 23:29:55 GMT):
Sorry, tapping this out from my phone. The credential definition contains an issuers public key and correctness proof for proving attributes in a credential. That is what I was referring to.

bbehlendorf (Fri, 08 Feb 2019 05:34:04 GMT):
Has joined the channel.

apspdfoknd (Fri, 08 Feb 2019 11:46:33 GMT):
Hi all, Is it possible to transform credentials (that are signed and stored in the wallet) to W3C verifiable claim model format (https://www.w3.org/TR/verifiable-claims-data-model/ example 3)?

apspdfoknd (Fri, 08 Feb 2019 11:46:33 GMT):
Hi all, Is it possible to transform credentials (that are signed and stored in the wallet) to W3C verifiable claim model format (https://www.w3.org/TR/verifiable-claims-data-model/)?

swcurran (Fri, 08 Feb 2019 18:31:47 GMT):
@apspdfoknd - transformation of credentials is extremely difficult because of the signatures/cryptography around the credential that make them verifiable. Transforming the data will likely invalidate the applied cryptography.

bsuichies (Fri, 08 Feb 2019 21:54:26 GMT):
Hello all - is there any good documentation on getting started with indy locally.. have the pool running, but am confused by the indy github documentation.. would like to setup a basic nodejs environment to play around, based on latest libindy

brentzundel (Fri, 08 Feb 2019 23:12:59 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=YTdupSv8axoHv3fKE) @apspdfoknd There are a number of us here who are participating in the definition of the W3C verifiable credential data model. We are also working out what it will take for Indy credentials to comply with the W3C verifiable credential data model. There should be a HIPE describing this in the next month or so.

bsuichies (Sat, 09 Feb 2019 07:24:26 GMT):
ccccccjfgdjgdrvtkcidingiinfbnkfktgnvlirbubbd

Sean_Bohan (Sat, 09 Feb 2019 21:29:58 GMT):
THIS THURSDAY, 2/14: Indy WG call: *The entire call will be a discussion about the Hyperledger Indy roadmap. * Please join us. When: 8am PT, 9am MT, 10am CT, 11amET, 3pmGMT Where: Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/232861185

haardikkk (Sun, 10 Feb 2019 05:11:16 GMT):
Has left the channel.

spregister (Mon, 11 Feb 2019 01:33:38 GMT):
Has joined the channel.

spregister (Mon, 11 Feb 2019 01:37:14 GMT):
Can Hyperledger-indy be used for enterprise use? Where so I start if I want to prototype?

cam-parra (Mon, 11 Feb 2019 15:12:16 GMT):
@spregister I believe so but @nage can confirm

colincmcc (Mon, 11 Feb 2019 16:14:35 GMT):
Has joined the channel.

kulkarnikk (Mon, 11 Feb 2019 16:19:10 GMT):
Has joined the channel.

nage (Mon, 11 Feb 2019 22:11:35 GMT):
Yes, it can. The best place to start is probably from the python reference agent. Take a look at https://github.com/hyperledger/indy-agent, https://github.com/sovrin-foundation/connector-app and https://vonx.io for some ideas on how to get going.

rekmarks (Mon, 11 Feb 2019 23:20:57 GMT):
Has joined the channel.

Alexi (Mon, 11 Feb 2019 23:23:53 GMT):
hello all, I had a question about when creating a credential: "Faber prepares both raw and encoded values for each attribute in the Transcript Credential Schema. Faber creates the Transcript Credential for Alice." ``` # note that encoding is not standardized by Indy except that 32-bit integers are encoded as themselves. IS-786 transcript_cred_values = json.dumps({ "first_name": {"raw": "Alice", "encoded": "1139481716457488690172217916278103335"}, "last_name": {"raw": "Garcia", "encoded": "5321642780241790123587902456789123452"}, "degree": {"raw": "Bachelor of Science, Marketing", "encoded": "12434523576212321"}, "status": {"raw": "graduated", "encoded": "2213454313412354"}, "ssn": {"raw": "123-45-6789", "encoded": "3124141231422543541"}, "year": {"raw": "2015", "encoded": "2015"}, "average": {"raw": "5", "encoded": "5"} }) ``` how are the values encoded in the examples? is there some sort of standard that has been used thus far?

vo2vo (Tue, 12 Feb 2019 01:36:11 GMT):
Hello. Anyone knows the relationship between Indy Crypto (https://github.com/hyperledger/indy-crypto) and Hyperledger Ursa? Trying to explain to some developers at my company about Hyperledger and Indy

dbluhm (Tue, 12 Feb 2019 03:26:03 GMT):
Indy Crypto will eventually be replaced by Ursa, which is a cross project shared crypto library

dbluhm (Tue, 12 Feb 2019 03:27:30 GMT):
@MALodder is heading up the Ursa project and can answer more questions

dbluhm (Tue, 12 Feb 2019 03:27:49 GMT):
@vo2vo ^^

MALodder (Tue, 12 Feb 2019 03:27:55 GMT):
indy-crypto is already in ursa

MALodder (Tue, 12 Feb 2019 03:28:52 GMT):
there is a [HIPE](https://github.com/hyperledger/indy-hipe/pull/87) to migrate to ursa

MALodder (Tue, 12 Feb 2019 03:29:09 GMT):
when that happens, indy-crypto will become archived

MALodder (Tue, 12 Feb 2019 03:29:32 GMT):
Ursa also includes other crypto primitives from other hyperledger projects like sawtooth and fabric

vsadriano (Tue, 12 Feb 2019 12:43:40 GMT):
Hi! Where can I get a specific documentation about `Sovrin Infrastructure Roles`?

lehors (Tue, 12 Feb 2019 13:03:05 GMT):
hello there

lehors (Tue, 12 Feb 2019 13:03:53 GMT):
I've seen highlevel presentations on Indy but I would like to know if there is any that describes the architecture and process flow

lehors (Tue, 12 Feb 2019 13:05:19 GMT):
something that would really show the different Indy components at play in the different scenarios: registration of some credential, verification, revocation

lehors (Tue, 12 Feb 2019 13:05:19 GMT):
something that would really show the different Indy components at play in different scenarios: registration of some credential, verification, revocation, ...

lehors (Tue, 12 Feb 2019 13:05:19 GMT):
something that would really show the different Indy components at play in different scenarios: issuance and registration of some credential, verification, revocation, ...

spregister (Tue, 12 Feb 2019 17:01:23 GMT):
spregpalspregister

spregister (Tue, 12 Feb 2019 17:03:47 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=RRjNZcMyfBtNAspfY) @nage Thank you @nage. I'll take a look at these links.

Alexi (Tue, 12 Feb 2019 17:04:08 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=TASfmoonYQwM38Jtc) was wondering if anyone had some insight on this? Want to know what most solutions have been using to encode the data when creating credentials, or at least how the data in the examples is being encoded

spregister (Tue, 12 Feb 2019 17:06:39 GMT):
I'm trying to go to the link (https://github.com/hyperledger/indy-sdk/blob/master/docs/ubuntu-build.md) mentioned on https://github.com/hyperledger/indy-agent. The link is broken and giving me 404. Anyone know the correct link?

dbluhm (Tue, 12 Feb 2019 17:48:23 GMT):
https://github.com/hyperledger/indy-sdk/blob/master/docs/build-guides/ubuntu-build.md

SilverSu (Wed, 13 Feb 2019 09:55:11 GMT):
Has joined the channel.

zoltan.lux (Wed, 13 Feb 2019 12:35:56 GMT):
Has joined the channel.

jenklogic (Wed, 13 Feb 2019 18:37:59 GMT):
Has joined the channel.

kpratihast (Thu, 14 Feb 2019 13:24:03 GMT):
Has joined the channel.

kpratihast (Thu, 14 Feb 2019 13:30:40 GMT):
Hi, I am persistently getting 114: 'CommonIOError' on calling even basic methods like creating new wallet. Any clue on how to resolve this ?

mwherman2000 (Thu, 14 Feb 2019 14:02:12 GMT):
Looking for feedback on the following Internet Naming Continuum chart (intended to help clarify the role of DIDs and the DID Resolution protocol relative to DNS, TCP-IP, etc.): https://github.com/mwherman2000/indy-arm/blob/master/README.md#appendix-c---internet-naming-continuum

dbluhm (Thu, 14 Feb 2019 14:23:09 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=X44XQx5Bq36J2zfxW) @kpratihast Sounds like you might be experiencing file permission issues

kpratihast (Thu, 14 Feb 2019 14:34:22 GMT):
@dbluhm Everything was working fine until a while ago, now suddenly none of the methods are co-operating anymore !

kpratihast (Thu, 14 Feb 2019 14:34:22 GMT):
@dbluhm Everything was working fine until a while ago, now suddenly some of the methods are co-operating anymore ! I am able to open and close existing wallets, however, I am getting new error on creating a new wallet. Is there any persistent file store in host machine where these wallets gets created and eventually referenced ?

kpratihast (Thu, 14 Feb 2019 14:34:22 GMT):
@dbluhm Everything was working fine until a while ago, now suddenly some of the methods aren't co-operating anymore ! I am able to open and close existing wallets, however, I am getting new error on creating a new wallet. Is there any persistent file store in host machine where these wallets gets created and eventually referenced ?

kpratihast (Thu, 14 Feb 2019 14:42:05 GMT):
Also, how can I retrieve Issuer's DID from a credential ?

xaviervila (Thu, 14 Feb 2019 15:20:51 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=fmqizFKvrgCkhBPeN) @Sean_Bohan For those interested near GMT, I guess that the call is scheduled to be at 4pmGMT

Sean_Bohan (Thu, 14 Feb 2019 15:22:42 GMT):
Thank you @xaviervila - that was a bad copy-paste

Sean_Bohan (Thu, 14 Feb 2019 15:22:52 GMT):
*4pm GMT*

esplinr (Thu, 14 Feb 2019 16:06:49 GMT):
Today's sprint demo from the Evernym Indy Team: https://www.youtube.com/watch?v=MWQRUHxEsPA&list=PLRp0viTDxBWGLdZk0aamtahB9cpJGV7ZF It includes a discussion of the recent releases of Indy Node and Indy SDK, some useful explanations about how the ledger works, a look at VCX message families, and a summary of work we are doing next.

esplinr (Thu, 14 Feb 2019 16:06:49 GMT):
Today's sprint demo from the Evernym Indy Team: https://www.youtube.com/watch?v=MWQRUHxEsPA&list=PLRp0viTDxBWGLdZk0aamtahB9cpJGV7ZF It includes a discussion of the recent releases of Indy Node and Indy SDK, some useful explanations about how the ledger works, a look at VCX message families, and a summary of work we are doing next.

esplinr (Thu, 14 Feb 2019 16:06:49 GMT):
Today's sprint demo from the Evernym Indy Team: https://www.youtube.com/watch?v=MWQRUHxEsPA&list=PLRp0viTDxBWGLdZk0aamtahB9cpJGV7ZF It includes a discussion of the recent releases of Indy Node and Indy SDK, some useful explanations about how the ledger works, a look at VCX message families, and a summary of work we are doing next.

ianco (Thu, 14 Feb 2019 19:33:26 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=tKCptrmr3EivybndY) @esplinr Thanks for posting these!

esplinr (Thu, 14 Feb 2019 19:44:40 GMT):
Thanks Ian for letting me know that they are valuable. It encourages us to keep it up.

nage (Thu, 14 Feb 2019 20:10:34 GMT):
The roadmap slides we used on today's Indy call https://docs.google.com/presentation/d/1pxtvZE4Z-gm8eyIK3W29oetAdogOPBJlGbDBqH9kt2o/edit?usp=sharing

Alexi (Thu, 14 Feb 2019 21:24:44 GMT):
hello all I was wondering if someone could tell me what the second value in a schema ID represents. in other words: issuer_did : ? : schema_name : schema_version for example: `4NiiH9wnpF9VKxw28fJj3h:2:test:0.1` what is the 2 ? it seems it always returns as two, is this the protocol version?

Alexi (Thu, 14 Feb 2019 22:17:31 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=EqtPfP6GcAGBjkfPq) so I found the source code for creating the id: ``` pub const SCHEMA_MARKER: &'static str = "2"; ... impl Schema { pub fn schema_id(did: &str, name: &str, version: &str) -> String { format!("{}{}{}{}{}{}{}", did, DELIMITER, SCHEMA_MARKER, DELIMITER, name, DELIMITER, version) } } ``` I see now that it is in fact always a 2. I am curious, I am assuming that given different id's the 2 simply represents that this id is one of a schema correct?

Alexi (Thu, 14 Feb 2019 22:17:31 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=EqtPfP6GcAGBjkfPq) so I found the source code for creating the schema id: ``` pub const SCHEMA_MARKER: &'static str = "2"; ... impl Schema { pub fn schema_id(did: &str, name: &str, version: &str) -> String { format!("{}{}{}{}{}{}{}", did, DELIMITER, SCHEMA_MARKER, DELIMITER, name, DELIMITER, version) } } ``` I see now that it is in fact always a 2. I am curious, I am assuming that given different id's the 2 simply represents that this id is one of a schema correct?

Alexi (Thu, 14 Feb 2019 22:43:51 GMT):
is it common practice to use the same credential definition for multiple credential offers? or is it best practice to make a new definition for each offer

dbluhm (Thu, 14 Feb 2019 22:55:07 GMT):
The same definition is typically used. If it's the same credential, using a different credential definition for each one is redundant. Someone else may be able to offer a more thorough explanation.

dbluhm (Thu, 14 Feb 2019 22:55:30 GMT):
@Alexi ^^

Alexi (Thu, 14 Feb 2019 23:01:39 GMT):
@dbluhm thanks! so even if an issuer is sending multiple credentials to different people, it would still be okay to use the same credential definition yes?

infominer33 (Fri, 15 Feb 2019 04:09:47 GMT):
I see hyperledger wiki got a face-lift. some broken links in the wake: https://wiki.hyperledger.org/projects/indy is now https://wiki.hyperledger.org/display/indy a re-direct would be nice. also, indy.readthedocs.io currently presents "this page does not exist yet"

infominer33 (Fri, 15 Feb 2019 04:09:47 GMT):
I see hyperledger wiki got a face-lift. some broken links in the wake: https://wiki.hyperledger.org/projects/indy is now https://wiki.hyperledger.org/display/indy a re-direct would be nice, or maybe some step in the process got misplaced... also, indy.readthedocs.io currently presents "this page does not exist yet"

infominer33 (Fri, 15 Feb 2019 04:10:20 GMT):
just thinking outloud

infominer33 (Fri, 15 Feb 2019 04:49:10 GMT):
in other news, I've been totally nerding out today, and whipped up a navigable version of Daniel Hardman's presentation on microledgers and edge-chains at HGF: https://github.com/infominer33/awesome-decentralized-id/blob/master/indy-sovrin-evernym/Microledgers-Edgechains-Hardman-HGF.md

HLFPOC (Fri, 15 Feb 2019 12:18:25 GMT):
Hello Team, Can anyone share some reference/pointers through which we can represent/demonstrate all the credential dependencies to get a claim. Like we have a dependency of Faber college transcript before applying for a job. Just wanted to know if there is a tool available to show this dependency in a flow-chart/graph ?

HLFPOC (Fri, 15 Feb 2019 12:18:25 GMT):
Hello Team, Can anyone share some reference/pointers through which we can represent/demonstrate all the credential dependencies to get a claim. Like we have a dependency of Faber college transcript before applying for a job. Just wanted to know if there is a tool available to show this dependency in a flow-chart/graph ? I have heard of dFlow app by bcgov, but not sure if same can be used in this case also.

PatrikStas (Fri, 15 Feb 2019 12:36:47 GMT):
Hi everyone, I'd like to share a hobby project I am/ve'been working on https://indyscan.io/ Link to github in top right corner ^_^ I'd call it very beta, there's a lot more things to improve - the graphs, transaction listing, filtering, fulltext search, main page, design etc. It's also easy to run this against your own indy pool, you can find instructions GitHub. Contributions are welcome!

PatrikStas (Fri, 15 Feb 2019 14:05:53 GMT):
https://indyscan.io

mgbailey (Fri, 15 Feb 2019 15:04:35 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=RhbYQrMM7g8qvtEnX) @PatrikStas @PatrikStas very cool! I'll be playing with this today.

mgbailey (Fri, 15 Feb 2019 15:04:35 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=RhbYQrMM7g8qvtEnX) @PatrikStas very cool! I'll be playing with this today.

Alexi (Fri, 15 Feb 2019 16:31:47 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=RhbYQrMM7g8qvtEnX) @PatrikStas very cool!

Alexi (Fri, 15 Feb 2019 16:31:47 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=RhbYQrMM7g8qvtEnX) @PatrikStas thats awesome, great job!

Alexi (Fri, 15 Feb 2019 16:33:56 GMT):
when I build a credential definition request and then sign and send it to the ledger I get the following response from the ledger: ``` {"reqId":1550247944140473701,"reason":"client request invalid: InvalidClientRequest('validation error [ClientClaimDefSubmitOperation]: cannot be smaller than 1 (ref=0)',)","identifier":"SZQGnjsDCvwxZFuB6YtpG5","op":"REQNACK"} ```

Alexi (Fri, 15 Feb 2019 16:33:56 GMT):
when I build a credential definition request and then sign and send it to the ledger I get the following response from the ledger: ``` {"reqId":1550247944140473701,"reason":"client request invalid: InvalidClientRequest('validation error [ClientClaimDefSubmitOperation]: cannot be smaller than 1 (ref=0)',)","identifier":"SZQGnjsDCvwxZFuB6YtpG5","op":"REQNACK"} ``` can someone help me in what this means, and show to fix it?

Alexi (Fri, 15 Feb 2019 16:33:56 GMT):
when I build a credential definition request and then sign and send it to the ledger I get the following response from the ledger: ``` {"reqId":1550247944140473701,"reason":"client request invalid: InvalidClientRequest('validation error [ClientClaimDefSubmitOperation]: cannot be smaller than 1 (ref=0)',)","identifier":"SZQGnjsDCvwxZFuB6YtpG5","op":"REQNACK"} ``` can someone help me in what this means, and how to fix it?

PatrikStas (Fri, 15 Feb 2019 17:34:41 GMT):
@Alexi Kinda wild guess, what value have you assigned to schema you're creating? I think I've bumped into this issue once.I feel like the version of the sceha must be higher than 1.0.0? Not sure though

PatrikStas (Fri, 15 Feb 2019 17:34:41 GMT):
@Alexi Kinda Thank you ! To your issue, a wild guess, what value have you assigned to schema you're creating? I think I've bumped into this issue once.I feel like the version of the sceha must be higher than 1.0.0? Not sure though

PatrikStas (Fri, 15 Feb 2019 17:34:41 GMT):
@Alexi Thank you ! To your issue, kinda wild guess, what value have you assigned to schema you're creating? I think I've bumped into this issue once.I feel like the version of the sceha must be higher than 1.0.0? Not sure though

PatrikStas (Fri, 15 Feb 2019 17:34:41 GMT):
@Alexi Thank you! To your issue, kinda wild guess, what value have you assigned to schema you're creating? I think I've bumped into this issue once.I feel like the version of the sceha must be higher than 1.0.0? Not sure though

Alexi (Fri, 15 Feb 2019 17:49:02 GMT):
@PatrikStas appreciate it, I was using a version lower than 1.0 (0.1.x) however changing it produced the same error

swcurran (Fri, 15 Feb 2019 19:58:41 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=RhbYQrMM7g8qvtEnX) @PatrikStas Awesome work! That's very cool.

Alexi (Fri, 15 Feb 2019 20:31:20 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=wxqoE7DCGxsyyM3BN) if anyone has some insight into this it would be great, been working on it all day and still cannot figure it out

nanspro (Sat, 16 Feb 2019 05:43:32 GMT):
Hi everyone, I am new to hyperledger and i am interested in learning/contributing to indy

nanspro (Sat, 16 Feb 2019 05:43:32 GMT):
Hi everyone, I am new to hyperledger and i am interested in learning/contributing to indy

nanspro (Sat, 16 Feb 2019 05:44:42 GMT):
I was trying to setup the development environment by following this https://github.com/hyperledger/indy-node/blob/master/docs/source/setup-dev.md

nanspro (Sat, 16 Feb 2019 05:45:43 GMT):
All the scripts ran successfully but on running tests i am getting same error in plenum, node, common, etc

nanspro (Sat, 16 Feb 2019 05:46:03 GMT):

Screenshot from 2019-02-16 11-08-47.png

pknowles (Sat, 16 Feb 2019 09:46:04 GMT):
I presume that someone on this channel will be aware of Shyft - https://www.shyft.network - My question is, are they working with HL Indy for the identity pieces?

pknowles (Sat, 16 Feb 2019 09:46:04 GMT):
I presume that someone on this channel will be aware of Shyft - https://www.shyft.network - My question is, are they working with HL Indy for any identity cogs in their architecture?

pknowles (Sat, 16 Feb 2019 09:46:04 GMT):
I presume that someone on this channel will be aware of Shyft - https://www.shyft.network - My question is, are they working with HL Indy for any identity cogs within their architecture?

pknowles (Sat, 16 Feb 2019 09:46:04 GMT):
I presume that someone on this channel will be aware of Shyft - https://www.shyft.network - My question is, are they working with HL Indy for the identity cogs within their architecture?

mwherman2000 (Sat, 16 Feb 2019 10:03:47 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=uqm3P8nvwY2FCnusJ) @pknowles Paul, I've had several conversations with the folks from Shyft (as well as SecureKey who are essentially in the same business): helping companies monetize their information assets with a particular focus on monetizing individuals personal identity data (with no value accruing/returning to the individual). I've written about both companies here: https://www.linkedin.com/pulse/canadian-banks-trafficking-digital-identities-canadians-herman/ I don't know anything about their technology platform but philosophically, Shyft is completely divergent (IMO) from the concepts behind SSI.

pknowles (Sat, 16 Feb 2019 10:04:51 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=jej82x7y5TMEzHjkH) @mwherman2000 Good to know. Thank you, Michael.

Alexi (Mon, 18 Feb 2019 17:09:10 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=wxqoE7DCGxsyyM3BN) hello, I apologize if I am being pesky. I still have not been able to find a solution to why i am getting this error. Does anyone have some insight into how to possibly fix this?

mgbailey (Mon, 18 Feb 2019 17:40:50 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=P2zwe5NYd7mHnYWCH) @Alexi For future reference, you may have better success with questions of this type in the indy-sdk channel. Can you give us a code snippet for better context? Are you running against a private test ledger, or the testnet?

Alexi (Mon, 18 Feb 2019 18:15:25 GMT):
@mgbailey thanks, just joined the channel. I am running it on a private test ledger (on a docker pool set up per the indy-sdk instructions). Here is the code leading i run testing the credential creation: ``` ... cred_tag = schema_name + '_credential' cred_def_config = { 'tag': cred_tag, 'type': 'CL', 'config': {"support_revocation": False} } print_log('\nCreating credential definition and storing it\n') cred_def_id, cred_def = \ await anoncreds.issuer_create_and_store_credential_def(wallet_handle, trust_anchor_did, created_schema, cred_def_config['tag'], cred_def_config['type'], json.dumps(cred_def_config['config'])) print_log('credential definition id: ', cred_def_id) print_log('credential definition: ', cred_def) cred_def_request = await ledger.build_cred_def_request(trust_anchor_did, cred_def) print_log('credential definition request: \n', cred_def_request) credential_response = await ledger.sign_and_submit_request(pool_handle, wallet_handle, trust_anchor_did, cred_def_request) print_log('credential definition response: \n', credential_response) ... ``` it is at this line that the response returns with the error: ``` credential_response = await ledger.sign_and_submit_request(pool_handle, wallet_handle, trust_anchor_did, cred_def_request) ``` appreciate the help

Alexi (Mon, 18 Feb 2019 18:15:25 GMT):
@mgbailey thanks, just joined the channel. I am running it on a private test ledger (on a docker pool set up per the indy-sdk instructions). Here is the code leading to the error that is testing the credential creation: ``` ... cred_tag = schema_name + '_credential' cred_def_config = { 'tag': cred_tag, 'type': 'CL', 'config': {"support_revocation": False} } print_log('\nCreating credential definition and storing it\n') cred_def_id, cred_def = \ await anoncreds.issuer_create_and_store_credential_def(wallet_handle, trust_anchor_did, created_schema, cred_def_config['tag'], cred_def_config['type'], json.dumps(cred_def_config['config'])) print_log('credential definition id: ', cred_def_id) print_log('credential definition: ', cred_def) cred_def_request = await ledger.build_cred_def_request(trust_anchor_did, cred_def) print_log('credential definition request: \n', cred_def_request) credential_response = await ledger.sign_and_submit_request(pool_handle, wallet_handle, trust_anchor_did, cred_def_request) print_log('credential definition response: \n', credential_response) ... ``` it is at this line that the response returns with the error: ``` credential_response = await ledger.sign_and_submit_request(pool_handle, wallet_handle, trust_anchor_did, cred_def_request) ``` appreciate the help

mgbailey (Mon, 18 Feb 2019 21:04:45 GMT):
@Alexi I see nothing obvious that is wrong with this code. Maybe the logged output would help?

kdenhartog (Tue, 19 Feb 2019 05:38:01 GMT):
@nanspro which version of python are you using? I use 3.5.2 for the most stable environment on the node.

kdenhartog (Tue, 19 Feb 2019 05:56:43 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=uqm3P8nvwY2FCnusJ) @pknowles From what I can deduce from their whitepaper, they're building on Ethereum and Bitcoin sidechains

Sreesha (Tue, 19 Feb 2019 12:08:09 GMT):
Has joined the channel.

Sreesha (Tue, 19 Feb 2019 12:34:28 GMT):
where can we get the msp for indy

ruggero.montalto-tno (Tue, 19 Feb 2019 13:04:51 GMT):
Has joined the channel.

ruggero.montalto-tno (Tue, 19 Feb 2019 13:05:12 GMT):
Hi! My name is Ruggero Montalto and I'm working on the implementation of a sovrin indy-node. Anyone here did the same?

ruggero.montalto-tno (Tue, 19 Feb 2019 14:18:50 GMT):
I use the stable Sovrin deb repo (` "deb https://repo.sovrin.org/deb xenial stable" `), but after running the command `sudo apt install -y sovrin` I obtain the following outcome: ``` Reading package lists... Done Building dependency tree Reading state information... Done Some packages could not be installed. This may mean that you have requested an impossible situation or if you are using the unstable distribution that some required packages have not yet been created or been moved out of Incoming. The following information may help to resolve the situation: The following packages have unmet dependencies: sovrin : Depends: indy-node (= 1.6.83) but it is not going to be installed Depends: sovtoken (= 0.9.9) but it is not going to be installed Depends: sovtokenfees (= 0.9.9) but it is not going to be installed Depends: libindy-crypto (= 0.4.5) but 0.5.1 is to be installed E: Unable to correct problems, you have held broken packages. ``` Any ideas why that might be?

ruggero.montalto-tno (Tue, 19 Feb 2019 14:18:50 GMT):
I use the stable Sovrin deb repo ( ` "deb https://repo.sovrin.org/deb xenial stable" ` ), but after running the command `sudo apt install -y sovrin` I obtain the following outcome: ``` Reading package lists... Done Building dependency tree Reading state information... Done Some packages could not be installed. This may mean that you have requested an impossible situation or if you are using the unstable distribution that some required packages have not yet been created or been moved out of Incoming. The following information may help to resolve the situation: The following packages have unmet dependencies: sovrin : Depends: indy-node (= 1.6.83) but it is not going to be installed Depends: sovtoken (= 0.9.9) but it is not going to be installed Depends: sovtokenfees (= 0.9.9) but it is not going to be installed Depends: libindy-crypto (= 0.4.5) but 0.5.1 is to be installed E: Unable to correct problems, you have held broken packages. ``` Any ideas why that might be?

ruggero.montalto-tno (Tue, 19 Feb 2019 14:49:36 GMT):
Found the solution: ``` ```

ruggero.montalto-tno (Tue, 19 Feb 2019 14:49:36 GMT):
Found the solution: ``` sudo apt-get install sovtokenfees=0.9.9 sovtoken=0.9.9 indy-node=1.6.83 indy-plenum=1.6.58 python3-indy-crypto=0.4.5 ```

Sandeep (Tue, 19 Feb 2019 16:40:43 GMT):
What is the difference between `cryptoAuthCrypt` and `cryptoAnonCrypt`

brentzundel (Tue, 19 Feb 2019 16:44:32 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=7iXvc33MotTfgsNDt) @Sandeep I believe the difference is as follows: `cryptoAuthCrypt` provides authenticated encryption (it is secret, and I can prove who encrypted it); `cryptoAnonCrypt provides encryption (it is secret, but I can't prove who encrypted it).

brentzundel (Tue, 19 Feb 2019 16:44:32 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=7iXvc33MotTfgsNDt) @Sandeep I believe the difference is as follows: `cryptoAuthCrypt` provides authenticated encryption (it is secret, and I can prove who encrypted it); `cryptoAnonCrypt` provides encryption (it is secret, but I can't prove who encrypted it).

nage (Tue, 19 Feb 2019 17:01:09 GMT):
The Sovrin Agent Connectathon is underway in Provo, UT. If you'd like to follow along, join us at https://chat.sovrin.org/channel/connectathon

nage (Tue, 19 Feb 2019 17:02:26 GMT):

IMG_4043.jpeg

pknowles (Tue, 19 Feb 2019 17:33:40 GMT):
This week's *Indy Semantics WG* call starts in 30 minutes. Zoom link: https://zoom.us/j/2157245727

noah4477 (Tue, 19 Feb 2019 20:32:48 GMT):
Has joined the channel.

noah4477 (Tue, 19 Feb 2019 20:36:11 GMT):
I'm trying to use cordentity and I've run into the issue that I can't build at all. It's because the indy-utils depends on indy and I can't seem to use maven to get indy. It tries using 'https://repo.evernym.com/artifactory/libindy-maven-local' but that doesn't connect to anything when it tries to get them pom or jar ('https://repo.evernym.com/artifactory/libindy-maven-local/org/hyperledger/indy/1.7.0/indy-1.7.0.pom') Does anyone have any idea of what I can do?

mgbailey (Tue, 19 Feb 2019 21:29:19 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=zHSJFhx7pshsbMaDS) @noah4477 Try again now. I see that Sovrin Foundation is trying to get it switched over to https://repo.sovrin.org/nexus, but they are not quite there yet.

ElSqueakador (Tue, 19 Feb 2019 22:12:24 GMT):
Has joined the channel.

GEEKTEDDY (Wed, 20 Feb 2019 02:10:46 GMT):
What’s the relationship between Indy and sovrin? sovrin is based on Indy or they are just paralleled

MALodder (Wed, 20 Feb 2019 03:16:44 GMT):
Sovrin is a global public utility based on Indy

MALodder (Wed, 20 Feb 2019 03:16:58 GMT):
Indy is just the ledger code that anyone can deploy as they see fit

MALodder (Wed, 20 Feb 2019 03:17:31 GMT):
Sovrin applies the Sovrin Trust Framework to Indy and has stewards that are volunteers that run the network

MALodder (Wed, 20 Feb 2019 03:19:17 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=7iXvc33MotTfgsNDt) @Sandeep Similar to what @brentzundel said. With `cryptoAuthcrypt`, the receiver knows who sent it, `cryptoAnonCrypt` the receiver does not know who sent it. With both, only the receiver can decrypt it

pknowles (Wed, 20 Feb 2019 05:19:58 GMT):
So... Luxoft's Cordentity integrates Indy with Corda. Can anyone on this channel point me to an identity integration program that integrates Indy with Ethereum? Thanks, channel!

DavidP (Wed, 20 Feb 2019 09:36:36 GMT):
I found the following description for cryptoAuthCrypt in the sdk: _ender can encrypt a confidential message specifically for Recipient, using Sender's public key. Using Recipient's public key, Sender can compute a shared secret key. Using Sender's public key and his secret key, Recipient can compute the exact same shared secret key. That shared secret key can be used to verify that the encrypted message was not tampered with, before eventually decrypting it._ My question is: How does it work? If you encrypt the message using sender's public key, only the sender could decrypt  the msg... I understand, that the sender add his public key into the message and generate a shared secret key from his private key and public key of the receipt. Is it right? Why you don’t sign the message with the sender private key and then encrypt with the receipt public key?

DavidP (Wed, 20 Feb 2019 09:36:48 GMT):
Thanks!

Sean_Bohan (Wed, 20 Feb 2019 14:26:15 GMT):
Due to the Sovrin Connect-a-Thon, and the number of Indy community members who are there, we will have the INDY WG call tomorrow and the agenda will be “Open Discussion” Where: https://zoom.us/j/232861185 When: 8am PT, 9am MT, 10am CT, 11amET, 4pmGMT

swcurran (Wed, 20 Feb 2019 15:29:20 GMT):
Hey folks - Indy Bootcamp Vancouver is on for March 11. Join us! For details, see: http://bit.ly/IndyBootcampVancouver

kdenhartog (Wed, 20 Feb 2019 16:33:17 GMT):
@DavidP Underneath, we're relying on libsodium's box and box_seal to handle this. When encrypting first thing we do is use the sender's private key and the receiver's public key to encrypt the message. We're then packaging the outputed ciphertext, the sender's public key, and the nonce into a structure called ComboBox. Then we're Anoncrypting (libsodium.box) this struct which requires only the receiver's public key because box is generating an ephemeral public key and appending the sender's public key to the ciphertext as the output of libsodium.box_seal(). Then for decrypting, we first call libsodium.box_seal_open() and the function knows how to parse the ephemeral key from the ciphertext and decrypt the message with the private key provided by the receiver in the API. This then reveals the ComboBox struct which then takes the information and calls libsodium.box_open() to decrypt the plaintext message provided in the authcrypt originally. This is finally returned with the sender's public key as well.

mboyd (Wed, 20 Feb 2019 16:39:36 GMT):
Hi all, I would like to publicly announce Hyperledger Indy's new documentation website: https://hyperledger-indy.readthedocs.io It is still a little rough around the edges, so please give your feedback on github at https://github.com/hyperledger/indy-docs/ Each of the individual repository documentation libraries can be found in the respective repositories under the `docs/` folders. While the documentation is currently very contributor oriented, we invite everyone to contribute to more developer and user facing documentation to drive adoption. To understand how to contribute to documentation, please review https://hyperledger-indy.readthedocs.io/en/latest/docs.html and the HIPE: https://hyperledger-indy.readthedocs.io/projects/hipe/en/latest/text/0025-indy-docs-framework/README.html I will be posting a short video tutorial later on these pages for greater clarity. Thanks for all of your great work!

Sean_Bohan (Wed, 20 Feb 2019 16:52:42 GMT):
Great work Michael & team

sacchit (Wed, 20 Feb 2019 21:42:48 GMT):
Has joined the channel.

sacchit (Wed, 20 Feb 2019 21:44:35 GMT):
Hi all, a quick question about Hyperledger Indy. When a verifier verifies a credential from a prover(for eg. a transcript) how is the private information in a credential (eg. name, ssn, gpa etc) verified in a certificate using the `schema id` and the `cred_def_id`? I tried looking into `anoncreds.verifier_verify_proof` function, but if someone could briefly explain it would help immensely. This is in reference with the getting_started guide for Indy. Thank you!

runiner (Thu, 21 Feb 2019 00:04:09 GMT):
Has joined the channel.

shankqr (Thu, 21 Feb 2019 02:26:58 GMT):
Has joined the channel.

shankqr (Thu, 21 Feb 2019 02:29:01 GMT):
Hi all. I am very grateful to discover the indy project. I am building a SSI for my government. Would love to hear your ideas :)

kpratihast (Thu, 21 Feb 2019 10:29:15 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=RZtNhJj8BrzhuPXb9) Can anyone please help me out with this...

Silona (Thu, 21 Feb 2019 11:33:30 GMT):
@mboyd the link to the Hyperledger wiki on the readme is broken.

feliperdelara (Thu, 21 Feb 2019 13:43:24 GMT):
Has joined the channel.

mwherman2000 (Thu, 21 Feb 2019 15:47:27 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=C7hHxxCtdHiDdQy53Y) @GEEKTEDDY At the highest level, think of Hyperledger Indy as a Self-Sovereign Identity (SSI) software platform and Sovrin as an SSI governance framework. They work hand-in-hand ...the "what" and the "how".

mwherman2000 (Thu, 21 Feb 2019 15:47:27 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=C7hHxxCtdHiDdQy53Y) @GEEKTEDDY At the highest level, think of Hyperledger Indy as a Self-Sovereign Identity (SSI) software platform and Sovrin as an SSI governance framework. They work hand-in-hand ...the "what" and the "how". It's possible to use Indy with your own governance framework but it would be a huge effort to develop your own.

mwherman2000 (Thu, 21 Feb 2019 15:47:27 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=C7hHxxCtdHiDdQy53Y) @GEEKTEDDY At the highest level, think of Hyperledger Indy as a Self-Sovereign Identity (SSI) software platform and Sovrin as an SSI governance framework. They work hand-in-hand ...the "what" and the "how". It's possible to use Indy with your own governance framework but it would be a huge effort to develop your own.

mwherman2000 (Thu, 21 Feb 2019 15:47:27 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=C7hHxxCtdHiDdQy53Y) @GEEKTEDDY At the highest level, think of Hyperledger Indy as a Self-Sovereign Identity (SSI) software platform and Sovrin as an SSI governance framework. They work hand-in-hand ...the "what" and the "how". It's possible to use Indy with your own governance framework but it would be a huge effort to develop your own. The Sovrin Foundation has also deployed/host its own instance of Indy/Sovrin platform.

mwherman2000 (Thu, 21 Feb 2019 15:47:27 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=C7hHxxCtdHiDdQy53Y) @GEEKTEDDY At the highest level, think of Hyperledger Indy as a Self-Sovereign Identity (SSI) software platform and Sovrin as an SSI governance framework. They work hand-in-hand ...the "what" and the "how". It's possible to use Indy with your own governance framework but it would be a huge effort to develop your own. The Sovrin Foundation has also deployed and hosts its own instance of Indy/Sovrin platform.

mwherman2000 (Thu, 21 Feb 2019 15:57:55 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=aKtQ2SXCfXhgpyhAL) @kdenhartog , it looks likes @davidp was quoting from some SDK documentation and on first reading it looks like the first sentence might not be correct? Can you confirm this sentence is/isn't correct? @DavidP : Do you have the link to where you found the italicized sentences?

nage (Thu, 21 Feb 2019 16:11:38 GMT):
https://docs.google.com/document/d/1dfGj5yjpSHzRF6UVPv7Ld1PTFme3fsXEwpLs0_e6NGA/edit?usp=sharing

nage (Thu, 21 Feb 2019 16:12:13 GMT):
Join us on https://chat.sovrin.org/channel/connectathon

nage (Thu, 21 Feb 2019 16:16:28 GMT):
https://docs.google.com/document/d/1x9O2m-jr282srH1BRcOJQjTROes8T7lnosI2_IPB26E/edit

Alexi (Thu, 21 Feb 2019 16:22:24 GMT):
Hey all quick question for when establishing connections: "A nonce is a random arbitrary number that can only be used one time. " what is the best practice that has been used when creating a nonce?

MALodder (Thu, 21 Feb 2019 16:49:56 GMT):
use a cryptographically secure pseudorandom number generator to generate the number

MALodder (Thu, 21 Feb 2019 16:50:14 GMT):
it also depends on where you are going to use it

MALodder (Thu, 21 Feb 2019 16:50:47 GMT):
the important item is it doesn't repeat

MALodder (Thu, 21 Feb 2019 16:51:18 GMT):
and its big enough like 80 bits or more

MALodder (Thu, 21 Feb 2019 16:51:27 GMT):
I prefer 128 bits

esplinr (Thu, 21 Feb 2019 17:05:45 GMT):
Great resources for getting started with Indy: 1. The README of the Indy SDK: https://github.com/hyperledger/indy-sdk/blob/master/README.md 2. For a deep dive, look at this list: https://github.com/infominer33/awesome-decentralized-id/tree/master/indy-sovrin-evernym

Alexi (Thu, 21 Feb 2019 17:14:39 GMT):
@MALodder thanks!

tenfinney (Thu, 21 Feb 2019 18:55:16 GMT):
Has joined the channel.

IanHarris (Thu, 21 Feb 2019 19:49:28 GMT):
Has joined the channel.

Rouven (Thu, 21 Feb 2019 20:11:43 GMT):
@Sean_Bohan - where can I find the recording from calls?

Sean_Bohan (Thu, 21 Feb 2019 20:17:55 GMT):
@Rouven - I will find out. We are migrating to the new wiki and Nathan has the recent recordings

Rouven (Thu, 21 Feb 2019 21:08:43 GMT):
Cool, thanks.

emilyz5 (Thu, 21 Feb 2019 21:33:54 GMT):
Has joined the channel.

Sreesha (Fri, 22 Feb 2019 05:18:22 GMT):
@danielhardman I downloaded the hyperledger indy folder structure but couldn't find anything related to msp Could you please tell me where can i find msp in the folder ?

danielhardman (Fri, 22 Feb 2019 05:24:24 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=MSfRfeiwM8ThMejRh) @Sreesha I don't know what "msp" means, so I'm not sure how to answer your question.

Sreesha (Fri, 22 Feb 2019 09:21:33 GMT):
@danielhardman what is the use of Jupyter notebook in indy

kpratihast (Fri, 22 Feb 2019 11:01:20 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=u7ar5w2uj8ZCM6Ppg) @kdenhartog

kdenhartog (Fri, 22 Feb 2019 11:04:58 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=7Eknzj95jJE9j5Z5f) @mwherman2000 This is a good catch. That statement is wrong and could use some clarification. Where is it at? I'll submit a PR to update it.

kdenhartog (Fri, 22 Feb 2019 11:06:08 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=Nw4D2AwMmwdt8sQoH) @Sreesha it's intended to be a way to setup a sandbox to learn new concepts and tools that exist in the Indy ecosystem

kdenhartog (Fri, 22 Feb 2019 11:07:59 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=2KQtsvkjENoRzq9tH) @kpratihast Can you clarify what you mean by this? Are you trying to receive the DID from the issued credential as the Holder, or are you trying to receive the Issuer's DID from the proof provided by the Holder as the Verifier?

kpratihast (Fri, 22 Feb 2019 11:12:00 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=us4JxG5dzNzm5E2FR) @kdenhartog @kdenhartog I am looking for a way as a Holder to extract Issuer's DID from my stored credentials which will be used to generate proof for the Verifier.

kdenhartog (Fri, 22 Feb 2019 11:13:40 GMT):
Gotcha. It's getting late for me, so I'm going to have to take a look at this one tomorrow and try to get back once I've finished investigating. Thank you for repinging me on this too.

ruggero.montalto-tno (Fri, 22 Feb 2019 11:20:16 GMT):
(I'm cross-posting from #indy-sdk ) Hi all, how can I join the global TestNet as a steward? I've got the whole procedure done up till generating a verkey, a BLS key and a PoP key by running the `sudo -i -u indy init_indy_node ` command. Do I need someone to accreditate me, or is there a way I can do it myself?

lynn.bendixsen (Fri, 22 Feb 2019 17:23:25 GMT):
Hi @ruggero.montalto-tno, since your company is already a Steward on the Sovrin MainNet all that is required for you to participate on the TestNet as a Steward is to add your Steward DID to the TestNet (send the request to support@sovrin.org). In other words, you do not have to have a validator node on the TestNet or any other Sovrin network, to be able to use that network for the purposes they were built for. That said, we are pleased to accept any offer of validator nodes for any of our networks as we are expanding rapidly and plan to provide and help maintain several networks for many different purposes. I have seen your chats in the Sovrin Rocket Chat Steward channel, but wanted to respond with more info here for the benefit of you and anyone else who might have the same question.

nage (Fri, 22 Feb 2019 17:23:30 GMT):
@ruggero.montalto-tno you will want to reach out to @lbendixsen. Most of that work happens at https://chat.sovrin.org/channel/steward

nage (Fri, 22 Feb 2019 17:23:30 GMT):
@ruggero.montalto-tno you will want to reach out to @lbendixsen. Most of that work happens at https://chat.sovrin.org/channel/stewards

lynn.bendixsen (Fri, 22 Feb 2019 17:23:36 GMT):
:)

MALodder (Fri, 22 Feb 2019 17:39:47 GMT):
Anoncred paper link https://github.com/hyperledger/ursa/blob/master/libursa/docs/AnonCred.pdf

MALodder (Fri, 22 Feb 2019 18:03:23 GMT):

AnonCrypto20.pdf

MALodder (Fri, 22 Feb 2019 18:03:25 GMT):
For those interested interested in anon creds 2.0 here is a draft of the paper

DuckLover (Fri, 22 Feb 2019 19:48:22 GMT):
Can anyone maybe provide a paper or article that compares the performance of differenct consens mechanism?

esplinr (Fri, 22 Feb 2019 20:32:06 GMT):
@DuckLover Lots of good discussion is here: https://forum.sovrin.org/t/researching-sovrin-ledger-2-0/1008

JoshCook (Fri, 22 Feb 2019 20:52:56 GMT):
Has joined the channel.

JoshCook (Fri, 22 Feb 2019 20:54:25 GMT):
Question about alice and faber transcript demo. When alice requests and than recieves her transcript, where exactly is Faber pulling this information for Alice's transcript? thanks

swcurran (Fri, 22 Feb 2019 21:13:33 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=oWYGKzsTh35rYtrs8) @JoshCook In the demo, the grades are just hardcoded in a JSON structure in the code. In a real implementation, the transcripts would come from some sort of backend system.

mwherman2000 (Sat, 23 Feb 2019 01:48:54 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=oWYGKzsTh35rYtrs8) @JoshCook In the GSG Python script, they're hardcoded: https://github.com/mwherman2000/indy-dev/blob/master/python/getting_started-verbose.py#L276-L284

nanspro (Sat, 23 Feb 2019 02:11:23 GMT):
Has left the channel.

pknowles (Sun, 24 Feb 2019 10:33:46 GMT):
In order to receive *Indy Semantics WG calendar invites*, make sure you've added your contact details to the following distribution list. https://docs.google.com/document/d/1NL36ZIksk4DmquRNvxpyZugWyjqCYa6n20FMzUnf-fY/edit?usp=sharing

kpratihast (Sun, 24 Feb 2019 11:11:39 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=FSS3GgvACJs9QRp2M) @kdenhartog Hi Kyle, did you get chance to have a look at it ?

kdenhartog (Sun, 24 Feb 2019 12:11:14 GMT):
I was able to look into this, and there's some functionality (such as Holder generating proof based on cred_def which is linked to issuer DID) but I'm not certain I understand exactly what you're looking to achieve yet. Could you provide a more specific scenario?

kpratihast (Sun, 24 Feb 2019 12:57:20 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=BucAKFaipnGsq6nXu) @kdenhartog Let's say the the Verifier has published a proof request wherein the 'requested_predicates' condition states that one of the attributes let's say 'age' has to be greater than 18 and there's a restriction applied that it should match with the credential issued by a Issuer on his/her DID. The Holder will pull the credential from his wallet which satisfies this proof request. The output of 'proverGetCredentials' method doesn't reveal Issuer's DID. So my question is there any method that Holder can use to fetch the Issuer's DID which he/she can eventually use to generate the proof?

Doug-K1 (Sun, 24 Feb 2019 19:03:59 GMT):
Can anyone tell me what are the trade-offs that come with using Indy as opposed to some other blockchain such as Quorum? My immediate application is focused on self-soverign identities, but in the future I want to expand our offerings to include functionality that is more like Quorum and I am worried about being constrained from building general purpose smart contracts.

Doug-K1 (Sun, 24 Feb 2019 19:23:01 GMT):
If Indy is purpose built just for Identity and a poor fit for other applications I suppose I could run a Quorum or other chain along with Indy in a configuration similar to The Ductchess Project http://bit.ly/2tBrVM8 - I would be interested to hear from anyone that is using a multi-chain strategy and why they chose it.

richzhao (Mon, 25 Feb 2019 02:42:24 GMT):
@mboyd I saw you created a indy Chinese Documentation Translation Session on bootcamp HK. I know there are people and companies are very interested in indy in Technical Working Group China. TWGC can help translating.

jljordan_bcgov (Mon, 25 Feb 2019 03:17:29 GMT):
Below is an excerpt from a draft readme for work we are doing in the Indy world. You can find full readme here -> https://github.com/bcgov/indy-catalyst/blob/master/README.md Also our open community website is https://vonx.io Main point is that Indy offers capability well beyond the narrow problem of personal identity. “It is important to note that while these emerging standards and technologies are being designed to tackle the very difficult challenges of secure and privacy respecting digital identity for people, they are not limited to the narrow context of personal identity. This new model can be applied to a broader set of use cases beyond those involving personally identifiable information. The model offers a generalized capability enabling highly secure entity to entity communications and it is this generalized capability that has led to the creation of Hyperledger Indy Catalyst. Indy Catalyst components enable enterprises to issue, hold and verify data about entities.” @Doug-K1 [ ](https://chat.hyperledger.org/channel/indy?msg=qHNe2kAqPYC9Hb7T4)

VinodValsan (Mon, 25 Feb 2019 10:02:24 GMT):
Has joined the channel.

VinodValsan (Mon, 25 Feb 2019 10:02:53 GMT):
can someone give me a link to the tutorial on hyperledger indy

vsadriano (Mon, 25 Feb 2019 10:23:01 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=EZa3q8E6mHeh7p8yF) @VinodValsan The `getting-started` is [here](https://github.com/hyperledger/indy-sdk/tree/master/docs/getting-started) and I liked of [this](https://github.com/hyperledger/education/tree/master/LFS171x/indy-material/nodejs).

Karlos (Mon, 25 Feb 2019 14:58:42 GMT):
Has joined the channel.

DanielHruby (Mon, 25 Feb 2019 15:13:21 GMT):
Has joined the channel.

DanielHruby (Mon, 25 Feb 2019 15:13:48 GMT):
Has left the channel.

Karlos (Mon, 25 Feb 2019 15:51:28 GMT):
Hi! does anyone know if there is a full sample implementation of an Indy Agent in python?

DuckLover (Mon, 25 Feb 2019 15:56:20 GMT):
Can someone help me to clarify what parts of Sovrin/indy are done and what concepts are in work?

OscarRoman (Mon, 25 Feb 2019 19:33:22 GMT):
Has joined the channel.

JoshCook (Mon, 25 Feb 2019 21:56:43 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=opnRfYsxgn5ypzrAy) @mwherman2000 Awesome Thank you very much

momoko8443 (Tue, 26 Feb 2019 00:42:26 GMT):
Hi Guys, I followed README.md to build an android client with indy sdk, but failed. I put libindy.so and indysdk1.81.jar into libs folder and set jniLibs.srcDirs = ['libs'], build phase is ok, but an exception throws at running time "java.lang.NoClassDefFoundError: Failed resolution of: Lcom/sun/jna/Callback; at org.hyperledger.indy.sdk.pool.Pool.createPoolLedgerConfig(Pool.java:93)"

jonathanreynolds (Tue, 26 Feb 2019 01:24:40 GMT):
Is anyone familiar with an SSI project to assist with identifying doctors working in UK hospitals? I'm hoping to get in touch with the team behind it but am struggling to find them! thanks

Sean_Bohan (Tue, 26 Feb 2019 03:20:40 GMT):
@jonathanreynolds - that sounds like Truu in the UK. This is the link to Manny’s presentation at the Global Forum in Dec: https://youtu.be/toS8n8hukuE Here is their URL: https://truu.id/

Sean_Bohan (Tue, 26 Feb 2019 03:20:40 GMT):
@jonathanreynolds - that sounds like Truu in the UK. This is the link to Manny’s presentation at the Global Forum in Dec: https://youtu.be/toS8n8hukuE Here is their URL: https://youtu.be/toS8n8hukuE

MALodder (Tue, 26 Feb 2019 04:34:38 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=xHYtvo9idsLs795aR) @DuckLover Anonymous credentials with CL signatures, range proofs with greater than or equal and non revocation are done

MALodder (Tue, 26 Feb 2019 04:34:57 GMT):
@DuckLover is there a specific feature(s) you are looking to use?

pknowles (Tue, 26 Feb 2019 05:43:54 GMT):
@jonathanreynolds - If Truu is the company you're looking for (as per @Sean_Bohan 's message), let me know and I can put you in touch with Manny. We chat every week.

DuckLover (Tue, 26 Feb 2019 11:10:03 GMT):
@MALodder i thought about what can change if i build a Sovrin Application now. I read about Microledgers and that soon everyone can write in the ledger for an example

captainsano (Tue, 26 Feb 2019 14:14:10 GMT):
Has joined the channel.

captainsano (Tue, 26 Feb 2019 14:47:01 GMT):
Hi guys, am trying to build an Android app here. In the docs, the maven repo at https://repo.evernym.com/artifactory/libindy-maven-local returns 502, is this an intermittent failure or there is another repo located elsewhere?

jonathanreynolds (Tue, 26 Feb 2019 16:10:20 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=wS3htpBv7a67wwcDR0) @Sean_Bohan thanks Sean!

jonathanreynolds (Tue, 26 Feb 2019 16:10:43 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=qm6tYFdGRwwSTQRvs) @pknowles thanks Paul. will DM

pknowles (Tue, 26 Feb 2019 17:30:10 GMT):
This week's ad hoc *Indy Semantics WG* call starts in 30 minutes. Zoom link: https://zoom.us/j/2157245727

mwherman2000 (Tue, 26 Feb 2019 21:42:12 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=umMpbbo7fM3CwfqZ4) @pknowles Awesome discussion this week Paul ...even thought it was an ad-hoc agenda ...nothing like a live demo to stir the imagination. :-)

pknowles (Tue, 26 Feb 2019 22:27:51 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=m2q37PWBBEcMCkyXm) @mwherman2000 The Overlays baby is venturing out is peering out of the womb! Dr. Mitwicki in his overalls.

pknowles (Tue, 26 Feb 2019 22:27:51 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=m2q37PWBBEcMCkyXm) @mwherman2000 The Overlays baby is peering out of the womb! Dr. Mitwicki in his overalls.

pknowles (Tue, 26 Feb 2019 22:27:51 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=m2q37PWBBEcMCkyXm) @mwherman2000 The Overlays baby is peering out of the womb! Dr. Mitwicki ( @mtfk ) in his overalls.

pknowles (Tue, 26 Feb 2019 22:27:51 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=m2q37PWBBEcMCkyXm) @mwherman2000 The Overlays piece peering out of the womb! Dr. Mitwicki ( @mtfk ) in his overalls.

Sandeep (Tue, 26 Feb 2019 22:39:38 GMT):
Do you have to be on the indy network to issue credentials?. Can credentials be issued if you can get hold of credential definition id of an existing credential definition on the network ?. Also, does the same thing apply for verification as well ?. Can the verifier submit a proof request and verify the received credentials if he knows the credential defintion id and credential schema id without being on the network ?

swcurran (Tue, 26 Feb 2019 23:40:45 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=7N3wN2hsNmv8WN9T6) @Sandeep You cannot use an existing CredDef to issue Credentials - you have to create your own. A CredDef includes public keys that are used to sign the claims and the CredDef creator must have the private keys for those public key. On the verification side - yes that is how it works. From the proof, the verifier gets the CredDef(s) ID and then gets the CredDef itself from the ledger. From that, the verifier can check the signatures on the claim (no tampering) and gets the DID on the ledger of the Issuer. The verifier can then investigate as necessary the Issuer to decide if the credential they issue can be trusted. It make take some work to make that "do I trust the issuer" work.

swcurran (Tue, 26 Feb 2019 23:40:45 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=7N3wN2hsNmv8WN9T6) @Sandeep You cannot use an existing CredDef to issue Credentials - you have to create your own. A CredDef includes public keys that are used to sign the claims and the CredDef creator must have the private keys for those public key. On the verification side - yes that is how it works. From the proof, the verifier gets the CredDef(s) ID and then gets the CredDef itself from the ledger. From that, the verifier can check the signatures on the claim (no tampering) and gets the DID on the ledger of the Issuer. The verifier can then investigate as necessary the Issuer to decide if the credential they issue can be trusted. It make take some effort to make that "do I trust the issuer" work.

swcurran (Tue, 26 Feb 2019 23:40:45 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=7N3wN2hsNmv8WN9T6) @Sandeep You cannot use an existing CredDef to issue Credentials - you have to create your own. A CredDef includes public keys that are used to sign the claims and the CredDef creator must have the private keys for those public key. On the verification side - yes that is how it works. From the proof, the verifier gets the CredDef(s) ID and then gets the CredDef itself from the ledger. From that, the verifier can check the signatures on the claim (no tampering) and gets the DID on the ledger of the Issuer. The verifier can then investigate as necessary the Issuer to decide if the credential they issue can be trusted. It make take some effort to to answer the "do I trust the issuer" question.

Sandeep (Tue, 26 Feb 2019 23:55:07 GMT):
thanks @swcurran.

Sandeep (Wed, 27 Feb 2019 00:10:01 GMT):
@swcurran . Without being on the network, how can the verifier create a proof request with a specific credential defintion id ?. Is there a discovery feature for seeing the available credential definitons on the network ?. Lets say Company B wants to create a proof request with credential defintion associated with Company A. Is there a way for Company B to see all the credential defintions created by Company A ?

daidoji (Wed, 27 Feb 2019 00:10:22 GMT):
Has joined the channel.

swcurran (Wed, 27 Feb 2019 00:18:10 GMT):
You have the right way it should work, but there are other ways to do it that sidestep the currently tricky problem of the lack of discovery on the ledger. The verifier could ask for any cred def that has a given schema, but that gets back to the same question - how to know what schema are on the ledger? The verifier can "know" it from other sources - e.g. someone could post a list of useful schema/cred defs (e.g. a curated list of Driver's Licences). This is a useful model as one also has to know what is the issuer behind a cred def - which means connecting to the DID and finding out who the issuer is and what is their process before issuing a Credential. Do you trust the issuer and it's processes? Discovery can be implemented in the same way that https://indyscan.io/ works - load the interesting ledger transactions into a ledger and track them. This is a pretty deep topic. Easy at first, but gets trickier. Doable and will be solved, but there is work to do.

swcurran (Wed, 27 Feb 2019 00:18:10 GMT):
You have the right way it should work, but there are other ways to do it that sidestep the currently tricky problem of the lack of discovery on the ledger. The verifier could ask for any cred def that has a given schema, but that gets back to the same question - how to know what schema are on the ledger? The verifier can "know" it from other sources - e.g. someone could post a list of useful schema/cred defs (e.g. a curated list of Driver's Licences). This is a useful model as one also has to know what is the issuer behind a cred def - which means connecting to the DID and finding out who the issuer is and what is their process before issuing a Credential. Do you trust the issuer and it's processes? Discovery can be implemented in the same way that https://indyscan.io/ works - load the interesting ledger transactions into a database and track them. This is a pretty deep topic. Easy at first, but gets trickier. Doable and will be solved, but there is work to do.

Sandeep (Wed, 27 Feb 2019 00:22:53 GMT):
Thanks again @swcurran. That was very helpful!!.

momoko8443 (Wed, 27 Feb 2019 06:14:54 GMT):
Does anybody successfully build an Android client with indy SDK? I always meet java.lang.UnsatisfiedLinkError: dlopen failed: library "libgnustl_shared.so" not found error when I launch AVD to debug...Who can save my day!!!

tangocqui (Wed, 27 Feb 2019 07:07:24 GMT):
Has joined the channel.

yahyaghardallou (Wed, 27 Feb 2019 10:03:11 GMT):
Has joined the channel.

yahyaghardallou (Wed, 27 Feb 2019 10:07:09 GMT):
Hello, How to explore the indy blockchain ?

yahyaghardallou (Wed, 27 Feb 2019 10:08:28 GMT):
ledger

MattRaffel (Wed, 27 Feb 2019 14:51:31 GMT):
@yahyaghardallou I suggest starting with this file: https://github.com/hyperledger/indy-sdk/blob/master/README.md it contains links to other documents including getting started etc...

cam-parra (Wed, 27 Feb 2019 15:33:09 GMT):
@yahyaghardallou also indy is a set of tools for implementing an indy ledger. You can see a working example at the Sovrin network or their test network

Sandeep (Wed, 27 Feb 2019 17:01:41 GMT):
@swcurran Can a holder of a credential initiate a credential sharing request with a verifier without the verifier submitting a proof request first ?

swcurran (Wed, 27 Feb 2019 17:06:08 GMT):
Yes - we have done that. They could generate the proof request themselves, and then just provide the proof request and proof to the verifier. We have considered this for the in-person proof scenario. The prover selects a standard proof request, generates a proof and sends it somehow (perhaps QR code) to the verifier - say a bartender. The verifier then can verify it.

Sandeep (Wed, 27 Feb 2019 17:13:04 GMT):
thanks @swcurran.

jljordan_bcgov (Wed, 27 Feb 2019 21:28:07 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=Mf3sNijdDMuRuGhR9) @yahyaghardallou Yes .. there are two ledger browsers ... https://indyscan.io/home/SOVRIN_MAINNET is a nice one you can browse Sovrin Main and Test networks We have one as well that we use for development -> http://prod.bcovrin.vonx.io We have an experimental version of our brower pointed at Sovrin Main Network -> https://sovrin-mainnet-browser.vonx.io

jljordan_bcgov (Wed, 27 Feb 2019 21:28:07 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=Mf3sNijdDMuRuGhR9) @yahyaghardallou Yes .. there are two ledger browsers ... https://indyscan.io/home/SOVRIN_MAINNET is a nice one you can browse Sovrin Main and Test networks We have one as well that we use for development -> http://prod.bcovrin.vonx.io We have an experimental version of our browser pointed at Sovrin Main Network -> https://sovrin-mainnet-browser.vonx.io

Sean_Bohan (Wed, 27 Feb 2019 21:41:15 GMT):
Hey everyone TOMORROW 2/28/2019 - Indy WG CALL!!!! Agenda: Connect-A-Thon Recap and INDY ROADMAP DISCUSSION CONTINUED Where: https://zoom.us/j/232861185 When: 8am PT, 9am MT, 10am CT, 11amET, 4pmGMT

DuckLover (Thu, 28 Feb 2019 02:42:51 GMT):
Hello, i got a pretty basic SSI/Sovrin question. Not sure if i misunderstood the papers or I'm not up-to-date but a Sovrin User need a trust anchor to get his first DID to use Sovrin. How is this self-sovereign? I need another authority to enable me a SSI.

richzhao (Thu, 28 Feb 2019 11:11:10 GMT):
yes, you are right, you need existing steward authority you at first @DuckLover

DuckLover (Thu, 28 Feb 2019 14:11:05 GMT):
Does anyone have information/resources about the upcoming ideas, that with the release of the Sovrin Token every Sovrin user can write to the ledger? I read an comment about it in an live document but cant find it or it is deleted.

DerrickL (Thu, 28 Feb 2019 14:14:54 GMT):
Has joined the channel.

sudomann (Thu, 28 Feb 2019 15:51:40 GMT):
@DuckLover there is no token

sudomann (Thu, 28 Feb 2019 15:52:11 GMT):
Could someone explain how applications interact with the Sovrin network?

sudomann (Thu, 28 Feb 2019 15:56:01 GMT):
i see that one would install a packaged binary with `sudo apt install sovrin`, but beyond that, a quick scan of their website doesn't show me documentation on how to use it. It is my understanding that it is merely built on top of Hyperledger Indy, so I have not bothered with an in depth study of Indy. :thinking:

DuckLover (Thu, 28 Feb 2019 16:00:46 GMT):
There is a Sovrin Token coming as far as i know. At least it is planned in the official Sovrin papers.

mgbailey (Thu, 28 Feb 2019 16:04:57 GMT):
@sudomann Interactions will be done using indy-sdk. The installation would be done using "sudo apt install libindy" instead of sovrin. More info in the hyperledger indy-sdk github, and the indy-sdk channel here.

nage (Thu, 28 Feb 2019 16:18:31 GMT):
Connectathon agenda (for those that missed it) https://docs.google.com/document/d/1dfGj5yjpSHzRF6UVPv7Ld1PTFme3fsXEwpLs0_e6NGA/edit?usp=sharing

nage (Thu, 28 Feb 2019 16:18:46 GMT):
videos and other resources coming in https://chat.sovrin.org/channel/connectathon

noah4477 (Thu, 28 Feb 2019 16:48:28 GMT):
Does anyone know if the indyutils repo was migrated? I'm trying to download the dependency and it's not working. I heard something a while ago about moving to sovrin but I don't have the link to where it might be

noah4477 (Thu, 28 Feb 2019 16:48:28 GMT):
Does anyone know if the libindy repo was migrated? I'm trying to download the dependency and it's not working. I heard something a while ago about moving to sovrin but I don't have the link to where it might be

noah4477 (Thu, 28 Feb 2019 17:27:04 GMT):
Is indy being migrated? I can't seem to get it from https://repo.evernym.com/artifactory/libindy-maven-local

mgbailey (Thu, 28 Feb 2019 17:57:56 GMT):
@noah4477 "all artifacts from old maven2 repo have been copied to new one, available now via `https://repo.sovrin.org/repository/maven-releases` or `https://repo.sovrin.org/repository/maven-public` endpoints"

esplinr (Thu, 28 Feb 2019 19:25:06 GMT):
Sprint demo video from the Evernym Indy team: https://www.youtube.com/watch?v=tdVAqJr9bUM&list=PLRp0viTDxBWGLdZk0aamtahB9cpJGV7ZF&index=1

esplinr (Thu, 28 Feb 2019 19:25:55 GMT):
Topics discussed: VCX error handling, audit ledger, authorization rules moving to the config ledger, and persisting instance change messages.

sudomann (Thu, 28 Feb 2019 21:45:16 GMT):
@mgbailey I see, thank you

sudomann (Thu, 28 Feb 2019 21:46:23 GMT):
@DuckLover yes, but nothing announced

sudomann (Thu, 28 Feb 2019 21:47:38 GMT):
If you joined their telegram at [this link](https://t.me/sovrin_foundation), you would see that Phil is constantly warding off people asking token, ICO, sale, etc.

cam-parra (Thu, 28 Feb 2019 22:23:37 GMT):
But the token code is open source and you're welcome to take a look at it and get it integrated with the sdk :)

cam-parra (Thu, 28 Feb 2019 22:23:37 GMT):
But the token code is open source and you're welcome to take a look at it and get it integrated with the sdk :) chat.sovrin.org is where you can ask more questions about the code

DuckLover (Thu, 28 Feb 2019 23:15:02 GMT):
I am more curious about the influence of the token to the current system. What would it change?

cam-parra (Thu, 28 Feb 2019 23:17:27 GMT):
@DuckLover @nage can answer this on chat.sovrin.org

esplinr (Fri, 01 Mar 2019 00:13:44 GMT):
@DuckLover The Sovrin whitepaper might answer your question: https://sovrin.org/library/sovrin-protocol-and-token-white-paper/

cam-parra (Fri, 01 Mar 2019 14:35:19 GMT):
Let's keep sovrin token discussions out of this channel please :)

Rouven (Fri, 01 Mar 2019 21:20:54 GMT):
Hi, where can I best learn more about the link secrets? E.g. I would like to understand what is required to change my link secret?

MALodder (Fri, 01 Mar 2019 22:03:17 GMT):
@Rouven Right now there isn't a good methodfor changing a link secret.

MALodder (Fri, 01 Mar 2019 22:03:17 GMT):
@Rouven Right now there isn't a good method for changing a link secret.

Rouven (Fri, 01 Mar 2019 22:13:06 GMT):
Thanks @MALodder - are there some ideas in discussion about potential future solutions?

MALodder (Fri, 01 Mar 2019 22:14:19 GMT):
Lots. We have discussed it quite a few times in the Sovrin Crypto Meeting that is held every Tuesday at 8am MST at https://zoom.us/j/567114224

Rouven (Fri, 01 Mar 2019 22:19:42 GMT):
ok, great - would love to learn more. Worth watching a specific recording?

MALodder (Fri, 01 Mar 2019 22:22:55 GMT):
no, not yet

MALodder (Fri, 01 Mar 2019 22:23:08 GMT):
we haven't recorded any of the sessions, but it might be a good idea

DuckLover (Sat, 02 Mar 2019 22:37:14 GMT):
Is someone more into the ledger stuff behind indy? Got a few questions about the storage :thinking:

kdenhartog (Sun, 03 Mar 2019 20:31:58 GMT):
@DuckLover pop over into the indy-node side of things. That's where that team typically is.

DuckLover (Sun, 03 Mar 2019 21:15:50 GMT):
@kdenhartog thanks, not even knew that that channel exists :-)

movee2005 (Mon, 04 Mar 2019 00:33:25 GMT):
Has joined the channel.

movee2005 (Mon, 04 Mar 2019 00:35:08 GMT):
I think I need some help with Indy - I am following https://github.com/hyperledger/indy-node/blob/stable/getting-started.md. I have installed the Indy CLI. I notice that the CLI commands do not work exactly like the documentation...

movee2005 (Mon, 04 Mar 2019 00:35:29 GMT):
ALICE> sratus Unknown group or command "sratus" Type "help" to display the help ALICE> status Unknown group or command "status" Type "help" to display the help ALICE>

movee2005 (Mon, 04 Mar 2019 00:36:40 GMT):
ALICE> new wallet Alice Unknown group or command "new" Type "help" to display the help ALICE> ALICE> new wallet Alice Unknown group or command "new" Type "help" to display the help ALICE>

DavidP (Mon, 04 Mar 2019 11:15:07 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=7Eknzj95jJE9j5Z5f) @mwherman2000 https://github.com/hyperledger/indy-sdk/blob/master/wrappers/nodejs/README.md -> look for: cryptoAuthCrypt

MALodder (Mon, 04 Mar 2019 15:01:12 GMT):
@mwherman2000 I think it should read like this: ``` Send can encrypt a confidential message intended only for Recipient. Sender computes the Diffie Hellman shared secret from his own private key and Recipients public key. Recipient can compute the exact same shared secret key by computing Diffie Hellman with his private key and Senders public key. The shared secret key verifies the encrypted message has not been tampered, then decrypts the message. ```

lorenz.loesch (Mon, 04 Mar 2019 15:35:27 GMT):
Has joined the channel.

mwherman2000 (Mon, 04 Mar 2019 15:56:29 GMT):
@MALodder Can you submit this change? ...it sounds good to me.

MALodder (Mon, 04 Mar 2019 15:57:00 GMT):
sure, we should make it so it doesn't pertain to just nodejs

Sean_Bohan (Mon, 04 Mar 2019 15:57:18 GMT):
Setting up the agendas for the next couple of weeks for the INDY call - if someone has something to demo on the call, please let me know!

shenoy (Mon, 04 Mar 2019 16:00:11 GMT):
Has joined the channel.

JanRzepecki (Tue, 05 Mar 2019 09:44:23 GMT):
Has joined the channel.

AbdullahAmin (Tue, 05 Mar 2019 12:59:48 GMT):
Has joined the channel.

AbdullahAmin (Tue, 05 Mar 2019 12:59:59 GMT):
Hello, I've created the binaries of libindy for android... I've successfully created the zip file for arm64-v8a as instructed in the url: https://indy.readthedocs.io/projects/sdk/en/latest/docs/build-guides/android-build.html#building-binaries-of-libnullpay-for-android I have also followed the steps to place the *.so files into jni lib... When I tried to build the apk on phone one-plus 5t(whose architecture is aarch64) the application is crashing... The error log is: java.lang.UnsatisfiedLinkError: dlopen failed: library "libgnustl_shared.so" not found While the supported arm64-v8a files are 1) libindy.so 2) libindy_shared.so

cam-parra (Tue, 05 Mar 2019 15:03:05 GMT):
@AbdullahAmin you don't need to build your own binaries unless you desire something different :) Here is the link to the artifact repo : https://repo.sovrin.org/android/

cam-parra (Tue, 05 Mar 2019 15:04:23 GMT):
@MattRaffel could also help you with some build of the build issues

pknowles (Tue, 05 Mar 2019 17:40:41 GMT):
This week's *Indy Semantics WG* call starts in 20 minutes. Zoom link: https://zoom.us/j/2157245727

dhuseby (Wed, 06 Mar 2019 02:26:23 GMT):

Nathan!

bharadwajambati (Wed, 06 Mar 2019 05:19:11 GMT):
Started using indy-cli based on the instructions provided in ``` https://github.com/hyperledger/indy-node/blob/master/getting-started.md ```

bharadwajambati (Wed, 06 Mar 2019 05:19:11 GMT):
Started using indy-cli based on the instructions provided in ``` https://github.com/hyperledger/indy-node/blob/master/getting-started.md ``` I launched the network and pool and prompted ALICE and tried to create a wallet for Alice using *new wallet Alice* command given in the docs. But getting an error. Can someone help me out with this?

ajmeraharsh (Wed, 06 Mar 2019 10:08:29 GMT):
Has left the channel.

infominer33 (Wed, 06 Mar 2019 10:56:39 GMT):
a little early, but... is anyone w BCGov around?

Diiaablo95 (Wed, 06 Mar 2019 11:46:04 GMT):
Hey all guys! I am trying to setup an Indy pool, where I only have a trustee, a steward and a validator node. When running the generate_indy_pool_transactions script, only the node keys are saved, this means that no more steward and (hence) nodes can be added to the pool. On the other side, I tried to generate the needed keys and start a pool with only one trustee, one steward and one validator, but by debugging the indy-cli client, I found that the validator node shows a public key of [0, 0, 0]..., and the assertion rc == 0 (src/curve_client.cpp:267) stops the execution.

Diiaablo95 (Wed, 06 Mar 2019 11:46:51 GMT):
Apparently, the generated test pool and domain transactions (with 1 node and 0 clients) and the my pool and domain transactions look the same in structure. Nevertheless, I cannot manage to connect the indy-cli client to the pool.

Diiaablo95 (Wed, 06 Mar 2019 11:46:59 GMT):
Do you guys have any suggestions concerning this?

Sean_Bohan (Wed, 06 Mar 2019 14:15:48 GMT):
Hey everyone TOMORROW 3/07/2019 - Indy WG CALL!!!! Community member Richard Esplin will be leading the call. Agenda for Thursday's Working Group call: Indy SDK! * Improving SDK governance * Managing SDK pull requests * Evolving the architecture * Testing the SDK * SDK CI / CD challenges * The future of wrappers * Bug bounties Where: https://zoom.us/j/232861185 When: 8am PT, 9am MT, 10am CT, 11amET, 4pmGMT

bharadwajambati1 (Wed, 06 Mar 2019 14:36:17 GMT):
Has joined the channel.

peteoleary (Wed, 06 Mar 2019 16:19:52 GMT):
Has joined the channel.

charliez (Thu, 07 Mar 2019 01:44:40 GMT):
Has joined the channel.

kyiopt (Thu, 07 Mar 2019 02:01:40 GMT):
Has joined the channel.

jdidelet (Thu, 07 Mar 2019 02:19:25 GMT):
Has joined the channel.

mattsauce2 (Thu, 07 Mar 2019 02:19:45 GMT):
Has joined the channel.

nage (Thu, 07 Mar 2019 02:19:46 GMT):
slides for the bootcamp https://docs.google.com/presentation/d/1KF-SWZ6M0T60lA3v5ahg2EIGL2L_XXblRhQgJ7Zlgxo/edit#slide=id.g52580e5378_10_72

nekia (Thu, 07 Mar 2019 02:19:49 GMT):
Has joined the channel.

iamsteveng (Thu, 07 Mar 2019 02:20:13 GMT):
Has joined the channel.

nage (Thu, 07 Mar 2019 02:20:35 GMT):
see my screen here https://zoom.us/j/8018881111

kclinux (Thu, 07 Mar 2019 02:21:17 GMT):
Has joined the channel.

davidpetit (Thu, 07 Mar 2019 02:21:20 GMT):
Has joined the channel.

itbill (Thu, 07 Mar 2019 02:30:21 GMT):
Has joined the channel.

mboyd (Thu, 07 Mar 2019 03:10:24 GMT):
Getting started with indy slides: https://docs.google.com/presentation/d/1mT_JKgvKNFhgt8d3oxuhclXnYx9bngQ95eApeeNbt-I/edit?usp=sharing

nitinvraj (Thu, 07 Mar 2019 03:20:16 GMT):
Has joined the channel.

MALodder (Thu, 07 Mar 2019 03:27:22 GMT):
This usually helps with for anyone who needs to compile libindy https://github.com/mikelodder7/indy-building

MALodder (Thu, 07 Mar 2019 03:27:37 GMT):
I’d like to get it merged into Indy-sdk

achn98 (Thu, 07 Mar 2019 03:28:24 GMT):
Has joined the channel.

mboyd (Thu, 07 Mar 2019 04:31:57 GMT):
https://uniresolver.io/

sprintwithcarlos (Thu, 07 Mar 2019 04:39:58 GMT):
Has joined the channel.

faisal00813 (Thu, 07 Mar 2019 05:30:07 GMT):
Some thing from Google https://www.xda-developers.com/google-android-digital-drivers-license/

mboyd (Thu, 07 Mar 2019 05:33:18 GMT):
Indy Chinese documentation translation: https://docs.google.com/presentation/d/178INe_FZiuKl8C4B9bf4apZb5WWCDK9iaIt6tWGTLzk/edit?usp=sharing

lidengjia (Thu, 07 Mar 2019 05:38:37 GMT):
Has joined the channel.

jackeyliliang (Thu, 07 Mar 2019 05:39:49 GMT):
Has joined the channel.

richzhao (Thu, 07 Mar 2019 05:50:45 GMT):
TWGC has completed a translation guideline at https://wiki.hyperledger.org/display/TWGC/Guideline . this is fabric, but it is suitable for indy too

mboyd (Thu, 07 Mar 2019 05:56:41 GMT):
Indy Walkthrough translation document: https://docs.google.com/document/d/1lMUxwX6j3E2qbN7PeLvPdNfm8pRe_i04bK_LxHDXkog/edit?usp=sharing

MayankArya19 (Thu, 07 Mar 2019 05:59:05 GMT):
Has joined the channel.

MayankArya19 (Thu, 07 Mar 2019 06:00:21 GMT):
hi, i am getting some problems while installing the setup

MayankArya19 (Thu, 07 Mar 2019 06:01:24 GMT):
docker exec -i -t indyclient bash this command returns no container indyclient found

MayankArya19 (Thu, 07 Mar 2019 06:01:33 GMT):
can someone tell me how to solve this issue

ltang007 (Thu, 07 Mar 2019 06:09:55 GMT):
Has joined the channel.

cjml1982 (Thu, 07 Mar 2019 06:20:32 GMT):
Has joined the channel.

Mindbloq (Thu, 07 Mar 2019 06:20:59 GMT):
Has joined the channel.

Mindbloq (Thu, 07 Mar 2019 06:21:14 GMT):
Hi For interoperability between consortium nodes, what data will be visible to other participating nodes and How do we protect our smart contracts. Is the code also shared if consortium is formed

mboyd (Thu, 07 Mar 2019 06:28:23 GMT):
Sovrin Video: - https://docs.google.com/document/d/1lMUxwX6j3E2qbN7PeLvPdNfm8pRe_i04bK_LxHDXkog/edit?usp=sharing Indy SDK README - https://docs.google.com/document/d/1lMUxwX6j3E2qbN7PeLvPdNfm8pRe_i04bK_LxHDXkog/edit?usp=sharing Indy Walkthrough - https://docs.google.com/document/d/1lMUxwX6j3E2qbN7PeLvPdNfm8pRe_i04bK_LxHDXkog/edit?usp=sharing Sovrin Glossary - https://docs.google.com/document/d/18XrPbST6wEVS_qIQvIgnBiFAx8DhttF1PA6R63OCmLY/edit (request for access)

mboyd (Thu, 07 Mar 2019 06:28:23 GMT):
Sovrin Video - Sovrin Video: - https://docs.google.com/document/d/1lMUxwX6j3E2qbN7PeLvPdNfm8pRe_i04bK_LxHDXkog/edit?usp=sharing Indy SDK README - https://docs.google.com/document/d/1lMUxwX6j3E2qbN7PeLvPdNfm8pRe_i04bK_LxHDXkog/edit?usp=sharing Indy Walkthrough - https://docs.google.com/document/d/1lMUxwX6j3E2qbN7PeLvPdNfm8pRe_i04bK_LxHDXkog/edit?usp=sharing Sovrin Glossary - https://docs.google.com/document/d/18XrPbST6wEVS_qIQvIgnBiFAx8DhttF1PA6R63OCmLY/edit (request for access)

nekia (Thu, 07 Mar 2019 06:32:12 GMT):
https://vimeo.com/305420834 I'm going to put a japanese subtitle for this video. Is there English script of this video already somewhere? @nage @mboyd

MayankArya19 (Thu, 07 Mar 2019 06:44:56 GMT):
is there any installation guide for windows?

jzhang681 (Thu, 07 Mar 2019 07:35:01 GMT):
Has joined the channel.

mboyd (Thu, 07 Mar 2019 07:36:12 GMT):
Indy agent: https://github.com/hyperledger/indy-agent

mboyd (Thu, 07 Mar 2019 07:36:12 GMT):
Streedcred agent-framework: https://github.com/streetcred-id/agent-framework

thinkpanda (Thu, 07 Mar 2019 07:42:26 GMT):
Has joined the channel.

iamsteveng (Fri, 08 Mar 2019 00:39:45 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=MC7PohWr4XGvXAYw5) @mboyd @mboyd The translation goes very well. Power of collaboration ;)

pawan_pandey (Fri, 08 Mar 2019 02:25:31 GMT):
Has joined the channel.

pawan_pandey (Fri, 08 Mar 2019 02:25:43 GMT):
@swcurran does Indy support DID Doc ?

swcurran (Fri, 08 Mar 2019 02:26:40 GMT):
@pawan_pandey - Indy supports the concept of DIDDoc, but it

swcurran (Fri, 08 Mar 2019 02:26:40 GMT):
@pawan_pandey - Indy supports the concept of DIDDoc, but it not as documented in the current spec. It is in the works. The DIDDoc spec is still evolving.

pawan_pandey (Fri, 08 Mar 2019 02:27:45 GMT):
@swcurran so how does resolution happen currently in Indy?

swcurran (Fri, 08 Mar 2019 02:28:01 GMT):
In the indy-agent world, support for off-ledger DIDDocs is being used and that will lead to full DIDDoc support on ledger.

swcurran (Fri, 08 Mar 2019 02:29:59 GMT):
As noted - the concept of resolution is the same. A DID is asked for of the ledger, and the ledger returns the attributes associated with that DID - the public key and endpoint. The Universal Resolver takes as input a DID in spec form, turns that into a query of the Indy ledger, and returns a formatted DIDDoc.

swcurran (Fri, 08 Mar 2019 02:31:04 GMT):
Here is an example of the universal resolver handling a DID on the Sovrin MainNet - https://uniresolver.io/#did=did:sov:HR6vs6GEZ8rHaVgjg2WodM

pawan_pandey (Fri, 08 Mar 2019 02:38:20 GMT):
@swcurran , can Indy handle a scenario where an issued credential has to be revalidated for example: e-Apostille

swcurran (Fri, 08 Mar 2019 02:38:52 GMT):
Don't know what that means?

swcurran (Fri, 08 Mar 2019 02:38:52 GMT):
Don't know what that means.

edmundto (Fri, 08 Mar 2019 02:39:05 GMT):
Has joined the channel.

pawan_pandey (Fri, 08 Mar 2019 02:40:43 GMT):
A use case where a degree issued by an educational institute of country A has to be notarized in order for it to be accepted by an educational institute in country B

swcurran (Fri, 08 Mar 2019 02:44:18 GMT):
Indy basically eliminates that scenario as the characteristics of the proof enables it to be used globally. If there was a desire to keep the ancient, paper-based process, there are several ways that could be done. For example, Holder proves the credential to a Notary, who issues an "It's Good!!!" credential to the Holder, who proves bother to the educational institution.

swcurran (Fri, 08 Mar 2019 02:44:18 GMT):
Indy basically eliminates that scenario as the characteristics of the proof enables it to be used globally. If there was a desire to keep the ancient, paper-based process, there are several ways that could be done. For example, Holder proves the credential to a Notary, who issues an "It's Good!!!" credential to the Holder, who proves both to the educational institution.

swcurran (Fri, 08 Mar 2019 02:44:44 GMT):
That process is just to mitigate a risk of forgery, and that's what Indy does.

pawan_pandey (Fri, 08 Mar 2019 02:47:48 GMT):
Great..Thanks

adriandotsoon (Fri, 08 Mar 2019 03:20:14 GMT):
Has joined the channel.

thinkpanda (Fri, 08 Mar 2019 05:50:49 GMT):
nage

kengeo (Fri, 08 Mar 2019 07:09:54 GMT):
Has joined the channel.

Rchauhan (Fri, 08 Mar 2019 12:03:59 GMT):
Has joined the channel.

Rchauhan (Fri, 08 Mar 2019 12:04:08 GMT):
Hello Friends We have seen your Alice example of Hyperledger Indy and appreciate your efforts. I am trying to understand the execution flow of the entire Alice example. but not able to understand. Kindly share something which can help to understand the execution flow of the project. I am checking the logs after ./manage build and ./manage up command but not getting fair idea to understand how the execution happening.

mwherman2000 (Fri, 08 Mar 2019 16:54:03 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=RFGyP6hcjbNrfb2Kc) @Rchauhan I'm madly trying to complete this: https://github.com/mwherman2000/indy-dev/blob/master/python/doc/getting_started-enterpise.md#hyperledger-indy-getting-started-guide-for-enterprise-architects-and-developers-indy-gsg-ea

PavanDurgadsimi (Fri, 08 Mar 2019 20:33:29 GMT):
Has joined the channel.

PavanDurgadsimi (Fri, 08 Mar 2019 20:34:45 GMT):
which is the best place to get started on the hyperledger indy project? Is it free or it is an enterprise related platform?

Sean_Bohan (Fri, 08 Mar 2019 21:24:07 GMT):
Hey @PavanDurgadsimi - Indy is open source software and one of the frameworks within Hyperledger. Anyone can use it or contribute to it and we welcome you to the community. Check out the wiki: https://wiki.hyperledger.org/plugins/servlet/mobile?contentId=2392563#content/view/2392563

Sean_Bohan (Fri, 08 Mar 2019 21:24:36 GMT):
And this: https://docs.google.com/presentation/d/1mT_JKgvKNFhgt8d3oxuhclXnYx9bngQ95eApeeNbt-I/mobilepresent?slide=id.p

mapa 5 (Sat, 09 Mar 2019 23:08:30 GMT):
Has joined the channel.

mapa 5 (Sat, 09 Mar 2019 23:08:35 GMT):
hello\

mapa 5 (Sat, 09 Mar 2019 23:08:37 GMT):
:D

mapa 5 (Sat, 09 Mar 2019 23:09:17 GMT):
im new year trying as a developer and ops guy

Rchauhan (Mon, 11 Mar 2019 05:39:19 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=z8Z24L4Ddtb9YRQJs) @mwherman2000 so have you completed this or working ?

davidpetit (Mon, 11 Mar 2019 07:06:42 GMT):
Has left the channel.

mwherman2000 (Mon, 11 Mar 2019 12:36:20 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=runZyogjX29mcwgSY) @Rchauhan Yes, all of the code works: 1. Use these instructions to setup your local environment: https://github.com/mwherman2000/indy-dev/blob/master/README.md#windows 2. Use the `getting_started-verbose.py` from here: https://github.com/mwherman2000/indy-dev/tree/master/python I'm still working on the Getting Started Guide (GSG-EA) ...hopefully completed by the end of the week. Progress was interrupted while I was working on the 2 roles and actors diagrams when I realized there wasn't a good resource for understanding the relationships between and across Sovrin roles. That led to me completing the first version of the Sovrin Governance Framework Comprehensive Architecture Reference Model (SOVRIN ARM): https://github.com/mwherman2000/sovrin-arm/blob/master/README.md#sovrin-arm---sovrin-governance-framework-comprehensive-architecture-reference-model-sovrin-arm

cam-parra (Mon, 11 Mar 2019 13:22:41 GMT):
@mapa 5 welcome! Feel free to ask any questions

cam-parra (Mon, 11 Mar 2019 13:22:41 GMT):
@mapa 5 welcome! Feel free to ask us any questions

DucaDellaForcoletta (Mon, 11 Mar 2019 13:29:01 GMT):
Has joined the channel.

DucaDellaForcoletta (Mon, 11 Mar 2019 13:30:38 GMT):
Hi everyone, I'm developing a POC on indy. In the issuer agent the error WalletAlredyOpen (206) is raised in case of two concurrent open_wallet. How do you manage concurrent open request? In the von-agent implementation example a lock is applied on each request but seems very strange in case of massive concurrent request. Thanks

cam-parra (Mon, 11 Mar 2019 13:46:26 GMT):
@mwherman2000 Are you set on the sovrin-arm name? It seems to me that the ARM TLA can be confusing to many developers. Since there are ARM architecture builds some developers may be mislead

lorenz.loesch (Mon, 11 Mar 2019 15:00:50 GMT):
hi all, we were trying to follow the getting started guide using the indy-sdk and got stuck here in step 2: https://github.com/hyperledger/indy-sdk/blob/master/docs/getting-started/indy-walkthrough.md#step-6-credential-definition-setup basically, our problem is that we don't know what the "primary key" is supposed to be. in the indy-sdk, a primary key in json format is a mandatory parameter. from the help function: `indy> ledger get-schema help` > `primary - Primary key in json format` we couldn't find any info on this so far. can anyone point us in the right direction? thanks!

lorenz.loesch (Mon, 11 Mar 2019 15:00:50 GMT):
hi all, we were trying to follow the getting started guide using the indy-cli and got stuck here in step 2: https://github.com/hyperledger/indy-sdk/blob/master/docs/getting-started/indy-walkthrough.md#step-6-credential-definition-setup basically, our problem is that we don't know what the "primary key" is supposed to be. in the indy-cli, a primary key in json format is a mandatory parameter. from the help function: `indy> ledger get-schema help` > `primary - Primary key in json format` we couldn't find any info on this so far. can anyone point us in the right direction? thanks!

lorenz.loesch (Mon, 11 Mar 2019 15:00:50 GMT):
hi all, we were trying to follow the getting started guide using the indy-cli and got stuck here in step 6.2: https://github.com/hyperledger/indy-sdk/blob/master/docs/getting-started/indy-walkthrough.md#step-6-credential-definition-setup basically, our problem is that we don't know what the "primary key" is supposed to be. in the indy-cli, a primary key in json format is a mandatory parameter. from the help function: `indy> ledger get-schema help` > `primary - Primary key in json format` we couldn't find any info on this so far. can anyone point us in the right direction? thanks!

mwherman2000 (Mon, 11 Mar 2019 16:14:02 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=uzrfCKwXdnHd7akja) @cam-parra I'm set on the concept of "Architecture Reference Models" ...the INDY ARM and SOVRIN ARM are two of the latest ARMs I've created: https://hyperonomy.com/?s=architecture+reference+model ...ARM falls out of that.

JasonCZM (Mon, 11 Mar 2019 21:39:09 GMT):
Has joined the channel.

jenklogic (Tue, 12 Mar 2019 00:17:54 GMT):
Hi all. I have been toying around with Indy for a few months now. I understand this is a whole new animal and docs/processes/whatever are ever-changing, but does anyone have a current, completely functional road map for installation and setup of the basic foundation of Indy? The docs currently out there are very difficult to follow and a lot of errors thrown (for me anyway) in a lot of different steps. Any help would be stupendously appreciated! :-)

msjeong (Tue, 12 Mar 2019 05:12:05 GMT):
Has joined the channel.

MALodder (Tue, 12 Mar 2019 13:57:52 GMT):
@jenklogic can you detail where you are facing difficulties? Is it working with Indy-SDK, understanding the entire architecture, or something else? I have written up my own compiling Indy-SDK docs [here](https://github.com/mikelodder7/indy-building) if that helps

lorenz.loesch (Tue, 12 Mar 2019 13:59:49 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=bPtxkhv7onyb3K5HL) any ideas concerning my question from yesterday?

ianco (Tue, 12 Mar 2019 14:27:55 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=58zzJ5vdicD6GPzdv) @lorenz.loesch Are you running an old version of indy-cli? Here is what I get from the cli "ledger get-schema help": indy> ledger get-schema help Command: ledger get-schema - Get Schema from Ledger. Usage: ledger get-schema did= name= version= Parameters are: did - DID of identity presented in Ledger name - Schema name version - Schema version Examples: ledger get-schema did=VsKV7grR1BUE29mG2Fm2kX name=gvt version=1.0

lorenz.loesch (Tue, 12 Mar 2019 14:43:22 GMT):
hi @ianco, thank you. I just noticed that i asked the wrong question. my question concerns the `ledger cred-def` command, which definitely has the mandatory `primary` parameter.

lorenz.loesch (Tue, 12 Mar 2019 14:47:29 GMT):
to clarify, the question is what the `primary` parameter is supposed to be and where we can get it. thank you for any input ;)

lorenz.loesch (Tue, 12 Mar 2019 14:48:31 GMT):
full output of the help call: (built from current master) ``` indy> ledger cred-def help Command: ledger cred-def - Send Cred Def transaction to the Ledger. Usage: ledger cred-def schema_id= signature_type= [tag=] primary= [revocation=] [fees_inputs=] [fees_outputs=] [extra=] Parameters are: schema_id - Sequence number of schema signature_type - Signature type (only CL supported now) tag - (optional) Allows to distinct between credential definitions for the same issuer and schema. Note that it is mandatory for indy-node version 1.4.x and higher primary - Primary key in json format revocation - (optional) Revocation key in json format fees_inputs - (optional) The list of source inputs fees_outputs - (optional) The list of outputs in the following format: (recipient, amount) extra - (optional) Optional information for fees payment operation Examples: ledger cred-def schema_id=1 signature_type=CL tag=1 primary={"n":"1","s":"2","rms":"3","r":{"age":"4","name":"5"},"rctxt":"6","z":"7"} ```

ianco (Tue, 12 Mar 2019 15:44:31 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=sqPBDjxJ2T7CHsBaE) @lorenz.loesch I'm not sure, but if you check the samples they create cred defs using the anoncreds api rather than the ledger directly: https://github.com/hyperledger/indy-sdk/blob/master/samples/python/src/getting_started.py#L222

ianco (Tue, 12 Mar 2019 15:44:56 GMT):
I don't think anoncreds is available through the cli?

jenklogic (Tue, 12 Mar 2019 16:19:01 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=KecqMYTFwMRyTRio6) @MALodder Hi MALodder. Let me get back to you on that later today when I go back through it and make notes. I will check out your docs as well, so thanks so much for providing those!'

jenklogic (Tue, 12 Mar 2019 16:19:01 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=KecqMYTFwMRyTRio6) @MALodder Hi MALodder. Let me get back to you on that later today when I go back through it and make notes. I will check out your docs as well, so thanks so much for providing those!

adityanalgework1 (Tue, 12 Mar 2019 18:41:29 GMT):
Has joined the channel.

msjeong (Wed, 13 Mar 2019 01:16:57 GMT):
Good morning ~ May I ask a question? I want to set up the Hyperledger Indy network, so I find the document that is getting-started(https://github.com/hyperledger/indy-node/blob/stable/getting-started.md#install-indy) but the document is so hard to setting the network. Could you tell me how to set up the network? or any document?

GEEKTEDDY (Wed, 13 Mar 2019 07:29:41 GMT):
Hi, I am wandering how to design the management services in my system(based on indy), there management services may include the features like issuers can manage the credentials they issued(etc revoc), the system admin add/delete an entity(maybe DID) and even the clients can manage their credentials in their wallets. So are there any modules or even ideas that Sovrin or Indy have to reach these requirement?

reithmayer (Wed, 13 Mar 2019 07:47:10 GMT):
Has joined the channel.

dl2017 (Wed, 13 Mar 2019 14:54:14 GMT):
Has joined the channel.

NasserRahal (Wed, 13 Mar 2019 16:00:40 GMT):
Has joined the channel.

JuanCamilo_0314 (Wed, 13 Mar 2019 19:19:44 GMT):
Has joined the channel.

Sean_Bohan (Wed, 13 Mar 2019 19:46:31 GMT):
Folks: Hey everyone TOMORROW 3/14/2019 - Indy WG CALL!!!! Agenda for Thursday's Working Group call: * Indy SDK Architecture * AGENTBOOK!!!! (from the BC.gov team) Where: https://zoom.us/j/232861185 When: 8am PT, 9am MT, 10am CT, 11amET, 4pmGMT

drummondreed (Wed, 13 Mar 2019 20:33:29 GMT):
@Sean_Bohan Thanks so much for keeping us current. I just heard about this "AgentBook" demo and don't to miss it (even though I'll be at the Linux Foundation Open Source Leadership Summit tomorrow).

Sean_Bohan (Wed, 13 Mar 2019 21:36:25 GMT):
Will be AGENTBOOK as well as recap of the INDY event this week in BC

drummondreed (Thu, 14 Mar 2019 02:14:53 GMT):
Really looking forward to it.

Rchauhan (Thu, 14 Mar 2019 06:12:50 GMT):
Hello Friends, I need some here, I am trying to create a wallet and want to store in my local file system with some information but I did't get success into this. Can anyone please help me into this or any sample which I can follow to do this. I want to do this using java language.

charliez (Thu, 14 Mar 2019 13:31:30 GMT):
Hey, channel, I'm testing the sovrin test network by replaying the AnoncredsRevoc example under indy-sdk repo, and I got the following error message: `Indy Error[113]: Error: Invalid structure, Caused by: RevocationRegistry not found for timestamp: 1552565450` Anyone has any idea about this error? Thanks

Sean_Bohan (Thu, 14 Mar 2019 14:48:50 GMT):
Folks: Due to a mistake I made regarding Daylight Savings Time in the US, the Hyperledger Indy call is happening 3pm GMT, not 4pm GMT Apologies for the confusion.

mxs1491 (Thu, 14 Mar 2019 16:02:49 GMT):
Bummer, just connected as the call was ending. @Sean_Bohan can you let me know when the recording is up loaded, I was really interested in the Agentbook demo.

mxs1491 (Thu, 14 Mar 2019 16:03:03 GMT):
Got bit by the time change...

Sean_Bohan (Thu, 14 Mar 2019 16:06:06 GMT):
Absolutely mxs1491 - on the road today and have to post it when i get decent, non-Starbucks WiFi (later today). Will share it right here

Sean_Bohan (Thu, 14 Mar 2019 16:06:20 GMT):
You getting bit was my fault

theoturner (Thu, 14 Mar 2019 16:52:52 GMT):
Hello all, is the term `master secret` or `link secret` correct to use going forward?

esplinr (Thu, 14 Mar 2019 17:28:45 GMT):
An update on Evernym's contributions to Indy over the past two weeks: https://www.youtube.com/watch?v=NxSwShIfaWw&list=PLRp0viTDxBWGLdZk0aamtahB9cpJGV7ZF&index=1

esplinr (Thu, 14 Mar 2019 17:29:03 GMT):
@lynn.bendixsen I call you out specifically in the video. grin

esplinr (Thu, 14 Mar 2019 17:30:24 GMT):
Key topics this round: SDK CLI commands for setting authorization rules on the config ledger, Catch-up improvements using the Audit Ledger, and changes to the release process and versioning to bring it closer to SemVer.

drummondreed (Fri, 15 Mar 2019 05:03:22 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=rA76ZS49zAFpw7mgC) Link secret

justinr3 (Fri, 15 Mar 2019 08:20:07 GMT):
Has joined the channel.

justinr3 (Fri, 15 Mar 2019 08:21:02 GMT):
Does anyone happen to know why one could be getting the following error? java.util.concurrent.ExecutionException: org.hyperledger.indy.sdk.InvalidParameterException: The value passed to parameter 4 is not valid. at java.util.concurrent.CompletableFuture.reportGet(CompletableFuture.java:357)

lidengjia (Sat, 16 Mar 2019 15:06:41 GMT):
hello,do someone have Indy's underlying architecture diagram?I aim a student who is studying indy.

lidengjia (Sat, 16 Mar 2019 15:07:02 GMT):
the

lidengjia (Sat, 16 Mar 2019 15:07:44 GMT):
i also want to know indy can run in the fabric network?

pieterp (Sun, 17 Mar 2019 03:53:49 GMT):
Has joined the channel.

pknowles (Sun, 17 Mar 2019 20:17:27 GMT):
Published article on the *Overlays data capture architecture* - https://www.dativa.com/introducing-overlays-data-capture-architecture/

bharadwajambati (Mon, 18 Mar 2019 04:26:41 GMT):
indyerror

lidengjia (Mon, 18 Mar 2019 05:13:09 GMT):
thank your sincerely help:blush:

Rchauhan (Mon, 18 Mar 2019 06:19:19 GMT):
Hi friends, I am trying to make build for indy java wrapper but when executing command mvn clean install facing issue relate to POM file "Could not resolve dependencies for project org.hyperledger:indy:jar:1.8.1: Failure to find org.hyperledger:indy:jar:1.8.1-dev-985 in https://repo.maven.apache.org/maven2 was cached in the local repository, resolution will not be reattempted until the update interval of central has elapsed or updates are forced " I also tried to make the changes as suggested in readme file like below. 1. Inside repositories tag block add: sovrin https://repo.sovrin.org/repository/maven-public 2. Inside dependencies tag block add: org.hyperledger indy 1.8.1-dev-985 but again facing same issue, can someone please suggest how to resolve this.

aguel (Mon, 18 Mar 2019 07:59:35 GMT):
Has joined the channel.

Rchauhan (Mon, 18 Mar 2019 11:15:53 GMT):
Hello Indy Friends, I want to join Indy call, please help to know the schedule for the INDY WG CALL

pieterp (Mon, 18 Mar 2019 11:18:44 GMT):
Hi @Rchauhan, see here for all details (scroll down to Meetings for details): https://wiki.hyperledger.org/display/indy/Hyperledger+Indy

Rchauhan (Mon, 18 Mar 2019 11:23:47 GMT):
thanks Pieterp I got it , just one information can we have a technical discussion in this call like how we can use indy in our usecase

Rchauhan (Mon, 18 Mar 2019 11:24:22 GMT):
because I am trying to use indy for our purpose but not able to do this

Rchauhan (Mon, 18 Mar 2019 11:36:21 GMT):
Pieterp, this indy java wrapper is stable to not ?

pieterp (Mon, 18 Mar 2019 11:36:45 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=a4B7LxBk7wu73Z8Bk) Not sure @Rchauhan, but I think you can raise it and if it cannot be answered, hopefully they can refer you to someone or somewhere. Good luck!

mwherman2000 (Mon, 18 Mar 2019 19:46:01 GMT):
ANNOUNCEMENT: The INDY ARM Interactive Explorer is available as of right now: https://github.com/mwherman2000/indy-arm/blob/master/README.md#indy-arm-interactive-explorer ...with about a dozen different viewpoints to start with. Click on any of the views on the left side of the web app to get going.

mwherman2000 (Mon, 18 Mar 2019 19:46:26 GMT):

indy-arm-explorer.png

JakeZheng (Tue, 19 Mar 2019 05:55:09 GMT):
Has joined the channel.

Rchauhan (Tue, 19 Mar 2019 06:20:48 GMT):
Hello Indy Friends need a help

Rchauhan (Tue, 19 Mar 2019 06:21:28 GMT):
when I am trying to execute cargo build command facing an error error: Edition 2018 is unstable and only available for nightly builds of rustc. error: Could not compile `zeroize`. can anyone suggest how to handle this error

pknowles (Tue, 19 Mar 2019 06:33:37 GMT):
Due to clock changes, the call time for today's *Indy Semantics WG* call remains the same for US participants but is an hour earlier than usual for European participants! For more information regarding the specifics, go to #indy-semantics

varadhbhatnagar (Tue, 19 Mar 2019 09:34:31 GMT):
Has joined the channel.

varadhbhatnagar (Tue, 19 Mar 2019 09:35:22 GMT):
Hello, I am new to Hyperledger Indy. Is it possible to implemet KYC on blockchain using Hyperledger Indy ?

mwherman2000 (Tue, 19 Mar 2019 17:03:20 GMT):
Fooling around with Facebook to see where someone can store a personal DID ...Other Names is too restrictive in terms of syntax. However, creating a Life Event is a pretty cool way/place to store your DID...

mwherman2000 (Tue, 19 Mar 2019 17:03:20 GMT):
Fooling around with Facebook to see where someone can store a personal DID ...Other Names is too restrictive in terms of syntax. However, creating a Life Event is a pretty cool way/place to store your DID... ...cool way to bootstrap.

mwherman2000 (Tue, 19 Mar 2019 17:03:29 GMT):

Clipboard - March 19, 2019 11:03 AM

mwherman2000 (Tue, 19 Mar 2019 17:29:12 GMT):
The Licenses and Certifications of your LinkedIn profile works perfectly....

mwherman2000 (Tue, 19 Mar 2019 17:30:26 GMT):

1. Store your person DID in your LinkedIn profile

mwherman2000 (Tue, 19 Mar 2019 17:30:26 GMT):

1. Store your person DID in your LinkedIn profile

mwherman2000 (Tue, 19 Mar 2019 17:30:26 GMT):

1. Store your person DID in your LinkedIn profile

mwherman2000 (Tue, 19 Mar 2019 17:32:30 GMT):

2. Your personal DID as part of your LinkedIn profile...

mwherman2000 (Tue, 19 Mar 2019 17:37:30 GMT):
The View Credential link also works ...it invokes whatever you provide as the Credential URL. MOOCs like edX use this same mechanism to (programmatically) post verifiable confirmations of course completions, etc.

mwherman2000 (Tue, 19 Mar 2019 17:38:00 GMT):
CC: @peacekeeper

vishalnigam (Tue, 19 Mar 2019 18:36:21 GMT):
Has joined the channel.

mwherman2000 (Wed, 20 Mar 2019 01:57:09 GMT):
#iDIDit ...apparently I found a bug in Facebook ...the video associated with my DID doesn't work: https://www.facebook.com/mwherman/posts/10156725176175932

mwherman2000 (Wed, 20 Mar 2019 01:57:09 GMT):
#iDIDit ...apparently I found a bug in Facebook ...the audio channel associated with the video associated with my DID doesn't work: https://www.facebook.com/mwherman/posts/10156725176175932

bharadwajambati (Wed, 20 Mar 2019 04:42:47 GMT):
Can someone help me out in creating a connection-request using indy-cli?

bharadwajambati (Wed, 20 Mar 2019 04:42:47 GMT):
Can someone help me out in creating a connection-request using indy-cli or how to se indy-cli?

Rchauhan (Wed, 20 Mar 2019 09:32:09 GMT):
Hello Indy Friends, please suggest how to change default wallet storage path

dbluhm (Wed, 20 Mar 2019 13:49:49 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=sbFMK9YwZQSieKzsq) @bharadwajambati Indy-cli does not have agent functionality. It is primarily used to interact with the ledger for storing DIDs, schemas, and credential definitions. Creating connections between one party and another is a function of agents. Today, there are a number of agent implementations being actively worked on. A small directory of at least some of those agents can be found at https://github.com/hyperledger/indy-agent/blob/master/README.md#known-agent-implementations

anillewis (Wed, 20 Mar 2019 16:06:25 GMT):
Has joined the channel.

bharadwajambati (Wed, 20 Mar 2019 19:12:56 GMT):
@dbluhm thanks a lot. Is there any documentation which corresponds to indy-cli?

dbluhm (Wed, 20 Mar 2019 19:25:28 GMT):
Assuming we're talking about the the CLI contained in the indy-sdk repo, it does a decent job of self documenting with the help commands. @bharadwajambati

dbluhm (Wed, 20 Mar 2019 19:27:56 GMT):
This design doc has a fair amount of information as well but I think most of that is encapsulated in the help commands: https://github.com/hyperledger/indy-sdk/tree/master/docs/design/001-cli

dbluhm (Wed, 20 Mar 2019 19:43:53 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=tTGnDZBKpHKaiFjNK) @Rchauhan Looks like you might need to update rust

dbluhm (Wed, 20 Mar 2019 19:45:28 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=R9hAW4wmXJB9g3zEq) @Rchauhan See `storage_config` parameter of create and open wallet sdk functions: https://github.com/hyperledger/indy-sdk/blob/master/wrappers/python/indy/wallet.py#L21

bharadwajambati (Wed, 20 Mar 2019 20:33:20 GMT):
thanks again @dbluhm

pioneer (Thu, 21 Mar 2019 03:07:19 GMT):
Has joined the channel.

Sean_Bohan (Thu, 21 Mar 2019 13:51:44 GMT):
Hey everyone TODAY 3/21/2019 - Indy WG CALL!!!! Agenda for Thursday's Working Group call: * Recaps from Agents, Semantics, URSA and Sovrin Crypto calls * BruceC shares his work with Pico Agents * Open Discussion Where: https://zoom.us/j/232861185 When: 8am PT, 9am MT, 10am CT, 11amET, 3pmGMT

randyz (Thu, 21 Mar 2019 14:28:03 GMT):
Has joined the channel.

anillewis (Thu, 21 Mar 2019 14:52:16 GMT):
Team...I have question on where the pair-wise DIDs are stored. As per this https://www.evernym.com/wp-content/uploads/2017/07/What-Goes-On-The-Ledger.pdf pairwise DID is stored off-ledger but as per this https://github.com/hyperledger/indy-sdk/blob/master/docs/getting-started/indy-walkthrough.md nym_request = await ledger.build_nym_request(steward['did'], decrypted_connection_response['did'], decrypted_connection_response['verkey'], None, role) await ledger.sign_and_submit_request(steward['pool'], steward['wallet'], steward['did'], nym_request) both faber and steward store the pairwise DID using the NYM transaction. Am I correct in assuming that pairwise DID is stored in the indy ledger?

AngeloBestetti (Thu, 21 Mar 2019 18:13:28 GMT):
Has joined the channel.

esune (Thu, 21 Mar 2019 20:37:28 GMT):
Has joined the channel.

dbluhm (Thu, 21 Mar 2019 21:12:01 GMT):
Anyone have a link to recent Indy WG call recordings?

nekia (Fri, 22 Mar 2019 07:32:25 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=WXC3f3ZrifgPbtuH3) https://docs.google.com/document/d/1xLjrPdk-1w8u9eJ7M6kWtZnN8ubpyclOnDYuEaJeLVo/edit?usp=sharing I've created the draft. In which projects should I issue a jira to proceed this task?

nekia (Fri, 22 Mar 2019 07:43:25 GMT):
@nage @mboyd

nekia (Fri, 22 Mar 2019 07:45:39 GMT):
Not Jira, Github issue.

BernardLin (Fri, 22 Mar 2019 14:29:05 GMT):
Has joined the channel.

nage (Fri, 22 Mar 2019 15:06:43 GMT):
Awesome! @mboyd will get this over to the video owner and we will get a ticket going somewhere on https://sovrin.atlassian.net/secure/Dashboard.jspa

Gruszka (Fri, 22 Mar 2019 21:43:22 GMT):
Has joined the channel.

Gruszka (Fri, 22 Mar 2019 22:40:38 GMT):
Hi guys, I’m Mateusz, new here, can you give me hits how can I help you with project? Do you have a list with ongoing tasks? I have also second question, I tried to play around with Indy and I stucked myself at connecting to pool. I have created it with ./start_pool script so I have nodes running in Docker containers. To connect to pool I found out that I need to create pool config, and pool config needs genesis transactions, but genesis transactions are stored in docker containers, so I’m not sure where I should run my IndyCLI. Should I try to connect in a) docker container aka Node1 or b) it can be completely other machine but copy genesis transaction file from one of the containers (I tried it, without success). My errors are probably TimeOuts: “Pool ‘myPool’ has not been connected” . Do you have any tips?

Gruszka (Fri, 22 Mar 2019 22:40:38 GMT):
Hi guys, I have a question, I tried to play around with Indy and I stucked myself at connecting to pool. I have created it with ./start_pool script so I have nodes running in Docker containers. To connect to pool I found out that I need to create pool config, and pool config needs genesis transactions, but genesis transactions are stored in docker containers, so I’m not sure where I should run my IndyCLI. Should I try to connect in a) docker container aka Node1 or b) it can be completely other machine but copy genesis transaction file from one of the containers (I tried it, without success). My errors are probably TimeOuts: “Pool ‘myPool’ has not been connected” . Do you have any tips?

kdenhartog (Fri, 22 Mar 2019 23:05:45 GMT):
On the front of running through the Getting started guide, I'd checkout https://github.com/kdenhartog/indy-dev

kdenhartog (Fri, 22 Mar 2019 23:06:36 GMT):
However, I think you'll find more value out of the AgentBook work which can be found here https://bit.ly/ibc_ab

kdenhartog (Fri, 22 Mar 2019 23:07:34 GMT):
This is another option I'd suggest that's in between the two and would be helpful https://github.com/hyperledger/education/tree/master/LFS171x/indy-material/nodejs

nekia (Sat, 23 Mar 2019 00:56:57 GMT):
Great! Thanks. @nage

Gruszka (Sat, 23 Mar 2019 19:44:47 GMT):
@kdenhartog Thanks!

mxs1491 (Mon, 25 Mar 2019 07:06:56 GMT):
Hi @Sean_Bohan did the recording of the AgentBook demo get uploaded? Sorry, if I missed the post...

NutecDev (Mon, 25 Mar 2019 10:06:38 GMT):
Has joined the channel.

ivanpagac (Mon, 25 Mar 2019 12:14:18 GMT):
Has joined the channel.

ivanpagac (Mon, 25 Mar 2019 12:14:36 GMT):
Hi guys, trying to use wrapper for iOS however I am getting the libswiftCore.dylib library not found referenced from Indy.framework. I already tried ebmbed swift libraries but no success. Could you at least navigate me here ? Thanks in advance. #indy

hamidm (Mon, 25 Mar 2019 12:14:52 GMT):
Has joined the channel.

hamidm (Mon, 25 Mar 2019 12:17:27 GMT):
HI, I'm trying to execute the how tos in C#/DotNet, and it seems I have a error code 308, when I try to open a pool.

hamidm (Mon, 25 Mar 2019 12:17:43 GMT):
string walletName = "myWallet"; string poolName = "pool"; string stewardSeed = "000000000000000000000000Steward1"; string poolConfig = "{\"genesis_txn\": \"/Users/hm/ws/SB/indy-sdk/cli/docker_pool_transactions_genesis\"}"; // Step 2 // Tell SDK which pool you are going to use. You should have already started // this pool using docker compose or similar. Console.WriteLine("Step 2 -- Creating a new local pool ledger configuration that can be used later to connect pool nodes."); Pool.CreatePoolLedgerConfigAsync(poolName, poolConfig); Console.WriteLine(" Open pool ledger and get the pool handle from libindy."); Pool pool = Pool.OpenPoolLedgerAsync(poolName, poolConfig).Result; Console.WriteLine(" Creates a new identity wallet."); Wallet.CreateWalletAsync(poolName, walletName, "default", string.Empty, string.Empty); Console.WriteLine(" Open identity wallet and get the wallet handle from libindy."); ` string walletName = "myWallet"; string poolName = "pool"; string stewardSeed = "000000000000000000000000Steward1"; string poolConfig = "{\"genesis_txn\": \"/Users/hm/ws/SB/indy-sdk/cli/docker_pool_transactions_genesis\"}"; // Step 2 // Tell SDK which pool you are going to use. You should have already started // this pool using docker compose or similar. Console.WriteLine("Step 2 -- Creating a new local pool ledger configuration that can be used later to connect pool nodes."); Pool.CreatePoolLedgerConfigAsync(poolName, poolConfig); Console.WriteLine(" Open pool ledger and get the pool handle from libindy."); Pool pool = Pool.OpenPoolLedgerAsync(poolName, poolConfig).Result; Console.WriteLine(" Creates a new identity wallet."); Wallet.CreateWalletAsync(poolName, walletName, "default", string.Empty, string.Empty); Console.WriteLine(" Open identity wallet and get the wallet handle from libindy."); string walletName = "myWallet"; string poolName = "pool"; string stewardSeed = "000000000000000000000000Steward1"; string poolConfig = "{\"genesis_txn\": \"/Users/hm/ws/SB/indy-sdk/cli/docker_pool_transactions_genesis\"}"; // Step 2 // Tell SDK which pool you are going to use. You should have already started // this pool using docker compose or similar. Console.WriteLine("Step 2 -- Creating a new local pool ledger configuration that can be used later to connect pool nodes."); Pool.CreatePoolLedgerConfigAsync(poolName, poolConfig); Console.WriteLine(" Open pool ledger and get the pool handle from libindy."); Pool pool = Pool.OpenPoolLedgerAsync(poolName, poolConfig).Result; Console.WriteLine(" Creates a new identity wallet."); Wallet.CreateWalletAsync(poolName, walletName, "default", string.Empty, string.Empty); Console.WriteLine(" Open identity wallet and get the wallet handle from libindy."); Wallet wallet = Wallet.OpenWalletAsync(walletName, string.Empty, string.Empty).Result;

hamidm (Mon, 25 Mar 2019 12:17:43 GMT):
string walletName = "myWallet"; string poolName = "pool"; string stewardSeed = "000000000000000000000000Steward1"; string poolConfig = "{\"genesis_txn\": \"/Users/hm/ws/SB/indy-sdk/cli/docker_pool_transactions_genesis\"}"; // Step 2 // Tell SDK which pool you are going to use. You should have already started // this pool using docker compose or similar. Console.WriteLine("Step 2 -- Creating a new local pool ledger configuration that can be used later to connect pool nodes."); Pool.CreatePoolLedgerConfigAsync(poolName, poolConfig); Console.WriteLine(" Open pool ledger and get the pool handle from libindy."); Pool pool = Pool.OpenPoolLedgerAsync(poolName, poolConfig).Result; Console.WriteLine(" Creates a new identity wallet."); Wallet.CreateWalletAsync(poolName, walletName, "default", string.Empty, string.Empty); Console.WriteLine(" Open identity wallet and get the wallet handle from libindy."); ` string walletName = "myWallet"; string poolName = "pool"; string stewardSeed = "000000000000000000000000Steward1"; string poolConfig = "{\"genesis_txn\": \"/Users/hm/ws/SB/indy-sdk/cli/docker_pool_transactions_genesis\"}"; // Step 2 // Tell SDK which pool you are going to use. You should have already started // this pool using docker compose or similar. Console.WriteLine("Step 2 -- Creating a new local pool ledger configuration that can be used later to connect pool nodes."); Pool.CreatePoolLedgerConfigAsync(poolName, poolConfig); Console.WriteLine(" Open pool ledger and get the pool handle from libindy."); Pool pool = Pool.OpenPoolLedgerAsync(poolName, poolConfig).Result; Console.WriteLine(" Creates a new identity wallet."); Wallet.CreateWalletAsync(poolName, walletName, "default", string.Empty, string.Empty); Console.WriteLine(" Open identity wallet and get the wallet handle from libindy."); Wallet wallet = Wallet.OpenWalletAsync(walletName, string.Empty, string.Empty).Result;`

hamidm (Mon, 25 Mar 2019 12:18:44 GMT):
` string walletName = "myWallet"; string poolName = "pool"; string stewardSeed = "000000000000000000000000Steward1"; string poolConfig = "{\"genesis_txn\": \"/Users/hm/ws/SB/indy-sdk/cli/docker_pool_transactions_genesis\"}"; // Step 2 // Tell SDK which pool you are going to use. You should have already started // this pool using docker compose or similar. Console.WriteLine("Step 2 -- Creating a new local pool ledger configuration that can be used later to connect pool nodes."); Pool.CreatePoolLedgerConfigAsync(poolName, poolConfig); Console.WriteLine(" Open pool ledger and get the pool handle from libindy."); Pool pool = Pool.OpenPoolLedgerAsync(poolName, poolConfig).Result; Console.WriteLine(" Creates a new identity wallet."); Wallet.CreateWalletAsync(poolName, walletName, "default", string.Empty, string.Empty); Console.WriteLine(" Open identity wallet and get the wallet handle from libindy."); Wallet wallet = Wallet.OpenWalletAsync(walletName, string.Empty, string.Empty).Result;`

hamidm (Mon, 25 Mar 2019 12:23:09 GMT):
Any workaround or fix ?

hamidm (Mon, 25 Mar 2019 12:29:27 GMT):
In addition, when I use another Nuget package, IndyDotNet, i get this error: IndyDotNet.Exceptions.InvalidStructureException: A value being processed is not valid.

hamidm (Mon, 25 Mar 2019 12:29:42 GMT):
The protocol version passed is 2.

Ryan2 (Tue, 26 Mar 2019 04:47:58 GMT):
Has joined the channel.

S.pradeepkumar (Tue, 26 Mar 2019 08:11:44 GMT):
Has joined the channel.

S.pradeepkumar (Tue, 26 Mar 2019 08:12:14 GMT):
Hi, i am unable t oconnect pool network

S.pradeepkumar (Tue, 26 Mar 2019 08:13:21 GMT):
Hi, i am unable to connect pool it says pool "testnet " has not connected

MichalRybarczyk (Tue, 26 Mar 2019 13:31:59 GMT):
Has joined the channel.

lepar (Tue, 26 Mar 2019 13:55:17 GMT):
Has joined the channel.

silviu (Tue, 26 Mar 2019 14:50:29 GMT):
Has joined the channel.

silviu (Tue, 26 Mar 2019 14:50:45 GMT):
Hi everyone. I'm trying to build a concept for federated identity on blockchain. would Hyperledger Indy be the right one to do that or should I use another Hyperledger framework ? Any suggestions?

PavanDurgadsimi (Tue, 26 Mar 2019 19:24:15 GMT):
Hello guys i just saw the overview video by Kyle DH, and i was bit in a confusion that how could you prove your assets to the other guys. In general senario, suppose if Alice wants to prove her degree certificate to the bank then how could she prove that? The video ends without the full demo, if anybody here knows how to do it then please share the link.

PavanDurgadsimi (Tue, 26 Mar 2019 19:26:09 GMT):
And how to pass on the operations into a production grade application such that the local ports are not used but real systems get involved in the process of exchanging data.

swcurran (Tue, 26 Mar 2019 19:41:33 GMT):
Hey Folks, BC Gov has posted a new lightweight *_Code With Us_* procurement opportunity (aka bounty) for "*Interoperable Hyperledger Indy Mobile Agent Capabilities*". Click here to see the details: https://bcdevexchange.org/opportunities/cwu/opp-interoperable-hyperledger-indy-mobile-agent-capabilities To apply for the opportunity, you must register with the BC Dev Exchange website (link above) and then submit a short proposal by April 1, 2019. The proposal evaluation criteria is provided in the posting. There is a github issue linked to the opoortunity for questions and answers, or you feel free to ask directly here. We'll repost any questions to the github issue. @SeanBohan_Sean @nage - please add this to the announcements on the Indy call this week.

swcurran (Tue, 26 Mar 2019 19:41:33 GMT):
Hey Folks, BC Gov has posted a new lightweight *_Code With Us_* procurement opportunity (aka bounty) for *Interoperable Hyperledger Indy Mobile Agent Capabilities*. Click here to see the details: https://bcdevexchange.org/opportunities/cwu/opp-interoperable-hyperledger-indy-mobile-agent-capabilities To apply for the opportunity, you must register with the BC Dev Exchange website (link above) and then submit a short proposal by April 1, 2019. The proposal evaluation criteria is provided in the posting. There is a github issue linked to the opoortunity for questions and answers, or you feel free to ask directly here. We'll repost any questions to the github issue. @Sean_Bohan @nage - please add this to the announcements on the Indy call this week.

swcurran (Tue, 26 Mar 2019 19:41:33 GMT):
Hey Folks, BC Gov has posted a new lightweight *_Code With Us_* procurement opportunity (aka bounty) for "*Interoperable Hyperledger Indy Mobile Agent Capabilities*". Click here to see the details: https://bcdevexchange.org/opportunities/cwu/opp-interoperable-hyperledger-indy-mobile-agent-capabilities To apply for the opportunity, you must register with the BC Dev Exchange website (link above) and then submit a short proposal by April 1, 2019. The proposal evaluation criteria is provided in the posting. There is a github issue linked to the opoortunity for questions and answers, or you feel free to ask directly here. We'll repost any questions to the github issue. @Sean_Bohan @nage - please add this to the announcements on the Indy call this week.

swcurran (Tue, 26 Mar 2019 19:41:33 GMT):
Hey Folks, BC Gov has posted a new lightweight *_Code With Us_* procurement opportunity (aka bounty) for "*Interoperable Hyperledger Indy Mobile Agent Capabilities*". Click here to see the details: https://bcdevexchange.org/opportunities/cwu/opp-interoperable-hyperledger-indy-mobile-agent-capabilities To apply for the opportunity, you must register with the BC Dev Exchange website (link above) and then submit a short proposal by April 1, 2019. The proposal evaluation criteria is provided in the posting. There is a github issue linked to the opoortunity for questions and answers, or you feel free to ask directly here. We'll repost any questions to the github issue. @Sean_Bohan @nage - please add this to the announcements on the Indy call this week.

nage (Wed, 27 Mar 2019 05:54:23 GMT):
I have just cross posted a message to the Hyperledger Indy and Hyperledger TSC lists announcing that with the CII badge now complete we are ready to move Indy from incubating to active status. Please reply with any additional information or support you think would be helpful. I'm hoping we can get to a :+1: from the TSC this week. (@SteveGoob I also added your google docs page to the wiki so everyone can see and edit it -- fixes and improvements welcome!)

ArushBartaria (Wed, 27 Mar 2019 07:07:53 GMT):
Has joined the channel.

ArushBartaria (Wed, 27 Mar 2019 07:08:33 GMT):
Hi all! I am a Python3 programmer. I want to contribute to the Indy project. Can you give me some leads on where to begin.

pknowles (Wed, 27 Mar 2019 07:19:31 GMT):
@pyraman The following link might shed some light on your query regarding the feasibility of implementing HL Indy with a smart contract engine. https://www.luxoft.com/videos/cordentity-power-of-self-sovereign-identity-and-verifiable-credentials-for-corda-smart-contracts/

pyraman (Wed, 27 Mar 2019 07:19:32 GMT):
Has joined the channel.

AW_test (Wed, 27 Mar 2019 09:06:07 GMT):
Has joined the channel.

AW_test (Wed, 27 Mar 2019 09:06:46 GMT):
Hi everybody I'm not sure if I'm right here, I'm trying to do the tutorial https://github.com/hyperledger/indy-sdk/blob/master/docs/getting-started/indy-walkthrough.md In step two I get that the command pool.create_pool_ledger_config is not known, better to say the second part of the command create_pool_ledger_config. As far as I know I have everything installed according to the prerequisite. Can anyone help me? Thank you

Maarep (Wed, 27 Mar 2019 10:00:31 GMT):
Has joined the channel.

dhuseby (Wed, 27 Mar 2019 17:06:54 GMT):
@here @nage @MALodder I'm trying to sign up to the sovrin.org gitlab server so that I can run through setting up a "volunteer" runner to help with ursa CI

dhuseby (Wed, 27 Mar 2019 17:07:15 GMT):
I'm getting an "ERROR for site owner: Invalid key type" message in the reCAPTCHA

dhuseby (Wed, 27 Mar 2019 17:07:24 GMT):
reCAPTCHA isn't working so I am unable to sign up

dhuseby (Wed, 27 Mar 2019 17:07:56 GMT):
can you guys fix that? I'd like to do this today to prep for the TSC talk tomorrow. I need to speak from an educated position on this or else the TSC will eat me for breakfast

MALodder (Wed, 27 Mar 2019 17:29:31 GMT):
@dhuseby I created an account for you

MALodder (Wed, 27 Mar 2019 17:29:37 GMT):
sent it to you over Signal

MALodder (Wed, 27 Mar 2019 17:29:37 GMT):
sent the credentials to you over Signal

lynn.bendixsen (Wed, 27 Mar 2019 18:15:27 GMT):
@ArushBartaria scroll down on https://github.com/hyperledger/indy-sdk to see tutorials and getting started instructions. Enjoy!

dhuseby (Wed, 27 Mar 2019 18:26:38 GMT):
@MALodder roger

MALodder (Wed, 27 Mar 2019 18:27:44 GMT):
@dhuseby what's our vector victor

dhuseby (Wed, 27 Mar 2019 18:42:51 GMT):
LOL

MALodder (Wed, 27 Mar 2019 18:49:28 GMT):
@dhuseby done was over under and under was over done

MALodder (Wed, 27 Mar 2019 18:49:28 GMT):
@dhuseby Dunn was under Over and I was under Dunn.

MALodder (Wed, 27 Mar 2019 18:50:50 GMT):
So, Dunn you were under Oveur and over Unger.

MALodder (Wed, 27 Mar 2019 18:51:02 GMT):
That's right. Dunn was over Under, and I was over Dunn.

MALodder (Wed, 27 Mar 2019 18:51:13 GMT):
So you see, both Dunn and I were under Over, even though I was under Dunn.

MALodder (Wed, 27 Mar 2019 18:51:21 GMT):
Dunn was over Under, and I was over Dunn.

dhuseby (Wed, 27 Mar 2019 18:51:24 GMT):
you're bored it seems

dhuseby (Wed, 27 Mar 2019 18:51:28 GMT):
: )

MALodder (Wed, 27 Mar 2019 18:51:35 GMT):
listening to lectures at DSU

MALodder (Wed, 27 Mar 2019 18:51:39 GMT):
this week

MALodder (Wed, 27 Mar 2019 18:52:26 GMT):
gotta pass on the legacy

MALodder (Wed, 27 Mar 2019 18:52:31 GMT):
its a classic

dhuseby (Wed, 27 Mar 2019 18:52:36 GMT):
@MALodder reminds me of this: https://www.youtube.com/watch?v=hoe24aSvLtw

dhuseby (Wed, 27 Mar 2019 18:52:55 GMT):
doctor, doctor, doctor, doctor, doctor

MALodder (Wed, 27 Mar 2019 18:52:59 GMT):
haha

dhuseby (Wed, 27 Mar 2019 19:21:11 GMT):
@MALodder ok, I've got an 8-core, 32 GB of RAM machine under my desk running Ubuntu 18.08

dhuseby (Wed, 27 Mar 2019 19:21:31 GMT):
it used to be the built bot for the Rust toolchain for *BSD and few other architectures

dhuseby (Wed, 27 Mar 2019 19:21:39 GMT):
but it's sitting idle at the moment

dhuseby (Wed, 27 Mar 2019 19:21:55 GMT):
I want to add it as a runner to the Ursa CI/CD pipeline

dhuseby (Wed, 27 Mar 2019 19:22:10 GMT):
I have logged in

MALodder (Wed, 27 Mar 2019 19:22:11 GMT):
you should be able to trigger the pipeline

dhuseby (Wed, 27 Mar 2019 19:22:13 GMT):
added an SSH key

dhuseby (Wed, 27 Mar 2019 19:22:21 GMT):
I've installed the gitlab-runner package and docker

dhuseby (Wed, 27 Mar 2019 19:22:24 GMT):
everything is there

dhuseby (Wed, 27 Mar 2019 19:22:32 GMT):
I just need to get a runner token

MALodder (Wed, 27 Mar 2019 19:22:41 GMT):
oh hold on

dhuseby (Wed, 27 Mar 2019 19:22:50 GMT):
everything says I should be able to create a token by going to the project page and clicking settings and then creating a runner token

dhuseby (Wed, 27 Mar 2019 19:22:57 GMT):
wait! don't create it for me

dhuseby (Wed, 27 Mar 2019 19:23:09 GMT):
the HL CI/CD proposal is to have this be self-service

dhuseby (Wed, 27 Mar 2019 19:23:23 GMT):
did you turn that off or is that how Gitlab just is?

MALodder (Wed, 27 Mar 2019 19:25:30 GMT):
I'm not creating it for you

MALodder (Wed, 27 Mar 2019 19:25:35 GMT):
you should be able to just see it then

MALodder (Wed, 27 Mar 2019 19:25:53 GMT):
Go to the settings -> CI/CD -> Runners and you should see the token

dhuseby (Wed, 27 Mar 2019 19:26:03 GMT):
settings?

dhuseby (Wed, 27 Mar 2019 19:26:11 GMT):
in the left column of buttons?

dhuseby (Wed, 27 Mar 2019 19:26:58 GMT):
there is no "settings" for me

dhuseby (Wed, 27 Mar 2019 19:27:24 GMT):
I can go to "user settings" and there is a thing for access tokens there

dhuseby (Wed, 27 Mar 2019 19:27:29 GMT):
but I don't think that is it

MALodder (Wed, 27 Mar 2019 19:28:29 GMT):
hold on, I can't log in, my 2FA keeps yielding a 500 error

dhuseby (Wed, 27 Mar 2019 19:28:46 GMT):

18PS4OOrqszkk5MksN9Z.jpeg

dhuseby (Wed, 27 Mar 2019 19:28:56 GMT):
That is what I am seeing

dhuseby (Wed, 27 Mar 2019 19:44:54 GMT):
@MALodder anything?

dhuseby (Wed, 27 Mar 2019 19:55:35 GMT):
@MALodder I think I know what's going on

dhuseby (Wed, 27 Mar 2019 19:55:41 GMT):
looking at the permissions of Gitlab: https://docs.gitlab.com/ee/user/permissions.html#gitlab-cicd-permissions

dhuseby (Wed, 27 Mar 2019 19:56:00 GMT):
only the maintainer and admin roles can add specific runners

dhuseby (Wed, 27 Mar 2019 19:56:21 GMT):
I want developers to be able to add runners as well

dhuseby (Wed, 27 Mar 2019 19:56:34 GMT):
I guess it is ok if the developer has to ping a maintainer first to get an access token

dhuseby (Wed, 27 Mar 2019 19:56:40 GMT):
I'm ok with that

MALodder (Wed, 27 Mar 2019 19:59:15 GMT):
okay try it now

MALodder (Wed, 27 Mar 2019 19:59:32 GMT):
you should it now if you refresh

dhuseby (Wed, 27 Mar 2019 20:10:16 GMT):
got it

MALodder (Wed, 27 Mar 2019 20:46:18 GMT):
:thumbsup:

dhuseby (Wed, 27 Mar 2019 21:15:15 GMT):
I can't get it to build ursa tho

dhuseby (Wed, 27 Mar 2019 21:15:26 GMT):
tried the rust:latest as the default docker image

dhuseby (Wed, 27 Mar 2019 21:15:46 GMT):
but the build yaml doesn't install libsodium-dev package

dhuseby (Wed, 27 Mar 2019 21:15:56 GMT):
you'll have to look at the jobs to see the messages

dhuseby (Wed, 27 Mar 2019 21:16:05 GMT):
I have my runner using docker as the executor

MALodder (Wed, 27 Mar 2019 23:14:42 GMT):
here is the Dockerfile I use to build an image that works for the runner

MALodder (Wed, 27 Mar 2019 23:14:47 GMT):
``` FROM ubuntu:18.04 LABEL maintainer="Michael Lodder " ENV PATH /root/.cargo/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin RUN apt-get update 2>&1 > /dev/null \ && apt-get install -qq -y git curl emscripten pkg-config libssl1.0.0 libssl-dev 2>&1 > /dev/null \ && cd /usr/lib/x86_64-linux-gnu \ && ln -s libssl.so.1.0.0 libssl.so.10 \ && ln -s libcrypto.so.1.0.0 libcrypto.so.10 \ && curl https://sh.rustup.rs -sSf | sh -s -- -y \ && curl https://rustwasm.github.io/wasm-pack/installer/init.sh -sSf | sh \ && rustup target add wasm32-unknown-unknown ```

dhuseby (Wed, 27 Mar 2019 23:18:45 GMT):
so why isn't that getting done when the job is run on my box?

dhuseby (Wed, 27 Mar 2019 23:18:53 GMT):
or do I have to build my own custom docker image

dhuseby (Wed, 27 Mar 2019 23:19:01 GMT):
and set that as the default image for my runner?

dhuseby (Wed, 27 Mar 2019 23:19:18 GMT):
if that's the case, shouldn't we check this dockerfile into the repo?

dhuseby (Wed, 27 Mar 2019 23:20:12 GMT):
or better yet, publish a docker image under the HL docker hub account so that we can all just set our runners to hyperledger:ursa-builder?

dhuseby (Wed, 27 Mar 2019 23:21:05 GMT):
@MALodder the gitlab docs suggest that the .gitlab-ci.yaml file can specify which image to use and the one in the runner's config is only used when one isn't specified in the CI yaml

dhuseby (Wed, 27 Mar 2019 23:21:21 GMT):
I think we should do the follow:

dhuseby (Wed, 27 Mar 2019 23:21:35 GMT):
1) check in the dockerfile into the project repo

dhuseby (Wed, 27 Mar 2019 23:22:24 GMT):
2) publish the docker image on docker hub under hyperledger:-builder (as opposed to hyperledger:-test)

dhuseby (Wed, 27 Mar 2019 23:22:44 GMT):
3) check the .gitlab-ci.yaml file into the repo

dhuseby (Wed, 27 Mar 2019 23:23:10 GMT):
4) document how to set up a runner on the wiki and in the readme file

MALodder (Wed, 27 Mar 2019 23:23:20 GMT):
yes it can but we haven't set up the registry yet

MALodder (Wed, 27 Mar 2019 23:23:38 GMT):
the .gitlab-ci-yaml will have it pull a prebuilt image when we finally get that running

MALodder (Wed, 27 Mar 2019 23:23:45 GMT):
then you won't have to build the image yourself

dhuseby (Wed, 27 Mar 2019 23:23:50 GMT):
gotcha

dhuseby (Wed, 27 Mar 2019 23:24:02 GMT):
so now I have to run that dockerfile and then change my runner.conf to use the built image?

dhuseby (Wed, 27 Mar 2019 23:24:12 GMT):
for now, I mean

MALodder (Wed, 27 Mar 2019 23:24:26 GMT):
then all we'll have to do is just add the `image: ubuntu-rust` name value pair at the top of the .yml file and runners will be all set

MALodder (Wed, 27 Mar 2019 23:24:31 GMT):
yes

MALodder (Wed, 27 Mar 2019 23:24:52 GMT):
or you can set the runner to use the shell executor

dhuseby (Wed, 27 Mar 2019 23:25:11 GMT):
I think we should be moving to publishing docker images under hyperledger:-builder on docker hub eventually

MALodder (Wed, 27 Mar 2019 23:25:20 GMT):
that would be awesome

dhuseby (Wed, 27 Mar 2019 23:25:27 GMT):
we have a docker hub account I think

MALodder (Wed, 27 Mar 2019 23:25:44 GMT):
I really want it so all docker images are pulled from predefined and accepted images

dhuseby (Wed, 27 Mar 2019 23:25:46 GMT):
if not, I think I might know somebody who is allowed to set that up for Hyperledger

vishalse (Wed, 27 Mar 2019 23:25:58 GMT):
Has joined the channel.

MALodder (Wed, 27 Mar 2019 23:26:12 GMT):
runners for windows and mac will have other requirements

vishalse (Wed, 27 Mar 2019 23:26:23 GMT):
HI I'd like to contribute to the hyperledger fabric python project and possibly come in as an intern this summer.

dhuseby (Wed, 27 Mar 2019 23:26:27 GMT):
why not change the dockerfile to start with rust:latest instead of ubuntu:18.04?

dhuseby (Wed, 27 Mar 2019 23:26:37 GMT):
the rust docker images are published by the rust team

vishalse (Wed, 27 Mar 2019 23:26:37 GMT):
Is there any open issue that needs to be tackled

dhuseby (Wed, 27 Mar 2019 23:26:53 GMT):
@vishalse hows it going?

dhuseby (Wed, 27 Mar 2019 23:26:56 GMT):
welcome!

dhuseby (Wed, 27 Mar 2019 23:27:07 GMT):
I think you're in the wrong chat room though : )

vishalse (Wed, 27 Mar 2019 23:27:11 GMT):
I am great thanks for asking @dhuseby .

MALodder (Wed, 27 Mar 2019 23:27:19 GMT):
I would like 4 linux runners ubuntu, centos, suse, android

vishalse (Wed, 27 Mar 2019 23:27:20 GMT):
Someone in general directed me here

MALodder (Wed, 27 Mar 2019 23:27:29 GMT):
what OS is rust:latest?

dhuseby (Wed, 27 Mar 2019 23:27:39 GMT):
I think debian

dhuseby (Wed, 27 Mar 2019 23:27:51 GMT):
which could be a problem

MALodder (Wed, 27 Mar 2019 23:27:59 GMT):
trying it now

dhuseby (Wed, 27 Mar 2019 23:28:15 GMT):
the debian packages tend to be overly conservative when it comes to tracking updates

dhuseby (Wed, 27 Mar 2019 23:28:23 GMT):
a lot of times debian packages are too old

kdenhartog (Wed, 27 Mar 2019 23:29:01 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=QjNpQqLnmy5fzMphv) @dhuseby The irony :laughing:

dhuseby (Wed, 27 Mar 2019 23:29:20 GMT):
@vishalse if you're interested in the fabric-python project, you should join the #fabric-sdk and #fabric and #fabric-questions channels on this server

dhuseby (Wed, 27 Mar 2019 23:29:28 GMT):
they will be able to give you more help I think

dhuseby (Wed, 27 Mar 2019 23:31:36 GMT):
@vishalse the person you should start talking to is @wangdong. he is the mentor for the fabric python sdk internship project.

dhuseby (Wed, 27 Mar 2019 23:31:45 GMT):
it would be a good thing to get to know him better

MALodder (Wed, 27 Mar 2019 23:32:22 GMT):
rust:latest uses debian 9

dhuseby (Wed, 27 Mar 2019 23:32:35 GMT):
@kdenhartog I so wish this was IRC. I'd sic the mockbot on you or /mute you or something petty like that

dhuseby (Wed, 27 Mar 2019 23:32:37 GMT):
: )

dhuseby (Wed, 27 Mar 2019 23:32:57 GMT):
@MALodder deb9 is fine

MALodder (Wed, 27 Mar 2019 23:36:25 GMT):
well I still think those 4 Dockers should be used

MALodder (Wed, 27 Mar 2019 23:36:49 GMT):
we could even add fedora, redhat, and debian but I think those 4 should pretty much cover it

vishalse (Wed, 27 Mar 2019 23:38:03 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=eKJogsMFfJT8peEue) @dhuseby thanks, just sent both wangdong and dexhunter a pm. Really appreciate the help @dhuseby

dhuseby (Thu, 28 Mar 2019 00:34:56 GMT):
@vishalse you're welcome! if you need anything else, find me.

JakeZheng (Thu, 28 Mar 2019 02:23:26 GMT):
missing 1.8.x indy-objective-c sdk in https://repo.sovrin.org/ios/libindy/stable/indy-objc/

infominer33 (Thu, 28 Mar 2019 03:21:03 GMT):
seems like dropping the wiki was premature :(

infominer33 (Thu, 28 Mar 2019 03:22:02 GMT):
at least I can still get to it via archive https://web.archive.org/web/20181207041937/https://wiki.hyperledger.org/projects/indy/documentation

Rchauhan (Thu, 28 Mar 2019 11:22:41 GMT):
Hello Indy friends, need a small help After the indy setup in my machine I am able to create and connect pool but after machine restart I am not able to connect with the pool. Anyone please suggest how to resolve this, Please friends reply into this Thanks

Rchauhan (Thu, 28 Mar 2019 12:27:39 GMT):
Hello friends anyone is der ?

dbluhm (Thu, 28 Mar 2019 13:33:49 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=pqZtdXcDr7yuwS5NG) @Rchauhan Sounds like you need to start the docker pool again. Unless you specifically instructed it to come back up after restart, I don't believe docker does that automatically.

nage (Thu, 28 Mar 2019 14:27:22 GMT):
The TSC just approved moving Indy from Incubating to Active! Thanks everyone for the work to make the proposal successful!

esplinr (Thu, 28 Mar 2019 21:26:40 GMT):
That's great news!

esplinr (Thu, 28 Mar 2019 21:28:47 GMT):
Bi-weekly update on Evernym's work for Hyperledger Indy and Sovrin: https://www.youtube.com/watch?v=isC86B2hhiI&list=PLRp0viTDxBWGLdZk0aamtahB9cpJGV7ZF Topics covered: * Nikita Khateev: Indy HIPE credential exchange proposal updates. * Artem Ivanov: Indy SDK architecture proposal updates. * Artem Ivanov: Indy SDK release 1.8.2. * Artem Obruchnikov: Indy Node uses the Audit Ledger to restore the three-phase-commit state after restart. * Brent Zundel: Indy HIPE for anoncreds 2.0 and Sovrin SIP for tokens.

eripi (Fri, 29 Mar 2019 02:18:17 GMT):
Has joined the channel.

kclinux (Fri, 29 Mar 2019 04:01:42 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=RFGyP6hcjbNrfb2Kc) @Rchauhan Hi @Rchauhan There are several versions of "Alice's example". I have come up an article showing the storyline and flow for indy-dev. See if this helps. https://medium.com/@kctheservant/exploring-hyperledger-indy-through-indy-dev-example-10075d2547ae

alokkv (Fri, 29 Mar 2019 05:56:06 GMT):
Has joined the channel.

salman4blochchain (Fri, 29 Mar 2019 06:18:54 GMT):
Has joined the channel.

JakeZheng (Fri, 29 Mar 2019 07:21:23 GMT):
Who can save my day!!! Indy-IOS-SDK missing 1.8.x version on private repo.....

lukasA (Fri, 29 Mar 2019 09:45:30 GMT):
Has joined the channel.

Rchauhan (Fri, 29 Mar 2019 11:37:52 GMT):
Thanks @dbluhm got the issue thanks for your kind support :)

MoonLee (Fri, 29 Mar 2019 14:26:10 GMT):
Has joined the channel.

silviu (Fri, 29 Mar 2019 19:51:07 GMT):
Hi everyone. I'm trying to build a concept for federated identity on blockchain. would Hyperledger Indy be the right one to do that or should I use another Hyperledger framework ? Any suggestions? Thank you

AnnaJ (Fri, 29 Mar 2019 19:58:57 GMT):
Has joined the channel.

esplinr (Sat, 30 Mar 2019 02:48:58 GMT):
@silviu That sounds like Indy's reason for existence.

esplinr (Sat, 30 Mar 2019 02:50:29 GMT):
Everyone: With the patient assistance of @rjones and @burdettadam , we streamlined the workflow for the Indy projects: Indy Node (INDY), Indy SDK (IS), and Indy Agent (IA).

rjones (Sat, 30 Mar 2019 02:50:30 GMT):
Has joined the channel.

esplinr (Sat, 30 Mar 2019 02:51:00 GMT):
They all had strange workflows that looked like at one time they were based on an internal Evernym Jira workflow that was poorly thought out.

esplinr (Sat, 30 Mar 2019 02:51:08 GMT):
Now they all share a common workflow.

esplinr (Sat, 30 Mar 2019 02:52:18 GMT):

Screenshot_20190329_205155.png

esplinr (Sat, 30 Mar 2019 04:26:20 GMT):
I wrote up an explanation of how it works here:

esplinr (Sat, 30 Mar 2019 04:26:20 GMT):
I wrote up an explanation of how it works here: https://wiki.hyperledger.org/display/indy/How+to+Contribute#HowtoContribute-CollaboratinginJira

esplinr (Sat, 30 Mar 2019 04:26:41 GMT):
Let me know if you have feedback.

kdenhartog (Sat, 30 Mar 2019 06:55:41 GMT):
@JakeZheng I don't think anyone has it if its not available in the repo.sovrin.org link where artifacts are posted to. My recommendation is to follow the guidelines included in the iOS build guide and self build it until the pipeline can include these new artifacts.

shenoy (Sat, 30 Mar 2019 12:58:08 GMT):
@Artemkaaas I am trying to follow the "How to Install" subsection in the Readme.md at https://github.com/hyperledger/indy-sdk/tree/master/wrappers/java and Maven throws me this Maven Dependency Problem error "Missing artifact org.hyperledger:indy:jar:1.8.1-dev-985" . I tried going through the Nexus repo at https://repo.sovrin.org/repository/maven-public/ as also searched org.hyperledger at https://mvnrepository.com/ and could not find anything that helped. Any idea how I can get my hands on the artifact org.hyperledger:indy:jar:1.8.1-dev-985?

silviu (Sat, 30 Mar 2019 21:35:53 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=BT7CGvSQD9gmxtxxY) @esplinr what i understand from what you're saying, is that self-sovereign identity is kind of federated identity on the blockchain. have i got this right ?

silviu (Sat, 30 Mar 2019 21:46:02 GMT):
and another question: what about the identity data, using indy, would it be stored on the blockchain or on the blockchain would be just the the metadata and the real identity information is stored in a wallet ? if so where can identity data usually besides wallets be stored on? thanks

shenoy (Sun, 31 Mar 2019 07:05:21 GMT):
@nage , @mboyd I am trying to follow the "How to Install" subsection in the Readme.md at https://github.com/hyperledger/indy-sdk/tree/master/wrappers/java and Maven throws me this Maven Dependency Problem error "Missing artifact org.hyperledger:indy:jar:1.8.1-dev-985" . I tried going through the Nexus repo at https://repo.sovrin.org/repository/maven-public/ as also searched org.hyperledger at https://mvnrepository.com/ and could not find anything that helped. Any idea how I can get my hands on the artifact org.hyperledger:indy:jar:1.8.1-dev-985?

Henrycoffin (Sun, 31 Mar 2019 09:28:37 GMT):
Has joined the channel.

Artemkaaas (Sun, 31 Mar 2019 09:59:21 GMT):
@shenoy look at pom.xml here: https://github.com/hyperledger/indy-sdk/blob/master/samples/java/pom.xml

shenoy (Sun, 31 Mar 2019 12:26:18 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=b4Cw5fYkrdzzo7XQK) @Artemkaaas @Artemkaaas Short reply: Thanks! It works! . Longer version: I switched the version to 1039 as in the pom.xml you pointed to and it worked. Thinking that version 985 might not exist, I reverted back to 985 and that works as well!! :astonished: I am trying to trace back my steps to see what I was doing wrong and I have no clue. Must be something in the pom for sure. Thanks for your help

Zohaib_Sohail (Mon, 01 Apr 2019 06:27:38 GMT):
Has joined the channel.

Zohaib_Sohail (Mon, 01 Apr 2019 07:06:10 GMT):
I am not getting the concept of indy-agents. In real app, every instance should have a indy-agent installed? if yes then it can only be done with desktop application because a web browser would not be able to handle all functionalities of indy-agent

Zohaib_Sohail (Mon, 01 Apr 2019 07:25:10 GMT):
Also, want some info about Agent providers and ready-to-use agent offerings?

carl.diclementi (Mon, 01 Apr 2019 14:44:54 GMT):
Has joined the channel.

carl.diclementi (Mon, 01 Apr 2019 14:46:24 GMT):
hey folks, noticed that the template.py link on the python documentation pages (such as https://hyperledger-indy.readthedocs.io/projects/sdk/en/latest/docs/how-tos/save-schema-and-cred-def/python/README.html) is a dead link. FYI!

mwherman2000 (Mon, 01 Apr 2019 18:35:47 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=ZDrWthKCKTEdED47S) @kclinux @kclinux @Rchauhan Checkout the following (it's back on the top of my stack to be completed)... *HyperLedger Indy Getting Started Guide for Enterprise Architects and Developers (INDY GSG-EA)* https://github.com/mwherman2000/indy-gsg-ea/blob/master/python/doc/getting_started-enterpise.md#hyperledger-indy-getting-started-guide-for-enterprise-architects-and-developers-indy-gsg-ea The HyperLedger Indy Getting Started Guide for Enterprise Architects and Developers (INDY GSG-EA) documents an end-to-end framework for analysing a business problem such as the Alice Buys a Car user story; then undertaking the design and implementation of an executable self-sovereign identity (SSI) solution that meets the requirements of this (or any similar purchasing) user story. This guide also introduces the use of several enterprise architecture concepts into the new world of SSI application analysis, design, and implementation. To achieve this goal, the guide uses the ArchiMate visual modeling language standard and the Archi open source, enterprise modeling tool for analysis and design. The implementation is a simple Python script. Architects and developers who are new to the HyperLedger Indy SSI software platform (Indy platform) and the Sovrin SSI governance framework (Sovrin framework) will gain significant new knowledge and understanding about the design and implementation of SSI solutions using the approach documented in this guide.

ZichengWang (Mon, 01 Apr 2019 20:46:01 GMT):
Has joined the channel.

mbanerjee (Mon, 01 Apr 2019 22:01:25 GMT):
Has joined the channel.

kclinux (Tue, 02 Apr 2019 03:15:45 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=B6t3frvAyWWK8XMKF) @mwherman2000 @mwherman2000 Thanks for the effort. I am happy to see that the overall flow is what I depict in my article (except the Red Lamborghini). I personally think these flow diagram is more for expert and professional. And that's why I come up with my work for explanation to who's new to the topic (me as well).

PooyanMehrparvar (Tue, 02 Apr 2019 18:32:46 GMT):
Has joined the channel.

VS09 1 (Wed, 03 Apr 2019 06:34:46 GMT):
Has joined the channel.

Sean_Bohan (Thu, 04 Apr 2019 01:16:24 GMT):
TOMORROW 04/052019 - Indy WG CALL!!!! Agenda for Thursday's Working Group call: ANVIL demo/discussion with the team from Outlier Ventures Check out the ANVIL launch blog which gives an overview (https://outlierventures.io/research/anvil-a-fetch-ai-and-sovrin-bridge/), and here is the recording + info of the webinar from last week: https://outlierventures.io/research/the-convergence-stack-webinar-featuring-anvil/ Where: https://zoom.us/j/232861185 When: 8am PT, 9am MT, 10am CT, 11amET, 4pmGMT

Karlos (Thu, 04 Apr 2019 09:56:08 GMT):
Hi all, I'm having problem to do validate proof. It constantly returns `false` and I don't know why. Do you know were can I find libindy log files? I'm using Ubuntu 16.04

Karlos (Thu, 04 Apr 2019 09:56:08 GMT):
Hi all, I'm having problems to validate generated proofs. It constantly returns `false` and I don't know why. Do you know were can I find libindy log files? I'm using Ubuntu 16.04

justinr3 (Thu, 04 Apr 2019 13:24:54 GMT):
Hi everyone, I'm having an issue with the void callback used for the Java wrapper in version Indy 1.7.0. It doesn't seem to pass check_useful_c_callback!(cb, ErrorCode::CommonInvalidParam4);

justinr3 (Thu, 04 Apr 2019 13:27:14 GMT):
And the end result is that I'm getting the InvalidParam4 exception, as expected if the callback didn't pass the check. From looking at the code, there really isn't anything I can do externally to prevent this error. I'd appreciate some advice!

justinr3 (Thu, 04 Apr 2019 13:43:16 GMT):
For reference, 1.5.0 seemed to work just fine for me. The difference I see is that 1.5.0 uses checkResult(result), while 1.7.0 uses checkResult(future, result) in the Java createWallet wrapper. The voidCb is exactly the same between versions.

pknowles (Thu, 04 Apr 2019 16:01:59 GMT):
Just jumped onto the *HL Indy WG* call. Outlier Ventures posted the GitHub links for *ANVIL* ... https://github.com/OutlierVentures/Sovrin-Springboard + https://github.com/OutlierVentures/ANVIL

aronvanammers (Thu, 04 Apr 2019 19:12:37 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=DwA5dj1XfNthez1aNO) @Sean_Bohan Hi @Sean_Bohan and attendees of the WG call, thanks again for hosting us today. It was a real pleasure and I hope you got something useful out of it. Feel free to reach out to @theoturner or myself. As for the GitHub links, what @pknowles said :point_up_2: :)

esplinr (Thu, 04 Apr 2019 19:40:47 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=kjqBinc2mHWyRCk6E) @silviu Sorry, I took a few days out of the office. Self sovereign identity means that the user is in control of their identity attributes. It can be used to meet federated identity use cases. In general, you don't want identity information on an immutable ledger because it will erode privacy. Users should store their own identity information which they exchange in a peer-to-peer fashion, and use the blockchain as a root of trust.

silviu (Thu, 04 Apr 2019 20:20:04 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=EPhHqqFqPfDJZ4z4P) @esplinr thank you very much for the answer. really helpful. Let's say, there are 2-3 companies, that would like to use a federated identity management based on the blockchain. But they want for example, to delete one employees identity from the system, when the work contract is terminated. Could this be done with SSI ? who hast the control in such a situation ?

esplinr (Thu, 04 Apr 2019 20:22:23 GMT):
The employer issues a credential to the employee saying that they are employed. The employee can share that credential with anyone they want, who can verify that it is valid using the employer's keys on the ledger. Other companies can use that credential to automate entering the employee into their vendor management systems.

esplinr (Thu, 04 Apr 2019 20:23:07 GMT):
The employer can revoke the credential. At this point, the employee can still share the credential, but the people who receive it will see that it is revoked when they try to validate it. They then know to remove the employee from their systems.

esplinr (Thu, 04 Apr 2019 20:23:45 GMT):
The Sovrin network doesn't currently solve the problem of notifying everyone that the credential was revoked.

esplinr (Thu, 04 Apr 2019 20:23:54 GMT):
That's a privacy concern.

esplinr (Thu, 04 Apr 2019 20:24:29 GMT):
Though businesses could use Sovrin to establish a trusted relationship with the employer and use that communication channel to receive the notification.

esplinr (Thu, 04 Apr 2019 20:24:33 GMT):
Does that help @silviu ?

silviu (Thu, 04 Apr 2019 20:28:34 GMT):
@esplinr yes, that's really useful information and makes sense. are we talking about a federated identity management system based on Sovrin or hyperledger indy in general ?

esplinr (Thu, 04 Apr 2019 20:29:20 GMT):
Sovrin is a deployment of Hyperledger Indy. The code is the same. You could do the same thing with Indy, but then you have to recruit everyone into your walled network instead of using the public utility that Sovrin provides.

silviu (Thu, 04 Apr 2019 20:33:48 GMT):
that means that, i could deploy my own version of "sovrin" in a federated way using hyperledger indy, based on a private network and the service would be used just for a conglomerate of organizations and their employees, am i right?

esplinr (Thu, 04 Apr 2019 20:34:32 GMT):
Correct. Others are doing that. But over time, I expect you'll find it's more trouble than its worth. Using Sovrin provides the same privacy guarantees without the headache of running your own consensus network.

silviu (Thu, 04 Apr 2019 20:35:47 GMT):
how would be access management managed in such a network? Smart contracts or it there a better way ?

silviu (Thu, 04 Apr 2019 20:35:47 GMT):
how would be access management be done in such a network? Smart contracts or it there a better way ?

silviu (Thu, 04 Apr 2019 20:35:47 GMT):
how would be access management be done in such a network? Smart contracts or is there a better way ?

esplinr (Thu, 04 Apr 2019 20:37:21 GMT):
Access to what? The employer only issues credentials to the people who should have them. The credential is provided by the employee to whomever they want. There is no global database to restrict access to.

silviu (Thu, 04 Apr 2019 20:38:41 GMT):
access to diverse services, or even internal resources for example

esplinr (Thu, 04 Apr 2019 20:39:21 GMT):
Sovrin doesn't manage internal resources. It manages the credentials the user uses to prove that they are authorized to access the resources. Once the credential is revoked, they lose access.

silviu (Thu, 04 Apr 2019 20:39:22 GMT):
let's say a university would want to make difference between professors and students and the services they want to access

silviu (Thu, 04 Apr 2019 20:39:22 GMT):
let's say a university would want to make difference between professors and students and the services they are allowed to access

esplinr (Thu, 04 Apr 2019 20:40:04 GMT):
It flips the access management model on its head. The user is in control of the information that they share.

silviu (Thu, 04 Apr 2019 20:40:52 GMT):
ok, i find this complicated and i'm not sure yet how access management be implemented

silviu (Thu, 04 Apr 2019 20:40:52 GMT):
ok, i find this complicated and i'm not sure yet how access management could be implemented

esplinr (Thu, 04 Apr 2019 20:41:15 GMT):
You grant access based on the user being able to provide a credential saying they should have access.

esplinr (Thu, 04 Apr 2019 20:41:23 GMT):
They have to provide the credential every time they want access.

esplinr (Thu, 04 Apr 2019 20:41:28 GMT):
If the credential is revoked, you don't provide access.

esplinr (Thu, 04 Apr 2019 20:42:14 GMT):
There is no central database to reference. The proof of the credential that the user provides has everything necessary to show it is valid, so long as the root keys are available on the global ledger.

silviu (Thu, 04 Apr 2019 20:44:54 GMT):
how are the grants managed ? does it need another service to take care of that ? because if the identity is valid, is doesn't mean a user can access all the services

silviu (Thu, 04 Apr 2019 20:45:05 GMT):
can access be granular managed ?

esplinr (Thu, 04 Apr 2019 20:45:41 GMT):
The person providing the service will need a system that checks the credential and decides what to do with that information. That is beyond the scope of Sovrin.

esplinr (Thu, 04 Apr 2019 20:46:33 GMT):
Sovrin solves the specific problem of proving something about an individual, organization, or thing in a decentralized manner. It doesn't have much to say about what you do with that information.

silviu (Thu, 04 Apr 2019 20:48:22 GMT):
ok, that's clear now. Any idea, if there's a blockchain based system, that could take over access management ?

esplinr (Thu, 04 Apr 2019 20:49:27 GMT):
There are efforts underway to implement DIDAuth and a DID-OpenID integration that could be used for this use case.

esplinr (Thu, 04 Apr 2019 20:50:39 GMT):
The key idea is that you don't want a blockchain handling access management. You want individuals to present the credentials that you then use to manage the access to your system. That provides flexibility and separation of concerns.

esplinr (Thu, 04 Apr 2019 20:50:55 GMT):
That is what we mean when we use terms like self-sovereignty and decentralization.

silviu (Thu, 04 Apr 2019 20:52:44 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=wNhKk8Qw9LThyeAm3) @esplinr is there any documentation on this matter ?

esplinr (Thu, 04 Apr 2019 20:53:20 GMT):
https://ssimeetup.org/introduction-did-auth-markus-sabadello-webinar-10/

silviu (Thu, 04 Apr 2019 20:55:56 GMT):
thank you very much for your time. everything is much clearer now. it helps a lot. it was worth the wait. if i'll have questions on the way , i'll let you know. thanks again

silviu (Thu, 04 Apr 2019 20:55:56 GMT):
thank you very much for your time. everything is much clearer now. it helps a lot. it was worth the wait. if i'll have questions on the way , i'll let you know. thanks again @esplinr

bart.cant@gmail.com (Thu, 04 Apr 2019 22:11:15 GMT):
Has joined the channel.

vikas.kundz (Fri, 05 Apr 2019 03:27:04 GMT):
Has joined the channel.

pknowles (Fri, 05 Apr 2019 06:04:51 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=NzRmay4cJKocHwqrE) @mtfk @mwherman2000 Identitiers for semantic objects. I know that @mtfk wants to approach this subject again with you. I'll also add this topic as an agenda item for the next *Indy Semantics WG* call on April 16th.

pknowles (Fri, 05 Apr 2019 06:04:51 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=NzRmay4cJKocHwqrE) @mwherman2000 Identitiers for semantic objects. I know that @mtfk wants to approach this subject again with you. I'll also add this topic as an agenda item for the next *Indy Semantics WG* call on April 16th

pknowles (Fri, 05 Apr 2019 06:08:57 GMT):
To avoid confusion with naming I started experimenting with the schema for the URN like: ``` did-schema:sov:1231415124151 did-overlay:sov:13215512312 did-schema-element:sov:5814108410924 ``` ` But I think everyone would agree that this is not the way to go. Personally I like the concept of reusing main @context in DID document to tell resolver what type of the content it is. Similar to the idea with MIME Type in the document resolved via HTTP/S So as soon as the I will try to resolve `did:sov:8291239123` I check the @context of the document to figure out with what I have to deal and for example it could be: ``` @context: ["https://domain.com/language/v1"] @context: ["https://domain.com/control/v1"] @context: ["https://domain.com/..../v1"] ``` I can't remember what other type of units we were talking about during the conf but those two are the most needed right now control is relation between did and public key, where language is did and data schema/language - something which describe how to talk not who controls it.

pknowles (Fri, 05 Apr 2019 06:08:57 GMT):
Posted by @mtfk back in December: "To avoid confusion with naming I started experimenting with the schema for the URN like: ``` did-schema:sov:1231415124151 did-overlay:sov:13215512312 did-schema-element:sov:5814108410924 ``` ` But I think everyone would agree that this is not the way to go. Personally I like the concept of reusing main @context in DID document to tell resolver what type of the content it is. Similar to the idea with MIME Type in the document resolved via HTTP/S So as soon as the I will try to resolve `did:sov:8291239123` I check the @context of the document to figure out with what I have to deal and for example it could be: ``` @context: ["https://domain.com/language/v1"] @context: ["https://domain.com/control/v1"] @context: ["https://domain.com/..../v1"] ``` I can't remember what other type of units we were talking about during the conf but those two are the most needed right now control is relation between did and public key, where language is did and data schema/language - something which describe how to talk not who controls it."

BhaskarRaju (Fri, 05 Apr 2019 08:28:57 GMT):
Has joined the channel.

DonClaude (Sat, 06 Apr 2019 11:42:27 GMT):
regasm

drummondreed (Sun, 07 Apr 2019 02:02:02 GMT):
@pknowles Paul, on this whole subject of addressing immutable content objects (like schema definitions) with DIDs, see the current state-of-play about how this will be done in the DID spec in these two Google docs. The first one describes the matrix parameters proposal: https://docs.google.com/document/d/1TctFY8euBH2wq7Z8c9KccICDZUGZplvhoqlHlFMahGk/edit?usp=sharing

drummondreed (Sun, 07 Apr 2019 02:03:59 GMT):
The second one is for contributors to write up their own use cases for DID URLs in the 3 syntaxes under consideration. Should only take any W3C Credentials Community Group member 5 mins to write up your own examples: https://docs.google.com/document/d/1VpYPvUw2o-01e727bCy2V-0MUEe0_NB-EqX677scQbI/edit?usp=sharing

pknowles (Sun, 07 Apr 2019 06:25:07 GMT):
Great stuff. Thanks, @drummondreed . I'll add this as an agenda item for the next *Indy Semantics WG* call. [Cc: @mtfk @mwherman2000 ]

pknowles (Sun, 07 Apr 2019 06:25:07 GMT):
Great stuff. Thanks, @drummondreed . I'll add this as an agenda item for the next #indy-semantics WG call. [Cc: @mtfk @mwherman2000 ]

pknowles (Sun, 07 Apr 2019 06:25:07 GMT):
Great stuff. Thx, @drummondreed . I'll add this as an agenda item for the next #indy-semantics WG call. [Cc: @mtfk @mwherman2000 ]

pknowles (Sun, 07 Apr 2019 06:25:07 GMT):
Great stuff. Thx, @drummondreed . I'll add this as an agenda item for the next #indy-semantics *WG* call. [Cc: @mtfk @mwherman2000 ]

AxelNennker (Sun, 07 Apr 2019 11:44:42 GMT):
It seems that https://build.sovrin.org/ is down. CI is not running.

AxelNennker (Sun, 07 Apr 2019 11:44:42 GMT):
I seems that https://build.sovrin.org/ is down. CI is not running.

tech9beta (Mon, 08 Apr 2019 05:24:09 GMT):
Has joined the channel.

Zohaib_Sohail (Mon, 08 Apr 2019 10:34:58 GMT):
what is the concept of master_secret or link_secret ? I am not getting this. Kindly help???

Andersonsmith (Mon, 08 Apr 2019 12:22:17 GMT):
Has joined the channel.

Andersonsmith (Mon, 08 Apr 2019 12:23:08 GMT):
How to fix `indy> pool connect sandbox protocol-version=2 timeout=60 Pool "sandbox" has not been connected.`

Zohaib_Sohail (Mon, 08 Apr 2019 15:03:41 GMT):
I want to code Alice, Faber and Acme scenario from scratch using NodeJS. From where should I start any suggestions/help would be appreciated?

swcurran (Mon, 08 Apr 2019 16:13:50 GMT):
You can look at the nodejs agent code that is in the indy-agent repo in the nodejs folder.

silviu (Mon, 08 Apr 2019 20:51:55 GMT):
Hello. i have a question to indy in combination with fabric. would it be possible to use the same blockchain for both of them ? let's say indy takes care of the authentication and fabric would take care of the authorization using smart contracts?

esplinr (Mon, 08 Apr 2019 21:33:36 GMT):
By "same blockchain" you mean the same distributed ledger?

esplinr (Mon, 08 Apr 2019 21:33:42 GMT):
That would be very hard with the current Indy technology.

esplinr (Mon, 08 Apr 2019 21:34:53 GMT):
But there are solutions that rely on running both as you said, Indy for tracking identities and Fabric or Burrow for the smart contracts.

tplooker (Tue, 09 Apr 2019 01:26:26 GMT):
Has joined the channel.

tplooker (Tue, 09 Apr 2019 01:27:24 GMT):
Hey guys we have a CI pipeline that rebuilds a custom node pool image for our agency and this morning we started getting the following failure ``` The following packages have unmet dependencies: indy-cli : Depends: libindy (= 1.8.2~1050) but 1.8.2~1052 is to be installed E: Unable to correct problems, you have held broken packages. ```

tplooker (Tue, 09 Apr 2019 01:27:48 GMT):
We inject indy-cli into the docker node pool image and it appears theres a dependency problem?

silviu (Tue, 09 Apr 2019 07:12:50 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=dKmLMap4pxjFdoqnt) @esplinr yes, that's what is meant. You mean, that there should be 2 different distributed ledgers, one for identity with indy and one for authorization with fabric or burrow.

silviu (Tue, 09 Apr 2019 07:12:50 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=dKmLMap4pxjFdoqnt) @esplinr yes, that's what is meant. You mean, that there should be 2 different distributed ledgers, one for identity with indy and one for authorization with fabric or burrow, right ?

silviu (Tue, 09 Apr 2019 07:13:12 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=ypEN277tpixRhbZgi) @esplinr can you name or or two please ?

wip-abramson (Tue, 09 Apr 2019 09:21:44 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=X6yThm5fJYR4bypW6) @Zohaib_Sohail Hi @Zohaib_Sohail , a master secret is a secret piece of information that a credential holder contributes to all their credentials. This gets signed by the issuer using a CL signature currently without them learning the value. The holder can then use knowledge of this secret to prove the credential was issued to them. I had a go a writing this up a while back - https://blog.goodaudience.com/cl-signatures-for-anonymous-credentials-93980f720d99 (may not be 100% accurate but its pretty close)

Zohaib_Sohail (Tue, 09 Apr 2019 11:44:44 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=e6gZFoQ856uD2jiMb) @wip-abramson Thanks i got the idea

lukasA (Tue, 09 Apr 2019 12:38:19 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=jEty8nQoTYTnjPbXL) i've got the same problem as @tplooker when trying to build this dockerfile: https://github.com/hyperledger/indy-sdk/blob/master/cli/cli.dockerfile

Karlos (Tue, 09 Apr 2019 13:15:20 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=sqfZ5EP38BTCqtC2M) It turns out I was using self attested attributes in the proof and that caused the problem with the validation. Are self attested attributes supported in the validation? If they are, is there something I have to do with them before validating? like store them with an ATTRIB transaction

VicCooper (Wed, 10 Apr 2019 01:07:26 GMT):
Has joined the channel.

pmaruindy (Wed, 10 Apr 2019 02:46:03 GMT):
Has joined the channel.

pranavkirtani (Wed, 10 Apr 2019 03:55:16 GMT):
Has joined the channel.

pranavkirtani (Wed, 10 Apr 2019 03:55:33 GMT):
is there any documentation on settting up indy-sdk on windows 10?

pranavkirtani (Wed, 10 Apr 2019 03:56:08 GMT):
I see the following error src\indy.cc(1133): error C3861: 'indy_pack_message': identifier not found [D:\indy-sdk\indy-sdk\samples\nodejs\node_ modules\indy-sdk\build\indynodejs.vcxproj]

nicola.attico (Wed, 10 Apr 2019 06:59:11 GMT):
Has joined the channel.

nicola.attico (Wed, 10 Apr 2019 07:04:24 GMT):
Hi, anyone can point me to the best tutorial to set up a dev environment either on docker or public cloud? I tried many but I always get stuck in the process...

agrawalamit2005 (Wed, 10 Apr 2019 07:09:34 GMT):
Has joined the channel.

agrawalamit2005 (Wed, 10 Apr 2019 07:10:09 GMT):
With indy SDK, is it supported to talk different blockchain network other than sovrin?

Zohaib_Sohail (Wed, 10 Apr 2019 07:13:32 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=z9S9GQypJaWDYvQNQ) @nicola.attico https://medium.com/@kctheservant/exploring-hyperledger-indy-through-indy-dev-example-10075d2547ae it may help

Zohaib_Sohail (Wed, 10 Apr 2019 08:19:44 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=CS8fPgojmF4LPPP3d) @swcurran basically right now i am looking at Hyperledger education and training material indy nodejs code. I want to ask that there are many "FIXME" comments in the code, what was the reason why the issues were not resolved?

andrej-zirko (Wed, 10 Apr 2019 14:10:00 GMT):
Has joined the channel.

swcurran (Wed, 10 Apr 2019 14:27:01 GMT):
@Zohaib_Sohail - the education repo is a snapshot of the indy-agent nodejs that was created for the course. So the indy-agent repo is where you want to look. In my opinion, where it needs to go is to along the lines of the indy-agent python implementation that is able to pass the tests in the indy-agent test suite. There are a couple of people that have contributed to the nodejs implementation that might be able to give you a bit of guidance.

Zohaib_Sohail (Wed, 10 Apr 2019 15:05:35 GMT):
Another question, in real world example institute that issue credentials would have its own validator node running or just it would send the transaction to any random validator node?

Zohaib_Sohail (Wed, 10 Apr 2019 15:05:50 GMT):
and what's the role of Trust Anchor?

hpduong (Wed, 10 Apr 2019 16:12:51 GMT):
Has joined the channel.

hpduong (Wed, 10 Apr 2019 16:12:56 GMT):
Hi all, I've been playing around with Indy, but how can upgrade to a new version of Hyperledger Indy without losing my data when restarting the network?

hpduong (Wed, 10 Apr 2019 16:12:56 GMT):
Hi all, I've been playing around with Indy, but how can I upgrade to a new version of Hyperledger Indy without losing my data when restarting the network? upgrade to a new version of Hyperledger Indy without losing my data when restarting the network?

hpduong (Wed, 10 Apr 2019 16:12:56 GMT):
Hi all, I've been playing around with Indy, but how can I upgrade to a new version of Hyperledger Indy without losing my data when restarting the network?

anikitinDSR (Wed, 10 Apr 2019 16:16:59 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=qTzo3zkwFG3AW36wS) @hpduong You can send a POOL_UPGRADE transaction as described in: https://github.com/hyperledger/indy-node/blob/master/docs/source/pool-upgrade.md Format of this txn you can see in this doc: https://github.com/hyperledger/indy-node/blob/master/docs/source/transactions.md#pool_upgrade

hpduong (Wed, 10 Apr 2019 16:18:41 GMT):
Thanks @anikitinDSR . It's a bit confusing because the keys are in tact, but the transactional data is gone.

daidoji (Wed, 10 Apr 2019 16:22:09 GMT):
@hpduong I believe that orgs might want to run their own validator nodes, but they could also just contract with a validator if they need that type of service. I think the Trust Anchor is just similar to a root_certificate holder in TLS system. Just a starting point (if you want it) to establish a chain of trust for your proof. Maybe others who are more informed can qualify that answer but that's the one I believe I've run into before.

daidoji (Wed, 10 Apr 2019 16:23:04 GMT):
there's also this forum posting https://forum.sovrin.org/t/on-trust-anchors/153

anikitinDSR (Wed, 10 Apr 2019 16:28:27 GMT):
@hpduong well, if you are worrying about your data, then you can backup '/var/lib/indy' directory. But, during upgrade procedure, some of data structure can be changed and corresponding migration scripts will be executed automatically. The steps like "backup data and reinstall indy-node to the latest version" does not guarantee ledger's compatibility. It's very important point if you want to upgrade from the very old version to latest. Also, upgrade procedure will make a backup of data and will restore it if something going wrong

swcurran (Wed, 10 Apr 2019 16:35:06 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=ahBPhSnkTjufhvY2c) @Zohaib_Sohail Once we get to steady state with a global implementation (e.g. Sovrin), the vast majority of issuers will not be validators. Stewards (as they are called in Sovrin) must go through an onboarding/certification process, so an issuer can't just one day say "I want to be a validator". It's basically a separate business question. The term "trust anchor" is being phased out in favour of "Transaction Endorser"). It basically means a DID that has permissioned write access to the ledger. See the V2 Sovrin Glossary - https://docs.google.com/document/d/1gfIz5TT0cNp2kxGMLFXr19x1uoZsruUe_0glHst2fZ8/edit#heading=h.i4y80elx0tpb

swcurran (Wed, 10 Apr 2019 16:35:06 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=ahBPhSnkTjufhvY2c) @Zohaib_Sohail Once we get to steady state with a global implementation (e.g. Sovrin), the vast majority of issuers will not be validators. Stewards (as they are called in Sovrin) must go through an onboarding/certification process, so an issuer can't just one day say "I want to be a validator". It's basically a separate business question. The term "trust anchor" is being phased out (in favour of "Transaction Endorser"). It basically means a DID that has permissioned write access to the ledger. See the V2 Sovrin Glossary - https://docs.google.com/document/d/1gfIz5TT0cNp2kxGMLFXr19x1uoZsruUe_0glHst2fZ8/edit#heading=h.i4y80elx0tpb

PatrikStas (Wed, 10 Apr 2019 19:02:59 GMT):
I have a fun new hobby nowadays to track number of stars on IndySDK github repo and it's growing steadily. To make things more fun, plotted together with a few uport repos I've found ^_^ https://star-history.t9t.io/#hyperledger/indy-sdk&uport-project/uport-credentials&uport-project/uport-connect&uport-project/ethr-did&uport-project/ethr-did-registry

mboyd (Wed, 10 Apr 2019 22:26:24 GMT):
cli

esplinr (Wed, 10 Apr 2019 23:18:13 GMT):
@PatrikStas This is great! Thank you for sharing.

aamdevan (Wed, 10 Apr 2019 23:55:34 GMT):
Has joined the channel.

aamdevan (Wed, 10 Apr 2019 23:56:45 GMT):
install

nage (Thu, 11 Apr 2019 04:49:32 GMT):
@all we are looking to name the new project proposal to split off the agent protocol (wallet, messaging and credentials exchange) project that has been incubated within Indy into its own project to help support standards compliance and protocol interoperability more broadly than just Indy. If you would like to help decide on the name, please respond to this survey sometime within the next 30 hours. https://forms.gle/5s879DB2Kp37DZYe7 I expect we will start discussing results here and in #indy-maintainers sometime Friday afternoon (US time) once most folks have had a chance to respond.

nage (Thu, 11 Apr 2019 14:39:30 GMT):
A few reminders for the poll: question 1 has an other option for suggestions, question 2 is for objections (it is harder to find new name suggestions buried in negative feedback). Also, please vote for *all* the options you can support, this poll doesn't have ranked voting but I am watching for names that have general acceptance and a second choice vote will help push the best names out in front. Finally, you should be able to edit your responses, as I review new suggested names those will be added to the poll, so feel free to come back and add checks to names you like or add objections to the new suggestions.

nage (Thu, 11 Apr 2019 14:41:54 GMT):
As you vet names, it is also a good idea to think about the repo and resume mechanics: "Review my PR on ______-sdk-python", "I am the maintainer of the ______-peer-resolver repository", "Google the ______ getting started guide...."

mauricio (Thu, 11 Apr 2019 15:08:16 GMT):
Has joined the channel.

Karlos (Thu, 11 Apr 2019 15:50:56 GMT):
Hi All! I'm trying to add an attribute to a NYM record, in this case to store the date when a credential was last issued to the DID. The problem is that I'm trying to add it with a Trust Anchor, as described in https://hyperledger-indy.readthedocs.io/projects/node/en/latest/transactions.html#attrib, but I'm constantly getting an error message: `client request invalid: UnauthorizedClientRequest('Only identity owner/guardian can add attribute for that identity',)` Do you know if this is supported?

esplinr (Thu, 11 Apr 2019 16:03:19 GMT):
Evernym's bi-weekly update on our efforts: https://www.youtube.com/watch?v=bcBeKilkink This time it's mostly about our work on Sovrin-specific functionality.

esplinr (Thu, 11 Apr 2019 16:03:19 GMT):
Evernym's bi-weekly update on our efforts: https://www.youtube.com/watch?v=bcBeKilkink&list=PLRp0viTDxBWGLdZk0aamtahB9cpJGV7ZF This time it's mostly about our work on Sovrin-specific functionality.

brentzundel (Thu, 11 Apr 2019 16:20:58 GMT):
Has left the channel.

Vanitha (Thu, 11 Apr 2019 16:59:53 GMT):
Has joined the channel.

twshelton (Thu, 11 Apr 2019 20:18:33 GMT):
Hi All! I'm updating an application and noticed a change here: vcx::api::issuer_credential::vcx_issuer_create_credential where previously a cred_def_id was needed to make the call and now a cred_def_handle is requested. I've looked through the API and am having trouble finding a way to retrieve the cred_def_handle of a given cred_def_id. Any thoughts on best way to go about this?

mgbailey (Thu, 11 Apr 2019 21:22:24 GMT):
@twshelton you get the handle by using the .handle method on the object. When you need to save and reuse it, serialize the object and save that. Then to retrieve it and get the handle, deserialize the saved serialized object, then get the handle of the restored object.

anillewis (Fri, 12 Apr 2019 14:23:06 GMT):
Team....I have a question on this agent policy register as described in https://sovrin.org/wp-content/uploads/2019/01/How-DIDs-Keys-Credentials-and-Agents-Work-Together-in-Sovrin-131118.pdf ..I am struggling to identify the indy-sdk api call for this....can someone please assist in identifying this?

justinr3 (Fri, 12 Apr 2019 14:39:07 GMT):
Hi everyone, I'm still having an issue with the void callback used for the Java wrapper in version Indy 1.7.0. It doesn't seem to pass check_useful_c_callback!(cb, ErrorCode::CommonInvalidParam4). Could this be Java version mismatch related issue?

justinr3 (Fri, 12 Apr 2019 14:40:46 GMT):
The code related to this issue looks to be the same in 1.8.2 as well, so I'm just wondering if anyone else has run into this issue when using the Java wrapper specifically.

mwherman2000 (Fri, 12 Apr 2019 18:13:05 GMT):
@nage Suggested additions: Hyperledger Choreographer or Hyperledger Choreography

mwherman2000 (Fri, 12 Apr 2019 18:13:05 GMT):
@nage Project name suggested additions: Hyperledger Choreographer or Hyperledger Choreography

mwherman2000 (Fri, 12 Apr 2019 18:14:36 GMT):
Background: https://www.youtube.com/watch?v=IAozsIsrlbU&list=PLU-rWqHm5p45dzXF2LJZjuNVJrOUR6DaD&index=2&t=240s

mwherman2000 (Fri, 12 Apr 2019 18:15:31 GMT):
...watch for about 4.5 minutes.

kdenhartog (Fri, 12 Apr 2019 18:28:06 GMT):
@justinr3 id suggest asking this in the #indy-sdk channel. Thats where the maintainers of the Java wrappers most often are.

justinr3 (Fri, 12 Apr 2019 18:43:35 GMT):
@kdenhartog Got it, thanks!

sumitkumarmb (Fri, 12 Apr 2019 19:13:07 GMT):
Has joined the channel.

elbaloo (Fri, 12 Apr 2019 19:59:35 GMT):
Has joined the channel.

Ala_Tian (Sat, 13 Apr 2019 03:28:43 GMT):
Has joined the channel.

pranavkirtani (Sat, 13 Apr 2019 04:13:16 GMT):
what is the benefit of having credential schema and credential definition?

pranavkirtani (Sat, 13 Apr 2019 04:13:24 GMT):
why not just have schema

jljordan_bcgov (Sat, 13 Apr 2019 13:47:19 GMT):
A schema is general and only has the claims that it is comprised of. A schema can be used by any DID owner. A credential definition is an instance of a schema created by a particular DID owner and annotated with public key material per claim. It is the data object that is used in the issue and verify sequences. @pranavkirtani [ ](https://chat.hyperledger.org/channel/indy?msg=ihXzGGh72FbZpCfDX)

jljordan_bcgov (Sat, 13 Apr 2019 13:49:09 GMT):
If you go here https://sovrin-mainnet-browser.vonx.io/browse/domain and use the Type filter you can see schema or cred defs on the Sovrin Mainnet

ayushprd (Sat, 13 Apr 2019 18:00:02 GMT):
Has joined the channel.

nikitabairagi (Sun, 14 Apr 2019 01:51:20 GMT):
Has joined the channel.

pranavkirtani (Mon, 15 Apr 2019 06:18:38 GMT):
I found a react native implementation of indy-agent, is there any demo UI where I can try issuing and verification flows?

pranavkirtani (Mon, 15 Apr 2019 06:18:40 GMT):
https://github.com/sovrin-foundation/connector-app

MoonLee (Mon, 15 Apr 2019 08:03:44 GMT):
Hello everybody I am working on a project. Is it possible to build my own indy network and store video information in it?

mikail 2 (Mon, 15 Apr 2019 10:45:03 GMT):
Has joined the channel.

mgbailey (Mon, 15 Apr 2019 14:02:38 GMT):
@MoonLee The indy software is purpose-built to hold DIDs, schemas, credential definitions and similar self-sovereign objects. There would be much better options than this for storing video data.

MoonLee (Tue, 16 Apr 2019 04:40:04 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=L2kehLLSMYJnpnqYj) @mgbailey Thanks for your answer

ranjan008 (Tue, 16 Apr 2019 08:41:50 GMT):
Has joined the channel.

itarunachalam (Tue, 16 Apr 2019 08:48:54 GMT):
Has joined the channel.

JeganSelvaraj (Tue, 16 Apr 2019 08:52:33 GMT):
Has joined the channel.

nage (Tue, 16 Apr 2019 15:02:05 GMT):
An update on the naming poll and project proposal: Hyperledger Aries was the clear leader in the results (by 3x). Most of the next tier of results were vetoed by large amounts of negative feedback (Union and Connector in particular). I am working on updates to the project proposal doc. Watch for updates here and in #indy-maintainers for requests to review.

MALodder (Tue, 16 Apr 2019 15:22:03 GMT):
++ for Aries

rjones (Tue, 16 Apr 2019 15:59:35 GMT):
I voted for it thirty times!

npc0405 (Wed, 17 Apr 2019 10:21:31 GMT):
Has joined the channel.

ONRising (Wed, 17 Apr 2019 13:42:48 GMT):
Has joined the channel.

ONRising (Wed, 17 Apr 2019 14:44:23 GMT):
Hello everyone, For our master thesis, we are studying solutions to support account migration on the Mastodon social network (https://github.com/tootsuite/mastodon/issues/177). In order to achieve this, we thought about distributed ledger technologies to build decentralized, human-meaningful and secured identifiers which we think could be a major update to increase the popularity of decentralized social networks. During our researches, we found the Hyperledger Indy project and we are wondering if this project could fit our needs. We would want to allow people to have a unique ID on the network which is independent of the instance they are registered on. In this case (according to our understanding), only one DID is needed per user (or more if if want to use more than one ID) and we need to link a human-meaningful name to it. This DID will allow transfers of personal data between two instances while letting the following list of other people following you unchanged. We would also want to know if we could connect to one instance using Indy keys instead of a password. I hope this description of our project is understandable. We would want to know if it is feasible and what you think.

esplinr (Wed, 17 Apr 2019 20:11:18 GMT):
@ONRising One of the main goals of Indy is to avoid third party collection of your personal data by correlating identifiers, so we make it really easy to create lots of global identifiers for individual connections. That would undermine your use case.

esplinr (Wed, 17 Apr 2019 20:11:54 GMT):
I'm curious to know what other people think about this question.

lijamie (Wed, 17 Apr 2019 22:31:45 GMT):
Has joined the channel.

lijamie (Wed, 17 Apr 2019 22:33:18 GMT):
I got an exception at: https://github.com/hyperledger/indy-plenum/blob/master/plenum/server/catchup/node_leecher_service.py Line#182.

lijamie (Wed, 17 Apr 2019 22:33:56 GMT):
When ledger_id is passed in as "str", the ledger object is None. When I casted it to int(ledger_id), it works.

lijamie (Wed, 17 Apr 2019 22:34:11 GMT):
How to report the bug?

rjones (Wed, 17 Apr 2019 22:35:53 GMT):
@lijamie please file a JIRA bug https://jira.hyperledger.org/projects/IS/

lijamie (Wed, 17 Apr 2019 22:38:01 GMT):
Don't see a place to signup Hyperledger JIRA.

esplinr (Wed, 17 Apr 2019 22:38:48 GMT):
@lijamie https://identity.linuxfoundation.org/

esplinr (Wed, 17 Apr 2019 22:38:59 GMT):
Click on "I need to create a Linux Foundation ID"

esplinr (Wed, 17 Apr 2019 22:39:18 GMT):
Also, check out https://wiki.hyperledger.org/display/indy/How+to+Contribute#HowtoContribute-ReportinganIssue

rjones (Wed, 17 Apr 2019 22:40:36 GMT):
@esplinr they have one :)

rjones (Wed, 17 Apr 2019 22:40:51 GMT):
@lijamie use the same account you used to log on here, to log in to JIRA. :)

esplinr (Wed, 17 Apr 2019 22:41:00 GMT):
_slaps forehead_

esplinr (Wed, 17 Apr 2019 22:41:01 GMT):
Of course

rjones (Wed, 17 Apr 2019 22:41:39 GMT):
the real hurdle is for people that create an account with a social login. that's when trouble starts

lijamie (Wed, 17 Apr 2019 22:44:50 GMT):
@rjones Can you help me Logged in now.

lijamie (Wed, 17 Apr 2019 22:45:07 GMT):
Wront message. Sorry.

lijamie (Wed, 17 Apr 2019 22:45:10 GMT):
I can login now.

rjones (Wed, 17 Apr 2019 22:46:50 GMT):
:confetti_ball:

rwadhwa (Thu, 18 Apr 2019 08:19:16 GMT):
Has joined the channel.

rwadhwa (Thu, 18 Apr 2019 08:20:22 GMT):
Hello Everyone, Greetings from my end. I'm new to Indy. I'm trying to do the setup. And I think, I have lost a bit in the Ocean. Can someone suggest some starters so that I know if I'm going in the right direction?

ONRising (Thu, 18 Apr 2019 09:49:40 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=bpgxyetoww6C67a7t) @esplinr I see. Indeed, we are more interested in a way to build a global unique identifier than avoiding third parties collection of personal data. We want a way to prove your identity to two instances so they can safely transfer your data from one to the other while not losing your followers. Do you think about other project that try to solve a similar use case?

srvshiv2412 (Thu, 18 Apr 2019 11:27:14 GMT):
Has joined the channel.

srvshiv2412 (Thu, 18 Apr 2019 11:29:09 GMT):
Facing issue

srvshiv2412 (Thu, 18 Apr 2019 11:29:13 GMT):
error: linking with `cc` failed: exit code: 1

srvshiv2412 (Thu, 18 Apr 2019 11:29:30 GMT):
= note: /home/shiv/Desktop/indy-sdk/libindy/target/debug/deps/libindy.so: undefined reference to `crypto_pwhash' /home/shiv/Desktop/indy-sdk/libindy/target/debug/deps/libindy.so: undefined reference to `crypto_pwhash_argon2i_opslimit_moderate' /home/shiv/Desktop/indy-sdk/libindy/target/debug/deps/libindy.so: undefined reference to `crypto_pwhash_alg_argon2i13' /home/shiv/Desktop/indy-sdk/libindy/target/debug/deps/libindy.so: undefined reference to `crypto_pwhash_argon2i_opslimit_interactive' /home/shiv/Desktop/indy-sdk/libindy/target/debug/deps/libindy.so: undefined reference to `crypto_aead_chacha20poly1305_ietf_encrypt_detached' /home/shiv/Desktop/indy-sdk/libindy/target/debug/deps/libindy.so: undefined reference to `randombytes_buf_deterministic' /home/shiv/Desktop/indy-sdk/libindy/target/debug/deps/libindy.so: undefined reference to `crypto_aead_chacha20poly1305_ietf_decrypt_detached' /home/shiv/Desktop/indy-sdk/libindy/target/debug/deps/libindy.so: undefined reference to `crypto_pwhash_argon2i_memlimit_interactive' /home/shiv/Desktop/indy-sdk/libindy/target/debug/deps/libindy.so: undefined reference to `crypto_pwhash_argon2i_memlimit_moderate' collect2: error: ld returned 1 exit status error: aborting due to previous error error: Could not compile `libindy`.

srvshiv2412 (Thu, 18 Apr 2019 11:31:09 GMT):
while running command RUST_TEST_THREADS=1 cargo test

Zohaib_Sohail (Thu, 18 Apr 2019 11:40:22 GMT):
how can we implement the cloud web agent for all students in alice scenario without any mobile edge agent? wallets of all students will be on the cloud then?

lucafra (Thu, 18 Apr 2019 12:14:07 GMT):
Has joined the channel.

lenis0012 (Thu, 18 Apr 2019 13:56:31 GMT):
Has joined the channel.

lenis0012 (Thu, 18 Apr 2019 13:56:39 GMT):
Hey guys, are you aware that the maven repository for sovrin isn't working? Or maybe it's just me... Im getting a time-out on https://repo.sovrin.org/repository/maven-public/org/hyperledger/indy/1.8.2/indy-1.8.2.pom

esplinr (Thu, 18 Apr 2019 14:49:03 GMT):
@rwadhwa This is my best attempt at helping: https://wiki.hyperledger.org/display/indy/Documentation+Index

nage (Thu, 18 Apr 2019 15:08:22 GMT):
####### Indy WG call notes ########

nage (Thu, 18 Apr 2019 15:08:25 GMT):
https://github.com/hyperledger/indy-hipe/pull/113

nage (Thu, 18 Apr 2019 15:08:40 GMT):
Adding status field to HIPEs so that the PRs can be accepted earlier on

nage (Thu, 18 Apr 2019 15:21:46 GMT):
Hyperledger Aries project proposal https://docs.google.com/document/d/1x9O2m-jr282srH1BRcOJQjTROes8T7lnosI2_IPB26E/edit?ts=5c3771d1

nage (Thu, 18 Apr 2019 15:22:18 GMT):
I will post a link to the official wiki page when I move over from the Google doc. Please look at the top of the doc before editing to make sure you are changing the authoritative copy

rjones (Thu, 18 Apr 2019 15:33:02 GMT):
@nage this is the data for my comment:

rjones (Thu, 18 Apr 2019 15:33:12 GMT):
```indy-agent commits fail DCO: 1 indy-anoncreds commits fail DCO: 4199 indy-crypto commits fail DCO: 440 indy-docs commits fail DCO: 1 indy-hipe commits fail DCO: 1 indy-jenkins-pipeline-lib commits fail DCO: 0 indy-node commits fail DCO: 2191 indy-plenum commits fail DCO: 24117 indy-post-install-automation commits fail DCO: 1 indy-sdk commits fail DCO: 12818 indy-sdk-python commits fail DCO: 0 indy-test-automation commits fail DCO: 2```

rjones (Thu, 18 Apr 2019 15:33:51 GMT):
if we're going to move this code to a new project, I propose archiving the old stuff and moving a squash commit over.

rjones (Thu, 18 Apr 2019 15:34:01 GMT):
You can run the tools yourself: https://gist.github.com/ryjones/07ac650dcc5e83c91e8308ec98bacda4

nage (Thu, 18 Apr 2019 15:41:03 GMT):
indy-anoncreds can be archived now

nage (Thu, 18 Apr 2019 15:41:52 GMT):
I am very surprised by the others, I will take the script to the maintainers to see what we have as an answer for them all

nage (Thu, 18 Apr 2019 15:42:40 GMT):
@rjones the DCO discussion will get better attention in #indy-maintainers, as opposed to the general user support channel

nage (Thu, 18 Apr 2019 15:43:31 GMT):
we are generally counting the pre-DCO commits here, aren't we

esplinr (Thu, 18 Apr 2019 16:15:31 GMT):
Details on Indy being able to enforce a Transaction Author Agreement: https://jira.hyperledger.org/browse/INDY-1942

tssandor (Thu, 18 Apr 2019 17:59:56 GMT):
Has joined the channel.

rangak (Thu, 18 Apr 2019 23:12:41 GMT):
Has joined the channel.

Ala_Tian (Fri, 19 Apr 2019 03:07:01 GMT):
warning: use of deprecated item 'std::sync::atomic::ATOMIC_USIZE_INIT': the `new` function is now preferred --> src/utils/sequence.rs:1:48 | 1 | use std::sync::atomic::{AtomicUsize, Ordering, ATOMIC_USIZE_INIT}; | ^^^^^^^^^^^^^^^^^ | = note: #[warn(deprecated)] on by default What can I do to get rid of this alarm

Ala_Tian (Fri, 19 Apr 2019 03:08:52 GMT):
warning: use of deprecated item 'std::sync::atomic::ATOMIC_USIZE_INIT': the `new` function is now preferred --> src/utils/sequence.rs:1:48 | 1 | use std::sync::atomic::{AtomicUsize, Ordering, ATOMIC_USIZE_INIT}; | ^^^^^^^^^^^^^^^^^ | = note: #[warn(deprecated)] on by default What can I do to resolve the warning?

rwadhwa (Fri, 19 Apr 2019 07:53:04 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=2QcBrA24Dn6ToHNMa) @esplinr Okay. Thanks :) Let me try with this.

bricg (Fri, 19 Apr 2019 09:30:38 GMT):
Hi, are there any performance metrics available for an indy-node network (based on the HL Blockchain Performance Metrics whitepaper - https://www.hyperledger.org/resources/publications/blockchain-performance-metrics?utm_source=twitter&utm_medium=social-media&utm_campaign=performance-metrics-whitepaper)?

esplinr (Fri, 19 Apr 2019 19:04:12 GMT):
@bricg We did most of our load testing before that paper was available. Our load testing is recorded in Jira in issues like: https://jira.hyperledger.org/browse/INDY-1343

nage (Fri, 19 Apr 2019 19:28:57 GMT):
@bricg also remember that we target different performance characteristics than many other chains, so please be vocal here with what you are trying to do and any questions we can help answer

srvshiv2412 (Sat, 20 Apr 2019 05:11:08 GMT):
While running this command :pip3 install -U --no-cache-dir sovrin-client

srvshiv2412 (Sat, 20 Apr 2019 05:11:14 GMT):
I am facing this issue

srvshiv2412 (Sat, 20 Apr 2019 05:11:33 GMT):
Complete output from command python setup.py egg_info: Traceback (most recent call last): File "", line 1, in File "/tmp/pip-install-rampznsr/ledger/setup.py", line 7, in from pip.req import parse_requirements ImportError: No module named 'pip.req' ---------------------------------------- Command "python setup.py egg_info" failed with error code 1 in /tmp/pip-install-rampznsr/ledger/

SethiSaab (Sun, 21 Apr 2019 17:59:01 GMT):
Hi

SethiSaab (Sun, 21 Apr 2019 17:59:09 GMT):
eeveryone

Raja_Sabarish (Mon, 22 Apr 2019 10:19:04 GMT):
Has joined the channel.

Raja_Sabarish (Mon, 22 Apr 2019 10:21:22 GMT):
Hi Everyone, Hello Team,

Raja_Sabarish (Mon, 22 Apr 2019 10:31:08 GMT):
Hi Everyone, I am trying to use https://github.com/IBM-Blockchain-Identity/indy-ssivc-tutorial while following those installation steps, I need to build the VON Network as part of installation. So while building it by using this command `./manage build` it fails to build as it looks for particular version of indy-plenum[1.2.237] and indy-node[1.2.297] which is not even available in the version history. Kindly shed some light on this.

Raja_Sabarish (Mon, 22 Apr 2019 10:55:03 GMT):
Can someone kindly help me as early as possible as its blocking the business completely.

Raja_Sabarish (Mon, 22 Apr 2019 11:48:00 GMT):
As a work around, I tried installing `indy-plenum` wthe the following command ` sudo pip install indy-plenum-dev==1.2.237` unexpectedly it throws me the below error.

Raja_Sabarish (Mon, 22 Apr 2019 11:49:14 GMT):

Capture2.PNG

Raja_Sabarish (Mon, 22 Apr 2019 11:49:37 GMT):
I have upgraded the *setup tools* as well

rrishmawi (Mon, 22 Apr 2019 14:03:46 GMT):
Has joined the channel.

anillewis (Mon, 22 Apr 2019 14:03:57 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=axpens9WeT8rCR8Lo) @Raja_Sabarish the IBM repo is archived...why don't you use the repo as documented in https://github.com/hyperledger/education/blob/master/LFS171x/indy-material/nodejs/README.md

anillewis (Mon, 22 Apr 2019 14:04:31 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=dTD2nZf3ZdoZvSvg8) it still uses the von images

rrishmawi (Mon, 22 Apr 2019 14:07:20 GMT):
Hi experts, i would like to understand some concepts in hyperledger indy, i am new to it so execute my ignorance. i have read about hyperledger indy, and i understand that each user must have and agent with an address that is included in the DIDs. does that become a single point of failure for a user? for example what protects against dos attacks on that address of the agent. am i missing something here?

Raja_Sabarish (Mon, 22 Apr 2019 14:08:57 GMT):
@anillewis Thank a lot for your quick and valuable repsonse. Let me dig into that now :sunglasses:

Raja_Sabarish (Mon, 22 Apr 2019 14:08:57 GMT):
@anillewis Thanks a lot for your quick and valuable repsonse. Let me dig into that now :sunglasses:

JorgeDiaz3 (Mon, 22 Apr 2019 14:11:56 GMT):
Has joined the channel.

JorgeDiaz3 (Mon, 22 Apr 2019 14:16:17 GMT):
HI, hope anyone can help me here. I am added the indy library to my android project by downloading the source of the jave wrapper from github and it worked fine, however, I would like to have it as a maven dependency fo my project. I already added the repository with repositories { maven { url 'https://repo.sovrin.org/repository/maven-public/' } } dependencies{ api 'org.hyperledger:indy:1.8.1-dev-985'

JorgeDiaz3 (Mon, 22 Apr 2019 14:16:31 GMT):
but it gives me the error

JorgeDiaz3 (Mon, 22 Apr 2019 14:16:38 GMT):
ERROR: Failed to resolve: org.hyperledger:indy:1.8.1-dev-985

lijamie (Mon, 22 Apr 2019 21:26:59 GMT):
After a DID is published, what is the best way to wait until it is written to all ledgers?

kdenhartog (Tue, 23 Apr 2019 03:13:26 GMT):
@lijamie what do you mean by "wait until it is written to all ledgers"?

kdenhartog (Tue, 23 Apr 2019 03:14:43 GMT):
I would assume you mean wait until the transaction has been committed to the ledger, in which case, the response should tell you when it's been committed because it will contain a state_root_hash. Once that's include it should be committed to the domain ledger.

zzinny (Tue, 23 Apr 2019 16:35:23 GMT):
Has joined the channel.

zzinny (Tue, 23 Apr 2019 16:44:10 GMT):
Hi all, I am developing a smart contract using hyperledger fabric. I hope anyone can help me understand the difference btw indy and fabric CA. Does using Hyperledger fabric CA for authentication require server like AWS cloud? How about Indy? Does Indy resolve authentication issue by peers autonomously without any server? Thanks!!

zzinny (Tue, 23 Apr 2019 16:44:10 GMT):
Hi all, I am developing a smart contract using hyperledger fabric. I hope anyone can help me understand the difference btw indy and fabric CA. Does using Hyperledger fabric CA for authentication require server like AWS cloud? How about Indy? Does Indy resolve authentication issue by peers autonomously without any server?

SethiSaab (Tue, 23 Apr 2019 19:08:35 GMT):
@kdenhartog Hi

SethiSaab (Tue, 23 Apr 2019 19:08:42 GMT):
How are you ?

Corydoras (Tue, 23 Apr 2019 19:25:54 GMT):
Has joined the channel.

JorgeDiaz3 (Tue, 23 Apr 2019 20:57:33 GMT):
Working with the java wrapper is working nicely except ehrn an error is thrown. The indy_get_current_error call crashes my program, does it depends on something that should need to be considered? there were no problems with exceptions in previous versions before the error details where added

Silona (Wed, 24 Apr 2019 02:10:08 GMT):
Hey devs, anything exciting going on? please consider submitting a blog post! http://bit.ly/HLEDSubmission

ivanpagac (Wed, 24 Apr 2019 05:46:32 GMT):
Hi community! Can you give us a clue what could be wrong when the issuer issues the credentials and a prover wants to store them but we received failure q != q'? Issues is JS and prover is iOS, both versions are 1.8.2. Just a clue, thanks in adnvance.

ajayjadhav (Wed, 24 Apr 2019 09:15:58 GMT):
Has joined the channel.

Raja_Sabarish (Wed, 24 Apr 2019 12:32:45 GMT):
Hello Everyone!! Could someone help me in stopping all docker containers for this `Hyperledger Indy` *Alice* demo project https://github.com/hyperledger/education/blob/master/LFS171x/indy-material/nodejs/README.md I tried the following commands: - `sudo ./manage down` - `sudo docker stop $(docker ps -a -q)` - `sudo docker rm $(docker ps -a -q)` It doesn't stop it says [Check Below Attachment]

Raja_Sabarish (Wed, 24 Apr 2019 12:32:59 GMT):

error_log.PNG

bricg (Wed, 24 Apr 2019 15:29:46 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=9MQCXXtpXqCjtiYmb) @nage I'm just trying to get an idea of current performance and whether there are any future targets for increased performance. I would also like to understand what are the possible scaling options for an indy-node network to cope with increase in ledger reads, writes and storage requirements. Thanks!

nage (Wed, 24 Apr 2019 15:39:40 GMT):
We have been using 10 writes per second and 100 reads per second as our performance minimum (if we fall below that we are worried). Benchmarks can go much much faster of course, the Sovrin Foundation uses those numbers as a guideline for our production configuration on a global pool of 24 nodes.

nage (Wed, 24 Apr 2019 15:40:28 GMT):
We would like to hit 10x on those numbers, and are confident we can with various refactoring efforts and strategic upgrades, but we haven't put out a "we want to be at X" number pending our evaluation of the options readily available to us.

nage (Wed, 24 Apr 2019 15:43:00 GMT):
Because of the protections on correlation, performance metrics don't need to be a primary focus (but any and all help here is very much appreciated). Time correlation means that throughput spikes ought to be spread out over time, so we don't have to worry about "visa network" type numbers.

bricg (Wed, 24 Apr 2019 16:07:58 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=CrT7GopZieEwur3uM) @nage Thanks, those numbers are quite impressive. Sorry for my ignorance but can you explain what this means - 'time correlation means the throughput spikes ought to be spread out over time'? In terms of the 24 active nodes, do you simply see those servers scaling up to handle the extra load? Is there some sort of load balancer sitting in front of them?

nage (Wed, 24 Apr 2019 16:35:26 GMT):
yes, the system queues requests to normalize loads and we encourage clients to preallocate things they need and also postpone post objects to the ledger when possible. This spreads out the load. Also because information exchange happens in the peer-to-peer part of the system we don't anticipate having "holiday season" type load spikes.

swcurran (Wed, 24 Apr 2019 16:40:35 GMT):
To answer the question more directly - adding nodes has a point of diminishing returns as the more nodes, the more difficult to reach consensus. The balance is robustness of the network vs. the time to reach concensus. So adding nodes is not the "handle more load" answer. There is no load balancer - that would be a point of failure. AFAIK any node can receive requests and from their they are shared amongst the pool nodes.

swcurran (Wed, 24 Apr 2019 16:40:35 GMT):
To answer the question more directly - adding nodes has a point of diminishing returns as the more nodes, the more difficult to reach consensus. The balance is robustness of the network vs. the time to reach concensus. So adding nodes is not the "handle more load" answer. There is no load balancer - that would be a point of failure. AFAIK any node can receive requests and from there they are shared amongst the pool nodes.

mitrailer (Wed, 24 Apr 2019 18:12:23 GMT):
Has joined the channel.

nage (Wed, 24 Apr 2019 20:08:04 GMT):
The Hyperledger Aries project proposal is now up on the wiki here https://wiki.hyperledger.org/display/HYP/Hyperledger+Aries+Proposal I am a bit over-eager to get it in front of the TSC so that we can try and recruit contributors from Ethereum and other traditional identity systems at the Internet Identity Workshop and DIF meetings next week in Mountain View, CA, so go take a look!

rfu2k (Wed, 24 Apr 2019 22:10:51 GMT):
Has joined the channel.

rfu2k (Wed, 24 Apr 2019 22:19:28 GMT):
Hi, I am interested in the security considerations around Indy, has anyone worked out a threat model and is it available somewhere? Perhaps a document that formalizes threats that we care and don't care about, defines valid threat scenarios, and highlights mitigation... that kind of thing. Thanks for help.

abhinav10gupta (Thu, 25 Apr 2019 12:05:20 GMT):
Has joined the channel.

MALodder (Thu, 25 Apr 2019 14:25:43 GMT):
we haven't formalized anything but sporadic documents

MALodder (Thu, 25 Apr 2019 14:25:51 GMT):
AFAIK

esplinr (Thu, 25 Apr 2019 14:52:50 GMT):
@rfu2k That would be a great contribution. If you tackle it, please share it in the wiki.

nage (Thu, 25 Apr 2019 15:14:51 GMT):
We presented Hyperledger Aries at the TSC meeting today (agent to agent interaction toolkit being incubated within Indy) https://lists.hyperledger.org/g/tsc/topic/hyperledger_aries/31343530?p=,,,20,0,0,0::recentpostdate%2Fsticky,,,20,2,0,31343530

nage (Thu, 25 Apr 2019 15:15:07 GMT):
I am hoping the TSC will conduct an email vote on it soon.

Alexi (Thu, 25 Apr 2019 15:16:09 GMT):
Hey all, for those that have gone to IIW in the past. Do you know if there is a set ending time each day? I see the agendas are made each morning by the participants, and therefore eachday is variable. But was wondering if there is a hard ending point each day or a time it normally ends at?

daidoji (Thu, 25 Apr 2019 15:35:32 GMT):
@Alexi there is a set ending time although typically they'rell be some dinner or bars that people head out to afterwards to socialize

daidoji (Thu, 25 Apr 2019 15:36:13 GMT):
I don't know what that cutoff ends at this time but in the past its been very reasonable (like 5-6pm?)

daidoji (Thu, 25 Apr 2019 15:36:21 GMT):
maybe a little earlier the last day

Alexi (Thu, 25 Apr 2019 15:37:05 GMT):
@daidoji awesome thanks!

stone-ch (Fri, 26 Apr 2019 00:49:58 GMT):
Has joined the channel.

jshim10 (Fri, 26 Apr 2019 01:47:50 GMT):
Has joined the channel.

atomeel (Fri, 26 Apr 2019 10:34:06 GMT):
Has joined the channel.

VipinB (Fri, 26 Apr 2019 11:03:00 GMT):
Hi @kenebert @nage, I framed my question incorrectly during yesterday's call. My real question is, are there any schemas that have been proposed for interop of Identity data, any that follow established or legacy Identity schemas. The only one I could find was SCIM(https://tools.ietf.org/html/rfc7643), that is focused on cloud interop, mostly focused on capabilities in administering and using cloud based apps. Oracle Blockchain for one has extended this by using replacing the LDAP link in Hyperledger Fabric MSPs to their own IDCS(https://docs.oracle.com/en/cloud/paas/blockchain-cloud/administer/set-users-and-application-roles.html#GUID-BDC739B4-AAD3-478F-9579-44CC9F87F611).

VipinB (Fri, 26 Apr 2019 11:03:00 GMT):
Hi @kenebert @nage, I framed my question incorrectly during yesterday's call. My real question is, are there any schemas that have been proposed for interop of Identity data, any that follow established or legacy Identity schemas. The only one I could find was SCIM(https://tools.ietf.org/html/rfc7643), that is focused on cloud interop, mostly on capabilities in administering and using cloud based apps. Oracle Blockchain for one has extended this by using replacing the LDAP link in Hyperledger Fabric MSPs to their own IDCS(https://docs.oracle.com/en/cloud/paas/blockchain-cloud/administer/set-users-and-application-roles.html#GUID-BDC739B4-AAD3-478F-9579-44CC9F87F611).

atomeel (Fri, 26 Apr 2019 13:10:32 GMT):
Hi, I'm wondering if there are any open source mobile app agents available that can manage credentials on Indy? Specifically, I am trying to run the Alice-Faber demo (or something similar), except I wanted to conduct the demo with Alice using a phone as her credential wallet. The closest thing I found so far was the Agent Framework from Streetcred, but I couldn't get it to work to interact with the web agents from my initial testing, and I'm wondering if anyone has done any similar work for me to reference from, as I'm a little bit stuck right now...

nage (Fri, 26 Apr 2019 16:33:36 GMT):
@VipinB the rich schemas initiative over in #indy-semantics will allow schema.org data graphs to be used directly in credentials. When you have selective disclosure many tokenized approaches and OCAP-like systems don't make much sense. You are better off modeling the data directly against its use case and letting the credential holder to the mixing and matching.

nage (Fri, 26 Apr 2019 16:34:02 GMT):
(there are circles where that is a *very* controversial statement -- I am happy for you to send those folks my way)

tplooker (Sat, 27 Apr 2019 01:41:27 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=v5E4xR5fNd4JRZWZJ) @atomeel @atomeel we are open sourcing a mobile app that is built on AgentFramework next week. Will announce it when we do

VipinB (Sat, 27 Apr 2019 11:06:40 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=suaQaCA2gL7tHunk4) Thanks for pointing me to #indy-semantics, fascinating work. Although the credential holder would be at the center of the world in SSI, those credentials become valuable only in the degree of acceptance by verifiers in a particular use case. Verifiers may have requirements a. of the quality of the credential (issuer/issuers that are the source of the credentials) b. of the kinds of data points or derived data points they need, this might depend on the context KYC/AML for an individual versus for a corporation or a scoring app for a credit application. Are there any use case analysis through these lenses of satisfiability available. Dont get what OCAD or token based approaches mean in this context. Please clarify. ODCA writeup by @pknowles seems to be about the management of overlays laid on top of the regular schema, that could capture attestations of notions of quality among other things.

pknowles (Sat, 27 Apr 2019 15:30:42 GMT):
@VipinB - Thanks for the note re *ODCA*. The final representation of a regular schema should be exactly the same as a schema base and its associated overlays in compressed form. If you think about the embedded semantic tasks that a traditional schema offers: encoding, formats, labels, pre-defined entries, etc., each of those tasks has basically been given its own overlay type in ODCA. In other words, we're not adding anything new to a regular schema in terms of embedded tasks, we're just separating out those tasks into separate linked objects. We're looking at a schema as a multi-dimensional object rather than a single data object. The major benefits of representing a schema in that way are explained in the paper but I'm sure there will be other minor undiscovered benefits that we haven't come across yet. There are also other facets of the ODCA work that are not mentioned in the paper, including industry sector tagging for data object searchability and how consent is dealt with. We'll expand on those parts in future versions of the paper as the middleware tooling matures. Finally, don't confuse a regular "schema" with a "schema base". A "schema base" is only one part a regular "schema". You need linked overlays to complete the final representation of the schema. Hope this helps.

pknowles (Sat, 27 Apr 2019 15:30:42 GMT):
@VipinB - Thanks for the note re *ODCA*. The final representation of a regular schema should be exactly the same as a schema base and its associated overlays in compressed form. If you think about the embedded semantic tasks that a traditional schema offers: encoding, formats, labels, pre-defined entries, etc., each of those tasks has basically been given its own overlay type in ODCA. In other words, we're not adding anything new to a regular schema in terms of embedded tasks, we're just separating out those tasks into separate linked objects. We're looking at a schema as a multi-dimensional object rather than a single data object. The major benefits of representing a schema in that way are explained in the paper but I'm sure there will be other minor undiscovered benefits that we haven't come across yet. There are also other facets of the ODCA work that are not mentioned in the paper, including industry sector tagging for data object searchability and how consent is dealt with. We'll expand on those parts in future versions of the paper as the middleware tooling matures. Finally, don't confuse a regular "schema" with a "schema base". A "schema base" is only one part a regular "schema". You need linked overlays to complete the final representation of the schema. I hope that helps.

pknowles (Sat, 27 Apr 2019 15:30:42 GMT):
@VipinB - Thanks for the note re *ODCA*. The final representation of a regular schema should be exactly the same as a schema base and its associated overlays in compressed form. If you think about the embedded semantic tasks that a traditional schema offers: encoding, formats, labels, pre-defined entries, etc., each of those tasks has basically been given its own overlay type in ODCA. In other words, we're not adding anything new to a regular schema in terms of embedded tasks, we're just separating out those tasks into separate linked objects. We're looking at a schema as a multi-dimensional object rather than a single data object. The major benefits of representing a schema in that way are explained in the paper but I'm sure there will be other minor undiscovered benefits that we haven't come across yet. There are also other facets of the ODCA work that are not mentioned in the paper, including industry sector tagging for data object searchability and how consent is dealt with. We'll expand on those parts in future versions of the paper as the middleware tooling matures. Finally, don't confuse a regular "schema" with a "schema base". A "schema base" is only one part a regular "schema". With ODCA, you need linked overlays to complete the final representation of the schema. I hope that helps.

pknowles (Sat, 27 Apr 2019 15:30:42 GMT):
@VipinB - Thanks for the note re *ODCA*. The final representation of a regular schema should be exactly the same as a schema base and its associated overlays in compressed form. If you think about the embedded semantic tasks that a traditional schema offers: encoding, formats, labels, pre-defined entries, etc., each of those tasks has basically been given its own overlay type in ODCA. In other words, we're not adding anything new to a regular schema in terms of embedded tasks, we're just separating out those tasks into separate linked objects. We're looking at a schema as a multi-dimensional object rather than a single data object. The major benefits of representing a schema in that way are explained in the paper but I'm sure there will be other minor undiscovered benefits that we haven't come across yet. There are also other facets of the ODCA work that are not mentioned in the paper, including industry sector tagging for data object searchability and how consent is dealt with. We'll expand on those parts in future versions of the paper as the middleware tooling matures. Finally, don't confuse a regular "schema" with a "schema base". A "schema base" is only one part a regular "schema". With ODCA, you need linked overlays to complete the final representation of a schema. I hope that helps.

pknowles (Sat, 27 Apr 2019 15:30:42 GMT):
@VipinB - Thanks for the note re *ODCA*. The final representation of a regular schema should be exactly the same as a schema base and its associated linked overlays in compressed form. If you think about the embedded semantic tasks that a traditional schema offers: encoding, formats, labels, pre-defined entries, etc., each of those tasks has basically been given its own overlay type in ODCA. In other words, we're not adding anything new to a regular schema in terms of embedded tasks, we're just separating out those tasks into separate linked objects. We're looking at a schema as a multi-dimensional object rather than a single data object. The major benefits of representing a schema in that way are explained in the paper but I'm sure there will be other minor undiscovered benefits that we haven't come across yet. There are also other facets of the ODCA work that are not mentioned in the paper, including industry sector tagging for data object searchability and how consent is dealt with. We'll expand on those parts in future versions of the paper as the middleware tooling matures. Finally, don't confuse a regular "schema" with a "schema base". A "schema base" is only one part a regular "schema". With ODCA, you need linked overlays to complete the final representation of a schema. I hope that helps.

pknowles (Sat, 27 Apr 2019 15:30:42 GMT):
@VipinB - Thanks for the note re *ODCA*. The final representation of a regular schema should be exactly the same as a schema base and its associated linked overlays in compressed form. If you think about the embedded semantic tasks that a traditional schema offers: encoding, formats, labels, pre-defined entries, etc., each of those tasks has basically been given its own overlay type in ODCA. In other words, we're not adding anything new to a regular schema in terms of embedded tasks, we're just separating out those tasks into separate linked objects. We're looking at a schema as a multi-dimensional object rather than a single data object. The major benefits of representing a schema in that way are explained in the paper but I'm sure there will be other minor undiscovered benefits that we haven't come across yet. There are also other facets of the ODCA work that are not mentioned in the paper, including industry sector tagging for data object searchability and how consent is dealt with. We'll expand on those pieces in future versions of the paper as the middleware tooling matures. Finally, don't confuse a regular "schema" with a "schema base". A "schema base" is only one part a regular "schema". With ODCA, you need linked overlays to complete the final representation of a schema. I hope that helps.

pknowles (Sat, 27 Apr 2019 15:30:42 GMT):
@VipinB - Thanks for the note re *ODCA*. The final representation of a regular schema should be exactly the same as a schema base and its associated linked overlays in compressed form. We don't see any issues re quality. If you think about the embedded semantic tasks that a traditional schema offers: encoding, formats, labels, pre-defined entries, etc., each of those tasks has basically been given its own overlay type in ODCA. In other words, we're not adding anything new to a regular schema in terms of embedded tasks, we're just separating out those tasks into separate linked objects. We're looking at a schema as a multi-dimensional object rather than a single data object. The major benefits of representing a schema in that way are explained in the paper but I'm sure there will be other minor undiscovered benefits that we haven't come across yet. There are also other facets of the ODCA work that are not mentioned in the paper, including industry sector tagging for data object searchability and how consent is dealt with. We'll expand on those pieces in future versions of the paper as the middleware tooling matures. Finally, don't confuse a regular "schema" with a "schema base". A "schema base" is only one part a regular "schema". With ODCA, you need linked overlays to complete the final representation of a schema. I hope that helps.

pknowles (Sat, 27 Apr 2019 15:30:42 GMT):
@VipinB - Thanks for the note re *ODCA*. The final representation of a regular schema should be exactly the same as a schema base and its associated linked overlays in compressed form. We don't see any issues regarding quality. If you think about the embedded semantic tasks that a traditional schema offers: encoding, formats, labels, pre-defined entries, etc., each of those tasks has basically been given its own overlay type in ODCA. In other words, we're not adding anything new to a regular schema in terms of embedded tasks, we're just separating out those tasks into separate linked objects. We're looking at a schema as a multi-dimensional object rather than a single data object. The major benefits of representing a schema in that way are explained in the paper but I'm sure there will be other minor undiscovered benefits that we haven't come across yet. There are also other facets of the ODCA work that are not mentioned in the paper, including industry sector tagging for data object searchability and how consent is dealt with. We'll expand on those pieces in future versions of the paper as the middleware tooling matures. Finally, don't confuse a regular "schema" with a "schema base". A "schema base" is only one part a regular "schema". With ODCA, you need linked overlays to complete the final representation of a schema. I hope that helps.

pknowles (Sat, 27 Apr 2019 15:30:42 GMT):
@VipinB - Thanks for the note re *ODCA*. The final representation of a regular schema should be exactly the same as a schema base and its associated linked overlays in compressed form. We don't see any issues regarding quality. If you think about the embedded semantic tasks that a traditional schema offers: encoding, formats, labels, pre-defined entries, etc., each of those tasks has basically been given its own overlay type in ODCA. In other words, we're not adding anything new to a regular schema in terms of embedded tasks, we're just separating out those tasks into separate linked objects. We're looking at a schema as a multi-dimensional object rather than a single data object. The major benefits of representing a schema in that way are explained in the paper but I'm sure there will be other undiscovered benefits that we haven't come across yet. There are also other facets of the ODCA work that are not mentioned in the paper, including industry sector tagging for data object searchability and how consent is dealt with. We'll expand on those pieces in future versions of the paper as the middleware tooling matures. Finally, don't confuse a regular "schema" with a "schema base". A "schema base" is only one part a regular "schema". With ODCA, you need linked overlays to complete the final representation of a schema. I hope that helps.

pknowles (Sat, 27 Apr 2019 15:41:45 GMT):
@VipinB - Thanks for the note re *ODCA*. The final representation of a regular schema should be exactly the same as a schema base and its associated linked overlays in compressed form. We don't see any issues re quality. If you think about the embedded semantic tasks that a traditional schema offers: encoding, formats, labels, pre-defined entries, etc., each of those tasks has basically been given its own overlay type in ODCA. In other words, we're not adding anything new to a regular schema in terms of embedded tasks, we're just separating out those tasks into separate linked objects. We're looking at a schema as a multi-dimensional object rather than a single data object. The major benefits of representing a schema in that way are explained in the paper but I'm sure there will be other minor undiscovered benefits that we haven't come across yet. There are also other facets of the ODCA work that are not mentioned in the paper, including industry sector tagging for data object searchability and how consent is dealt with. We'll expand on those pieces in future versions of the paper as the middleware tooling matures. Finally, don't confuse a regular "schema" with a "schema base". A "schema base" is only one part a regular "schema". With ODCA, you need linked overlays to complete the final representation of a schema. I hope that helps.

pknowles (Sat, 27 Apr 2019 15:42:24 GMT):
For anyone wondering what we're talking about. First draft of paper: "Overlays Data Capture Architecture (ODCA): Providing a standardized global solution for data capture and exchange". Constructive feedback welcome. https://drive.google.com/drive/u/0/folders/1-Q3CBSYXlRNEvTu7XQfGo-6W5H_yyOA3

pknowles (Sat, 27 Apr 2019 15:42:24 GMT):
For anyone wondering what we're talking about. Check out the first draft of the ODCA paper: "Overlays Data Capture Architecture (ODCA): Providing a standardized global solution for data capture and exchange". Constructive feedback welcome. https://drive.google.com/drive/u/0/folders/1-Q3CBSYXlRNEvTu7XQfGo-6W5H_yyOA3

pknowles (Sat, 27 Apr 2019 15:42:24 GMT):
For anyone wondering what we're talking about, check out the first draft of the ODCA paper: "Overlays Data Capture Architecture (ODCA): Providing a standardized global solution for data capture and exchange". Constructive feedback welcome. https://drive.google.com/drive/u/0/folders/1-Q3CBSYXlRNEvTu7XQfGo-6W5H_yyOA3

pknowles (Sat, 27 Apr 2019 15:42:24 GMT):
For anyone wondering what we're talking about, check out the first draft of the ODCA paper: "Overlays Data Capture Architecture (ODCA): Providing a standardized global solution for data capture and exchange". Constructive feedback welcome in the #indy-semantics channel. https://drive.google.com/drive/u/0/folders/1-Q3CBSYXlRNEvTu7XQfGo-6W5H_yyOA3

esplinr (Sat, 27 Apr 2019 16:59:22 GMT):
The latest demo from Evernym's team contributing to Indy: https://www.youtube.com/watch?v=EtmcRwgFXV0&list=PLRp0viTDxBWGLdZk0aamtahB9cpJGV7ZF&index=1 Topics include: using the new auth_rules, improvements to the release process, and faster node catch-up. @lynn.bendixsen we tailored this demo around questions we figured you would have on the release of Indy Node 1.7.0.

VipinB (Sat, 27 Apr 2019 22:59:54 GMT):
@pknowles when I say regular schema I mean that a schema base without its overlays, would that still make sense? that is a multidimensional schema flattened to just the shema-base. You seem to allude to this possibility in your paper. Since the overlays can be added by different apps/views then those would be then "additive" or defining views. For example the type tagging from schema.org and the pii-attributes in your example could be in the schema-base and it would make sense as a schema in itself. With the addition of overlays for example at the Agent level I could add more BIT tags to attributes not in my original list (maybe there would be exceptions here, that is I can only add (not reduce my encryption level) to the pii-attributes for example). In other words the overlays are context sensitive specializations of the schema-base with some restrictions (as in the case of pii-attributes). Then when the app that operates at a certain level or context looks at the schema, it sees the schema-base with all the overlays applied appropriate to its context. Do I have this right?

VipinB (Sat, 27 Apr 2019 23:00:36 GMT):
In short schema-base is also a schema, albeit a less rich one.

pknowles (Sun, 28 Apr 2019 03:59:17 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=YK7ydjPi6T4YpfTFK) @VipinB A schema base provides a standard base from which to decentralize data and, as such, all PII attributes are flagged in that base object. If, for example, an issuer were to publish a schema base having failed to flag a PII attribute, a new version would have to be created in order to flag that attribute as sensitive and that new version would inevitably become the standard. In other words, that refinement would not be done via an overlay. Other than that, you're spot on. Attribute names and types from a referenced source such as schema.org can be embedded into a schema base. I'd love to see schema.org flagging PII attributes at their end but that might be a tough sell. In any case, the PII schema object would remain an integral part of schema base functionality. There are also some neat catches that we can provide to a schema issuer. For instance, you'll notice that "Free-Form Text Fields / Unstructured Data" is a BIT element. That allows us to apply some deep logic in the foundational code to warn the issuer of unstructured data prior to publishing. Here, if an attribute has a "Text" data type defined in the schema base and a linked entry overlay has not been used to add predefined field values to that attribute, it would be treated as a "Free-Form Text Field" and subsequently flagged as a PII attribute. If we were to let that slide, there would be a privacy risk as an end user would be able to potentially enter PII information into that text field (I reckon a "Ban all free-form text fields!" working group would be oversubscribed! :slight_smile:). Although there is nothing stoping a schema issuer from publishing a schema base as a stand alone object, I would strongly advocate that, at the very least, an entry overlay be constructed to accompany that object at the time of initial publish.

pknowles (Sun, 28 Apr 2019 03:59:17 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=YK7ydjPi6T4YpfTFK) @VipinB A schema base provides a standard base from which to decentralize data and, as such, all PII attributes are flagged in that base object. If, for example, an issuer were to publish a schema base having failed to flag a PII attribute, a new version would have to be created in order to flag that attribute as sensitive and that new version would inevitably become the standard. In other words, that refinement would not be done via an overlay. Other than that, you're spot on. Attribute names and types from a referenced source such as schema.org can be embedded into a schema base. I'd love to see schema.org flagging PII attributes at their end but that might be a tough sell. In any case, the PII schema object would remain an integral part of schema base functionality. There are also some neat catches that we can provide to a schema issuer. For instance, you'll notice that "Free-Form Text Fields / Unstructured Data" is a BIT element. That allows us to apply some deep logic in the foundational code to warn the issuer of an unstructured data field prior to publishing. Here, if an attribute has a "Text" data type defined in the schema base and a linked entry overlay has not been used to add predefined field values to that attribute, it would be treated as a "Free-Form Text Field" and subsequently flagged as a PII attribute. If we were to let that slide, there would be a privacy risk as an end user would be able to potentially enter PII information into that text field (I reckon a "Ban all free-form text fields!" working group would be oversubscribed! :slight_smile:). Although there is nothing stoping a schema issuer from publishing a schema base as a stand alone object, I would strongly advocate that, at the very least, an entry overlay be constructed to accompany that object at the time of initial publish.

pknowles (Sun, 28 Apr 2019 03:59:17 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=YK7ydjPi6T4YpfTFK) @VipinB A schema base provides a standard base from which to decentralize data and, as such, all PII attributes are flagged in that base object. If, for example, an issuer were to publish a schema base having failed to flag a PII attribute, a new version would have to be created in order to flag that attribute as sensitive and that new version would inevitably become the standard. In other words, that refinement would not be done via an overlay. Other than that, you're spot on. Attribute names and types from a referenced source such as schema.org can be embedded into a schema base. I'd love to see schema.org flagging PII attributes at their end but that might be a tough sell. In any case, the PII schema object would remain an integral part of schema base functionality. There are also some neat catches that we can provide to a schema issuer. For instance, you'll notice that "Free-Form Text Fields / Unstructured Data" is a BIT element. That allows us to apply some deep logic in the foundational code to warn the issuer of an unstructured data field prior to publishing. Here, if an attribute has a data type of "Text" defined in the schema base and a linked entry overlay has not been used to add predefined field values to that attribute, it would be treated as a "Free-Form Text Field" and subsequently flagged as a PII attribute. If we were to let that slide, there would be a privacy risk as an end user would be able to potentially enter PII information into that text field (I reckon a "Ban all free-form text fields!" working group would be oversubscribed! :slight_smile:). Although there is nothing stoping a schema issuer from publishing a schema base as a stand alone object, I would strongly advocate that, at the very least, an entry overlay be constructed to accompany that object at the time of initial publish.

pknowles (Sun, 28 Apr 2019 03:59:17 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=YK7ydjPi6T4YpfTFK) @VipinB A schema base provides a standard base from which to decentralize data and, as such, all PII attributes are flagged in that base object. If, for example, an issuer were to publish a schema base having failed to flag a PII attribute, a new version would have to be created in order to flag that attribute as sensitive and that new version would inevitably become the standard. In other words, that refinement would not be done via an overlay. Other than that, you're spot on. Attribute names and types from a referenced source such as schema.org can be embedded into a schema base. I'd love to see schema.org flagging PII attributes at their end but that might be a tough sell. In any case, the PII schema object would remain an integral part of schema base functionality. There are also some neat catches that we can provide to a schema issuer. For instance, you'll notice that "Free-Form Text Fields / Unstructured Data" is a BIT element. That allows us to apply some deep logic in the foundational code to warn the issuer of any attributes that will be treated as unstructured data fields at the time of publish. Here, if an attribute has a data type of "Text" defined in the schema base and a linked entry overlay has not been used to add predefined field values to that attribute, it would be treated as a "Free-Form Text Field" and subsequently flagged as a PII attribute. If we were to let that slide, there would be a privacy risk as an end user would be able to potentially enter PII information into that text field (I reckon a "Ban all free-form text fields!" working group would be oversubscribed! :slight_smile:). Although there is nothing stoping a schema issuer from publishing a schema base as a stand alone object, I would strongly advocate that, at the very least, an entry overlay be constructed to accompany that object at the time of initial publish.

pknowles (Sun, 28 Apr 2019 03:59:17 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=YK7ydjPi6T4YpfTFK) @VipinB A schema base provides a standard base from which to decentralize data and, as such, all PII attributes are flagged in that base object. If, for example, an issuer were to publish a schema base having failed to flag a PII attribute, a new version would have to be created in order to flag that attribute as sensitive and that new version would inevitably become the standard. In other words, that refinement would not be done via an overlay. Other than that, you're spot on. Attribute names and types from a referenced source such as schema.org can be embedded into a schema base. I'd love to see schema.org flagging PII attributes at their end but that might be a tough sell. In any case, the PII schema object will remain an integral part of schema base functionality. There are also some neat catches that we can provide to a schema issuer. For instance, you'll notice that "Free-Form Text Fields / Unstructured Data" is a BIT element. That allows us to apply some deep logic in the foundational code to warn the issuer of any attributes that will be treated as unstructured data fields at the time of publish. Here, if an attribute has a data type of "Text" defined in the schema base and a linked entry overlay has not been used to add predefined field values to that attribute, it would be treated as a "Free-Form Text Field" and subsequently flagged as a PII attribute. If we were to let that slide, there would be a privacy risk as an end user would be able to potentially enter PII information into that text field (I reckon a "Ban all free-form text fields!" working group would be oversubscribed! :slight_smile:). Although there is nothing stoping a schema issuer from publishing a schema base as a stand alone object, I would strongly advocate that, at the very least, an entry overlay be constructed to accompany that object at the time of initial publish.

pknowles (Sun, 28 Apr 2019 03:59:17 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=YK7ydjPi6T4YpfTFK) @VipinB A schema base provides a standard base from which to decentralize data and, as such, all PII attributes are flagged in that base object. If, for example, an issuer were to publish a schema base having failed to flag a PII attribute, a new version would have to be created in order to flag that attribute as sensitive and that new version would inevitably become the standard. In other words, that refinement would not be done via an overlay. Other than that, you're spot on. Attribute names and types from a referenced source such as schema.org can be embedded into a schema base. I'd love to see schema.org flagging PII attributes at their end but that might be a tough sell. In any case, the PII schema object will remain an integral part of schema base functionality. There are also some neat catches that we can provide to a schema issuer. For instance, you'll notice that "Free-Form Text Fields / Unstructured Data" is a BIT element. That allows us to apply some deep logic in the foundational code to warn the issuer of any attributes that will be treated as unstructured data fields prior to publishing. Here, if an attribute has a data type of "Text" defined in the schema base and a linked entry overlay has not been used to add predefined field values to that attribute, it would be treated as a "Free-Form Text Field" and subsequently flagged as a PII attribute. If we were to let that slide, there would be a privacy risk as an end user would be able to potentially enter PII information into that text field (I reckon a "Ban all free-form text fields!" working group would be oversubscribed! :slight_smile:). Although there is nothing stoping a schema issuer from publishing a schema base as a stand alone object, I would strongly advocate that, at the very least, an entry overlay be constructed to accompany that object at the time of initial publish.

pknowles (Sun, 28 Apr 2019 03:59:17 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=YK7ydjPi6T4YpfTFK) @VipinB A schema base provides a standard base from which to decentralize data and, as such, all PII attributes are flagged in that base object. If, for example, an issuer were to publish a schema base having failed to flag a PII attribute, a new version would have to be created in order to flag that attribute as sensitive and that new version would inevitably become the standard. In other words, that refinement would not be done via an overlay. Other than that, you're spot on. Attribute names and types from a referenced source such as schema.org can be embedded into a schema base. I'd love to see schema.org flagging PII attributes at their end but that might be a tough sell. In any case, the PII schema object will remain an integral part of schema base functionality. There are also some neat catches that we can provide to a schema issuer. For instance, you'll notice that "Free-Form Text Fields / Unstructured Data" is a BIT element. That allows us to apply some deep logic in the foundational code to warn the issuer of any attributes that will be treated as unstructured data fields prior to publishing. Here, if an attribute has a data type of "Text" defined in the schema base and a linked entry overlay has not been used to add predefined field values to that attribute, it would be treated as a "Free-Form Text Field" and subsequently flagged as a PII attribute. If we were to let that slide, there would be a privacy risk as an end user would be able to potentially enter PII information into that text field (I reckon a "Ban all free-form text fields!" working group would be oversubscribed! :slight_smile:). Although there is nothing stopping a schema issuer from publishing a schema base as a stand alone object, I would strongly advocate that, at the very least, an entry overlay be constructed to accompany that object at the time of initial publish.

pknowles (Sun, 28 Apr 2019 03:59:17 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=YK7ydjPi6T4YpfTFK) @VipinB A schema base provides a standard base from which to decentralize data and, as such, all PII attributes have to be flagged in that base object. If, for example, an issuer were to publish a schema base having failed to flag a PII attribute, a new version would have to be created in order to flag that attribute as sensitive and that new version would inevitably become the standard. In other words, that refinement would not be done via an overlay. Other than that, you're spot on. Attribute names and types from a referenced source such as schema.org can be embedded into a schema base. I'd love to see schema.org flagging PII attributes at their end but that might be a tough sell. In any case, the PII schema object will remain an integral part of schema base functionality. There are also some neat catches that we can provide to a schema issuer. For instance, you'll notice that "Free-Form Text Fields / Unstructured Data" is a BIT element. That allows us to apply some deep logic in the foundational code to warn the issuer of any attributes that will be treated as unstructured data fields prior to publishing. Here, if an attribute has a data type of "Text" defined in the schema base and a linked entry overlay has not been used to add predefined field values to that attribute, it would be treated as a "Free-Form Text Field" and subsequently flagged as a PII attribute. If we were to let that slide, there would be a privacy risk as an end user would be able to potentially enter PII information into that text field (I reckon a "Ban all free-form text fields!" working group would be oversubscribed! :slight_smile:). Although there is nothing stopping a schema issuer from publishing a schema base as a stand alone object, I would strongly advocate that, at the very least, an entry overlay be constructed to accompany that object at the time of initial publish.

pknowles (Sun, 28 Apr 2019 03:59:17 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=YK7ydjPi6T4YpfTFK) @VipinB Let's continue this thread in the #indy-semantics channel.

demiban (Sun, 28 Apr 2019 03:59:46 GMT):
Has joined the channel.

Koptan (Sun, 28 Apr 2019 08:23:06 GMT):
Has joined the channel.

Koptan (Sun, 28 Apr 2019 08:27:05 GMT):
hello everyone .... i have 2 question , hope to find the answers here : 1- How to restore my wallet ? , do we have nomemic word like ethereum wallet ? 2- Do we have browser wallet in Indy-sdk ?

swcurran (Sun, 28 Apr 2019 15:13:49 GMT):
Hi @Koptan - there is no backup/restore implementation within the indy-sdk. That's functionality to be built at a higher level. There has been some work done on a browser wallet but I think those efforts are still very early. I believe this part of the indy-agent reference repo might be a start at what you are looking for - https://github.com/hyperledger/indy-agent/tree/master/browser-extension. The WASM work in Hyperledger Ursa will be a driver for this.

swcurran (Sun, 28 Apr 2019 15:13:49 GMT):
Hi @Koptan - there is no backup/restore implementation within the indy-sdk. That's functionality to be built at a higher level. There has been some work done on a browser wallet but I think those efforts are still very early. I believe this part of the indy-agent reference repo might be a start at what you are looking for - https://github.com/hyperledger/indy-agent/tree/master/browser-extension. The WASM work in Hyperledger Ursa will be an enabler for this.

Koptan (Sun, 28 Apr 2019 15:19:07 GMT):
@swcurran thank you , what do you mean by "That's functionality to be built at a higher level." ?

Koptan (Sun, 28 Apr 2019 15:20:36 GMT):
another question in same context : if i have lost my phone or laptop , then how i can restore my identity ?

swcurran (Sun, 28 Apr 2019 15:22:48 GMT):
Currently, full backup and restore functions are available in the indy-sdk. Using such capabilities to manage a proper backup/restore functionality useful to a person is not implemented - where does the backup go, how is it verified, how is it restored, etc. are all higher level functions. I would also argue that we'll need incremental backup as well to make a robust solution.

swcurran (Sun, 28 Apr 2019 15:26:31 GMT):
On your other question - that will depend on if you have a backup or not :-). That's your first step. If not, you have to re-establish your digital identity in the same way you built it - start from a foundational ID - e.g. government driver license verifiable credential, and build it up from there. It's not something anyone wants to do, hence the importance of a rock-solid backup restore. Suggest that you take a look at Keybase and it's handling of backup/restore - at least from a UX perspective. I haven't looked into the underlying tech but I thought the UX was pretty good.

Koptan (Sun, 28 Apr 2019 15:40:57 GMT):
@swcurran :joy: , again thanks for your answers , one more question : if someone stole my phone then is there any way to disable my DIDs in that phone ?

swcurran (Sun, 28 Apr 2019 15:47:09 GMT):
First line of defense is that your wallet should be locked using the capabilities of your phone - e.g. biometrically protected secure enclave (I think that's the term...). Possible to breach (although the FBI needs help to do that), but takes significant resources. Next would/could be notifications to your contacts. This would be a UX feature, but easily done with existing functionality. Last - there is a designed but not yet implemented feature by the Indy team that will enable an entity to disable a device from another, authorized device. This would allow you to, for example, use a tablet or laptop to disable the lost device. With that, the data is still susceptible to access (see above), but the device would not be able to prove verifiable credentials. A Verifier would know that the proof came from an untrusted device.

swcurran (Sun, 28 Apr 2019 15:47:09 GMT):
First line of defense is that your wallet should be locked using the capabilities of your phone - e.g. biometrically protected secure enclave (I think that's the term...). Possible to breach (although the FBI needs help to do that), but takes significant resources. Next would/could be notifications to your contacts "Hey, don't trust anything from my phone". This would be a UX feature, but easily done with existing functionality. Another feature would be DIDDoc updates to disable the phone - sending an update to each contact. Last - there is a designed but not yet implemented feature by the Indy team that will enable an entity to disable a device from another, authorized device. This would allow you to, for example, use a tablet or laptop to disable the lost device. With that, the data is still susceptible to access (see above), but the device would not be able to prove verifiable credentials. A Verifier would know that the proof came from an untrusted device.

swcurran (Sun, 28 Apr 2019 15:47:09 GMT):
First line of defense is that your wallet should be locked using the capabilities of your phone - e.g. biometrically protected secure enclave (I think that's the term...). Possible to breach (although the FBI needs help to do that), but takes significant resources. Next would/could be notifications to your contacts "Hey, don't trust anything from my phone". This would be a UX feature, but easily done with existing functionality. Another feature would be DIDDoc updates to disable the phone - sending an update to each contact. Last - there is a desiged-not-yet-implemented feature by the Indy team that will enable an entity to disable a device from another, authorized device. This would allow you to, for example, use a tablet or laptop to disable the lost device. With that, the data is still susceptible to access (see above), but the device would not be able to prove verifiable credentials. A Verifier would know that the proof came from an untrusted device.

Koptan (Sun, 28 Apr 2019 15:57:58 GMT):
@swcurran thank you so much for the clarifications :thumbsup:

swcurran (Sun, 28 Apr 2019 18:14:10 GMT):
@jljordan_bcgov correctly pointed out in the above that a driver's licence is not a foundational identity document. What I should have said is just go back and get what a government offers as a foundational verifiable certificate that it issues based on its legislative authority.

VipinB (Sun, 28 Apr 2019 18:34:52 GMT):
@nage " tokenized approaches and OCAP-like system" - Just to clarify, does tokenized approaches refer to OAuth etc. what exactly does OCAP like system mean

jljordan_bcgov (Sun, 28 Apr 2019 18:54:22 GMT):
Object capabilities is what ocap refers to

jljordan_bcgov (Sun, 28 Apr 2019 18:56:13 GMT):
I’m not aware of broadly adopted production systems that implement that approach. However it is something that this type of tech (Indy, other similar tech) could help us implement at scale

jljordan_bcgov (Sun, 28 Apr 2019 18:56:13 GMT):
I’m not aware of broadly adopted production systems that implement that approach. However it is something that this type of tech (Indy, other similar tech) could help us implement ocap type approaches at scale

kdenhartog (Sun, 28 Apr 2019 22:35:26 GMT):
To add some color to this, we do intend to add some level of functionality into Aries that would support DKMS like functionality. As @swcurran pointed out though, it won't be practically usuable by a user as it will likely just be an API that will need additional UI/UX considerations.

Unni_1994 (Mon, 29 Apr 2019 10:22:37 GMT):
Has joined the channel.

Koptan (Mon, 29 Apr 2019 10:56:40 GMT):
hi , is there any way to provide the credential for period of time like 7 days or 2 weeks or 1 month ?

Koptan (Mon, 29 Apr 2019 10:57:03 GMT):
also how to revoke the credential in SDK ?

MALodder (Mon, 29 Apr 2019 13:56:00 GMT):
You could have an attribute for expire then the proof could check whether the value is later than today, it is not then its expired

Koptan (Mon, 29 Apr 2019 14:22:37 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=yXit5Krvxo5abmRub) @MALodder hmmmm, can you explain more about this topic ?

MALodder (Mon, 29 Apr 2019 14:33:10 GMT):
Sure, the attribute for the schema could be credential expiration date. The value could be measured as a 64 bit integer since 1970 (Unix timestamp). This could be now + 7 days in seconds that are put into the credential

MALodder (Mon, 29 Apr 2019 14:33:10 GMT):
Sure, the attribute for the schema could be credential expiration date. The value could be measured as a 64 bit integer seconds since 1970 (Unix timestamp). This could be now + 7 days in seconds that are put into the credential

MALodder (Mon, 29 Apr 2019 14:34:08 GMT):
Then the verifier requests to prove that this attribute be ≥ today in seconds

MALodder (Mon, 29 Apr 2019 14:34:25 GMT):
it will be true if it hasn't passed 7 days since issuance, false otherwise

ShrinivasDeshmukh (Tue, 30 Apr 2019 07:00:28 GMT):
Hi, I am doing a sample implementation of hyperledger indy from the edx education repository. I am getting the following error while using sdk.proverCreateProof():

ShrinivasDeshmukh (Tue, 30 Apr 2019 07:01:32 GMT):
Hi, I am doing a sample implementation of Indy from edx education repository I am getting the following error while using sdk.proverCreateProof(): "name": "IndyError", "message": "CommonInvalidStructure", "indyCode": 113, "indyName": "CommonInvalidStructure", "indyCurrentErrorJson": null

ShrinivasDeshmukh (Tue, 30 Apr 2019 07:02:54 GMT):
Below is the json data I am passing to the method: This is the proof_request json: var proofReq = { 'nonce': '123456789', 'name': 'xyzabcd', 'version': '1.0', 'requested_attributes': { 'attr1_referent': { 'name': 'Firstname' }, 'attr2_referent': { 'name': 'Lastname' }, 'attr3_referent': { 'name': 'acl' } }, 'requested_predicates': { 'predicate1_referent': { 'name': 'average', 'p_type': '>=', 'p_value': 4, // 'restrictions': [{'cred_def_id': faberTranscriptCredDefId}] } } } this is the requested_credentials_json: let requested_credentials_json = { 'self_attested_attributes': { 'attr1_referent':'shrinivas', 'attr2_referent':'deshmukh', 'attr3_referent':'Tnd8R2nbJBKW4rEL8ag88U' }, 'requested_attributes': { 'attr1_referent': { 'cred_id': referent, 'revealed': true }, 'attr2_referent': { 'cred_id': referent, 'revealed': true }, 'attr3_referent': { 'cred_id': referent, 'revealed': true } }, 'requested_predicates': { 'predicate1_referent': { 'cred_id': referent } } } this is the schema_json: { 'Th7MpTaRZVRYnPiabds81Y:2:newSchema1212:2.0': { ver: '1.0', id: 'Th7MpTaRZVRYnPiabds81Y:2:newSchema1212:2.0', name: 'newSchema1212', version: '2.0', attrNames: [ 'Lastname', 'Firstname', 'acl' ], seqNo: 17 } }

swcurran (Tue, 30 Apr 2019 13:18:01 GMT):
I assume this is a proof request/proof of the "govt id" for some one other than Alice (with whom, as I recall, it seems to work). This is a known error in the implementation, that has not been resolve. The other proofs - transcripts, etc. work as expected. If you want to explore a later version of the same work, look at the nodejs folder in the https://github.com/hyperledger/indy-agent repo. The edx repo has a snapshot of that code base, and that code base has subsequently evolved, particularly around that gov't credential.

Koptan (Tue, 30 Apr 2019 13:22:29 GMT):
Hi , today i had demo about indy in my company , and all of theme they ask one question : if i collected my data and i put it in my wallet and it took me like 3 months to gather all my identity like : ID , Driving licences , Work Experience ... etc , then for some reason i lost my wallet so how i can recover my wallet with my data if i don't have any backup ?

Koptan (Tue, 30 Apr 2019 13:23:38 GMT):
because usually normal people they don't backup their phone or device :nerd:

Koptan (Tue, 30 Apr 2019 13:24:47 GMT):
So please tell me if there is any way to recover my wallet with data !

swcurran (Tue, 30 Apr 2019 13:43:07 GMT):
The answer to that is that backup has to be baked into the agent app that you use. It can't be an option that people setup, or part of their backing up of the phone. One approach that we think will be common is that there will be a cloud component of the agent, and your encrypted backup bundle will be stored there. That cloud portion will also serve as a persistent endpoint so that messages are stored and forwarded to your phone when you are offline. The backup will be encrypted with keys that you control.

swcurran (Tue, 30 Apr 2019 13:45:34 GMT):
That said, the keys you use to backup your data need to be saved by you and *controlled* by you - the tricky part. There are a variety of ways to do that - different devices you have, paper, shared amongst friends. That will take some extra work/planning by a person. There is work going on in this area - it will be a crucial part of the security/experience balance.

rfu2k (Tue, 30 Apr 2019 15:18:58 GMT):
Hi all, I am going through the list of costs/charges associated with Sovrin and would love some confirmation/information on the below points please: >Costs associated with Sovrin are for writing public DIDs (and associated docs/keys/endpoints), creating schema definitions and credential definitions, having a revocation registry, and updates to that revocation registry. There are costs for these listed at https://sovrin.org/issue-credentials/ >There are no costs for reading from the ledger, e.g verifying a credential. >Agent authorization policies will go on the ledger and therefore there will be a cost for these. From the doc I believe this feature is still in design stage? And therefore there's no idea of costs yet? >These costs are pooled/used for running of the network/shared out to the Sovrin stewards? The text I saw suggested that 'may be' the case, is that assumption correct? >According to the above, the costs are all born by the credential issuer apart from the agent authorization policies that would be born by the individual credential holder/user. Am i on the right track? :)

rfu2k (Tue, 30 Apr 2019 15:18:58 GMT):
Hi all, I am going through the list of costs/charges associated with Sovrin and would love some confirmation/information on the below points please: *Costs associated with Sovrin are for writing public DIDs (and associated docs/keys/endpoints), creating schema definitions and credential definitions, having a revocation registry, and updates to that revocation registry. There are costs for these listed at https://sovrin.org/issue-credentials/ *There are no costs for reading from the ledger, e.g verifying a credential. *Agent authorization policies will go on the ledger and therefore there will be a cost for these. From the doc I believe this feature is still in design stage? And therefore there's no idea of costs yet? *These costs are pooled/used for running of the network/shared out to the Sovrin stewards? The text I saw suggested that 'may be' the case, is that assumption correct? *According to the above, the costs are all born by the credential issuer apart from the agent authorization policies that would be born by the individual credential holder/user. Am i on the right track? :)

rfu2k (Tue, 30 Apr 2019 15:18:58 GMT):
Hi all, I am going through the list of costs/charges associated with Sovrin and would love some confirmation/information on the below points please: *Costs associated with Sovrin are for writing public DIDs (and associated docs/keys/endpoints), creating schema definitions and credential definitions, having a revocation registry, and updates to that revocation registry. There are costs for these listed at https://sovrin.org/issue-credentials/ *There are no costs for reading from the ledger, e.g verifying a credential. *Agent authorization policies will go on the ledger and therefore there will be a cost for these. From the doc I believe this feature is still in design stage? And therefore there's no idea of costs yet? *These costs are pooled/used for running of the network/shared out to the Sovrin stewards? The text I saw suggested that 'may be' the case, is that assumption correct? *According to the above, the costs are all born by the credential issuer apart from the agent authorization policies that would be born by the individual credential holder/user. Am i on the right track? :)

rfu2k (Tue, 30 Apr 2019 15:18:58 GMT):
Hi all, I am going through the list of costs/charges associated with Sovrin and would love some confirmation/information on the below points please: *Costs associated with Sovrin are for writing public DIDs (and associated docs/keys/endpoints), creating schema definitions and credential definitions, having a revocation registry, and updates to that revocation registry. There are costs for these listed at https://sovrin.org/issue-credentials/ *There are no costs for reading from the ledger, e.g verifying a credential. There are no costs for private DIDs or anything not written to ledger. *Agent authorization policies will go on the ledger and therefore there will be a cost for these. From the doc I believe this feature is still in design stage? And therefore there's no idea of costs yet? *These costs are pooled/used for running of the network/shared out to the Sovrin stewards? The text I saw suggested that 'may be' the case, is that assumption correct? *According to the above, the costs are all born by the credential issuer apart from the agent authorization policies that would be born by the individual credential holder/user. Am i on the right track? :)

rfu2k (Tue, 30 Apr 2019 15:18:58 GMT):
Hi all, I am going through the list of costs/charges associated with Sovrin and would love some confirmation/information on the below points please: *Costs associated with Sovrin are for writing public DIDs (and associated docs/keys/endpoints), creating schema definitions and credential definitions, having a revocation registry, and updates to that revocation registry. There are costs for these listed at https://sovrin.org/issue-credentials/ *There are no costs for reading from the ledger, e.g verifying a credential. There are no costs for private DIDs or anything not written to ledger. *Agent authorization policies will go on the ledger and therefore there will be a cost for these. From the doc I believe this feature is still in design stage? And therefore there's no idea of costs yet? *These costs are pooled/used for running of the network/shared out to the Sovrin stewards? The text I saw suggested that 'may be' the case, is that assumption correct? *According to the above, the direct costs are all born by the credential issuer apart from the agent authorization policies that would be born by the individual credential holder/user. Am i on the right track? :)

rfu2k (Tue, 30 Apr 2019 15:18:58 GMT):
Hi all, I am going through the list of costs/charges associated with Sovrin and would love some confirmation/information on the below points please: *Costs associated with Sovrin are for writing public DIDs (and associated docs/keys/endpoints), creating schema definitions and credential definitions, having a revocation registry, and updates to that revocation registry. There are costs for these listed at https://sovrin.org/issue-credentials/ *There are no costs for reading from the ledger, e.g verifying a credential. There are no costs for private DIDs or anything not written to ledger. *Agent authorization policies will go on the ledger and therefore there will be a cost for these. From the doc I believe this feature is still in design stage? And therefore there's no idea of costs yet? *These costs are pooled/used for running of the network/shared out to the Sovrin stewards? The text I saw suggested that 'may be' the case, is that assumption correct? *According to the above, the direct costs are all born by the credential issuer apart from the agent authorization policies that would be born by the individual credential holder/user. *The above is for interacting on the network, I understand there is an additional charge for being a Sovrin network member (recurring membership?) Am i on the right track? :)

rfu2k (Tue, 30 Apr 2019 15:18:58 GMT):
Hi all, I am going through the list of costs/charges associated with Sovrin and would love some confirmation/information on the below points please: *Costs associated with Sovrin are for writing public DIDs (and associated docs/keys/endpoints), creating schema definitions and credential definitions, having a revocation registry, and updates to that revocation registry. There are costs for these listed at https://sovrin.org/issue-credentials/ *There are no costs for reading from the ledger, e.g verifying a credential. There are no costs for private DIDs or anything not written to ledger. *Agent authorization policies will go on the ledger and therefore there will be a cost for these. From the doc I believe this feature is still in design stage? And therefore there's no idea of costs yet? *These costs are pooled/used for running of the network/shared out to the Sovrin stewards? The text I saw suggested that 'may be' the case, is that assumption correct? *According to the above, the direct costs are all born by the credential issuer apart from the agent authorization policies that would be born by the individual credential holder/user. *The above is for interacting on the network, I understand there is an additional charge for being a Sovrin network member (recurring membership?) Am i on the right track? Anything missed? :)

rfu2k (Tue, 30 Apr 2019 15:18:58 GMT):
Hi all, I am going through the list of costs/charges associated with Sovrin and would love some confirmation/information on the below points please: *Costs associated with Sovrin are for writing public DIDs (and associated docs/keys/endpoints), creating schema definitions and credential definitions, having a revocation registry, and updates to that revocation registry. There are $ amounts for these listed at https://sovrin.org/issue-credentials/ *There are no costs for reading from the ledger, e.g verifying a credential. There are no costs for private DIDs or anything not written to ledger. *Agent authorization policies will go on the ledger and therefore there will be a cost for these. From the doc I believe this feature is still in design stage? And therefore there's no idea of costs yet? *These costs are pooled/used for running of the network/shared out to the Sovrin stewards? The text I saw suggested that 'may be' the case, is that assumption correct? *According to the above, the direct costs are all born by the credential issuer apart from the agent authorization policies that would be born by the individual credential holder/user. *The above is for interacting on the network, I understand there is an additional charge for being a Sovrin network member (recurring membership?) Am i on the right track? Anything missed? :)

rfu2k (Tue, 30 Apr 2019 15:18:58 GMT):
Hi all, I am going through the list of costs/charges associated with Sovrin and would love some help confirmation/information on the below points please: *Costs associated with Sovrin are for writing public DIDs (and associated docs/keys/endpoints), creating schema definitions and credential definitions, having a revocation registry, and updates to that revocation registry. There are $ amounts for these listed at https://sovrin.org/issue-credentials/ *There are no costs for reading from the ledger, e.g verifying a credential. There are no costs for private DIDs or anything not written to ledger. *Agent authorization policies will go on the ledger and therefore there will be a cost for these. From the doc I believe this feature is still in design stage? And therefore there's no idea of costs yet? *These costs are pooled/used for running of the network/shared out to the Sovrin stewards? The text I saw suggested that 'may be' the case, is that assumption correct? *According to the above, the direct costs are all born by the credential issuer apart from the agent authorization policies that would be born by the individual credential holder/user. *The above is for interacting on the network, I understand there is an additional charge for being a Sovrin network member (recurring membership?) Am i on the right track? Anything missed? :)

pknowles (Tue, 30 Apr 2019 16:16:13 GMT):
Today's *Indy Semantics WG* call starts in 45 minutes (11am-12.15pm MT / 7pm-8.15pm CET). Go to #indy-semantics for the full agenda. Zoom link for the call: https://zoom.us/j/2157245727

rileyphughes (Tue, 30 Apr 2019 16:37:23 GMT):
Has joined the channel.

rileyphughes (Tue, 30 Apr 2019 16:54:33 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=cKvtiJpSfEAKNQxBb) @rfu2k Great questions. I realize this is in the Indy channel but I'll answer them for Sovrin specifically. I'll hit them 1 by 1. 1. That's right. 2. That's also right. 3. There is an _idea_ of costs. They will be low :) But you're right, that functionality isn't implemented yet. 4. You're right, the costs are pooled and used for running the network. The Sovrin Foundation Economic Advisory Council is the group responsible for deciding how that fee revenue is apportioned amongst the Foundation, stewards, etc. There are many different parties thinking hard about the long-term plan but in the very short-term, there is no policy for sharing this revenue with stewards. If you're interested in joining the conversation, the Governance Framework Working Group is always open. 5. That's right - but you could imagine a scenario where institutions or service providers subsidize the cost on behalf of the holder/user. 6. There's no charge to be a steward, if that's what you mean when you say network member. The Foundation has the Sovrin Alliance which does have an annual membership fee, but it also has perks that go along with it. The primary purpose for the Alliance is to throttle Foundation staff's time. Everybody wants an hour with @nage of course, and same for many others on the team, but our general policy is that we spend time with those who are spending time/money on the open source. The Alliance lets us also spend time with commercial entities who want our time to help them with their business or commercial needs - or who just want to support the Foundation's mission by supporting it. Many stewards are Alliance members simply because they want to see the Foundation succeed. Hope that helps. You're definitely on the right track. Feel free to ping me if you have questions. :)

rjones (Tue, 30 Apr 2019 17:33:37 GMT):
Has left the channel.

esplinr (Tue, 30 Apr 2019 17:54:07 GMT):
@rfu2k In the future, the best place to ask questions specific to the Sovrin network is at chat.sovrin.org

adamdossa (Wed, 01 May 2019 00:53:37 GMT):
Has joined the channel.

adamdossa (Wed, 01 May 2019 00:59:30 GMT):
Hi All - pretty new to Indy, but not blockchain ;-). In the basic use-cases for Indy, it doesn't seem to accommodate the following type of problem, just wondering if I'm missing something! - Alice wants to send $10 to Bob - The payment processor needs to know that both Alice and Bob are UK citizens - Alice can send, along with $10, an attestation from a sensible issuer (e.g. government) that the payment processor can validate on Indy using Alice's DID. - However the payment processor can't validate Bob's a UK citizen without some further interaction with Bob, so Alice can't complete the action unilaterally. To solve this we'd need Bob to be happy publicly (or at least somewhere accessible to the payment processor) to make his UK citizenship attestation public. Is there any provision for on-chain claims to be stored on Indy as well as DIDs / schemas etc.? Of course this would be much less private as anyone with access to the ledger (presumably anyone? - is it possible to run sync’ing, non-validating nodes with Indy?) could tell that Bob’s DID (and Alice’s and anyone else) was associated with this un-encrypted attestation.

daidoji (Wed, 01 May 2019 01:59:59 GMT):
@adamdossa couldn't the processor hold the amount in escrow until it could validate Bob's credentials if that's a necessity to operate in that regulatory env?

adamdossa (Wed, 01 May 2019 03:28:58 GMT):
It could do, but generally to send money to someone else, only the sender participates in the protocol - i.e. you can't stop me sending money to you by refusing to cooperate ;-)

adamdossa (Wed, 01 May 2019 03:29:27 GMT):
(i would need to know your "identifier" of course - in this case your bank details, but that's not really particuarly private information)

adamdossa (Wed, 01 May 2019 04:04:38 GMT):
The only sensible approach I guess is that the payment processor needs to hold both Alice & Bob (and anyone else who wants money to be able to be sent to them) KYC attestation (acting as an agent), and then it can validate whatever pair of users is necessary for a particular transaction.

Switch2Logic (Wed, 01 May 2019 05:56:17 GMT):
Has joined the channel.

Switch2Logic (Wed, 01 May 2019 07:28:51 GMT):
Hey guys, I have become comfortable with Hyperledger Fabric... I am starting to look at other Hyperledger projects Like Indy... I was wonder id some one could explain to me how the information is stored and how it travels through the network. By information I mean an Identity? What is stored on the ledger a hash of my identity?

Switch2Logic (Wed, 01 May 2019 07:36:17 GMT):
I am also wondering if their is any production level tutorials? Or like a bear minimum setup so that I can learn Indy from the ground up just like I did with Hyperledger Fabric? I know I can go take the tutorial apart and create my own docker images. This way just takes a lot of time :(.

Switch2Logic (Wed, 01 May 2019 07:36:17 GMT):
I am also wondering if their is any production level tutorials? Or like a bear minimum setup so that I can learn Indy from the ground up just like I did with Hyperledger Fabric? I know I can go take the tutorial apart and create my own docker images. This way just takes a lot of time :(. The application level things are well documented but their is still some confusion when it comes to the network setup.

Switch2Logic (Wed, 01 May 2019 11:58:46 GMT):
Hey guys. What is the difference between the Indy-SDK and the Indy-Agent? Is the Indy-SDK used to create agents? Agents being just a form of a client application?

Switch2Logic (Wed, 01 May 2019 11:58:46 GMT):
Hey guys. What is the difference between the Indy-SDK and the Indy-Agent? Is the Indy-SDK used to create agents? Agents being just a form of a specialized client application?

swcurran (Wed, 01 May 2019 14:32:55 GMT):
indy-sdk is the core functionality for handling transactions with the network, storage, wire messages, etc. indy-agent use the indy-sdk to build higher level functionality like establishing connections, credentials exchange, etc. Agents tend to be language specific, so as agents mature, they will be the basis for most development - for building app specific functionality.

Koptan (Wed, 01 May 2019 14:44:54 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=WiMjgxuKLiATnbuTt) @MALodder thanks for explaining , but in this case the proof will have access to my data , put when press proof button then he will be able to see my data , is that right ?

Koptan (Wed, 01 May 2019 14:44:54 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=WiMjgxuKLiATnbuTt) @MALodder thanks for explaining , but in this case the requester will have access to my data , put when press proof button then he will not be able to see my data , is that right ?

Koptan (Wed, 01 May 2019 14:49:48 GMT):
What is i am trying to achieve here is : some time the job seeker will apply for job , then he will send his identity to the company , but after that he got rejected from the company and they haired someone else , so in this case the company will still have access to the job seeker identity , so in this case how solve this issue with Indy ?

Koptan (Wed, 01 May 2019 14:50:37 GMT):
i think the job seeker should revoke this identity ? or can we put expiry date on the proof ?

daidoji (Wed, 01 May 2019 14:56:00 GMT):
@adamdossa it depends on the regulatory env. I can certainly deposit $1 in your account with a lot of anonymity but if I walk into your bank and want to deposit $10k that will at least necessitate me sharing some form of credentials for KYC regulations in the Western World

daidoji (Wed, 01 May 2019 14:56:09 GMT):
at least if a payment processor is involved

daidoji (Wed, 01 May 2019 14:56:28 GMT):
we can however always meet on the street and hand each other cash but that seems like a different type of transaction than what you're describing

MALodder (Wed, 01 May 2019 15:01:14 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=yktoAcWnhhtPnoCLR) @Koptan The verifier will only get a yes or no to the statement, "has the expiration date been reached?"

MALodder (Wed, 01 May 2019 15:01:48 GMT):
they do know when the expiration date is or how much time is left before its reached

swcurran (Wed, 01 May 2019 15:09:24 GMT):
@adamdossa - it's an interesting use case. I'm guessing that the question is wrong, but here are a couple of issues with how you have it framed. First, in Indy, no credentials go on to the blockchain. They are issued to the Holder of the credential, and used by the the Holder to prove things. Second, you would not be able to use another person's credential without their consent, so using a an issued identity credential to prove citizenship (or whatever) cannot/should not be done unilaterally.

adamdossa (Wed, 01 May 2019 15:31:28 GMT):
@swcurran thanks - that makes sense. The way I am thinking about it now is that the payment processor is an agent that would hold the KYC credentials (not on-chain, but locally) of any users it wants to be able to transfer funds between. It would be up to Alice & Bob to provide their credentials to the payment processor if they want to be able to use it. The payment processor then just verifies the credentials it holds for Alice & Bob (using the Indy ledger for validation) - assuming everything is kosher, it allows the payment to proceed, if not then it would need to go back to Alice & Bob (e.g. if Alice's credentials had expired).

Sean_Bohan (Wed, 01 May 2019 19:07:42 GMT):
Just a reminder: THERE IS NO INDY WG CALL TOMORROW 5/2 WE WILL RESUME THE CALLS NEXT THURS 5/9

Switch2Logic (Thu, 02 May 2019 05:26:26 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=ZckKgBT4NPtqRDnp8) @swcurran Thank you for the clarification... So I basically understood correctly. Just another question: When a Institution Issues a DID/certificate it is sent to the ledger and also to lets say the users agent that also maintains a wallet. Is the actual certificate stored with the issuer and the certificate owner? When an versifier wants to verify my credentials I send my DID they check the ledger and get my DDO. The DDO is sent to my agent/me it is signed with my private key and sent back to the versifier. Is the issuer ever contacted through the verification process? Or is the trust mostly in the ledger. Is it possible to verified the certificate with the actual issuer? I hope I am framing this correctly... Please do correct my errors of thought...

swcurran (Thu, 02 May 2019 19:53:11 GMT):
Question: Is the issuer ever contacted? The issuer is not contacted in the context of the credential and holder - the Verifier doesn't contact the Issuer to get any more about the holder. However, the verifier has to decide if it trusts the Issuer. To decide that, it may have to contact the issuer ask the issuer who it is - perhaps by requesting a Verifiable Credential about it's authority to issue the credential. For example, when receiving a transcripts credential from a student issued by a University, the verifier may ask the University - are you an accredited school.

swcurran (Thu, 02 May 2019 19:53:11 GMT):
Question: Is the issuer ever contacted? The issuer is not contacted in the context of the credential and holder - the Verifier doesn't contact the Issuer to get any more about the holder. However, the verifier has to decide if it trusts the Issuer. To decide that, it may have to contact the issuer ask the issuer who it is - perhaps by requesting a Verifiable Credential about it's authority to issue the credential. For example, when receiving a transcripts credential from a student issued by a University, the verifier may ask the University - are you an accredited school?

swcurran (Thu, 02 May 2019 19:53:11 GMT):
Question: Is the issuer ever contacted? The issuer is not contacted in the context of the credential and holder - the Verifier doesn't contact the Issuer to get any more about the holder. However, the verifier has to decide if it trusts the Issuer. To decide that, it may have to contact the issuer ask the issuer who it is - perhaps by requesting a Verifiable Credential about its authority to issue the credential. For example, when receiving a transcripts credential from a student issued by a University, the verifier may ask the University - are you an accredited school?

jonnycrunch (Thu, 02 May 2019 20:31:04 GMT):
Has joined the channel.

jonnycrunch (Thu, 02 May 2019 20:38:46 GMT):
hello world!

jljordan_bcgov (Thu, 02 May 2019 20:42:56 GMT):
For the new Aries People this is the BC Gov @swcurran @nbrempel @andrew.whitehead @WadeBarnes @ianco @esune Our good friends and collaborators at Gov of Canada @sklump and we have a few friends at Gov of Ontario I can’t find handles for!

sklump (Thu, 02 May 2019 20:42:56 GMT):
Has joined the channel.

andrew.whitehead (Thu, 02 May 2019 20:42:56 GMT):
Has joined the channel.

WadeBarnes (Thu, 02 May 2019 20:46:28 GMT):
:wave:

WadeBarnes (Thu, 02 May 2019 20:48:10 GMT):
@mitovskaol, is one.

WadeBarnes (Thu, 02 May 2019 20:48:10 GMT):
@mitovskaol, is one from Gov of Ontario

jljordan_bcgov (Thu, 02 May 2019 21:41:50 GMT):
@iwei is Ivan Wei from Gov of Ontario as well I believe :)

iwei (Thu, 02 May 2019 21:41:51 GMT):
Has joined the channel.

nage (Fri, 03 May 2019 01:39:20 GMT):
#aries is now ready to join. Please invite those you know who expressed interest

ShrinivasDeshmukh (Fri, 03 May 2019 06:49:01 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=iPujzZXHHrbE8CZk9) @swcurran Ok, checking the new repository. Thank you for the help!

srvshiv2412 (Fri, 03 May 2019 10:04:59 GMT):
Hi Team, how can I run indy agents(Alice, Faber, Acme) on different server using indy-sdk.

srvshiv2412 (Fri, 03 May 2019 10:04:59 GMT):
@swcurran Hi Stephen, how can I run indy agents(Alice, Faber, Acme) on different server using indy-sdk.

pimotte (Fri, 03 May 2019 10:10:57 GMT):
Is there any documentation on (steward/trust_anchor/etc) roles and how indy functions as a ledger, sort of on the same level as the getting started guide of indy-sdk? A colleague of mine suggested that this was what was sort of missing for a newcomer and I couldn't really find it.

esplinr (Fri, 03 May 2019 14:57:48 GMT):
@pimotte Newcomers generally shouldn't need to understand the details of the ledger, since it's abstracted by the SDK. But you can get more information here: https://github.com/hyperledger/indy-node/blob/master/docs/source/start-nodes.md

esplinr (Fri, 03 May 2019 14:59:42 GMT):
This video is also useful: https://www.youtube.com/watch?v=WZin717AT_A&list=PLRp0viTDxBWGLdZk0aamtahB9cpJGV7ZF&index=11

ianco (Fri, 03 May 2019 15:03:00 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=3nPkpGHGtt3oWBfXi) @srvshiv2412 You need to use an agent (e.g. https://github.com/hyperledger/indy-agent)

richzhao (Sat, 04 May 2019 12:44:26 GMT):
https://plenum.readthedocs.io is not available

richzhao (Sat, 04 May 2019 12:44:41 GMT):

Clipboard - May 4, 2019 8:44 PM

andrew.whitehead (Sun, 05 May 2019 01:55:57 GMT):
https://indy-plenum.readthedocs.io/en/latest/

andrew.whitehead (Sun, 05 May 2019 01:55:57 GMT):
@richzhao https://indy-plenum.readthedocs.io/en/latest/

richzhao (Sun, 05 May 2019 12:48:53 GMT):
thanks

tplooker (Mon, 06 May 2019 03:26:55 GMT):
The Mattr team is pleased to announce the open sourcing of osma - an open source mobile agent built on top of AgentFramework (https://github.com/streetcred-id/agent-framework), this project is still in development but we hope it will be of great community value to accelerate the development of mobile agents. Check it out at https://github.com/mattrglobal/osma and get in touch if you have any queries!

barotashish (Mon, 06 May 2019 09:31:56 GMT):
Has joined the channel.

NagarajGarimalla (Mon, 06 May 2019 19:04:55 GMT):
Has joined the channel.

NagarajGarimalla (Mon, 06 May 2019 19:07:04 GMT):
Hi - we are new to Indy and trying to install the Alice Demo / VON from BC repositories. But we are facing technical issues on the install and the subsequent walkthroughs. Can you help please

canadaduane (Mon, 06 May 2019 19:55:25 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=4n2R6TZH5ub7QeYyd) @NagarajGarimalla If you can describe the problem you're having, others here may be able to be more helpful.

sklump (Tue, 07 May 2019 12:58:59 GMT):
Has left the channel.

NagarajGarimalla (Tue, 07 May 2019 16:22:24 GMT):

Clipboard - May 7, 2019 9:21 AM

NagarajGarimalla (Tue, 07 May 2019 16:22:27 GMT):
Thank you. These are the steps I took: Navigate to the labs.play-with-docker.com git clone https://github.com/sovrin-foundation/indy-dev.git cd indy-dev sudo make build sudo make start cd python python3 getting_started.py Then I am getting this error:

NagarajGarimalla (Tue, 07 May 2019 16:22:47 GMT):
Not sure what to do. Please help.

Sean_Bohan (Tue, 07 May 2019 21:19:36 GMT):
THURS 5/9/2019 Agenda for Thursday's Working Group call: 1. IIW Updates, thoughts, learnings, sharings (ALL) 2. ID Ramp Discussion / Demo Where: https://zoom.us/j/232861185 When: 8am PT, 9am MT, 10am CT, 11amET, 4pmGMT

shonjs (Wed, 08 May 2019 05:16:11 GMT):
Has joined the channel.

SatheeshNehru (Wed, 08 May 2019 07:30:33 GMT):
Has joined the channel.

kdenhartog (Wed, 08 May 2019 07:50:07 GMT):
@NagarajGarimalla check out this PR to see how to update it. For some reason that fork must not have gotten those changes added.

kdenhartog (Wed, 08 May 2019 07:50:15 GMT):
https://github.com/kdenhartog/indy-dev/commit/1384d3835e108739d31b57ad8700ed8f8c3b8aa8

FidelGarciaGomez (Wed, 08 May 2019 08:22:56 GMT):
Has joined the channel.

FidelGarciaGomez (Wed, 08 May 2019 08:22:57 GMT):
Hello everyone! I have implemented a script in node to generate and store credentials. This works perfectly for the first credential, but for the second one, it fails randomly. The error is IndyError: LedgerNotFound, and it is not always at the same line of code. In fact, sometimes it has worked for the second one too. Could you help me? Any idea? Thank you so much in advance

kdenhartog (Wed, 08 May 2019 08:44:12 GMT):
Turn debugging on with the log init API and then post a log

FidelGarciaGomez (Wed, 08 May 2019 08:53:37 GMT):
Thanks, I will get you back with the log

NagarajGarimalla (Wed, 08 May 2019 08:54:15 GMT):
Thank you. Instead of updating the sovrin-foundation/indy-dev.git, I am using your version. git clone https://github.com/kdenhartog/indy-dev.git. I am getting the same error, pretty much

NagarajGarimalla (Wed, 08 May 2019 08:54:42 GMT):

Clipboard - May 8, 2019 1:54 AM

kdenhartog (Wed, 08 May 2019 08:59:02 GMT):
@NagarajGarimalla ill try running this tomorrow to see if I can figure out why this is happening again

NagarajGarimalla (Wed, 08 May 2019 08:59:27 GMT):
Thank you sir. Much obliged.

FidelGarciaGomez (Wed, 08 May 2019 09:26:06 GMT):
libindy said: 5 indy::services::ledger parse_response >>> response "{\"op\":\"REPLY\",\"result\":{\"reqId\":1557306987868408300,\"tag\":\"TAG2\",\"txnTime\":null,\"signature_type\":\"CL\",\"type\":\"108\",\"seqNo\":null,\"origin\":\"VxJQbmEbji3mB2bsCjCCQf\",\"ref\":26,\"state_proof\":{\"root_hash\":\"AFwKupveWAsFNS9gX6MJMY6717oW6n6KGnUmEWf9nsfc\",\"proof_nodes\":\"+QHH+FGAoNr+DB6BsYHEZH1nUobA8LXZjXtv/Y4njk6R+vivG4Z4gICAgICAgKDX2CxtLp3UYznk/T7Vvwoh2gC3aqee/8Q2mBUmys/1OICAgICAgID5AXGg5pFQZDAE6OlyctGe0bmAQezf5uY8jPTGli6kfOWggeiAgKCtKxjVbiNo9foeVyH/K4zLhGtcUJdbE97lAEWtjOnAz6C8F8RU7+I6W9c/olmjD/B8aTKFzBGbFRg+wtO+Kf3M5aACAl3Nzd6LaTgs48BCLhqyjJQOsXXhaAg6cx9JMIB/FqBm/GBgXNig6hWXfSiPzxOpDtrJHZJhS+c/WtIr8+mP76Bnq00w7eU8rDDLLy7g9B7emHV/2tBrg8b+jaIK8ZQymYCAoFNZD8tvj9TNXl+s1JinRkl7/pUIvHlbNLQglPQFEVSAoCRyJt6FDmJQ60JcPeVA5EXTnNrRiLBPYNq9dgI6SYlCgKBTN2yBgrZdY8SNjpbsUTdw6/M9kZMhqU9r0Pj7UP60FqD4tG6xe4ScZl2BxBLXqT5zydiIKea4Z+YINtM7LvrlOqA7Ohl8US/Ipw9m90WNb/7n5LAxzanxSmitjBCIzSGcGoA=\",\"multi_signature\":{\"participants\":[\"Node4\",\"Node2\",\"Node3\"],\"value\":{\"txn_root_hash\":\"GyVp7dG9seWvuUAYcT3ts1dB5VUF6hBBMg5Ph8mJQKEd\",\"pool_state_root_hash\":\"2C4YKAeEisEQmcridUiGDNM5985Z7PaGKqr3YN1PjcWf\",\"ledger_id\":1,\"timestamp\":1557306987,\"state_root_hash\":\"AFwKupveWAsFNS9gX6MJMY6717oW6n6KGnUmEWf9nsfc\"},\"signature\":\"RWASvUkMqDMFcRofWwaqQvhwokRXiZH8CckjuGHAeZmgHCdMdFcsKtXJPH8CD8548rJ4XGxoGomKRmF4HC1KMru9JfDu85AgbuTQaVfFCPojwaUPgXPhELBp2sehFvCjNDFcDYv9F2AGdeSzSnBgTgMzKXWpfiu8dUCbtWEKjNJhEw\"}},\"identifier\":\"Dms9TQChnKbVvHpngx2rGE\",\"data\":null}}" indy::services::ledger src/services/ledger/mod.rs 663 holder-service | libindy said: 5 indy::api::ledger prepare_result_2: >>> Err(IndyError { inner: Error("data did not match any variant of untagged enum Reply", line: 0, column: 0) holder-service | holder-service | Structure doesn't correspond to type. Most probably not found holder-service | holder-service | Item not found on ledger }) indy::api::ledger src/api/ledger.rs 891 holder-service | libindy said: 5 indy::api::ledger indy_parse_get_cred_def_response: cred_def_id: "", cred_def_json: "" indy::api::ledger src/api/ledger.rs 892 holder-service | (node:256) UnhandledPromiseRejectionWarning: IndyError: LedgerNotFound

FidelGarciaGomez (Wed, 08 May 2019 09:31:23 GMT):
Any idea? it is the part of the code when we are trying to create a credential definition from a schema. Note that, this fails from the second execution onward. The first one is ok

Zohaib_Sohail (Wed, 08 May 2019 11:26:16 GMT):
how we have to pay for did,credentials etc on sovrin network? is there any tokens or through banking channels?

Diiaablo95 (Wed, 08 May 2019 12:58:20 GMT):
Hey people! I have a question regarding Verifiable Credentials and Zero-Knowledge proofs: let's say I wanna prove to my employer that I am an employee of the company without revealing who I am. I wanna do that because I can have a 10% discount. Is it possible for me to collude with another person, who is not an employee of the same company, to pass my credentials or my proof to that person so that he/she gets the same discount?

Diiaablo95 (Wed, 08 May 2019 12:58:20 GMT):
Hey people! I have a question regarding *Verifiable Credentials and Zero-Knowledge proofs*: let's say I wanna prove to my employer that I am an employee of the company without revealing who I am. I wanna do that because I can have a 10% discount. Is it possible for me to collude with another person, who is not an employee of the same company, to pass my credentials or my proof to that person so that he/she gets the same discount?

Diiaablo95 (Wed, 08 May 2019 13:02:52 GMT):
In other words: is there a way to enforce something similar to Mandatory Access Control for credentials and derived proofs?

Diiaablo95 (Wed, 08 May 2019 13:02:52 GMT):
Hey people! A new question (I am working with Indy and underlying concepts in my M.Sc. thesis). *Is it possible to use ZKP with my identity provider, without revealing my identity?* I mean, can I prove that I am a member of that "club", for instance, without revealing what my identity is? So that I do not need to generate a new DID everytime I interact with my identity provider? Thanks! :)

FidelGarciaGomez (Wed, 08 May 2019 13:19:57 GMT):
@kdenhartog , do you have any idea to solve the error uploaded above?

FidelGarciaGomez (Wed, 08 May 2019 13:40:42 GMT):
I fixed the problem. The log was very helpful. Thank you so much

kdenhartog (Wed, 08 May 2019 16:08:18 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=tay47zhbBzwxMJrbB) @Diiaablo95 Not without also providing them with your entire identity. You must share all of your credentials or none of yours which becomes a major deterrent because that would mean sharing your medical credentials, financial credentials, etc.

jaromil (Wed, 08 May 2019 16:21:32 GMT):
how comes a single credential is dependent from others? I'm not familiar with this implementation, working with one where credentials are independent from one another.

kdenhartog (Wed, 08 May 2019 16:25:36 GMT):
We bind credentials with a link secret rather than to a DID. This means that you need to use a common link secret for all of your credentials so that you can combine the individual credentials together during a proof offer.

jaromil (Wed, 08 May 2019 16:31:48 GMT):
ACK. ingenious! is there a paper that references this? I don't understand it fully yet, while it interests me to understand indy better. what prevents me from sharing a common link secret and a single credential? is it a new computation necessary every time there is a new credential?

jaromil (Wed, 08 May 2019 16:35:45 GMT):
is the link secret common to all DIDs? a crypto paper would help

esplinr (Wed, 08 May 2019 18:08:22 GMT):
https://sovrin.org/issue-credentials/

kdenhartog (Wed, 08 May 2019 18:10:39 GMT):
https://docs.google.com/document/d/1pODumTPhScMTgEi1HHXjlMvQCn_U25vbyduC78aiqGA/edit

kdenhartog (Wed, 08 May 2019 18:12:52 GMT):
If you're looking for full on math, check out this paper https://github.com/hyperledger/indy-crypto/blob/master/libindy-crypto/docs/AnonCred.pdf

jaromil (Wed, 08 May 2019 18:51:05 GMT):
I was looking for the math, thanks! will spend some time on it, looks like its still a work in progress, in particular I miss references and bibliography

jaromil (Wed, 08 May 2019 18:56:38 GMT):
I'm fascinated by the challenge. AFAIK the other way to avoid credential trading is biometrics.

dbluhm (Wed, 08 May 2019 19:03:38 GMT):
Indy Agent call happening now! Feel free to join us: Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/856588081 Meeting Agenda and Notes: https://wiki.hyperledger.org/display/indy/Indy+Agent+Working+Group Or iPhone one-tap : US: +16699006833,,856588081# or +16465588656,,856588081# Or Telephone: Dial(for higher quality, dial a number based on your current location): US: +1 669 900 6833 or +1 646 558 8656 Meeting ID: 856 588 081 International numbers available: https://zoom.us/u/akZ4IVIpQ

Beluosa (Wed, 08 May 2019 21:38:29 GMT):
Has joined the channel.

kdenhartog (Thu, 09 May 2019 07:48:11 GMT):
just pushed an update that got the getting started guide working locally and it should also work in browser too.

Switch2Logic (Thu, 09 May 2019 07:59:59 GMT):
Good day, I am having a little trouble understanding some concept about DID's. I understand each party has a public DID. and then they share a private relational DID. so they both can identify each other. From the documentation it seems like the public DID's are stored on the public ledger. When party would like to connect to another they do so with their public did? I am confused with the initial interaction just before they exchange private DISs of ledger... Could both parties not expose each others public DIDS? Does this matter?

Switch2Logic (Thu, 09 May 2019 07:59:59 GMT):
Good day, I am having a little trouble understanding some concept about DID's. I understand each party has a public DID. and then they share a private relational DID. so they both can identify each other. From the documentation it seems like the public DID's are stored on the public ledger. When party would like to connect to another they do so with their public did? I am confused with the initial interaction just before they exchange private DIDs of ledger... Could both parties not expose each others public DIDS? Does this matter?

Switch2Logic (Thu, 09 May 2019 08:03:47 GMT):
Who adds public DIDs into the ledger?

Switch2Logic (Thu, 09 May 2019 08:14:03 GMT):
Would you give each Agent a public DID?

NagarajGarimalla (Thu, 09 May 2019 10:46:30 GMT):
Thank you. This works for me now. I am going to proceed with the next steps. How do I create new anchors and new credentials? Is there a guide for that or can you share the steps please.

jljordan_bcgov (Thu, 09 May 2019 14:15:38 GMT):
There are a few docs that may help ... I think you are considering the initial invitation process, a good question and an interesting problem that is context dependent. Are the parties known to either other, in physical proximity, is it a well known digital service, and other such considerations. https://openssi.github.io/peer-did-method-spec/index.html#cross-registration https://github.com/hyperledger/indy-hipe/blob/b0708395fd1669df33a9619efa7770a20c97006e/text/0003-did-comm/README.md

kdenhartog (Thu, 09 May 2019 14:55:30 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=zDx8MxJ9d93CZJrzK) @NagarajGarimalla It's all included in the how to guides

esplinr (Thu, 09 May 2019 14:59:34 GMT):
Evernym's latest sprint demo introduces Indy's support for enforcing Transaction Author Agreements: https://www.youtube.com/watch?v=q-lB6Z3eE7o&list=PLRp0viTDxBWGLdZk0aamtahB9cpJGV7ZF

nage (Thu, 09 May 2019 20:09:02 GMT):
Today's WG call has been uploaded here https://drive.google.com/open?id=1AwHWN95KmSEi5fijraID0tFFMzYHoMwt

BillBarnhill (Thu, 09 May 2019 22:28:48 GMT):
Has joined the channel.

drummondreed (Fri, 10 May 2019 05:05:25 GMT):
It's because when you use ZKP (specifically Camenisch-Lysyanskaya proofs), all of your creds are tied to the same underlying link secret.

pknowles (Fri, 10 May 2019 05:27:22 GMT):
Google is looking to add Electronic ID support so developers can build mobile apps that can be securely used as an ID. https://venturebeat.com/2019/05/09/google-is-bringing-electronic-ids-to-android/

iamtxena (Fri, 10 May 2019 07:15:20 GMT):
Has joined the channel.

SatheeshNehru (Mon, 13 May 2019 06:15:42 GMT):
1) is there any doc on how cl signature works? 2)to disclose specific set of attribute does the holder need to contact issuer??

AbhishekAadi (Mon, 13 May 2019 06:59:53 GMT):
Has joined the channel.

AbhishekAadi (Mon, 13 May 2019 06:59:55 GMT):
Hello Guys, Does anybody have documents related to blockchain Security and Hyperledger Indy??

MALodder (Mon, 13 May 2019 13:58:09 GMT):
https://github.com/hyperledger/indy-hipe/blob/c761c583b1e01c1e9d3ceda2b03b35336fdc8cc1/text/anoncreds-protocol/README.md

MALodder (Mon, 13 May 2019 13:58:37 GMT):
Is there something specific that you are looking for?

AlexanderVtyurin (Mon, 13 May 2019 15:49:29 GMT):
Hello everyone. I have some problem with revocation. Use-case is: issuer issues 2 similar credentials (using the same schema, credDef and revocation registry) and verifier tries to verify both. Verification of the second credential fails if we don't revoke the first one before it. Do you have any test which cover this situation? Is it correct use-case at all (do Indy allows to perform such flow)? Thanks in advance.

tomislav (Mon, 13 May 2019 18:29:23 GMT):
You should be able to verify both credentials without revoking either of them. Are you writing the updated revocation registry to the ledger after each credential issuance?

manuvarghese (Mon, 13 May 2019 18:41:13 GMT):
Trying to set up an instance of OrgBook (VONX) and getting error in starting with initial seed.

manuvarghese (Mon, 13 May 2019 18:41:15 GMT):

Clipboard - May 13, 2019 2:41 PM

manuvarghese (Mon, 13 May 2019 18:42:08 GMT):
seems like a generic python error but unable to resolve - help ?

manuvarghese (Mon, 13 May 2019 18:42:08 GMT):
seems like a generic python error but unable to resolve - help pls ?

manuvarghese (Mon, 13 May 2019 18:50:36 GMT):
vonx

andrew.whitehead (Mon, 13 May 2019 19:51:17 GMT):
It's trying to connect to the ledger, try running a local von-network instance first (https://github.com/bcgov/von-network)

AbhishekAadi (Mon, 13 May 2019 20:36:32 GMT):
@MALodder yeah I'm searching for the proper documentation on Hyperledger Indy , just like hyperledger Fabric.

manuvarghese (Tue, 14 May 2019 02:18:42 GMT):
Just saw this. Yes, tried that & it worked - was able to set up all 3 components. Much appreciated

MALodder (Tue, 14 May 2019 02:21:02 GMT):
I'm not familiar with how Fabric does there documentation

MALodder (Tue, 14 May 2019 02:21:02 GMT):
I'm not familiar with how Fabric does their documentation

manuvarghese (Tue, 14 May 2019 02:29:22 GMT):
As of now its a blank state - can you give me pointers on how to populate the network? PS: Saw a dataLoad script in OrgBook - any other curated way to load custom data ?

andrew.whitehead (Tue, 14 May 2019 02:58:26 GMT):
You might want to try the Greenlight demo for setting up a custom issuer: https://github.com/bcgov/greenlight/blob/master/docker/VONQuickStartGuide.md

AbhishekAadi (Tue, 14 May 2019 06:33:16 GMT):
Oh.. Okay no problem. Thanks.. :relaxed:

AlexanderVtyurin (Tue, 14 May 2019 08:01:47 GMT):
Could you provide an example in any language, please? Maybe I'm doing something wrong. Yes, I'm writing a new revocation registry delta after each issuance/revocation.

AbhishekAadi (Tue, 14 May 2019 12:23:05 GMT):
Hello Guys, Does anyone having documents or video tutorials for the installation of Hyperledger Indy on Windows, Ubuntu & MAC?

luckycharms810 (Tue, 14 May 2019 15:33:09 GMT):
Has joined the channel.

vsadriano (Tue, 14 May 2019 16:50:44 GMT):
Hi! Where can I read about operating system compatibility?

vsadriano (Tue, 14 May 2019 17:01:06 GMT):
I would like to build a validator node debian sid based. Is it possible?

nicola.attico (Tue, 14 May 2019 17:23:58 GMT):
Hi @Zohaib_Sohail this was very helpful and I'm able to run it end to end. Now, is there any equivalent for nodejs? :grinning:

jadhavajay (Tue, 14 May 2019 20:00:32 GMT):
Hi all,

jadhavajay (Tue, 14 May 2019 20:03:05 GMT):
This is probably a very basic question. Can two different Indy networks talk to each other? For ex. In a scenario, an Issuer issues a credential on Network A, to the holder on Network B?

jadhavajay (Tue, 14 May 2019 20:03:21 GMT):
Is it possible?

jadhavajay (Tue, 14 May 2019 20:05:09 GMT):
Assuming that the public DIDs of these two organizations are on different network.

vindard (Tue, 14 May 2019 20:22:18 GMT):
Has joined the channel.

vindard (Tue, 14 May 2019 20:25:46 GMT):
Hey all, a quick question about creating schema definitions. Is this completely up to the discretion of whoever's making these right now, or are there a set of reference/recommended definitions for certain common types of credentials (e.g. driver's license, passport)?

dbluhm (Tue, 14 May 2019 20:41:26 GMT):
This falls more into the territory of agents. Issuers don't need to have any knowledge of Network X to issue a credential with their public DID anchored to Network Z. Holders don't need strong ties to any given Network. The greatest burden is placed on the Verifier who would have to know how to resolve a DID on the Network from which the credential is issued. This applies to much more than just Indy. The Hyperledger Aries project aims to bridge the gap between many networks by providing a generic "DID resolver" interface. Aries and Agents are under active development. A more in depth explanation of agents can be found here: https://github.com/hyperledger/indy-hipe/tree/master/text/0002-agents

jadhavajay (Tue, 14 May 2019 20:46:53 GMT):
Thanks @dbluhm for the explanation. I will check the details in Aries.

dbluhm (Tue, 14 May 2019 20:56:13 GMT):
To this point (and others may need to correct me), I think schema definitions have organically been added according to the issuers' needs. Over time I think the vision is that, in similar fashion to establishing which DIDs are accepted issuers of credential X, trust frameworks will delineate what schema's ought to be used for a driver's license, passport, etc.

jwow (Wed, 15 May 2019 00:56:25 GMT):
Has joined the channel.

jwow (Wed, 15 May 2019 00:56:26 GMT):
daniel

jwow (Wed, 15 May 2019 00:59:17 GMT):
My goal (however quixotic) is to develop a discoverable and extensible taxonomy of credentials that permits a common semantic meaning despite spelling or even language differences.

jwow (Wed, 15 May 2019 01:05:28 GMT):
daniel

athulramesh (Wed, 15 May 2019 11:34:30 GMT):
Has joined the channel.

Unni_1994 (Wed, 15 May 2019 11:58:14 GMT):
Hi all , How the steward is added in the indy network for the first time?

dbluhm (Wed, 15 May 2019 13:58:29 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=vporAGWvv96axoj7Q) @jwow That sounds a lot like the schema overlays @pknowles has been working on

pknowles (Wed, 15 May 2019 14:15:40 GMT):
*ODCA paper* on medium.com - https://medium.com/@paul.knowles_52509/overlays-data-capture-architecture-odca-providing-a-standardized-global-solution-for-data-caeb1679137a

vindard (Wed, 15 May 2019 14:35:34 GMT):
@dbluhm ok yea that makes sense. I suspected as much, thanks

vindard (Wed, 15 May 2019 14:36:17 GMT):
@jwow oh that sounds promising! Will keep an eye out for that body of work

dbluhm (Wed, 15 May 2019 14:37:44 GMT):
This was mentioned in another thread but just to make sure @vindard sees it too: you should look into @pknowles work on schema overlays which sounds very similar to @jwow's goal

vindard (Wed, 15 May 2019 14:38:44 GMT):
Thanks for the heads up. I actually just caught that medium post you shared (looking into it now as well)

vindard (Wed, 15 May 2019 14:39:40 GMT):
I also saw mention of some "Schema 2.0" work a little higher up in one of these channels. Would that also factor in to what we're trying to accomplish here? (.. and how would things like schema.org fit into all of this?)

dbluhm (Wed, 15 May 2019 14:44:53 GMT):
I'm probably not the best person to answer that question (@kenebert can give a better explanation) but Schema 2.0 is the effort for supporting more complex data structures in a way that still allows selective disclosure using zero knowledge proofs. Part of that is explicitly defining the data type of a schema attribute. I think schema overlays and schema 2.0 have a lot to say about the other but I haven't been involved in those conversations. I believe @kenebert, @pknowles and others meet regularly to discuss their work.

dbluhm (Wed, 15 May 2019 14:46:30 GMT):
Schema 2.0 would allow usage of schema.org schemas? I'm really getting out of my area of expertise lol

dbluhm (Wed, 15 May 2019 14:46:30 GMT):
I think Schema 2.0 would allow usage of schema.org schemas? I'm really getting out of my area of expertise lol

vindard (Wed, 15 May 2019 15:12:07 GMT):
That's fine lol. The direction you're providing is still really helpful, much appreciated :pray:

pknowles (Wed, 15 May 2019 17:20:55 GMT):
@dbluhm Nothing you said is inaccurate :slight_smile: !!! I can confirm that @kenebert and I catchup every fortnight to discuss both the ODCA and the Rich Schema work. At some stage, I envisage these initiatives will align.

pknowles (Wed, 15 May 2019 17:20:55 GMT):
@dbluhm Nothing you said is inaccurate :slight_smile: !!! I can confirm that the #indy-semantics WG attendees catchup every fortnight to discuss both the ODCA and the Rich Schema work. Both @kenebert and I are engaged on those calls. At some point in the near future, I envisage these initiatives will align.

mauricio (Thu, 16 May 2019 03:06:51 GMT):
Has left the channel.

RameshT (Thu, 16 May 2019 12:19:56 GMT):
Has joined the channel.

kenebert (Thu, 16 May 2019 16:50:12 GMT):
I agree with pknowles on the eventual alignment of the overlays with the schema work.

axel (Thu, 16 May 2019 17:10:30 GMT):
hi everyone, on the call Sean mentioned the European Identity and Cloud conference as being next week

axel (Thu, 16 May 2019 17:10:38 GMT):
but according to the website I'm finding (https://www.kuppingercole.com/events/eic2019) it seems it's happening right now (14 - 17 May)

axel (Thu, 16 May 2019 17:10:45 GMT):
am I getting this wrong?

dbluhm (Thu, 16 May 2019 17:55:25 GMT):
@axel everything I'm seeing is pointing to those dates as well; looks like @Sean_Bohan may have misspoke or was referring to hearing back from people attending EIC in next weeks Indy WG call

esplinr (Thu, 16 May 2019 19:02:27 GMT):
I believe that he was referring to hearing back from conference attendees.

esplinr (Thu, 16 May 2019 19:03:02 GMT):
Indy Node currently only supports Ubuntu 18.04. If you get it working somewhere else, it would be great if you could submit a pull request with the changes and instructions.

axel (Thu, 16 May 2019 22:03:25 GMT):
makes sense then, thanks

mgbailey (Thu, 16 May 2019 22:18:59 GMT):
16.04, that is

esplinr (Thu, 16 May 2019 22:43:57 GMT):
Yes, @mgbailey is correct. Indy Node currently only compiles on Ubuntu 16.04. (Thanks for fixing that!)

Sean_Bohan (Fri, 17 May 2019 00:27:58 GMT):
@dbluhm is right - I thought I had said on the call I would have someone next week give us a report from EIC

gsantos (Fri, 17 May 2019 13:15:01 GMT):
Has joined the channel.

gsantos (Fri, 17 May 2019 13:15:14 GMT):
Hi all, how can I get the schemas that are written on the ledger without having their id's? That is, if one entity writes a schema, how can another entity have access to it without knowing it's id?

swcurran (Fri, 17 May 2019 14:21:12 GMT):
@gsantos - discovery is not yet built into the ledger. Current solutions would be the following. For the Issuer to publish what credentials they are issuing (schema and credential defs) in some way (e.g. tell the other party). Ledger browsers have been built that monitor the ledger, load the transactions into a database for discovery (this has been done a couple of times - this one https://indyscan.io/home/SOVRIN_MAINNET and this BC Gov one: https://sovrin-mainnet-browser.vonx.io/browse/domain). With these you can do searching. In the future, I would expect there will be registries of these.

swcurran (Fri, 17 May 2019 14:21:12 GMT):
@gsantos - discovery is not yet built into the ledger. Current solutions would be the following. For the Issuer to publish what credentials they are issuing (schema and credential defs) in some way (e.g. tell the other party). Ledger browsers have been built that monitor the ledger, load the transactions into a database for discovery (this has been done a couple of times - this one https://indyscan.io/home/SOVRIN_MAINNET and this BC Gov one: https://sovrin-mainnet-browser.vonx.io/browse/domain). With these you can do searching. In the future, I would expect there will be registries of schema and more ledger query tools for discovery.

gsantos (Fri, 17 May 2019 14:30:50 GMT):
@swcurran thank you very much! :)

kclinux (Fri, 17 May 2019 14:42:18 GMT):
Hi team, I would like to hear from your view about Microsoft's recently announced Identity Overlay Network (ION), which is using Sidetree protocol with bitcoin network as the trust anchor. Any comparison between Indy's approach and Sidetree's? Thanks.

daidoji (Fri, 17 May 2019 14:48:25 GMT):
yeah I saw that this morning, might be too early to have digested it though. didn't it just drop today? https://news.slashdot.org/story/19/05/16/1926237/microsoft-launches-decentralized-identity-tool-on-bitcoin-blockchain

kdenhartog (Fri, 17 May 2019 14:48:26 GMT):
One of the biggest difference to the two is that with ION you would have probablistic finality of the DID, where as with Indy you get deterministic finality. More realistically what this means is that if a DID gets updated during block X you should probably wait 6 blocks to make sure the changes don't get reorged out. There's only limited number of edge cases where this actually could be a concern. For the most part, you can trust it once it's hit the mempool. Another difference is that with a specific implementation of Indy (Sovrin) there's governance layers that exist on top that are more familiar to business organizations.

kdenhartog (Fri, 17 May 2019 14:50:26 GMT):
Some other difference is that Indy supports ZKPs out of the box, where as ION doesn't have support for this right now. I suspect someone will come around and build things like that in the future. The implications of this is that Indy can support issuing to holders that don't require an anchored DID.

kdenhartog (Fri, 17 May 2019 14:51:48 GMT):
Meaning if I want to get a credential as a holder, I would need to get a DID anchored to the ledger so that the issuer can issue to a specific credential. This is considered less private because it can become a globally correlatable identifier

kclinux (Fri, 17 May 2019 14:54:06 GMT):
Thanks @kdenhartog ! I don't mean to create big competition. But it's good if Hyperledger Indy (and Sovrin) can develop some material addressing this. It is some very important points that people should know.

kdenhartog (Fri, 17 May 2019 15:06:50 GMT):
Definitely need material on it, but given things are shifting so quickly it won't remain relevant long.

jwow (Sat, 18 May 2019 05:51:56 GMT):
That's an excellent concept but from my experience I see a number of deficiencies, especially in creating a schema that is extensible, discoverable and support multi-language. Additionally, the conditional use statements as shown are very limiting. I have a number of other concerns but suspect a lot of others really want to get down to the nitty-gritty of this. I wonder what is the best avenue for a deeper & richer discussion?

pknowles (Sat, 18 May 2019 06:33:40 GMT):
@jwow Head over to #indy-semantics for a deeper discussion. All of the main ODCA contributors are members of that channel. There is also a fortnightly *Indy Semantics WG* call every other Tuesday for semantics development discussions. The published *ODCA paper* only contains the basic elements of the architecture and the concept. There are a number of other parts that are already fairly strong conceptually that are not in the paper including Industry classification code tagging for data object searchability and personal data processing/consent association to published schema families. ODCA has been in constant development by members of the community for over a year now so the conceptual architecture is fairly mature now. ODCA really deserves a book rather than a simple paper! :slight_smile:

pknowles (Sat, 18 May 2019 06:33:40 GMT):
@jwow Head over to #indy-semantics for a deeper discussion. All of the main ODCA contributors are members of that channel. There is also a fortnightly *Indy Semantics WG* call every other Tuesday for semantics development discussions. The published *ODCA paper* only contains the basic elements of the architecture and the concept. There are a number of other parts that are already fairly strong conceptually that are not in the paper including Industry classification code tagging for data object searchability and personal data processing/consent association to published schema families. ODCA has been in constant development by members of the community for over a year now so the conceptual architecture is fairly mature now. ODCA really deserves a book rather than just a simple paper! :slight_smile:

pknowles (Sat, 18 May 2019 06:33:40 GMT):
@jwow Head over to #indy-semantics for a deeper discussion. All of the main ODCA contributors are members of that channel. There is also a fortnightly *Indy Semantics WG* call every other Tuesday for semantics development discussions. The published *ODCA paper* only contains the basic building blocks of the architecture and the concept. There are a number of other parts that are already fairly strong conceptually that are not in the paper including Industry classification code tagging for data object searchability and personal data processing/consent association to published schema families. ODCA has been in constant development by members of the community for over a year now so the conceptual architecture is fairly mature now. ODCA really deserves a book rather than just a simple paper! :slight_smile:

pknowles (Sat, 18 May 2019 06:33:40 GMT):
@jwow Head over to #indy-semantics for a deeper discussion. All of the main ODCA contributors are members of that channel. There is also a fortnightly *Indy Semantics WG* call every other Tuesday for semantics development discussions. The published *ODCA paper* only contains the basic building blocks of the architecture and the concept. There are a number of other parts that are already fairly strong conceptually that are not in the paper including Industry classification code tagging for data object searchability and personal data processing/consent association to published schema families. ODCA has been in constant development by members of the community for over a year so the conceptual architecture is fairly mature now. ODCA really deserves a book rather than just a simple paper! :slight_smile:

circlespainter (Sat, 18 May 2019 07:34:51 GMT):
Has joined the channel.

PhilippH (Sat, 18 May 2019 16:58:15 GMT):
Has joined the channel.

PhilippH (Sat, 18 May 2019 17:16:16 GMT):
Hi everyone, Your expert opinion is needed! I am very interested in self-sovereign identity and therefore investigating this topic as part of my bachelor's thesis. I would like to invite you to take part in my survey, which you can reach via the following link: https://ubayreuthmarketing.qualtrics.com/jfe/form/SV_39O8XuUAgYEUWDH I know how important your time is - so I highly appreciate your participation. The survey is computer and mobile compatible and takes about 5 minutes. I am happy about everyone who participates and supports my research! Thank you in advance

PhilippH (Sat, 18 May 2019 17:16:16 GMT):
Hi everyone, Your expert opinion is needed! I am very interested in self-sovereign identity and therefore investigating this topic as part of my bachelor's thesis. I would like to invite you to take part in my survey, which you can reach via the following link: https://ubayreuthmarketing.qualtrics.com/jfe/form/SV_39O8XuUAgYEUWDH I know how important your time is - so I highly appreciate your participation. The survey is computer and mobile compatible and takes about 5 minutes. I am happy about everyone who participates and supports my research! Thank you in advance

hamza8070 (Tue, 21 May 2019 21:14:47 GMT):
Has joined the channel.

hamza8070 (Tue, 21 May 2019 21:14:52 GMT):
Hi hope everyone doing good. If anyone have worth reading material on Indy sdk and its architecture kindly share with me thanks!

esplinr (Tue, 21 May 2019 21:19:17 GMT):
https://hyperledger-indy.readthedocs.io/projects/sdk/en/latest/

esplinr (Tue, 21 May 2019 21:19:40 GMT):
https://wiki.hyperledger.org/display/indy/Documentation+Index

hamza8070 (Tue, 21 May 2019 21:20:39 GMT):
Thanks espiinr sir i will read these docs.

nnao (Tue, 21 May 2019 23:08:35 GMT):
Has joined the channel.

ilya1w (Wed, 22 May 2019 03:51:20 GMT):
Has joined the channel.

vidhi24 (Wed, 22 May 2019 09:30:43 GMT):
Has joined the channel.

vidhi24 (Wed, 22 May 2019 09:30:43 GMT):
Hi i am newbie to hyperledger i was trying to connect to pool but facing problem ``` Pool "test" has not been connected Thanks,Vidhi ```

vidhi24 (Wed, 22 May 2019 10:02:00 GMT):
Getting this error while connecting to pool``` thread '' panicked at 'FIXME: IndyError { inner: Invalid argument IO error }', src/libcore/result.rs:1009:5 note: Run with `RUST_BACKTRACE=1` for a backtrace. ```

jucah (Wed, 22 May 2019 17:04:19 GMT):
Has joined the channel.

kdenhartog (Wed, 22 May 2019 17:49:13 GMT):
@vidhi24 what tests are you trying to run?

kdenhartog (Wed, 22 May 2019 17:50:47 GMT):
@esplinr and @MALodder what was our thinking with asking Indy questions on stack overflow? Should we redirect @vidhi24 question and if so where should we be redirecting people to?

MALodder (Wed, 22 May 2019 17:53:09 GMT):
Stackoverflow is fine

cam-parra (Wed, 22 May 2019 17:53:33 GMT):
Stackoverflow would be great!

kdenhartog (Wed, 22 May 2019 17:56:44 GMT):
With that, @vidhi24 can you post it on Stackoverflow and I'll answer it there? We're trying to get a knowledge base of questions built up that's indexable by google, and this is a good example of one that should be addressed there.

cam-parra (Wed, 22 May 2019 18:00:23 GMT):
Looks like there is an existing tag

cam-parra (Wed, 22 May 2019 18:00:23 GMT):
Looks like there is an existing tag for hyperledger-indy so if @vidhi24 could use that and all future questions

EdEykholt (Thu, 23 May 2019 02:51:21 GMT):
Has joined the channel.

vidhi24 (Thu, 23 May 2019 05:32:47 GMT):
okay sure @kdenhartog i was trying to build simple digital identity network

vidhi24 (Thu, 23 May 2019 05:43:06 GMT):
I have posted the question on stackoverflow under hyperledger-indy tag please answer asap...``` Link:https://stackoverflow.com/questions/56268693/cant-connect-to-pool-in-hyperledger-indy-sdk ```

pranavkirtani (Fri, 24 May 2019 05:00:49 GMT):
how scalable is indy? how many transaction per second can it handle?

EdCurran (Fri, 24 May 2019 13:10:06 GMT):
Has joined the channel.

EdCurran (Fri, 24 May 2019 13:10:07 GMT):
As a verifier I can create a predicate and the prover can prove they have a hiddden attribute that satisfies it. A predicate allows (>,<, >=, <=) operations. Can I (as verifier) create some kind of predicate, more an expression at this point, where a prover must prove the sum of two hidden fields satisfies a predicate? (Or multiplication, subtraction etc), whilst maintaining all the guarantees you would hope for? e.g. the input values are unchanged since issuance, no opportunity for the prover to fabricate the intermediate value without the corresponding (hidden) input values applied to the correct expression? From what I can tell indy can't do this at the moment, My question is more how feasible is this? My thinking is it's incredibly hard as you're in the territory of secure multiparty computation.

MALodder (Fri, 24 May 2019 16:13:57 GMT):
@EdCurran what is the semantic meaning of the two attributes being used? Is this in the context where two attributes represent monetary values and you want to prove the sum is greater than a threshold?

MALodder (Fri, 24 May 2019 16:15:09 GMT):
For the current system this isn't implemented but its something we could consider for anoncreds .0 which we are actively working on

adamdossa (Fri, 24 May 2019 19:02:58 GMT):
I had a question about the Alice, Faber, Acme, Thift example. Faber & Acme need to have their DIDs / DID Documents recorded on the Indy ledger (since they are providing attestations about Alice which need to be verifiable) but do Alice and Thift bank need to have their DIDs stored on-chain? e.g. Alice needs to generate pairwise pseudononymous DIDs to establish secure connections with Faber & Acme, but I don't believe you need to store the DID on-chain to do this (Alice can contact Faber through their endpoint which is on-chain, and then establish a secure connection). However in the example it looks like Alice's DIDs are stored on the Indy ledger. Is anyone able to clarify this?

mgbailey (Fri, 24 May 2019 19:29:24 GMT):
You are correct. Pairwise DIDs do not need to be stored on the ledger. More current implementations only share the DIDs between the connecting parties, which are responsible for storing them.

adamdossa (Fri, 24 May 2019 19:31:04 GMT):
Thanks!

EdCurran (Sat, 25 May 2019 04:28:02 GMT):
Okay let's use subtraction. If I'm a lender that gives loans I want to perform an affordability calculation. I want to know income - expentiure is greater than the required monthly payments. In the spirit of selective disclosure can I do this without revealing exact income and expenditure (no i can't at the moment)

EdCurran (Sat, 25 May 2019 04:28:02 GMT):
Okay let's use subtraction. If I'm a lender that gives loans I want to perform an affordability calculation. I want to know income - expenditure is greater than the required monthly payments. In the spirit of selective disclosure can I do this without revealing exact income and expenditure (no i can't at the moment)

AbhishekAadi (Sat, 25 May 2019 10:09:30 GMT):
Hello Guys, I want to learn about Hyperledger Indy from the scratch.. Can anybody please provide me links of the video tutorials, documents or anything related to that????????? I'll be grateful to you. Thanks.

MALodder (Sat, 25 May 2019 13:22:35 GMT):
This type of proof would be possible with the changes we are proposing for 2.0. The feature would probably go in 2.1. Cryptographically this is possible and not too difficult

kevv (Sat, 25 May 2019 14:01:51 GMT):
Has joined the channel.

EdCurran (Sat, 25 May 2019 15:01:59 GMT):
I think it would be very useful. What I think would be really cool is to create an evaluator for FEEL (see DMN: https://en.wikipedia.org/wiki/Decision_Model_and_Notation) that can be run against verifiable credentials. Including being run over hidden attributes on the provers end. From that you could execute DMN straight against credentials. You'd need some some kind of extra construct defining which issuers you accept credentials from within the DMN. Then bam easy repeatable decision making on trusted data. At the moment you could evaluate unary operators against hidden fields (which do make up a big chunk of use cases) but would have to run full expressions on the verifier side against fields revealed to you from the prover.

MALodder (Sat, 25 May 2019 15:10:15 GMT):
Yes the trust framework would define the acceptable issuers. I’ve put this feature in for 2.1

EdCurran (Sat, 25 May 2019 15:21:49 GMT):
That's excellent news thank you. Where can I read about anoncreds 2.0? One day i'll dive into how that crypto actually works :)

MALodder (Sat, 25 May 2019 17:43:18 GMT):
We’re about 85% done with the write up. I can send you the current draft if you want

hamza8070 (Sat, 25 May 2019 19:43:22 GMT):
i am following this link for learning hyperledger Indy but things are vague here https://hyperledger-indy.readthedocs.io/en/latest/

hamza8070 (Sat, 25 May 2019 19:43:52 GMT):
If anyone have more precise information and docs about Indy do share with me thanks

Sean_Bohan (Mon, 27 May 2019 18:22:02 GMT):
Gang: Community calls this week: Sovrin Crypto: Tues 5/28: 3pmBST, 10amET, 8amMT HL URSA: Weds 5/29: 3pmBST, 10amET, 8amMT Indy Semantics: Tues 5/28: 6pmBST, 1pmET, 11amMT HL Identity WG1: Weds 5/29: 5pmBST, 12nET, 10amMT Aries WG: Weds 5/29: 8pmBST, 3pmET, 1pmMT HL Identity WG2: Weds 5/29: 2amBST, 9pmET, 7pmMT Indy WG: Thurs 5/30: 4pmBST, 11amET, 9amMT Indy Maintainers: Mon 6/3 Details here: https://wiki.hyperledger.org/plugins/servlet/mobile?contentId=327683#content/view/2392132

Diiaablo95 (Tue, 28 May 2019 06:20:29 GMT):
Hey people! A new question (I am working with Indy and underlying concepts in my M.Sc. thesis).* Is it possible to use ZKP with my identity provider, without revealing my identity?* I mean, can I prove that I am a member of that "club", for instance, without revealing what my identity is? So that I do not need to generate a new DID everytime I interact with my identity provider? Thanks! 🙂

kdenhartog (Tue, 28 May 2019 06:36:35 GMT):
yes it is possible to prove possession of a credential without revealing it's data. Also, What do you mean by "do not need to generate a new DID everytime..."?

Diiaablo95 (Tue, 28 May 2019 07:00:39 GMT):
I mean exactlty that. If somehow a ZKP presented to my issuer leaks identity information, because he is my issues, then I would need to generate a new identity everytime I interact with it and link the proof to that identity instead of the original one, otherwise the issuer, also a verifier in this case, would be able to understand who I am independently from the ZKP.

Diiaablo95 (Tue, 28 May 2019 07:00:39 GMT):
I mean exactlty that. If somehow a ZKP presented to my issuer leaks identity information, because he is my issuer, then I would need to generate a new identity everytime I interact with it and link the proof to that identity instead of the original one, otherwise the issuer, also a verifier in this case, would be able to understand who I am independently from the ZKP.

Diiaablo95 (Tue, 28 May 2019 07:03:14 GMT):
I am just trying to solve this small bit, because when I told my advisor that a ZKP can be presented to the original issuer, it does not leak any information about the credential and the original DID it was issued to. But he said it was not true, and that for ZKPs to work you have to have of course the credential, but also to generate everytime a new DID so that you can present the ZKP with that DID and not the original one. And I started experiencing and existential crisis :sob:

EdCurran (Tue, 28 May 2019 09:05:58 GMT):
Yes please that would be great, thank you.

MALodder (Tue, 28 May 2019 12:09:29 GMT):
First off, issuers know to whom the credential is issued. They are attesting to that fact by agreeing to sign some attributes about them. We use what’s called a link secret to bind the credentials together so the holder can prove all of them belong to the same identity. This is never revealed to anyone so even if an issuer is compromised then you don’t have to change it. The link secret is blinded before the issuer signs it. During a proof this value also stays hidden and to a verifier it looks like a different presentation every time because proofs have randomness to them. For example, if the verifier just asked to prove they are a member of the group and nothing else, then they don’t learn anything else. That’s what’s zero knowledge means. The verifier has a very low chance of correlating you. Now if you reveal your name then correlation risks rise a lot. The point is you don’t need to change information for a value that nobody knows if the issuer is compromised. Indy doesn’t typically issue credentials to a DID, it’s to a link secret. That way holders can use the credentials across DIDs. Another piece of information that does leak during a proof presentation is who the issuers are because, the verifier needs to know which public keys to use for verification and which revocation registries to check.

DarionHernandez (Tue, 28 May 2019 12:26:11 GMT):
Has joined the channel.

Diiaablo95 (Tue, 28 May 2019 20:06:20 GMT):
Yes of course the issuer must know who it is issuing a credential to. My question concerns the next steps. If I get credential A for my DID did:sovrin:abcd, can I then create zero-knowledge proofs regarding that credential and present them to the issuer, acting as verifier, without revealing what my DID is? Because my next question would then be how can I actually present a proof without leaking my DID in a DID-based communication system such as Indy? Should I use an external communication infrastructure that does not rely on DIDs? What about a Tor network then, since typical internet-based communication leaks other type of information such as IP addresses and MAC addresses? I might have come up with a way to use ZKPs without generating new DIDs, and it relies on being able to communicate with the verifier-issuer using HTTPS (only verifier authenticates itself) over Tor network, so that proof requests and proofs are part of the HTTP header or body payloads. Does it make sense? :D

MALodder (Tue, 28 May 2019 20:38:38 GMT):
In Indy ecosystem, you have a pairwise DID to talk to each party. This makes them contextual. You do not have to use DIDs at all but they certainly help. You can do anonymous credentials without DIDs as I was saying. You can prove the link secret without revealing it each time to a verifier and to them it will look like a different person each interaction. To correlate you, they would have to use other means like behavior patterns or metadata.

MALodder (Tue, 28 May 2019 20:39:13 GMT):
If the verifier allows it, you could communicate with them over Tor and use another discovery mechanism than DIDs. Its up to you

swcurran (Tue, 28 May 2019 20:53:29 GMT):
@Diiaablo95 - in Indy DIDs and credentials are (mostly) separate. The only DID involved in a credential exchange is the DID of the issuer must be public and associated with an issued credential. Other than that, DIDs are used strictly has (as Mike says, optional) communication channels, and in general are done with non-public (not on the ledger) pairwise DIDs. That is, for every relationship you have, you create a DID and give the associated DIDDoc directly to the other party and they do the same for you. The only purpose of that process is to give each other a secure, private channel to exchange messages.

kdenhartog (Tue, 28 May 2019 21:57:42 GMT):
The problem I believe @Diiaablo95 is highlighting is that if I get a credential issued to me using DID1 with Issuer1 and then I reuse the same DID so that Issuer1 can verify me at a later date, but rather than providing all attributes in the credential I only provide proof of possession. In this case when I provide proof later, If I use DID1 again, the issuer could correlate me by use of the DID.

kdenhartog (Tue, 28 May 2019 21:58:47 GMT):
However, if I create DID2 for the verification they wouldn't be able to. Furthermore, given that the DID2 has only been used for verification, it could be used again if you only wanted to provide proof of possession at a later verification time.

kdenhartog (Tue, 28 May 2019 21:59:45 GMT):
Also, would you mind posting this question to Stack Overflow? This is a good question that will get brought up more, and I'd like to have an answer that's indexed and findable through google.

swcurran (Tue, 28 May 2019 22:01:45 GMT):
I don't think its a common use case. In today's world it's called having two logins to one service. You can do that with Indy as well. Don't know why you would but have at it. I think the only important thing out of this is the separation of DIDs and Credentials - the comms channel is not tied to the credential.

swcurran (Tue, 28 May 2019 22:01:45 GMT):
I don't think it's a common use case. In today's world it's called having two logins to one service. You can do that with Indy as well. Don't know why you would but have at it. I think the only important thing out of this is the separation of DIDs and Credentials - the comms channel is not tied to the credential.

kdenhartog (Tue, 28 May 2019 22:03:31 GMT):
I agree it's not common at this point. Given we trust the issuer to issue a credential, I'm not sure how valuable reducing the correlation would be for something like that. However, it's possible to resolve by creating only 1 new DID for verification.

MALodder (Tue, 28 May 2019 22:08:50 GMT):
First of all, you’re forgetting that the DIDs can be blinded before being signed therefore no correlation from that will happen. Correlation will happen via other methods

kdenhartog (Tue, 28 May 2019 22:57:39 GMT):
Can you explain what you mean? What I mean is if I setup a channel for communication to receive a credential and then reuse that channel to verify the credential the DID is a point of correlation as I understand it.

AbhishekAadi (Wed, 29 May 2019 04:20:46 GMT):
Hello Guys, i have a doubt regarding the interoperability of blockchain platforms. Can we merge Hyperledger Fabric, Sawtooth & Indy together?

iamShashvat (Wed, 29 May 2019 06:12:21 GMT):
Has joined the channel.

iamShashvat (Wed, 29 May 2019 06:12:22 GMT):
Hey! I have one doubt take a scenario in which: Alice has the credentials from Faber college(which is digitally signed by Faber college to proof their Identity) according to the transcript schema, let say college provide the required 5 values, and after getting credential it stored in the wallet, some times later Alice is applying for a new job and have to provide her educational proof, so Can she use same credentials provided by Faber college earlier? If yes, let say she wants to provide only 3 values (in her credentials there are 5 values currently) so how can she hide 2 values? If she tries to hide the data, the total hash value will change, which shows tamper with data. Please provide some light in this area. Resource: https://medium.com/@kctheservant/exploring-hyperledger-indy-through-indy-dev-example-10075d2547ae (For in-depth scenario Knowledge)

vidhi24 (Wed, 29 May 2019 11:57:46 GMT):
Hello guys ! I have a question on indy basically indy-cli. Is it possible to send transaction on ledger for example if I have scenario of Alice and School so is it possible to send a transaction on Alice ledger that alice has been student of school since 7years through School ?

MALodder (Wed, 29 May 2019 12:16:34 GMT):
There is no hash. We signatures that support hiding attributes and selective disclosure. CL sigs is what is currently in Indy but we plan to add BBS+ and/or Pointcheval-Saunders.

MALodder (Wed, 29 May 2019 12:16:34 GMT):
There is no hash. Indy uses signatures that support hiding attributes and selective disclosure. CL sigs is what is currently in Indy but we plan to add BBS+ and/or Pointcheval-Saunders.

josh.graber (Wed, 29 May 2019 12:46:56 GMT):
Has joined the channel.

Diiaablo95 (Wed, 29 May 2019 13:25:24 GMT):
Hey guys! Thanks all, these are all valuable and interesting inputs. My main point, as keen… was pointing out, is that DIDs in Indy are mainly used as communication channels, and hence are susceptible to the same information leak that takes place if you use the same IP address over time. No matter what info you are sending, if you use the same IP address all the different requests are linked to you as, for instance, 10.20.30.40. Now, I do not fully understand the question of a second DID for verification, especially who should create it and who should interact with it. The main reason why I am asking this question is that we have a pilot project in which we have a fleet manager and a series of electric vehicles. Upon registration, electric vehicles need to get a credential for the fleet manager saying that they are indeed customers of that fleet manager. At a later stage, whenever an electric vehicle goes to a charging station to charge, they would like to get an authorization token from the fleet manager service, and in order to do so they have to prove they are customers. Revealing the DID or somehow correlating that would make it possible for the fleet manager to (approximately) know the location of the electric vehicles, that is why de-correlation is possible. Always using new DIDs means that a new connection with the fleet manager must be established every time, and this can be time consuming. So what I would like to do is to get the initial credential from the fleet manager (the DID is not even relevant, but I honestly didn’t know it could be blinded), and after that be able, every time I need to charge my car at a charging station, proving to the fleet manager my membership in order to get the authorization token to charge my car. So how could it be prevented to correlate and de-anonymize the driver?

Diiaablo95 (Wed, 29 May 2019 13:25:24 GMT):
Hey guys! Thanks all, these are all valuable and interesting inputs. My main point, as keen… was pointing out, is that DIDs in Indy are mainly used as communication channels, and hence are susceptible to the same information leak that takes place if you use the same IP address over time. No matter what info you are sending, if you use the same IP address all the different requests are linked to you as, for instance, 10.20.30.40. Now, I do not fully understand the question of a second DID for verification, especially who should create it and who should interact with it. The main reason why I am asking this question is that we have a pilot project in which we have a fleet manager and a series of electric vehicles. Upon registration, electric vehicles need to get a credential for the fleet manager saying that they are indeed customers of that fleet manager. At a later stage, whenever an electric vehicle goes to a charging station to charge, they would like to get an authorization token from the fleet manager service, and in order to do so they have to prove they are customers. Revealing the DID or somehow correlating that would make it possible for the fleet manager to (approximately) know the location of the electric vehicles, that is why de-correlation is possible. Always using new DIDs means that a new connection with the fleet manager must be established every time, and this can be time consuming. So what I would like to do is to get the initial credential from the fleet manager (the DID is not even relevant, but I honestly didn’t know it could be blinded), and after that be able, every time I need to charge my car at a charging station, proving to the fleet manager my membership in order to get the authorization token to charge my car. So how could it be prevented to correlate and de-anonymize the driver?

Diiaablo95 (Wed, 29 May 2019 13:25:24 GMT):
Hey guys! Thanks all, these are all valuable and interesting inputs. My main point, as @kdenhartog was pointing out, is that DIDs in Indy are mainly used as communication channels, and hence are susceptible to the same information leak that takes place if you use the same IP address over time. No matter what info you are sending, if you use the same IP address all the different requests are linked to you as, for instance, 10.20.30.40. Now, I do not fully understand the question of a second DID for verification, especially who should create it and who should interact with it. The main reason why I am asking this question is that we have a pilot project in which we have a fleet manager and a series of electric vehicles. Upon registration, electric vehicles need to get a credential for the fleet manager saying that they are indeed customers of that fleet manager. At a later stage, whenever an electric vehicle goes to a charging station to charge, they would like to get an authorization token from the fleet manager service, and in order to do so they have to prove they are customers. Revealing the DID or somehow correlating that would make it possible for the fleet manager to (approximately) know the location of the electric vehicles, that is why de-correlation is possible. Always using new DIDs means that a new connection with the fleet manager must be established every time, and this can be time consuming. So what I would like to do is to get the initial credential from the fleet manager (the DID is not even relevant, but I honestly didn’t know it could be blinded), and after that be able, every time I need to charge my car at a charging station, proving to the fleet manager my membership in order to get the authorization token to charge my car. So how could it be prevented to correlate and de-anonymize the driver?

MALodder (Wed, 29 May 2019 13:56:27 GMT):
So I believe there is a disconnect between what you understand of Indy and anonymous credentials.I’ll attempt help with that. DIDs can’t be correlated directly because the idea is they are pairwise. Establishing new connections is cheap. I just create a new DID and send it to another party. New connection done secure channel setup. Now comes the issue of trust. DIDs don’t establish that, they just make sure the channel is secure. What is nice about pairwise DIDs is that Bob knows Alice as DID "abc" and Charlie knows Alice as DID "def". Unless Charlie tells Bob that Alice is also known as "def", Bob would not be able to correlate Alice by DID alone. This makes DIDs contextual. This mimics the real world in the sense that you are known by different things in different contexts like with family you are a sibling, son, grandson and work you are an employee, boss, owner. Anonymous credentials help establish trust in those contexts. This is similar to meeting someone new in person. That's the easy part. But trusting that person doesn't come until you get to know more about them. To get to know someone digitally requires trustable credentials. The two parties can share as much information as they want and they both can verify it using Indy which until now could not be done outside of special deployments. Anonymous credentials are not issued to DIDs. They are issued to an identity that lives behind multiple DIDs. Alice can use her credentials to make proofs to multiple DID channels. When a holder wants to begin receiving credentials, the holder creates a link secret that is a random number always kept private. The link secret is blinded before the issuer signs it, so the issuer never knows the true value. Even if the holder goes to the same issuer, this number appears unique every time due to how the blinding is done. The issuer signs the value without knowing its true value, called a blinded signature. The remaining attributes are known to the issuer and signed as well. To make a proof, the holder must know every attribute value and the signature. The dislcosure proof, allows Alice to choose to reveal attributes while the remaining are kept hidden. The verifier receives a proof that hidden attributes exist and Alice knows what the signed values are. So for your example of a fleet, an electric vehicle could establish a new DID with each charging station. The credential could prove its a member of the fleet without revealing which member. The only thing the charging station learns is DID B arrived at HH:mm:SS on YYYY-MM-DD and proved to be a valid fleet member and stayed for 45 minutes to charge. That's it unless the proof description requires the vehicle to reveal more information like a fleet number or vehicle number. ZKPs are not perfect in that they eliminate correlation entirely but they enhance privacy where before they had to reveal everything or implement complex methods to perform selective disclosure.

Sean_Bohan (Wed, 29 May 2019 16:23:50 GMT):
Hey everyone: Tomorrow’s INDY WG call - Thurs 5/30: 4pmBST, 11amET, 9amMT Zoom: https://zoom.us/j/232861185 Agenda: 1. Verifiable Credentials/Schema2.0 with Ken and Brent 2. DID:GIT with DaveH

drummondreed (Wed, 29 May 2019 17:05:40 GMT):
Alice will be able to us the ZKP crypto that Indy credentials use to share only the values that any particular verifier needs to know. And Alice does not need to "hide any data" to do that. Selective disclosure is the primary reason that Indy (and now Aries) is using the ZKP format for verifiable credentials.

RodrigoMedeiros (Wed, 29 May 2019 17:10:26 GMT):
Has joined the channel.

SeanBohan (Wed, 29 May 2019 17:49:06 GMT):
Has joined the channel.

iamShashvat (Thu, 30 May 2019 03:48:36 GMT):
Thanks alot!

pknowles (Thu, 30 May 2019 05:23:26 GMT):
The *Blinding Identity Taxonomy* (BIT) aims to provide a needed common standards to help protect the privacy of _personally identifiable information_ (PII) about people, organizations, or things. Note that the current version of the BIT does not identify correlation patterns. That may be considered in future versions as correlation patterns emerge. The latest version of the BIT resides with Kantara Initiative’s _Consent & Information Sharing WG_. Here is a link to the latest version of the BIT ... https://kantarainitiative.org/confluence/display/infosharing/Blinding+Identity+Taxonomy During this week’s *Hyperledger Indy Semantics WG* call, I put the spotlight on two contained elements that are integral to SSI that had been included in the taxonomy at the time of conception. Namely, … - _Self-sovereign Key Identifiers_ - _Decentralised Identifiers (DIDs)_ Now that SSI technology is more mature, I believe that those elements can be removed and replaced by the following two elements … - _Private keys / master keys_ - _Public keys / symmetric keys_ I had mentioned various DID types and DID Docs as potential BIT elements during the call but, on closer revision, I would suggest that only the above keys would actually constitute PII. I'm keen to hear people's thoughts of this.

pknowles (Thu, 30 May 2019 05:23:26 GMT):
The *Blinding Identity Taxonomy* (BIT) aims to provide a needed common standards to help protect the privacy of _personally identifiable information_ (PII) about people, organizations, or things. Note that the current version of the BIT does not identify correlation patterns. That may be considered in future versions as correlation patterns emerge. The latest version of the BIT resides with Kantara Initiative’s _Consent & Information Sharing WG_. Here is a link to the latest version ... https://kantarainitiative.org/confluence/display/infosharing/Blinding+Identity+Taxonomy During this week’s *Hyperledger Indy Semantics WG* call, I put the spotlight on two contained elements that are integral to SSI that had been included in the taxonomy at the time of conception. Namely, … - _Self-sovereign Key Identifiers_ - _Decentralised Identifiers (DIDs)_ Now that SSI technology is more mature, I believe that those elements can be removed and replaced by the following two elements … - _Private keys / master keys_ - _Public keys / symmetric keys_ I had mentioned various DID types and DID Docs as potential BIT elements during the call but, on closer revision, I would suggest that only the above keys would actually constitute PII. I'm keen to hear people's thoughts of this.

pknowles (Thu, 30 May 2019 05:23:26 GMT):
The *Blinding Identity Taxonomy* (BIT) aims to provide a needed common standards to help protect the privacy of _personally identifiable information_ (PII) about people, organizations, or things. Note that the current version of the BIT does not identify correlation patterns. Correlation may be considered in future versions as patterns emerge. The latest version of the BIT resides with Kantara Initiative’s _Consent & Information Sharing WG_. Here is a link to the latest version ... https://kantarainitiative.org/confluence/display/infosharing/Blinding+Identity+Taxonomy During this week’s *Hyperledger Indy Semantics WG* call, I put the spotlight on two contained elements that are integral to SSI that had been included in the taxonomy at the time of conception. Namely, … - _Self-sovereign Key Identifiers_ - _Decentralised Identifiers (DIDs)_ Now that SSI technology is more mature, I believe that those elements can be removed and replaced by the following two elements … - _Private keys / master keys_ - _Public keys / symmetric keys_ I had mentioned various DID types and DID Docs as potential BIT elements during the call but, on closer revision, I would suggest that only the above keys would actually constitute PII. I'm keen to hear people's thoughts of this.

pknowles (Thu, 30 May 2019 05:23:26 GMT):
The *Blinding Identity Taxonomy* (BIT) aims to provide a needed common standards to help protect the privacy of _personally identifiable information_ (PII) about people, organizations, or things. Note that the current version of the BIT does not identify correlation patterns. Correlation may be considered in future versions as patterns emerge. The latest version of the BIT resides with Kantara Initiative’s _Consent & Information Sharing WG_ ... https://kantarainitiative.org/confluence/display/infosharing/Blinding+Identity+Taxonomy During this week’s *Hyperledger Indy Semantics WG* call, I put the spotlight on two contained elements that are integral to SSI that had been included in the taxonomy at the time of conception. Namely, … - _Self-sovereign Key Identifiers_ - _Decentralised Identifiers (DIDs)_ Now that SSI technology is more mature, I believe that those elements can be removed and replaced by the following two elements … - _Private keys / master keys_ - _Public keys / symmetric keys_ I had mentioned various DID types and DID Docs as potential BIT elements during the call but, on closer revision, I would suggest that only the above keys would actually constitute PII. I'm keen to hear people's thoughts of this.

pknowles (Thu, 30 May 2019 05:23:26 GMT):
The *Blinding Identity Taxonomy* (BIT) aims to provide a needed common standards to help protect the privacy of _personally identifiable information_ (PII) about people, organizations, or things. Note that the current version of the BIT does not identify correlation patterns. Correlation may be considered in future versions as patterns emerge. The latest version of the BIT resides with Kantara Initiative’s _Consent & Information Sharing WG_ ... https://kantarainitiative.org/confluence/display/infosharing/Blinding+Identity+Taxonomy During this week’s *Hyperledger Indy Semantics WG* call, I put the spotlight on two contained elements that are integral to SSI that had been included in the taxonomy at the time of conception. Namely, … - _Self-sovereign Key Identifiers_ - _Decentralised Identifiers (DIDs)_ Now that SSI technology is more mature, I believe that those elements can be removed and replaced by the following two elements … - _Private keys / master keys_ - _Public keys / symmetric keys_ I had mentioned various DID types and DID Docs as potential BIT elements during the call but, on closer revision, I would suggest that only the above keys would actually constitute PII. I'm keen to hear people's thoughts on this.

pknowles (Thu, 30 May 2019 05:23:26 GMT):
The *Blinding Identity Taxonomy* (BIT) aims to provide a needed common standards to help protect the privacy of _personally identifiable information_ (PII) about people, organizations, or things. Note that the current version of the BIT does not identify correlation patterns. Correlation may be considered in future versions as patterns emerge. The latest version of the BIT resides with Kantara Initiative’s _Consent & Information Sharing WG_ ... https://kantarainitiative.org/confluence/display/infosharing/Blinding+Identity+Taxonomy During this week’s *Hyperledger Indy Semantics WG* call, I put the spotlight on two contained elements that are integral to SSI that had been included in the taxonomy at the time of conception. Namely, … - _Self-sovereign Key Identifiers_ - _Decentralised Identifiers (DIDs)_ Now that SSI technology is more mature, I believe that those elements can be removed and replaced by the following two elements … - _Private keys / master keys_ - _Public keys / symmetric keys_ I had mentioned various DID types and DID Docs as potential BIT elements during the call but, on closer revision, I would suggest that only the keys actually constitute PII. I'm keen to hear people's thoughts on this.

pknowles (Thu, 30 May 2019 05:23:26 GMT):
The *Blinding Identity Taxonomy* (BIT) aims to provide a needed common standards to help protect the privacy of _personally identifiable information_ (PII) about people, organizations, or things. Note that the current version of the BIT does not identify correlation patterns. Correlation may be considered in future versions as patterns emerge. The latest version of the BIT resides with Kantara Initiative’s _Consent & Information Sharing WG_ ... https://kantarainitiative.org/confluence/display/infosharing/Blinding+Identity+Taxonomy During this week’s *Hyperledger Indy Semantics WG* call, I put the spotlight on two contained elements that are integral to SSI that had been included in the taxonomy at the time of conception. Namely, … - _Self-sovereign Key Identifiers_ - _Decentralised Identifiers (DIDs)_ Now that SSI technology is more mature, I believe that those elements can be removed and replaced by the following two elements … - _Private Keys / Master Keys_ - _Public Keys / Symmetric Keys_ I had mentioned various DID types and DID Docs as potential BIT elements during the call but, on closer revision, I would suggest that only the keys actually constitute PII. I'm keen to hear people's thoughts on this.

pknowles (Thu, 30 May 2019 05:23:26 GMT):
The *Blinding Identity Taxonomy* (BIT) aims to provide a needed common standards to help protect the privacy of _personally identifiable information_ (PII) about people, organizations, or things. Note that the current version of the BIT does not identify correlation patterns. Correlation may be considered in future versions as patterns emerge. The latest version of the BIT resides with Kantara Initiative’s _Consent & Information Sharing WG_ ... https://kantarainitiative.org/confluence/display/infosharing/Blinding+Identity+Taxonomy During this week’s *Hyperledger Indy Semantics WG* call, I put the spotlight on two contained elements that are integral to SSI that had been included in the taxonomy at the time of conception. Namely, … - _Self-sovereign Key Identifiers_ - _Decentralised Identifiers (DIDs)_ Now that SSI technology is more mature, I believe that those elements can be removed and replaced by the following two elements … - _Private Keys / Master Keys_ - _Public Keys / Symmetric Keys_ I had mentioned various DID types and DID Docs as potential BIT elements during the call but, on closer revision, I would suggest that only the _keys_ actually constitute PII. I'm keen to hear people's thoughts on this.

iamShashvat (Thu, 30 May 2019 06:35:26 GMT):
Layman Question: Why do we require Agents? Don't you think increasing intermediaries(agents) will unstable the architecture?

Diiaablo95 (Thu, 30 May 2019 08:21:29 GMT):
Hey @MALodder ! Thanks, this clarifies a bit more. The only fix is that every time the user does not talk to the charging station, which are not Indy-enabled, but to the same fleet manager that issued the initial DID and the relative credential stating that the user is a customer of that fleet manager. The charging station are only able to read and validate authorization tokens, which are obtained by the user on a per-charge basis, upon verification from its fleet manager that he/she is indeed a customer. But anyways, you are saying that generating new DIDs is not time-consuming (this would happen on a recent mobile device). So how would the new connection be established in this way? Is there need for a dynamic QRCode (for instance) generated by the charging station that would allow the user to create the connection with the fleet manager? Otherwise, again, downloading a connection request from the fleet manager website, for instance, would allow the fleet manager to track which user DID created which connection with which fleet manager DID, bringing some correlation attacks, isn’t that correct? And after connection establishment, basically the fleet manager would publish a proof request and then the usual flow takes place, right?

SethiSaab (Thu, 30 May 2019 11:27:51 GMT):
Anyone customized universal resolver ?

SethiSaab (Thu, 30 May 2019 11:27:51 GMT):
Has Anyone customized universal resolver ?

MALodder (Thu, 30 May 2019 12:27:01 GMT):
Connections can be created however you want but they don’t require the fleet managers knowledge. QR codes are just one method. Basically the car creates connections with whoever it’s owner wants it can do any proofs it wants without any involvement from the fleet manager. Once the fleet manager issues the credential, there is no need to talk to him any more

ptab-pawan (Thu, 30 May 2019 13:18:09 GMT):
Has joined the channel.

ptab-pawan (Thu, 30 May 2019 13:18:45 GMT):
hi

ptab-pawan (Thu, 30 May 2019 13:18:52 GMT):
I am trying to build VCX on macOS using manual steps mentioned in Url : https://github.com/hyperledger/indy-sdk/tree/master/vcx Referring to Section: OSX To build libvcx for OSX and iOS using scripts do the following steps -- I am able to execute steps from 1 to 8. For step 9. ./mac.06.libvcx.build.sh > mac.06.libvcx.build.sh.out 2>&1 After executing above command and checking mac.06.libvcx.build.sh.out I get error : ld: symbol(s) not found for architecture x86_64 clang: error: linker command failed with exit code 1 (use -v to see invocation) error: aborting due to previous error error: Could not compile `libvcx`. So I ran mac.build.and.install.rust.tools.sh  script and executed all steps again But still I see the same error. Also In step 11. mentions a file “./mac.07.libvcx.prepare.ios.deps.sh” which doesn’t exist in macOS GitHub. Has anyone able to build VCX for macOSX and iOS ?

ptab-pawan (Thu, 30 May 2019 13:19:41 GMT):

Screen Shot 2019-05-30 at 6.36.00 PM.png

ptab-pawan (Thu, 30 May 2019 13:20:29 GMT):
I am trying to build VCX on macOS using manual steps mentioned in Url : https://github.com/hyperledger/indy-sdk/tree/master/vcx Referring to Section: OSX To build libvcx for OSX and iOS using scripts do the following steps -- I am able to execute steps from 1 to 8. For step 9. ./mac.06.libvcx.build.sh > mac.06.libvcx.build.sh.out 2>&1 After executing above command and checking mac.06.libvcx.build.sh.out I get error : ld: symbol(s) not found for architecture x86_64 clang: error: linker command failed with exit code 1 (use -v to see invocation) error: aborting due to previous error error: Could not compile `libvcx`. So I ran mac.build.and.install.rust.tools.sh  script and executed all steps again But still I see the same error. Also In step 11. mentions a file “./mac.07.libvcx.prepare.ios.deps.sh” which doesn’t exist in macOS GitHub. Has anyone able to build VCX for macOSX and iOS ?

luckycharms810 (Thu, 30 May 2019 13:45:39 GMT):
Hello, I've only recently been looking in to Indy the last few months. I am currently prototyping a multi-tenant agent. Are there any resources around proper wallet management ? Specifically, when is it appropriate to use the non_secrets portion of the wallet ? I've been looking at the example here -> https://github.com/hyperledger/indy-agent/blob/master/python/indy_sdk_utils.py and it seems to store a reverse index from the ver-key to the did in the non_secrets portion. When is it appropriate to do that vs storing metadata for each key ? Also are there conventions around metadata for public-did's, credentials, pairwise dids ?

swcurran (Thu, 30 May 2019 14:01:14 GMT):
Good questions. I believe the teams working on agents (e.g. indy-agent, indy-catalyst and I believe agent-framework) are all using non-secrets for storing relationship metadata and protocol state information in a similar way, but there has not been formal discussions on exactly what to store. This will be necessary for portability (being able to pick up your data from one agent to another), so these discussions need to happen relatively soon. @andrew.whitehead might want to weigh in here and add/correct anything here.

luckycharms810 (Thu, 30 May 2019 14:06:03 GMT):
Yep, makes sense. I definitely want to make sure that the wallets we manage are portable. Also want to make sure we arent defeating the purpose of certain aspects of DKMS by storing information in 'non_secrets'

dbluhm (Thu, 30 May 2019 14:59:24 GMT):
Just as an aside, everything stored in the `non-secrets` is still encrypted, including the tags and types used to retrieve the data (with one exception of tags explicitly marked as not encrypted).

dbluhm (Thu, 30 May 2019 14:59:24 GMT):
Just as an aside, everything stored in through the `non-secrets` API is still encrypted, including the tags and types used to retrieve the data (with one exception of tags explicitly marked as not encrypted).

dbluhm (Thu, 30 May 2019 14:59:24 GMT):
Just as an aside, everything stored through the `non-secrets` API is still encrypted, including the tags and types used to retrieve the data (with one exception of tags explicitly marked as not encrypted).

luckycharms810 (Thu, 30 May 2019 15:00:09 GMT):
Thanks, very helpful to know!

luckycharms810 (Thu, 30 May 2019 15:00:44 GMT):
i think for now I'm going to use metadata to tag my did's and use non-secrets for everything else ( requests, messages )

pknowles (Thu, 30 May 2019 16:13:20 GMT):
Courtesy of some valuable input from MALodder , a new version of the *BIT* has been submitted and subsequently accepted by Kantara Initiative's _Consent & Information Sharing WG_. The new version has been uploaded to the following HL Indy shared drive: https://drive.google.com/drive/u/0/folders/1gSD1b70OySIUKNOQTSbQ7khq9oy1V8UP

pknowles (Thu, 30 May 2019 16:13:20 GMT):
Courtesy of some valuable input from @MALodder , a new version of the *BIT* has been submitted and subsequently accepted by Kantara Initiative's _Consent & Information Sharing WG_. The new version has been uploaded to the following HL Indy shared drive: https://drive.google.com/drive/u/0/folders/1gSD1b70OySIUKNOQTSbQ7khq9oy1V8UP

pknowles (Thu, 30 May 2019 16:13:20 GMT):
Courtesy of some valuable input from @MALodder , a new version of the *BIT* has been submitted to (and, subsequently, accepted by) Kantara Initiative's _Consent & Information Sharing WG_. The new version has been uploaded to the following HL Indy shared drive: https://drive.google.com/drive/u/0/folders/1gSD1b70OySIUKNOQTSbQ7khq9oy1V8UP

pknowles (Thu, 30 May 2019 16:13:20 GMT):
Courtesy of some valuable input from @MALodder , a new version of the *BIT* has been submitted to (and subsequently accepted by) Kantara Initiative's _Consent & Information Sharing WG_. The new version has been uploaded to the following HL Indy shared drive: https://drive.google.com/drive/u/0/folders/1gSD1b70OySIUKNOQTSbQ7khq9oy1V8UP

rchristman (Thu, 30 May 2019 16:17:10 GMT):
Has joined the channel.

rchristman (Thu, 30 May 2019 16:21:39 GMT):
Does anyone have the link to the Rich Schema HIPE document that was discussed in this morning's working group call?

dbluhm (Thu, 30 May 2019 16:24:54 GMT):
https://github.com/hyperledger/indy-hipe/blob/b915e3a2426bd7d5da4982ab477f2dc4bbf1641e/text/rich-schemas/README.md

dbluhm (Thu, 30 May 2019 16:25:32 GMT):
It's in a PR against indy-hipe: https://github.com/hyperledger/indy-hipe/pull/119

luckycharms810 (Thu, 30 May 2019 17:02:21 GMT):
This might be a fairly silly question, but what is exactly the string called 'did'

luckycharms810 (Thu, 30 May 2019 17:02:51 GMT):
My understanding is that the ver key is what is used to verify the possession of the corresponding private key

luckycharms810 (Thu, 30 May 2019 17:03:12 GMT):
Is the did purely just a random string that's used as an address, and the private key never really leaves the wallet ?

luckycharms810 (Thu, 30 May 2019 17:03:34 GMT):
* or i should say it never is accessible outside the indy-sdk

dbluhm (Thu, 30 May 2019 17:16:06 GMT):
You can read up on Decentralized Identifiers in the DID spec. The abstract gives a pretty good, short description of what a DID is and why it's needed: https://w3c-ccg.github.io/did-spec/

dbluhm (Thu, 30 May 2019 17:19:12 GMT):
A simple/incomplete answer would be that a DID allows someone to maintain a digital identity through key rotations and such while also providing a spec and a method for resolving DIDs on a ledger

luckycharms810 (Thu, 30 May 2019 17:20:44 GMT):
Yep, i think i understand that portion

luckycharms810 (Thu, 30 May 2019 17:20:58 GMT):
So basically a did - allows you to query the blockchain for a ver key

luckycharms810 (Thu, 30 May 2019 17:21:19 GMT):
and verify when things are signed by the corresponding private key

luckycharms810 (Thu, 30 May 2019 17:21:44 GMT):
but when i do a list_my_dids_with_meta

luckycharms810 (Thu, 30 May 2019 17:21:49 GMT):
and see something like ``` { "metadata": null, "tempVerkey": null, "verkey": "HZEhH8inNyY25EwqHEDfKqoY5BeSL7mP4QzqgAQcRnPf", "did": "XNntkYo6mtciW2ZsgRrJ37"```

luckycharms810 (Thu, 30 May 2019 17:22:02 GMT):
is that string "did" a private key ?

dbluhm (Thu, 30 May 2019 17:24:34 GMT):
No, it is essentially just a random string that identifies the verkey, as you already seem to have figured out :slight_smile: On create_and_store_my_did, the DID is usually just the first 16 bytes (might need to verify that though it's not a super important detail) of the generated verkey

dbluhm (Thu, 30 May 2019 17:24:55 GMT):
But after key rotations and such, that can change

dbluhm (Thu, 30 May 2019 17:25:14 GMT):
(Sorry, may have misunderstood your first question)

dbluhm (Thu, 30 May 2019 17:25:14 GMT):
(Sorry, may have misunderstood your question)

dbluhm (Thu, 30 May 2019 17:26:35 GMT):
> the private key never really leaves the wallet? That is correct, the indy-sdk never exposes the private key

luckycharms810 (Thu, 30 May 2019 17:26:50 GMT):
Ah perfect, thanks!

esplinr (Thu, 30 May 2019 18:00:34 GMT):
I was discussing with @lynn.bendixsen the new auth_rules functionality. If we haven't defined any custom auth_rules, and run a get_auth_rules, it doesn't return anything. Does that mean that the default auth_rules aren't treated the same as custom auth_rules? cc @ashcherbakov @Toktar @Artemkaaas

iamShashvat (Thu, 30 May 2019 18:52:22 GMT):
With Indy infrastructure, the service endpoint for a DID will point to an agent. what if someone get the link/IP of agent, Can he eavesdrop? Is there any precautions for that?

esplinr (Thu, 30 May 2019 18:53:00 GMT):
The messages are encrypted.

iamShashvat (Thu, 30 May 2019 19:05:59 GMT):
Apart from this is there any more way to secure services endpoints?

swcurran (Thu, 30 May 2019 19:11:04 GMT):
That's a pretty deep question. Agents are just services (code), so all approaches to securing services apply. Messages are encrypted at the message protocol level, and so are not dependent on underlying transport. So basically, you have all of the existing mechanisms, plus encryption where the communicating parties have exchanged keys.

iamShashvat (Thu, 30 May 2019 19:34:53 GMT):
Thank you! But still I have some doubt. Correct me if I am wrong, I'll elaborate: let say Alice and Faber College establish the connecting using DID and Keys. Agents of Alice and Faber college makes a connection using keys. Now by some source, I got the DID Document or we can say Indy ledger's data where I found Service endpoints: https://exp/api/123231/etcetc basically it is pointing to an Alice's Agent. Dont you think If I have the link, I can access to the dycrypted data of their conversation? ------X-------X------ Possible answer-- The communication between the agents only decrypt by the private key, which Alice have in her Wallet, Is it correct? P.S. I am not the cryptography expert, it might possible I am missing out some point, Please do correct me If I am wrong. ------X------X------ Also please tell me, Can anybody be access to my DID Document? By any source, if somebody gets my DID Document, is there any problem ?

iamShashvat (Thu, 30 May 2019 19:48:36 GMT):
Thank you! But still I have some doubt. Correct me if I am wrong, I'll elaborate: let say Alice and Faber College establish the connecting using DID and Keys. Agents of Alice and Faber college makes a connection using keys. Now by some source, I got the DID Document or we can say Indy ledger's data where I found Service endpoints: https://exp/api/123231/etcetc basically it is pointing to an Alice's Agent. Dont you think If I have the link, I can access to the dycrypted data of their conversation? ------X-------X------ Possible answer-- The communication between the agents only decrypt by the private key, which Alice have in her Wallet, Is it correct? P.S. I am not the cryptography expert, it might possible I am missing out some point, Please do correct me If I am wrong. ------X------X------ Also please tell me, Can anybody be access to my DID Document? By any source, if somebody gets my DID Document, is there any problem ?

wang305305 (Thu, 30 May 2019 20:17:41 GMT):
Has joined the channel.

wang305305 (Thu, 30 May 2019 20:17:42 GMT):
Hi I'm trying to run the sample python code write_nym_and_query_verkey and I have this error: ErrorCode.PoolLedgerTimeout: 307. I followed the instruction and ran method 3 according to this page(https://github.com/hyperledger/indy-sdk/blob/master/README.md#how-to-start-local-nodes-pool-with-docker) but it still does not work. can somebody help me with it?

ptab-pawan (Thu, 30 May 2019 20:26:49 GMT):
Hi kdenhartog, Can someone assist or give a quick clue on how to build the VCX for iOS? I want to generate the VCX pod for iOS. I followed the instructions in : https://hyperledger-indy.readthedocs.io/projects/sdk/en/latest/vcx/wrappers/ios/README.html But get the below error : ld: library not found for -lvcx clang: error: linker command failed with exit code 1 (use -v to see invocation) Used git repos from : 1.https://github.com/hyperledger/indy-sdk/ 2.https://github.com/evernym/indy-sdk 3.https://github.com/evernym/sdk

ptab-pawan (Thu, 30 May 2019 20:26:49 GMT):
Hi kdenhartog, I am exploring Indy for consuming in iOS . I was able to build libIndy and libnullPay libraries. But I suppose I cant proceed without building libVCX? Can someone assist or give a quick clue on how to build the VCX for iOS? I want to generate the VCX pod for iOS. I followed the instructions in : https://hyperledger-indy.readthedocs.io/projects/sdk/en/latest/vcx/wrappers/ios/README.html But get the below error : ld: library not found for -lvcx clang: error: linker command failed with exit code 1 (use -v to see invocation) Used git repos from : 1.https://github.com/hyperledger/indy-sdk/ 2.https://github.com/evernym/indy-sdk 3.https://github.com/evernym/sdk

kdenhartog (Thu, 30 May 2019 20:47:50 GMT):
You should be able to build an iOS app independent of using libVCX. You would have to make direct calls to the indySDK rather than using libVCX. If you want something quick to experiment with take a look at https://github.com/mattrglobal/osma which doesn't use VCX or https://github.com/sovrin-foundation/connector-app/ which does. Some other options may be listed here https://github.com/hyperledger/indy-agent

wang305305 (Thu, 30 May 2019 20:49:06 GMT):
Hi kdenhartog, could you please help me with my problem?:smile:

ianco (Thu, 30 May 2019 20:51:01 GMT):
You will get this error if your indy nodes are running on a docker network (i.e. method 3) and your python script is just running on localhost

ianco (Thu, 30 May 2019 20:51:18 GMT):
Try starting up your docker network using method 1

ianco (Thu, 30 May 2019 20:51:28 GMT):
https://github.com/hyperledger/indy-sdk/blob/master/README.md#1-starting-the-test-pool-on-localhost

wang305305 (Thu, 30 May 2019 20:52:29 GMT):
yeah I tried method 1 too

wang305305 (Thu, 30 May 2019 20:52:45 GMT):
same error

ianco (Thu, 30 May 2019 20:53:21 GMT):
try "rm ~/.indy*" to clean out any old cache

ianco (Thu, 30 May 2019 20:53:56 GMT):
the error means that your script can't connect to the nodes, which could be: can't connect to the IP's, or pool config is wrong, or something like that

wang305305 (Thu, 30 May 2019 20:58:00 GMT):
do you know the equivalent command in windows?

ianco (Thu, 30 May 2019 20:58:27 GMT):
no, I'm not sure where Indy stores its files in Windows

wang305305 (Thu, 30 May 2019 20:58:44 GMT):
ok

wang305305 (Thu, 30 May 2019 21:00:07 GMT):
you said "Try starting up your docker network using method 1" but I think method 1 doesn't start a network

ianco (Thu, 30 May 2019 21:00:27 GMT):
It starts it on localhost

ianco (Thu, 30 May 2019 21:00:35 GMT):
that's the one I usually use

ptab-pawan (Thu, 30 May 2019 21:02:02 GMT):
@kdenhartog : I had a look at Sovrin foundation/connector-app, but it doesnt build due to libVCX.So you mean we can run all indy sdk functionality without libVCX ?

wang305305 (Thu, 30 May 2019 21:02:23 GMT):
ok

kdenhartog (Thu, 30 May 2019 21:02:36 GMT):
yup. LibVCX is a library built on top.

kdenhartog (Thu, 30 May 2019 21:03:13 GMT):
Ahh the classic 307 error. Almost everytime it's an issue of IP addresses in Genesis TXN files

kdenhartog (Thu, 30 May 2019 21:04:13 GMT):
What OS are you using?

wang305305 (Thu, 30 May 2019 21:04:23 GMT):
windows 10

kdenhartog (Thu, 30 May 2019 21:05:10 GMT):
The OSMA app is an example of an app built without using LibVCX. It consumes the agent framework instead. If you wanted to you could directly consume IndySDK as well.

ptab-pawan (Thu, 30 May 2019 21:06:40 GMT):
ok, thanks @kdenhartog . I am going to use libIndy wrapper for iOS and build react native app.I am aware I need to write the bindings from iOS native code to React-Native . Hope this approach will also work without libVCX

kdenhartog (Thu, 30 May 2019 21:07:29 GMT):
Can you post the output for `docker network ls`

wang305305 (Thu, 30 May 2019 21:08:26 GMT):
NETWORK ID NAME DRIVER SCOPE af3c776057fe indy_pool_network nat local 587ea48f2f66 nat nat local 3bded4185e1c none null local

iamShashvat (Thu, 30 May 2019 21:08:43 GMT):
@kdenhartog could you please help?

kdenhartog (Thu, 30 May 2019 21:09:07 GMT):
ok now run `docker network inspect af3c776057fe`

wang305305 (Thu, 30 May 2019 21:11:59 GMT):
[ { "Name": "indy_pool_network", "Id": "af3c776057fe079f2a7229bd127a4c2272c6fdeafadb16d4d6592b664da4cfd3", "Created": "2019-05-29T15:18:07.3788299-04:00", "Scope": "local", "Driver": "nat", "EnableIPv6": false, "IPAM": { "Driver": "windows", "Options": {}, "Config": [ { "Subnet": "10.0.0.0/8" } ] }, "Internal": false, "Attachable": false, "Ingress": false, "ConfigFrom": { "Network": "" }, "ConfigOnly": false, "Containers": {}, "Options": { "com.docker.network.windowsshim.hnsid": "30016A41-23ED-4C94-8263-C9CFB7FA10EA" }, "Labels": {} } ]

kdenhartog (Thu, 30 May 2019 21:14:04 GMT):
Each agent independently controls their own key. These keys are used to encrypt messages for specific agents. If the message that is encrypted for Alice's agent possesses one of the keys then she can decrypt the message sent to her endpoint. Let's say Bob messes up and sends a message to Alice's endpoint but doesn't include her key. Then in that case Alice cannot decrypt the message just like everyone inbetween Alice and Bob (think network providers like ISPs) cannot decrypt the message.

kdenhartog (Thu, 30 May 2019 21:14:35 GMT):
perfect, looks like you're running your ledger at 10.0.0.0 local ip.

kdenhartog (Thu, 30 May 2019 21:16:42 GMT):
I'm looking into what may be different about how Windows 10 uses docker.

wang305305 (Thu, 30 May 2019 21:16:53 GMT):
because in method 3, "docker network create --subnet 10.0.0.0/8 indy_pool_network"

wang305305 (Thu, 30 May 2019 21:16:54 GMT):
ok

kdenhartog (Thu, 30 May 2019 21:17:02 GMT):
The genesis txn file should be using 10.0.0.2

wang305305 (Thu, 30 May 2019 21:17:10 GMT):
yes

kdenhartog (Thu, 30 May 2019 21:19:23 GMT):
Did you do the second step of doing `docker build --build-arg pool_ip=10.0.0.2 -f ci/indy-pool.dockerfile -t indy_pool .`

wang305305 (Thu, 30 May 2019 21:20:37 GMT):
yes

wang305305 (Thu, 30 May 2019 21:20:50 GMT):
i did all 3 steps in method 3

swcurran (Thu, 30 May 2019 21:25:34 GMT):
These gets more into Agents and how they managed, but the "endpoint" in the DID is not going to be directly to Alice, but rather to an Agency operated by someone ("Agents-R-Us") that do that on behalf of many Alices. Using the "DIDComm" protocols, the Agency will know how to route the message to Alice. It works a lot like the post office - put the letter in a postbox and through a series of steps it gets to Alice's front door. Those in between can't see what's in the envelope (or even the final destination). They can just see enough to pass it along the next step. Yes, the core message is encrypted based on a private key held by Alice. Anyone can see a public DIDDocument (e.g. one on a ledger), so no problem with that. There are also private DIDs shared between two parts (Alice and Faber) and while unlikely anyone could get to those DIDDocs. Worst case would be them taking over someone's wallet and pretending to be them, but there are ways to protect against that. Again - lots of details once you get into it. Too much for rocketchat. Read in the Aries RFCs and the Indy HIPEs: https://github.com/hyperledger/indy-hipe https://github.com/hyperledger/aries-rfcs

kdenhartog (Thu, 30 May 2019 21:31:36 GMT):
Hmm I'm not entirely sure then. See if you can get this to work https://github.com/kdenhartog/indy-dev

kdenhartog (Thu, 30 May 2019 21:33:07 GMT):
I work on mac normally, so I'm not always as familiar with the Windows based approaches. I can tell you with certainty though that it has something to do with how the IndySDK is finding the nodes and what IP the nodes are running on.

wang305305 (Thu, 30 May 2019 21:46:25 GMT):
ok thanks

iamShashvat (Fri, 31 May 2019 02:58:24 GMT):
Thank you Soo much swcurran! I'll clear my concepts asap and start contributing soon.

ptab-pawan (Fri, 31 May 2019 05:38:33 GMT):
@kdenhartog , Why do I need agent framework for building iOS/ android mobile app? If required which agent framework can be used for iOS and android ?

kdenhartog (Fri, 31 May 2019 06:24:19 GMT):
you don't consuming IndySDK directly works just as well. In terms of why you might want to consider something like it though would be to get quick support of protocols that other agents are supporting. However, if you want to write your own protocol stuff over time using the SDK that works just as well. Are you looking to build a native iOS app or use something like react native?

kdenhartog (Fri, 31 May 2019 06:24:19 GMT):
you don't. consuming IndySDK directly works just as well. In terms of why you might want to consider something like it though would be to get quick support of protocols that other agents are supporting. However, if you want to write your own protocol stuff over time using the SDK that works just as well. Are you looking to build a native iOS app or use something like react native?

ptab-pawan (Fri, 31 May 2019 06:25:58 GMT):
built react native app for iOS and android. thanks for clearing the doubt. Not finding a good documentation to use libindy methods directly . DO you have a good read?

kdenhartog (Fri, 31 May 2019 06:31:36 GMT):
The how-to guides help somewhat although they are somewhat outdated. E.g. they don't show how to use pack/unpack rather than authcrypt/anoncrypt. My suggestion would be if you go that route to try and build a library independent of the app itself that handles the API calls. Then consume the library in the app. That was the thinking behind AgentFramework and libVCX because the IndySDK calls are fairly primitive compared to what makes sense in an agent. I assume you would get a few people willing to help on the library/framework if you went that route given the plethora of people who work with js.

ptab-pawan (Fri, 31 May 2019 06:44:20 GMT):
@kdenhartog : So you mean I cant write a JS bridging class that exposes indy calls to react native ? was referring to :https://jira.hyperledger.org/browse/IS-975

ptab-pawan (Fri, 31 May 2019 06:47:13 GMT):
@kdenhartog : I am not very clear on how I can consume libIndy directly. Can you please connect me to few people who can help here ? We have built a webDapp for SSI with Indy Node.JS . So the next step is to develop a mobile app which will communicate with webDapp.

kdenhartog (Fri, 31 May 2019 07:36:43 GMT):
Essentially what you would do is write a JS library that consumes the JS wrapper of IndySDK. Then you would consume that library in your react native app and your webDapp.

kdenhartog (Fri, 31 May 2019 07:46:11 GMT):
So for example one of the API calls in the JS library would be create connection. In the AgentFramework this is how they do it: https://github.com/streetcred-id/agent-framework/blob/2ad84bca8877d00d23356f970891178220446533/src/AgentFramework.Core/Contracts/IConnectionService.cs#L41

kdenhartog (Fri, 31 May 2019 07:46:11 GMT):
So for example one of the API calls in the JS library would be create connection. In the AgentFramework this is the contract for creating a connection: https://github.com/streetcred-id/agent-framework/blob/2ad84bca8877d00d23356f970891178220446533/src/AgentFramework.Core/Contracts/IConnectionService.cs#L41

kdenhartog (Fri, 31 May 2019 07:49:07 GMT):
Here's an API of how it's being done in Python too: https://github.com/bcgov/indy-catalyst/blob/eb5261cec36666e986fefcedb8cc3ac4d04ac553/agent/indy_catalyst_agent/messaging/connections/manager.py#L76

paliwalg (Fri, 31 May 2019 08:22:26 GMT):
Has joined the channel.

ptab-pawan (Fri, 31 May 2019 09:07:18 GMT):
@kdenhartog : As a first approach I had tried installing the npm module for Node SDK. But faced a lot of issues with bindings > binding.js library integrating with react native. Couldnt get it working. So took second approach to try native iOS wrappers .

kdenhartog (Fri, 31 May 2019 09:08:44 GMT):
In that case you could do everything natively in the App. You'll end up rewriting a lot of code in the two of them.

kdenhartog (Fri, 31 May 2019 09:09:13 GMT):
If this is just PoC, then that won't be too big of a deal. If you're working on a production codebase, I'd suggest finding out how to get the JS stuff working.

ptab-pawan (Fri, 31 May 2019 09:18:20 GMT):
@kdenhartog : It would be more than a POC so need it to be flexible and easily maintainable with Indy new features. Will analyse how can I get VCX or NodeJS working. As implementing agent framework would be writing all code as you explained.

axel (Fri, 31 May 2019 10:50:33 GMT):
hi everyone, sorry if this isn't a good place to ask, there was a Sovrin meet-up in London 10 days ago (https://www.eventbrite.co.uk/e/sovrin-london-meetup-3-tickets-60808777691) I was wondering if anyone went and what came out of it

axel (Fri, 31 May 2019 10:50:46 GMT):
wasn't able to make it and wanted to catch up on it

pknowles (Fri, 31 May 2019 13:27:21 GMT):
The best place to direct that question is probably at http://chat.sovrin.org

Diiaablo95 (Fri, 31 May 2019 13:27:33 GMT):
@MALodder I agree with you there's no need to, but the use case requires the user to ask for a token to its fleet manager every time he needs to charge. And in order to get the token, he needs to prove to the manager that he is a customer. Hence, everytime the user needs to create a connection witht the fleet manager and then send to proof over such connection?

MALodder (Fri, 31 May 2019 13:55:06 GMT):
To clear up a confusion with terminology, creating a connection in this space means to establish a new relationship with another party. When you say create a connection I believe you are meaning the user communicates with the fleet manager to get the token. So with that in mind, the user does not need to create a new connection but communicates with the fleet manager when he needs a token. Indeed he'll just send a proof that he's a customer and that's it. Now, if you don't want the fleet manager to have any correlation whatsoever, you do not have to this over a DID connection and can be completely anonymous. However, there is nothing that guarantees no one has tampered with the messages. Say Alice is a customer and Bob is the fleet manager. Alice can send a valid proof to Bob but without a DID connection, Mallory can change the proof to another valid one without Alice or Bob knowing. Alice's proof could be valid but Mallory sends another proof for a revoked customer. Bob won't be able to tell the difference. With a DID, Bob only knows that DID 'abc' sent a proof and needs a token. Small price to pay for security.

Diiaablo95 (Fri, 31 May 2019 14:23:27 GMT):
Yes! That is exactly what I mean. Without the establishment of a new connection between a new, ephimeral DID from user's side and the fleet manager, there is no possiblity to reliably send a proof. Exactly. That is why I was thinking to use HTTPS over Tor to achieve such functionality. In this way there is no need to create a new connection each time, but still all the guarantees offered by both Tor and HTTPS are available. Now my work could also be a performance comparison between the two options, but I think I should choose only one otherwise it would take too much resources.

Diiaablo95 (Fri, 31 May 2019 14:23:27 GMT):
Yes @MALodder ! That is exactly what I mean. Without the establishment of a new connection between a new, ephimeral DID from user's side and the fleet manager, there is no possiblity to reliably send a proof. Exactly. That is why I was thinking to use HTTPS over Tor to achieve such functionality. In this way there is no need to create a new connection each time, but still all the guarantees offered by both Tor and HTTPS are available. Now my work could also be a performance comparison between the two options, but I think I should choose only one otherwise it would take too much resources.

Diiaablo95 (Fri, 31 May 2019 14:28:31 GMT):
I also think HTTP would be too much overhead and the entire thing could be implemented at a lower level in the stack, like a secure TCP connection or using zeromq, but I do not anything about those.

MALodder (Fri, 31 May 2019 14:29:29 GMT):
Ephemeral connections are not hard to create and can be done without DIDs. HTTPS isn't usually too heavy but would require each hop to have a Certificate and be protected by TLS

axel (Fri, 31 May 2019 14:31:00 GMT):
thanks a lot!

Diiaablo95 (Fri, 31 May 2019 14:31:59 GMT):
Could you point me to some resources showing the creation of ephimeral, privacy-preserving connections? I guess they would be using the _endpoint_ part of the verinym fleet manager DDO, right?

MALodder (Fri, 31 May 2019 14:34:00 GMT):
DDO is an old term and is now called a DID Doc. If you want ephemeral connections, don't bother with DIDs. You would basically do a key exchange and have authentication from the fleet manager end then send the proof

swcurran (Fri, 31 May 2019 14:36:28 GMT):
There is an indy-hipe about ephemeral connections that Mattr has done and it has the concepts you are talking about. Might help you - I'll find the link

swcurran (Fri, 31 May 2019 14:36:57 GMT):
https://github.com/hyperledger/indy-hipe/pull/95

Diiaablo95 (Fri, 31 May 2019 14:41:44 GMT):
@MALodder that is actually a good point. My use case indeed would mostly require anoncreds and ZKPs rather than the whole Indy protocol. Surely Indy makes the entire thing easy, but there surely are libraries for just anoncreds and ZKPs. @swcurran thanks for the heads up! This means that I am not the only poor soul thinking about use cases like this :joy:

MALodder (Fri, 31 May 2019 14:42:40 GMT):
Sure you can use indy-sdk to do anoncreds and ZKPs.

Diiaablo95 (Fri, 31 May 2019 14:43:29 GMT):
Ok, I am going through all the APIs right now. So I might finally find what I am looking for. But in the meanwhile, thanks very much! :smile:

VipinB (Fri, 31 May 2019 15:22:36 GMT):
Are there open source connectors to Active directory/ LDAP/ Open ID Connect-Oauth from Indy? Do you have any github references to these? I know that "Legacy systems" are anathema and DID based solutions are the new way, we do need an adoption ramp as we bring these solutions to the Enterprise. Any response on this would be appreciated.

Diiaablo95 (Fri, 31 May 2019 16:46:22 GMT):
So, technically there is no need to create new connections each time, if users need to interact with the public identity of the fleet manager, right? I mean, users can just create a new DID every time, without creating any pairwise, and authcrypt with their single-use DID and the fleet manager public DID. What would be the the security implications of that? If we assume the fleet manager rotates the keys associated with its verinym on a regular basis?

Diiaablo95 (Fri, 31 May 2019 16:46:22 GMT):
So, technically there is no need to create new connections each time, if users need to interact with the public identity of the fleet manager, right? I mean, users can just create a new DID every time, without creating any pairwise connection, and authcrypt with their single-use DID and the fleet manager public DID. What would be the the security implications of that? If we assume the fleet manager rotates the keys associated with its verinym on a regular basis?

MALodder (Fri, 31 May 2019 16:49:05 GMT):
What you want is anoncrypt. Yes you can anoncrypt to the fleet managers public DID. The implication is controlling spam to the fleet manager as he cannot tell you is sending the message. This is usually managed by some proof of work like a captcha. Otherwise you establish a connection so the fleet manager knows who sent the message like from DID "abc".

hamza8070 (Fri, 31 May 2019 19:13:02 GMT):
Hi i have read the docs about hyperledger Indy and now can anyone share a proper use case with me which is implemented from start to end so that i relate the theory with practical example thanks.

lynn.bendixsen (Fri, 31 May 2019 20:53:13 GMT):
I am testing auth_rules and ran into a snag. Can someone help me with the constraints section of the command that will allow a STEWARD to ADD a node but only one node to a pool?

kdenhartog (Fri, 31 May 2019 23:11:30 GMT):
IDRamp has done work to address logging in to a rocketchat channel using VCs and SAML. I've seen them on the sovrin rocketchat, but I'm not sure what their handle on HL rocketchat is.

Dubh3124 (Sat, 01 Jun 2019 02:37:00 GMT):
I am trying to validate a person's DID is on the ledger before storing into a wallet, however, it seems when calling `did.key_for_did` the DID gets stored because when I call `did.store_their_did` I get Wallet item already exist error. Is validating a DID before storing the wrong way to go about it? Link to code https://codeshare.io/50Z6ln

Dubh3124 (Sat, 01 Jun 2019 02:37:00 GMT):
I am trying to validate a user's DID is on the ledger before storing into a wallet, however, it seems when calling `did.key_for_did` the DID gets stored because when I call `did.store_their_did` I get Wallet item already exist error. Is validating a DID before storing the wrong way to go about it? Link to code https://codeshare.io/50Z6ln

kdenhartog (Sat, 01 Jun 2019 21:55:22 GMT):
I had posted a comment about this on the PR, but I didn't see it within the APIs. Maybe @Artemkaaas or @sergey.minaev would be able to provide more info on it.

jljordan_bcgov (Sun, 02 Jun 2019 13:19:22 GMT):
We are also beginning to investigate this area ... as a means to jump off enterprise login. We are using Keycloak as our RBAC SSO system ...

VipinB (Sun, 02 Jun 2019 15:50:48 GMT):
Thanks guys

glotov (Mon, 03 Jun 2019 09:03:17 GMT):
Has joined the channel.

glotov (Mon, 03 Jun 2019 09:25:09 GMT):
Hi! How can I install libindy to Ubuntu 18.04 bionic? Indy-sdk's readme only refers 16.04 xenial. Should I try to build it from source for 18.04?

vardan10 (Mon, 03 Jun 2019 10:26:19 GMT):
Has joined the channel.

ivanovv (Mon, 03 Jun 2019 13:45:38 GMT):
Has joined the channel.

cam-parra (Mon, 03 Jun 2019 14:01:46 GMT):
@glotov It should work on 18,04 if you build it from source. I don't see why you couldn't

iamShashvat (Mon, 03 Jun 2019 17:00:17 GMT):
node gyp build error while installing Indy-sdk in ubuntu machine! Please Help

iamShashvat (Mon, 03 Jun 2019 17:04:26 GMT):
gyp

andrew.whitehead (Mon, 03 Jun 2019 17:24:50 GMT):
Are you trying to build the nodejs wrapper?

iamShashvat (Mon, 03 Jun 2019 18:01:40 GMT):
just tried : \

iamShashvat (Mon, 03 Jun 2019 18:01:59 GMT):
yes yes: npm install --save indy-sdk

andrew.whitehead (Mon, 03 Jun 2019 18:06:56 GMT):
Try the node-gyp install instructions, it needs to be installed globally I think: https://github.com/nodejs/node-gyp#installation

wang305305 (Mon, 03 Jun 2019 19:16:16 GMT):
Hi everyone I changed the IPs in docker_pool_transactions_genesis, but the pool.txn still regenerate the wrong ip. where does the pool.txn get the ip?

cam-parra (Mon, 03 Jun 2019 20:07:52 GMT):
@kdenhartog may know more but I believe that when the docker is spun up in that command you give it the IPs you want

wang305305 (Mon, 03 Jun 2019 20:15:00 GMT):
I did give the IP when building image and running the docker container.

cam-parra (Mon, 03 Jun 2019 20:19:08 GMT):
```generate_indy_pool_transactions --nodes 4 --clients 5 --nodeNum 1 2 3 4 --ips="$pool_ip,$pool_ip,$pool_ip,$pool_ip"```

cam-parra (Mon, 03 Jun 2019 20:19:22 GMT):
Is it this command that you changed?

wang305305 (Mon, 03 Jun 2019 20:20:16 GMT):
I changed this too. in the dockerfile

cam-parra (Mon, 03 Jun 2019 20:20:55 GMT):
Weird. Can you post the result genesis file?

wang305305 (Mon, 03 Jun 2019 20:21:19 GMT):
{"reqSignature":{},"txn":{"data":{"data":{"alias":"Node1","blskey":"4N8aUNHSgjQVgkpm8nhNEfDf6txHznoYREg9kirmJrkivgL4oSEimFF6nsQ6M41QvhM2Z33nves5vfSn9n1UwNFJBYtWVnHYMATn76vLuL3zU88KyeAYcHfsih3He6UHcXDxcaecHVz6jhCYz1P2UZn2bDVruL5wXpehgBfBaLKm3Ba","blskey_pop":"RahHYiCvoNCtPTrVtP7nMC5eTYrsUA8WjXbdhNc8debh1agE9bGiJxWBXYNFbnJXoXhWFMvyqhqhRoq737YQemH5ik9oL7R4NTTCz2LEZhkgLJzB3QRQqJyBNyv7acbdHrAT8nQ9UkLbaVL9NBpnWXBTw4LEMePaSHEw66RzPNdAX1","client_ip":"127.0.0.1","client_port":9702,"node_ip":"127.0.0.1","node_port":9701,"services":["VALIDATOR"]},"dest":"Gw6pDLhcBcoQesN72qfotTgFa7cbuqZpkX3Xo6pLhPhv"},"metadata":{"from":"Th7MpTaRZVRYnPiabds81Y"},"type":"0"},"txnMetadata":{"seqNo":1,"txnId":"fea82e10e894419fe2bea7d96296a6d46f50f93f9eeda954ec461b2ed2950b62"},"ver":"1"} {"reqSignature":{},"txn":{"data":{"data":{"alias":"Node2","blskey":"37rAPpXVoxzKhz7d9gkUe52XuXryuLXoM6P6LbWDB7LSbG62Lsb33sfG7zqS8TK1MXwuCHj1FKNzVpsnafmqLG1vXN88rt38mNFs9TENzm4QHdBzsvCuoBnPH7rpYYDo9DZNJePaDvRvqJKByCabubJz3XXKbEeshzpz4Ma5QYpJqjk","blskey_pop":"Qr658mWZ2YC8JXGXwMDQTzuZCWF7NK9EwxphGmcBvCh6ybUuLxbG65nsX4JvD4SPNtkJ2w9ug1yLTj6fgmuDg41TgECXjLCij3RMsV8CwewBVgVN67wsA45DFWvqvLtu4rjNnE9JbdFTc1Z4WCPA3Xan44K1HoHAq9EVeaRYs8zoF5","client_ip":"127.0.0.1","client_port":9704,"node_ip":"127.0.0.1","node_port":9703,"services":["VALIDATOR"]},"dest":"8ECVSk179mjsjKRLWiQtssMLgp6EPhWXtaYyStWPSGAb"},"metadata":{"from":"EbP4aYNeTHL6q385GuVpRV"},"type":"0"},"txnMetadata":{"seqNo":2,"txnId":"1ac8aece2a18ced660fef8694b61aac3af08ba875ce3026a160acbc3a3af35fc"},"ver":"1"} {"reqSignature":{},"txn":{"data":{"data":{"alias":"Node3","blskey":"3WFpdbg7C5cnLYZwFZevJqhubkFALBfCBBok15GdrKMUhUjGsk3jV6QKj6MZgEubF7oqCafxNdkm7eswgA4sdKTRc82tLGzZBd6vNqU8dupzup6uYUf32KTHTPQbuUM8Yk4QFXjEf2Usu2TJcNkdgpyeUSX42u5LqdDDpNSWUK5deC5","blskey_pop":"QwDeb2CkNSx6r8QC8vGQK3GRv7Yndn84TGNijX8YXHPiagXajyfTjoR87rXUu4G4QLk2cF8NNyqWiYMus1623dELWwx57rLCFqGh7N4ZRbGDRP4fnVcaKg1BcUxQ866Ven4gw8y4N56S5HzxXNBZtLYmhGHvDtk6PFkFwCvxYrNYjh","client_ip":"127.0.0.1","client_port":9706,"node_ip":"127.0.0.1","node_port":9705,"services":["VALIDATOR"]},"dest":"DKVxG2fXXTU8yT5N7hGEbXB3dfdAnYv1JczDUHpmDxya"},"metadata":{"from":"4cU41vWW82ArfxJxHkzXPG"},"type":"0"},"txnMetadata":{"seqNo":3,"txnId":"7e9f355dffa78ed24668f0e0e369fd8c224076571c51e2ea8be5f26479edebe4"},"ver":"1"} {"reqSignature":{},"txn":{"data":{"data":{"alias":"Node4","blskey":"2zN3bHM1m4rLz54MJHYSwvqzPchYp8jkHswveCLAEJVcX6Mm1wHQD1SkPYMzUDTZvWvhuE6VNAkK3KxVeEmsanSmvjVkReDeBEMxeDaayjcZjFGPydyey1qxBHmTvAnBKoPydvuTAqx5f7YNNRAdeLmUi99gERUU7TD8KfAa6MpQ9bw","blskey_pop":"RPLagxaR5xdimFzwmzYnz4ZhWtYQEj8iR5ZU53T2gitPCyCHQneUn2Huc4oeLd2B2HzkGnjAff4hWTJT6C7qHYB1Mv2wU5iHHGFWkhnTX9WsEAbunJCV2qcaXScKj4tTfvdDKfLiVuU2av6hbsMztirRze7LvYBkRHV3tGwyCptsrP","client_ip":"127.0.0.1","client_port":9708,"node_ip":"127.0.0.1","node_port":9707,"services":["VALIDATOR"]},"dest":"4PS3EDQ3dW1tci1Bp6543CfuuebjFrg36kLAUcskGfaA"},"metadata":{"from":"TWwCRQRZ2ZHMJFn9TzLp7W"},"type":"0"},"txnMetadata":{"seqNo":4,"txnId":"aa5e817d7cc626170eca175822029339a444eb0ee8f0bd20d3b0b76e566fb008"},"ver":"1"}

wang305305 (Mon, 03 Jun 2019 20:21:57 GMT):
I changed all IPs to 10.0.0.2

cam-parra (Mon, 03 Jun 2019 20:22:36 GMT):
Did you do that when setting up the individual nodes? Can you share link to docker image?

wang305305 (Mon, 03 Jun 2019 20:23:22 GMT):
https://github.com/hyperledger/indy-sdk/blob/master/ci/indy-pool.dockerfile

wang305305 (Mon, 03 Jun 2019 20:24:00 GMT):
I did: docker network create --subnet 10.0.0.0/8 indy_pool_network docker build --build-arg pool_ip=10.0.0.2 -f ci/indy-pool.dockerfile -t indy_pool . docker run -d --ip="10.0.0.2" --net=indy_pool_network indy_pool

cam-parra (Mon, 03 Jun 2019 20:26:45 GMT):
https://github.com/hyperledger/indy-node/blob/master/docs/source/start-nodes.md#remote-test-network-example

cam-parra (Mon, 03 Jun 2019 20:27:44 GMT):
This might help but also it might be that individual creation doesn't include those IPs so it errors and uses the default

wang305305 (Mon, 03 Jun 2019 20:31:41 GMT):
thanks I'll try it out

cam-parra (Mon, 03 Jun 2019 20:32:08 GMT):
Let me know if this works. Interested in what the problem could be

wang305305 (Mon, 03 Jun 2019 20:45:09 GMT):
I did this:

wang305305 (Mon, 03 Jun 2019 20:45:10 GMT):
RUN generate_indy_pool_transactions --nodes 4 --clients 5 --nodeNum 1 --ips="$pool_ip,$pool_ip,$pool_ip,$pool_ip" RUN generate_indy_pool_transactions --nodes 4 --clients 5 --nodeNum 2 --ips="$pool_ip,$pool_ip,$pool_ip,$pool_ip" RUN generate_indy_pool_transactions --nodes 4 --clients 5 --nodeNum 3 --ips="$pool_ip,$pool_ip,$pool_ip,$pool_ip" RUN generate_indy_pool_transactions --nodes 4 --clients 5 --nodeNum 4 --ips="$pool_ip,$pool_ip,$pool_ip,$pool_ip"

wang305305 (Mon, 03 Jun 2019 20:45:20 GMT):
still no luck

wang305305 (Mon, 03 Jun 2019 20:45:44 GMT):
pool.txn still has the wrong IP

wang305305 (Mon, 03 Jun 2019 20:46:16 GMT):
I found that it modifies the txn everytime I run the python sample

wang305305 (Mon, 03 Jun 2019 20:46:47 GMT):
write_nym_and_query_verkey.py

luckycharms810 (Tue, 04 Jun 2019 02:09:46 GMT):
What is currently seen as the standard for onboarding new users ? The method described in the HIPE here -> https://github.com/hyperledger/indy-hipe/blob/master/text/0031-connection-protocol/README.md, looks similar to what is implemented in the indy-agent. However the getting started guide (https://github.com/hyperledger/indy-sdk/blob/master/samples/python/src/getting_started.py) show's a much less sophisticated method of creating pairwise did's. It involves writing the pairwise did's to the ledger, and much smaller messages.

mboyd (Tue, 04 Jun 2019 02:51:26 GMT):
Hi @luckycharms810, The getting started guide as it currently stands is outdated. We are working to update it to the most current connection protocol as defined in the HIPE that you linked. We no longer write pairwise DID's to the ledger, which minimizes our write transactions and preserves confidentiality for peer to peer communication. There is more explanation on this communication protocol in https://github.com/hyperledger/indy-hipe/tree/master/text/0028-wire-message-format as well. Does that answer your question?

luckycharms810 (Tue, 04 Jun 2019 03:03:40 GMT):
Yep, that is quite helpful! Is there anything preventing users from using the old style of pairwise did's on the chain apart from cost ?

luckycharms810 (Tue, 04 Jun 2019 03:05:05 GMT):
My understanding is that the changes are only in the protocol, but there arent any changes in indy-node that would prevent the additional nym transactions ?

caveman7 (Tue, 04 Jun 2019 08:41:04 GMT):
Has joined the channel.

caveman7 (Tue, 04 Jun 2019 08:42:01 GMT):
hello, newbie here. is there any indy docker images available?

ibrahimel (Tue, 04 Jun 2019 08:51:16 GMT):
Has joined the channel.

YassineAmor (Tue, 04 Jun 2019 09:16:50 GMT):
Has joined the channel.

YassineAmor (Tue, 04 Jun 2019 09:19:24 GMT):
Hello, does anyone know how to go about linking HL Fabric with HL Indy? I found this project https://wiki.hyperledger.org/display/INTERN/Extending+HL+Fabric+for+connecting+with+HL+Indy and it states that writing your own custom MSP is the solution, but I have no idea how to write my own MSP. Can anybody help point me in the right direction? Thanks

swcurran (Tue, 04 Jun 2019 14:08:22 GMT):
Nothing to prevent that, but is discouraged. Either you are using the DID pairwise only but sharing it with the world, or you are using the same DID for many users - which you don't want to do for connections. So you are paying for the DIDs to be public and not getting any benefit.

DougKing (Tue, 04 Jun 2019 14:22:26 GMT):
Has joined the channel.

luckycharms810 (Tue, 04 Jun 2019 14:51:30 GMT):
Thanks @swcurran, Just trying to understand if there is a path for me to gradually make the switch in the future. Is there anywhere I can find information on cost of creating transactions on the Sovrin chain ?

swcurran (Tue, 04 Jun 2019 15:12:25 GMT):
Here is the pricing - https://sovrin.org/issue-credentials/

luckycharms810 (Tue, 04 Jun 2019 15:17:27 GMT):
Thank you!

dhuseby (Tue, 04 Jun 2019 16:08:27 GMT):
git-did

SeanBohan (Tue, 04 Jun 2019 17:51:03 GMT):
Hey everyone - if you have something to share on this Thursday’s INDY WG call, please let me know so I can add you to the agenda

rfu2k (Tue, 04 Jun 2019 20:35:55 GMT):
Hi @ptab-pawan and @kdenhartog I am also starting to try build a React Native mobile app, but for Android, and came across the post at https://jira.hyperledger.org/browse/IS-975 and I'm very pleased to find your conversation above. I am also very lost where to begin with this being new to Indy and React Native lol. @ptab-pawan I hope you will soon have good success and when things begin working for you please do a write-up or some small guide or helps for people like me and others coming behind, we will appreciate :)

lijamie (Tue, 04 Jun 2019 22:47:44 GMT):
Hello, does anyone know when I call DidSigner(seed=b'000000000000000000000000Trustee1') and create_and_store_my_did(seed="the hex seed from DidSigner"), I got the same "did" but different verkey?

lijamie (Tue, 04 Jun 2019 22:48:18 GMT):
I suspect I have two inconsistent versions of SDK and plenum. But I cant find the version compatibility document.

lijamie (Tue, 04 Jun 2019 22:48:51 GMT):
I am running plenum==1.8.1, node=1.8.0, sdk==1.8.3RC (this is from the commit log).

lijamie (Tue, 04 Jun 2019 22:48:58 GMT):
Any hint will be helpful.

lijamie (Tue, 04 Jun 2019 22:59:03 GMT):
Here is the Python code

lijamie (Tue, 04 Jun 2019 22:59:04 GMT):
import asyncio import base64 import json from indy import IndyError from indy.did import create_and_store_my_did from indy.wallet import delete_wallet, create_wallet, open_wallet, close_wallet from plenum.common.signer_did import DidSigner from stp_core.common.util import adict def test_did_signer(): signer = DidSigner(seed=b'000000000000000000000000Trustee1') trustee_def = adict() trustee_def.name = 'Hello' trustee_def.nym = signer.identifier # type: str trustee_def.verkey = signer.verkey # type: str trustee_def.seed_base64 = base64.b64encode(signer.seed).decode('utf-8') # type: str return trustee_def async def test_indy_sdk(trustee_def): wallet_id = 'EXAMPLE_WALLET' new_wallet_key = { 'key_derivation_method': 'RAW', 'key': 'yeiA3wc2tdCtyqdmvCRJxQYHmbKYHtEHbyBDssQWpWa' } try: await delete_wallet(config=json.dumps({"id": wallet_id}), credentials=json.dumps(new_wallet_key)) except IndyError: print('No wallet to delete') await create_wallet(config=json.dumps({"id": wallet_id}), credentials=json.dumps(new_wallet_key)) h_wallet = await open_wallet(config=json.dumps({"id": wallet_id}), credentials=json.dumps(new_wallet_key)) (did, verkey) = await create_and_store_my_did(h_wallet, json.dumps({'seed': trustee_def.seed_base64})) close_wallet(h_wallet) return did, verkey trustee_def = test_did_signer() loop = asyncio.get_event_loop() did, verkey = loop.run_until_complete(loop.create_task(test_indy_sdk(trustee_def))) assert trustee_def['verkey'] == verkey

lijamie (Tue, 04 Jun 2019 22:59:48 GMT):
import asyncio import base64 import json from indy import IndyError from indy.did import create_and_store_my_did from indy.wallet import delete_wallet, create_wallet, open_wallet, close_wallet from plenum.common.signer_did import DidSigner from stp_core.common.util import adict def test_did_signer(): signer = DidSigner(seed=b'000000000000000000000000Trustee1') trustee_def = adict() trustee_def.name = 'Hello' trustee_def.nym = signer.identifier # type: str trustee_def.verkey = signer.verkey # type: str trustee_def.seed_base64 = base64.b64encode(signer.seed).decode('utf-8') # type: str return trustee_def async def test_indy_sdk(trustee_def): wallet_id = 'EXAMPLE_WALLET' new_wallet_key = { 'key_derivation_method': 'RAW', 'key': 'yeiA3wc2tdCtyqdmvCRJxQYHmbKYHtEHbyBDssQWpWa' } try: await delete_wallet(config=json.dumps({"id": wallet_id}), credentials=json.dumps(new_wallet_key)) except IndyError: print('No wallet to delete') await create_wallet(config=json.dumps({"id": wallet_id}), credentials=json.dumps(new_wallet_key)) h_wallet = await open_wallet(config=json.dumps({"id": wallet_id}), credentials=json.dumps(new_wallet_key)) (did, verkey) = await create_and_store_my_did(h_wallet, json.dumps({'seed': trustee_def.seed_base64})) close_wallet(h_wallet) return did, verkey trustee_def = test_did_signer() loop = asyncio.get_event_loop() did, verkey = loop.run_until_complete(loop.create_task(test_indy_sdk(trustee_def))) assert trustee_def['verkey'] == verkey

lijamie (Tue, 04 Jun 2019 23:15:28 GMT):
If I generate my own "genesis" domain transaction file, should I use the abbrev verkey or full verkey? Why there are two forms of the verkeys and how to use them?

luckycharms810 (Wed, 05 Jun 2019 19:05:53 GMT):
Could someone help me understand how the revocation registry uses the tails_writer ?

swcurran (Wed, 05 Jun 2019 19:08:29 GMT):
Have you read the revocation HIPE? That's got a good overview.

luckycharms810 (Wed, 05 Jun 2019 19:08:54 GMT):
Nope, I can definitely do that. Appreciate the pointer

swcurran (Wed, 05 Jun 2019 19:09:16 GMT):
Stephen Klump has done a tails server and likely the most advanced implementation.

swcurran (Wed, 05 Jun 2019 19:09:19 GMT):
Hang on...

swcurran (Wed, 05 Jun 2019 19:09:56 GMT):
https://github.com/hyperledger/indy-hipe/tree/master/text/0011-cred-revocation

luckycharms810 (Wed, 05 Jun 2019 19:11:30 GMT):
Ah okay, so I guess this is using some sort of multiplicative homomorphic property ?

luckycharms810 (Wed, 05 Jun 2019 19:12:40 GMT):
I guess the part of implementation that has been more relevant to my question is a more general question about the SDK

luckycharms810 (Wed, 05 Jun 2019 19:12:54 GMT):
In the python version of the SDK, there seems to be a lot of parameters that are JSON dumped strings

luckycharms810 (Wed, 05 Jun 2019 19:13:11 GMT):
Generally these seem to contain path's to files.

luckycharms810 (Wed, 05 Jun 2019 19:14:13 GMT):
I find that this style has kind of narrowed down ways in which to implement a wrapper, especially because I cant just pass the SDK some file handles instead.

luckycharms810 (Wed, 05 Jun 2019 19:14:35 GMT):
Is there some sort of work around for this ?

SeanBohan (Wed, 05 Jun 2019 20:19:55 GMT):
Hey everyone: Tomorrow’s INDY WG call - Thurs 5/30: 4pmBST, 11amET, 9amMT Zoom: https://zoom.us/j/232861185 Agenda: 1. Audit ledger, auth_rules and token rollouts on the Sovrin BuilderNet with @lynn.bendixsen and @alexander.shcherbakov & team 2. Open Discussion

SeanBohan (Wed, 05 Jun 2019 20:19:55 GMT):
Hey everyone: Tomorrow’s INDY WG call - Thurs 6/6: 4pmBST, 11amET, 9amMT Zoom: https://zoom.us/j/232861185 Agenda: 1. Audit ledger, auth_rules and token rollouts on the Sovrin BuilderNet with @lynn.bendixsen and @alexander.shcherbakov & team 2. Open Discussion

amanji (Wed, 05 Jun 2019 20:41:20 GMT):
Has joined the channel.

luckycharms810 (Thu, 06 Jun 2019 02:57:59 GMT):
@swcurran, would love to hear your thoughts on this.

lreinink (Thu, 06 Jun 2019 10:51:49 GMT):
Has joined the channel.

lreinink (Thu, 06 Jun 2019 10:51:51 GMT):
I have a question about Indy Node. In the Bitcoin network, each node is connected to 8 peers. It can also be configured to allow for more connections. How is this done in Indy Node?

SeanBohan (Thu, 06 Jun 2019 14:14:18 GMT):
Hey everyone: In 45 minutes - INDY WG call - Thurs 6/6: 4pmBST, 11amET, 9amMT Zoom: https://zoom.us/j/232861185 Agenda: https://wiki.hyperledger.org/pages/viewpage.action?pageId=13862386

dbluhm (Thu, 06 Jun 2019 15:25:23 GMT):
@ibrahimel Welcome! As @drummondreed said, we're all really excited for your work!

drummondreed (Thu, 06 Jun 2019 15:26:07 GMT):
Super excited. I'm the co-editor of the DID spec at W3C, so if you end out having any questions about the DID spec, ping me.

drummondreed (Thu, 06 Jun 2019 15:26:52 GMT):
Also, Markus Sabadello ( @peacekeeper ) is the primary author of the DID Resolution spec, so he can help answer questions too.

swcurran (Thu, 06 Jun 2019 15:46:20 GMT):
Sorry - been offline for a bit. Stephen Klump is the best person to ask about that - he's done the best implementation. With anoncreds 2.0, the revo. method will change, and that might change this.

SeanBohan (Thu, 06 Jun 2019 16:07:40 GMT):
Hyperledger Code of Conduct https://wiki.hyperledger.org/display/HYP/Hyperledger+Code+of+Conduct Linux Foundation Antitrust Policy https://www.linuxfoundation.org/antitrust-policy/ Hyperledger Calendar of Public Meetings https://wiki.hyperledger.org/display/HYP/Calendar+of+Public+Meetings Indy Node Improvements deck: https://wiki.hyperledger.org/download/attachments/13862386/2019H1_Indy-Node-Improvements.pdf?api=v2

esplinr (Thu, 06 Jun 2019 16:10:17 GMT):
Evernym's contributions to Indy over the past two weeks: https://www.youtube.com/watch?v=_gplljzJ3Ls&list=PLRp0viTDxBWGLdZk0aamtahB9cpJGV7ZF * Hot fix for Sovrin Builder Net * Improvements to AUTH_RULES

esplinr (Thu, 06 Jun 2019 16:10:17 GMT):
Evernym's contributions to Indy over the past two weeks: https://www.youtube.com/watch?v=_gplljzJ3Ls&list=PLRp0viTDxBWGLdZk0aamtahB9cpJGV7ZF * Hot fix for Sovrin Builder Net * Improvements to AUTH_RULES * Plans for pluggable request handlers

nage (Thu, 06 Jun 2019 16:13:12 GMT):
The Call recording is uploading now

nage (Thu, 06 Jun 2019 16:14:46 GMT):
https://drive.google.com/open?id=1rXeFjUajY-kbwxN7bXnkvTQJrTGFmyy9

iamtxena (Fri, 07 Jun 2019 05:58:18 GMT):
Hi! Was this session recorded? Where can I find the recording? Thanks!

iamtxena (Fri, 07 Jun 2019 05:59:13 GMT):
Got it, two messages downs. :blush:

iamtxena (Fri, 07 Jun 2019 05:59:13 GMT):
Got it, two messages down. :blush:

esplinr (Fri, 07 Jun 2019 21:55:37 GMT):
Incidentally, the Sovrin Foundation has started posting the recordings to YouTube: https://www.youtube.com/watch?v=t2d_g3gXZjc&list=PLsbl4bRWoC_AV6E3QN1p8Kb1AukONxbSU&index=4

mboyd (Sat, 08 Jun 2019 05:54:56 GMT):
Does anyone know where Indy's most current thinking on an open credential's exchange protocol is located? Are we pretty much following the book of the [W3C verifiable credentials data model](https://w3c.github.io/vc-data-model/#use-cases-and-requirements), or are there significant differences?

TelegramSam (Sat, 08 Jun 2019 14:24:38 GMT):
https://github.com/hyperledger/aries-rfcs/tree/master/features/0036-issue-credential

TelegramSam (Sat, 08 Jun 2019 14:24:45 GMT):
https://github.com/hyperledger/aries-rfcs/tree/master/features/0037-present-proof

TelegramSam (Sat, 08 Jun 2019 14:24:51 GMT):
@mboyd ^

mboyd (Sat, 08 Jun 2019 21:34:46 GMT):
@TelegramSam thank you!

drummondreed (Sun, 09 Jun 2019 19:37:47 GMT):
FYI, there is no W3C verifiable credential exchange protocol. It's explicitly out-of-scope for the W3C Verifiable Claims Working Group. So different developer communities are creating defacto protocols. In this community it's the Aries RFCs that @TelegramSam pointed you to.

mboyd (Sun, 09 Jun 2019 19:47:26 GMT):
That makes sense. I suppose my second question should have been: Has the Indy community decided to format our credentials with the W3C data model, or is that out of the scope of the Indy project to specify?

drummondreed (Sun, 09 Jun 2019 19:49:28 GMT):
So first let me clarify that the portion of the "Indy community" working on credentials is now the "Aries community". Secondly, I believe the answer to conference with the W3C Verifiable Credential Data Model is yes, but specifically with the ZKP format. @brentzundel is the expert on the details.

mboyd (Sun, 09 Jun 2019 19:52:53 GMT):
Thanks, @drummondreed. I'll suppose I'll go pick on other channels for a bit :)

mboyd (Sun, 09 Jun 2019 19:55:07 GMT):
I haven't attended a WG call in a bit. From what I understand, Indy is now focused primarily on the distributed ledger portion (plenum, node, etc) and the "client" side work (sdk, agent, etc) is now part of Aries?

drummondreed (Sun, 09 Jun 2019 19:56:31 GMT):
Yes. An easy way to put it is the in the four-layer SSI architecture diagram (Appendix D of the Sovrin Glossary), Indy now covers layer 1 and Aries layers 2 and 3. https://docs.google.com/document/d/1gfIz5TT0cNp2kxGMLFXr19x1uoZsruUe_0glHst2fZ8/edit?pli=1#heading=h.n5k2u6fkhf74

mboyd (Sun, 09 Jun 2019 19:58:01 GMT):
Ah, wonderful. Thanks for the help

mboyd (Sun, 09 Jun 2019 20:07:59 GMT):
just updated the indy wiki with a note to help guide more devs interested in application development to the aries project. https://wiki.hyperledger.org/display/indy/Hyperledger+Indy

drummondreed (Sun, 09 Jun 2019 20:09:03 GMT):
Wise, thanks

swcurran (Sun, 09 Jun 2019 21:54:58 GMT):
@mboyd - there is a mix of contexts in that question. W3C doesn't talk about exchange protocols, just about the data model, so there is no "following the book" wrt to W3C credentials. Currently, there is some hope that the W3C data model will include elements to support ZKPs and the Indy approach to revocation, but it's not clear that will happen. Time is getting very short in the current cycle and I'm not feeling particuarly optimistic that there will be time to get those into the this pass of the W3C credentials data model.

swcurran (Sun, 09 Jun 2019 21:54:58 GMT):
@mboyd - there is a mix of contexts in that question. W3C doesn't talk about exchange protocols, just about the data model, so there is no "following the book" wrt to W3C credentials. Currently, there is some hope that the W3C data model will include elements to support ZKPs and the Indy approach to revocation, but it's not clear that will happen. However, time is getting very short in the current cycle and I'm not feeling particuarly optimistic that there will be time to get those into the this pass of the W3C credentials data model.

OmprakashSah (Mon, 10 Jun 2019 06:19:03 GMT):
Has joined the channel.

OmprakashSah (Mon, 10 Jun 2019 06:19:04 GMT):
Hello. Where can I find technical paper on Hyperledger Indy?

mboyd (Mon, 10 Jun 2019 07:05:59 GMT):
Welcome. I'd read the [Indy Walkthrough](https://github.com/hyperledger/indy-sdk/blob/master/docs/getting-started/indy-walkthrough.md) first. It describes a potential use case that Indy can accomplish. Then I would peruse the documents in these two repositories to get a better understanding of our underlying concepts: http://github.com/hyperledger/indy-hipe/ https://github.com/hyperledger/aries-rfcs/

Zohaib_Sohail (Mon, 10 Jun 2019 11:31:55 GMT):
can someone please tell me the difference between cred_schema and cred_definition and what's the need of both like can't we use just one of them by merging?

mboyd (Mon, 10 Jun 2019 13:15:14 GMT):
The short answer is that splitting up schema and credential_defs allow for better interoperable exchange and flexibility of credentials. A single schema can be used across multiple different organization's cred_defs to maximize reuse of the same data types. Multiple schemas could also be combined in one cred_def as well. You can learn more by reading [What goes on the ledger? (From Evernym, Sovrin Foundation)](https://sovrin.org/wp-content/uploads/2018/10/What-Goes-On-The-Ledger.pdf) for recent thinking on this.

luckycharms810 (Mon, 10 Jun 2019 14:17:19 GMT):
During the creation of a new did, is there a way for me change the DID string ? I'd love to be able to specify the string itself, so that I can create DID's that can partially identify the Trust Anchors. ( For e.g. I've seen an example did that starts with -> sov:did:123456.... )

swcurran (Mon, 10 Jun 2019 15:03:16 GMT):
In theory, I believe that can be done - at least for pariwise-DIDs (not sure what indy-node enforces). However, it is not considered as secure an approach. You should check the did:peer method spec: https://dhh1128.github.io/peer-did-method-spec/index.html

luckycharms810 (Mon, 10 Jun 2019 15:41:04 GMT):
Ah, i guess I'm more interested in the Public did string. So this would be more specifically for a Trust Anchor's did. If i understand correctly, trust anchors that issue credentials will have a public did. They then introduce the credential transaction using that did, and folks can be assured when verifying the credentials that they were in-fact issued by the Trust Anchor.

luckycharms810 (Mon, 10 Jun 2019 15:41:23 GMT):
Does that make sense @swcurran ?

swcurran (Mon, 10 Jun 2019 15:46:04 GMT):
For folks to be assured about (aka trust) the issuer, they need to investigate the authority of the issuer to issue credentials. They don't care that the DID is called "Harvard University" - they find out from an authoratative source that Harvard is an Accredited University and Colleges as deemed by the US Government. They request verifiable credentials to find ones that assert who the is issuer. This can happen in various ways - here is a paper we wrote a long time ago about that: https://docs.google.com/document/d/1nYq0iakgtyC21oUGWa5hLuJUoKeJFpURtGz6HcLIltY/edit

luckycharms810 (Mon, 10 Jun 2019 15:48:33 GMT):
Got it, yeah i think this would be more for the sake of convenience / for folks to gain some simple metadata by looking at the did. SInce the did string itself doesnt have any ties to the key I thought it might be a nice to have.

swcurran (Mon, 10 Jun 2019 15:49:19 GMT):
That said - I believe in indy-sdk/indy-node, the DID must be derived from the initial verkey (public key) of the keypair.

luckycharms810 (Mon, 10 Jun 2019 15:50:14 GMT):
Got it, makes sense. Thanks for a link to the paper!

luckycharms810 (Mon, 10 Jun 2019 15:51:34 GMT):
Out of curiosity, are there any other major instances of an Indy network other than Sovrin ?

swcurran (Mon, 10 Jun 2019 15:58:32 GMT):
Not production instances. I've heard of some plans, but nothing is there.

arner (Mon, 10 Jun 2019 18:59:38 GMT):
Has joined the channel.

luckycharms810 (Mon, 10 Jun 2019 23:41:50 GMT):
Are there any thoughts around maintaining an 'issuance' registry rather than a revocation registry ? I'm only curious because if your wallet is compromised and someone issues creds on your behalf, there would be no way to know that it's happening.

luckycharms810 (Mon, 10 Jun 2019 23:42:21 GMT):
I understand an issuance registry would need much more frequent updates, but if you are willing to accept a delay it could be done once a day or so.

esplinr (Tue, 11 Jun 2019 00:52:41 GMT):
There have been discussions about how to address the problem you are mentioning, but no concrete proposals as of yet.

drummondreed (Tue, 11 Jun 2019 02:05:00 GMT):
@luckycharms810 On the issue of human-friendly names for DIDs, that subject has been discussed at great length at the Rebooting the Web of Trust conferences for the past four years. It's also mentioned directly in section 9.9 of the current W3C DID spec: 9.9 The Role of Human-Friendly Identifiers DIDs achieve global uniqueness without the need for a central registration authority. This comes, however, at the cost of human memorability. The algorithms capable of generating globally unique identifiers automatically produce random strings of characters that have no human meaning. This demonstrates the axiom about identifiers known as Zooko's Triangle: "human-meaningful, decentralized, secure—pick any two". There are of course many use cases where it is desirable to discover a DID when starting from a human-friendly identifier—a natural language name, a domain name, or a conventional address for a DID controller such as a mobile telephone number, email address, Twitter handle, or blog URL. However, the problem of mapping human-friendly identifiers to DIDs (and doing so in a way that can be verified and trusted) is out-of-scope for this specification. Solutions to this problem (and there are many) should be defined in separate specifications that reference this specification. It is strongly recommended that such specifications carefully consider: (a) the numerous security attacks based on deceiving users about the true human-friendly identifier for a target entity, and (b) the privacy consequences of using human-friendly identifiers that are inherently correlatable, especially if they are globally unique.

drummondreed (Tue, 11 Jun 2019 02:05:00 GMT):
@luckycharms810 On the issue of human-friendly names for DIDs, that subject has been discussed at great length at the Rebooting the Web of Trust conferences for the past four years. It's also mentioned directly in section 9.9 of the current W3C DID spec: ``` 9.9 The Role of Human-Friendly Identifiers DIDs achieve global uniqueness without the need for a central registration authority. This comes, however, at the cost of human memorability. The algorithms capable of generating globally unique identifiers automatically produce random strings of characters that have no human meaning. This demonstrates the axiom about identifiers known as Zooko's Triangle: "human-meaningful, decentralized, secure—pick any two". There are of course many use cases where it is desirable to discover a DID when starting from a human-friendly identifier—a natural language name, a domain name, or a conventional address for a DID controller such as a mobile telephone number, email address, Twitter handle, or blog URL. However, the problem of mapping human-friendly identifiers to DIDs (and doing so in a way that can be verified and trusted) is out-of-scope for this specification. Solutions to this problem (and there are many) should be defined in separate specifications that reference this specification. It is strongly recommended that such specifications carefully consider: (a) the numerous security attacks based on deceiving users about the true human-friendly identifier for a target entity, and (b) the privacy consequences of using human-friendly identifiers that are inherently correlatable, especially if they are globally unique. ```

drummondreed (Tue, 11 Jun 2019 02:05:13 GMT):
Source: https://w3c-ccg.github.io/did-spec/

drummondreed (Tue, 11 Jun 2019 02:07:47 GMT):
There are specs under development to create cryptographically verifiable links from human-friendly names (e.g., DNS records, web pages) to DIDs and vice versa. Also, we anticipate that verifiable credentials will be used to verifiably assert the human-friendly name for a public DID. GLEIF, for example, is already working on this for businesses. https://www.gleif.org/

drummondreed (Tue, 11 Jun 2019 02:08:28 GMT):
And BC Gov's Orgbook project is setting up searchable, human-readable credential registries for finding businesses by any attribute in verifiable credentials for their business registration. https://vonx.io/

luckycharms810 (Tue, 11 Jun 2019 02:10:44 GMT):
Ah, that makes complete sense. Could definitely see how using a human friendly convention can lead to typo-squatting.

luckycharms810 (Tue, 11 Jun 2019 02:15:41 GMT):
So basically there is a need for a sort of "yellow pages".

drummondreed (Tue, 11 Jun 2019 02:16:26 GMT):
Yes, a "yellow pages" is certainly one type of solution. To be clear, we (the DID community) are NOT trying to avoid the very challenging problem of mapping human-friendly names to DIDs. We are just trying to maintain the correct architectural layering of the solution, so that we can keep those hard issues of human-friendly naming at a separate layer from DID architecture. Credit goes to Christopher Allen (founder of Rebooting the Web of Trust) for that insight—and for sticking to it over the past four years. It has provided to be very wise.

luckycharms810 (Tue, 11 Jun 2019 02:20:26 GMT):
Understood, i guess having a verified credential of identity could be one method. Although that still requires some trusted party / quorum of parties to independently offer you identifying credentials. Very helpful nonetheless

luckycharms810 (Tue, 11 Jun 2019 02:31:40 GMT):
@drummondreed, just realized you are the person who has the YT video on DKMS, and cloud agents!

luckycharms810 (Tue, 11 Jun 2019 02:32:45 GMT):
Thanks for the video, it was very informative!

mustafahusain (Tue, 11 Jun 2019 06:36:09 GMT):
Has joined the channel.

gnom1gnom (Tue, 11 Jun 2019 11:54:10 GMT):
Has joined the channel.

gnom1gnom (Tue, 11 Jun 2019 11:54:11 GMT):
Hi! Does anyone can explain how can the credential issuer request a signature from the credential owner (prover)? Is it possible to use the proverStoreCredential for this?

gnom1gnom (Tue, 11 Jun 2019 11:54:11 GMT):
Hi! Does anyone can explain how can the credential issuer request a signature from the credential owner (prover)? Is it possible to use the proverStoreCredential for this? Or should the prover_create_proof be used instead by the prover as a form of acknowledgment that the certificate was "read, understood and accepted" by the prover?

gnom1gnom (Tue, 11 Jun 2019 11:54:11 GMT):
Hi! Does anyone can explain how can the credential issuer request a signature from the credential owner (prover)? Is it possible to use the proverStoreCredential for this? Or should the prover_create_proof be used instead by the prover as a form of acknowledgment that the certificate was "read, understood and accepted" by the prover? I'm looking for something analogical to uport.requestVerificationSignature ...

swcurran (Tue, 11 Jun 2019 14:01:35 GMT):
In Indy, the Holder/Prover provides the issuer with a blinded link secret that is embedded in the credential, and that enables the proof that the credential was issued to the holder. That data is provided in the request_credential message/process. As I understand it, the link secret in Indy uses a scheme based on a Pederson Commitment. A Holder uses the same link secret for all the credentials that it holds, to show that when the claims from different credentials are proven together, they were all issued to the same entity.

swcurran (Tue, 11 Jun 2019 14:02:18 GMT):
https://en.wikipedia.org/wiki/Commitment_scheme

MALodder (Tue, 11 Jun 2019 14:26:32 GMT):
@swcurran you are correct about Pedersen commitments. It won't change for Anoncreds 2.0 either. The same blinding mechanism still applies

sukalpomitra (Wed, 12 Jun 2019 03:31:11 GMT):
Has joined the channel.

drummondreed (Wed, 12 Jun 2019 05:16:29 GMT):
@luckycharms810 My pleasure—that's why we do 'em!

SethiSaab (Wed, 12 Jun 2019 06:39:50 GMT):
Hi Team i have question

SethiSaab (Wed, 12 Jun 2019 06:40:01 GMT):
please help me to get more knowledge on this

SethiSaab (Wed, 12 Jun 2019 06:40:02 GMT):
Suppose a DID document of a device is there and this device also have verifiable credentials Now questions is ----> What kind of service does a service end point of that device offers and if a verifier has to check the authneticity of the device using service endpoint how will he because the verifiable credentials are already available and it can easily verify those by comparing signatures

SethiSaab (Wed, 12 Jun 2019 06:40:50 GMT):
what additional information we can get from service end point about authenticity of anything even when we have verifiable credentials available for that

SethiSaab (Wed, 12 Jun 2019 06:41:56 GMT):
----------------------------------------------------- question 2

SethiSaab (Wed, 12 Jun 2019 06:42:41 GMT):
as of now there is no global schema available ... suppose 2 devices are providing data to server and they are using their own key words to define data like t for temprature

SethiSaab (Wed, 12 Jun 2019 06:43:10 GMT):
is it a good practice to put these kinds of details to the metadata associated with dids or endpoints

gnom1gnom (Wed, 12 Jun 2019 08:22:04 GMT):
Perhaps I haven't made myself clear. I want to understand how should I implement following two use cases: 1. An issuer gets 5$ for every credential issued. To get this 5$ he/she needs to prove to the payer that the credential was accepted by the holder. 2. Suppose the certificate issued to a holder is a proof that the issuer and the holder signed a contract with each other. Could it be that the acceptance of the certificate by the holder is a form of digital signature? Can the issuer request that the holder digitally signs a hash of the received certificate and sends it back to the issuer? Could it be a part of the certificate issuing protocol? And finally, is there any chance that the holder could revoke this certificate without the issuer acting on his behalf?

AbhishekAadi (Wed, 12 Jun 2019 11:04:39 GMT):
Hello guys, I have a question regarding Hyperledger Indy. What is the need for Hyperledger Indy if identity management can also be done by other Hyperledger projects? Is there any special thin different in Hyperledger Indy from other Hyperledger projects?

AbhishekAadi (Wed, 12 Jun 2019 11:06:34 GMT):
Also, Can anybody share me some good questions on Hyperledger Indy that can be asked in any interview? Especially Technical questions. Thanks.

Zohaib_Sohail (Wed, 12 Jun 2019 11:40:19 GMT):
how can I find that a DID belongs to faber/alice?

Zohaib_Sohail (Wed, 12 Jun 2019 11:50:53 GMT):
Also I am getting confuse in how proof requests and validation works? currently what I understand is that verifier will verify data from credential_definitions and schemas from the ledger but how the verifier would get to know that the issuer of the credential is authenticated and trusted?

luckycharms810 (Wed, 12 Jun 2019 15:31:39 GMT):
Is there anywhere i can find some more information on DKMS. Specifically I'd love to be able to explore the underlying schema's of the sqlite wallet. Furthermore the SDK seems to only return JSON strings of the contents of the wallet. Is their a way to access the wallet directly so i can do things like paginate and order results ?

SeanBohan (Wed, 12 Jun 2019 16:35:05 GMT):
@AbhishekAadi - Identity management can be done a lot of ways, with or without Hyperledger projects. Fabric has Fabric CA (https://hyperledger-fabric-ca.readthedocs.io/en/release-1.4/) or you could build something from scratch with Sawtooth or Iroha. Of the Hyperledger Frameworks, INDY was purpose-built for identity - not supply chain, remittances, track trace or other use cases.

SeanBohan (Wed, 12 Jun 2019 16:46:24 GMT):
INDY does one thing savagely well: IDENTITY. It was purpose-built for identity use cases and designed with emerging open standards like DIDs, Verifiable Credentials, DKMS, etc. to make sure INDY can be interoperable with other decentralized identity systems regardless of what blockchain they are built on (Fabric, Ethereum, Bitcoin, etc.). An INDY implementation could be public, permissioned or public/permissioned like the Sovrin DLT and provide identities for people, organizations and things. There are a number of integrations out there that use INDY as an identity layer integrated into a larger implementation: CORDA+INDY (Tieto & Luxsoft), FETCH+INDY (Anvil), even Fabric+INDY (Chris Ferriss @ IBM mentioned opportunities for INDY as a membership plugin at the Hyperledger Hackfest AMS last year. I am not an engineer, but if I was interviewing, I would ask the following questions: Why is correlation-resistance important? How is INDY Interoperable with other Decentralized Identity efforts? How does revocation work within INDY? Where are the open standards communities where INDY is engaged? (Decentralized Identity Foundation, W3C, Rebooting Web of Trust, Oasis)

sukalpomitra (Wed, 12 Jun 2019 16:51:21 GMT):
@AbhishekAadi if I may I think this course on Indy designed by @swcurran can give you more direction to what you are looking for. "https://github.com/swcurran/education/blob/master/LFS171x/docs/introduction-to-hyperledger-indy.md"

AbhishekAadi (Wed, 12 Jun 2019 18:46:09 GMT):
Thanks @SeanBohan .. I'll definitely look on this.

AbhishekAadi (Wed, 12 Jun 2019 19:39:18 GMT):
Sure i'll go through it.. Thanks @sukalpomitra

valentin994 (Thu, 13 Jun 2019 08:20:10 GMT):
Has joined the channel.

FarhanShafiq (Thu, 13 Jun 2019 08:24:32 GMT):
Has joined the channel.

AndrewNikitin (Thu, 13 Jun 2019 09:05:14 GMT):
Has joined the channel.

iamtxena (Thu, 13 Jun 2019 16:15:27 GMT):
hey @kdenhartog , you pointed to the Rich Schema page on the indy call, could you share the link again, I cannot find it. sorry.

kdenhartog (Thu, 13 Jun 2019 18:44:26 GMT):
Here it is: https://github.com/hyperledger/indy-hipe/pull/119/files?short_path=f514c91#diff-f514c911b6ef8f09d85af0cd3c122a11

BreizhIndy (Thu, 13 Jun 2019 21:44:26 GMT):
Hello, I'm looking for hyperledger indy performance test

BreizhIndy (Thu, 13 Jun 2019 21:44:36 GMT):
do you know where I can find it

BreizhIndy (Thu, 13 Jun 2019 21:47:52 GMT):
scalability test is more accurate

BreizhIndy (Thu, 13 Jun 2019 21:47:52 GMT):
scalability test to be more accurate

SethiSaab (Fri, 14 Jun 2019 05:11:51 GMT):
Hi Guys U have a question .

SethiSaab (Fri, 14 Jun 2019 05:12:57 GMT):
As per my knowledge . when we create a DID document it gets saved in parts ... in case of Uport and BItcoin and we get block id and transaction ID . to fetch complete DID Document

SethiSaab (Fri, 14 Jun 2019 05:13:27 GMT):
but I am confused ... like when we search a DID Document using DID... .how will that work

SethiSaab (Fri, 14 Jun 2019 05:13:29 GMT):
?

SethiSaab (Fri, 14 Jun 2019 05:13:49 GMT):
Hi Guys I have a question *

SethiSaab (Fri, 14 Jun 2019 05:20:04 GMT):
I have seen that in case of DID based on bctr . it actually uses the 3rd part (xxxx-xxxx-xxxx) of did did:bctr:xxxx-xxxx-xxxx to reach at the correct block and transaction and that is how it gets the complete document but how does that works in INDY

FarhanShafiq (Fri, 14 Jun 2019 07:35:44 GMT):
Hey Guys, I'm new to indy and installing indy through docker setup given by https://github.com/hyperledger/indy-node/blob/master/environment/docker/pool/README.md but i'm facing error: generate_indy_pool_transactions: error: argument --clients: invalid int value: '127.0.0.1:2122,127.0.0.1:2123,127.0.0.1:2124,127.0.0.1:2125'

FarhanShafiq (Fri, 14 Jun 2019 07:35:44 GMT):
Hey Guys, I'm new to indy and installing indy through docker setup given by https://github.com/hyperledger/indy-node/blob/master/environment/docker/pool/README.md but facing error: generate_indy_pool_transactions: error: argument --clients: invalid int value: '127.0.0.1:2122,127.0.0.1:2123,127.0.0.1:2124,127.0.0.1:2125'

BreizhIndy (Fri, 14 Jun 2019 08:42:05 GMT):
no scalablity/performance test have been performed yet ? I can't find any

FarhanShafiq (Fri, 14 Jun 2019 09:34:45 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=rDY7WdrpxtpnZLsvL) Issue resolved by removing port from ips.

FarhanShafiq (Fri, 14 Jun 2019 09:36:28 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=y7CBAhPgY5Putqxqh) I changed ips from 10.0.0.1-4 to localhost but nodes are still running on 10.0.01-4. Is there any way these nodes on local ips ?

FarhanShafiq (Fri, 14 Jun 2019 09:38:49 GMT):
command :- ./pool_start.sh 4 127.0.0.1,127.0.0.1,127.0.0.1,127.0.0.1 nodes running status:- Node node4 started on 10.0.0.5

SethiSaab (Fri, 14 Jun 2019 12:32:29 GMT):
I have a question regarding construction of DID documents. I have seen that on some platform did documents are getting generated by extracting data from multiple transactions. like in BTCR did method and in uport did method. Could someone help to understand that how is this working?

ashcherbakov (Fri, 14 Jun 2019 12:47:33 GMT):
The DID DOC is not supported in Indy yet in the form described in https://github.com/sovrin-foundation/sovrin/blob/master/spec/did-method-spec-template.html. But this is on the Roadmap (this year). However, as of now one can achieve DID DOC behavior when saving a DID with a public key (verkey) by NYM txn, and other metadata (like `endpoints`) by ATTRIB txn (using the `raw` version). You can find more info about txn and request format here: https://github.com/hyperledger/indy-node/blob/master/docs/source/transactions.md https://github.com/hyperledger/indy-node/blob/master/docs/source/requests.md

SethiSaab (Fri, 14 Jun 2019 12:49:11 GMT):
THanks alot Sir

SethiSaab (Fri, 14 Jun 2019 12:49:45 GMT):
I have seen the platforms and they are getting complete data from different documents and merging that into 1 complete DID document

SethiSaab (Fri, 14 Jun 2019 12:49:56 GMT):
so is this doc contain info about this ?

SethiSaab (Fri, 14 Jun 2019 12:50:08 GMT):
I will be reading this to understand it properly

FarhanShafiq (Fri, 14 Jun 2019 14:03:57 GMT):
I'm new to indy and running pool using this guide: https://hyperledger-indy.readthedocs.io/projects/node/en/latest/start-nodes.html But indy-cli doesn't show any pool/open pool.

FarhanShafiq (Fri, 14 Jun 2019 14:04:21 GMT):
How to check whether pool is running or not ??

FarhanShafiq (Fri, 14 Jun 2019 14:09:47 GMT):
systemctl status indy-node stop with exit-code systemctl status indy-node � indy-node.service - Indy Node Loaded: loaded (/etc/systemd/system/indy-node.service; enabled; vendor preset : enabled) Active: activating (auto-restart) (Result: exit-code) since Fri 2019-06-14 14 :06:14 UTC; 309ms ago Process: 10192 ExecStart=/usr/bin/env python3 -O /usr/local/bin/start_indy_nod e ${NODE_NAME} ${NODE_IP} ${NODE_PORT} ${NODE_CLIENT_IP} ${NODE_CLIENT_PORT} (code=exited, status=1/FAILURE) Main PID: 10192 (code=exited, status=1/FAILURE) Jun 14 14:06:14 494d2e919c35 systemd[1]: indy-node.service: Unit entered failed state. Jun 14 14:06:14 494d2e919c35 systemd[1]: indy-node.service: Failed with result 'exit-code'.

FarhanShafiq (Fri, 14 Jun 2019 14:09:47 GMT):
> systemctl status indy-node � indy-node.service - Indy Node Loaded: loaded (/etc/systemd/system/indy-node.service; enabled; vendor preset : enabled) Active: activating (auto-restart) (Result: exit-code) since Fri 2019-06-14 14 :06:14 UTC; 309ms ago Process: 10192 ExecStart=/usr/bin/env python3 -O /usr/local/bin/start_indy_nod e ${NODE_NAME} ${NODE_IP} ${NODE_PORT} ${NODE_CLIENT_IP} ${NODE_CLIENT_PORT} (code=exited, status=1/FAILURE) Main PID: 10192 (code=exited, status=1/FAILURE) Jun 14 14:06:14 494d2e919c35 systemd[1]: indy-node.service: Unit entered failed state. Jun 14 14:06:14 494d2e919c35 systemd[1]: indy-node.service: Failed with result 'exit-code'.

luckycharms810 (Fri, 14 Jun 2019 14:21:04 GMT):
You might want to check out /var/log/indy//.log

luckycharms810 (Fri, 14 Jun 2019 14:21:38 GMT):
Generally it has to do with some missing / incorrect configuration @FarhanShafiq

FarhanShafiq (Fri, 14 Jun 2019 14:22:33 GMT):
data node1_info.json domain_transactions_genesis node1_version_info.json keys pool_transactions_genesis node1_additional_info.json don't have log extension file.

FarhanShafiq (Fri, 14 Jun 2019 14:22:33 GMT):
data node1_info.json domain_transactions_genesis node1_version_info.json keyspool_transactions_genesis node1_additional_info.json don't have log extension file.

FarhanShafiq (Fri, 14 Jun 2019 14:22:33 GMT):
data node1_info.json domain_transactions_genesis node1_version_info.json keys pool_transactions_genesis node1_additional_info.json don't have log extension file.

ashcherbakov (Fri, 14 Jun 2019 14:32:26 GMT):
Indy doesn't have a method to merge this info. But it can be done on app level.

luckycharms810 (Fri, 14 Jun 2019 14:45:18 GMT):
I think you checked "/var/lib/indy" .. you want "/var/log/indy"

BreizhIndy (Fri, 14 Jun 2019 15:54:10 GMT):
hello, where can I find info on how to perform performance test on my implementation ?

brentzundel (Fri, 14 Jun 2019 23:02:57 GMT):
Has joined the channel.

brentzundel (Fri, 14 Jun 2019 23:02:57 GMT):
The HIPE for Rich Schemas has entered the final comment period: https://github.com/hyperledger/indy-hipe/pull/119

sid555 (Sat, 15 Jun 2019 19:14:16 GMT):
Has joined the channel.

davidkhala (Sun, 16 Jun 2019 09:33:43 GMT):
Has joined the channel.

davidkhala (Sun, 16 Jun 2019 09:33:45 GMT):
Hi, Could anyone notify me which id is Nathan George?

Nawinbit (Sun, 16 Jun 2019 10:06:03 GMT):
Has joined the channel.

Nawinbit (Sun, 16 Jun 2019 10:50:51 GMT):
Hi everyOne, I am new to the Indy. I have three queries while going through the indy walkthrough doc 1. At the beginning, while faber communicates with steward, why faber needs a different DID for verinym, it can use the same DID which faber used for connection request?. Any way faber doesn't share any private Keys to the steward right?. Generating new DID and keypair, increase the wallet memory , if we going for mobile app isn't it?. 2. While sharing the proof, it is mentioned like acme raise the proof request and alice would share the credential for the proof. How this flow happening?. How alice knows that, she get the proof request?. Is the request directly comes to her wallet or is there any database is to store and process the request? 3. Upon receiving the proof, what acme will do after verification?. Is acme store the alice credential in their wallet. If yes, can Alice able to restrict the access to her credential after it was shared?.

drummondreed (Sun, 16 Jun 2019 16:07:05 GMT):
@nage. But I'm not getting a match on that right now, so something might be wrong with the Rocketchat directory.

ShrinivasDeshmukh (Mon, 17 Jun 2019 06:25:52 GMT):
Hi everyone, I want to add a new indy node to an existing network. I am using "bcgovimages/von-image:py35-1.6-8" this bcgov image. I added the pool_genesis and domain_genesis file to the new node. I generated keys for new node using init_indy_keys. But I am unable to join the new node to the network. I followed this documentation: https://indy-node.readthedocs.io/en/latest/add-node.html

andrew.whitehead (Mon, 17 Jun 2019 07:00:48 GMT):
We’ve never tested that scenario as we only use that image for testing and demo purposes, but I’d be interested to see how this can be accomplished. What are you using this deployment for?

ShrinivasDeshmukh (Mon, 17 Jun 2019 07:11:42 GMT):
I'm building a webapp through which 'TRUST_ANCHORS' can have the option to host an indy node, and join an existing network

ShrinivasDeshmukh (Mon, 17 Jun 2019 07:14:17 GMT):
I'm able to spin up the network at once, given the IPs on which nodes must be hosted. But am unable to make a node join an existing network

stefan.vogl (Mon, 17 Jun 2019 13:37:59 GMT):
hello, is there a technical description of the zero-knowledge payment protocol mentioned in the sovrin whitepaper?

FarhanShafiq (Mon, 17 Jun 2019 13:47:25 GMT):
Unable to connect indy-node with indy-cli. Is there any configuring for indy-cli to detect indy-node ? indy-node is currently working fine by checking systemctl status and node monitor status is also good. > indy> pool list There are no pools defined > indy> pool connect pool1 Pool "pool1" does not exist.

Pradeep_Pentakota (Mon, 17 Jun 2019 16:37:17 GMT):
Has joined the channel.

Silona (Mon, 17 Jun 2019 20:57:12 GMT):
no rocket chat's search is horrid...

Rick (Tue, 18 Jun 2019 09:36:06 GMT):
Has joined the channel.

Rick (Tue, 18 Jun 2019 09:36:07 GMT):
hello, can a verifier reuse a proof of a holder to forward to another verifier? would this be valid actually?

FarhanShafiq (Tue, 18 Jun 2019 11:07:06 GMT):
what is the integer role sequence used in domain_transaction_genesis ??

FarhanShafiq (Tue, 18 Jun 2019 11:07:48 GMT):
Test network comes with 5 entities and only one has role '0' and rest are '2'.

FarhanShafiq (Tue, 18 Jun 2019 11:08:07 GMT):
Can anyone tell me what role '0' is and rest ??

FarhanShafiq (Tue, 18 Jun 2019 11:27:46 GMT):
I used one of the did from the domain_transaction_genesis for adding new node as a validator but got error: Transaction has been rejected: client request invalid: InsufficientCorrectSignatures(0, 1)

BreizhIndy (Tue, 18 Jun 2019 12:53:20 GMT):
Hello, I'm looking for performance evaluation of hyperledger indy. Where can I find it ?

yutopyan (Tue, 18 Jun 2019 14:37:26 GMT):
Has joined the channel.

FarhanShafiq (Tue, 18 Jun 2019 14:55:19 GMT):
@BreizhIndy is there anyway to recover the seed of a STEWARD DID ??

FarhanShafiq (Tue, 18 Jun 2019 14:58:10 GMT):
I got it, I need to use seed when using DID which recover signing key of STEWARD.

BreizhIndy (Tue, 18 Jun 2019 14:58:18 GMT):
it is in your genesis file i think

FarhanShafiq (Tue, 18 Jun 2019 14:58:27 GMT):
Thanks let me check.

FarhanShafiq (Tue, 18 Jun 2019 15:01:13 GMT):
It's not in domain_transaction_genesis but there is verkey.

FarhanShafiq (Tue, 18 Jun 2019 15:01:22 GMT):
What does verkey does ??

FarhanShafiq (Tue, 18 Jun 2019 15:01:22 GMT):
What verkey does ??

danielhardman (Tue, 18 Jun 2019 15:06:00 GMT):
@BreizhIndy Indy's performance will depend hugely on how big the pool of validators is, how much RAM and compute power they have, and how fast their network connections are. Thus, there isn't a general "Indy performance" study. However, I know that the team that builds Indy does evaluate Indy in a specific configuration to guarantee acceptable performance before each release. They are trying to prove sustained reads of thousands of reads per second and >10 writes per second, with a validator pool of 25-ish, and they have consistently achieved those metrics. Less validators would increase write performance significantly. (10 writes per second sounds slow to me, although it is twice as fast as the global bitcoin network. However, Indy has a number of performance optimizations that are still teed up, that should easily allow this number to be 1-3 orders of magnitude faster when we need it. It just hasn't been a focus for now, partly because of how skewed traffic is toward reading.) The bottom line is that Indy's performance is predictable, and fast enough, for the use cases we have today. And it can be increased substantially when we need to.

FarhanShafiq (Tue, 18 Jun 2019 15:07:17 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=8Dnm7HkmfBvQdJ6ch) Do you know where can i get seed of steward DID for recovering sign_key to perform transaction??

FarhanShafiq (Tue, 18 Jun 2019 15:08:16 GMT):
@danielhardman Do you know from where i can get seed of steward DID to recover signing key ?

danielhardman (Tue, 18 Jun 2019 15:18:19 GMT):
Seeds are never recorded by anybody but the steward themselves. So if you are talking about a steward DID that you created a long time ago, only you would have that seed. If you have lost a steward key, the steward identity will probably have to be abandoned and replaced with a new one. On the production network, this will be a bother; on the test networks, it will be easier. If you are talking about where you can get a new seed for a new steward DID, I would use a secure random number generator (e.g., on java, java.security.SecureRandom) or similar.

BreizhIndy (Tue, 18 Jun 2019 15:44:32 GMT):
Thanks @danielhardman, I have to set up an experiment with 1000 wallet and different number of node. How can I evaluate red and write ? Do I have to send a big amount of transaction read/write and wait to see when it finish ?

danielhardman (Tue, 18 Jun 2019 15:57:20 GMT):
Yes, you would have to simply time the transactions. However, i don't understand what you mean about 1000 wallets. Did you mean 1000 writes? You could try contacting @ashcherbakov to see what script is being used by testers on his team.

BreizhIndy (Tue, 18 Jun 2019 16:10:00 GMT):
I mean simulate 1000 independent users exchanging credential, of course IRL they would not do the transactions at the exact same time, but I still need to check if it's possible

BreizhIndy (Tue, 18 Jun 2019 16:14:22 GMT):
There is multiple read and write per credentials exchange

BreizhIndy (Tue, 18 Jun 2019 16:15:35 GMT):
to simplify my experiment I can maybe do for example: 1 university send 1000 "degree" credentials to the students who graduated from the university

BreizhIndy (Tue, 18 Jun 2019 16:15:43 GMT):
then try with 100 and 10000

BreizhIndy (Tue, 18 Jun 2019 16:16:00 GMT):
and with different number of nodes

danielhardman (Tue, 18 Jun 2019 16:42:10 GMT):
Yes, that should easily work well.

danielhardman (Tue, 18 Jun 2019 16:45:40 GMT):
Exchanging credentials does not involve the ledger very much, though. Issuance doesn't require any write with each credential--only a write of the schema, cred def, and revocation registry that they all use. So you could issue 1000 or 10M credentials, and you still only have 3 ledger writes and no ledger reads (except maybe to populate a cache). When the credential is used to present proof, the verifier and the holder/prover each do a ledger read to check an accumulator value. These reads are likely to be cached for the holder, if they are proving regularly; in real usage, only the verifier's reads will probably have a linear correlation to ledger traffic.

lijamie (Tue, 18 Jun 2019 19:29:59 GMT):
I ran into dependency hell in building libindy. When I add "mongo_driver = "0.12.1"" into [dependencies] of Cargo.toml under libindy project, the signature will stop to verify.

lijamie (Tue, 18 Jun 2019 19:30:23 GMT):
The following is the segment in indy-sdk/libindy/Cargo.tomo

lijamie (Tue, 18 Jun 2019 19:30:27 GMT):
[dependencies]

lijamie (Tue, 18 Jun 2019 19:30:36 GMT):
mongo_driver = 0.12.1

lijamie (Tue, 18 Jun 2019 19:30:40 GMT):
int-traits = {......

lijamie (Tue, 18 Jun 2019 19:32:37 GMT):
The following is the log.

lijamie (Tue, 18 Jun 2019 19:32:38 GMT):
2019-06-18 19:23:21,256|INFO|message_processor.py|trustee_02 discarding message ({'operation': {'type': '1', 'verkey': 'Yf1dYJqsARCNHG1KnLfS2JgfyTtUEjZghJaHX9McpBH', 'dest': '215coCn9k8QLaQnNNqtyiu', 'role': '2'}, 'reqId': 1560885801245853708, 'identifier': 'V4SGRU86Z58d6TV7PBUe6f', 'signature': '5u67A2G1NEgxUSpfbyW5qkrxC6P5gSmFDZhvsHcw695sGR9aPiTWhVH14KcJZuUdz3cndCohFTgTAK4wk2DoGzjE', 'protocolVersion': 2}, b'8.X%DrZfysKb@tJd2-Qp@0)#3S4:60xiHEeQ4!.7') because InsufficientCorrectSignatures(0, 1)

lijamie (Tue, 18 Jun 2019 19:32:53 GMT):
Anyone has any good ideas?

drummondreed (Tue, 18 Jun 2019 20:20:59 GMT):
The crypto devs like @brentzundel or @MALodder can confirm, but I pretty sure the answer is no. That itself would be a correlation risk.

MALodder (Tue, 18 Jun 2019 20:22:16 GMT):
No, the verifier can only show that someone proved something to him, no impersonation unlike full disclosure proofs which reveal even the signature then the verifier can impersonate. But not with indy

raj_shekhar (Wed, 19 Jun 2019 05:29:01 GMT):
Has joined the channel.

raj_shekhar (Wed, 19 Jun 2019 05:29:46 GMT):
Hi, How diff is Aries from Indy

raj_shekhar (Wed, 19 Jun 2019 05:29:48 GMT):
?

FarhanShafiq (Wed, 19 Jun 2019 06:29:54 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=o9caCaxEaBoYsoDbe) Thanks @danielhardman, i created network again but from where can i get the seed of STEWARD DID ??

FarhanShafiq (Wed, 19 Jun 2019 06:29:54 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=o9caCaxEaBoYsoDbe) Thanks @danielhardman, i created network again but from where can i get the seed of STEWARD DID ?? There is also nodes seed when generating pool transactions.

danielhardman (Wed, 19 Jun 2019 06:33:04 GMT):
Aries will include the DID communication and wallet stuff from indy-sdk, plus additional functionality to support other blockchains. It will not include anything to run blockchains--only to use them. So indy-node will never move into Aries. If you are using indy-sdk today, keep using it, and in a few months it might be suitable to use aries-sdk instead. If you are trying to run an indy-blockchain, keep using indy-node; Aries will never replace that.

danielhardman (Wed, 19 Jun 2019 06:35:45 GMT):
Farhan: If you added a steward to your network, you can do it one of two ways: with a seed or without a seed. If you did it with a seed, then the seed will appear on the commandline where you added that NYM, as a named arg, like so: "seed=xyz". If you did it without a seed, then there is no way to look it up.

danielhardman (Wed, 19 Jun 2019 06:35:45 GMT):
Farhan: If you added a steward to your network, you might have done it one of two ways: with a seed or without a seed. If you added the steward with a seed, then the seed will appear on the commandline where you added that steward (a NYM transaction), as a named arg, like so: "seed=xyz". If you did it without a seed, then there is no way to look it up.

FarhanShafiq (Wed, 19 Jun 2019 06:40:07 GMT):
There are some Steward DID that come up with test network as registered in domain_transaction_genesis so, can i use these DID to add a node ?

danielhardman (Wed, 19 Jun 2019 06:40:36 GMT):
yes

tergos (Wed, 19 Jun 2019 06:41:25 GMT):
Has joined the channel.

tergos (Wed, 19 Jun 2019 06:41:27 GMT):
hey i was trying to get the alice tutorial example working , but when i type in the command "sudo make indy-base" it gives me the error "reuesting key 68DB5E88 from hkp server keyserver.ubuntu.com"

tergos (Wed, 19 Jun 2019 06:41:44 GMT):
gpg:keyserver timed out

FarhanShafiq (Wed, 19 Jun 2019 06:42:24 GMT):
I used one of domain_transaction_genesis Steward DID and use in wallet and perform a transaction but it says : Transaction has been rejected: client request invalid: InsufficientCorrectSignatures

danielhardman (Wed, 19 Jun 2019 06:42:46 GMT):
Are you on a large corporate network where handling of DNS for keyserver.ubuntu.com might be customized by the internal network?

FarhanShafiq (Wed, 19 Jun 2019 06:43:50 GMT):
Do i need to use DID with seed in wallet to get the signing key to perform transaction? if yes, how do i get these DID seed?

tergos (Wed, 19 Jun 2019 06:43:51 GMT):
i am using my college network.

danielhardman (Wed, 19 Jun 2019 06:44:43 GMT):
That's related to a recent change that was made in Indy to handle multi-signature stuff. You may need to sign the transaction or you may need to reconfigure the network not to use multisig. This is a problem outside my current knowledge. I suggest that you contact @ashcherbakov with your question.

danielhardman (Wed, 19 Jun 2019 06:44:59 GMT):
He is in Russia so it is the beginning of his work day.

tergos (Wed, 19 Jun 2019 06:45:07 GMT):
by internal network u mean ,

danielhardman (Wed, 19 Jun 2019 06:45:08 GMT):
Maybe try him in an hour or so.

danielhardman (Wed, 19 Jun 2019 06:45:23 GMT):
yes

danielhardman (Wed, 19 Jun 2019 06:46:41 GMT):
Someone else recently reported this same problem, and I heard the theory that there was something about the network configuration that was interfering with interactions with keyserver.ubuntu.com. But I don't know if the theory proved to be true or not.

danielhardman (Wed, 19 Jun 2019 06:47:31 GMT):
I suggest that you contact @ashcherbakov with your question. He is in Russia, so it is the beginning of his workday. Maybe try him in an hour.

tergos (Wed, 19 Jun 2019 06:47:40 GMT):
is there any way that i can get around it

tergos (Wed, 19 Jun 2019 06:47:56 GMT):
this happens in step 4 of 16

FarhanShafiq (Wed, 19 Jun 2019 06:49:26 GMT):
Last question, do each DID have it's own signing key and saved in wallet to perform transaction? because i shouldn't be able use any DID to perform transaction on owner behave unless i have authentication keys to perform transaction.

FarhanShafiq (Wed, 19 Jun 2019 06:49:26 GMT):
Last question, do each DID have it's own signing key and saved in wallet to perform transaction? because i shouldn't be able use any DID to perform transaction on owner behalf unless i have authentication keys to perform transaction.

danielhardman (Wed, 19 Jun 2019 06:49:45 GMT):
I think this is a question to ask your local sysadmins or network admins. I believe that what's happening is that the code is attempting to verify the signature on some software, but your network may have access to keyserver.ubuntu.com blocked. There probably is a workaround, but I don't know it.

danielhardman (Wed, 19 Jun 2019 06:51:29 GMT):
Every DID has its own private+public keypair (also called the "signing key" and "verkey", respectively). All keypairs are stored in a wallet. In some of the getting started workflows, there is more than one actor (e.g., Faber, Thrift, Alice); each one will have its own wallet.

danielhardman (Wed, 19 Jun 2019 06:52:33 GMT):
Using a seed allows you to force a wallet on device 1 and a wallet on device 2 to contain the same keypair; if a command to create a DID and/or keypair uses the same seed, it ends up with the same keypair values.

danielhardman (Wed, 19 Jun 2019 06:55:45 GMT):
I'm sorry I'm not more help. It's 1 am for me and this is a workflow I haven't done for a long time, so I think you'll be better off talking to someone who's more current and more on what it's doing.

danielhardman (Wed, 19 Jun 2019 06:56:06 GMT):
I'm sorry I'm not more help. It's 1 am for me and this is a workflow I haven't done for a long time, so I think you'll be better off talking to someone who's more current and more on what it's doing.

danielhardman (Wed, 19 Jun 2019 06:56:06 GMT):
I'm sorry I'm not more help. It's 1 am for me and this is a workflow I haven't done for a long time, so I think you'll be better off talking to someone who's more current and more deep on what it's doing.

tergos (Wed, 19 Jun 2019 07:07:38 GMT):
Thank you sir.

FarhanShafiq (Wed, 19 Jun 2019 07:10:57 GMT):
Thank you so much @danielhardman :)

FarhanShafiq (Wed, 19 Jun 2019 07:12:34 GMT):
It's help a lot to me

raj_shekhar (Wed, 19 Jun 2019 07:14:32 GMT):
thanks Daniel

Rick (Wed, 19 Jun 2019 07:18:15 GMT):
Hi, for indy's credential definition is there any 'in-the-box' migration way that i could use when i want to move a credential definition from the old definition migrating to the new cred def?

BrajeshKumar (Wed, 19 Jun 2019 08:56:25 GMT):
Has joined the channel.

Rick (Wed, 19 Jun 2019 09:16:56 GMT):
ok. noted with thanks.

FarhanShafiq (Wed, 19 Jun 2019 09:21:43 GMT):
"domain genesis transactions define initial trusted trustees and stewards" as mentioned in transactions documentation: https://github.com/hyperledger/indy-node/blob/master/docs/source/transactions.md But from where i can get signing_key/seed of these DID's to perform transaction?

FarhanShafiq (Wed, 19 Jun 2019 09:23:46 GMT):
Or how to create a initial trustee DID to add in genesis transaction from which i can create Steward and other DID?

FarhanShafiq (Wed, 19 Jun 2019 09:35:57 GMT):
How to create an initial Indy Steward for the test network ?

FarhanShafiq (Wed, 19 Jun 2019 09:35:57 GMT):
How to create an initial Indy Steward for the test network using indy-cli?

SeanBohan (Wed, 19 Jun 2019 14:40:10 GMT):
Hey everyone: Tomorrow’s INDY WG call - Thurs 6/20: 4pmBST, 11amET, 9amMT Zoom: https://zoom.us/j/232861185 Agenda: 1. Indy-SDK Credential Attribute Tag Policy with SKlump 2. Cached Credentials for offline proving in INDY SDK with DarkoK 3. Changes to the Thursday Calls? 4. Open Discussion

FarhanShafiq (Thu, 20 Jun 2019 08:38:17 GMT):
What is the Trust Anchor Role? is it Endorser or Steward?

FarhanShafiq (Thu, 20 Jun 2019 08:59:13 GMT):
Is there tool to read the indy block to detect newly register DID?

AbhishekAadi (Thu, 20 Jun 2019 10:43:11 GMT):
Hello guys, can anybody please provide me the resources on themis protocol?

ShrinivasDeshmukh (Thu, 20 Jun 2019 15:36:53 GMT):
Hi everyone, please can someone explain the process of adding new indy-node to existing network in detail? I am following the indy documentation, but am unable to add the node

kdenhartog (Thu, 20 Jun 2019 16:26:05 GMT):
@ShrinivasDeshmukh Can you post more details about the issue you're encountering? @mgbailey is the most knowledgeable about these issues typically.

ShrinivasDeshmukh (Thu, 20 Jun 2019 16:32:24 GMT):
When I followed the Indy document, I get an error 'removing bid' when new node is trying to connect to the pool

ShrinivasDeshmukh (Thu, 20 Jun 2019 16:33:29 GMT):
I then followed these steps: 1. Create a did in the new node, 2. Add did to the ledger and assign steward role to it

ShrinivasDeshmukh (Thu, 20 Jun 2019 16:34:39 GMT):
I am getting the following error: Transaction has been rejected: client request invalid: InsufficientCorrectSignatures(0, 1)

esplinr (Thu, 20 Jun 2019 20:18:59 GMT):
https://www.reddit.com/r/rust/comments/c20aed/facebook_just_pick

esplinr (Thu, 20 Jun 2019 20:20:13 GMT):
Today's demo from the team at Evernym covers the fixes included in the June release of Indy Node, including renaming of Trust Anchor to Endorser (cc @FarhanShafiq ). We also discuss the implementation of PBFT View Change. https://www.youtube.com/watch?v=fCmsgzBKxEk&list=PLRp0viTDxBWGLdZk0aamtahB9cpJGV7ZF&index=1

esplinr (Thu, 20 Jun 2019 20:24:42 GMT):
Past demo videos are here: https://www.reddit.com/r/rust/comments/c20aed/facebook_just_picked_rust_to_implement_their_new/erhsz9q/

esplinr (Thu, 20 Jun 2019 20:24:42 GMT):
Past demo videos are here: https://wiki.hyperledger.org/display/indy/Evernym+Sprint+Demos

FarhanShafiq (Fri, 21 Jun 2019 06:08:47 GMT):
Thanks @esplinr

zixian5 (Fri, 21 Jun 2019 06:13:31 GMT):
hi,Does anyone konw how to use sdk to reverse the revocation?

Heena078 (Fri, 21 Jun 2019 06:31:06 GMT):
Has joined the channel.

ShrinivasDeshmukh (Fri, 21 Jun 2019 07:03:43 GMT):
@mgbailey Hello Mike, Please can you help me find a solution?

Zohaib_Sohail (Fri, 21 Jun 2019 07:18:45 GMT):
can someone please tell me the proper way to test predicates while presenting a proof?

Zohaib_Sohail (Fri, 21 Jun 2019 07:18:45 GMT):
can someone please tell me the proper way to test predicates while presenting a proof? because the predicate is always happened to be true while the condition is false. maybe i am missing something but dont know what!

Zohaib_Sohail (Fri, 21 Jun 2019 07:59:35 GMT):
i am using nodejs agent

Heena078 (Fri, 21 Jun 2019 10:15:29 GMT):
Does anyone have idea about this indy-ssivc-tutorial tutorial i tried lot but i am getting issue in Prerequisite .....................???

Zohaib_Sohail (Fri, 21 Jun 2019 13:58:45 GMT):
that repo is archived. I would suggest you to look into this tutorial https://github.com/hyperledger/education/tree/master/LFS171x/indy-material/nodejs, which is also not maintained now

Zohaib_Sohail (Fri, 21 Jun 2019 13:58:45 GMT):
that repo is archived. I would suggest you to look into this tutorial https://github.com/hyperledger/education/tree/master/LFS171x/indy-material/nodejs, which is also not maintained now but its actually quite helpful to get you started

swcurran (Fri, 21 Jun 2019 14:46:45 GMT):
I was going to say the same. We were that - glad you like it. It the concepts are pretty good, but the implementation is behind. Don't spend a lot of time on the code with that.

swcurran (Fri, 21 Jun 2019 14:46:45 GMT):
I was going to say the same. We wrote the tutorial for that, and BYU wrote the code - glad you like it. It the concepts are pretty good, but the implementation is behind. Don't spend a lot of time on the code with that.

mgbailey (Fri, 21 Jun 2019 21:56:46 GMT):
@ShrinivasDeshmukh You need to be a Trustee to add a Steward. In the CLI follow the steps shown in slide 3 of https://docs.google.com/presentation/d/1s0FH7GibFgFkSojwyh_cYEA_A2BUlP8WsEnmpEVHdrA/edit#slide=id.g4e379abf57_0_6

ShrinivasDeshmukh (Sat, 22 Jun 2019 02:46:42 GMT):
Thank you @mgbailey . Will follow the steps.

morphatic (Sun, 23 Jun 2019 16:48:24 GMT):
Has joined the channel.

drummondreed (Sun, 23 Jun 2019 22:43:34 GMT):
@FarhanShafiq I don't see that anyone's answered this, so in brief: the Trust Anchor role was designed before there was a clear separation of Steward and Endorser. The goal is that Steward is defined as a specific role for operating an Indy validator node, and Endorser is a specific role for approving transactions on on the ledger. I believe (but do not know for sure) that the code is evolving to separate these roles.

sukalpomitra (Mon, 24 Jun 2019 03:30:52 GMT):
hi can anyone guide in how to run indy-cat-agent. The readme says "Instructions forthcoming. indy_catalyst_agent will be made available in the future as a python package at pypi.org."

sukalpomitra (Mon, 24 Jun 2019 04:21:44 GMT):
got it running now. thanks to andrew.whitehead

zixian5 (Mon, 24 Jun 2019 05:42:22 GMT):
hi,can anyone tell me why this method (indy_issuer_recover_credentia)is forbidden to use?

andrew.whitehead (Mon, 24 Jun 2019 06:02:05 GMT):
I’m curious as well, maybe something to do with how the accumulator is calculated. Revocation in general is under active development for ‘anoncreds 2.0’ so I’m not sure much is changing in the short term

zixian5 (Mon, 24 Jun 2019 06:13:32 GMT):
:neutral_face: Amazing! I tested the method. It could recover a revocation normally!

andrew.whitehead (Mon, 24 Jun 2019 06:18:12 GMT):
Strange, I looked at the rust code and it was commented out. Did you recompile it?

zixian5 (Mon, 24 Jun 2019 06:20:00 GMT):
Yes! not too difficult

FarhanShafiq (Mon, 24 Jun 2019 06:29:28 GMT):
Thanks @drummondreed , i found that Trust anchor named has been changed into endorser while the role remains same.

FarhanShafiq (Mon, 24 Jun 2019 12:31:41 GMT):
Is there any way to revoke the identity of a user ??

hanubc7743 (Tue, 25 Jun 2019 12:47:16 GMT):
Has joined the channel.

hanubc7743 (Tue, 25 Jun 2019 13:10:45 GMT):
Hi We are looking for the hyperledger indy experienced developers so who has interest they can share the profile to hanubc7743@gmail.com

esplinr (Tue, 25 Jun 2019 21:44:11 GMT):
Another good place to post is #jobs on chat.sovrin.org

davology (Wed, 26 Jun 2019 09:11:57 GMT):
Has joined the channel.

davology (Wed, 26 Jun 2019 09:14:04 GMT):
Hello, I’ve been using Indy in AWS private network successfully

davology (Wed, 26 Jun 2019 09:15:57 GMT):
However, when I try to access from a client via the public IP I hit problems because the default genesis block uses the private IP addresses. I’m using the docker images. Please can anyone tell me how to set-up for public access in AWS? Thanks

ap (Wed, 26 Jun 2019 12:14:08 GMT):
Has joined the channel.

ap (Wed, 26 Jun 2019 12:14:09 GMT):
HI, I need help to export indy wallet from android app

ap (Wed, 26 Jun 2019 12:14:46 GMT):
i did get any response after hitting *Wallet.exportWallet(myWallet, String.valueOf(exportConfigJson));* this command

swcurran (Wed, 26 Jun 2019 16:20:07 GMT):
FYI - BC Gov just posted a new Code With Us bounty about integrating an Indy/Aries agent as an IdP with Red Hat's Keycloak (and other) OpenID Connect IAM - https://www.bcdevexchange.org/opportunities/cwu/opp-create-a-red-hat-keycloak-identity-provider--idp--capable-of-processing-verifiable-credentials-using-decentralized-identity-technology-created-by-bc-gov-to-authorize-access-to-a-bc-government-digital-service-

MarcoPasotti (Wed, 26 Jun 2019 16:29:25 GMT):
Has joined the channel.

pknowles (Wed, 26 Jun 2019 17:57:32 GMT):
Has anybody come across *BigID* before? If so, any feedback? https://bigid.com

swcurran (Thu, 27 Jun 2019 02:33:11 GMT):
Indy Working Group Call Agenda/Notes/Links Skip to end of metadata Created by Sean Bohan, last modified by Stephen Curran just a moment ago Go to start of metadata When: Every Thurs 4pmBST, 11amET, 9amMT, 8amPT Where: https://zoom.us/j/232861185

swcurran (Thu, 27 Jun 2019 02:35:36 GMT):
Indy Working Group Call Agenda/Notes/Links When: Every Thurs 4pmBST, 11amET, 9amMT, 8amPT Where: https://zoom.us/j/232861185 This week's call - hosted by and with a presentation from Stephen Curran - "Flipping our thinking about the DID Ledger Layer"

swcurran (Thu, 27 Jun 2019 02:35:36 GMT):
*Thursday Morning: Indy Working Group Call* When: Every Thurs 4pmBST, 11amET, 9amMT, 8amPT Where: https://zoom.us/j/232861185 This week's call - hosted by and with a presentation from Stephen Curran - "Flipping our thinking about the DID Ledger Layer"

Heena078 (Thu, 27 Jun 2019 04:27:05 GMT):
hey thank you will go through it...

Rick (Thu, 27 Jun 2019 09:18:14 GMT):
Hi for Indy it seems that nothing much goes on the chain only (cred def, schema & revocation registry) and there is no chaincode hence what is the consensus mechanism for/doing? AFAIK consensus in other chains i.e. fabric is to make sure nothing is double spent after running the chaincode before sending to the orderer. Is Indy's consensus mechanism just ordering the blocks?

Thomas_the_Consultant (Thu, 27 Jun 2019 14:05:54 GMT):
Has joined the channel.

Thomas_the_Consultant (Thu, 27 Jun 2019 14:06:00 GMT):
Hello :) I'd have a basic question... Can a user be issued with a credential of “You are 6ft tall and weigh 80kg”, and then create a proof (or rather verified presentation) from this that they weigh over 30kg and are over 4ft tall, with the specific number being kept confidential? If this is currently working, is it restricted to range proofs, or can it do more sophisticated assertions?

Thomas_the_Consultant (Thu, 27 Jun 2019 14:06:00 GMT):
Hello :) I have a basic question... Can a user be issued with a credential of “You are 6ft tall and weigh 80kg”, and then create a proof (or rather verified presentation) from this that they weigh over 30kg and are over 4ft tall, with the specific number being kept confidential? If this is currently working, is it restricted to range proofs, or can it do more sophisticated assertions?

ShrinivasDeshmukh (Thu, 27 Jun 2019 15:58:29 GMT):
Hi, is there any provision in fabric/indy to add hyperledger Indy's trust anchor as an OrgMSP for fabric?

rohit_kumar (Thu, 27 Jun 2019 16:27:43 GMT):
Has joined the channel.

rohit_kumar (Thu, 27 Jun 2019 16:28:45 GMT):
I want to ask a basic question. When Faber College issue identity to Alice then faber college and Alice both know the identity of Alice. It means Faber College maintaning local database of students and assign identity to its students. Which students kept in their wallet. Then how identity is decentralize? Faber college local database might be a honey pot for attackers.

rohit_kumar (Thu, 27 Jun 2019 16:29:04 GMT):
Then how identity is decetralize?

rohit_kumar (Thu, 27 Jun 2019 16:30:22 GMT):
Acme corporation can directly ask identity of Alice from Faber college. Instead of her

DucaDellaForcoletta (Fri, 28 Jun 2019 09:45:00 GMT):
Hi Daniel, Actually a verifier can only check globally the presentation proof of a prover. Actually a verifier can only state "this credential is not valid" without know "is invalid because revoked" or "is invalid because you have tampered it". This info is very important in a business scenario.

DucaDellaForcoletta (Fri, 28 Jun 2019 09:45:00 GMT):
Hi Daniel, Actually a verifier can only check globally the presentation proof of a prover. Actually a verifier can only states "this credential is not valid" without know "is invalid because revoked" or "is invalid because you have tampered it". This info is very important in a business scenario.

DucaDellaForcoletta (Fri, 28 Jun 2019 09:45:00 GMT):
Hi Daniel, Actually a verifier can only check globally the presentation proof of a prover. Actually a verifier can only states "this credential is not valid" without know "is invalid because revoked" or "is invalid because you have tampered it". This info is very important in a business scenario. I'm referring to the actual implementation of the framework

DucaDellaForcoletta (Fri, 28 Jun 2019 09:45:00 GMT):
Hi Daniel, Actually a verifier can only check globally the presentation proof of a prover. Actually a verifier can only states "this credential is not valid" without know "is invalid because revoked" or "is invalid because you have tampered it". This info is very important in a business scenario. I'm referring to the actual implementation of the indy framework

uNmAnNeR (Fri, 28 Jun 2019 12:14:15 GMT):
Has joined the channel.

uNmAnNeR (Fri, 28 Jun 2019 12:24:07 GMT):
Hello everyone! My question is about indy-agent. I wondering if there single protocol for indy-agent, or there are many different implementations of it (described https://github.com/hyperledger/indy-agent#known-agent-implementations). For instance https://github.com/evernym/connectathon-agent/ has different message format then indy-agent. So is there some standard way to communicate between indy actors which is supported by majority? Or indy-agent implementation is just an example how it can be done? Is indy-agent a production solution? And is it still under development/supported?

swcurran (Fri, 28 Jun 2019 14:12:18 GMT):
We are getting to a single set of protocols - the ones in that table marked "Passed" work together. Evernym was a pioneer in agents and have to do some retrofitting to get them aligned with the current protocols. The discussions on the protocols are happening on the Wed. Aries calls and are documented in the aries-rfcs repo at Hyperledger. A new agent that evolved from the `indy-catalyst` agent will be transferred to Hyperledger this coming Monday (or so) - `aries-cloudagent-python` - that implements many of the known protocols. It is currently at `bcgov` with the same name. More docs today and over the weekend. It exposes a REST interface and uses webhooks to allow you to write a controller web app in whatever language you want to use to inject the business rules for an instance of the agent.

uNmAnNeR (Fri, 28 Jun 2019 14:14:07 GMT):
thanks for explanation, will take a look at cloudagent closely

mossmanpete (Sat, 29 Jun 2019 00:04:34 GMT):
Has left the channel.

rasviitanen (Sun, 30 Jun 2019 07:33:21 GMT):
Has joined the channel.

tplooker (Sun, 30 Jun 2019 19:22:18 GMT):
Hey guys, at Mattr, we're growing and on the lookout for new members in the developer community that are interested in an opportunity with us. If you are or know anyone that could be interested please let either kdenhartog or I know, thanks.

n0rbs (Mon, 01 Jul 2019 02:53:41 GMT):
Has joined the channel.

jandayanan (Mon, 01 Jul 2019 07:15:58 GMT):
Has joined the channel.

ademcaglin (Mon, 01 Jul 2019 10:53:07 GMT):
Has joined the channel.

ademcaglin (Mon, 01 Jul 2019 10:53:08 GMT):
Hi community, i am learning about self soverign identity, decentralized idenfiers and hyperledger indy. First i started to learn specifications about these concepts. Now i have some question to ask. This platform has good community to find answers for my questions. But i like stackoverflow way to ask a question. I searched some questions about these subjects, i found a few questions. So i asked a question about DIDs and i need help. Could you answer my question. Here is link: https://stackoverflow.com/questions/56834159/does-decentralized-identifiers-cover-decentralized-pki

josh.graber (Mon, 01 Jul 2019 13:07:05 GMT):
What is the latest word on the authorization policy registry? I heard that this is not actively supported by Hyperledger Indy yet but is being worked on?

MALodder (Mon, 01 Jul 2019 14:31:02 GMT):
I

MALodder (Mon, 01 Jul 2019 14:31:02 GMT):
I'm actively working on it. We discuss it in the Sovrin Crypto Calls which happen at 9am MDT every Monday

MALodder (Mon, 01 Jul 2019 14:32:24 GMT):
We might be discussing it in depth today if we have time

MALodder (Mon, 01 Jul 2019 14:32:42 GMT):
The meeting id for that is Description:Agenda: https://docs.google.com/document/d/1M3zgsRoOxlkjNPbDp4_7YWxuqO2EyefOT3Bt7ZgGIDY/edit?usp=sharing Join from PC, Mac, Linux, iOS or Android: https://zoom.us/j/567114224

josh.graber (Mon, 01 Jul 2019 14:35:24 GMT):
What time?

MALodder (Mon, 01 Jul 2019 14:35:52 GMT):
at 9AM MDT

josh.graber (Mon, 01 Jul 2019 14:36:51 GMT):
Thanks

josh.graber (Mon, 01 Jul 2019 14:49:27 GMT):
I have a conflict, unfortunately. Interested in hearing how this plays out... We'll likely have some stuff riding on this in the near future

SteveGoob (Mon, 01 Jul 2019 22:52:19 GMT):
As some of you may know, indy packages have been having troubles installing and publishing this weekend. We found out this was an expired gpg key that prevented debian packages from being signed. We have now rotated the keys successfully, but there is an issue with ubuntu's keyserver not resolving the abbreviated key properly. To work around that issue, anyone wishing to install any indy packages on debian-based systems (including ubuntu) needs to run: ``` apt-key adv --keyserver keyserver.ubuntu.com --recv-keys CE7709D068DB5E88 ``` We're sorry for any inconvenience and confusion this caused.

drummondreed (Mon, 01 Jul 2019 23:28:37 GMT):
@Rick The short answer is yes. Also, the validator nodes ensure that any DID (NYM transaction) can only be entered once.

drummondreed (Mon, 01 Jul 2019 23:32:24 GMT):
@MALodder or the other crypto gurus can answer this. Currently I believe it is restricted to range proofs, but the Hyperledger Ursa team is working on the second generation that will be more sophisticated.

MALodder (Mon, 01 Jul 2019 23:33:53 GMT):
@drummondreed is correct. For it’s just range proofs but next iteration you can do range proofs and set membership tests. What other proofs would you do with height and weight?

drummondreed (Mon, 01 Jul 2019 23:36:58 GMT):
So you are correct that Faber College will likely still keep a local database of the attributes it used to issue credentials. However Faber College does not have (and will never have) the private key used to *produce a proof* of those credentials after they are issued. Only the holder has that private key. Since with ZKP credentials, verifiers only ask for proofs, the value of the raw identity attributes is literally *drained away* by SSI architecture. The value is now in being able to produce the proofs...which requires the private keys...which is decentralized. And which should finally solve the honey pot issue for identity data.

n0rbs (Tue, 02 Jul 2019 06:14:17 GMT):
Hi everyone. I'm trying to deploy the indy nodes in K8s. I'm using nodePort to forward the requests to the container nodes via the `32xxx` ports. I'm using minikube to run it locally. How do we customize the pool_genesis_transaction config so that we can use the nodePort, 32701-32708, instead of the default ports, 7901-7908, and send the transactions?

n0rbs (Tue, 02 Jul 2019 07:38:20 GMT):
We've provisioned pods for each nodes (node1, node2, node3, node4) but after inspecting the pods, they seem can't to find the other nodes. Connection seems to be refused from the other nodes

Thomas_the_Consultant (Tue, 02 Jul 2019 09:24:08 GMT):
Interesting. I've been told by commercial organisation that this isn't yet implemented. For now, we should be fine with range proofs for our purposes.

Thomas_the_Consultant (Tue, 02 Jul 2019 09:26:20 GMT):
If range-proofs are in the current implementation, then that's a big plus for our super secret business plans :)

ShrinivasDeshmukh (Tue, 02 Jul 2019 09:32:17 GMT):
Hi everyone, I have a question The edge agent(mobile) will hold the wallet and all the data in it, and cloud agent will hold proofs, creds, etc data in it. As soon as any activity like cred issuance, etc happen on edge agent, this data should be pushed to the cloud agent (all sharable data). Is this the right approach? Can someone please suggest me the correct way of achieving it?

Thomas_the_Consultant (Tue, 02 Jul 2019 10:01:13 GMT):
If we could work across matrixes of values, that would be great too...

kdenhartog (Tue, 02 Jul 2019 10:18:44 GMT):
For now it's suggested that all proofs, credential data, and private keys are kept in edge wallets. Otherwise you're trusting that a cloud agent won't impersonate you and generate proofs or sign without your consent.

ShrinivasDeshmukh (Tue, 02 Jul 2019 10:46:57 GMT):
Ok! Thank you for the help!

Thomas_the_Consultant (Tue, 02 Jul 2019 11:56:42 GMT):
Or summations, ex. normalised SAT score plus FICO score is greater than 0.9. That sort of thing...

MALodder (Tue, 02 Jul 2019 12:32:09 GMT):
ZKPs aren’t meant to handle complex equations unless you’re willing to wait minutes to compute them in zero knowledge. Simple equations like the sum of two attributes is less than a threshold or the difference between two attributes are just fine

sheru (Tue, 02 Jul 2019 14:41:58 GMT):
Has joined the channel.

sheru (Tue, 02 Jul 2019 14:42:00 GMT):
hi

esplinr (Tue, 02 Jul 2019 16:25:57 GMT):
I'm not sure which "commercial organization" you are referring to, but having a capability in Ursa or Indy SDK doesn't mean that it is exposed through all the layers in the application stack.

esplinr (Tue, 02 Jul 2019 16:26:02 GMT):
So you will want to confirm.

MALodder (Tue, 02 Jul 2019 16:27:10 GMT):
The range proof '>=' has always been supported. Ursa supports '<','<=','>' but I'm not sure if indy-sdk supports those three yet

swcurran (Tue, 02 Jul 2019 17:06:21 GMT):
FYI - BC Gov has posted this "Code With Us" (bounty) opportunity for a developer/team to build an Aries-based Identity Provider (IdP) for an IAM solution (Red Hat's Keycloak). The opportunity closes at 4PM Pacific on July 3, 2019 so if you are interested, please get your response in ASAP. https://www.bcdevexchange.org/opportunities/cwu/opp-create-a-red-hat-keycloak-identity-provider--idp--capable-of-processing-verifiable-credentials-using-decentralized-identity-technology-created-by-bc-gov-to-authorize-access-to-a-bc-government-digital-service-

Thomas_the_Consultant (Tue, 02 Jul 2019 18:12:20 GMT):
@MALodder Thanks! That clears things up a lot :+1:

Thomas_the_Consultant (Tue, 02 Jul 2019 18:16:45 GMT):
Ideally we'd like to be able to have a proof that the user's credential point is within a defined area in an arbitrarily dimensioned space. (It's a financial application.) But '>=' gets us a long way already.

SeanBohan (Tue, 02 Jul 2019 20:41:24 GMT):
Folks: NO INDY CALL THIS WEEK, JULY 4th. See you next Thursday, July 11!!!

Rick (Wed, 03 Jul 2019 05:55:19 GMT):
any tips/sources/how-to to implement indy agent/ aries using serverless/lambda? is this possible

esplinr (Wed, 03 Jul 2019 20:01:17 GMT):
Update from the Evernym developers on our contributions to Indy. Topics include: * Release status * Proposal for separating #HyperledgerAries from Indy * Property based testing for view change * Bug fix for node crash Previous topics and videos: https://wiki.hyperledger.org/display/indy/Evernym+Sprint+Demos

esplinr (Wed, 03 Jul 2019 20:01:17 GMT):
Update from the Evernym developers on our contributions to Indy. https://www.youtube.com/watch?v=UvP6RGYy4Yo&list=PLRp0viTDxBWGLdZk0aamtahB9cpJGV7ZF Topics include: * Release status * Proposal for separating #HyperledgerAries from Indy * Property based testing for view change * Bug fix for node crash Previous topics and videos: https://wiki.hyperledger.org/display/indy/Evernym+Sprint+Demos

esplinr (Wed, 03 Jul 2019 20:01:17 GMT):
Update from the Evernym developers on our contributions to Indy. https://www.youtube.com/watch?v=UvP6RGYy4Yo&list=PLRp0viTDxBWGLdZk0aamtahB9cpJGV7ZF Topics include: * Release status * Proposal for separating #HyperledgerAries from Indy * Property based testing for view change * Bug fix for node crash Previous topics and videos: https://wiki.hyperledger.org/display/indy/Evernym+Sprint+Demos

esplinr (Wed, 03 Jul 2019 20:01:17 GMT):
Update from the Evernym developers on our contributions to Indy. https://www.youtube.com/watch?v=UvP6RGYy4Yo&list=PLRp0viTDxBWGLdZk0aamtahB9cpJGV7ZF Topics include: * Release status * Proposal for separating #HyperledgerAries from Indy * Property based testing for view change * Bug fix for node crash Previous topics and videos: https://wiki.hyperledger.org/display/indy/Evernym+Sprint+Demos

lachid (Wed, 03 Jul 2019 21:06:20 GMT):
Has joined the channel.

lachid (Wed, 03 Jul 2019 21:06:21 GMT):
how should i use different thandefault storage type for the wallet?

Diiaablo95 (Thu, 04 Jul 2019 12:00:06 GMT):
Hi people! Do you know if images from Hyperledger repositories can be used in a M.Sc. thesis, if properly cited?

s.weidenbach (Thu, 04 Jul 2019 18:47:50 GMT):
Has joined the channel.

esplinr (Fri, 05 Jul 2019 15:47:55 GMT):
@Diiaablo95 They are Apache licensed. I'm not aware of any concerns.

AndrewNikitin (Fri, 05 Jul 2019 16:20:37 GMT):
We change tag representation for indy-core-baseci docker's image. For now image with master's repository is hyperledger/indy-core-baseci:0.0.3-master and for stable: hyperledger/indy-core-baseci:0.0.3-stable All of corresponded images was compiled with latest gpg keys and pushed into hyperledger's dockerhub.

ramxis (Mon, 08 Jul 2019 10:04:09 GMT):
Has joined the channel.

ramxis (Mon, 08 Jul 2019 10:05:11 GMT):
is hyperledger indy the same as Soverin Network, I am assuming they are related like Ethereum protocol is related to main net and ropsten, i.e. Soverin is an instantiation of the indy codebase / protocol ?

ramxis (Mon, 08 Jul 2019 10:12:09 GMT):
My second question is how does the Evernym product suite relates to the Soverin network, I am researching idenity solutions for a pilot project. So far I find the requirement of onboarding government agencies such as trust anchors or issuers as hugely problematic and a barrier for any project in the near future. My question is how can evernym product suite e.g. Verity, Verity:auth, Verity:Onboarding, and connect.me can help an organization wishing to use Soverin.

Thomas_the_Consultant (Mon, 08 Jul 2019 11:20:59 GMT):
I have a build error for libindy on mac osx. Where should I report it?

ramxis (Mon, 08 Jul 2019 12:02:03 GMT):
can anyone give a quick update regarding these questions ?

jljordan_bcgov (Mon, 08 Jul 2019 13:23:41 GMT):
Yes that is correct. Sovrin is one instance of an Indy distributed ledger and is governed by the Sovrin governance framework.

Thomas_the_Consultant (Mon, 08 Jul 2019 14:00:52 GMT):
Sovrin is mainnet. Indy is the EVM. Evernym is the Redhat of Indy. Verity and connect.me are commercial components built on Indy.

WadeBarnes (Mon, 08 Jul 2019 14:03:43 GMT):
@Thomas_the_Consultant, I believe connect.me is built on Hyperledger, but not Hyperledger-Indy specifically.

WadeBarnes (Mon, 08 Jul 2019 14:03:43 GMT):
@Thomas_the_Consultant, ~I believe connect.me is built on Hyperledger, but not Hyperledger-Indy specifically.~ Sorry I was thinking about Verified.me.

SeanBohan (Mon, 08 Jul 2019 17:43:00 GMT):
Folks: If you have a demo you want to share or discussion you want to spark on this Thursday’s INDY call, please let me know here or DM me!

pknowles (Mon, 08 Jul 2019 18:26:52 GMT):
@janl has just finished a really cool consent lifecycle demo built on HL Indy. The demo has already been endorsed by Mark Lizar, CEO at Open Consent Group. Jan - Reach out to Sean!

pknowles (Mon, 08 Jul 2019 18:26:52 GMT):
@janl has just finished a really cool consent lifecycle demo built on HL Indy. The demo has already been endorsed by Mark Lizar, CEO at Open Consent Group. Jan - Reach out to @SeanBohan !

pknowles (Tue, 09 Jul 2019 11:32:31 GMT):
Cause for concern? https://medium.com/@jonathan.leitschuh/zoom-zero-day-4-million-webcams-maybe-an-rce-just-get-them-to-visit-your-website-ac75c83f4ef5

Thomas_the_Consultant (Tue, 09 Jul 2019 14:09:12 GMT):
Yes. All my SSI call have been on Zoom :-| xD

AmanAgrawal (Wed, 10 Jul 2019 10:13:50 GMT):
Has joined the channel.

AmanAgrawal (Wed, 10 Jul 2019 10:13:56 GMT):
Hey guys, I have recently started learning indy, and I have successfully recreated use case shared with link https://github.com/swcurran/education/tree/master/LFS171x/indy-material/nodejs I am now trying to replicate the above use case with 3 node agents from my use case. For this I need to change make changes in docker-compose.yaml with my server IP and Ports as I want to host these agents on my server and access these agents using those IP. If I change the IP and ports and now and deploy the application stops working. Could someone please guide here as to how to I can achieve this? Thanks :)

swcurran (Wed, 10 Jul 2019 11:09:25 GMT):
Hi @AmanAgrawal - I'm not able to help for a bit as I'm teaching an Indy course for the next couple of days. I'm not quite clear what you are trying to do - just change the config so that the UI is visible on other machines? I would not recommend investing much time in that, if you are trying to do "real things" with Indy/Aries. If you are just make a minor extension to it, that's fine, but if you want to go even deeper, you might look at the latest work being done in Aries. There is a new guide that gets you coding in Aries that has just become available (link below). It starts with the same content you have just looked at and goes from there. https://github.com/hyperledger/aries-cloudagent-python/tree/master/docs/GettingStartedAriesDev

ravip (Wed, 10 Jul 2019 14:29:52 GMT):
Has joined the channel.

ravip (Wed, 10 Jul 2019 14:29:52 GMT):
Hello everyone, so I am learning about the concepts of wallets in indy and had some questions. I hope I could get some input from the folks here or a direction to proceed. I apologize if I have not used the correct technical vocabulary to ask the question, I am still learning.

ravip (Wed, 10 Jul 2019 14:30:06 GMT):
This is one of the questions: Can you and should you store IPFS links to PII images in user's indy wallet?

luckycharms810 (Wed, 10 Jul 2019 20:44:09 GMT):
This might be a silly question, but is there a way to issue a proof of credential in a way such that it stop's working after a period of time ?

kdenhartog (Wed, 10 Jul 2019 21:05:01 GMT):
Kinda, but the reaction has been a bit hyperbolic. In order for it to work, the app needs to open, so it's not really like the User is unaware of it. Plus the mac camera light still turns on.

nhrishi (Thu, 11 Jul 2019 06:28:52 GMT):
Has joined the channel.

drummondreed (Thu, 11 Jul 2019 06:55:33 GMT):
Yes, there are various ways to approach including images and links in credentials. Probably best to ask that question in the #aries channel as that's now where agent/wallet work is happening.

drummondreed (Thu, 11 Jul 2019 06:57:20 GMT):
The easy answer is to include an expiration date as a claim in the credential, which would simply tell verifiers that it is no longer valid after that timestamp. By "not working", do you mean that the holder would no longer be able to even create a proof for it?

yogeshquick (Thu, 11 Jul 2019 07:31:22 GMT):
Has joined the channel.

Thomas_the_Consultant (Thu, 11 Jul 2019 09:05:00 GMT):
I'm just learning, and I feel "IndyError: CommonInvalidStructure" and its friends are very overused in in the wrapper APIs 🙂 That's my main observation so far!

Zohaib_Sohail (Thu, 11 Jul 2019 10:33:18 GMT):
how to run indy-catalyst agent?

Thomas_the_Consultant (Thu, 11 Jul 2019 11:47:08 GMT):
I'm trying to enable logging output from libindy like this `RUST_LOG=trace RUST_BACKTRACE=1 node src/myscript.js` but I'm not getting the hale of logging messages I'd expect 🤨 Am I missing something?

swcurran (Thu, 11 Jul 2019 12:21:55 GMT):
@Zohaib_Sohail - What was previously indy-catalyst agent is now aries-cloudagent-python and part of Hyperledger. https://github.com/hyperledger/aries-cloudagent-python Is that what you are looking for? There is a Getting Started Guide in the readme, and there are some good labs/demos for learning in the `demo` folder.

luckycharms810 (Thu, 11 Jul 2019 12:41:47 GMT):
Ah no, I guess more that a holder would be able to create a proof that would stop working for the verifier after a certain amount of time. A sort of TTL for the created proof.

luckycharms810 (Thu, 11 Jul 2019 12:42:30 GMT):
It would not so much have an impact on the credential itself, but the verifier's ability to check it.

nage (Thu, 11 Jul 2019 20:43:46 GMT):
For those looking to start something in the SSI space http://www.globenewswire.com/news-release/2019/07/11/1881713/0/en/Technology-Innovators-Launch-New-Incubator-to-Accelerate-Adoption-of-Self-Sovereign-Identity.html

shashanksaxena18 (Fri, 12 Jul 2019 07:51:16 GMT):
Has joined the channel.

shashanksaxena18 (Fri, 12 Jul 2019 07:51:17 GMT):
I am working on hyperledger indy playform, while development i need to run /manage build and /manage up command after any changes in the code, anyone have solution for this tedious and slow process???

Zohaib_Sohail (Fri, 12 Jul 2019 09:04:00 GMT):
Got it Thanks. I will thoroughly look into this now.

BrajeshKumar (Fri, 12 Jul 2019 10:59:58 GMT):
From Where should I start Indy as a beginner ? Any suggested material ?

shashanksaxena18 (Fri, 12 Jul 2019 11:17:42 GMT):
https://github.com/hyperledger/education

BrajeshKumar (Fri, 12 Jul 2019 13:01:32 GMT):
Thank you

trinayanbhatt (Fri, 12 Jul 2019 13:01:58 GMT):
Has joined the channel.

trinayanbhatt (Fri, 12 Jul 2019 13:01:59 GMT):
Hey brajesh is it working?

BrajeshKumar (Fri, 12 Jul 2019 13:03:07 GMT):
Yes @trinayanbhatt its running, need to explore more to know the flow.

BrajeshKumar (Fri, 12 Jul 2019 13:03:59 GMT):
Yes @trinayanbhatt its running, need to explore more to know the flow.

trinayanbhatt (Fri, 12 Jul 2019 13:04:29 GMT):
okay will look into it and will ask for help if any required

yogeshquick (Fri, 12 Jul 2019 13:05:18 GMT):
Hey trinayanbhatt i just go through this example

yogeshquick (Fri, 12 Jul 2019 13:05:54 GMT):
i want to modify the schema can i do as all the images are pointing to specific url

jljordan_bcgov (Fri, 12 Jul 2019 13:10:39 GMT):
You might want to check out https://github.com/hyperledger/aries-cloudagent-python/blob/master/docs/GettingStartedAriesDev/README.md if you are looking to understand how to use verifiable credentials with your application

yogeshquick (Fri, 12 Jul 2019 13:12:55 GMT):
thanks @jljordan_bcgov . Is any running example in which we can update the schema to understand the indy concepts

jljordan_bcgov (Fri, 12 Jul 2019 13:18:30 GMT):
Quite a few different demos and info at that link By updating the schema - if you mean a credential schema - there are some examples of that in our “Greenlight” demo -

jljordan_bcgov (Fri, 12 Jul 2019 13:18:35 GMT):
https://github.com/hyperledger/aries-cloudagent-python/blob/master/docs/GettingStartedAriesDev/DecentralizedIdentityDemos.md

jljordan_bcgov (Fri, 12 Jul 2019 13:18:46 GMT):
Links in the above page

yogeshquick (Fri, 12 Jul 2019 13:22:59 GMT):
thanks @jljordan_bcgov i will go through above link and let you know if i face any issue thanks once again :)

yogeshquick (Fri, 12 Jul 2019 13:31:56 GMT):
@jljordan_bcgov is there any node js based demo available?

jljordan_bcgov (Fri, 12 Jul 2019 13:44:05 GMT):
Not a recent one

jljordan_bcgov (Fri, 12 Jul 2019 13:44:44 GMT):
There is one in the education course above ... however it is quite dated and likely to cause some confusion around the concepts of Peer DIDs and agent to agent messaging

jljordan_bcgov (Fri, 12 Jul 2019 13:45:12 GMT):
Which are core to Aries / Indy based systems

swcurran (Fri, 12 Jul 2019 15:05:13 GMT):
The aries-cloudagent-python model is that you build a controller (your app) as a pair with an agent. The controller is like a webapp, can be in any language you want. The agent sends HTTP events to the controller, and exposes an HTTP JSON API that the controller can call to control the agent. You put your business rules, UI, etc. into the controller.

swcurran (Fri, 12 Jul 2019 15:05:13 GMT):
The aries-cloudagent-python model is that you build a controller (your app) that is paired with an agent. The controller is like a webapp, can be in any language you want. The agent sends HTTP events to the controller, and exposes an HTTP JSON API that the controller can call to control the agent. You put your business rules, UI, etc. into the controller.

swcurran (Fri, 12 Jul 2019 15:06:56 GMT):
This covered in the docs @jljordan_bcgov referenced, but is a good way to get going in any language you want for a non-mobile agent.

Thomas_the_Consultant (Fri, 12 Jul 2019 15:15:48 GMT):
Is there a good overview of the messaging aspects of Indy/Ariel? Relationships between Didcomm and A2A etc.

jljordan_bcgov (Fri, 12 Jul 2019 15:19:18 GMT):
a good RFC to start with https://github.com/hyperledger/aries-rfcs/blob/master/concepts/0005-didcomm/README.md

jljordan_bcgov (Fri, 12 Jul 2019 15:19:32 GMT):
the rest are https://github.com/hyperledger/aries-rfcs/blob/master/index.md

jljordan_bcgov (Fri, 12 Jul 2019 15:27:52 GMT):
the rest are https://github.com/hyperledger/aries-rfcs/blob/master/index.md

smithsamuelm (Fri, 12 Jul 2019 16:23:44 GMT):
tried building indy-sdx on macOS https://github.com/hyperledger/indy-sdk/blob/master/docs/build-guides/mac-build.md but ran into difficulty one of the rust dependencies would not build. All the other rust dependencies built fine. The error is in the issue. but reproduced here I created a jira issuer for indy-sdk Here is the issue IS-1314 $ cargo build Compiling rusqlite v0.13.0 error[E0119]: conflicting implementations of trait `std::convert::From<&_>` for type `types::to_sql::ToSqlOutput<'_>`: --> /Users/samuel/.cargo/registry/src/github.com-1ecc6299db9ec823/rusqlite-0.13.0/src/types/to_sql.rs:26:1 | 18 | / impl<'a, T: ?Sized> From<&'a T> for ToSqlOutput<'a> 19 | | where &'a T: Into> 20 | | { 21 | | fn from(t: &'a T) -> Self { 22 | | ToSqlOutput::Borrowed(t.into()) 23 | | } 24 | | } | |_- first implementation here 25 | 26 | impl<'a, T: Into> From for ToSqlOutput<'a> { | ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ conflicting implementation for `types::to_sql::ToSqlOutput<'_>` | = note: downstream crates may implement trait `std::convert::From<&_>` for type `types::value::Value` error: aborting due to previous error For more information about this error, try `rustc --explain E0119`. error: Could not compile `rusqlite`. To learn more, run the command again with --verbose.

andrew.whitehead (Fri, 12 Jul 2019 16:34:07 GMT):
Somebody else reported this in the #indy-sdk channel, are you using the latest rust release?

andrew.whitehead (Fri, 12 Jul 2019 16:47:47 GMT):
I think it's a rusqlite issue that wasn't flagged in older versions of the compiler, but is fixed in 1.19.0: https://github.com/jgallagher/rusqlite/issues/313

andrew.whitehead (Fri, 12 Jul 2019 16:47:47 GMT):
I think it's a rusqlite issue that wasn't flagged in older versions of the compiler, but is fixed in 0.19.0: https://github.com/jgallagher/rusqlite/issues/313

andrew.whitehead (Fri, 12 Jul 2019 16:47:47 GMT):
I think it's an outdated rusqlite issue that wasn't flagged in older versions of the compiler, but is fixed in 0.19.0: https://github.com/jgallagher/rusqlite/issues/313

ianco (Fri, 12 Jul 2019 17:23:43 GMT):
FYI I'm able to successfully build with: rustc --version rustc 1.34.0 (91856ed52 2019-04-10)

andrew.whitehead (Fri, 12 Jul 2019 17:24:22 GMT):
I confirmed von-image is having the same issue building the SDK. Trying the latest rust nightly

andrew.whitehead (Fri, 12 Jul 2019 17:24:22 GMT):
I confirmed von-image is having the same issue building the SDK. Trying the latest rust nightly (reports as 1.38.0)

trinayanbhatt (Fri, 12 Jul 2019 17:50:21 GMT):
Yes, this dependency is also causing problem in linux machine while von-network

trinayanbhatt (Fri, 12 Jul 2019 17:50:21 GMT):
Yes, this dependency is also causing problem in linux machine while using von-network

andrew.whitehead (Fri, 12 Jul 2019 17:52:48 GMT):
No luck with the nightly but it appears to work fine with 1.34.0 (`./rustup --default-toolchain 1.34.0 -y`)

smithsamuelm (Fri, 12 Jul 2019 18:03:18 GMT):
$ rustc --version rustc 1.36.0

smithsamuelm (Fri, 12 Jul 2019 18:20:21 GMT):
So I am using a later rust release. Should I downgrade

smithsamuelm (Fri, 12 Jul 2019 18:22:36 GMT):
rustc version 1.36 should I downgrade?

andrew.whitehead (Fri, 12 Jul 2019 18:30:50 GMT):
@smithsamuelm That's my thinking, yes

MALodder (Fri, 12 Jul 2019 20:07:33 GMT):
I'm not seeing this problem and I'm running rust 1.36

MALodder (Fri, 12 Jul 2019 20:07:49 GMT):
I think this might be related to what version of libsqlite.so you have installed for linux

MALodder (Fri, 12 Jul 2019 20:09:02 GMT):
trying installing sqlite libraries in a different folder and setting the SQLITE3_LIB_DIR environment variable to that

smithsamuelm (Fri, 12 Jul 2019 20:28:31 GMT):
2 @malodder. I am on macOS What version of SQLite should I be using or what are you using?

smithsamuelm (Fri, 12 Jul 2019 20:31:53 GMT):
I will play around with SQLite thanks

smithsamuelm (Fri, 12 Jul 2019 21:23:27 GMT):
The default sqlite3 for macOS mojave is 3.24. The latest homebrew install is 3.28. It installs keg only because of the conflict with the macos version. However I overrode the symlinks. as follows:

smithsamuelm (Fri, 12 Jul 2019 21:24:48 GMT):
echo 'export PATH="/usr/local/opt/sqlite/bin:$PATH"' >> ~/.bash_profile

smithsamuelm (Fri, 12 Jul 2019 21:25:02 GMT):
``` export LDFLAGS="-L/usr/local/opt/sqlite/lib" export CPPFLAGS="-I/usr/local/opt/sqlite/include" ```

smithsamuelm (Fri, 12 Jul 2019 21:25:16 GMT):
export PKG_CONFIG_PATH="/usr/local/opt/sqlite/lib/pkgconfig"

smithsamuelm (Fri, 12 Jul 2019 21:25:18 GMT):
$ export SQLITE3_LIB_DIR="/usr/local/opt/sqlite/lib" $ echo $SQLITE3_LIB_DIR /usr/local/opt/sqlite/lib

smithsamuelm (Fri, 12 Jul 2019 21:25:48 GMT):
But I get the same error.

smithsamuelm (Fri, 12 Jul 2019 21:26:09 GMT):
@malodder What version of sqlite3 are you using?

smithsamuelm (Fri, 12 Jul 2019 21:26:25 GMT):
@MALodder

smithsamuelm (Fri, 12 Jul 2019 21:27:40 GMT):
Not sure if the others besides SQLITE3_LIB_DIR matter for rust build

smithsamuelm (Fri, 12 Jul 2019 22:03:10 GMT):
@MALodder I tried rustc 1.34 and it built without the error. So how would you like it documented. I assume there is some compatiblity issue with rusqlite and the version of sqilite and rustc. Based on the error its an illegal trait definition that has crept in someplace.

smithsamuelm (Fri, 12 Jul 2019 22:03:40 GMT):
I got some warning messages when building the cli. Do you care about those?

smithsamuelm (Fri, 12 Jul 2019 22:03:58 GMT):
``` Compiling indy v1.10.0 (/Data/Code/public/hyperledger/indy/indy-sdk/wrappers/rust) warning: variable does not need to be mutable --> src/commands/ledger.rs:1026:21 | 1026 | let mut sources: Vec = serde_json::from_str(&sources_json) | ----^^^^^^^ | | | help: remove this `mut` | = note: #[warn(unused_mut)] on by default warning: variable does not need to be mutable --> src/commands/ledger.rs:1085:21 | 1085 | let mut receipts: Vec = serde_json::from_str(&receipts_json) | ----^^^^^^^^ | | | help: remove this `mut` warning: variable does not need to be mutable --> src/commands/ledger.rs:1130:21 | 1130 | let mut fees: HashMap = serde_json::from_str(&fees_json) | ----^^^^ | | | help: remove this `mut` warning: variable does not need to be mutable --> src/commands/ledger.rs:1133:21 | 1133 | let mut fees = | ----^^^^ | | | help: remove this `mut` warning: variable does not need to be mutable --> src/commands/payment_address.rs:70:21 | 70 | let mut payment_addresses: Vec = serde_json::from_str(&payment_addresses_json) | ----^^^^^^^^^^^^^^^^^ | | | help: remove this `mut` Finished dev [unoptimized + debuginfo] target(s) in 45.29s ```

sumodgeorge (Sat, 13 Jul 2019 03:04:42 GMT):
Has joined the channel.

sumodgeorge (Sat, 13 Jul 2019 03:04:46 GMT):
Hi, I am trying to run the TheOrgBook using Docker compose, I followed the steps documented in https://github.com/bcgov/TheOrgBook/blob/master/docker/README.md. It went well till I reached the step - Building postgresql image. This has been running for past 30 mins, wanted to know if it does take that long to run

jljordan_bcgov (Sat, 13 Jul 2019 03:25:05 GMT):
@WadeBarnes? ^^ @sumodgeorge should not take 30 mins that is for sure ... our devs would not be happy!

sumodgeorge (Sat, 13 Jul 2019 03:27:16 GMT):
Yes Wade, that is what I also felt. I am not sure what could be triggering this to consume such a long time

sumodgeorge (Sat, 13 Jul 2019 03:31:09 GMT):
Is it possible that the source-to-image version is having some issues. I used the version https://github.com/openshift/source-to-image/releases/download/v1.1.13/source-to-image-v1.1.13-b54d75d3-darwin-amd64.tar.gz

sumodgeorge (Sat, 13 Jul 2019 03:31:25 GMT):
I am using mac

WadeBarnes (Sat, 13 Jul 2019 03:35:35 GMT):
@sumodgeorge, can you PM me the build logs and indicate where it’s hanging. The angular build typically takes the longest and that does not even come close to 30 minutes to build.

sumodgeorge (Sat, 13 Jul 2019 03:35:52 GMT):
Okay sure

WadeBarnes (Sat, 13 Jul 2019 03:36:16 GMT):
I’ll have a look at that when I can.

sumodgeorge (Sat, 13 Jul 2019 03:36:33 GMT):
Thanks Wade, Appreciate your help

MALodder (Sat, 13 Jul 2019 05:02:51 GMT):
I'm using 3.24.0

MALodder (Sat, 13 Jul 2019 05:03:01 GMT):
3.24.0

MALodder (Sat, 13 Jul 2019 05:06:08 GMT):
3.24

MALodder (Sat, 13 Jul 2019 05:07:47 GMT):
I'm on rust stable 1.35 sqlite3 3.24 on mac os x 10.14.5 and libindy builds just fine for me

MALodder (Sat, 13 Jul 2019 05:08:21 GMT):
I even ran `cargo clean` first

MALodder (Sat, 13 Jul 2019 05:08:52 GMT):
is it trying to compile rusqlite v0.13.0?

MALodder (Sat, 13 Jul 2019 05:10:02 GMT):
I'm wondering if its another dependency that sqlite requires

MALodder (Sat, 13 Jul 2019 05:11:18 GMT):
what version of xcode do you have installed

MALodder (Sat, 13 Jul 2019 05:11:48 GMT):
I'm running xcode-select version 2354.

andrew.whitehead (Sat, 13 Jul 2019 11:48:51 GMT):
The error was occurring when trying to compile rusqlite. I'm pretty sure it's related to this issue (https://github.com/jgallagher/rusqlite/issues/313), which is fixed in 0.19.0. I haven't tried updating just that dependency though, it could be the API has changed since then.

smithsamuelm (Sat, 13 Jul 2019 16:11:01 GMT):
I was using rust 1.36 which did not work. But rust 1.34 did. I did not try rust 1.35. On mojave 10.14.5. I first tried the built in sqlite3 3.24 but then based on your suggestion upgraded to the brew keg sqlite 3.26. My dev tools are

smithsamuelm (Sat, 13 Jul 2019 16:12:23 GMT):
$ system_profiler SPDeveloperToolsDataType Developer: Developer Tools: Version: 10.2.1 (10E1001) Location: /Applications/Xcode.app Applications: Xcode: 10.2.1 (14490.122) Instruments: 10.2.1 (64490.66) SDKs: iOS: 12.2: (16E226) iOS Simulator: 12.2: (16E226) macOS: 10.14: (18E219) tvOS: 12.2: (16L225) tvOS Simulator: 12.2: (16L225) watchOS: 5.2: (16T224) watchOS Simulator: 5.2: (16T224)

smithsamuelm (Sat, 13 Jul 2019 16:12:31 GMT):
$ xcode-select --version xcode-select version 2354.

smithsamuelm (Sat, 13 Jul 2019 16:14:53 GMT):
@andrew.whitehead it is a problem with rusqlite 0.13.0 but I am using rustc 1.36.0 so the problem if it is the same one that was fixed before is not reocurring.

smithsamuelm (Sat, 13 Jul 2019 16:19:05 GMT):
The error is a trait definition error conflict. I don't know rust so I don't know where to look to fix it. I suspect that the error has nothing to do with the sqlite3 library version 1.34 etc but is a problem with rusqlite code itself and it is conflicting with something in the rustc library that rustc 1.36 introduced. Maybe tighter enforcement of the trait conflicts.

smithsamuelm (Sat, 13 Jul 2019 16:24:39 GMT):
@MALodder I just tried building with rustc 1.35 and it built fine. So could you confirm by building with rustc 1.36 on your machine to see if you get the error?

smithsamuelm (Sat, 13 Jul 2019 16:27:38 GMT):
It appears to be a new bug with rusqlite 0.13.0 and rustc stable 1.36

MALodder (Sat, 13 Jul 2019 16:28:06 GMT):
Sure I’ll try now

MALodder (Sat, 13 Jul 2019 16:29:28 GMT):
Yep I get the error now

smithsamuelm (Sat, 13 Jul 2019 16:29:40 GMT):
:sweat_smile:

smithsamuelm (Sat, 13 Jul 2019 16:30:15 GMT):
:innocent:

MALodder (Sat, 13 Jul 2019 16:30:41 GMT):
so its a rust 1.36 incompatibility with rusqlite

smithsamuelm (Sat, 13 Jul 2019 16:30:45 GMT):
What to do now with it?

MALodder (Sat, 13 Jul 2019 16:30:57 GMT):
I would use rust 1.35

smithsamuelm (Sat, 13 Jul 2019 16:31:28 GMT):
Should report it to rust?

MALodder (Sat, 13 Jul 2019 16:31:29 GMT):
unless there is a rust 1.36 specific feature you are needing

MALodder (Sat, 13 Jul 2019 16:31:41 GMT):
no report it to the rusqlite crate owner

smithsamuelm (Sat, 13 Jul 2019 16:31:46 GMT):
No nothing that I know of.

MALodder (Sat, 13 Jul 2019 16:32:04 GMT):
but it looks like rusqlite might already be aware of it

smithsamuelm (Sat, 13 Jul 2019 16:32:41 GMT):
Oh the joys of building with the latest versions.

smithsamuelm (Sat, 13 Jul 2019 16:33:19 GMT):
Thanks for all the help

mkdogan (Sun, 14 Jul 2019 11:49:25 GMT):
Has joined the channel.

mkdogan (Sun, 14 Jul 2019 11:49:27 GMT):
@smithsamuelm did downgrading fix the issue?

sanket1211 (Mon, 15 Jul 2019 07:43:31 GMT):
Has joined the channel.

Vikrum (Mon, 15 Jul 2019 11:50:27 GMT):
Has left the channel.

santmukh (Mon, 15 Jul 2019 13:39:54 GMT):
Has joined the channel.

ra_w (Mon, 15 Jul 2019 16:44:32 GMT):
Has joined the channel.

Thomas_the_Consultant (Mon, 15 Jul 2019 17:09:26 GMT):
So I shouldn't compare what's happening in Connect.me or Aries to Ethereum's Whisper network? It's all different transports and the default is HTTPS. Is it usually HTTPS?

smithsamuelm (Mon, 15 Jul 2019 18:08:16 GMT):
@mkdogan downgrading rustc to 1.35 or 1.34 works. @malodder confirmed bug with 1.36. So its problem between rustc 1.36 and rusqlite 0.13

luckycharms810 (Mon, 15 Jul 2019 20:04:13 GMT):
Not sure if this is the right channel for this, but I found a really weird bug in the Indy python SDK when I used it w/ python3.6

luckycharms810 (Mon, 15 Jul 2019 20:05:28 GMT):
Basically I found that calling ```wallet.close_wallet(handle) ```, did not close any of the open file handles to the sqlite.db file, and furthermore they did not collapse the changes from the WAL file back in to the main file.

luckycharms810 (Mon, 15 Jul 2019 20:05:40 GMT):
I also noticed the -shm file also hangs around too.

alank9 (Mon, 15 Jul 2019 23:14:36 GMT):
Has joined the channel.

CarlosAvim (Tue, 16 Jul 2019 06:35:16 GMT):
Hello all, looking for an advice about which wrapper to choose. I know how to code in Java, Rust and Python (I don't mind learning a new) so I am wondering which one of Indy SDK's wrapper is the most "complete" and maintained (up to date with latest dev). Thanks!

andrew.whitehead (Tue, 16 Jul 2019 07:15:43 GMT):
As far as I know they have equivalent functionality at the moment

CarlosAvim (Tue, 16 Jul 2019 09:16:45 GMT):
I see, thanks.

lucafra (Tue, 16 Jul 2019 10:16:11 GMT):
Hi All, i am having problem in setting up the "python ref agent" in my AWS instance. I am running python 3.6. The docker build fails because of this: ModuleNotFoundError: No module named 'python_agent_utils'. I cannot find this missing module anywhere..Someone can help?

Thomas_the_Consultant (Tue, 16 Jul 2019 11:46:19 GMT):
Am I right in thinking that the way did:peer generally works is... 1. User Agent resolves DID of a credential issuer on Sovrin chain to an HTTPS end-point. 2. Introduces itself to the HTTPS end-point of the issuer, passing over a pair-wise DID whom's own end-point is the User Agent's Cloud Agent. 3. The response from the credential issuer is passed to the User's Cloud Agent using the A2A protocol (I presume that's what Zero MQ libs are there for). This response will be encrypted with Authcrypt and not readable to the Cloud Agent. 4. Cloud Agent will send a mobile push to the User Agent, or alternatively the User Agent will just poll to collect the response. ? Sorry if this is basic, I'm just find that there's a wall of docs out there when I Google. Everything seems possible, somewhat less seems documented :)

ravip (Tue, 16 Jul 2019 12:45:37 GMT):
How do I clear the log file that is generated at http://localhost:9000/ledger/domain? It seems to have reached the max file size limit and any new transactions that are created on the ledger are not visible.

drummondreed (Tue, 16 Jul 2019 13:38:52 GMT):
discovery with did:peer:

swcurran (Tue, 16 Jul 2019 15:16:21 GMT):
The bigger question is why start with the indy-sdk. Depending on what you are doing, a more productive route is to build an agent based on Agent code - https://github.com/hyperledger/aries-cloudagent-python Check out the Getting Started Guide link in that repo.

swcurran (Tue, 16 Jul 2019 15:16:21 GMT):
The bigger question is why start with the indy-sdk at all? Depending on what you are doing, a more productive route is to build an agent based on Agent code (which embeds the indy-sdk) - https://github.com/hyperledger/aries-cloudagent-python Check out the Getting Started Guide link in that repo.

bbehlendorf (Tue, 16 Jul 2019 17:30:22 GMT):
If anyone has any deployments either launched or ready to launch that have any sort of public-facing side, consider submitting them for an ID2020 Certification Mark - it's something I advised them on, and they're collecting an initial set of certified deployments to demonstrate its value. See: https://docs.google.com/forms/d/e/1FAIpQLSe2F5IVTmNypE05LmzacWEMWlN9UEyRIUanKDBK5FhvzjZ-Lw/viewform?embedded=true and https://medium.com/id2020/id2020-launches-technical-certification-mark-e6743d3f70fd

ravip (Tue, 16 Jul 2019 20:23:59 GMT):
Let's say, there is an application that uses decentralized identity mechanism to recognize the participants and it is a requirement that user wallet needs to be portable, then where does the cloud wallet (cloud agent) live in this case? I mean who hosts the individual user's cloud wallets?

andrew.whitehead (Wed, 17 Jul 2019 00:24:37 GMT):
Are you sure you `await`ed the result? :)

CarlosAvim (Wed, 17 Jul 2019 05:22:38 GMT):
Hey @swcurran I wasn't aware of that project. I think that it's what I was looking for! Thanks alot.

AvikHazra-klizos (Wed, 17 Jul 2019 06:02:10 GMT):
Has joined the channel.

biml4 (Wed, 17 Jul 2019 13:40:31 GMT):
Has joined the channel.

biml4 (Wed, 17 Jul 2019 13:40:32 GMT):
Hi there, I am new to Hyperledger Indy and I was wondering if someone can guide on where to start please? Thanks :)

luckycharms810 (Wed, 17 Jul 2019 14:22:40 GMT):
Yep! I basically have been running the same agent code across a ubuntu 16.04 server and a 18.04 server

luckycharms810 (Wed, 17 Jul 2019 14:23:48 GMT):
Running on 18.04, the same functions fail to close the wallet

luckycharms810 (Wed, 17 Jul 2019 14:24:01 GMT):
18.04 packages python3.6 as the default python3

luckycharms810 (Wed, 17 Jul 2019 14:24:10 GMT):
16.04 packages python3.5

luckycharms810 (Wed, 17 Jul 2019 14:24:33 GMT):
I've been using both lsof to check open file handles, and the presence of the shm and wal files.

luckycharms810 (Wed, 17 Jul 2019 14:24:33 GMT):
I've been using both`lsof` to check open file handles, and the presence of the shm and wal files.

swcurran (Wed, 17 Jul 2019 14:51:58 GMT):
Aries is an agent that runs on top of indy-sdk and can be used to implement applications. Here is a "0 to coding" guide for Indy/Aries. https://github.com/hyperledger/aries-cloudagent-python/tree/master/docs/GettingStartedAriesDev

Thomas_the_Consultant (Wed, 17 Jul 2019 15:55:45 GMT):
Do pairwise DIDs have DID documents in any meaningful way? And, if so, what should be in them?

swcurran (Wed, 17 Jul 2019 16:05:52 GMT):
Yes. You can look at the did:peer spec (https://openssi.github.io/peer-did-method-spec/index.html) and specifically for DIDcomm, take a look at the RFC - https://github.com/hyperledger/aries-rfcs/tree/master/features/0067-didcomm-diddoc-conventions

swcurran (Wed, 17 Jul 2019 16:07:09 GMT):
There is an article in the Aries Getting Started guide that talks about Routing that goes over it as well.

luckycharms810 (Wed, 17 Jul 2019 16:33:10 GMT):
Oh yeah, I guess an important piece of information is that the wallet does close on ubuntu 16.04 + python3.5

lachid (Wed, 17 Jul 2019 16:35:04 GMT):
is Ledger.BuildGetDdoRequestAsync working?? if so.. how is the correct use_? im getting an InvalidClientRequest('validation error [ClientAuthRuleOperation]: missed fields - field, auth_action, new_value, constraint, auth_type. ',)

andrew.whitehead (Wed, 17 Jul 2019 16:35:34 GMT):
You could try setting RUST_LOG=DEBUG for additional output on the SDK side, or PYTHONASYNCIODEBUG=1 to check the Python asyncio side

ViaSky (Wed, 17 Jul 2019 17:23:12 GMT):
Has joined the channel.

SeanBohan (Wed, 17 Jul 2019 19:04:19 GMT):
Folks: Indy Working Group Call When: Thurs 7/18 at 4pmBST, 11amET, 9amMT, 8amPT Where: https://zoom.us/j/232861185 Agenda: consent Receipt demo with JanL & Open Discussion

lynn.bendixsen (Wed, 17 Jul 2019 22:58:55 GMT):
I am having some recurring warnings in my Nodes log file that I would like to understand better. Questions: 1. What is batched.py? 2. Why is it writing to my log file ~8 times per minute (removing the same items over and over)? ```2019-07-17 22:51:09,122|WARNING|batched.py|CONNECTION: FoundationBuilder has removed rid DacmhXMW9xYRCy6R3fAj54iCKCD8BdJnwyMGgt329cdy 2019-07-17 22:51:20,323|WARNING|batched.py|CONNECTION: FoundationBuilder has removed rid 5AcGx1KMbranvdoyPJtME1K7q9dKbP6oCBPCkDQjkyTU 2019-07-17 22:51:20,733|INFO|zstack.py|FoundationBuilder reconnecting to danube:HA(host='173.249.14.196', port='9701') 2019-07-17 22:51:25,008|WARNING|batched.py|CONNECTION: FoundationBuilder has removed rid DacmhXMW9xYRCy6R3fAj54iCKCD8BdJnwyMGgt329cdy 2019-07-17 22:51:36,282|WARNING|batched.py|CONNECTION: FoundationBuilder has removed rid 5AcGx1KMbranvdoyPJtME1K7q9dKbP6oCBPCkDQjkyTU 2019-07-17 22:51:39,137|WARNING|batched.py|CONNECTION: FoundationBuilder has removed rid DacmhXMW9xYRCy6R3fAj54iCKCD8BdJnwyMGgt329cdy 2019-07-17 22:51:50,338|WARNING|batched.py|CONNECTION: FoundationBuilder has removed rid 5AcGx1KMbranvdoyPJtME1K7q9dKbP6oCBPCkDQjkyTU 2019-07-17 22:51:50,749|INFO|zstack.py|FoundationBuilder reconnecting to danube:HA(host='173.249.14.196', port='9701') 2019-07-17 22:51:55,005|WARNING|batched.py|CONNECTION: FoundationBuilder has removed rid DacmhXMW9xYRCy6R3fAj54iCKCD8BdJnwyMGgt329cdy 2019-07-17 22:52:06,300|WARNING|batched.py|CONNECTION: FoundationBuilder has removed rid 5AcGx1KMbranvdoyPJtME1K7q9dKbP6oCBPCkDQjkyTU 2019-07-17 22:52:09,145|WARNING|batched.py|CONNECTION: FoundationBuilder has removed rid DacmhXMW9xYRCy6R3fAj54iCKCD8BdJnwyMGgt329cdy 2019-07-17 22:52:20,362|WARNING|batched.py|CONNECTION: FoundationBuilder has removed rid 5AcGx1KMbranvdoyPJtME1K7q9dKbP6oCBPCkDQjkyTU 2019-07-17 22:52:20,761|INFO|zstack.py|FoundationBuilder reconnecting to danube:HA(host='173.249.14.196', port='9701') 2019-07-17 22:52:25,013|WARNING|batched.py|CONNECTION: FoundationBuilder has removed rid DacmhXMW9xYRCy6R3fAj54iCKCD8BdJnwyMGgt329cdy 2019-07-17 22:52:29,117|WARNING|zstack.py|Remote danube is not connected - message will not be sent immediately.If this problem does not resolve itself - check your firewall settings 2019-07-17 22:52:29,243|WARNING|zstack.py|Remote danube is not connected - message will not be sent immediately.If this problem does not resolve itself - check your firewall settings 2019-07-17 22:52:29,341|INFO|replica.py|FoundationBuilder:0 set last ordered as (11, 10737) 2019-07-17 22:52:29,342|INFO|replica.py|FoundationBuilder:0 ordered batch request, view no 11, ppSeqNo 10737, ledger 2, state root bngFNnXiy24vZhkbrc1vCCiAdtBrreKwMtiErc5484z, txn root DZa2yp6V7emd71tvYYz3T7pwydx7fqDkyJ6ebnAF5ocq, audit root CLdw9ixLApGqtS9TSfy9eCw4Sur2Go3pohzPskCH2dYb, requests ordered 0, discarded 0 2019-07-17 22:52:36,312|WARNING|batched.py|CONNECTION: FoundationBuilder has removed rid 5AcGx1KMbranvdoyPJtME1K7q9dKbP6oCBPCkDQjkyTU 2019-07-17 22:52:39,146|WARNING|batched.py|CONNECTION: FoundationBuilder has removed rid DacmhXMW9xYRCy6R3fAj54iCKCD8BdJnwyMGgt329cdy```

Thomas_the_Consultant (Wed, 17 Jul 2019 23:11:11 GMT):
Supposing I have a wallet (Streetcred/Connect.me) installed on a smart phone alongside the Faber University app... I want the user to click a button on the Faber app, and then be switched into wallet to receive a Credential. The Faber App (which doesn't know the user's DID/public key etc) authenticates with a backend using a shared secret, the Faber Agent [having a DIDComm end-point and the ability to issue credentials] will need this shared secret to authenticate on the user's behalf and find out the grades for the credential. What's the sensible way to do this? How do we generally pass cookies around to re-identify users between the HTTPS/REST and DIDComm/Indy world?

swcurran (Wed, 17 Jul 2019 23:31:23 GMT):
This has been discussed in a previous Aries (or Indy Agent call) that I missed. If the user is on a desktop, a QR code can be used. Doesn't work well on a phone :-). As I recall, there are two options - a registered file type (or protocol??) handler that is associated with the Wallet app, or a deeplink to a specific app. As I understand it, the problem with the deeplink is that it doesn't work across mobile eco-systems, or across apps - it is to a specific app. So it seems like the "registered file type" seems to be the best approach, but I think there are problems with that as well. Note that this isn't a DIDComm specific issue - it's any link in a mobile browser/mobile app that is expected to open a class of apps. Further, you really want it to work such if you don't have any app of that class installed, the link triggers a way for you to install an app...

PaulA (Thu, 18 Jul 2019 05:37:05 GMT):
Has joined the channel.

Thomas_the_Consultant (Thu, 18 Jul 2019 07:22:22 GMT):
I'd be great if I could suggest to the wallet that it make a certain Credentials Request including some sort of metadata that's meaningful to the issuer.

Thomas_the_Consultant (Thu, 18 Jul 2019 07:25:34 GMT):
In the QR code on a displayed on a desktop case, what data does the QR code contain?

ashcherbakov (Thu, 18 Jul 2019 07:49:50 GMT):
> 1. What is batched.py? This is a wrapper class for sending messages by batches on the transport layer (to be more efficient when dealing with a lot of small messages)

andrew.whitehead (Thu, 18 Jul 2019 07:50:45 GMT):
I think what you’re suggesting might be covered by the credential proposals covered here? https://github.com/hyperledger/aries-rfcs/blob/master/features/0036-issue-credential/README.md As to the QR code, in that case I think it would contain a presentation request? Although I’m not totally clear on the use case

ashcherbakov (Thu, 18 Jul 2019 07:56:48 GMT):
> 2. Why is it writing to my log file ~8 times per minute (removing the same items over and over)? This warning indicates that a Node can not send a Reply to a Remote with the given ID. There can be different reasons why Remote is removed: a Node has been demoted, Node's IP/port/keys have been changed, etc. As for why it repeatedly logs this, it needs to be investigated. Can you please create a ticket and put a full log there with a description of the use case where it's reproduced? Additional information about changing the pool state will be very useful as well (for example whether some nodes have been demoted, etc.)

Thomas_the_Consultant (Thu, 18 Jul 2019 08:05:32 GMT):
Hmm, thinking about it, the ideal would be if I could pass an additional parameter to the Issuing Agent in the connection request. So... ``` { "@type": "did:sov:BzCbsNYhMrjHiqZDTUASHg;spec/didexchange/1.0/invitation", "@id": "12345678900987654321", "label": "Alice", "recipientKeys": ["8HH5gYEeNc3z7PYXmd54d4x6qAfCNrqQqEB3nS7Zfu7K"], "serviceEndpoint": "https://example.com/endpoint", "routingKeys": ["8HH5gYEeNc3z7PYXmd54d4x6qAfCNrqQqEB3nS7Zfu7K"], "some_bearer_token_I_hope_wont_break_the_validation": "133t_t0k3n" } ``` And then `133t_t0k3n` can get pulled out by the issuing agent to work out who is making the request.

Thomas_the_Consultant (Thu, 18 Jul 2019 08:06:32 GMT):
Before I was allowed to talk to outside ppl, I was sent an email with 24-point red text telling me not to be clear about the use-case :) So apologise for the vagueness.

andrew.whitehead (Thu, 18 Jul 2019 08:07:16 GMT):
That's usually covered by the recipient key in a connection request, which can be tied to a specific connection invitation (which might have special privileges)

andrew.whitehead (Thu, 18 Jul 2019 08:09:00 GMT):
I didn't know there was a welcome email with big red text now :)

Thomas_the_Consultant (Thu, 18 Jul 2019 08:13:52 GMT):
Doesn't the request key need to be the agent's public key? How do I add information to that?

andrew.whitehead (Thu, 18 Jul 2019 08:27:46 GMT):
That’s the sender key in this case

Thomas_the_Consultant (Thu, 18 Jul 2019 10:57:47 GMT):
So where would I include `133t_t0k3n` in the connection request to ensure it's picked up and included in the exchange request? In the service end-point URL? Or is this completely the wrong way to think about it, and what I want to do is pass the `133t_t0k3n` to the Issuer out of band, who will then pass back a private DID specifically for this client?

Thomas_the_Consultant (Thu, 18 Jul 2019 10:58:08 GMT):
The big red text is from my product lead xD

nage (Thu, 18 Jul 2019 14:17:47 GMT):
A reminder that there will be an Indy WG call in 45 minutes here https://zoom.us/j/232861185 See https://wiki.hyperledger.org/pages/viewpage.action?pageId=13862386

mattatkiva (Thu, 18 Jul 2019 16:04:48 GMT):
Has joined the channel.

mattatkiva (Thu, 18 Jul 2019 16:05:44 GMT):
I'm looking for a sample or documentation for creating and updating an AUTH NYM transaction (preferrably using indy-cli but not required)

mattatkiva (Thu, 18 Jul 2019 16:05:44 GMT):
I'm looking for a sample or documentation for creating and updating an AUTH_RULES NYM transaction (preferrably using indy-cli but not required)

kdenhartog (Thu, 18 Jul 2019 18:21:10 GMT):
@mattatkiva this is the only documentation that I'm aware of https://github.com/sovrin-foundation/libsovtoken/blob/master/doc/fees.md

swcurran (Thu, 18 Jul 2019 18:22:47 GMT):
You don't need to include the extra token because you (the issuer) created the receipientKey and can use that to get back to the context of to whom you sent that invitation. An example is the email verification service we (BC Gov) built. At the service, you put in your email, we send you an email with a link to an invitation. When you use that invitation to connect, we can associate that back to the email address you entered. https://email-verification.vonx.io/

swcurran (Thu, 18 Jul 2019 18:24:20 GMT):
It usually contains the invitation itself. It might be wrapped in a way to trigger your User Agent to process it in a certain way - e.g. to have it open a certain app to process the request. For example, the invitation might be embedded in a URL.

lynn.bendixsen (Thu, 18 Jul 2019 18:36:25 GMT):
Hi @mattatkiva, I don't know of any better documentation that what Kyle shared, but I can help with some samples as I have been doing quite a bit of auth_rules setting and testing on the Sovrin Foundation's BuilderNet. For example the auth_rule that I set for a privileged network was ```ledger auth-rule action=ADD field=role txn_type=NYM old_value=* constraint="{"constraint_id":"OR","auth_constraints":[{"sig_count":1,"role":"0","constraint_id":"ROLE","need_to_be_owner":false},{"sig_count":1,"role":"2","constraint_id":"ROLE","need_to_be_owner":false},{"sig_count":1,"role":"101","constraint_id":"ROLE","need_to_be_owner":false}]}"``` That command sets it up so that one signature that has role Trustee(0), Steward(2), OR Endorser(101) can add an unprivileged nym to the ledger(no role). I can send a sample of what I used for a privileged ledger also (fees enabled) if that helps, or I can answer more questions that you might have.

lynn.bendixsen (Thu, 18 Jul 2019 18:36:25 GMT):
Hi @mattatkiva, I don't know of any better documentation than what Kyle shared, but I can help with some samples as I have been doing quite a bit of auth_rules setting and testing on the Sovrin Foundation's BuilderNet. For example the auth_rule that I set for a privileged network was ```ledger auth-rule action=ADD field=role txn_type=NYM old_value=* constraint="{"constraint_id":"OR","auth_constraints":[{"sig_count":1,"role":"0","constraint_id":"ROLE","need_to_be_owner":false},{"sig_count":1,"role":"2","constraint_id":"ROLE","need_to_be_owner":false},{"sig_count":1,"role":"101","constraint_id":"ROLE","need_to_be_owner":false}]}"``` That command sets it up so that one signature that has role Trustee(0), Steward(2), OR Endorser(101) can add an unprivileged nym to the ledger(no role). I can send a sample of what I used for a privileged ledger also (fees enabled) if that helps, or I can answer more questions that you might have.

lynn.bendixsen (Thu, 18 Jul 2019 18:36:25 GMT):
Hi @mattatkiva, I don't know of any better documentation than what Kyle shared, but I can help with some samples as I have been doing quite a bit of auth_rules setting and testing on the Sovrin Foundation's BuilderNet. For example, the auth_rule that I set for a privileged network was ```ledger auth-rule action=ADD field=role txn_type=NYM old_value=* constraint="{"constraint_id":"OR","auth_constraints":[{"sig_count":1,"role":"0","constraint_id":"ROLE","need_to_be_owner":false},{"sig_count":1,"role":"2","constraint_id":"ROLE","need_to_be_owner":false},{"sig_count":1,"role":"101","constraint_id":"ROLE","need_to_be_owner":false}]}"``` That command sets it up so that one signature that has role Trustee(0), Steward(2), OR Endorser(101) can add an unprivileged nym to the ledger(no role). I can send a sample of what I used for a privileged ledger also (fees enabled) if that helps, or I can answer more questions that you might have.

nage (Thu, 18 Jul 2019 18:42:47 GMT):
The implementation work for identity inside of Hyperledger has gotten to a point where there is a significant amount of cross-project collaboration. In particular we have a lot of Indy+Aries+Ursa discussions about API surfaces, release planning and use cases that has always been going on in the Thursday Indy WG call. Since the nature of that collaboration now spans multiple projects with the three we mentioned, we are moving that call underneath the Identity WG's management and officially opening it up to all Identity Implementation work (we didn't want it to distract from the higher-level whitepaper or use case work already going on in the bi-weekly call we have now). So if you're writing identity code for a Hyperledger project, or anything that intends to be interoperable with a Hyperledger project, please join us for the first Identity Implementers call next Thursday at 9 AM mountain time. We will be discussing whether to have the call every week or every other week, and looking for any other working groups from the projects that would like to report status or discuss progress.

mattatkiva (Thu, 18 Jul 2019 18:43:53 GMT):
thank you :) I am specifically looking to create a nym transaction that sets the number trustees required for consensus. Do you know how to do that?

mattatkiva (Thu, 18 Jul 2019 18:44:52 GMT):
will you be sending out a meeting invite for that WG?

nage (Thu, 18 Jul 2019 18:45:15 GMT):
we will be updating the calendar appointment in the Hyperledger Community Calendar

nage (Thu, 18 Jul 2019 18:45:54 GMT):
the zoom info should be the same as the existing Indy WG call

swcurran (Thu, 18 Jul 2019 18:49:10 GMT):
An observation: While it definitely makes sense for combining the three projects into a single call, focusing the name on identity vs. the broader verifiable credentials/trusted data use cases is a bit limiting. We spend a lot of energy pushing on the possibilities beyond identity.

lynn.bendixsen (Thu, 18 Jul 2019 18:54:53 GMT):
Yes, I think so. (If I understand you correctly.)

nage (Thu, 18 Jul 2019 18:57:54 GMT):
we could take up a WG rebranding, but it seems a bit obtuse. I'm trying to work with the groups we have

swcurran (Thu, 18 Jul 2019 18:58:46 GMT):
Makes sense. Hence the observation :-). Just making sure the message is there.

wadafud (Fri, 19 Jul 2019 02:48:46 GMT):
Has joined the channel.

lucafra (Fri, 19 Jul 2019 09:54:42 GMT):
Hi All, is there a way to continue with this Agent demo sending credentials? https://github.com/hyperledger/indy-agent/blob/master/python/README.md

dbluhm (Fri, 19 Jul 2019 11:48:17 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=z8mN7fzQoxpToQx2f) @lucafra I highly recommend you check out the aries-cloudagent-python repository and use that agent over the Python Indy reference agent. We haven't officially added notices to the indy-agent repo yet (we need to) but it has been superseded by the Aries project and most components in there have corresponding (and much improved) code in the Aries project.

ShrinivasDeshmukh (Fri, 19 Jul 2019 14:12:02 GMT):
Hi everyone, I am not getting exactly how will the edge agent and cloud agent will bind (or tunnel). What I am trying to implement is that the cloud agent will receive the message, check the format (protocol) and forward it to the edge agent. Edge agent will decrypt the message using his private key stored in his wallet. Is my understanding correct? Please can someone help me with this?

dbluhm (Fri, 19 Jul 2019 14:37:08 GMT):
In most cases, an edge agent will not have a persistent endpoint that the cloud agent can connect to at will; for example, a mobile phone does not have a consistent IP Address and a web server running, accepting POST requests. The edge agent must initiate the connection to the cloud agent in order for the cloud agent to deliver messages to the edge that it received on it's behalf. The edge agent alone holds the keys to decrypt the message, as you correctly pointed out. These are the kinds of discussions that are being held in the #aries channel which is the cross-ledger interoperable agents project Hyperledger Aries :slight_smile:

dbluhm (Fri, 19 Jul 2019 14:38:22 GMT):
I also highly recommend you look into the aries-rfcs repository for more reading on agents and how they communicate: https://github.com/hyperledger/aries-rfcs

ShrinivasDeshmukh (Fri, 19 Jul 2019 14:43:23 GMT):
Thank you @dbluhm. So the cloud agent will hold the messages until the edge initiates the connection. And once edge agent initiates the connection, all the queued message will be delivered to the edge agent at once right? And the connection has to be initiated by the edge agent only, since there cannot be persistent binding because of no static IP of the edge agent. Is this correct?

dbluhm (Fri, 19 Jul 2019 14:58:14 GMT):
We've discussed a few ways this could be done. One way is the edge agent initiates the connection and the cloud agent bundles all queued messages into a single response, as you've described. Another way that could be beneficial for very simple agents (static agents, for instance, perhaps running on an IoT device) is the cloud agent responds with a single message each time but indicates that more messages are pending so the edge agent continues to post a message that triggers the sending of another message in the pending queue until the pending count reaches 0. This is described in detail in the [transport return route Aries RFC](https://github.com/hyperledger/aries-rfcs/tree/master/features/0092-transport-return-route). Another option is that the edge agent opens a WebSocket or other bidirectional transport to the cloud agent and all the messages could be transferred over the socket individually.

ShrinivasDeshmukh (Fri, 19 Jul 2019 15:06:41 GMT):
Thanks a lot! We'll proceed with the websockets implementation for now.!

Thomas_the_Consultant (Fri, 19 Jul 2019 15:09:48 GMT):
Ah, request rather than recipient.

swcurran (Fri, 19 Jul 2019 15:28:31 GMT):
@ShrinivasDeshmukh - if you want to see an example of this, there is a demo "performance.py" in the aries-cloudagent-python `demo` folder. See the docs for how to run it - you have to pass into the startup a third parameter set to True. When you do that, the demo spins up three agents - Alice, Alice's routing agent and Faber. Faber then issues and Alice receives 300 verifiable credentials and timing information is generated. In the code you can see how Alice and her routing agent collaborate. The demo readme shows only how to run the demo in default (Alice - Faber directly), but you can tweak the startup to add the third parameter. Let me know if you have trouble with that and we can help. The easiest way to just run it is to use the "In Your Browser" method in the readme. As you get deeper into the code, you can use the "Running in Docker" approach, or if you really want - running locally.

swcurran (Fri, 19 Jul 2019 15:29:18 GMT):
Link to readme - https://github.com/swcurran/aries-cloudagent-python/tree/master/demo#performance-demo

ShrinivasDeshmukh (Fri, 19 Jul 2019 15:35:40 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=3CfabCTmEsfY2Tk8Z) @swcurran Thanks @swcurran , I've cloned the repository and have started with the code. Will definitely go through the performance.py and get back to you!

ShrinivasDeshmukh (Fri, 19 Jul 2019 15:37:34 GMT):
Also, kudos to the aries team. The code is very well structured and easy to understand. Thanks a lot!

swcurran (Fri, 19 Jul 2019 16:48:35 GMT):
FYI - the BC Gov team has initiated an open source project (via an open source "Code With Us" bounty) to implement a OpenIDConnect Identity Provider that works using Verifiable Credentials. Mattr Global is doing the work with BC Gov. The initial repo for the effort has been created and a first draft of a document that outlines many of the details of how it is going to work is in the repo. Check it out here - https://github.com/bcgov/vc-authn-oidc. Please feel free to help with this effort - add issues, ask questions and add pull requests. We're hoping to have something that can be easily used with any IAM in any deployment.

daidoji (Fri, 19 Jul 2019 20:57:55 GMT):
Did this edx course disappear? https://www.edx.org/course/blockchain-for-business-an-introduction-to-hyperledger-technologies

daidoji (Fri, 19 Jul 2019 20:58:06 GMT):
its 404'ing now

daidoji (Fri, 19 Jul 2019 20:59:13 GMT):
oh whoops, guess it moved URLS

daidoji (Fri, 19 Jul 2019 20:59:13 GMT):
https://www.edx.org/course/blockchain-for-business-an-introduction-to-hyperledger-technologies-2

swcurran (Fri, 19 Jul 2019 20:59:25 GMT):
Funny - I just discovered that. It's still there - https://www.edx.org/course/blockchain-for-business-an-introduction-to-hyperledger-technologies-2

daidoji (Fri, 19 Jul 2019 21:01:10 GMT):
@swcurran do you know if we can get the certificate after we audit? Or is it required we decide at this time?

swcurran (Fri, 19 Jul 2019 21:02:43 GMT):
I'm not certain. I would bet that you can audit and then pay, redo the tests and get the certificate, but not certain.

daidoji (Fri, 19 Jul 2019 21:04:06 GMT):
roger

swcurran (Fri, 19 Jul 2019 21:04:39 GMT):
I did some of the content, but am not involved at that level. Perhaps @Silona knows. Or knows someone that knows :-)

daidoji (Fri, 19 Jul 2019 21:05:13 GMT):
haha no worries, I'll just purchase and then take my chances with our company's reimbursement procedure :)

daidoji (Fri, 19 Jul 2019 21:05:20 GMT):
it should be all right

daidoji (Fri, 19 Jul 2019 21:05:29 GMT):
thanks

andrew.whitehead (Sat, 20 Jul 2019 03:23:10 GMT):
I also wrote a draft of a message queue protocol (should probably be called a queue management protocol): https://hackmd.io/tdMkVAcGTS-nAAX1j6W7ZQ

samaganamkarthik (Sat, 20 Jul 2019 09:24:35 GMT):
Has joined the channel.

kukkig (Mon, 22 Jul 2019 01:57:14 GMT):
Has joined the channel.

kukkig (Mon, 22 Jul 2019 01:57:15 GMT):
Hi everyone, I have a question. From what I understand, when the wallet is created there is a key for a wallet right. So what i want to know is where does this key is being stored ? is it inside the wallet DB or it is else where ? And another thing that I am confusing, does the credential has a key like a wallet ?

kukkig (Mon, 22 Jul 2019 01:57:15 GMT):
Hi everyone, I have a question. From what I understand, when the wallet is created there is a key for a wallet right. So what i want to know is where does this key is being stored ? is it inside the wallet DB or it is else where ? And another thing that I am confusing, does the credential has a key like a wallet ? One more thing, when there is new credential issued by an issuer, then how does the credential is being recorded into the wallet ? like.. how the key-value pair is being recorded into each tables inside the wallet DB ?

lucafra (Mon, 22 Jul 2019 08:20:53 GMT):
Thanks, I will definitely check it out

ShrinivasDeshmukh (Mon, 22 Jul 2019 09:40:30 GMT):
So if we want to use this protocol, can we include the protocol headers as a part of message payload while using AMQP of kafka?

ShrinivasDeshmukh (Mon, 22 Jul 2019 09:40:30 GMT):
So if we want to use this protocol, can we include the protocol headers as a part of message payload while using AMQP or kafka?

biml4 (Mon, 22 Jul 2019 09:53:01 GMT):
Hi everyone, wanted to ask if there is a way of using Hyperledger Indy with Hyperledger Fabric or is there a way I can store identities (DID`s) on the Fabric ledger? Thanks.

lachid (Mon, 22 Jul 2019 12:47:31 GMT):
Hi please i need some help.. I get an InvalidStructureException when I try to CreateAndStoreMyDid with a custom seed. Can anyone help_?

swcurran (Mon, 22 Jul 2019 13:50:51 GMT):
There has been talk of integrating Indy and Fabric, but AFAIK there is not an implementation. I'm curious as to whether you have an use cases in mind for integrating the two? Is there a specific goal for having DIDs on the Fabric ledger? I don't know that much about Fabric - just the main concepts at a high level.

SeanBohan (Mon, 22 Jul 2019 14:13:31 GMT):
@kdenhartog - Kuzma’s work last year was with INDY and Fabric?

biml4 (Mon, 22 Jul 2019 14:21:11 GMT):
I want to have the ability of creating an identity and trading assets with other identities.(Identities with Indy and trading assets with Fabric)

RaviJha (Mon, 22 Jul 2019 17:52:42 GMT):
Has joined the channel.

RaviJha (Mon, 22 Jul 2019 17:52:44 GMT):
Hi. Is only PostgreSQL is supported with Indy or Aries? If yes, why?

swcurran (Mon, 22 Jul 2019 17:56:31 GMT):
SQLite is also supported (it was the original) and Postgres added. The database is pluggable so others can be added. Postgres was added because we (BC Gov) needed it.

RaviJha (Mon, 22 Jul 2019 18:01:49 GMT):
Only RDBMS type DB is supported, is it so?

swcurran (Mon, 22 Jul 2019 18:03:55 GMT):
You would have to look at the API to determine if other models could be used. I would not be surprised, but don't know. The data model is very simple - key-value pairs for the most part. The only "tricky" part is the encryption and tagging.

RaviJha (Mon, 22 Jul 2019 18:10:15 GMT):
can we save multiple credentials on file system without any db?

swcurran (Mon, 22 Jul 2019 18:12:09 GMT):
If you want to do that, I would use the SQLite implementation. You need to have a database for querying and so on.

RaviJha (Mon, 22 Jul 2019 18:14:30 GMT):
Okay.. Got it.. Thanks.. In alice faber demo how do I configure PostgreSQL and store credentials in them?

swcurran (Mon, 22 Jul 2019 18:25:06 GMT):
The indy-sdk demo? Not sure of that. I think you would have to look at the postgres docs and figure out how to launch the agents talking to a database.

swcurran (Mon, 22 Jul 2019 18:25:34 GMT):
We have some other demos in other repos that do that, but not sure how it would be done from that indy-sdk repo.

kdenhartog (Mon, 22 Jul 2019 18:27:52 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=CpvDzVkHr7Oci92YA3) @SeanBohan He was an intern that worked on agents (now Aries), but also did a hackathon at the Hyperledger global forum that integrated the two projects together.

kdenhartog (Mon, 22 Jul 2019 18:28:29 GMT):
He sent me access to the code a while back, but it's not publicly viewable from what I remember.

StephenJackson (Mon, 22 Jul 2019 19:17:02 GMT):
Has joined the channel.

StephenJackson (Mon, 22 Jul 2019 19:17:04 GMT):
Not sure if this is the right place for this or if I should make an issue on the Jira board - I'm currently trying to create a revocation registry for a credential definition and I am getting a "CommonInvalidStructure" error. I am passing in the issuer's wallet handle and verinym, a null value for the revocDefType, a tag, an object with max_cred_num: 1 for the config, and a blobStorageHandle. Any advice?

StephenJackson (Mon, 22 Jul 2019 20:35:11 GMT):
Further context - I'm running the Indy SDK in an Express app inside of a Docker container, and something about this structure is making it so that the openBlobStorageWriter function isn't creating a tails folder and blob storage file when called. Trying to track down why, but I presume the issue is related to that.

lynn.bendixsen (Mon, 22 Jul 2019 23:05:22 GMT):
The error you are seeing is sometimes returned when one of your parameters is invalid, and it looks to me like revocDefType isn't supposed to be null. (I am not an expert, just googled around a bit and have seen that error for other calls...) ```revocDefType (enum string): Revocation Registry type (only CL_ACCUM is supported for now).```

lynn.bendixsen (Mon, 22 Jul 2019 23:05:22 GMT):
The error you are seeing is sometimes returned when one of your parameters is invalid, and it looks to me like revocDefType isn't supposed to be null. (I am not an expert on revocation registries, just googled around a bit and have seen that error for other calls...) ```revocDefType (enum string): Revocation Registry type (only CL_ACCUM is supported for now).```

dgunseli (Tue, 23 Jul 2019 13:43:31 GMT):
Has joined the channel.

mattatkiva (Tue, 23 Jul 2019 14:18:43 GMT):
I was wondering if I could get some performance data on indy. how many transactions etc? what kinds of transactions is that? etc. any info would be helpful

swcurran (Tue, 23 Jul 2019 14:37:11 GMT):
The best implementation of full revocation handling that I know of is in von_anchor - https://von-anchor.readthedocs.io

swcurran (Tue, 23 Jul 2019 14:37:30 GMT):
Suggest you review the code and the tests in that repo for guidance.

Silona (Tue, 23 Jul 2019 17:32:55 GMT):
https://www.hyperledger.org/blog/2019/07/23/rhythm-and-melody-how-hubs-and-agents-rock-together

dgunseli (Tue, 23 Jul 2019 20:00:22 GMT):
I have some issues when I try to install libindy on linux mint I got "E: Unable to locate package libindy" error

dbluhm (Tue, 23 Jul 2019 20:03:29 GMT):
As Linux Mint is based on Ubuntu, the packages built for Ubuntu may work but you will likely have to manually download them and install with `deb`. If that doesn't work, you will have to build from source for now.

dbluhm (Tue, 23 Jul 2019 20:05:17 GMT):
I seem to remember @esplinr mentioning a StackOverflow question about this topic but I can't seem to find it

dgunseli (Tue, 23 Jul 2019 20:09:31 GMT):
Actually I'm trying to install libindy to docker

dgunseli (Tue, 23 Jul 2019 20:10:15 GMT):
ok, done I think, change repository in dockerfile :)

mattatkiva (Tue, 23 Jul 2019 20:12:51 GMT):
hmmm. a quick search on [hyperledger-indy] didn't bring up any posts titled about performance.

esplinr (Tue, 23 Jul 2019 21:46:48 GMT):
I suggested to someone that they post on Stack Overflow, but I don't think they did.

esplinr (Tue, 23 Jul 2019 21:47:27 GMT):
@mattatkiva Please post on Stack Overflow, and I'll answer there. Tag with #hyperledger-indy

esplinr (Tue, 23 Jul 2019 21:47:45 GMT):
We've discussed this in this channel a few times this year, so I think we should get the answer somewhere more durable.

jljordan_bcgov (Tue, 23 Jul 2019 23:19:55 GMT):
Hi Fellow Community Members! We, the folks from the BC gov VON (https://vonx.io) team, would like to share an opportunity with members of open communities to work with us to do some focused collaboration around mobile agent, Aries agents and aries-cloudagent-python interoperability. We have a pilot production service planned for a significant professional accreditation organization within our province coming up in the fall. This pilot has a requirement for mobile agents to be able to hold and offer verifiable credentials issued and verified by services operating instances of aries-cloudagent-python. Details are in this Open Collaboration MOU: https://drive.google.com/open?id=1GPZ4p4zM36AN3DcHJ13s_IisSrT4F3S0 More details around the use case and how we intend to work with the community are in the document. Our offer is to work closely, in the open, with groups that we believe are close to having a releasable mobile agent such that their agent could be one of the listed compatible agents when this new service goes live. Our expectation is for the production pilot to occur this fall. If you are interested and think you have a mobile agent that is close to release, please review this Open Collaboration MOU, and contact us via Hyperledger Rocket.Chat or via my email available in the MOU. We would expect to have all participates identified with signed MOUs by Aug 6th 2019.

simranvc (Wed, 24 Jul 2019 09:22:20 GMT):
Has joined the channel.

simranvc (Wed, 24 Jul 2019 09:31:24 GMT):

Screenshot from 2019-07-24 15-00-00.png

Shubham-koli (Wed, 24 Jul 2019 09:47:42 GMT):
Has joined the channel.

Shubham-koli (Wed, 24 Jul 2019 09:47:43 GMT):
help this poor soul

Rick (Wed, 24 Jul 2019 11:40:52 GMT):
can we configure for the agent to target a specific node in the indy pool to communicate to instead of the indy-node-pool?

rhall9090 (Wed, 24 Jul 2019 13:35:55 GMT):
Has joined the channel.

lorenz.loesch (Wed, 24 Jul 2019 14:25:01 GMT):
hi guys. I am trying to verify a proof on android. I am able to retrieve the schema and credDefs from the ledger and I am able to run `Anoncreds.verifierVerifyProof` but the result is always `false` - even though I am certain it should be `true`. can you give me any hints how i could debug this issue? is there any way I can find out *why* it failed? here are some logs: ``` I/commands: src/commands/mod.rs:124 | AnoncredsCommand command received I/*creds_command_executor: src/commands/anoncreds/mod.rs:58 | Verifier command received I/*ifier_command_executor: src/commands/anoncreds/verifier.rs:38 | VerifyProof command received I/anoncreds_service: /home/indy/.cargo/registry/src/github.com-1ecc6299db9ec823/ursa-0.1.1/src/cl/verifier.rs:246 | Verifier verify proof -> done I/HyperLedgerIndyDriver: Verified Proof: false ``` any help is appreciated

samurai (Wed, 24 Jul 2019 20:04:42 GMT):
Has joined the channel.

samurai (Wed, 24 Jul 2019 20:04:43 GMT):
Hi all,

samurai (Wed, 24 Jul 2019 20:04:47 GMT):
I would please want to know, how i could use Hyperledger Indy to build a blockchain, that would be able to host user data.

dbluhm (Wed, 24 Jul 2019 20:15:42 GMT):
Hyperledger Indy is not a blockchain for storing user data but rather a blockchain designed to allow users to hold their own data and selectively disclose as much or as little information as they desire with those that ask for it. This might be a good article to read as a primer on self-sovereign identity: https://medium.com/@AlexPreukschat/self-sovereign-identity-a-guide-to-privacy-for-your-digital-identity-5b9e95677778

samurai (Wed, 24 Jul 2019 20:18:52 GMT):
Thanks for the response

luckycharms810 (Thu, 25 Jul 2019 14:21:27 GMT):
Hello, I have a couple of questions about the revocation flow that I'd love if anyone could help me with. The process of issuing a credential creates 3 artifacts: - cred_json - This one is somewhat self explanatory and is securely sent to the credential holder. - cred_revoc_id - The docstring for this var says -> `local id for revocation info (Can be used for revocation of this cred)` - revoc_reg_delta_json - This seems to be the only artifact necessary to create the revocation registry entry request My understanding is that given the `cred_revoc_id`, I can run `anoncreds.issuer_revoke_credential` to re-create my `revoc_reg_delta_json`. * So out of curiosity, is there currently a standard way of storing the `cred_revoc_id`s in a manner that is correlated to the credential as well as the holder within the wallet ? * Is there a difference between the `revoc_reg_delta_json` that is created at the point of credential issuance, and the `revoc_reg_delta_json` created later ?

nage (Thu, 25 Jul 2019 14:36:11 GMT):
https://wiki.hyperledger.org/display/IWG/2019-07-25+Identity+WG+Implementers+Call

rchristman (Thu, 25 Jul 2019 16:05:23 GMT):
@nage Nathan, if no one else volunteers to host the Identity WG call, I would like to do it.

nage (Thu, 25 Jul 2019 16:59:56 GMT):
Thanks! Would you care to “audition” by hosting next weeks call? ( will get the agenda setup and have someone from the Sovrin Foundation manage the recording.)

mattatkiva (Thu, 25 Jul 2019 19:16:08 GMT):
any recommendations on diagnosising indysdk error 309 - LedgerNotFound

drummondreed (Thu, 25 Jul 2019 19:30:49 GMT):
That is excellent. It is a really good way to get to know a lot of community members.

simranvc (Fri, 26 Jul 2019 13:05:24 GMT):
how to add multiple users at run time and store in indy

lynn.bendixsen (Fri, 26 Jul 2019 16:09:32 GMT):
You can run indy-cli in batch mode to add multiple new users "all at once". For more details use ```indy-cli --help```

lynn.bendixsen (Fri, 26 Jul 2019 16:09:32 GMT):
You can run indy-cli in batch mode to add multiple new DIDs "all at once". For more details use ```indy-cli --help```

rchristman (Fri, 26 Jul 2019 16:52:14 GMT):
Sure, an audition on Aug 1 is a good idea. You guys do a great job leading the discussions! On Aug 15, I'll be on vacation and unable to lead the call.

nage (Fri, 26 Jul 2019 17:05:00 GMT):
we have @rchristman hosting next Thursday and @mccown two weeks after that. I need to get the calendar appointment updated to every other week and make sure a SF staff member can pass host over to the right person and that auto-record is setup properly.

nikos.triantafyllou (Mon, 29 Jul 2019 09:55:04 GMT):
Has joined the channel.

nikos.triantafyllou (Mon, 29 Jul 2019 09:57:25 GMT):
Hi all, I am trying to connect to the Indy STN but I am getting an internal server error. The url I tried is https://s3.us-east-2.amazonaws.com/evernym-cs/sovrin-STNnetwork/www/trust-anchor.html do you know if the registration url has changed?

ShrinivasDeshmukh (Mon, 29 Jul 2019 17:50:03 GMT):
Hi @SteveGoob am unable to pull libindy using the current keys. Please can you help?

SteveGoob (Mon, 29 Jul 2019 18:17:59 GMT):
@ShrinivasDeshmukh Could you clarify a bit for me? Are you trying to install libindy on your machine? Pull the repository? What OS/distro are you running?

ShrinivasDeshmukh (Mon, 29 Jul 2019 18:20:17 GMT):
I am installing in nodejs docker container, Ubuntu 18.04 docker image

Shubham-koli (Tue, 30 Jul 2019 13:25:07 GMT):
you should use ubuntu 16.04. I've faced lot of errors while trying to start on ubuntu 18.04

Shubham-koli (Tue, 30 Jul 2019 13:30:44 GMT):
I was trying to create multiple users in indy... but somehow all the users that I create with my code has same DID and Key. I'm not able to understand why is this happening.

Shubham-koli (Tue, 30 Jul 2019 13:31:03 GMT):
I'm using the code in `https://github.com/hyperledger/indy-sdk/blob/master/samples/nodejs/src/gettingStarted.js`

Shubham-koli (Tue, 30 Jul 2019 13:31:27 GMT):

imgpsh_mobile_save2.jpeg

Shubham-koli (Tue, 30 Jul 2019 13:31:52 GMT):

imgpsh_mobile_save.jpeg

mattatkiva (Tue, 30 Jul 2019 15:56:21 GMT):
@Shubham-koli I dont have enough information to see what you see. Dids are created with this command: `createAndStoreMyDid` so make sure the inputs are unique.

mattatkiva (Tue, 30 Jul 2019 15:56:21 GMT):
@Shubham-koli I dont have enough information to see what you see. Dids are created with this command: `createAndStoreMyDid` so make sure the inputs are unique (and that command also saves them in the wallet so make sure you are using the correct wallet).

cam-parra (Tue, 30 Jul 2019 23:06:12 GMT):
is the TAA part of indy-sdk or is it a plugin?

nage (Wed, 31 Jul 2019 03:11:37 GMT):
The agreement itself is a configurable option, the implementation is part of the sdk (and node configuration)

simranvc (Wed, 31 Jul 2019 05:34:49 GMT):
@mattatkiva i am getting an error when i pass the random deed value of other user

simranvc (Wed, 31 Jul 2019 05:35:06 GMT):

Screenshot from 2019-07-31 11-03-49.png

simranvc (Wed, 31 Jul 2019 05:45:26 GMT):
@cam-parra yes its part of indy-sdk plugin

ShrinivasDeshmukh (Wed, 31 Jul 2019 06:02:04 GMT):
I cloned the repository and built libindy in the container. I was able to run the code!

ShrinivasDeshmukh (Wed, 31 Jul 2019 06:02:04 GMT):
Hi @SteveGoob, I cloned the repository and built libindy in the container. I was able to run the code!

ShrinivasDeshmukh (Wed, 31 Jul 2019 06:09:43 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=uJWxh4ZArWrxGGmJ5) @simranvc The seed should be 32 characters. I think you are passing 33 characters in the seed. Can you please check?

sauveergoel (Wed, 31 Jul 2019 06:10:12 GMT):
Has joined the channel.

sauveergoel (Wed, 31 Jul 2019 06:10:14 GMT):
Hi, i am facing the following issue: indy.error.IndyError: (, {'backtrace': '', 'message': 'Error: Invalid library state\n Caused by: Ledger merkle tree is not acceptable for current tree.\n'})

ShrinivasDeshmukh (Wed, 31 Jul 2019 06:18:01 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=PBoYS96YLnHR33FvT) @sauveergoel I had this error when my pool config was inconsistent with the Genesis transaction file. Can you check your pool config and Genesis txn file?

sauveergoel (Wed, 31 Jul 2019 06:27:00 GMT):
I am trying to connect my aries agent to indy pool where i am supplying the pool_genesis_transaction file from the container

simranvc (Wed, 31 Jul 2019 06:52:56 GMT):
@ShrinivasDeshmukh yes its solved thanks

gcsfred (Wed, 31 Jul 2019 14:34:19 GMT):
Has joined the channel.

gcsfred (Wed, 31 Jul 2019 14:34:21 GMT):
I ran and followed the Indy Agent Demonstration ( https://github.com/hyperledger/education/tree/master/LFS171x/indy-material/nodejs , https://github.com/hyperledger/education/blob/master/LFS171x/indy-material/nodejs/AgentDemoScript.md ). Does anyone have pointers to the code that show some key parts of the demo? (Say, where the proof requests and send credential offer are in the code ) Links to a repo will help, or even function fragments.

swcurran (Wed, 31 Jul 2019 14:48:34 GMT):
That was a plan for that code base, but it is now rather dated. As your next steps in exploring, you might want to look at aries-cloudagent-python, which has similar demos, and some additional resources to learn the code. The best place to get started is here - https://github.com/hyperledger/aries-cloudagent-python/tree/master/docs/GettingStartedAriesDev It's first step takes you to what you probably have looked at the - Indy chapter from the edX.

swcurran (Wed, 31 Jul 2019 14:50:19 GMT):
We recently did an Architectural Deep Dive into that code and the associated recording and slide presentation is available. I'll add that to the aries-cloudagent-python in the next couple of days.

swcurran (Wed, 31 Jul 2019 14:50:19 GMT):
We recently did an Architectural Deep Dive into that code and the associated recording and slide presentation is available. I'll add that to the aries-cloudagent-python documentation in the next couple of days.

AlexITC (Wed, 31 Jul 2019 15:03:58 GMT):
Has joined the channel.

AlexITC (Wed, 31 Jul 2019 15:05:10 GMT):
Hello, I'm following the walkthrough, specifically, this https://github.com/hyperledger/indy-sdk/blob/master/docs/getting-started/indy-walkthrough.md#step-2-connecting-to-the-indy-nodes-pool

AlexITC (Wed, 31 Jul 2019 15:05:37 GMT):
it gives you some code samples but doesn't tell you how to actually run, any guidance is appreciated

dbluhm (Wed, 31 Jul 2019 15:10:50 GMT):
I think the code in the walkthrough is taken from the `getting-started.py` in this directory: https://github.com/hyperledger/indy-sdk/tree/master/samples/python Some very basic instructions for running those components are in the README in that directory but if that's not working for you, please reach out. I also highly recommend looking into the Aries Cloud Agent python getting started guide linked above by @swcurran

dbluhm (Wed, 31 Jul 2019 15:11:10 GMT):
https://github.com/hyperledger/aries-cloudagent-python/tree/master/docs/GettingStartedAriesDev

AlexITC (Wed, 31 Jul 2019 15:12:30 GMT):
I will try it soon, thanks

gcsfred (Wed, 31 Jul 2019 15:16:14 GMT):
Thank you. I actually tried to follow https://github.com/hyperledger/aries-cloudagent-python/tree/master/demo for Aries yesterday, but there was an error, so I switched to the Indy demo. ( I'm doing a Lunch & Learn session at my company today. Too bad the Aries demo didn't work. )

gcsfred (Wed, 31 Jul 2019 15:16:14 GMT):
@swcurran Thank you. I actually tried to follow https://github.com/hyperledger/aries-cloudagent-python/tree/master/demo for Aries yesterday, but there was an error, so I switched to the Indy demo. ( I'm doing a Lunch & Learn session at my company today. Too bad the Aries demo didn't work. )

swcurran (Wed, 31 Jul 2019 15:17:26 GMT):
Arrggh...sorry about that. The Play-with-VON site had some issues, I hear. I'm on vacation this week, so just checking in periodically.

gcsfred (Wed, 31 Jul 2019 15:18:51 GMT):
@swcurran In fact, Play-with-VON is down now, and my demo would start in 15 minutes :(

WadeBarnes (Wed, 31 Jul 2019 15:19:17 GMT):
on it

arsulegai (Wed, 31 Jul 2019 15:35:07 GMT):
Has joined the channel.

WadeBarnes (Wed, 31 Jul 2019 15:54:20 GMT):
Back up as of 8:37AM Pacific.

ianco (Wed, 31 Jul 2019 15:58:35 GMT):
you can also use play with docker https://labs.play-with-docker.com/

arsulegai (Wed, 31 Jul 2019 16:05:08 GMT):
Hello all, I can't seem to find one. Is there a PPT on Indy in the Hyperledger website or elsewhere?

ravip (Wed, 31 Jul 2019 16:47:00 GMT):
Problem: PoolLedgerTimeout error when calling createPoolLedgerConfig Initial diagnosis: I read somewhere that the cause might be that the IP addresses in the pool configuration and genesis file might not match. Environment: I have an agent (a nodejs web app) running in localhost and I have a pool of four indy validator nodes running in a docker container. I am using Mac OS. The docker containers have IP address ranging from 173.17.0.101 to 173.17.0.104 and the ports ranging from 9701 to 9708 and there is port forwarding to host enabled. The ledger browser is at 173.17.0.101:9000 and I can access the same using localhost:9000. In the genesis file that I supply to agent, I change the 173.17 ip addresses to 127.0.0.1. What I did to connect: I made the genesis file (that was created during the start of the pool) available to the agent by passing the path of the file to createPoolLedgerConfig method as an argument. Can anyone please point out if I am missing something?

sukalpomitra (Wed, 31 Jul 2019 16:49:52 GMT):
does the genesis file have the correct client and node ips?

AlexITC (Wed, 31 Jul 2019 16:50:21 GMT):
while trying to run the python samples, I get `ModuleNotFoundError: No module named 'indy'`, I believe it needs to be installed manually

sukalpomitra (Wed, 31 Jul 2019 16:50:56 GMT):
why dont u try the network at http://138.197.138.255/

sukalpomitra (Wed, 31 Jul 2019 16:50:56 GMT):
why dont u try the network at http://138.197.138.255/ and then if it works, u know its something wrong with the ip addresses

dbluhm (Wed, 31 Jul 2019 16:54:14 GMT):
Try creating a virtual environment (`python3 -m venv env`), activating it (`source env/bin/activate`), and installing indy with `pip install python3-indy` then give running the sample another shot

AlexITC (Wed, 31 Jul 2019 16:55:13 GMT):
I'll try that, thanks

AlexITC (Wed, 31 Jul 2019 16:55:48 GMT):
then, I get `ImportError: cannot import name 'anoncreds'`

AlexITC (Wed, 31 Jul 2019 16:56:01 GMT):
isn't there a `requirements` file to install everything at once?

AlexITC (Wed, 31 Jul 2019 16:56:09 GMT):
or are these dependencies listed anywhere?

dbluhm (Wed, 31 Jul 2019 16:57:28 GMT):
Typically there is. Off the top of my head, I'm not sure the samples directory does this. Try `pip install -e .`

AlexITC (Wed, 31 Jul 2019 16:58:06 GMT):
I guess that's likely on the actual python wrapper

dbluhm (Wed, 31 Jul 2019 17:00:36 GMT):
I just did a quick sanity check on my machine and was able to run the getting started guide with these steps: ``` python3 -m venv env source env/bin/activate pip install -e . python src/getting-started.py ``` So installing python3-indy may not be necessary

AlexITC (Wed, 31 Jul 2019 17:01:19 GMT):
hmm, I was trying to run just python by using pyenv

AlexITC (Wed, 31 Jul 2019 17:01:33 GMT):
let me try that one

AlexITC (Wed, 31 Jul 2019 17:02:30 GMT):
that does the trick

AlexITC (Wed, 31 Jul 2019 17:03:42 GMT):
thanks!

dbluhm (Wed, 31 Jul 2019 17:03:51 GMT):
No problem!

AlexITC (Wed, 31 Jul 2019 17:04:22 GMT):
last time I tried, the libindy rust tests failed, do I need to launch a node to actually run them?

AlexITC (Wed, 31 Jul 2019 17:04:45 GMT):
I'm running them again to gather the errors

dbluhm (Wed, 31 Jul 2019 17:05:19 GMT):
I do believe this is the case. You should be able to follow these instructions to get a local test pool: https://github.com/hyperledger/indy-sdk#how-to-start-local-nodes-pool-with-docker

AlexITC (Wed, 31 Jul 2019 17:05:49 GMT):
yes, I'm able to get that, but first I compiled libindy and ran the tests

mattatkiva (Wed, 31 Jul 2019 17:05:55 GMT):
agreed. the pools are required

AlexITC (Wed, 31 Jul 2019 17:05:59 GMT):
I'll try to run them having the docker images running

AlexITC (Wed, 31 Jul 2019 17:06:19 GMT):
these are the failures ``` failures: high_cases::cred_def_cache::indy_get_cred_def_cache_works high_cases::cred_def_cache::indy_get_cred_def_empty_options high_cases::cred_def_cache::indy_get_cred_def_min_fresh_works high_cases::cred_def_cache::indy_get_cred_def_no_cache_works high_cases::cred_def_cache::indy_get_cred_def_no_store_works high_cases::cred_def_cache::indy_get_cred_def_only_cache_no_cached_data high_cases::schema_cache::indy_get_schema_cache_works high_cases::schema_cache::indy_get_schema_empty_options high_cases::schema_cache::indy_get_schema_empty_options_for_unknown_id high_cases::schema_cache::indy_get_schema_min_fresh_works high_cases::schema_cache::indy_get_schema_no_cache_works high_cases::schema_cache::indy_get_schema_no_store_works high_cases::schema_cache::indy_get_schema_only_cache_no_cached_data ```

AlexITC (Wed, 31 Jul 2019 17:07:41 GMT):
I don't think this is related to the pools, `thread 'main' panicked at 'called `Result::unwrap()` on an `Err` value: IndyError { error_code: WalletAlreadyExistsError, message: "Error: Wallet with this name already exists\n Caused by: Wallet database file already exists: \"/home/dell/.indy_client/wallet/default-wallet_id-indy_get_cred_def_cache_works-2/sqlite.db\"\n", indy_backtrace: Some("") }', src/libcore/result.rs:999:5`

dbluhm (Wed, 31 Jul 2019 17:10:20 GMT):
Looks like cleanup failed from the last test. This could use some work from the SDK team... A quick way to remove the wallet would be to just `rm -r ~/.indy_client/wallet/default-wallet-id-indy_get_cred_def_Cahce_works-2`

dbluhm (Wed, 31 Jul 2019 17:10:44 GMT):
Some typos in that path but you get the point lol...

AlexITC (Wed, 31 Jul 2019 17:12:48 GMT):
they succeed now, thanks!

ravip (Wed, 31 Jul 2019 17:18:55 GMT):
@sukalpomitra So I tried the genesis file of the network at http://138.197.138.255/

ravip (Wed, 31 Jul 2019 17:19:14 GMT):
I am still getting the same error 'Pool Ledger Timeout'

sukalpomitra (Wed, 31 Jul 2019 17:19:48 GMT):
which agent are you using?

ravip (Wed, 31 Jul 2019 17:21:20 GMT):
What I did: I downloaded the genesis file from http://138.197.138.255/ and pasted it in a local directory in my nodejs web app agent and passed the path of that file to

ravip (Wed, 31 Jul 2019 17:21:22 GMT):
let poolGenesisTxnPath = '/Users/newschool/blockchain_project/stand/test/gtf' let poolConfig = { "genesis_txn": poolGenesisTxnPath }; await sdk.createPoolLedgerConfig(config.poolName, poolConfig);

ravip (Wed, 31 Jul 2019 17:21:22 GMT):
let poolGenesisTxnPath = '/Users/me/blockchain_project/dir/test/gtf' let poolConfig = { "genesis_txn": poolGenesisTxnPath }; await sdk.createPoolLedgerConfig(config.poolName, poolConfig);

ravip (Wed, 31 Jul 2019 17:22:35 GMT):
I am using nodejs agent

dbluhm (Wed, 31 Jul 2019 17:24:35 GMT):
From https://github.com/hyperledger/indy-agent ?

sukalpomitra (Wed, 31 Jul 2019 17:25:49 GMT):
@dbluhm is the right person to answer you. I have not used that code. But I have used through aries cloud agent and agent-framework and I did not have issues

sukalpomitra (Wed, 31 Jul 2019 17:27:41 GMT):
btw

ravip (Wed, 31 Jul 2019 17:28:00 GMT):
@dbluhm : From this link : https://github.com/hyperledger/education.git

sukalpomitra (Wed, 31 Jul 2019 17:28:05 GMT):
I think you need to pass the file name with the path not only the [ath "/Users/me/blockchain_project/dir/test/gtf"

ravip (Wed, 31 Jul 2019 17:28:32 GMT):
I guess both the links have the same nodejs agent?

sukalpomitra (Wed, 31 Jul 2019 17:28:51 GMT):
try passing the file name too

dbluhm (Wed, 31 Jul 2019 17:31:39 GMT):
@sukalpomitra 's suggestion may work but I very highly recommend you instead look into the Aries Cloud Agent - Python project for an up to date introduction into the Indy/Aries communities. The Node.js agent contains a lot of early thinking that has been greatly improved on since. A good getting started guide for the Aries Cloud Agent (ACA) can be found here: https://github.com/hyperledger/aries-cloudagent-python/tree/master/docs/GettingStartedAriesDev

dbluhm (Wed, 31 Jul 2019 17:33:01 GMT):
The node.js agent is essentially deprecated and unmaintained

ravip (Wed, 31 Jul 2019 17:43:04 GMT):
So I tried @sukalpomitra 's suggestion. but no luck trying to connect.

dbluhm (Wed, 31 Jul 2019 17:50:44 GMT):
Again, I encourage you to look into ACA-Py. I believe your time will be far better spent there. As for the issues you're experiencing, @swcurran may be the best person to answer your questions as he's the original author of that particular demo, borrowing code from the Node.js agent.

ravip (Wed, 31 Jul 2019 18:42:21 GMT):
Thanks for pointing that out @dbluhm

ravip (Wed, 31 Jul 2019 18:59:50 GMT):
@swcurran Whenever you are free, can you share your insight on what might be the cause of this.

ianco (Wed, 31 Jul 2019 19:47:08 GMT):
"In the genesis file that I supply to agent, I change the 173.17 ip addresses to 127.0.0.1." - this will not work. You need to have matching IP's in the node and client. You need to either run your app in a docker container (that recognizes the nodes' IP's) or else run a set of nodes that are listening on 127.0.0.1

ravip (Wed, 31 Jul 2019 20:13:13 GMT):
@ianco : thanks for the response, so you are saying the genesis file at both locations (pool and agent) should have the same IPs? What if the indy nodes are run in a docker container? Isnt is possible that they have IPs which are not recognizable by an agent running elsewhere? Unless yes they are resolved to same host address. But again in this case genesis file at both the locations could have different IPs?

ianco (Wed, 31 Jul 2019 20:17:16 GMT):
As far as I know the IP's have to match, so the nodes need to be running on publicly accessible IP's, or else the agent (or other client app) needs to be running on the same docker network.

ianco (Wed, 31 Jul 2019 20:18:13 GMT):
There is a script in the indy-sdk that will start up a set of nodes on 127.0.0.1 https://github.com/hyperledger/indy-sdk#1-starting-the-test-pool-on-localhost

ianco (Wed, 31 Jul 2019 20:18:42 GMT):
Or else von-network will run a set of nodes that are accessible inside a docker network https://github.com/bcgov/von-network

dbluhm (Wed, 31 Jul 2019 20:19:47 GMT):
Identity WG Implementers call is happening tomorrow! @SteveGoob will be demoing the GitLab CI/CD Migration and we'll have time for open discussion. If anyone has other topics they'd like to bring up on the call tomorrow, let me know! Agenda (Zoom link, call time, and meeting notes included): https://wiki.hyperledger.org/x/ehP5

dbluhm (Wed, 31 Jul 2019 20:19:47 GMT):
Identity WG Implementers call is happening tomorrow! @SteveGoob will be demoing the Indy GitLab CI/CD Migration and we'll have time for open discussion. If anyone has other topics they'd like to bring up on the call tomorrow, let me know! @rchristman has graciously volunteered to host the call for us. Agenda (Zoom link, call time, and meeting notes included): https://wiki.hyperledger.org/x/ehP5

ianco (Wed, 31 Jul 2019 20:20:33 GMT):
... or else you need publicly accessible IP's. The BC Gov runs a test network (BCovrin) - you can see the ledger browser here: http://test.bcovrin.vonx.io/ (I think this is the one referenced in some of our tutorials)

ianco (Wed, 31 Jul 2019 20:21:04 GMT):
You can connect to the Sovrin Staging Network, but you need to get someone at Sovrin to create your DID's for you

ravip (Wed, 31 Jul 2019 20:27:44 GMT):
Thanks @ianco for your valuable suggestions. Can I send you a pm? I am having some issues locally connecting agent and nodes

ianco (Wed, 31 Jul 2019 20:28:21 GMT):
Sure

AlexITC (Thu, 01 Aug 2019 03:37:32 GMT):
so, I ran the python samples, all at once with `python -m src.main`, and it ends with an error: ``` INFO:indy.libindy.native.indy.commands.pool: src/commands/pool.rs:74 | OpenAck handle 200, pool_id 200, result Ok(()) INFO:indy.libindy.native.indy.commands: src/commands/mod.rs:148 | WalletCommand command received INFO:indy.libindy.native.indy.commands: src/commands/mod.rs:148 | WalletCommand command received INFO:indy.libindy.native.indy.commands: src/commands/mod.rs:148 | WalletCommand command received INFO:indy.libindy.native.indy.commands: src/commands/mod.rs:148 | WalletCommand command received INFO:indy.libindy.native.indy.commands: src/commands/mod.rs:148 | WalletCommand command received INFO:indy.libindy.native.indy.commands: src/commands/mod.rs:148 | WalletCommand command received INFO:indy.libindy.native.indy.commands: src/commands/mod.rs:144 | DidCommand command received INFO:indy.libindy.native.indy.commands.did: src/commands/did.rs:125 | CreateAndStoreMyDid command received Traceback (most recent call last): File "/home/dell/.pyenv/versions/3.6.8/lib/python3.6/runpy.py", line 193, in _run_module_as_main "__main__", mod_spec) File "/home/dell/.pyenv/versions/3.6.8/lib/python3.6/runpy.py", line 85, in _run_code exec(code, run_globals) File "/home/dell/projects/indy/indy-sdk/samples/python/src/main.py", line 17, in run_coroutine(main) File "/home/dell/projects/indy/indy-sdk/samples/python/src/utils.py", line 47, in run_coroutine loop.run_until_complete(coroutine()) File "/home/dell/.pyenv/versions/3.6.8/lib/python3.6/asyncio/base_events.py", line 484, in run_until_complete return future.result() File "/home/dell/projects/indy/indy-sdk/samples/python/src/main.py", line 14, in main await txn_author_agreement.demo() File "/home/dell/projects/indy/indy-sdk/samples/python/src/txn_author_agreement.py", line 56, in demo aml_req = await ledger.build_acceptance_mechanisms_request(trustee['did'], AttributeError: module 'indy.ledger' has no attribute 'build_acceptance_mechanisms_request' ```

CarstenStahlschmidt (Thu, 01 Aug 2019 06:23:06 GMT):
Has joined the channel.

simranvc (Thu, 01 Aug 2019 12:20:33 GMT):

Screenshot from 2019-08-01 17-49-24.png

simranvc (Thu, 01 Aug 2019 12:21:05 GMT):
i am getting an error in pool handle is expecting number me also pass the function`openpoolledger`

simranvc (Thu, 01 Aug 2019 12:21:23 GMT):
but not getting the key of pool handle in number

dbluhm (Thu, 01 Aug 2019 14:38:15 GMT):
Apologies; last minute Zoom room changes for the Implementers call (happening in about 20 minutes) to make sure we can record the call. The updated link can be found on the Agenda: https://wiki.hyperledger.org/x/ehP5

ShrinivasDeshmukh (Thu, 01 Aug 2019 15:09:38 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=ENujCko93uvng5hAu) @simranvc Line 27 of your code, you should probably use await indy.openPoolLedger

AlexITC (Thu, 01 Aug 2019 15:36:43 GMT):
I found this while following the walkthrough, does anyone knows the algorithm name? ``` Faber authenticates and encrypts the message by calling crypto.auth_crypt function, which is an implementation of the authenticated-encryption schema. Authenticated encryption is designed for sending of a confidential message specifically for the Recipient. The Sender can compute a shared secret key using the Recipient's public key (verkey) and his secret (signing) key. The Recipient can compute exactly the same shared secret key using the Sender's public key (verkey) and his secret (signing) key. That shared secret key can be used to verify that the encrypted message was not tampered with, before eventually decrypting it. ```

mattatkiva (Thu, 01 Aug 2019 17:12:05 GMT):
is there a way to recover a credential definition if the wallet is deleted (accidentally)? its easy to create a new wallet, recreate the DID using the same seed.....I am not able to create the same credential definition as the public part is on the ledger.

andrew.whitehead (Thu, 01 Aug 2019 17:17:58 GMT):
It’s not, there’s private keys associated with the cred

andrew.whitehead (Thu, 01 Aug 2019 17:17:58 GMT):
It’s not, there’s private keys associated with the cred def

sukalpomitra (Thu, 01 Aug 2019 17:29:36 GMT):
but if I am not wrong you can again create the cred def schema using the schema id, a ltransaction is written in the ledger, but the cred def id remains the same

sukalpomitra (Thu, 01 Aug 2019 17:29:39 GMT):
http://138.197.138.255/browse/domain?page=1&query=lrole&txn_type=102

swcurran (Thu, 01 Aug 2019 18:03:16 GMT):
Yes, you can create it again. But you can't reuse the old one - I believe you have to bump the verison. Unlike with DIDs, you can't pass in a seed for a Cred Def, so you can't recreate an identical one. It would make development a lot easier if that was possible, but alas, no. :-(

mattatkiva (Thu, 01 Aug 2019 18:09:58 GMT):
what happens if a cred def is created, credentials are made against the def, the wallet creating the cred def is deleted..... are the existing credentials now invalid?

swcurran (Thu, 01 Aug 2019 18:16:10 GMT):
No - the existing credentials will still work. No new credentials can be issued, but existing ones will continue to be provable.

mattatkiva (Thu, 01 Aug 2019 18:16:38 GMT):
can they be revoked (assuming the revocation entries existed)?

swcurran (Thu, 01 Aug 2019 18:16:58 GMT):
But, like with losing your wallet containing credentials and keys - don't do it. Backups, restoration are crucial in any key-based systems.

swcurran (Thu, 01 Aug 2019 18:18:25 GMT):
It depends exactly what you lost. I believe if you ONLY lost the cred def, you could revoke them (it makes sense...), but if you also lost your private key for writing to the ledger, you would be stuck. If you lost all the information about the credentials you have issued, you would be lost.

swcurran (Thu, 01 Aug 2019 18:19:06 GMT):
Not 100% sure on the revocation, but my guess is because of the rest of that, it's likely moot.

mattatkiva (Thu, 01 Aug 2019 18:21:45 GMT):
yes, that makes sense. I wanted to confirm. we will need to use the wallet backup functionality. We discovered this problem while building out an environment. So I thought I would ask, just in case we were missing something in our thinking

Yano (Thu, 01 Aug 2019 18:23:52 GMT):
Has joined the channel.

swcurran (Thu, 01 Aug 2019 18:24:11 GMT):
We also found it early in our development and led to us largely using per developer ledger instances for testing. Shared testing ledgers are harder. We're now shifting to random DIDs vs. set DID seeds, which actually makes it easier - share a schema, but make new DIDs and Cred Defs on every run. Our devs at BC Gov might have a better view.

mattatkiva (Thu, 01 Aug 2019 18:57:25 GMT):
related to that: if we rotate a key used to create a cred def, will the previously issued credentials still work?

swcurran (Thu, 01 Aug 2019 18:59:41 GMT):
There is no way to rotate the cred def keys (there is one key per claim). To "rotate" the keys, you would version the cred def - create a new one. The existing credentials would still work and use the old version, new credentials would use the new version. The issuer can rotate the key of their DID, but not of a cred def.

mattatkiva (Thu, 01 Aug 2019 19:01:59 GMT):
great. That makes sense. thank you

esplinr (Thu, 01 Aug 2019 23:42:05 GMT):
https://speakerdeck.com/resplin/distributed-ledgers-finally-brought-me-a-usable-digital-identity-9a60c10c-8f31-45a1-b892-3347c86d4922

esplinr (Thu, 01 Aug 2019 23:44:44 GMT):
And a recording of the presentation to go along with the deck: https://fosdem.org/2019/schedule/event/ledger_identity/

esplinr (Thu, 01 Aug 2019 23:45:05 GMT):
Of course there are lots of other presentations about Indy, depending on your audience. This is just one that I recently gve.

esplinr (Thu, 01 Aug 2019 23:45:05 GMT):
Of course there are lots of other presentations about Indy, depending on your audience. This is just one that I recently gave.

esplinr (Thu, 01 Aug 2019 23:47:09 GMT):
This afternoon, Evernym brings you a double feature: two sprint demos that cover most of the features in this week's release of Indy Node and Indy SDK: Sprint EV 19.14 https://www.youtube.com/watch?v=XvEgMm1F02c&list=PLRp0viTDxBWGLdZk0aamtahB9cpJGV7ZF&index=2 Sprint EV 19.15 https://www.youtube.com/watch?v=DPlKwtNwo2Q&list=PLRp0viTDxBWGLdZk0aamtahB9cpJGV7ZF&index=1

esplinr (Thu, 01 Aug 2019 23:47:31 GMT):
cc @lynn.bendixsen @burdettadam

simranvc (Fri, 02 Aug 2019 04:41:20 GMT):
(node:10178) UnhandledPromiseRejectionWarning: TypeError: signAndSubmitRequest expects Number for wh

simranvc (Fri, 02 Aug 2019 04:42:00 GMT):
hello guys i am gettiing this error in index.js file when i am run the` signandsubmitrequest` function

simranvc (Fri, 02 Aug 2019 04:42:55 GMT):
anyone has any idea???

shantanu_kadu (Fri, 02 Aug 2019 09:36:30 GMT):
Has joined the channel.

simranvc (Fri, 02 Aug 2019 11:36:29 GMT):
on single pool we cant create issuer,proover and verifier

simranvc (Fri, 02 Aug 2019 11:36:44 GMT):
if we dont want multiple ledger

simranvc (Fri, 02 Aug 2019 11:36:47 GMT):
???

AlexITC (Sat, 03 Aug 2019 02:50:54 GMT):
I'm trying to follow the [write-did-and-query-verkey](https://github.com/hyperledger/indy-sdk/tree/master/docs/how-tos/write-did-and-query-verkey) guide in Java but I'm getting this exception: ``` Exception in thread "main" java.util.concurrent.ExecutionException: org.hyperledger.indy.sdk.IOException: Error: IO error Caused by: Can't open genesis txn file Caused by: No such file or directory (os error 2) at java.util.concurrent.CompletableFuture.reportGet(CompletableFuture.java:357) at java.util.concurrent.CompletableFuture.get(CompletableFuture.java:1895) at io.iohk.indy.PoC$.main(PoC.scala:38) at io.iohk.indy.PoC.main(PoC.scala) Caused by: org.hyperledger.indy.sdk.IOException: Error: IO error Caused by: Can't open genesis txn file Caused by: No such file or directory (os error 2) at org.hyperledger.indy.sdk.IndyException.fromSdkError(IndyException.java:124) at org.hyperledger.indy.sdk.IndyJava$API.checkResult(IndyJava.java:92) at org.hyperledger.indy.sdk.pool.Pool.access$400(Pool.java:20) at org.hyperledger.indy.sdk.pool.Pool$2.callback(Pool.java:70) at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method) at sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:62) at sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43) at java.lang.reflect.Method.invoke(Method.java:498) at com.sun.jna.CallbackReference$DefaultCallbackProxy.invokeCallback(CallbackReference.java:520) at com.sun.jna.CallbackReference$DefaultCallbackProxy.callback(CallbackReference.java:551) ```

AlexITC (Sat, 03 Aug 2019 02:51:55 GMT):
I wonder about this config entry for the pool, which is a hardcoded path to something that doesn't exists on my system, I was thinking that it is possibly related to the docker pool but I haven't been able to find if that's true, `"genesis_txn": "/home/vagrant/code/evernym/indy-sdk/cli/docker_pool_transactions_genesis"`

AlexITC (Sun, 04 Aug 2019 05:19:35 GMT):
I was able to update the path to the correct one but I get a timeout after the next step, running the same how-to with python leads to a timeout too, `Error occurred: (, {'backtrace': '', 'message': 'Error: Pool timeout\n Caused by: Pool timeout\n'})`, the pool is supposed to be running properly because the libindy rust tests are passing and they use the pool

sukalpomitra (Sun, 04 Aug 2019 11:31:34 GMT):
generally in my experience I have seen the pool timeout comes when the clientip/nodeip are not resolvable/reachable

AlexITC (Sun, 04 Aug 2019 21:07:56 GMT):
do you have any advice to fix it? I can't even run the how-tos

AlexITC (Sun, 04 Aug 2019 21:08:23 GMT):
I use this command to run the pool locally `docker run -it -p 9701-9708:9701-9708 indy_pool`

bootstrapsp (Sun, 04 Aug 2019 22:27:01 GMT):
Just open sourced Miu : https://github.com/bootstrapsp/miu , it's a gRPC based py server that implements Indy SDK py wrapper. So now with the help of gRPC protofile we can write clients in all the languages supported by gRPC, like Golang :D

bootstrapsp (Sun, 04 Aug 2019 22:27:01 GMT):
Just open sourced Miu : https://github.com/bootstrapsp/miu , it's a gRPC based py server that implements Indy SDK py wrapper. So now with the help of gRPC protofile we can write clients in all the languages supported by gRPC, like Golang :D . May be of some help to people looking to write clients in Golang

bootstrapsp (Sun, 04 Aug 2019 22:27:01 GMT):
Just open sourced Miu : https://github.com/bootstrapsp/miu , it's a gRPC based py server that implements Indy SDK py wrapper. So now with the help of gRPC protofile we can write clients in all the languages supported by gRPC, like Golang :D . May be of some help to people looking to write clients in Golang . Of course this is just a pre-release, any help is appreciated.

sukalpomitra (Mon, 05 Aug 2019 01:33:37 GMT):
maybe try using a public hosted ledger - http://138.197.138.255

AlexITC (Mon, 05 Aug 2019 02:31:17 GMT):
how do you tell the library to use that host? possibly an environment variable?

sukalpomitra (Mon, 05 Aug 2019 03:05:19 GMT):
sorry i dont know what you are using. For the aries cloud agent you just pass it through genesis-url param.

AlexITC (Mon, 05 Aug 2019 03:05:49 GMT):
I'm following https://github.com/hyperledger/indy-sdk/tree/master/docs/how-tos/write-did-and-query-verkey

sukalpomitra (Mon, 05 Aug 2019 03:10:27 GMT):
Sorry I have not gone through these how-to

AlexITC (Mon, 05 Aug 2019 03:11:32 GMT):
what would you recommend to get started then?

sukalpomitra (Mon, 05 Aug 2019 03:12:10 GMT):
I am more familiar with aries-python-cloud-agent , osma and agent-framework repos

AlexITC (Mon, 05 Aug 2019 03:12:41 GMT):
I'll look into those ones, thanks

ShrinivasDeshmukh (Mon, 05 Aug 2019 05:25:31 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=3zJzqkz6ALgTBWXJA) @AlexITC In the how tos file, there is a utils.py, wherein Genesis transaction is written hard coded. Here is the link: https://github.com/hyperledger/indy-sdk/blob/master/docs/how-tos/write-did-and-query-verkey/python/utils.py

ShrinivasDeshmukh (Mon, 05 Aug 2019 05:28:30 GMT):
In the file itself, there is an env variable TEST_POOL_IP, you can try setting the variable to your node's IP

simranvc (Mon, 05 Aug 2019 07:57:06 GMT):

Screenshot from 2019-08-05 12-51-43.png

AlexITC (Mon, 05 Aug 2019 13:31:42 GMT):
I have saw that, I was following the Java example, sadly, neither the python one works, I wonder how can I know which IP is the pool running in

ShrinivasDeshmukh (Mon, 05 Aug 2019 15:03:21 GMT):
Its your machine's IP (192.168 or 10.20.x.x). if your machine is behind firewall, your local IP might not be resolvable. You can try docker inspect and get ip of the containers and give them in the pool config.

ShrinivasDeshmukh (Tue, 06 Aug 2019 09:26:38 GMT):
Hi everyone, I am getting following error while submitting cred def to the ledger: `Transaction has been rejected: validation error [ClientClaimDefSubmitOperation]: cannot be smaller than 1 (ref=0)` the `buildCredDefRequest` method returns json, with ref=0. Any solutions?

ShrinivasDeshmukh (Tue, 06 Aug 2019 09:29:11 GMT):
Also, I am getting `commonInvalidStructure` error while using `issuerCreateAndStoreRevocReg`. I've passed parameters as per npm's indy-sdk document. I think the error might be with tailsWriterHandle, what is this param, and what should I pass into it?

sanchopansa (Tue, 06 Aug 2019 09:50:35 GMT):
Has joined the channel.

sauveergoel (Tue, 06 Aug 2019 16:00:41 GMT):
Hi can anyone help me with understanding credential schema and credential definition. the difference between them and the relationship between them

sukalpomitra (Tue, 06 Aug 2019 17:00:38 GMT):
As far I understand them, a schema can be issued by any issuer but makes more sense when a standardising body creates it. All issuers then follow this schema to create their credential def schema. A schema can have n attributes while a cred def schema will have the same n attributes and may have some more. A credential definition schema is tagged to an issuer and can only be used by the issuer who has created it. So to my understanding a schema is more for standardising attributes that should be used if an issuer wants to issue a credential. A credential definition schema on the other hand gives the credential to add more info on top of it and for a verifier to say that he only trusts Issuer A and will only verify credentials issued by Issuer A or credential definition schema A.

sauveergoel (Tue, 06 Aug 2019 17:16:06 GMT):
Thank you for your quick response as I understood your explaination, you are saying that goverment should create all schemas and organizations eg. Faber, Acme should create a mashup of theses schemas to generate a desired credential definition. if that is the case then can credential definition be created by a normal user(who is not a trust anchor)?

sukalpomitra (Tue, 06 Aug 2019 17:17:21 GMT):
no afaik to create a credential definition schema or a schema definition you need to have a role of TRUST ANCHOR.

sauveergoel (Tue, 06 Aug 2019 17:18:05 GMT):
ok, then why do we need credentail definition why cant we upgrade the existin schema or create a new schema?

sukalpomitra (Tue, 06 Aug 2019 17:18:55 GMT):
well as ai said schema is for all to use , while a credential definition schema is tagged to a definite issuer(TRUST ANCHOR)

sauveergoel (Tue, 06 Aug 2019 17:20:20 GMT):
if i understand correctly then you mean that if faber has created a credential def then acme cannot use the same def even if it is requred to g?enerate a def with the same attributes

sukalpomitra (Tue, 06 Aug 2019 17:20:39 GMT):
yes

sukalpomitra (Tue, 06 Aug 2019 17:21:24 GMT):
and then a verifier using that cred def can say a holder that i only accept Faber or Acme's credentials or both

sauveergoel (Tue, 06 Aug 2019 17:21:25 GMT):
but the cred def is written to the ledger that mean its public data

sauveergoel (Tue, 06 Aug 2019 17:22:05 GMT):
then how are we restricting the sharing usage of the cred def?

swcurran (Tue, 06 Aug 2019 17:23:20 GMT):
FYI - a Cred Def does not have other attributes. It references a schema previously written to the ledger. Those are the only attributes in credentials issued related to that Cred Def.

sauveergoel (Tue, 06 Aug 2019 17:24:39 GMT):
is it possible to refer to multiple schemas and restrict some attributes?

sukalpomitra (Tue, 06 Aug 2019 17:25:48 GMT):
oh I always thought a cred def should always have all the attributes in the schema plus it can add its own

sukalpomitra (Tue, 06 Aug 2019 17:27:28 GMT):
the cred def when you write to the ledger is alsio written to the wallet. You cannot issue a credential with a cred def that is not present in your wallet

sauveergoel (Tue, 06 Aug 2019 17:28:36 GMT):
then what can we do with the cred def if we are reading it from the ledger?

sukalpomitra (Tue, 06 Aug 2019 17:29:35 GMT):
I guess you can query the ledger and get the attributes it supports and write a proof request using them

sukalpomitra (Tue, 06 Aug 2019 17:29:35 GMT):
I guess as a verifier you can query the ledger and get the attributes it supports and write a proof request using them

sauveergoel (Tue, 06 Aug 2019 17:30:48 GMT):
but as far as I know the proof request has a whole new structure which comprises of self attested, verifyable and zkp values?

sukalpomitra (Tue, 06 Aug 2019 17:32:06 GMT):
A proof request consists of requested"_attributes, requested_predicates and self_attested_attributes. These are all objects that has a name property that requires the attribute name

sauveergoel (Tue, 06 Aug 2019 17:32:43 GMT):
didnt understand

sauveergoel (Tue, 06 Aug 2019 17:32:47 GMT):
:(

sukalpomitra (Tue, 06 Aug 2019 17:34:10 GMT):
A proof request json has property names such as requested_attributes , the value for which is a list of object. the object then has a property called name, whose value are the attribute names

sauveergoel (Tue, 06 Aug 2019 17:36:20 GMT):
umm ok but where is the cred def in this picture?

sukalpomitra (Tue, 06 Aug 2019 17:36:48 GMT):
that same object also has one more property called restrictions which ios a list of creed def id

sukalpomitra (Tue, 06 Aug 2019 17:36:48 GMT):
that same object also has one more property called restrictions which is a list of cred def id

sauveergoel (Tue, 06 Aug 2019 17:38:58 GMT):
but isnt it a use of cred def? my question was around what is cred def and how is it different from cred schema

sukalpomitra (Tue, 06 Aug 2019 17:41:52 GMT):
maybe go here and see how a cred def looks like in ledger. And then change the type to schema and query to see how it looks like. http://138.197.138.255/browse/domain?page=1&query=lrole&txn_type=102

sukalpomitra (Tue, 06 Aug 2019 17:43:49 GMT):
Also when the verifier sets restrictions with a list of cred def, it actually says i need proof only from issuers who have written these cred def. The cred def id also has the public DID for the issuer

swcurran (Tue, 06 Aug 2019 17:46:48 GMT):
Yes. And a public key per attribute used for signing the claim in the credential.

swcurran (Tue, 06 Aug 2019 17:47:04 GMT):
A schema is just a list of attributes.

swcurran (Tue, 06 Aug 2019 17:48:05 GMT):
Cred Def also links to the optional revocation registry for the credentials issued against the Cred Def.

sauveergoel (Tue, 06 Aug 2019 17:52:44 GMT):
thanks for this clarification. If i understand correctly then it means that a cred def is a usable instance of a schema that consists of the schema itself plus revocation registry details and a public key (of the issuer). if this is true then why do we need schemas and not just use cred def instead?

swcurran (Tue, 06 Aug 2019 17:54:42 GMT):
So that we get common use of schema. We don't want every issuer to make up there own schema. They can, but that does not lead to interoperability. For example, we'd like every university to use the same format for credentials, so verifiers would know how to handle the credential regardless of the source. Not likely - so we can't force it - but an ideal to aim for and make possible.

sauveergoel (Tue, 06 Aug 2019 17:56:28 GMT):
got it...that means the a credential def is the draft of a credential where as schemas are guidelines..correct?

swcurran (Tue, 06 Aug 2019 17:58:12 GMT):
Not really. The attributes listed in the schema are the attributes in the credentials that are issued. The cred def uses the list of attributes to know what key pairs to make - one for each attribute. So they are all tied together.

sauveergoel (Tue, 06 Aug 2019 18:02:08 GMT):
ok...so i can say that a cred def is the structural base of a cred that is refering to one or more schemas plus other usable information such as revocation and keys?

sauveergoel (Tue, 06 Aug 2019 18:02:51 GMT):
which will be useful to validate the proof attributes

sauveergoel (Tue, 06 Aug 2019 18:04:37 GMT):
just more one quick question, what is signifance of the class atrribute in the cred def? as I could see only "CL" as a value being supplied to that attribute

swcurran (Tue, 06 Aug 2019 18:05:30 GMT):
One schema right now. Next version of schema has another layer of indirection, but current has only one schema per Cred Def.

sauveergoel (Tue, 06 Aug 2019 18:06:09 GMT):
ok

simran (Wed, 07 Aug 2019 10:22:11 GMT):
Has joined the channel.

simran (Wed, 07 Aug 2019 10:22:11 GMT):
can anyone created multiple issuer and one prover on differnt ports no and they communicating each other

dgunseli (Wed, 07 Aug 2019 11:00:00 GMT):
Hi everyone

dgunseli (Wed, 07 Aug 2019 11:00:09 GMT):
I get "/usr/lib/libindy.so: undefined symbol: indy_set_logger" error

dgunseli (Wed, 07 Aug 2019 11:00:16 GMT):
when using Aries agent

simran (Wed, 07 Aug 2019 12:36:33 GMT):

Screenshot from 2019-08-07 18-05-13.png

AlexITC (Wed, 07 Aug 2019 14:12:49 GMT):
how do you enable debug/trace logs for libindy? and where do you see those logs?

andrew.whitehead (Wed, 07 Aug 2019 15:05:17 GMT):
I think you probably have an older library installed. Try installing 1.9.0

andrew.whitehead (Wed, 07 Aug 2019 15:05:17 GMT):
I think you probably have an older Indy SDK library installed. Try installing 1.9.0

dgunseli (Wed, 07 Aug 2019 16:20:49 GMT):
Yes, I’ve updated the libindy library and the problem has been solved

mattatkiva (Wed, 07 Aug 2019 18:09:12 GMT):
Are there any size limitation to data in a credential? EG: is it good practice to put images in credential?

mattatkiva (Wed, 07 Aug 2019 18:09:12 GMT):
Are there any size limitation to data in a credential? EG: is it good practice to put images in credential? I thought I heard recommendations in the past to avoid that practice. I wanted to confirm if that is the best practice.

swcurran (Wed, 07 Aug 2019 20:37:08 GMT):
Not certain, but I'm guessing it is the signature being used. "CL" is the cryptographic scheme - CL Signature - used by HL Indy.

swcurran (Wed, 07 Aug 2019 20:37:31 GMT):
I'm guessing they've added an attribute so that they can change in the scheme in the future.

AlexITC (Wed, 07 Aug 2019 23:04:10 GMT):
turns out that the logs are exposed to the wrappers, in which case, the java wrapper already logs them

esplinr (Wed, 07 Aug 2019 23:05:03 GMT):
There is a limit to how big a message can be before timeouts are exceeded. This prevents messages from containing large images.

esplinr (Wed, 07 Aug 2019 23:05:47 GMT):
We have put small images in credentials, but the correct solution is to follow the emerging W3C standard for Verifiable Credentials which Brent and Ken have been working to bring to Indy.

swcurran (Wed, 07 Aug 2019 23:40:35 GMT):
@sklump is working on that use case and has been successful at putting in an image into a credential. Definitely works!

sklump (Wed, 07 Aug 2019 23:40:35 GMT):
Has joined the channel.

AlexITC (Thu, 08 Aug 2019 04:06:42 GMT):
I finally solved my issue for connecting to the docker pool using the java wrapper, I was trying to using this config https://github.com/hyperledger/indy-sdk/blob/master/cli/docker_pool_transactions_genesis but it actually has ips like 10.* instead of localhosts ones, I followed one file generated by the rust tests and I was able to get it working, the only problem was that file

LiuyangRiver (Thu, 08 Aug 2019 06:40:52 GMT):
Has joined the channel.

sklump (Thu, 08 Aug 2019 10:56:28 GMT):
The other consideration is semantics: is the image a PNG, BMP, JPEG, etc? The issuer needs to know what he's been proposed to issue, and the holder agent has to know how to render it. So there is metadata per credential and attribute to store in the holder wallet.

sklump (Thu, 08 Aug 2019 10:56:28 GMT):
The other consideration is semantics: is the image a PNG, BMP, JPEG, etc? Is the value base64 (base64URL? base58? base85? base32? ASN.1-88? 94? BER? DER? etc?)-encoded? The issuer needs to know what the holder has proposed to issue, and the holder agent has to know how to render it. So there is metadata per credential and (non-'text/plain') attribute to store in the holder wallet.

sklump (Thu, 08 Aug 2019 12:12:33 GMT):
Also, a cleverer cryptographer than I may find a way to attack naïve image attribute propagation with steganography. What if the error-check bits all resolve to some obscene image? Does it matter? When I was at Entrust there was a hue and cry when our random one-time-registration-password generator threw up a code with very bad embedded word in an early version for a high-visibility customer. There was trouble, and the next version added some checks.

sklump (Thu, 08 Aug 2019 12:12:33 GMT):
Also, a cleverer cryptographer than I may find a way to attack naïve image attribute propagation with steganography. What if the error-check bits all resolve to some obscene image? Does it matter? When I was at Entrust there was a hue and cry when our random one-time-registration-password generator threw up a code with very bad embedded word in an early version for a high-visibility customer in a demo for some bigwigs. There was trouble, and the next version added some checks.

esplinr (Thu, 08 Aug 2019 14:22:05 GMT):
That's a great story.

mattatkiva (Thu, 08 Aug 2019 15:32:54 GMT):
definitely. worthy to note.

mattatkiva (Thu, 08 Aug 2019 15:59:32 GMT):
this is related to my question about having images in credentials. I think I know the answer and I want to confirm I understand correctly: once a credential is created, it is immutable. is that correct?

mattatkiva (Thu, 08 Aug 2019 15:59:32 GMT):
this is related to my question about having images in credentials. I think I know the answer and I want to confirm I understand correctly: once a credential is created, it is immutable. is that correct? to change the credential, I would have to create a new credential....?

swcurran (Thu, 08 Aug 2019 16:07:51 GMT):
Yes and Yes. The Credential is created and delivered from the Issuer to the Holder/Prover and the Credential and claims are signed - cannot be changed. If you want to change the data in the credential, you have to create and provide a new one - and you will probably revoke the old one.

akuklin (Thu, 08 Aug 2019 17:16:20 GMT):
Has joined the channel.

raj_shekhar (Fri, 09 Aug 2019 10:13:04 GMT):
Hi I have just started with Indy and getting below error while installing it. error: failed to run custom build command for `libsodium-sys v0.0.16`

raj_shekhar (Fri, 09 Aug 2019 10:13:04 GMT):
Hi I have just started with Indy and getting below error while installing it. error: failed to run custom build command for `libsodium-sys v0.0.16` I am following this document- https://hyperledger-indy.readthedocs.io/projects/sdk/en/latest/docs/build-guides/ubuntu-build.html

raj_shekhar (Fri, 09 Aug 2019 10:13:04 GMT):
Hi I have just started with Indy and getting below error while running this command - cargo build. error: failed to run custom build command for `libsodium-sys v0.0.16` I am following this document- https://hyperledger-indy.readthedocs.io/projects/sdk/en/latest/docs/build-guides/ubuntu-build.html

carl.diclementi (Fri, 09 Aug 2019 13:41:26 GMT):
Where is the best place for me to read up on the types of nodes in an Indy network and the roles that they play?

esplinr (Fri, 09 Aug 2019 16:21:21 GMT):
There is currently only one type of node: a validator node. In the future we would like to implement some read-only observer nodes, but those don't exist yet.

carl.diclementi (Fri, 09 Aug 2019 16:22:12 GMT):
Oh I thought they existed already, got it! Thanks.

carl.diclementi (Fri, 09 Aug 2019 16:22:12 GMT):
Oh I thought they existed already, got it! That makes sense why I was having a hard time finding documentation about it. Thanks.

TimBenke (Mon, 12 Aug 2019 08:04:48 GMT):
Has joined the channel.

TimBenke (Mon, 12 Aug 2019 08:13:42 GMT):
Hello, I'm trying to compare different SSI solutions. We've built something based on uport's libs, but we wanted to go the *source* with indy to check how it compares. I've already read a lot of documentation, but I was wondering how to obtain a DID document? I've only seen the examples to get basic DID infos or specific attributes, but I was wondering if there's an easy method to get a complete JSON-LD DID document, such as with uport's DID resolver: https://github.com/uport-project/did-resolver

sklump (Mon, 12 Aug 2019 11:06:02 GMT):
https://github.com/hyperledger/aries-cloudagent-python/pull/130 replaces and supersedes PR#60. It includes support for Aries#0037. I can demo some good bits any time.

dbluhm (Mon, 12 Aug 2019 13:12:25 GMT):
The Indy ledger actually predates DID Documents; as you've pointed it out, it uses "attributes" instead right now but I believe full DID Doc support is a high priority, especially with the migration of the Indy Agent work to Hyperledger Aries.

TimBenke (Mon, 12 Aug 2019 13:17:13 GMT):
yes, we're considering Indy because it seems very central in the SSI community. We've also discovered that e.g. uport seems to be missing important parts like revocation lists/registries that Indy already has.

sklump (Mon, 12 Aug 2019 13:42:04 GMT):
VON anchor includes DID doc support for parsing and serialization: https://github.com/PSPC-SPAC-buyandsell/von_anchor/blob/master/von_anchor/a2a/diddoc.py, https://github.com/PSPC-SPAC-buyandsell/von_anchor/blob/master/test/test_a2a.py aca-py includes the above.

TimBenke (Mon, 12 Aug 2019 14:27:34 GMT):
cool, thank you

AlexITC (Mon, 12 Aug 2019 15:53:32 GMT):
I wonder if any of you have used indy to work without ZK-proofs, the idea is that the complete credentials are shared and the issuers have revocation lists

swcurran (Mon, 12 Aug 2019 16:50:26 GMT):
I don't think anyone has done that. Pretty much everything that the community has done has used core indy-sdk capabilities of which both of those are a part.

dbluhm (Mon, 12 Aug 2019 16:55:11 GMT):
As a reminder, even when using Indy Credentials using ZKP, it is possible to simply reveal the attributes certified in a credential.

AlexITC (Mon, 12 Aug 2019 16:56:17 GMT):
yes, i know those details are on the core design but it seems that our client isn't ready for such a modern approach

DibbsZA (Mon, 12 Aug 2019 22:38:41 GMT):
Where would I find documentation regarding schema definitions? Specifically the message format of a schema definition

raj_shekhar (Tue, 13 Aug 2019 05:50:27 GMT):
Hi ,

raj_shekhar (Tue, 13 Aug 2019 05:52:11 GMT):
Hi, Anybody tried this tutorial, is it working?? https://github.com/hyperledger/education/tree/master/LFS171x/indy-material/nodejs

manavgupta (Tue, 13 Aug 2019 05:54:33 GMT):
Has joined the channel.

TimBenke (Tue, 13 Aug 2019 06:21:38 GMT):
Hi, are you looking for this? https://hyperledger-indy.readthedocs.io/projects/node/en/latest/requests.html#schema

doyajii1 (Tue, 13 Aug 2019 07:02:36 GMT):
Has joined the channel.

doyajii1 (Tue, 13 Aug 2019 07:03:46 GMT):
just wondering is there reason why indy java sdk starts with 1.8?

manavgupta (Tue, 13 Aug 2019 07:04:46 GMT):
Is there any sample mobile application which can connect to alice-Faber demo??

dbluhm (Tue, 13 Aug 2019 13:27:13 GMT):
https://github.com/hyperledger/education/commit/f4cdceea2483aaa784c983d835e72e2c3a0429c8 The indy materials in that repository should have links to other more up to date resources now

dbluhm (Tue, 13 Aug 2019 13:30:00 GMT):
Use of Alice and Faber in demos is so common now that you'll have to give a little more info about which demo you're referring to :slight_smile: There is a mobile app that could be a good starting point: https://github.com/mattrglobal/osma It is based on the dotnet Agent framework which should be compatible with the Aries Cloud Agent and other dotnet agent framework based apps

mattatkiva (Tue, 13 Aug 2019 14:31:56 GMT):
starts with? I'm assuming your asking about the jdk version?

Ladar (Tue, 13 Aug 2019 15:21:18 GMT):
Has joined the channel.

Ladar (Tue, 13 Aug 2019 15:21:18 GMT):
If I want to use the *indy_register_wallet_storage* to register a custom wallet storage implementation, is it possible to access the function through NodeJS. It doesn't appear accessible through the provided SDK.

mattatkiva (Tue, 13 Aug 2019 16:16:47 GMT):
the method you mentioned is designed to be called by a "plugin" (I use the word loosely) that handles custom wallet storage. Consumers of a plugin will call the plugin initialization method (and the plugin itself calls indy_register_wallet_storage). Check out the postgres plugin. If you're feeling thats backwards, I join you. :D That's how it works.

gcsfred (Tue, 13 Aug 2019 18:53:05 GMT):
I was running the Alice <-> ACME <-> Faber demo. ACME sends a proof of request for transcript data to Alice and she accepts it. Where exactly is the transcript data living and in which message does it flow from Alice to ACME?

gcsfred (Tue, 13 Aug 2019 18:54:48 GMT):
Link: https://github.com/hyperledger/education/blob/master/LFS171x/indy-material/nodejs/AgentDemoScript.md

gcsfred (Tue, 13 Aug 2019 19:22:24 GMT):
Another question: do credentials hold the very data?

swcurran (Tue, 13 Aug 2019 19:39:21 GMT):
The transcript data was given by Faber to Alice when it issued the credential in the Faber-Alice part of the demo. It is stored by Alice in the credential from Faber in her wallet. The claims from that credential flows from Alice to ACME in the proof that is sent in the "Credential-Presentation" message.

gcsfred (Tue, 13 Aug 2019 19:42:32 GMT):
Is the data encrypted in the credential message? Is the encrypted data stored in the blockchain?

swcurran (Tue, 13 Aug 2019 19:46:53 GMT):
In Aries, the message itself is encrypted from the sender (Alice) to the receiver (ACME). Once the message is decrypted, the presentation JSON with the claims is in plaintext. No data is sent to the Blockchain. Ever. It is only used for DIDs, schema/credential metadata and revocation information.

swcurran (Tue, 13 Aug 2019 19:47:16 GMT):
The things you need to verify the proof.

gcsfred (Tue, 13 Aug 2019 19:49:18 GMT):
Ok. Can a credential be revoked?

swcurran (Tue, 13 Aug 2019 21:46:00 GMT):
Yes-ish. Indy supports revocation. I'm not sure that any of the Aries implementations support revocation. ACA-Py does not, but it's predecessor, von_agent does.

swcurran (Tue, 13 Aug 2019 21:47:38 GMT):
Issuer can revoke at anytime. The Prover does a zero-knowledger "prove of non-revocation" to show that at some point in time, the credential was not revoked. The verifier defines to the prover an acceptable period for the proof.

sklump (Wed, 14 Aug 2019 13:10:00 GMT):
correction: von_anchor

sklump (Wed, 14 Aug 2019 13:10:00 GMT):
correction: von_anchor, not von_agent is current and allows for issuers to revoke. Revocation is under redesign for anoncreds-2.0.

sklump (Wed, 14 Aug 2019 13:10:00 GMT):
correction: von_anchor, not von_agent is current and allows for issuers to revoke. Revocation is under redesign at the indy level for anoncreds-2.0.

sklump (Wed, 14 Aug 2019 13:10:00 GMT):
correction: von_anchor, not von_agent is current and allows for issuers to revoke. Revocation is under redesign at the ursa and indy levels for anoncreds-2.0.

AlexITC (Wed, 14 Aug 2019 14:43:57 GMT):
I found this presentation and I wonder if this means that indy is going to be rewritten, or abandoned for a new project: https://docs.google.com/presentation/d/e/2PACX-1vQetphlWf2TEb-zarMnJFbvkK4r2USIgAYOyeKAMzCL5EU1nUMdZB4q8BSY6bJGpYi8D5TIB2ezEOHo/pub#slide=id.g4245334205_0_1344

dbluhm (Wed, 14 Aug 2019 14:53:40 GMT):
Indy will not be abandoned. The Consensus algorithm noted as having need for improvement and potentially replacement is truthfully a very small part of Indy. I believe those conversations eventually resolved to, rather than replace Plenum, improve it's view change strategy to that used by PBFT I think. The Indy team may be able to add more info there.

dbluhm (Wed, 14 Aug 2019 14:55:03 GMT):
It's also worth mentioning that what Indy is responsible for is evolving with the split off of the Indy Agent community into the Hyperledger Aries project.

AlexITC (Wed, 14 Aug 2019 14:56:36 GMT):
thanks!

mattatkiva (Wed, 14 Aug 2019 16:00:59 GMT):
For performance reasons we are considering cacheing (holding in memory) schema and credential definitions. is there any reason we should never cache a credential definition (thats the result of parseGetCredDefResponse)?

dbluhm (Wed, 14 Aug 2019 16:11:05 GMT):
I believe caching was recently added as a feature of the Indy SDK but I'm not sure on the details. Just making sure you're aware.

ShrinivasDeshmukh (Wed, 14 Aug 2019 16:11:57 GMT):
Hi @swcurran, while going through aries code for revocation, I found revocation is not supported. Here is the link: https://github.com/hyperledger/aries-cloudagent-python/blob/02050bd1d32f84807d8bc270318aee55949e8390/aries_cloudagent/holder/indy.py#L86 . Just curious to know if there is any specific reason for this?

ianco (Wed, 14 Aug 2019 17:55:21 GMT):
We (bc gov) cache credential definitions, I'm not sure offhand about the specifics

swcurran (Wed, 14 Aug 2019 18:01:25 GMT):
Yes - time :-). We're starting to talk about adding support for revocation now. We had hoped that the AnonCreds 2.0 work with the re-vamped revocation mechanism be in place before we needed it, but we suspect that we'll go with the existing mechanism. We're thinking that from the agent side, the 1.0 to 2.0 change will not be that significant vs. the overall effort to support revocation. The plan is to use as much of what we can from https://von-anchor.readthedocs.io, and in particular it's Tails server implementation - https://readthedocs.org/projects/von-tails/

swcurran (Wed, 14 Aug 2019 18:02:26 GMT):
Tails files are a big part of what makes the AnonCreds 1.0 implementation hard, and if we can just spin up a single tails server and use it, the 1.0 to 2.0 delta will be simpler.

esplinr (Thu, 15 Aug 2019 02:50:54 GMT):
Aries is the protocol for agent to agent communication. It is becoming widely adopted and is not going away any time soon. Indy is one method of storing the schemas and public keys necessary for a third party to verify Aries credentials. We don't expect it to go away any time soon.

esplinr (Thu, 15 Aug 2019 02:51:26 GMT):
But as @dbluhm explained, the consensus algorithm will need to evolve in order to support the scale we expect to need with the Sovrin network.

esplinr (Thu, 15 Aug 2019 02:52:18 GMT):
Our work to replace PBFT view change will address the biggest current problem, but there are many more problems that will need to be addressed over time. That presentation is to raise awareness, so that others will help in finding solutions.

esplinr (Thu, 15 Aug 2019 02:54:41 GMT):
The SDK credential caching was added in https://jira.hyperledger.org/browse/IS-1075

esplinr (Thu, 15 Aug 2019 02:54:55 GMT):
We did a demo in the Indy Working Group call on June 20.

esplinr (Thu, 15 Aug 2019 02:55:10 GMT):
https://wiki.hyperledger.org/pages/viewpage.action?pageId=13862386&preview=/13862386/16323578/GMT20190620-150649_Agent-Work_1920x1080.mp4

esplinr (Thu, 15 Aug 2019 02:55:42 GMT):
Start at minute 31.

StefanK (Thu, 15 Aug 2019 08:27:40 GMT):
Has joined the channel.

StefanK (Thu, 15 Aug 2019 08:27:41 GMT):
Hey Indyguys, iam stefan and writing my these at bosch.. i have about 45 days remaining x) .... i face a problem in the indy samples, i updatedet indy-lib to the newest version also the sdk and if i try to run the nodejs sample i get the following error. Can someone help ? :) Need to start programming my PoC >.< "Alice" -> Authdecrypt "Job-Application" Proof Request from Acme "Alice" -> Get credentials for "Job-Application" Proof Request "Alice" -> Get Schema from Ledger (node:27922) UnhandledPromiseRejectionWarning: IndyError: LedgerNotFound at Object.callback (/root/indy-sdk/samples/nodejs/node_modules/indy-sdk/src/wrapIndyCallback.js:15:10) (node:27922) UnhandledPromiseRejectionWarning: Unhandled promise rejection. This error originated either by throwing inside of an async function without a catch block, or by rejecting a promise which was not handled with .catch(). (rejection id: 1)

StefanK (Thu, 15 Aug 2019 08:29:39 GMT):
Also the Ledger is running :) the docker pool which is descripted in the indy-sdk

SushilsharmaHyper (Thu, 15 Aug 2019 13:17:33 GMT):
Has joined the channel.

SushilsharmaHyper (Thu, 15 Aug 2019 13:17:44 GMT):
Hello Team, I want create SSI project based on Indy,,any help would be highly appreciated..thank you

ShrinivasDeshmukh (Thu, 15 Aug 2019 13:40:24 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=dJTe8re629Y5gTzbo) @swcurran That's cool! Thanks for the update!

sklump (Thu, 15 Aug 2019 13:49:38 GMT):
Just a brief word of caution against irrational exuberance. Nothing about tails files is as straightforward as it looks, hence the hairball of https://raw.githubusercontent.com/PSPC-SPAC-buyandsell/von_anchor/master/docs/source/pic/dir_tails.png and rrbuilder.py. Expect delays.

dbluhm (Thu, 15 Aug 2019 14:45:12 GMT):
Identity WG Implementer's call will be starting in about 15 minutes! Vladimir Shishkin of Evernym will be talking about Indy performance testing after our regular project updates. Meeting agenda with zoom link: https://wiki.hyperledger.org/x/zhP5

esplinr (Thu, 15 Aug 2019 16:28:23 GMT):
This week's demo video from Evernym describing the new features in Indy that will be released later this month and next month: https://www.youtube.com/watch?v=Y7_XA15T_mg&list=PLRp0viTDxBWGLdZk0aamtahB9cpJGV7ZF * Discussion of the ongoing view change improvements * Demo of how to prove possession of payment address * Demo of how to endorse third party transactions * New Indy SDK images for Ubuntu 18.04

nicothnr (Thu, 15 Aug 2019 18:09:05 GMT):
Has joined the channel.

nicothnr (Thu, 15 Aug 2019 18:09:06 GMT):
Hi all, I'm conducting a survey for a research project examining the acceptance of self-sovereign identity (SSI) and the influence of digital privacy. I would be very grateful if you could take a few minutes to answer the questionnaire. The survey should take about 8 minutes to complete. I highly appreciate your support! Feel free to contact me if you have any questions. https://ubayreuthmarketing.qualtrics.com/jfe/form/SV_1H8sBpmvirZJKJL

RedKnight13 (Fri, 16 Aug 2019 06:33:19 GMT):
Has joined the channel.

AlexITC (Sun, 18 Aug 2019 04:11:32 GMT):
I wonder if I is possible to issue a credential without requiring the credential subject to accept it, for example, suppose that the subject is offline when the credential is issued and the subject doesn't have any agent that's online

sukalpomitra (Sun, 18 Aug 2019 05:52:08 GMT):
The acceptance comes at the offer state. Once the holder accepts, then credential is issued straight. For the point you mentioned, that is why a cloud agent should be used. So that when the edge agent comes online it can get the messages from the cloud agent.

kingpinXD (Sun, 18 Aug 2019 12:34:17 GMT):
Has joined the channel.

kingpinXD (Sun, 18 Aug 2019 12:34:18 GMT):
Hi all , I am just trying to learn more about SSI and how I can leverage INDY , I went over some documents, I am just confused about the relationship between DID and the public/private keys .

kingpinXD (Sun, 18 Aug 2019 12:35:56 GMT):
When I basically store a credential definition , do I use my DID to sign the transaction which goes to the ledger ?

AlexITC (Sun, 18 Aug 2019 15:23:02 GMT):
that's what I was wondering if can be avoided, the offering step

doyajii1 (Mon, 19 Aug 2019 00:07:34 GMT):
yes

doyajii1 (Mon, 19 Aug 2019 00:07:34 GMT):
Yes, I read that "Java wrapper currently requires Java 8" and I was wondering if there will be backward compatibility maybe up to java 1.7.

doyajii1 (Mon, 19 Aug 2019 00:07:34 GMT):
Yes, I read that "Java wrapper currently requires Java 8" and I was wondering if there will be backward compatibility maybe up to java 1.7 or so.

manavgupta (Mon, 19 Aug 2019 05:39:03 GMT):
Hi everyone, I am starting to work on indy. In indy i have setup the indy-node and now how should i start writing the codes ? Can anyone help me find the best possible way out

caveman7 (Mon, 19 Aug 2019 07:31:27 GMT):
From what I understand, the credential holder generates proof and the verifier perform proof validation without interacting with the Indy network. in the Verifiable Credentials ecosystem (https://github.com/WebOfTrustInfo/rwot7-toronto/blob/master/topics-and-advance-readings/verifiable-credentials-primer.md) it is mentioned that `Verifier verifies identifier ownership against identifier registry`. Can anybody clarify this step?

adityasingh177 (Mon, 19 Aug 2019 09:24:28 GMT):
Has joined the channel.

MALodder (Mon, 19 Aug 2019 11:55:07 GMT):
@caveman7 the holder and verifier will need to interact with the Indy network to use the public keys and check the revocation registry

simran (Mon, 19 Aug 2019 12:26:58 GMT):
`keyfordid`is working sometime or not why is it so ??? anyone have any idea about this

simran (Mon, 19 Aug 2019 12:28:07 GMT):
what i am lacking can you please give me some idea

Zohaib_Sohail (Mon, 19 Aug 2019 14:36:45 GMT):
i want to study indy-crypto in detail from where should i start please refer me some good documentation ? @WadeBarnes

WadeBarnes (Mon, 19 Aug 2019 14:40:37 GMT):
@Zohaib_Sohail, @swcurran, @sklump, would either of you be able to point @Zohaib_Sohail in the right direction?

MALodder (Mon, 19 Aug 2019 14:43:06 GMT):
@Zohaib_Sohail indy-crypto is deprecated. take a look at https://github.com/hyperledger/ursa

swcurran (Mon, 19 Aug 2019 14:47:37 GMT):
In Indy, the credential offer is required. You can read more about the process in the RFC on issuing a credential, with a link to the code in Indy. https://github.com/hyperledger/aries-rfcs/tree/master/features/0036-issue-credential#offer-credential

swcurran (Mon, 19 Aug 2019 14:50:19 GMT):
There are lots of confused statements in what you've written here - suggest that you keep reading. Perhaps this document might give you the background - it used to be a chapter in the edX Blockchain for Business course - https://github.com/swcurran/education/blob/master/LFS171x/docs/introduction-to-hyperledger-indy.md

swcurran (Mon, 19 Aug 2019 14:53:01 GMT):
A quick try at your comments: - Your cred def contains your DID and goes to the ledger - that's right. - a DID is resolved to a DID Doc. The DID Doc contains the public key, and you hold the private key - a Cred Def also contains a set of public keys that are used to sign the claims within the credentials that will be issued against the cred def.

swcurran (Mon, 19 Aug 2019 14:54:52 GMT):
Depends on your goal. A good way to get started is to use the `aries-cloudagent-python` to get started. This guide gives you a starting point. If you know enough about Indy, you can skip the first (although it's good if you are just getting started). The rest summarize getting going in development: https://github.com/hyperledger/aries-cloudagent-python/tree/master/docs/GettingStartedAriesDev

swcurran (Mon, 19 Aug 2019 14:57:22 GMT):
Both the holder and the verifier interact with the Indy network to do the proof and verification, respectively. In Indy, that statement from the RWoT paper is not correct. Indy uses a different approach to prove that the holder owns the credential - a ZKP that does not disclose correlating data about the holder.

kingpinXD (Mon, 19 Aug 2019 20:04:27 GMT):
okay the last point is what made it all clear

kingpinXD (Mon, 19 Aug 2019 20:08:21 GMT):
Hey Guys , I was wondering, what's your opinion about using indy for enterprise blockchain. Where the members of the system are different applications, which are onboarded on the platform by a master application. These applications are then given a role of trust anchors, they can then issue credentials to their users. The users can then use the credentials to when they want to use the the rest apis exposed by these applications. Effectively removing the need for the user to login, they sort of just send the zkp in the header. Is anybody working on a project like this? does an agent like this exist which enables an application ( Basically a webserver with certain routes to be onboarded ?

galaxystar (Tue, 20 Aug 2019 01:26:16 GMT):
Has joined the channel.

esplinr (Tue, 20 Aug 2019 03:59:50 GMT):
How would the holder save the credential if they can't receive it?

James_Manger (Tue, 20 Aug 2019 04:34:20 GMT):
Has joined the channel.

rcherukuri (Tue, 20 Aug 2019 05:47:29 GMT):
Has joined the channel.

AlexITC (Tue, 20 Aug 2019 16:38:23 GMT):
when issuing a credential, first a credential offer is sent which requires access to the issuer wallet, I wonder if creating this offer requires the private key that will be used for actually issuing the credential

swcurran (Tue, 20 Aug 2019 16:43:59 GMT):
You need to look through the SDK for that. Based on the wording in the RFC for issuing a credential, yes - the issuer is making a cryptographic commitment. There is a link later in the section to the IndySDK code. https://github.com/hyperledger/aries-rfcs/tree/master/features/0036-issue-credential#offer-credential

AlexITC (Tue, 20 Aug 2019 16:45:54 GMT):
that's my understanding, is there a way to use diffferent keys? like issuing a key that issues certificates, an another key that only signs credential offers

swcurran (Tue, 20 Aug 2019 16:54:24 GMT):
Sorry - don't know it that level of detail. I think that's getting into Ursa territory. Perhaps @MALodder might be able to answer that.

AlexITC (Tue, 20 Aug 2019 16:54:50 GMT):
thanks

MALodder (Tue, 20 Aug 2019 17:19:43 GMT):
The key used for signing credentials can't be used for signing anything else. I'm not clear on what you mean by signing certificates

AlexITC (Tue, 20 Aug 2019 17:21:28 GMT):
there are two phases for issuing a credential, a credential offer is send, and when the subject replies with the credential request, the issuer issues the credential

AlexITC (Tue, 20 Aug 2019 17:22:20 GMT):
My problem is that I want to issue a bunch of credentials and the subject may not be available to reply to the credential offer, which would lead me to keep the issuer key hot for a good amount of time instead of just issuing the batch and move the key to cold storage

AlexITC (Tue, 20 Aug 2019 17:37:51 GMT):
I see two potential solutions to that problem, the one stated above, or, a cloud agent acting on the subject behalf that accepts credential offers and replies with credential request, but I would only consider that if the agent can't do more actions than that, which I don't know if is possible

MALodder (Tue, 20 Aug 2019 17:39:02 GMT):
agents can do just that phase

AlexITC (Tue, 20 Aug 2019 18:56:00 GMT):
that may work, I will research if they can just do that and nothing else, thanks!

EtricKombat (Wed, 21 Aug 2019 05:59:48 GMT):
Has joined the channel.

EtricKombat (Wed, 21 Aug 2019 05:59:50 GMT):
I have a question related to an npm error , this happens when i try to run an Indy project . Error message : npm ERR! code ECONNRESET npm ERR! errno ECONNRESET . If someone could shred some light it would be helpful. I configured npm proxy already in this format ---> npm config set https-proxy http://:@:

ShrinivasDeshmukh (Wed, 21 Aug 2019 13:41:56 GMT):
Hi, I see there is a method `buildGetRevocRegDeltaRequest` in the npm indy-sdk, but we are not writing the revocRegDelta to the ledger. I cannot find a method in the npm indy-sdk to write the `revocRegDelta` to the ledger. As per my understanding, `issuerCreateCredential` returns revocRegDelta as well. Once the credential is saved to wallet, how should we fetch the revocRegDelta? I need the `revocRegDelta` for creating revocation state.

EtricKombat (Thu, 22 Aug 2019 03:10:13 GMT):
Hi Folks.... While i tries to start up TheOrgBook instance i've been getting the below listed errror for the past 2 days....any idea why this happens? Recreating 774248dac3ef_tob_tob-api_1 ... error ERROR: for tob-api Cannot start service tob-api: driver failed programming external connectivity on endpoint tob_tob-api_1 (c866bc0703b7c134551a7ffa0630cc86a052c599bb9041b929aa9129821c7c3b): Error starting userland proxy: listen tcp 0.0.0.0:3000: bind: address already in use ERROR: Encountered errors while bringing up the project.

EtricKombat (Thu, 22 Aug 2019 03:11:30 GMT):
i don't use proxy for this one...as i was trying it out from my home network.

andrew.whitehead (Thu, 22 Aug 2019 03:51:41 GMT):
It looks like something else is using port 3000. ‘docker ps’ might provide more info

esune (Thu, 22 Aug 2019 16:57:36 GMT):
What @andrew.whitehead said seems correct. I have also seen this issue (on Windows) when the Docker daemon got itself in trouble for some reason. Restarting Docker fixed it for me - assuming the port is really not in use already.

EtricKombat (Thu, 22 Aug 2019 21:03:23 GMT):
docker ps``` ``` ``` ``` CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES``` ``` ``` ``` b76260a0d894 solr "docker-entrypoint.s…" 46 seconds ago Up 41 seconds 0.0.0.0:8983->8983/tcp tob_tob-solr_1``` ``` 09b1a1eb22b2 postgresql "container-entrypoin…" 46 seconds ago Up 41 seconds 0.0.0.0:5433->5432/tcp tob_wallet-db_1``` ``` ``` ``` 14f9fbdacdd4 postgresql "container-entrypoin…" 46 seconds ago Up 41 seconds 0.0.0.0:5432->5432/tcp tob_tob-db_1

EtricKombat (Thu, 22 Aug 2019 21:03:23 GMT):
docker ps CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES b76260a0d894 solr "docker-entrypoint.s…" 46 seconds ago Up 41 seconds 0.0.0.0:8983->8983/tcp tob_tob-solr_1 09b1a1eb22b2 postgresql "container-entrypoin…" 46 seconds ago Up 41 seconds 0.0.0.0:5433->5432/tcp tob_wallet-db_1 14f9fbdacdd4 postgresql "container-entrypoin…" 46 seconds ago Up 41 seconds 0.0.0.0:5432->5432/tcp tob_tob-db_1

EtricKombat (Thu, 22 Aug 2019 21:03:23 GMT):
docker ps CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES b76260a0d894 solr "docker-entrypoint.s…" 46 seconds ago Up 41 seconds 0.0.0.0:8983->8983/tcp tob_tob-solr_1``` ``` 09b1a1eb22b2 postgresql "container-entrypoin…" 46 seconds ago Up 41 seconds 0.0.0.0:5433->5432/tcp tob_wallet-db_1``` ``` 14f9fbdacdd4 postgresql "container-entrypoin…" 46 seconds ago Up 41 seconds 0.0.0.0:5432->5432/tcp tob_tob-db_1

EtricKombat (Thu, 22 Aug 2019 21:03:23 GMT):
docker ps CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES``` ``` ``` ``` b76260a0d894 solr "docker-entrypoint.s…" 46 seconds ago Up 41 seconds 0.0.0.0:8983->8983/tcp tob_tob-solr_1``` ``` 09b1a1eb22b2 postgresql "container-entrypoin…" 46 seconds ago Up 41 seconds 0.0.0.0:5433->5432/tcp tob_wallet-db_1``` ``` 14f9fbdacdd4 postgresql "container-entrypoin…" 46 seconds ago Up 41 seconds 0.0.0.0:5432->5432/tcp tob_tob-db_1

EtricKombat (Thu, 22 Aug 2019 21:19:26 GMT):
Hope this is helpful. Checked ist: -tried restarting each of these containers. -docker login - checked & confirmed -There is no proxy configuration for this issue (In issue mentioned above by me related to npm error i 'm using an proxy enivronment.

ianco (Thu, 22 Aug 2019 23:40:57 GMT):
For this error `Error starting userland proxy: listen tcp 0.0.0.0:3000: bind: address already in use` the offending port is `3000`, and I'm not sure what's running on this port. If you comment out the port mapping in the docker-compose.yml file `#- 3000:3000` everything seems to still work. @andrew.whitehead any thoughts?

andrew.whitehead (Thu, 22 Aug 2019 23:53:02 GMT):
That port is only used by the ptvsd debugger, so likely fine to disable the mapping

andrew.whitehead (Thu, 22 Aug 2019 23:53:02 GMT):
That port is only used by the ptvsd debugger, so likely fine to disable the mapping. Not sure what else uses that port, could be a virus scanner or something

sukalpomitra (Fri, 23 Aug 2019 02:41:51 GMT):
@ianco you can use netstat to find what is running on 3000 and then use the pid to kill the process

ianco (Fri, 23 Aug 2019 04:26:19 GMT):
It would be something running inside the docker container though ...

sukalpomitra (Fri, 23 Aug 2019 04:59:27 GMT):
I doubt it as your docker-compose is failing while starting the container.. so I think something is already running and if you get the pid you can kill it

ShashankKulkarni (Fri, 23 Aug 2019 10:12:01 GMT):
Has joined the channel.

ShashankKulkarni (Fri, 23 Aug 2019 10:13:18 GMT):
Hi everyone. I am always getting false response from verifierVerifyProof indy-sdk method. I am fetching schema, cred-defs, revocRegDef, revocRegDelta from the ledger. Still getting the response as false. Can anyone help?

ShashankKulkarni (Fri, 23 Aug 2019 10:13:18 GMT):
Hi everyone. I am always getting false response from verifierVerifyProof indy-sdk method. I am fetching schema, cred-defs, revocRegDef, revocRegDelta from the ledger. Still getting the response as false. @swcurran Can you please help?

DennisM330 (Fri, 23 Aug 2019 11:43:20 GMT):
Has joined the channel.

swcurran (Fri, 23 Aug 2019 16:41:36 GMT):
@andrew.whitehead @nbrempel ^^^ Nick - is this the same thing you are seeing?

nbrempel (Fri, 23 Aug 2019 16:49:14 GMT):
Maybe, @ShashankKulkarni can you paste the full error including stacktrace?

nage (Fri, 23 Aug 2019 19:59:31 GMT):
If you have a contribution to Hyperledger code bases in the past 12 months (Indy, Aries, Ursa, and more) and haven't received an email with the subject "[Hyperledger Contributors] 2019-2020 TSC elections starting soon" then you didn't catch the email with instructions for being included in the upcoming Hyperledger elections or may have a problem that prevents you from voting. PM me if this is you, and I will see what we can do to help.

jeffanon (Sat, 24 Aug 2019 23:36:29 GMT):
Has joined the channel.

ShashankKulkarni (Mon, 26 Aug 2019 12:30:22 GMT):
There is no error message, just the verification method returns False. Here are the steps I am doing: 1. Create schema and write to ledger 2. create cred def and write to ledger 3. create revoc reg and write the delta and revoc reg to ledger 4. create cred offer 5. issuer cred 5. Save cred to wallet 6. generate proof_request 7. Fetch revocRegDelta and revocRegDef from ledger 8. generate proof 9. Fetch revocRegDelta and revocRegDef from ledger 10. Verify proof

ShashankKulkarni (Mon, 26 Aug 2019 12:30:37 GMT):
verify proof methods returns false everytime

ShashankKulkarni (Mon, 26 Aug 2019 12:31:33 GMT):
@nbrempel This is my proof request `{ "nonce": "25474750", "name": "Income", "version": "1.0", "requested_attributes": { "attr1_referent": { "name": "Salary" }, "attr2_referent": { "name": "Firstname" } }, "requested_predicates": {}, "non_revoked": { "from": 1566817711, "to": nowSeconds } }`

ShashankKulkarni (Mon, 26 Aug 2019 12:31:33 GMT):
@nbrempel This is my proof request ```{ "nonce": "25474750", "name": "Income", "version": "1.0", "requested_attributes": { "attr1_referent": { "name": "Salary" }, "attr2_referent": { "name": "Firstname" } }, "requested_predicates": {}, "non_revoked": { "from": 1566817711, "to": nowSeconds } }```

ShashankKulkarni (Mon, 26 Aug 2019 12:31:33 GMT):
@nbrempel This is my proof request ```{ "nonce": "25474750", "name": "Income", "version": "1.0", "requested_attributes": { "attr1_referent": { "name": "Salary" }, "attr2_referent": { "name": "Firstname" } }, "requested_predicates": {}, "non_revoked": { "from": 0, "to": nowSeconds } }```

ShashankKulkarni (Mon, 26 Aug 2019 12:32:45 GMT):
This is revocRegDef: ```{ "D29a5Qew7ZNkNiq64Qu5q2:4:D29a5Qew7ZNkNiq64Qu5q2:3:CL:24:income1:CL_ACCUM:55090": { "ver": "1.0", "id": "D29a5Qew7ZNkNiq64Qu5q2:4:D29a5Qew7ZNkNiq64Qu5q2:3:CL:24:income1:CL_ACCUM:55090", "revocDefType": "CL_ACCUM", "tag": "55090", "credDefId": "D29a5Qew7ZNkNiq64Qu5q2:3:CL:24:income1", "value": { "issuanceType": "ISSUANCE_ON_DEMAND", "maxCredNum": 500, "publicKeys": { "accumKey": { "z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} }, "tailsHash": "EUp9qrW1L6kFC1J5ju3n6UxmoHTvmxi4MMCe4gfwea8C", "tailsLocation": "/home/ayanworks-14/indy-solution/protech-indy-solution/Server/revoc/cred-def-income1/EUp9qrW1L6kFC1J5ju3n6UxmoHTvmxi4MMCe4gfwea8C" } } }```

ShashankKulkarni (Mon, 26 Aug 2019 12:32:45 GMT):
This is revocRegDef: ```{ "D29a5Qew7ZNkNiq64Qu5q2:4:D29a5Qew7ZNkNiq64Qu5q2:3:CL:24:income1:CL_ACCUM:55090": { "ver": "1.0", "id": "D29a5Qew7ZNkNiq64Qu5q2:4:D29a5Qew7ZNkNiq64Qu5q2:3:CL:24:income1:CL_ACCUM:55090", "revocDefType": "CL_ACCUM", "tag": "55090", "credDefId": "D29a5Qew7ZNkNiq64Qu5q2:3:CL:24:income1", "value": { "issuanceType": "ISSUANCE_ON_DEMAND", "maxCredNum": 500, "publicKeys": { "accumKey": { "z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} }, "tailsHash": "EUp9qrW1L6kFC1J5ju3n6UxmoHTvmxi4MMCe4gfwea8C", "tailsLocation": "/home/ayanworks-14/indy-solution/PATH_TO_SERVER/Server/revoc/cred-def-income1/EUp9qrW1L6kFC1J5ju3n6UxmoHTvmxi4MMCe4gfwea8C" } } }```

ShashankKulkarni (Mon, 26 Aug 2019 12:32:45 GMT):
This is revocRegDef: ```{ "D29a5Qew7ZNkNiq64Qu5q2:4:D29a5Qew7ZNkNiq64Qu5q2:3:CL:24:income1:CL_ACCUM:55090": { "ver": "1.0", "id": "D29a5Qew7ZNkNiq64Qu5q2:4:D29a5Qew7ZNkNiq64Qu5q2:3:CL:24:income1:CL_ACCUM:55090", "revocDefType": "CL_ACCUM", "tag": "55090", "credDefId": "D29a5Qew7ZNkNiq64Qu5q2:3:CL:24:income1", "value": { "issuanceType": "ISSUANCE_ON_DEMAND", "maxCredNum": 500, "publicKeys": { "accumKey": { "z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} }, "tailsHash": "EUp9qrW1L6kFC1J5ju3n6UxmoHTvmxi4MMCe4gfwea8C", "tailsLocation": "PATH_TO_SERVER/revoc/cred-def-income1/EUp9qrW1L6kFC1J5ju3n6UxmoHTvmxi4MMCe4gfwea8C" } } }```

ShashankKulkarni (Mon, 26 Aug 2019 12:35:23 GMT):
@nbrempel This is my revocDelta: ```{ '1566817711': { ver: '1.0', value: { prevAccum: '1 0000000000000000000000000000000000000000000000000000000000000000 1 0000000000000000000000000000000000000000000000000000000000000000 2 095E45DDF417D05FB10933FFC63D474548B7FFFF7888802F07FFFFFF7D07A8A8 1 0000000000000000000000000000000000000000000000000000000000000000 1 0000000000000000000000000000000000000000000000000000000000000000 1 0000000000000000000000000000000000000000000000000000000000000000', accum: '1 0000000000000000000000000000000000000000000000000000000000000000 1 0000000000000000000000000000000000000000000000000000000000000000 2 095E45DDF417D05FB10933FFC63D474548B7FFFF7888802F07FFFFFF7D07A8A8 1 0000000000000000000000000000000000000000000000000000000000000000 1 0000000000000000000000000000000000000000000000000000000000000000 1 0000000000000000000000000000000000000000000000000000000000000000' } } } ```

ShashankKulkarni (Mon, 26 Aug 2019 12:35:23 GMT):
@nbrempel This is my revocDelta: ```{ 'D29a5Qew7ZNkNiq64Qu5q2:4:D29a5Qew7ZNkNiq64Qu5q2:3:CL:24:income1:CL_ACCUM:55090': {'1566817711': { ver: '1.0', value: { prevAccum: '1 0000000000000000000000000000000000000000000000000000000000000000 1 0000000000000000000000000000000000000000000000000000000000000000 2 095E45DDF417D05FB10933FFC63D474548B7FFFF7888802F07FFFFFF7D07A8A8 1 0000000000000000000000000000000000000000000000000000000000000000 1 0000000000000000000000000000000000000000000000000000000000000000 1 0000000000000000000000000000000000000000000000000000000000000000', accum: '1 0000000000000000000000000000000000000000000000000000000000000000 1 0000000000000000000000000000000000000000000000000000000000000000 2 095E45DDF417D05FB10933FFC63D474548B7FFFF7888802F07FFFFFF7D07A8A8 1 0000000000000000000000000000000000000000000000000000000000000000 1 0000000000000000000000000000000000000000000000000000000000000000 1 0000000000000000000000000000000000000000000000000000000000000000' } } } } ```

ShashankKulkarni (Mon, 26 Aug 2019 13:15:03 GMT):

ShashankKulkarni - Mon Aug 26 2019 18:44:54 GMT+0530 (India Standard Time).txt

ShashankKulkarni (Mon, 26 Aug 2019 13:15:03 GMT):

ShashankKulkarni - Mon Aug 26 2019 18:44:54 GMT+0530 (India Standard Time).txt

ShashankKulkarni (Mon, 26 Aug 2019 13:15:03 GMT):

ShashankKulkarni - Mon Aug 26 2019 18:44:54 GMT+0530 (India Standard Time).txt

ShashankKulkarni (Mon, 26 Aug 2019 13:15:03 GMT):

ShashankKulkarni - Mon Aug 26 2019 18:44:54 GMT+0530 (India Standard Time).txt

ShashankKulkarni (Mon, 26 Aug 2019 13:16:24 GMT):
@nbrempel That is my proof_json

esplinr (Mon, 26 Aug 2019 15:00:54 GMT):
We are about to start the Indy Maintainers call: https://zoom.us/j/615818107

mattatkiva (Tue, 27 Aug 2019 18:40:01 GMT):
I'm trying to run the performance scripts (at ) I am getting this error: ```root@b7c5ed6ca7cc:/src/ignore/scripts/perf_load# python3 perf_processes.py -c 100 -n 10 -k get_schema -g ../pool_transactions_genesis Traceback (most recent call last): File "perf_processes.py", line 18, in from indy import pool ImportError: No module named 'indy'``` I do have indy installed: ```apt-get install libindy Reading package lists... Done Building dependency tree Reading state information... Done libindy is already the newest version (1.11.0~1294).``` any other ideas what is wrong?

mattatkiva (Tue, 27 Aug 2019 18:40:01 GMT):
I'm trying to run the performance scripts (at https://github.com/hyperledger/indy-node/tree/master/scripts/performance/perf_load) I am getting this error: ```root@b7c5ed6ca7cc:/src/ignore/scripts/perf_load# python3 perf_processes.py -c 100 -n 10 -k get_schema -g ../pool_transactions_genesis Traceback (most recent call last): File "perf_processes.py", line 18, in from indy import pool ImportError: No module named 'indy'``` I do have indy installed: ```apt-get install libindy Reading package lists... Done Building dependency tree Reading state information... Done libindy is already the newest version (1.11.0~1294).``` any other ideas what is wrong?

mattatkiva (Tue, 27 Aug 2019 18:48:47 GMT):
the answer is: https://pypi.org/project/python3-indy/

mattatkiva (Tue, 27 Aug 2019 18:57:37 GMT):
however I get an additional error: I have this error now ImportError: No module named 'perf_load'

mattatkiva (Tue, 27 Aug 2019 18:57:37 GMT):
however I get an additional error: I have this error now `ImportError: No module named 'perf_load'`

andrew.whitehead (Tue, 27 Aug 2019 19:36:48 GMT):
Try `pip3 install -e .` from the parent directory

AlexITC (Wed, 28 Aug 2019 16:38:32 GMT):
from my understanding, issuing credentials doesn't need to store a proof in the ledger, is there a simple way to get a proof on the ledger? or is it designed this way?

dbluhm (Wed, 28 Aug 2019 16:43:25 GMT):
Can you clarify what your intent is for having a proof on the ledger? A really short answer would be yes it is designed this way

AlexITC (Wed, 28 Aug 2019 20:44:38 GMT):
the reason is to be able to detect fake credentials in case of the key getting compromised, without they being in the ledger, there is no way to detect them

dbluhm (Wed, 28 Aug 2019 20:50:16 GMT):
Calling in @MALodder to check my answer but (assuming the key you're talking about is the issuer's key being compromised) when the issuer discovers that it's issuing key has been compromised, the issuer rotates the key. Since issued credentials are tied to the DID and not the key of the issuer in order to verify it, rotating and updating the DIDDoc with the new key would effectively invalidate all credentials issued with the previous key.

AlexITC (Wed, 28 Aug 2019 20:50:51 GMT):
so, rotating the key means that you need to reissue everything

dbluhm (Wed, 28 Aug 2019 20:52:43 GMT):
I believe so but I'm not an expert on this

AlexITC (Wed, 28 Aug 2019 20:53:18 GMT):
in fact, it makes sense, I think I remember I saw something like that in the source code

AlexITC (Wed, 28 Aug 2019 20:53:20 GMT):
thanks

dbluhm (Wed, 28 Aug 2019 20:53:37 GMT):
:thumbup:

brentzundel (Wed, 28 Aug 2019 22:06:09 GMT):
to clarify: the public keys for an issuer's credential signature are in the cred def. The cred def is linked to the issuer's DID, but the DID keys are not otherwise related to the keys used to issue a credential, i.e., they aren't stored in the DIDDoc. Rotation of the keys in a cred def would make all previously issued credentials unverifiable, essentially revoking them all and requiring a re-issuance of credentials using the new keys.

dbluhm (Wed, 28 Aug 2019 22:25:00 GMT):
Thanks for the clarification!

manavgupta (Thu, 29 Aug 2019 03:47:45 GMT):
hi everyone, I was executing the build android script for libnullpay in indy-sdk. I am facing some error i.e.. ("linking with `/tmp/android_build/toolchains/linux/x86/bin/i686-linux-android-clang` failed: exit code: 1" and "Could not compile 'null-payment-method'") so if anyone could help me with this??

EtricKombat (Thu, 29 Aug 2019 12:31:22 GMT):
I've been getting below listed errors for the web UI(angular-on-nginx image) when i try building the TheOrgBook using "./manage build" command for the past few days npm ERR! code ECONNRESET npm ERR! errno ECONNRESET npm ERR! network request to https://github.com/cywolf/bootstrap-theme/releases/download/v4.1.0-rc0/bcgov-bootstrap-theme-4.1.0-rc0.tgz failed, reason: read ECONNRESET Is there any specific location i need to setup proxy again? am i missing any specific location? I have set the proxy config correctly for listed (verified) 1) git proxy 2) apt proxy 3) docker daemon proxy 4) docker config proxy 5) Environment proxy 6) Exported proxy 7) NPM proxy config

ptab-pawan (Thu, 29 Aug 2019 13:20:36 GMT):
Performed the below steps on ubuntu VM : docker build --build-arg pool_ip=192.168.179.90 -f ci/indy-pool.dockerfile -t indy_pool . docker run -itd -p 192.168.179.90:9701-9708:9701-9708 indy_pool Gave the Ubuntu VM IP address in the docker_pool_transaction_genesis file. I created a pool on ubuntu VM : default-pool Took the config.json file of the default-pool : {"genesis_txn":"/root/indy_sdk_1.11.0/indy-sdk/cli/docker_pool_transactions_genesis"} Used Java sample from below location in react native android app : https://github.com/hyperledger/indy-sdk/tree/master/samples/java/src/main/java I am using libIndy 1.11.0 and able to load libIndy successfully in android app. : Used PoolUtils example from https://github.com/hyperledger/indy-sdk/blob/master/wrappers/java/src/test/java/org/hyperledger/indy/sdk/utils/PoolUtils.java String poolConfig = "{\"genesis_txn\": \"/root/indy_sdk_1.11.0/indy-sdk/cli/docker_pool_transactions_genesis\"}"; Pool.setProtocolVersion(2).get(); String poolName = PoolUtils.createPoolLedgerConfig(); Log.i("indy_sdk", poolName); System.out.println("\n2. Open pool ledger and get the pool handle from libindy.\n"); Pool pool = Pool.openPoolLedger(poolName, poolConfig).get(); System.out.println("\n2. Open pool ledger.\n"); Error : indy_sdk: org.hyperledger.indy.sdk.InvalidStateException: The SDK library experienced an unexpected internal error. If a pool is already created on ubuntu VM how can I connect to the existing pool by giving VM ipadress? As using another example I tried creating pool from mobile app and that throws error pool Error: org.hyperledger.indy.sdk.pool.PoolConfigNotCreatedException: The requested pool cannot be opened because it does not have an existing configuration.

ptab-pawan (Thu, 29 Aug 2019 13:23:27 GMT):
@sukalpomitra : How to specify IP address to a Pool when the pool already exists and you just want to connect to it ?

dbluhm (Thu, 29 Aug 2019 13:33:17 GMT):
Identity WG Implementers Call will be happening in about an hour and half! Agenda: https://wiki.hyperledger.org/x/MSD5 We will have our project updates with a more in depth update on the Aries SDK Architecture discussions. If you have other topics you would like to add to the agenda, feel free to edit the agenda yourself or contact me or @nage.

ShrinivasDeshmukh (Thu, 29 Aug 2019 14:16:53 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=Rm3PutHe3xeNngs43) @ptab-pawan You need to copy the pool.txn as well. It's better if you create a pool config on mobile side, using the pool.txn as gen txn file.

swcurran (Thu, 29 Aug 2019 15:23:19 GMT):
We'll take a look at this. The URL you gave is suspect, and that might be the issue. @andrew.whitehead - can you please also take a look?

esplinr (Thu, 29 Aug 2019 15:57:30 GMT):
Evernym's demo of improvements in Indy Node and Indy SDK: https://www.youtube.com/watch?v=bPppdiMBJ_I&list=PLRp0viTDxBWGLdZk0aamtahB9cpJGV7ZF&index=1 * New architecture documentation for Indy SDK * Fixes for GET_TXN * Upcoming releases, including support for Ubuntu 18.04 and CentOS

tarun.sharma (Thu, 29 Aug 2019 20:48:37 GMT):
Hi, I installed libindy but using shell script in indy-sdk but I still do not see libindy.dylib in my /usr/local/lib/. Is there a reason why? I got libindy installed in green on my terminal.

tarun.sharma (Thu, 29 Aug 2019 20:48:37 GMT):
Hi, I installed libindy using shell script in indy-sdk but I still do not see libindy.dylib in my /usr/local/lib/. Is there a reason why? I got libindy installed in green on my terminal.

sukalpomitra (Fri, 30 Aug 2019 02:55:00 GMT):
Hi Tarun if you are in Macbook then only you will get a dylib. In Ubuntu its a so file. Look for them at libindy target folder. You should see it where you have kept your cargo.toml and you have run the cargo build command

AlexITC (Fri, 30 Aug 2019 04:27:37 GMT):
is there any hardware wallet supporting indy?

manavgupta (Fri, 30 Aug 2019 06:01:18 GMT):
hi everyone I am trying to execute Connector app of sovrin foundation. I am facing problem in executing the build-android file for libnullpay. The error is as followed ("linking with /tmp/android_build/toolchains/linux/x86/bin/i686-linux-android-clang failed: exit code: 1" and "Could not compile 'null-payment-method'") Anyone could help??

adave (Fri, 30 Aug 2019 14:00:42 GMT):
Has joined the channel.

esplinr (Sat, 31 Aug 2019 05:40:57 GMT):
Not yet.

esplinr (Sat, 31 Aug 2019 05:41:37 GMT):
Have you installed LibIndy and LibNulPay, or LibIndy and LibSovToken?

manavgupta (Mon, 02 Sep 2019 04:48:07 GMT):
I have provided the path to the so files. Do we need to install in the system??

ShashankKulkarni (Mon, 02 Sep 2019 07:20:21 GMT):
@Artemkaaas can you please explain how timestamp should used in case of revocation

ShashankKulkarni (Mon, 02 Sep 2019 07:20:21 GMT):
@Artemkaaas can you please explain how timestamp should be used in case of revocation

nitika-goel (Mon, 02 Sep 2019 08:21:26 GMT):
Has joined the channel.

nitika-goel (Mon, 02 Sep 2019 08:21:28 GMT):
Hi, I have been through the demo at https://github.com/hyperledger/education/tree/master/LFS171x/indy-material/nodejs. I was able to successfully launch the demo, but this installs 4 nodes on the fly along with 5 agents. I need the following: 1. Launch nodes on the server 2. Add agents as and when required based on sign up functionality. 3. Able to add new nodes on another server which can sync with the current existing nodes. Can someone help me with these please?

swcurran (Mon, 02 Sep 2019 17:38:50 GMT):
@nitika-goel - that demo is pretty old and I would not recommend using that for anything more than understanding the concepts in self-sovereign identity. If you are just getting started in the space, you should be looking at Aries code, such as aries-cloudagent-python. This getting started guide in that repo extends the edX blockchain course that (likely) led you to that demo - https://github.com/hyperledger/aries-cloudagent-python/tree/master/docs/GettingStartedAriesDev There is guidance in the von-network repo to help you run the nodes on a server. https://github.com/bcgov/von-network

RameshT (Tue, 03 Sep 2019 18:17:57 GMT):
Hi, from what I understand from your query, the example which you are running is on Docker instances which doesn't give us control on the code. What you can do is , seperate the NodeJS code and try to run as independent service rather that in the Docker instance. This will give you more control over the code. We have done the same and it works.

RameshT (Tue, 03 Sep 2019 18:17:57 GMT):
Hi, from what I understand from your query, the example which you are running is on Docker instances which doesn't give us control on the code. What you can do is , seperate the NodeJS code and try to run as independent service rather that in the Docker instance. This will give you more control over the code for customization as per your requirements. We have done the same and it works.

VenkateshSYS (Wed, 04 Sep 2019 11:25:50 GMT):
Has joined the channel.

doyajii1 (Fri, 06 Sep 2019 02:36:36 GMT):
do anyone know what's the maximum message length for anonCrypt function?

doyajii1 (Fri, 06 Sep 2019 02:36:36 GMT):
do anyone know what's the maximum message length for anonCrypt function? https://github.com/hyperledger/indy-sdk/blob/adfdec0ddaee158060f822c8f0810d8f286ae7ac/wrappers/java/src/main/java/org/hyperledger/indy/sdk/crypto/Crypto.java

VipinB (Fri, 06 Sep 2019 17:18:37 GMT):
Is Self-Sovereign Identity training available through Hyperledger or any of the MooC platforms?

Dan (Fri, 06 Sep 2019 17:34:25 GMT):
*Maintainers* I don't see any of you registered: https://wiki.hyperledger.org/display/events/Maintainer+Summit+October+8-10,+2019 (except Nathan)

drewau (Fri, 06 Sep 2019 20:31:38 GMT):
Has joined the channel.

VenkateshSorapalli (Fri, 06 Sep 2019 20:44:53 GMT):
Has joined the channel.

VenkateshSYS (Sat, 07 Sep 2019 07:02:07 GMT):
Hello Guys, From this url of hyperledger-indy demo, https://github.com/hyperledger/education/tree/master/LFS171x/indy-material/nodejs all the nodes and agents (which are static) are running in the docker container automatically when runs the docker-compose.yml file. Username and Password are defined in the docker-compose.yml file in the demo which is static process How to do the sign up functionality dynamically ?

zac (Sat, 07 Sep 2019 15:14:10 GMT):
Has joined the channel.

kukkig (Mon, 09 Sep 2019 12:38:27 GMT):
hi everyone i have a simple question here, can a holder revoke sharing proof after we have shared it to a verifier ?

sklump (Mon, 09 Sep 2019 12:43:25 GMT):
@kukkig Revocation operates on credentials, not proofs, and only the issuer can revoke a credential.

ptab-pawan (Mon, 09 Sep 2019 13:09:27 GMT):
Hi @ShrinivasDeshmukh : I am facing issue : indy_sdk: org.hyperledger.indy.sdk.InvalidStateException: The SDK library experienced an unexpected internal error. What is going wrong. using the latest indy framework files.

ShrinivasDeshmukh (Mon, 09 Sep 2019 13:51:40 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=dH4gEY6qMF3vAvhQ2) @ptab-pawan When did you face this error? This error occurred to me when the pool.txn had some erroneous data, and I was trying to connect using the same pool config

esplinr (Mon, 09 Sep 2019 15:02:32 GMT):
Starting the Indy Contributors call now: https://zoom.us/j/615818107

cam-parra (Mon, 09 Sep 2019 18:43:39 GMT):
@ashcherbakov do you know where in the plenum codebase a libsodium dependency is introduced?

yutopyan (Mon, 09 Sep 2019 19:22:17 GMT):
openid

cam-parra (Mon, 09 Sep 2019 20:19:19 GMT):
@esplinr @ashcherbakov could your team please prioritize this PR: https://github.com/hyperledger/indy-node/pull/1443 . It contains the changes to support bionic but I need help from your devops to see if I configured everything correctly

ptab-pawan (Tue, 10 Sep 2019 07:01:30 GMT):
When i am trying to call the method :Pool.createPoolLedgerConfig(DEFAULT_POOL_NAME, createPoolLedgerConfigJSONParameter.toJson()).get();. I have a ubuntu VM on the same N/W and i have done indy installation and setup a local pool there. Now from my macbook -> React native android I am trying to connect to Ubuntu VM pool. I have given the Ubuntu IP in the client_ip and node_ip. Now I changed the code to write genesis content to the pool.txn file and pass the GenesisTxnFile.getAbsolutePath() to PoolJSONParameters.CreatePoolLedgerConfigJSONParameter. Output is indy_sdk_genesisTxnFilePath: /data/user/0/com.indyssid/cache/indy/pool12.txn indy_sdk_createPoolLedgerConfigJSONParameter: {"genesis_txn":"/data/user/0/com.indyssid/cache/indy/pool12.txn"} indy_sdk: ================== Some Functionality Failed ================== indy_sdk: org.hyperledger.indy.sdk.pool.PoolLedgerConfigExistsException: A pool ledger configuration already exists with the specified name. What I am doing wrong ?

ashcherbakov (Tue, 10 Sep 2019 07:07:31 GMT):
@cam-parra Thank you for the contribution. Please have a look at my comment. I think much more work is required for Ubuntu 18.04 support than currently in the PR. Please consider starting with a PoA and have a look at the tasks in https://jira.hyperledger.org/browse/INDY-2186

ShrinivasDeshmukh (Tue, 10 Sep 2019 07:08:12 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=Xcxcty7CdDDTDdXQA) @ptab-pawan The error says, pool with the name you have specified in the DEFAULT_POOL_NAME param, already exists. Try changing the DEFAULT_POOL_NAME value.

ptab-pawan (Tue, 10 Sep 2019 09:39:12 GMT):
@ShrinivasDeshmukh : Able to open the pool and get the crypto , anocreds , ledger sample working. But when I check on the ubuntu VM using indy-cli i cant see the pool/wallet etc.

ptab-pawan (Tue, 10 Sep 2019 09:39:12 GMT):
@ShrinivasDeshmukh : Able to open and connect to the pool and get the crypto , anocreds , ledger sample working. But when I check on the ubuntu VM using indy-cli i cant see the pool/wallet created. why ?

ptab-pawan (Tue, 10 Sep 2019 09:39:12 GMT):
@ShrinivasDeshmukh : Able to see the adb logcat od android app that the it has connected to the pool and created wallet etc . Used the crypto , anocreds , ledger samples :https://github.com/hyperledger/indy-sdk/tree/master/samples/java/src/main/java . . But when I check on the ubuntu VM using indy-cli i cant see the pool/wallet created. ool the IP addresses in /var/lib/indy/sandbox/pool_transactions_genesis (in docker) and the pool configuration you use in your mobile app

ptab-pawan (Tue, 10 Sep 2019 09:39:12 GMT):
@ShrinivasDeshmukh : Able to see the adb logcat od android app that the it has connected to the pool and created wallet etc . Used the crypto , anocreds , ledger samples :https://github.com/hyperledger/indy-sdk/tree/master/samples/java/src/main/java . . But when I check on the ubuntu VM using indy-cli i cant see the pool/wallet created. IP addresses in /var/lib/indy/sandbox/pool_transactions_genesis (in docker) and the pool configuration in my mobile app do match.

janb (Thu, 12 Sep 2019 07:37:02 GMT):
Has joined the channel.

janb (Thu, 12 Sep 2019 07:37:03 GMT):
In the current indy implementation it's possible to implement the storage of the wallet - will there be a similar "plugability" for signing? Right now the private keys are all within the software, if you wanted to move them into a HSM will that be made available and if so what are the timelines?

MALodder (Thu, 12 Sep 2019 13:41:49 GMT):
I believe that is a goal for Aries

iainbarclay (Thu, 12 Sep 2019 16:11:31 GMT):
Has joined the channel.

domwoe (Sun, 15 Sep 2019 12:55:03 GMT):
Has joined the channel.

domwoe (Sun, 15 Sep 2019 12:58:32 GMT):
Is there some explanation anywhere why credential schemes and definitions are written to the ledger?

anwen (Sun, 15 Sep 2019 20:34:03 GMT):
Has joined the channel.

anwen (Sun, 15 Sep 2019 20:34:04 GMT):
Hello! I'm trying to find out if the Indy Plenum Ledger is a blockchain or some different type of DLT. Can anyone here provide further information on that?

swcurran (Sun, 15 Sep 2019 22:00:57 GMT):
In the Indy ZKP ecosystem they are necessary components for issuing and proving credentials, they must be globally accessible and immutable. The ledger was a therefore a useful place to them.

doyajii1 (Mon, 16 Sep 2019 12:18:56 GMT):
I think message size doesn't matter since the function takes care of it.

gvuk (Mon, 16 Sep 2019 14:30:23 GMT):
Has joined the channel.

gvuk (Mon, 16 Sep 2019 14:30:25 GMT):
Hey guys, I'm trying to run the sample Hyperledger-indy on my Mac. I had success with it on Windows (I know lol). I am having the issue where it keep throwing: java.lang.NullPointerException: null at org.hyperledger.indy.sdk.pool.Pool.setProtocolVersion(Pool.java:246) ~[indy-1.8.1-dev-1039.jar:na] and it throws it on the line: Pool.setProtocolVersion(PROTOCOL_VERSION).get(); Does this mean my libindy is set-up incorrectly on the Mac?

esplinr (Mon, 16 Sep 2019 21:52:54 GMT):
Indy uses Byzantine Fault Tolerance. See: https://github.com/hyperledger/indy-plenum/

VenkateshSYS (Tue, 17 Sep 2019 06:26:19 GMT):
Hello guys, How to debug the indy application https://github.com/hyperledger/education/tree/master/LFS171x/indy-material/nodejs in google chrome?

VenkateshSYS (Tue, 17 Sep 2019 06:29:02 GMT):
debug

echoharker (Tue, 17 Sep 2019 15:34:47 GMT):
Has joined the channel.

agiledeveloper (Tue, 17 Sep 2019 22:16:37 GMT):
Has joined the channel.

mwherman2000 (Tue, 17 Sep 2019 22:37:09 GMT):
Hi, LTNS, I'm trying to come with a set of trust levels or or trust categories for Digital Identifiers (DID) and I'm wondering if there is an existing HYPE or other document that might help me. At the highest level, is the DID backed by a distributed ledger (or not)? o.e. the Indy ledger. Has anyone seen a categorization/taxonomy/set of trust levels for DIDs? Here's some "back of the envelope" notes I've been working on... Defining Axes of Trust in Something (data/information) or Someone Secure - ? - not easily disambiguated Reliable - ? - available? - not easily disambiguated Historized - lives on a transaction journal or database Auditable - lives on a transaction journal or database, projections = ledgers Verifiable - lives on a transaction journal (preferably, a trusted transaction journal) Permanent - live on a permanent transaction journal Immutable - live on a write-once, read-only, trusted transaction journal, or decentralized transaction journal (e.g. blockchain) Cryptographically Verifiable - lives on a decentralized transaction journal (e.g. blockchain)

mwherman2000 (Tue, 17 Sep 2019 22:37:09 GMT):
Hi, LTNS, I'm trying to come with a set of trust levels or or trust categories for Digital Identifiers (DID) and I'm wondering if there is an existing HYPE or other document that might help me. At the highest level, is the DID backed by a distributed ledger (or not)? o.e. the Indy ledger. Has anyone seen a categorization/taxonomy/set of trust levels for DIDs? Here's some "back of the envelope" notes I've been working on... Defining Axes of Trust in Something (data/information) or Someone Secure - ? - not easily disambiguated Reliable - ? - available? - not easily disambiguated Historized - lives on a transaction journal or database Auditable - lives on a transaction journal or database, projections = ledgers Verifiable - lives on a transaction journal (preferably, a trusted transaction journal) Permanent - live on a permanent transaction journal Immutable - live on a write-once, read-only, trusted transaction journal, or decentralized transaction journal (e.g. blockchain) Cryptographically Verifiable - lives on a decentralized transaction journal (e.g. blockchain) CC: @danielhardman

mwherman2000 (Tue, 17 Sep 2019 22:37:09 GMT):
Hi, LTNS, I'm trying to come up with a set of trust levels or trust categories for Digital Identifiers (DIDs) and I'm wondering if there is an existing HYPE or other document that might help me. At the highest level, for example, is the DID backed by a distributed ledger (or not)? i.e. the Indy ledger. Has anyone seen a categorization/taxonomy/set of trust levels for DIDs? Here's some "back of the envelope" notes I've been working on... Defining Axes of Trust in Something (data/information) or Someone Secure - ? - not easily disambiguated Reliable - ? - available? - not easily disambiguated Historized - lives on a transaction journal or database Auditable - lives on a transaction journal or database, projections = ledgers Verifiable - lives on a transaction journal (preferably, a trusted transaction journal) Permanent - live on a permanent transaction journal Immutable - live on a write-once, read-only, trusted transaction journal, or decentralized transaction journal (e.g. blockchain) Cryptographically Verifiable - lives on a decentralized transaction journal (e.g. blockchain) CC: @danielhardman

mwherman2000 (Tue, 17 Sep 2019 22:37:09 GMT):
Hi, LTNS, I'm trying to come up with a set of trust levels or trust categories for Digital Identifiers (DIDs) and I'm wondering if there is an existing HYPE or other document that might help me. At the highest level, for example, is a DID backed by a distributed ledger (or not)? i.e. the Indy ledger. Has anyone seen a categorization/taxonomy/set of trust levels for DIDs? Here's some "back of the envelope" notes I've been working on... Defining Axes of Trust in Something (data/information) or Someone Secure - ? - not easily disambiguated Reliable - ? - available? - not easily disambiguated Historized - lives on a transaction journal or database Auditable - lives on a transaction journal or database, projections = ledgers Verifiable - lives on a transaction journal (preferably, a trusted transaction journal) Permanent - live on a permanent transaction journal Immutable - live on a write-once, read-only, trusted transaction journal, or decentralized transaction journal (e.g. blockchain) Cryptographically Verifiable - lives on a decentralized transaction journal (e.g. blockchain) CC: @danielhardman

mwherman2000 (Tue, 17 Sep 2019 22:37:09 GMT):
Hi, LTNS, I'm trying to come up with a set of trust levels or trust categories for Digital Identifiers (DIDs) and I'm wondering if there is an existing HYPE or other document that might help me. At the highest level, for example, is a DID backed by a distributed ledger (or not)? i.e. the Indy ledger. Has anyone seen a categorization/taxonomy/set of trust levels for DIDs? Here's some "back of the envelope" notes I've been working on... Defining Axes of Trust in Something (data/information) or Someone [Update: Technically, I'm really asking about *Trust in a Digital Identity defined by it's associated DID and the Credential (set of Claims) associated with the DID*. I'm clarifying this because when your start to research trust, trustworthiness, trust levels, etc., you end up coming across a lot of different articles about the trustworthiness of the content in Wikipedia, for example.] Secure - ? - not easily disambiguated Reliable - ? - available? - not easily disambiguated Historized - lives on a transaction journal or database Auditable - lives on a transaction journal or database, projections = ledgers Verifiable - lives on a transaction journal (preferably, a trusted transaction journal) Permanent - live on a permanent transaction journal Immutable - live on a write-once, read-only, trusted transaction journal, or decentralized transaction journal (e.g. blockchain) Cryptographically Verifiable - lives on a decentralized transaction journal (e.g. blockchain) CC: @danielhardman

anwen (Wed, 18 Sep 2019 08:39:58 GMT):
So it is not an actual blockchain?

dbluhm (Wed, 18 Sep 2019 13:58:06 GMT):
Indy can be used to make a distributed immutable ledger of hash-linked data which by some definitions is a blockchain. It really depends on what definition you're going by since "blockchain" often implies a lot more within some communities.

daidoji (Wed, 18 Sep 2019 14:36:23 GMT):
Is there any SSI ecosystem blog posts or road maps of things that need to exist for SSI that aren't actively being worked on right now?

daidoji (Wed, 18 Sep 2019 14:36:46 GMT):
Like "it would be great if we had this but we don't" kind of posts or lists of tools?

mattatkiva (Wed, 18 Sep 2019 19:28:39 GMT):
for nodejs setLogger command: is there some documentation on the level parameter: 4 means? 5 means what? ```setLogger ( logFn ) Set a function to be called every time a log is emitted from libindy. logFn: Function(Int level, String target, String message, String module_path, String file, Int line) Example: indy.setLogger(function (level, target, message, modulePath, file, line) { console.log('libindy said:', level, target, message, modulePath, file, line) })```

echoharker (Wed, 18 Sep 2019 22:37:23 GMT):
When proving non-revocation of a credential, how does the prover not reveal their factor and the witness to the accumulator? ZKPs?

echoharker (Wed, 18 Sep 2019 22:37:23 GMT):
When proving non-revocation of a credential, how does the prover not reveal their factor and the witness to the verifier? ZKPs?

swcurran (Wed, 18 Sep 2019 23:00:09 GMT):
@echoharker - yes. The best description of the process (apart from digging into the code and academic papers) is here https://github.com/hyperledger/indy-hipe/tree/master/text/0011-cred-revocation

MALodder (Thu, 19 Sep 2019 03:34:38 GMT):
A ZKP provides a pedersen commitment to the provers value in the accumulator, and a pedersen commitment to the witness accumulator that is the accumulator with every value minus the prover's value. These commitments are sent to the verifier who can check the actual accumulator value from the ledger matches. The commitment to the prover's value is also an attribute in his credential. Effectively, the holder is proving that the issuer signed the attribute in the primary credential, it hasn't changed, and is in the non-revocation registry

VenkateshSYS (Thu, 19 Sep 2019 05:27:12 GMT):
Hi Guys, How to access the pool transaction genesis file, it show the path./home/indy/ledger/sandbox/pool_transactions_genesis, but I don’t see it in ls command How can I get it?

andrew.whitehead (Thu, 19 Sep 2019 05:37:29 GMT):
Sounds like you're referring to one in a docker container. You would need to use `docker cp` to copy it out of the container

ap (Thu, 19 Sep 2019 10:11:01 GMT):
Hi all, We are not able to install libindy, any one can you please help out for this

dbluhm (Thu, 19 Sep 2019 13:31:18 GMT):
Anything more you can provide such as the scripts and/or commands you are running and the output of those commands will help diagnose the issue

ap (Thu, 19 Sep 2019 14:24:23 GMT):
Fixed...Thanks for your time

jessie (Thu, 19 Sep 2019 20:20:45 GMT):
Has joined the channel.

esplinr (Thu, 19 Sep 2019 23:06:08 GMT):
Evernym's latest sprint demo video shows the new support for Fully Qualified DIDs in Indy SDK, and a detailed discussion about the architecture of LibIndy, plus some bug fixes: https://www.youtube.com/watch?v=NmSLOmkLBtM&list=PLRp0viTDxBWGLdZk0aamtahB9cpJGV7ZF

sauveergoel (Fri, 20 Sep 2019 05:34:49 GMT):
Hi how can i reduce the size of the wallet in postgres?

VenkateshSorapalli (Fri, 20 Sep 2019 12:03:20 GMT):
when runs the docker-compose file of Hyperledger-indy, the necessary indy-sdk package will be installed from python script. What's the location of the indy-sdk package files?

VenkateshSorapalli (Fri, 20 Sep 2019 12:03:20 GMT):
When runs the docker-compose file of Hyperledger-indy, the necessary indy-sdk package will be installed from python script. What's the location of the indy-sdk package files?

VenkateshSorapalli (Fri, 20 Sep 2019 12:03:20 GMT):
When runs the docker-compose file of Hyperledger-indy, the necessary indy-sdk package will be installed from dockerfile. What's the location of the indy-sdk package files?

VenkateshSorapalli (Fri, 20 Sep 2019 12:03:20 GMT):
When runs the docker-compose file of Hyperledger-indy, the necessary indy-sdk package will be installed from .dockerfile. What's the location of the indy-sdk package files?

donqui (Fri, 20 Sep 2019 12:21:17 GMT):
They should be located here https://repo.sovrin.org/sdk/ (deb https://repo.sovrin.org/sdk/deb xenial master)

VenkateshSorapalli (Fri, 20 Sep 2019 12:24:16 GMT):
How to access the node modules for new js file

VenkateshSorapalli (Fri, 20 Sep 2019 12:28:16 GMT):
If I again npm install to generate node modules for new js file, already created endpoint for the agent via .dockerfile and access the agent endpoint through new node modules differs.. how to get the endpoint for tge agent as same as at the creation time??

VenkateshSorapalli (Fri, 20 Sep 2019 12:28:16 GMT):
If I again npm install to generate node modules for new js file, already created endpoint for the agent via .dockerfile and access the agent endpoint through new node modules differs.. how to get the endpoint for the agent as same as at the creation time??

swcurran (Fri, 20 Sep 2019 18:51:01 GMT):
It's not Postgres specific - will be the same for any wallet. We've done some assessment on our use case, but we'd be interested in what you see. Every time you create a credential, the credential is stored in the `item` table and 2*# of claims in credential+6 encrypted `tags` are automatically created. The six extra is an approximation right now based on our calculations - not looking at code. The claim tags are needed if you want to use search queries to find credentials in your wallet with specific claim names and specific claim values. If you know you don't need to search for credentials, you can set a credential tag policy in Indy that allows you to NOT create some or all of the claim tags. We'll have numbers soon, but we think we'll reduce tag storage by about 75-80%. We found that the tags for the credentials we are using are about 0.5kb each, so depending on the credentials, it can take a lot of storage. We also found that by default, our Aries agent is storing an item for each protocol state record. So, for each issuing of a credential, the credential exchange record is stored completely for the full sequence of calls (offer, request, issue), which is quite a lot of data. We are trying a few things to deal with that: - deleting the protocol records after the protocol completes, where it makes sense for our use case - reducing the data that we are collecting in the protocol record as the state of the protocol instance moves forward. For example, after the Offer-Credential message is processed and the Request-Credential message is sent, the data about the Offer can be reduced/eliminated as we no longer need it. Both of those somewhat depend on the use case (e.g. the need for auditing) so we are looking at ways to make that configurable if it makes sense. Our use case we don't need that, so we are just using methods to minimize storage. @cam-parra - this note might be of interest to you.

VenkateshSYS (Sat, 21 Sep 2019 14:24:14 GMT):
can anybody tell me what are the background process when executes docker-compose build and docker-compose up in hyperledger indy

VenkateshSYS (Sat, 21 Sep 2019 14:24:14 GMT):
can anybody tell me what are the background process when executes docker-compose build and docker-compose up in hyperledger indy?

VenkateshSYS (Sun, 22 Sep 2019 14:58:44 GMT):
How to run custom nodejs express api (get/post) file in running container and how to access it?

VenkateshSYS (Sun, 22 Sep 2019 14:58:44 GMT):
How to run custom nodejs express api (get/post) file in running docker container and how to access it?

VenkateshSYS (Sun, 22 Sep 2019 15:49:00 GMT):
"scripts": { "startfile" : "node file.js" } How to write this in dockerfile CMD []

volume_zero (Sun, 22 Sep 2019 18:43:08 GMT):
Has joined the channel.

PoojaJagtap (Mon, 23 Sep 2019 09:03:56 GMT):
Has joined the channel.

mwherman2000 (Mon, 23 Sep 2019 15:37:25 GMT):
@kdenhartog and co. Hi, I've got a bit of a time-limited "emergency" where I need to get `git clone https://github.com/kdenhartog/indy-dev.git` to build ...and I'm running into an `E: There were unauthenticated packages and -y was used without --allow-unauthenticated` issue: ``` The following packages will be upgraded: python3-setuptools python3-six 2 upgraded, 41 newly installed, 0 to remove and 38 not upgraded. Need to get 43.8 MB of archives. After this operation, 148 MB of additional disk space will be used. WARNING: The following packages cannot be authenticated! libindy-crypto python3-pygments python3-six python3-setuptools libpbc0 python3-charm-crypto python3-base58 indy-anoncreds python3-jsonpickle python3-rlp python3-sha3 python3-ioflo python3-semver python3-orderedset python3-sortedcontainers python3-psutil python3-portalocker python3-pyzmq python3-libnacl python3-intervaltree python3-indy-crypto python3-rocksdb python3-dateutil python3-pympler indy-plenum python3-timeout-decorator indy-node E: There were unauthenticated packages and -y was used without --allow-unauthenticated The command '/bin/sh -c apt-get update -y && apt-get install -y indy-plenum=${indy_plenum_ver} indy-anoncreds=${indy_anoncreds_ver} indy-node=${indy_node_ver} python3-indy-crypto=${python3_indy_crypto_ver} libindy-crypto=${indy_crypto_ver} vim' returned a non-zero code: 100 ``` Any ideas on how to quickly patch this? ...whatever way is simplest. Thank you. ...needed for a conference demo.

mwherman2000 (Mon, 23 Sep 2019 15:37:25 GMT):
@kdenhartog and co. Hi, I've got a bit of a time-limited "emergency" where I need to get `git clone https://github.com/kdenhartog/indy-dev.git` to build ...and I'm running into an `E: There were unauthenticated packages and -y was used without --allow-unauthenticated` issue when `indy_pool` is being built... ``` The following packages will be upgraded: python3-setuptools python3-six 2 upgraded, 41 newly installed, 0 to remove and 38 not upgraded. Need to get 43.8 MB of archives. After this operation, 148 MB of additional disk space will be used. WARNING: The following packages cannot be authenticated! libindy-crypto python3-pygments python3-six python3-setuptools libpbc0 python3-charm-crypto python3-base58 indy-anoncreds python3-jsonpickle python3-rlp python3-sha3 python3-ioflo python3-semver python3-orderedset python3-sortedcontainers python3-psutil python3-portalocker python3-pyzmq python3-libnacl python3-intervaltree python3-indy-crypto python3-rocksdb python3-dateutil python3-pympler indy-plenum python3-timeout-decorator indy-node E: There were unauthenticated packages and -y was used without --allow-unauthenticated The command '/bin/sh -c apt-get update -y && apt-get install -y indy-plenum=${indy_plenum_ver} indy-anoncreds=${indy_anoncreds_ver} indy-node=${indy_node_ver} python3-indy-crypto=${python3_indy_crypto_ver} libindy-crypto=${indy_crypto_ver} vim' returned a non-zero code: 100 ``` Any ideas on how to quickly patch this? ...whatever way is simplest. Thank you. ...needed for a conference demo.

mwherman2000 (Mon, 23 Sep 2019 15:37:25 GMT):
@kdenhartog and co. Hi, I've got a bit of a time-limited "emergency" where I need to get `git clone https://github.com/kdenhartog/indy-dev.git` to build ...and I'm running into an `E: There were unauthenticated packages and -y was used without --allow-unauthenticated` issue when `indy_pool` is being built... ``` The following packages will be upgraded: python3-setuptools python3-six 2 upgraded, 41 newly installed, 0 to remove and 38 not upgraded. Need to get 43.8 MB of archives. After this operation, 148 MB of additional disk space will be used. WARNING: The following packages cannot be authenticated! libindy-crypto python3-pygments python3-six python3-setuptools libpbc0 python3-charm-crypto python3-base58 indy-anoncreds python3-jsonpickle python3-rlp python3-sha3 python3-ioflo python3-semver python3-orderedset python3-sortedcontainers python3-psutil python3-portalocker python3-pyzmq python3-libnacl python3-intervaltree python3-indy-crypto python3-rocksdb python3-dateutil python3-pympler indy-plenum python3-timeout-decorator indy-node E: There were unauthenticated packages and -y was used without --allow-unauthenticated The command '/bin/sh -c apt-get update -y && apt-get install -y indy-plenum=${indy_plenum_ver} indy-anoncreds=${indy_anoncreds_ver} indy-node=${indy_node_ver} python3-indy-crypto=${python3_indy_crypto_ver} libindy-crypto=${indy_crypto_ver} vim' returned a non-zero code: 100 ``` Any ideas on a quick workaround for this? ...whatever way is simplest. Thank you. ...needed for a conference demo.

mwherman2000 (Mon, 23 Sep 2019 15:37:55 GMT):
Here's my script...

mwherman2000 (Mon, 23 Sep 2019 15:37:55 GMT):
Here's my [Windows] script...

mwherman2000 (Mon, 23 Sep 2019 15:39:03 GMT):
``` mkdir c:\INDY c: cd c:\INDY rm indy-dev -r -f git clone https://github.com/kdenhartog/indy-dev.git rem git clone https://github.com/mwherman2000/indy-dev.git echo Press Enter to continue pause cd indy-dev docker build -f indy-pool.dockerfile -t indy_pool . docker build -f indy-dev.dockerfile -t indy_dev . docker images echo Press Enter to continue pause ```

mwherman2000 (Mon, 23 Sep 2019 15:57:09 GMT):
I'm trying `--allow-unauthenticated` on the `apt-get install` command #fingerscrossed

VithushanJegatheeswaran (Tue, 24 Sep 2019 09:40:17 GMT):
Has joined the channel.

VithushanJegatheeswaran (Tue, 24 Sep 2019 09:40:18 GMT):
hey guys i dont understand the documentation for ```def get_wallet_record(wallet_handle: int, type_: str, id: str, options_json: str) -> str:``` ``` Get an wallet record by id :param wallet_handle: wallet handler (created by open_wallet). :param type_: allows to separate different record types collections :param id: the id of record :param options_json: //TODO: FIXME: Think about replacing by bitmask { retrieveType: (optional, false by default) Retrieve record type, retrieveValue: (optional, true by default) Retrieve record value, retrieveTags: (optional, true by default) Retrieve record tags } :return: wallet record json: { id: "Some id", type: "Some type", // present only if retrieveType set to true value: "Some value", // present only if retrieveValue set to true tags: , // present only if retrieveTags set to true } ```

VithushanJegatheeswaran (Tue, 24 Sep 2019 09:40:48 GMT):
what is `type_` and `id`

sukalpomitra (Tue, 24 Sep 2019 09:41:40 GMT):
I guess type is ConnectionRecord or CredentialRecord etc

VithushanJegatheeswaran (Tue, 24 Sep 2019 09:41:44 GMT):
where do i get these information ? I already have `wallet_handle`

sukalpomitra (Tue, 24 Sep 2019 09:41:49 GMT):
id is the primary key

VithushanJegatheeswaran (Tue, 24 Sep 2019 09:42:25 GMT):
primary key of what ? :D sorry I'm new to all of this

VithushanJegatheeswaran (Tue, 24 Sep 2019 09:43:05 GMT):
... and there is no example code for this function :D

sukalpomitra (Tue, 24 Sep 2019 09:45:33 GMT):
for the wallet record..

VithushanJegatheeswaran (Tue, 24 Sep 2019 09:47:27 GMT):
hmm ok .. is there a function which list all the data/content of a wallet ?

VithushanJegatheeswaran (Tue, 24 Sep 2019 09:49:22 GMT):
I just want to check if i recieved a `verifiable claim` or not

VithushanJegatheeswaran (Tue, 24 Sep 2019 09:49:22 GMT):
I just want to check if i received a `verifiable claim` or not

sukalpomitra (Tue, 24 Sep 2019 10:00:11 GMT):
there is... I dont remember at the top of my head currently

VithushanJegatheeswaran (Tue, 24 Sep 2019 11:03:53 GMT):
I found this function ``` async def export_wallet(handle: int, export_config_json: str) -> None: """ Exports opened wallet to the file. ``` Do you mean this ?

sklump (Tue, 24 Sep 2019 11:06:49 GMT):
These are non-secret records - neither credentials nor cred defs nor schemas, etc. You can search non-secret records with empty filters to get all non-secret records (e.g., https://github.com/hyperledger/indy-sdk/blob/aa6da11c5385a79e038cec4942b37b25883da914/wrappers/python/tests/non_secrets/test_wallet_search.py#L46). But this search won't find credentials, only non-secret records. You want https://github.com/hyperledger/indy-sdk/blob/aa6da11c5385a79e038cec4942b37b25883da914/wrappers/python/tests/anoncreds/test_prover_search_credentials.py#L24.

sklump (Tue, 24 Sep 2019 11:06:49 GMT):
These are non-secret records - neither credentials nor cred defs nor schemas, etc. You can search non-secret records with empty filters to get all non-secret records (e.g., https://github.com/hyperledger/indy-sdk/blob/aa6da11c5385a79e038cec4942b37b25883da914/wrappers/python/tests/non_secrets/test_wallet_search.py#L46). But this search won't find credentials, only non-secret records. You want `prover_search_credentials()`, `prover_fetch_credentials()`, `prover_close_credentials_search()` as per https://github.com/hyperledger/indy-sdk/blob/aa6da11c5385a79e038cec4942b37b25883da914/wrappers/python/tests/anoncreds/test_prover_search_credentials.py#L24.

nitika-goel (Tue, 24 Sep 2019 13:55:53 GMT):
Hi, we are working on a Hyperledger Indy project. Have followed the demo on https://github.com/hyperledger/education/tree/master/LFS171x/indy-material/nodejs. The demo works perfectly but runs the application on different ports for different users. Is it possible that the agents run on different ports but the application can run one a single port. On login, it can connect to the agent port required.

nitika-goel (Tue, 24 Sep 2019 13:55:53 GMT):
Hi, we are working on a Hyperledger Indy project. Have followed the demo on https://github.com/hyperledger/education/tree/master/LFS171x/indy-material/nodejs. The demo works perfectly but runs the application on different ports for different users. Is it possible that the agents run on different ports but the application can run on a single port. On login, it can connect to the agent port required.

nitika-goel (Tue, 24 Sep 2019 13:56:40 GMT):
Can somebody please help me.

ezamp (Tue, 24 Sep 2019 19:25:49 GMT):
Has joined the channel.

ezamp (Tue, 24 Sep 2019 19:25:49 GMT):
Hi, I'm working on a project based on Hyperledger Indy. Actually I'm trying to import the indy framework into a new project on Xcode 11.0. I'm using cocapods but i have some problems with Podfile and 'libindy-objc' version. Can somebody help me?

grossetti (Tue, 24 Sep 2019 20:54:49 GMT):
Has joined the channel.

grossetti (Tue, 24 Sep 2019 20:54:51 GMT):
Hi everyone

grossetti (Tue, 24 Sep 2019 20:54:52 GMT):
I am trying to understand the difference in between Indy and Iroha in regards to Identity

grossetti (Tue, 24 Sep 2019 21:01:08 GMT):
Can someone please explain the differences? I was not able to find any comparsons

grossetti (Tue, 24 Sep 2019 21:01:08 GMT):
Can someone please explain the differences? I was not able to find any comparisons

VenkateshSYS (Wed, 25 Sep 2019 06:04:23 GMT):
Can anybody tell me why this error occurs while executing this line? await sdk.createPairwise(await indy.wallet.get(), theirDid, myDid, meta); >> INDY error 212. what happens in the background of this code line?

jakubkoci (Wed, 25 Sep 2019 06:51:22 GMT):
This error code means WalletItemNotFound. If I understand the Rust code in `create_pairwise`, the wallet tries to find dids and that's what probably throws an error. Have you created theses dids by `createAndStoreMyDid` or similar before calling createPairwise?

jakubkoci (Wed, 25 Sep 2019 06:51:22 GMT):
This error code means WalletItemNotFound. If I understand the Rust code in `create_pairwise`, the wallet tries to find dids in the wallet and that's what probably throws an error. Have you created theses dids by `createAndStoreMyDid` or similar before calling createPairwise?

jakubkoci (Wed, 25 Sep 2019 07:04:37 GMT):
Discussion moved to indy-sdk :)

VithushanJegatheeswaran (Wed, 25 Sep 2019 08:08:38 GMT):
`Python wrapper` ``` print(alice['pool']) print(alice['did_for_Company']) print(authdecrypted_license_cred_offer['cred_def_id']) ``` OUTPUT: ``` 2 HPSerat64pwMrvZAqpcwGr TcZchwR5EFFK212gQAu17E:3:CL:TcZchwR5EFFK212gQAu17E:2:License:1.2:TAG1 ``` ``` print("\"Alice\" -> Get \"Company License\" Credential Definition from Ledger") (alice['Company_license_cred_def_id'], alice['Company_license_cred_def']) = await get_cred_def(alice['pool'], alice['did_for_Company'],alice['license_cred_def_id']) ``` OUTPUT ``` "Alice" -> Get "Company License" Credential Definition from Ledger --------------------------------------------------------------------------- CommonInvalidStructure Traceback (most recent call last) in async-def-wrapper() in get_cred_def(pool_handle, _did, cred_def_id) 106 107 async def get_cred_def(pool_handle, _did, cred_def_id): --> 108 get_cred_def_request = await ledger.build_get_cred_def_request(_did, cred_def_id) 109 get_cred_def_response = await ledger.submit_request(pool_handle, get_cred_def_request) 110 return await ledger.parse_get_cred_def_response(get_cred_def_response) /usr/local/lib/python3.5/dist-packages/indy/ledger.py in build_get_cred_def_request(submitter_did, id_) 605 c_submitter_did, 606 c_id, --> 607 build_get_cred_def_request.cb) 608 609 res = request_json.decode() /usr/lib/python3.5/asyncio/futures.py in __iter__(self) 359 if not self.done(): 360 self._blocking = True --> 361 yield self # This tells Task to wait for completion. 362 assert self.done(), "yield from wasn't used with future" 363 return self.result() # May raise too. /usr/lib/python3.5/asyncio/tasks.py in _wakeup(self, future) 294 def _wakeup(self, future): 295 try: --> 296 future.result() 297 except Exception as exc: 298 # This may also be a cancellation. /usr/lib/python3.5/asyncio/futures.py in result(self) 272 self._tb_logger = None 273 if self._exception is not None: --> 274 raise self._exception 275 return self._result 276 CommonInvalidStructure: ```

VithushanJegatheeswaran (Wed, 25 Sep 2019 08:10:38 GMT):
How to resolve this `CommonInvalidStructure` Error ? What is the issue here ? I'm desperate :confounded:

VithushanJegatheeswaran (Wed, 25 Sep 2019 08:12:53 GMT):
``` async def get_cred_def(pool_handle, _did, cred_def_id): get_cred_def_request = await ledger.build_get_cred_def_request(_did, cred_def_id) get_cred_def_response = await ledger.submit_request(pool_handle, get_cred_def_request) return await ledger.parse_get_cred_def_response(get_cred_def_response) ```

VithushanJegatheeswaran (Wed, 25 Sep 2019 08:14:57 GMT):
Here is the Method from [getting-started.ipub](https://github.com/hyperledger/indy-sdk/blob/master/docs/getting-started/getting-started.ipynb)

dbluhm (Wed, 25 Sep 2019 13:38:29 GMT):
What version of libindy are you using? Can you enable logs and share the output? You should be able to enable logs by adding the following to your python script: ``` import logging logging.getLogger('indy').setLevel(logging.DEBUG) ```

mattatkiva (Wed, 25 Sep 2019 13:59:50 GMT):
https://github.com/hyperledger/indy-sdk/pull/1897

VithushanJegatheeswaran (Wed, 25 Sep 2019 14:56:40 GMT):
OMG first of all thank you for this tip :clap: ``` DEBUG:indy.ledger:build_get_cred_def_request: >>> submitter_did: '6qubttPZawxFegkxvKfhc9', id: 'GodJAa4B8MuU1qspG5UTs2:3:CL:GodJAa4B8MuU1qspG5UTs2:2:License:1.2:TAG1' DEBUG:indy.libindy:do_call: >>> name: indy_build_get_cred_def_request, args: (c_char_p(140693269640872), c_char_p(140693263741736), ) INFO:indy.libindy.native.indy.commands: src/commands/mod.rs:136 | LedgerCommand command received DEBUG:indy.libindy:do_call: Function indy_build_get_cred_def_request returned err: 0 INFO:indy.libindy.native.ledger_command_executor: src/commands/ledger.rs:374 | BuildGetCredDefRequest command received DEBUG:indy.libindy:do_call: <<< DEBUG:indy.libindy.native.indy.commands.ledger: src/commands/ledger.rs:797 | build_get_cred_def_request >>> submitter_did: Some("6qubttPZawxFegkxvKfhc9"), id: "GodJAa4B8MuU1qspG5UTs2:3:CL:GodJAa4B8MuU1qspG5UTs2:2:License:1.2:TAG1" INFO:indy.libindy.native.indy.services.ledger: src/services/ledger/mod.rs:162 | build_get_cred_def_request() => Err(IndyError { inner: ParseIntError { kind: InvalidDigit } Schema ID is invalid number in: GodJAa4B8MuU1qspG5UTs2:3:CL:GodJAa4B8MuU1qspG5UTs2:2:License:1.2:TAG1 Invalid structure }) DEBUG:indy.libindy:_get_error_details: >>> DEBUG:indy.libindy:_get_error_details: <<< error_details: {'backtrace': '', 'message': 'Error: Invalid structure\n Caused by: Schema ID is invalid number in: GodJAa4B8MuU1qspG5UTs2:3:CL:GodJAa4B8MuU1qspG5UTs2:2:License:1.2:TAG1\n Caused by: invalid digit found in string\n'} DEBUG:indy.libindy:_indy_callback: >>> command_handle: 65, err , args: (b'',) DEBUG:indy.libindy:_indy_callback: <<< DEBUG:indy.libindy:_indy_loop_callback: >>> command_handle: 65, err , args: (b'',) WARNING:indy.libindy:_indy_loop_callback: Function returned error DEBUG:indy.libindy:_indy_loop_callback <<< ``` It seams like the `Schema ID` is invalid :/

VithushanJegatheeswaran (Wed, 25 Sep 2019 14:56:40 GMT):
OMG first of all thank you for this answer :clap: ``` DEBUG:indy.ledger:build_get_cred_def_request: >>> submitter_did: '6qubttPZawxFegkxvKfhc9', id: 'GodJAa4B8MuU1qspG5UTs2:3:CL:GodJAa4B8MuU1qspG5UTs2:2:License:1.2:TAG1' DEBUG:indy.libindy:do_call: >>> name: indy_build_get_cred_def_request, args: (c_char_p(140693269640872), c_char_p(140693263741736), ) INFO:indy.libindy.native.indy.commands: src/commands/mod.rs:136 | LedgerCommand command received DEBUG:indy.libindy:do_call: Function indy_build_get_cred_def_request returned err: 0 INFO:indy.libindy.native.ledger_command_executor: src/commands/ledger.rs:374 | BuildGetCredDefRequest command received DEBUG:indy.libindy:do_call: <<< DEBUG:indy.libindy.native.indy.commands.ledger: src/commands/ledger.rs:797 | build_get_cred_def_request >>> submitter_did: Some("6qubttPZawxFegkxvKfhc9"), id: "GodJAa4B8MuU1qspG5UTs2:3:CL:GodJAa4B8MuU1qspG5UTs2:2:License:1.2:TAG1" INFO:indy.libindy.native.indy.services.ledger: src/services/ledger/mod.rs:162 | build_get_cred_def_request() => Err(IndyError { inner: ParseIntError { kind: InvalidDigit } Schema ID is invalid number in: GodJAa4B8MuU1qspG5UTs2:3:CL:GodJAa4B8MuU1qspG5UTs2:2:License:1.2:TAG1 Invalid structure }) DEBUG:indy.libindy:_get_error_details: >>> DEBUG:indy.libindy:_get_error_details: <<< error_details: {'backtrace': '', 'message': 'Error: Invalid structure\n Caused by: Schema ID is invalid number in: GodJAa4B8MuU1qspG5UTs2:3:CL:GodJAa4B8MuU1qspG5UTs2:2:License:1.2:TAG1\n Caused by: invalid digit found in string\n'} DEBUG:indy.libindy:_indy_callback: >>> command_handle: 65, err , args: (b'',) DEBUG:indy.libindy:_indy_callback: <<< DEBUG:indy.libindy:_indy_loop_callback: >>> command_handle: 65, err , args: (b'',) WARNING:indy.libindy:_indy_loop_callback: Function returned error DEBUG:indy.libindy:_indy_loop_callback <<< ``` It seams like the `Schema ID` is invalid :/

sklump (Wed, 25 Sep 2019 15:12:03 GMT):
Did you write the schema to the ledger, then read it back to get the sequence number before creating the cred def? If not, the cred def id will be in long format (`issuer-did:3:origin-did:schema-name:schema-version:cred-def-tag`) rather than in short format (`issuer-did:3:schema-seq-no:cred-def-tag`). Perhaps some processing further down the line assumes the more typical short format, but I the long format is also licit, accommodated here: https://github.com/hyperledger/indy-sdk/blob/da4da126c5fb81fa2bb2d6b8215374a8e53ce73a/libindy/src/commands/anoncreds/issuer.rs#L353. To get back the cred def id short format on cred def write, be sure to request and parse the schema from the ledger before creating a cred def on its transaction number.

sklump (Wed, 25 Sep 2019 15:12:03 GMT):
Did you read the schema from the ledger, then parse it get the sequence number before creating the cred def? If not, the cred def id will be in long format (`issuer-did:3:origin-did:schema-name:schema-version:cred-def-tag`) rather than in short format (`issuer-did:3:schema-seq-no:cred-def-tag`). Perhaps some processing further down the line assumes the more typical short format, but I the long format is also licit, accommodated here: https://github.com/hyperledger/indy-sdk/blob/da4da126c5fb81fa2bb2d6b8215374a8e53ce73a/libindy/src/commands/anoncreds/issuer.rs#L353. To get back the cred def id short format on cred def write, be sure to request and parse the schema from the ledger before creating a cred def on its transaction number.

sklump (Wed, 25 Sep 2019 15:12:03 GMT):
Did you read the schema from the ledger, then parse it get the sequence number before creating the cred def? If not, the cred def id will be in long format (`issuer-did:3:origin-did:schema-name:schema-version:cred-def-tag`) rather than in short format (`issuer-did:3:schema-seq-no:cred-def-tag`). Perhaps some processing further down the line assumes the more typical short format, but the long format is also licit, accommodated here: https://github.com/hyperledger/indy-sdk/blob/da4da126c5fb81fa2bb2d6b8215374a8e53ce73a/libindy/src/commands/anoncreds/issuer.rs#L353. To get back the cred def id short format on cred def write, be sure to request and parse the schema from the ledger before creating a cred def on its transaction number.

sklump (Wed, 25 Sep 2019 15:12:03 GMT):
Did you read the schema from the ledger, then parse it get the sequence number before creating the cred def? If not, the cred def id will be in long format (`issuer-did:3:origin-did:schema-name:schema-version:cred-def-tag`) rather than in short format (`issuer-did:3:schema-seq-no:cred-def-tag`). Perhaps some processing further down the line assumes the more typical short format, but the long format is also licit, accommodated here: https://github.com/hyperledger/indy-sdk/blob/da4da126c5fb81fa2bb2d6b8215374a8e53ce73a/libindy/src/commands/anoncreds/issuer.rs#L353. To get back the cred def id short format on cred def write, be sure to request and parse the schema from the ledger before creating a cred def on its transaction number. Whichever format you get, you are stuck with it forever. Once it's written, it's written.

sklump (Wed, 25 Sep 2019 15:12:03 GMT):
Did you read the schema from the ledger, then parse it get the sequence number before creating the cred def? If not, the cred def id will be in long format (`issuer-did:3:origin-did:schema-name:schema-version:cred-def-tag`) rather than in short format (`issuer-did:3:schema-seq-no:cred-def-tag`). Perhaps some processing further down the line assumes the more typical short format, but the long format is also licit, accommodated here: https://github.com/hyperledger/indy-sdk/blob/da4da126c5fb81fa2bb2d6b8215374a8e53ce73a/libindy/src/commands/anoncreds/issuer.rs#L353. To get back the cred def id short format on cred def write, be sure to request and parse the schema from the ledger before creating a cred def on its transaction number. Whichever format you get, you are stuck with it forever. Once it's written, it's written. This is a good thing, since otherwise credential searches would be ambiguous and buggy on cred def id.

lynn.bendixsen (Wed, 25 Sep 2019 16:03:14 GMT):
I recently worked through some similar issues to what you are mentioning @sklump. What you are saying is a bit confusing to me, so maybe its not related, but I thought I would mention my experience since it is at least close. More details available if needed, but we had to make sure that "protocol_version" was set to 2 to get the "long version" of the cred_def_id, otherwise we would get the short version. I am pretty sure the long version should be used for all recent versions of Indy (since moving to protocol_version 2 over a year ago). The problem we had was that "protocol_version" is not automatically set. To set it we had to either call ~"set_protocol_version" or connecting to a pool that had "protocol_version=2" would somehow set the version globally for us for future sdk calls. https://github.com/hyperledger/indy-sdk/blob/master/libindy/src/domain/anoncreds/credential_definition.rs#L123

sklump (Wed, 25 Sep 2019 16:24:43 GMT):
The protocol version refers to the ledger protocol. The code doesn't lie: the indy `IssuerCommandExecutor._prepare_create_and_store_credential_definition()` call tries to parse the schema seq number, and if it can't, substitutes the schema id.

lynn.bendixsen (Wed, 25 Sep 2019 16:51:04 GMT):
Sounds like my issue was something different, sorry for the noise.

sklump (Wed, 25 Sep 2019 16:58:37 GMT):
It's good to read others' experience in dealing with toolkit intricacies.

sklump (Wed, 25 Sep 2019 16:58:37 GMT):
It's good to read others' experience in dealing with toolkit intricacies. Ledger protocol version 2 did introduce the trailing tag to the cred def id.

sklump (Wed, 25 Sep 2019 16:58:37 GMT):
It's good to read others' experience in dealing with toolkit intricacies. @lynn.bendixsen - Ledger protocol version 2 did introduce the trailing tag to the cred def id, which made it longer than it had been by a colon and the tag.

VithushanJegatheeswaran (Thu, 26 Sep 2019 09:49:52 GMT):
`Did you read the schema from the ledger, then parse it get the sequence number before creating the cred def?` This was the cause of my error. THANK YOU SO MUCH I created a Schema like this and got `license_schema_id and license_schema` ``` (Company['license_schema_id'], Company['license_schema']) = \ await anoncreds.issuer_create_schema(Company['did'], license['name'], license['version'], json.dumps(license['attributes'])) ``` Then sent the Schema to Ledger ``` await send_schema(Company['pool'], Company['wallet'], Company['did'], Company['license_schema']) time.sleep(1) ``` My mistake was that I skipped the next step where I got the `license_schema_id and license_schema` from the ledger. I thought that I don't need to get these information from the ledger, since I already have them. Here is the difference: ``` Before: {"ver":"1.0","id":"DNRackiEMapJPxTvzGPWLW:2:License:1.2","name":"License","version":"1.2","attrNames":["ssn","status","degree","last_name","first_name","average","year"],"seqNo":null} ``` ``` After: {"ver":"1.0","id":"DNRackiEMapJPxTvzGPWLW:2:License:1.2","name":"License","version":"1.2","attrNames":["year","first_name","ssn","degree","average","status","last_name"],"seqNo":19} ``` The `seqNo` was set to `null` and caused the problem :D This one Method solved the problem: ``` (Company['license_schema_id'], Company['license_schema']) = \ await get_schema(Company['pool'], Company['did'], Company['license_schema_id']) ``` I overwrote/updated `license_schema` and got the missing `seqNo` P.S. English is not my mother tongue; please excuse any errors on my part.

sklump (Thu, 26 Sep 2019 10:45:28 GMT):
The schema can't have a sequence number until it is written to the ledger, which is obvious when you think about it - but only stepping into the trap makes one think about it. Your English is fine, carry on and good luck.

sklump (Thu, 26 Sep 2019 10:45:28 GMT):
The schema can't have a sequence number until it is written to the ledger, which is obvious when you think about it - but only stepping into the trap makes one think about it. The von anchor code base (https://github.com/PSPC-SPAC-buyandsell/von_anchor/tree/master/von_anchor) implements approaches for dealing with indy-sdk and avoiding subtle traps like this. Your English is fine, carry on and good luck.

VithushanJegatheeswaran (Thu, 26 Sep 2019 14:16:05 GMT):
I finally issued a `verifiable claim` :man_dancing: :man_dancing: :man_dancing: ``` (search_handle, total_count) = await prover_search_credentials(alice['wallet'],'{}') print('search_handle:',search_handle, 'total_count:',total_count) credentials = json.loads( await prover_fetch_credentials(search_handle, total_count)) pp.pprint(credentials[0]) ``` ``` search_handle: 61 total_count: 1 { 'attrs': { 'average': '5', 'degree': 'Bachelor of Science, Marketing', 'first_name': 'Alice', 'last_name': 'Garcia', 'ssn': '123-45-6789', 'status': 'graduated', 'year': '2015'}, 'cred_def_id': 'DNRackiEMapJPxTvzGPWLW:3:CL:19:TAG1', 'cred_rev_id': None, 'referent': '5348515c-fdb3-4fec-9dc6-b30e0aa1f98d', 'rev_reg_id': None, 'schema_id': 'DNRackiEMapJPxTvzGPWLW:2:License:1.2'} ``` What is the purpose of `referent` ? Is it getting used to verify this claim ?

sklump (Thu, 26 Sep 2019 14:37:24 GMT):
The referent is just a handle in the wallet uniquely identifying the credential.

lawkim (Thu, 26 Sep 2019 14:42:45 GMT):
Has joined the channel.

Henrycoffin (Thu, 26 Sep 2019 14:49:22 GMT):
Hi everyone. I have a very general question surrounding the initial connection between an identity holder and an Issuer. Is there anything within Indy to help prove that the person trying to connect is in fact who they say they are? Interested to hear how others are dealing with this issue.

VithushanJegatheeswaran (Thu, 26 Sep 2019 14:51:58 GMT):
I think ... the holder encrypts the message with his private key and with the issuers public key

VithushanJegatheeswaran (Thu, 26 Sep 2019 14:53:27 GMT):
then the issuer can decrypt the message using the holders public key --> that proves that the message is really coming from the holder

Henrycoffin (Thu, 26 Sep 2019 15:02:41 GMT):
Does that not just prove that the message is coming form the person claiming to be the holder? I understand how everything works once that pairwise connection has been established. It's just the implementation of that initial connection that I need to get my head around. (mainly so I can explain it to business people). If the answer is that we would have to follow any existing KYC process then that is obviously fine. I was just wondering if Indy provided anything that would prevent a bad actor from being able to create a connection and be issued with a credential? I appreciate this is probably a real edge case and I struggle to think of scenarios where this might happen but, I need to provide assurances to the business.

sklump (Thu, 26 Sep 2019 15:03:27 GMT):
But the (putative) holder's private key (and DID) are not public, and can be anything. So can the (putative) issuer's. Only when the holder verifies the credential offer does the holder have any confidence that the issuer controls the private key as anchored on the ledger. At least this is as far as I understand at this time.

sklump (Thu, 26 Sep 2019 15:03:27 GMT):
But the (putative) holder's key (and DID) are not on the ledger, and can be anything. So can the (putative) issuer's. Only when the holder verifies the credential offer does the holder have any confidence that the issuer controls the private key as anchored on the ledger. At least this is as far as I understand at this time.

sklump (Thu, 26 Sep 2019 15:03:27 GMT):
But the (putative) holder's key (and DID) are not on the ledger, and can be anything. So can the (putative) issuer's. Only when the holder verifies the credential offer does the holder have any confidence that the issuer controls the private key as anchored on the ledger. At least this is as far as I understand at this time. As far as the issuer knowing the holder's identity, off-line processes have to vet. E.g., present proof of foundational identity.

sklump (Thu, 26 Sep 2019 15:03:27 GMT):
But the (putative) holder's key (and DID) are not on the ledger, and can be anything. So can the (putative) issuer's pairwise DID and key pair for the interaction. Only when the holder verifies the credential offer does the holder have any confidence that the issuer controls the private key as anchored on the ledger. At least this is as far as I understand at this time. As far as the issuer knowing the holder's identity, off-line processes have to vet. E.g., present proof of foundational identity.

VithushanJegatheeswaran (Thu, 26 Sep 2019 15:03:34 GMT):
only the holder will know his private key

dbluhm (Thu, 26 Sep 2019 15:04:26 GMT):
Identity WG Implementers call happening now: https://zoom.us/j/232861185

fhmarino (Thu, 26 Sep 2019 16:22:17 GMT):
Hi guys! I'm just trying to create an aca-py using the following command

fhmarino (Thu, 26 Sep 2019 16:28:24 GMT):
Hi guys! I'm just trying to create an aca-py with the following command: aca-py start --inbound-transport http 0.0.0.0 8100 \ --outbound-transport http \ --log-level debug \ --endpoint http://localhost:8100 \ --label back-office \ --seed 10d38d0ea70ca428b98d95ce6888151a \ --genesis-url http://localhost:9000/genesis \ --ledger-pool-name localindypool \ --wallet-key 123456 --wallet-name backofficewallet --wallet-type indy \ --admin 0.0.0.0 8150 --admin-insecure-mode \ --public-invites --auto-accept-invites --auto-accept-requests --auto-ping-connection --auto-respond-messages \ --auto-respond-credential-offer --auto-respond-presentation-request --auto-verify-presentation --auto-store-credential \ --auto-respond-presentation-proposal --auto-respond-credential-request --auto-respond-credential-proposal \ --debug-connections --debug-credentials --debug-presentations But rising this error: aries_cloudagent.conductor ERROR Unable to start administration API ... AttributeError: 'MarshmallowPlugin' object has no attribute 'openapi'

amundson (Thu, 26 Sep 2019 17:16:33 GMT):
Has left the channel.

swcurran (Thu, 26 Sep 2019 20:00:34 GMT):
@andrew.whitehead @esune ^^^^

swcurran (Thu, 26 Sep 2019 20:02:06 GMT):
This question was answered on the #aries channel.

esplinr (Thu, 26 Sep 2019 22:35:46 GMT):
This week's update on Indy from Evernym: * The September releases, * Updated PyZMQ, * Progress on improved view change algorithm https://www.youtube.com/watch?v=qFXSWCA1GIg&list=PLRp0viTDxBWGLdZk0aamtahB9cpJGV7ZF&index=1

VithushanJegatheeswaran (Fri, 27 Sep 2019 10:29:39 GMT):
How can I connect to a remote Indy pool ? for example: I want to create a indy pool on a Server and want to develop on my local machiene is that possible ?

VithushanJegatheeswaran (Fri, 27 Sep 2019 10:29:39 GMT):
How can I connect to a remote Indy pool ? for example: I want to create a indy pool on a remote Server and want to develop on my local machiene is that possible ?

VithushanJegatheeswaran (Fri, 27 Sep 2019 10:31:54 GMT):
This is how i currently create and connect to a pool, which is running localy on my machine ``` await pool.set_protocol_version(2) pool_ = { 'name': 'pool1', 'config': json.dumps({"genesis_txn": '/home/indy/sandbox/pool_transactions_genesis'}) } print("Open Pool Ledger: {}".format(pool_['name'])) try: await pool.create_pool_ledger_config(pool_['name'], pool_['config']) except IndyError as ex: if ex.error_code == ErrorCode.PoolLedgerConfigAlreadyExistsError: pass pool_['handle'] = await pool.open_pool_ledger(pool_['name'], None) ```

VithushanJegatheeswaran (Fri, 27 Sep 2019 10:34:20 GMT):
this code snippet is from https://github.com/hyperledger/indy-sdk/blob/master/docs/getting-started/getting-started.ipynb

sukalpomitra (Fri, 27 Sep 2019 14:37:24 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=eN7PgiHrqxY2xLJHF) @VithushanJegatheeswaran You just need to use the genesis file of the ledger you wanna connect to

VithushanJegatheeswaran (Fri, 27 Sep 2019 14:49:43 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=3989eca2-4220-409c-ac64-3a853a8a6333) ohhhh thank you

liormargalit (Sat, 28 Sep 2019 19:03:24 GMT):
Has joined the channel.

liormargalit (Sat, 28 Sep 2019 19:03:27 GMT):
Hi, running the command "sudo docker build -f ci/indy-pool.dockerfile -t indy_pool ." result in the following error: The following packages have unmet dependencies: indy-plenum : Depends: python3-pyzmq (= 17.0.0) but 18.1.0 is to be installed

VenkateshSYS (Sun, 29 Sep 2019 19:17:50 GMT):
Hey guys! Can anybody tell why this error comes when running the node file which is depending on indy-sdk.. ```Error: Could not locate the bindings file. Tried: → /home/ubuntu/**************/node_modules/indy-sdk/build/indynodejs.node → /home/ubuntu/********************/node_modules/indy-sdk/build/Debug/indynodejs.node → /home/ubuntu/********************/node_modules/indy-sdk/build/Release/indynodejs.node → /home/ubuntu/********************/node_modules/indy-sdk/out/Debug/indynodejs.node → /home/ubuntu/********************/node_modules/indy-sdk/Debug/indynodejs.node → /home/ubuntu/********************/node_modules/indy-sdk/out/Release/indynodejs.node → /home/ubuntu/********************/node_modules/indy-sdk/Release/indynodejs.node → /home/ubuntu/********************/node_modules/indy-sdk/build/default/indynodejs.node → /home/ubuntu/********************/node_modules/indy-sdk/compiled/10.16.3/linux/x64/indynodejs.node → /home/ubuntu/********************/node_modules/indy-sdk/addon-build/release/install-root/indynodejs.node → /home/ubuntu/********************/node_modules/indy-sdk/addon-build/debug/install-root/indynodejs.node → /home/ubuntu/********************/node_modules/indy-sdk/addon-build/default/install-root/indynodejs.node → /home/ubuntu/********************/node_modules/indy-sdk/lib/binding/node-v64-linux-x64/indynodejs.node at bindings (/home/ubuntu/********************/node_modules/bindings/bindings.js:126:9) at Object. (/home/ubuntu/********************/node_modules/indy-sdk/src/index.js:3:31) at Module._compile (internal/modules/cjs/loader.js:778:30) at Object.Module._extensions..js (internal/modules/cjs/loader.js:789:10) at Module.load (internal/modules/cjs/loader.js:653:32) at tryModuleLoad (internal/modules/cjs/loader.js:593:12) at Function.Module._load (internal/modules/cjs/loader.js:585:3) at Module.require (internal/modules/cjs/loader.js:692:17) at require (internal/modules/cjs/helpers.js:25:18) at Object. (/home/ubuntu/********************/indy/src/wallet/index.js:2:13) ```

Abhishekkishor (Mon, 30 Sep 2019 03:45:11 GMT):
Has joined the channel.

RohitShitre (Mon, 30 Sep 2019 06:52:11 GMT):
Has joined the channel.

VithushanJegatheeswaran (Mon, 30 Sep 2019 07:48:57 GMT):
Hey I never had this issue but I would try to remove '/home/ubuntu/********************/node_modules' and reinstall all modules using `npm install`

vikrumd (Mon, 30 Sep 2019 08:15:13 GMT):
Has joined the channel.

agiledeveloper (Mon, 30 Sep 2019 12:25:06 GMT):
Hi Experts; Can we create a credential offer from the CLI ?

agiledeveloper (Mon, 30 Sep 2019 12:25:06 GMT):
Hi Experts; Can we create a credential offer from the CLI ? @all

agiledeveloper (Mon, 30 Sep 2019 12:25:06 GMT):
Hi Experts; Can we create a credential offer from the CLI ?

Henrycoffin (Mon, 30 Sep 2019 13:14:59 GMT):
Should I be using Indy or Aries for verifiable credentials? Indy is saying most of the functionality has been moved to aries but, aries is mostly saying that it is not production ready yet?????

MALodder (Mon, 30 Sep 2019 14:21:20 GMT):
@Henrycoffin for now, use Indy

Henrycoffin (Mon, 30 Sep 2019 14:39:06 GMT):
Thanks

rkreutzer (Mon, 30 Sep 2019 16:33:33 GMT):
Has joined the channel.

manavgupta (Tue, 01 Oct 2019 04:10:20 GMT):
Hi Everyone, Can I create a custom genesis file for the indy pool?

aguel (Tue, 01 Oct 2019 05:23:18 GMT):
I have the same issue, anyone?

bansatya (Tue, 01 Oct 2019 07:08:29 GMT):
Has joined the channel.

bansatya (Tue, 01 Oct 2019 07:08:31 GMT):
HI, running Indy node pool using docker which start 4 nodes successfully(log shows all are syncing each other). However while connecting from Indy-cli, the pool is not getting connected. Enabled the log for the client but log shows nothing Only "Pool "local-pool" has not been connected" message is displayed.

bansatya (Tue, 01 Oct 2019 07:08:31 GMT):
HI, running Indy node pool using docker which start 4 nodes successfully(log shows all are syncing each other). However while connecting from Indy-cli, the pool is not getting connected. Enabled the log for the client but log shows nothing Only "local-pool" has not been connected" message is displayed.

donqui (Tue, 01 Oct 2019 12:29:31 GMT):
pyzmq package was update and now apt is complaining because he does not know how to resolve the dependency. If you add `python3-pyzmq=17.0.0` to the apt command in the dockerfile it should help

vbhagg (Tue, 01 Oct 2019 14:28:10 GMT):
Has joined the channel.

liormargalit (Tue, 01 Oct 2019 15:58:24 GMT):
It was fixed in https://jira.hyperledger.org/browse/IS-1382

lynn.bendixsen (Tue, 01 Oct 2019 18:03:54 GMT):
Yes, for your own pools you can use the following to create the genesis files. If you are wondering about creating a custom genesis file for an existing pool, the no, that won't work. https://github.com/sovrin-foundation/steward-tools/tree/master/create_genesis

lynn.bendixsen (Tue, 01 Oct 2019 18:03:54 GMT):
Yes, for your own pools you can use the following to create the genesis files. If you are wondering about creating a custom genesis file for an existing pool, then no, that won't work. https://github.com/sovrin-foundation/steward-tools/tree/master/create_genesis

lynn.bendixsen (Tue, 01 Oct 2019 18:09:45 GMT):
Hi @bansatya, There are a few reasons why the pool might not connect. The first thing I would look at is your genesis file. Check to make sure that the ip addresses and ports listed in there are correct and that your firewall is configured correctly.

acaldas18 (Tue, 01 Oct 2019 22:53:32 GMT):
Has joined the channel.

agiledeveloper (Wed, 02 Oct 2019 00:43:32 GMT):
Second Try, Can we create a credential offer from the CLI ?

eduelias (Wed, 02 Oct 2019 11:21:54 GMT):
Has joined the channel.

swcurran (Wed, 02 Oct 2019 14:36:50 GMT):
@agiledeveloper no you cannot. You need at minimum a small script.

swcurran (Wed, 02 Oct 2019 14:38:00 GMT):
I'm traveling and can't send a script, but suggest looking at the Indy-sdk Python test scripts.

agiledeveloper (Wed, 02 Oct 2019 14:59:08 GMT):
thanks, I am looking at the java wrapper for the SDK. pretty straight forward...thanks alot

habermas (Thu, 03 Oct 2019 00:53:29 GMT):
Has joined the channel.

habermas (Thu, 03 Oct 2019 00:53:31 GMT):
hello, I am studying indy for hours but I ended up being totally lost :( All I am trying to do is a simple website in php/html that will utilize indy to provide identities (DIDs)... I am searching for a simple example but I can not find anything... could anyone point me to any good tutorial or give me any advice?

habermas (Thu, 03 Oct 2019 00:54:48 GMT):
Just the most basic implementation of website using indy to provide IDs, but I am feeling helpess with the official documentation...

manavgupta (Thu, 03 Oct 2019 07:08:20 GMT):
Thanks it would be helpful for me

MALodder (Thu, 03 Oct 2019 14:01:51 GMT):
@agiledeveloper No

dalesupa22 (Thu, 03 Oct 2019 19:58:09 GMT):
Has joined the channel.

anwen (Fri, 04 Oct 2019 10:53:22 GMT):
Where can I find a quick overview of what devices and technical environments indy is compatible with?

cgeene (Fri, 04 Oct 2019 17:52:54 GMT):
Has joined the channel.

tomislav (Sat, 05 Oct 2019 01:52:28 GMT):
@anwen You can compile it to pretty much any architecture. Any specific runtimes you're looking into?

agiledeveloper (Sat, 05 Oct 2019 14:02:15 GMT):
Any suggestions where to check ? `java.util.concurrent.ExecutionException: org.hyperledger.indy.sdk.anoncreds.ProofRejectedException: The proof has been rejected. error. `

anwen (Sun, 06 Oct 2019 20:46:51 GMT):
is there a publicly shared distributed ledger for everyone, or a seperate ledger for each application?

andrew.whitehead (Sun, 06 Oct 2019 23:31:36 GMT):
There’s the Sovrin MainNet ledger as well as a couple development ledgers for testing deployments (BuilderNet and StagingNet), or you can install a test ledger locally

esplinr (Mon, 07 Oct 2019 15:04:07 GMT):
Starting Indy Contributors coordination meeting: https://zoom.us/j/615818107

gaila (Tue, 08 Oct 2019 12:42:37 GMT):
Has joined the channel.

VithushanJegatheeswaran (Tue, 08 Oct 2019 13:00:53 GMT):
Hey guys I'm stuck again I want to run this cmmand: ``` alice['job_application_proof'] = \ await anoncreds.prover_create_proof(alice['wallet'], alice['job_application_proof_request'], alice['job_application_requested_creds'], alice['master_secret_id'], alice['schemas'], alice['cred_defs'], alice['revoc_states']) ``` This is the debug message: `_indy_loop_callback: Function returned error ` This is the error message: ```--------------------------------------------------------------------------- CommonInvalidStructure Traceback (most recent call last) in async-def-wrapper() 34 35 print("\"Alice\" -> Authcrypt \"Job-Application\" Proof for TBSP") ---> 36 alice['authcrypted_job_application_proof'] = \ 37 await crypto.auth_crypt(alice['wallet'], alice['key_for_tbsp'], alice['tbsp_key_for_alice'], 38 alice['job_application_proof'].encode('utf-8')) /usr/local/lib/python3.5/dist-packages/indy/anoncreds.py in prover_create_proof(wallet_handle, proof_req_json, requested_credentials_json, master_secret_name, schemas_json, credential_defs_json, rev_states_json) 1410 c_credential_defs_json, 1411 c_rev_infos_json, -> 1412 prover_create_proof.cb) 1413 1414 res = proof_json.decode() /usr/lib/python3.5/asyncio/futures.py in __iter__(self) 359 if not self.done(): 360 self._blocking = True --> 361 yield self # This tells Task to wait for completion. 362 assert self.done(), "yield from wasn't used with future" 363 return self.result() # May raise too. /usr/lib/python3.5/asyncio/tasks.py in _wakeup(self, future) 294 def _wakeup(self, future): 295 try: --> 296 future.result() 297 except Exception as exc: 298 # This may also be a cancellation. /usr/lib/python3.5/asyncio/futures.py in result(self) 272 self._tb_logger = None 273 if self._exception is not None: --> 274 raise self._exception 275 return self._result 276 CommonInvalidStructure: ```

VithushanJegatheeswaran (Tue, 08 Oct 2019 13:03:13 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=HG993C8yJykLWaosF)
VithushanJegatheeswaran - Tue Oct 08 2019 15:02:56 GMT+0200 (Central European Summer Time).txt

manavgupta (Wed, 09 Oct 2019 06:28:48 GMT):
Hi everyone I am getting an error org.hyperledger.indy.sdk.InvalidStateException: The SDK library experienced an unexpected internal error while running the command Pool.openPoolLedger. Can anyone tell me the solution for that??

manavgupta (Wed, 09 Oct 2019 09:40:41 GMT):
Hi, I have created the custom pool genesis now when i am connecting the indy pool with client it gives an error:- org.hyperledger.indy.sdk.InvalidStateException: The SDK library experienced an unexpected internal error Can you help me solve this error

Koptan (Wed, 09 Oct 2019 12:03:00 GMT):
Hi , i have very basic question , let say I am patient and I want to share my medical data with doctor , So the question is , 1- Can i put duration for the sharing of data like 7 days or 14 days ? 2- Can i NOT allowing the doctor or hospital from saving my data on their system ?

Koptan (Wed, 09 Oct 2019 12:03:00 GMT):
Hi , i have very basic question , let say I am patient and I want to share my medical data with doctor , So the question is , 1- Can i put duration for the sharing of data like 7 days or 14 days ? 2- Can i NOT allowing the doctor or hospital from saving my data on their system after i shared my medical data ?

DibbsZA (Wed, 09 Oct 2019 12:29:52 GMT):
As I understand the question, the answer is NO to both. Once data is shared, it is shared and I don't see a way for data to magically dissapear on expiry. Whether the data is legally *CONSIDERED VALID* & *CONSENTED* is a different matter, and one that can only be resolved in the offline world (courts)

DibbsZA (Wed, 09 Oct 2019 12:33:02 GMT):
Also I think there would be legal rights of the doctor to retain that information if they had rendered any medical advice or services based on the information.

VithushanJegatheeswaran (Wed, 09 Oct 2019 14:55:33 GMT):
hey guys do I have to create a "Sovrin Steward" user ?

VithushanJegatheeswaran (Wed, 09 Oct 2019 14:55:33 GMT):
hey guys do I have to create a "Sovrin Steward" user ? if yes.. why ? what is the reason for that ?

kevinvanderpoll (Wed, 09 Oct 2019 15:40:54 GMT):
Has joined the channel.

kevinvanderpoll (Wed, 09 Oct 2019 15:40:55 GMT):
Hello All! Can someone share me some resources to learn the INDY CLI.

cam-parra (Wed, 09 Oct 2019 15:43:48 GMT):
@kevinvanderpoll the indy cli has built in documentation. Just type help as a command and it will help you started

kevinvanderpoll (Wed, 09 Oct 2019 15:47:32 GMT):
Thanks @cam-parra. I'm surely gonna have a few questions, thanks for getting back to me.

lynn.bendixsen (Wed, 09 Oct 2019 15:52:14 GMT):
You are getting in pretty deep now :) I would need to have several more details to be able to make suggestions. To begin with: Where are your nodes hosted? (locally?, multiple remote machines?, Docker?) Are the nodes in your genesis file able to communicate with each other (run `sudo validator-info` from each nodes command prompt) ? If you can ping (or netcat) all of your nodes IP addresses/ports from the client using the values in the genesis file, then post your genesis file here and I can have a look to see if I see any obvious problems.

lynn.bendixsen (Wed, 09 Oct 2019 15:52:14 GMT):
You are getting in pretty deep now :) I would need to have several more details to be able to make suggestions. To begin with: Where are your nodes hosted? (locally?, multiple remote machines?, Docker?) Are the nodes in your genesis file able to communicate with each other (run `sudo validator-info` from each nodes command prompt)? If you can ping (or netcat) all of your nodes IP addresses/ports from the client using the values in the genesis file, then post your genesis file here and I can have a look to see if I see any obvious problems.

lynn.bendixsen (Wed, 09 Oct 2019 16:09:11 GMT):
No, creating a "Sovrin Steward" DID (user) is not required to be able to use a Sovrin network. Creating a Sovrin Steward user is only required for Sovrin Stewards. This is because each Steward is required to host a validator node in one of the Sovrin Networks. In the past, the steward role could do more than just host a node, but with recent changes, the steward role does not have any other privileges on a Sovrin network. Generically, each node in a default indy network must be added by a different steward DID so you will need at least one per node.

Ladar (Wed, 09 Oct 2019 18:18:44 GMT):
Is there a Chat Room for the IOS build. I think the Podfile needs to be updated on the repository end.

Ladar (Wed, 09 Oct 2019 18:19:37 GMT):
The Podfile is linking to https://repo.sovrin.org/ios/libindy/stable/indy-objc/1.10.0/libindy-objc.zip which is not built currently

manavgupta (Thu, 10 Oct 2019 04:14:23 GMT):
I am running all the nodes on single remote machine. The nodes are communicating with each other and I am able to ping IPaddress and port from the client as well. When I run the default scripts of indy node on the same remote machine the client gets connected to indy pool easily but in custom pool genesis it does not. Here is the genesis which i built

manavgupta (Thu, 10 Oct 2019 04:16:02 GMT):

domain_transactions_genesis.txt

manavgupta (Thu, 10 Oct 2019 04:16:04 GMT):

pool_transactions_genesis.txt

manavgupta (Thu, 10 Oct 2019 04:24:24 GMT):
Here is what i get after running sudo validator-info

manavgupta (Thu, 10 Oct 2019 04:24:44 GMT):

sudo-validator-info.txt

FarhanShafiq (Thu, 10 Oct 2019 07:51:13 GMT):
Can someone help me understand why pairwise DID are important? What if i only use one DID for all connection?

sukalpomitra (Thu, 10 Oct 2019 07:53:40 GMT):
so that your data is not co-related later

FarhanShafiq (Thu, 10 Oct 2019 08:01:22 GMT):
Thanks but i didn't understand co-related?

sukalpomitra (Thu, 10 Oct 2019 08:03:20 GMT):
well if u use the same identifier everywhere, u can get data from diff sites and co-relate this data to create a persona. For example u interact with Bank B with identifier 1 and Bank B also with identifier 1. Then someone can co-relate and find out you have loan from Bank A and indurance from Bank B. Hope I am making sense

FarhanShafiq (Thu, 10 Oct 2019 08:07:27 GMT):
Thank you so much, i want to know how to find co-relate data. Is there any link or document about it?

sukalpomitra (Thu, 10 Oct 2019 08:08:13 GMT):
well you can google about exactis or cambridge analytica. In netflix, there is a great documentary about it called "The Great Hack"

FarhanShafiq (Thu, 10 Oct 2019 08:13:43 GMT):
Thank you so much :)

FarhanShafiq (Thu, 10 Oct 2019 08:58:36 GMT):
How is it possible to manage all pairwise DIDs for user and what will be the recovery process?

esplinr (Thu, 10 Oct 2019 15:07:46 GMT):
Identity Implementers call is starting: https://zoom.us/j/232861185

cam-parra (Thu, 10 Oct 2019 15:43:03 GMT):
@esplinr @MALodder Hey could this be run https://github.com/hyperledger/indy-plenum/pull/1369 on the pipeline? This will enable plenum to use ursa instead of crypto

lynn.bendixsen (Thu, 10 Oct 2019 16:07:28 GMT):
Thanks for the details. The files look like they are correct. A couple of possible issues I see are 1. The validator-info outputs show that all nodes are stopped. Were they running when you saw the error?

lynn.bendixsen (Thu, 10 Oct 2019 16:09:26 GMT):
2. One thing you said made me wonder if the files you sent match the similar files on your nodes. I am pretty sure that the nodes pool_transaction_genesis file needs to match what you are using in the client for it to work

lynn.bendixsen (Thu, 10 Oct 2019 16:25:37 GMT):
3. Another issue you might be having is that the es that are in your pool file need to be the Public addresses (I have seen people mistakenly use private IP's here)

Fosol (Thu, 10 Oct 2019 22:50:07 GMT):
Has joined the channel.

Fosol (Thu, 10 Oct 2019 22:50:07 GMT):
Hi, I have a general question about management of Wallets. Currently they are only accessible via an Agent (Cloud or Edge). They are possibly encrypted through the owner's key-phrase. What if the Agent becomes unavailable? Is there a way to regenerate a Wallet from nothing? Is there a built in way to share/copy/move a Wallet from one Agent to another?

dbluhm (Thu, 10 Oct 2019 23:38:50 GMT):
You bring up several related but distinct topics: identity recovery, wallet backup, synchronizing wallets across multiple agents, and wallet portability. Each topic is fairly deep and the solutions for them are in varying stages of implementation. Plans and designs for identity recovery can be found in the [DKMS RFC](https://github.com/hyperledger/aries-rfcs/blob/master/concepts/0051-dkms/README.md). [A fairly detailed discussion on wallet backup and portability](https://chat.hyperledger.org/channel/aries?msg=8Lno4JQa2MQKNALCt) recently occurred in the #aries channel. Wallet synchronization is something we know we want but specific implementations have not yet been made. We anticipate potentially relying on the services something like a DIF Identity Hub. For more conceptual details on how we think that could happen, you might find this [article on the similarities and differences of DIF Identity Hubs and Agents](https://medium.com/decentralized-identity/rhythm-and-melody-how-hubs-and-agents-rock-together-ac2dd6bf8cf4) interesting.

dbluhm (Thu, 10 Oct 2019 23:49:10 GMT):
The discussion began at the linked message and continued in the main thread for a while

manavgupta (Fri, 11 Oct 2019 03:38:10 GMT):
The nodes are running on a separate remote machine at the time of error. The pool genesis file is same at node and client i have checked that too earlier only And the IP address which i am using is not private as well

manavgupta (Fri, 11 Oct 2019 03:41:52 GMT):
When i start the nodes by default process of indy node on that machine and tries to connect client from different machine then the client gets connected but in the case of custom genesis it doses not

lynn.bendixsen (Fri, 11 Oct 2019 03:44:19 GMT):
Looks like you have covered all of the common cases... Maybe send a copy of the pool genesis file that works for comparison? Sorry, but i am running out of suggestions.

manavgupta (Fri, 11 Oct 2019 03:44:31 GMT):
And one thing in the second point you said the file should match with the client, which means the genesis transaction we give in pool config ledger?

lynn.bendixsen (Fri, 11 Oct 2019 03:48:04 GMT):
The pool genesis file you use for client connection, should match the one you had on the nodes when you started the network pool. (/var/lib/indy//pool_transactions_genesis)

manavgupta (Fri, 11 Oct 2019 03:48:06 GMT):
I just give the genesis transaction same as given in Pool genesis transaction while i run the command Pool.createPoolLedgerConfig

manavgupta (Fri, 11 Oct 2019 03:49:15 GMT):
let me send you the client file as well

manavgupta (Fri, 11 Oct 2019 03:50:34 GMT):

clientPool.txt

manavgupta (Fri, 11 Oct 2019 03:54:38 GMT):
Is there anything I am missing in client Pool genesis?

lynn.bendixsen (Fri, 11 Oct 2019 03:55:46 GMT):
At a quick glance, that looks like it matches the pool genesis you sent earlier. I will look at it a little bit more. Can you send a copy of the pool file that worked? (the one that was used when you started the pool "by a default process" as you mentioned earlier)

manavgupta (Fri, 11 Oct 2019 03:57:43 GMT):

poolDefault.txt

manavgupta (Fri, 11 Oct 2019 03:58:17 GMT):
here it is

lynn.bendixsen (Fri, 11 Oct 2019 04:04:09 GMT):
Hmmm, looks like the working one has different node names and keys. The quickest way for me to help further might be to get on a zoom chat and have a quick chat with you while you share your screen and we look at a few files on one of your nodes. Is that's something you would like to do right now?

manavgupta (Fri, 11 Oct 2019 04:05:39 GMT):
yeah sure can you send me the link of zoom chat

lynn.bendixsen (Fri, 11 Oct 2019 04:06:05 GMT):
https://zoom.us/j/9953953572

manavgupta (Fri, 11 Oct 2019 04:06:06 GMT):
so i can join the call in that

lynn.bendixsen (Fri, 11 Oct 2019 04:39:06 GMT):
@mgbailey @sergey.khoroshavin I got on a zoom chat with @manavgupta but was still unable to find the issue he is having. Can you have a look at this thread and see if you can help? Manav is running 4 nodes on the same remote server with a custom genesis file. (Works with the default genesis file using different nodes on the same server, but the custom genesis file fails) Genesis files and other details are included above.

VithushanJegatheeswaran (Fri, 11 Oct 2019 09:29:41 GMT):
thank you :)

VithushanJegatheeswaran (Fri, 11 Oct 2019 09:30:08 GMT):
``` { "ver": "1.0", "id": "XrArmxdbKfXSVfkirzJP1N:3:CL:19:TAG1", "schemaId": "19", "type": "CL", "tag": "TAG1", "value": { "primary": { "n": "1099372606.....", "s": "6357232250...", "r": { "status": "81671543509947563268464806064120713570033515792669824775266654299913867077811631688710421197846814917413236588843529482167046336086789659711351380359035039937553698009739619574934558906195887910468454505996188701385969508762232548422378508097381334275730680099894635264263072355781645557815484547651534159500435113276221779613434850314579292762452096047380738039807518329983137400709255040903372845066701545719406159633551087247647633967928236280285518385874942381190525986140763734833635906354852867035092022040774639004109570120770440303559389956472635882529692778225903313706907217765272548766819383582427657488034", "last_name": "107917864017833261055140547062250579150609821457593855985535857848981045820852513685499193575930081841996055936874640335263958291492834056128124702962086867299081156099579942647625735599780742168218329619688105936530629884024432891019051127536583155582904307953108978583403697691336355369313800752461661979257525059292657778064035578716529060341343407414523030025767993884487161600429944543259761181573943365530811139153869451370920953012699444285821221450479476872503441362711747691326302185103604249599964504329396031461800133437276218697710485271238490794630625382544529032960634922504058823888814426393456799324193", "master_secret": "7237009419008290381057263977321197143877579325205355626664313050364977353032138100783449010090447318672069968680749274194346940642600205231830969628149404133537973741075783579229185480684515287444870514126445128345944706944013232190034001608976067091351664011120203637640271384550217696064163611168403593703142862657281897315685109028127142877389087568548958483013423241183264111865696913495368919447461708257843193518492847517048146247827300633777714858145473988542630651000568643399332800747100568398131405250687570859978266398858386237483800756896551458444903339156170491834217516642296149488218343472628958108786", "first_name": "67448034366297581295079732212688140034572986256064482669473757103003990953676763141645233147891151359756486327960296778408293075948303464283649441549026041051219777698122339896027186383131907506646276745470473188174301766951268622441178576019867113260755016268591010705521021892426980087879624560943449520607986526239950666929786403620284053575739104289091211921980028387844774630643678017653350134504676622622707697174967304876660914856780682985236635729669012338897730472077363674790625766146208042086149402597889956060362265572797200314650115121385013692490648719121342277367520609685814529826664405077367511528706" }, "rctxt": "1031562977...", "z": "7233198996...." } } } ```

VithushanJegatheeswaran (Fri, 11 Oct 2019 09:30:08 GMT):
``` { "ver": "1.0", "id": "XrArmxdbKfXSVfkirzJP1N:3:CL:19:TAG1", "schemaId": "19", "type": "CL", "tag": "TAG1", "value": { "primary": { "n": "1099372606.....", "s": "6357232250...", "r": { "status": "81671543509947563268464806064120713570033515792669824775266654299913867077811631688710421197846814917413236588843529482167046336086789659711351380359035039937553698009739619574934558906195887910468454505996188701385969508762232548422378508097381334275730680099894635264263072355781645557815484547651534159500435113276221779613434850314579292762452096047380738039807518329983137400709255040903372845066701545719406159633551087247647633967928236280285518385874942381190525986140763734833635906354852867035092022040774639004109570120770440303559389956472635882529692778225903313706907217765272548766819383582427657488034", "last_name": "107917864017833261055140547062250579150609821457593855985535857848981045820852513685499193575930081841996055936874640335263958291492834056128124702962086867299081156099579942647625735599780742168218329619688105936530629884024432891019051127536583155582904307953108978583403697691336355369313800752461661979257525059292657778064035578716529060341343407414523030025767993884487161600429944543259761181573943365530811139153869451370920953012699444285821221450479476872503441362711747691326302185103604249599964504329396031461800133437276218697710485271238490794630625382544529032960634922504058823888814426393456799324193", "master_secret": "7237009419008290381057263977321197143877579325205355626664313050364977353032138100783449010090447318672069968680749274194346940642600205231830969628149404133537973741075783579229185480684515287444870514126445128345944706944013232190034001608976067091351664011120203637640271384550217696064163611168403593703142862657281897315685109028127142877389087568548958483013423241183264111865696913495368919447461708257843193518492847517048146247827300633777714858145473988542630651000568643399332800747100568398131405250687570859978266398858386237483800756896551458444903339156170491834217516642296149488218343472628958108786", "first_name": "67448034366297581295079732212688140034572986256064482669473757103003990953676763141645233147891151359756486327960296778408293075948303464283649441549026041051219777698122339896027186383131907506646276745470473188174301766951268622441178576019867113260755016268591010705521021892426980087879624560943449520607986526239950666929786403620284053575739104289091211921980028387844774630643678017653350134504676622622707697174967304876660914856780682985236635729669012338897730472077363674790625766146208042086149402597889956060362265572797200314650115121385013692490648719121342277367520609685814529826664405077367511528706" }, "rctxt": "1031562977...", "z": "7233198996...." } } } ```

VithushanJegatheeswaran (Fri, 11 Oct 2019 09:30:08 GMT):
``` { "ver": "1.0", "id": "XrArmxdbKfXSVfkirzJP1N:3:CL:19:TAG1", "schemaId": "19", "type": "CL", "tag": "TAG1", "value": { "primary": { "n": "1099372606.....", "s": "6357232250...", "r": { "status": "8167154350...", "last_name": "1079178640...", "master_secret": "7237009419...", "first_name": "6744803436..." }, "rctxt": "1031562977...", "z": "7233198996...." } } } ```

VithushanJegatheeswaran (Fri, 11 Oct 2019 09:30:08 GMT):
This is my generated credential definition: ``` { "ver": "1.0", "id": "XrArmxdbKfXSVfkirzJP1N:3:CL:19:TAG1", "schemaId": "19", "type": "CL", "tag": "TAG1", "value": { "primary": { "n": "1099372606.....", "s": "6357232250...", "r": { "status": "8167154350...", "last_name": "1079178640...", "master_secret": "7237009419...", "first_name": "6744803436..." }, "rctxt": "1031562977...", "z": "7233198996...." } } } ``` What do these keys (`n, s, r, rctxt, z`) stand for ?

VithushanJegatheeswaran (Fri, 11 Oct 2019 11:12:41 GMT):
can a issuere use the same credential definition multiple times ?

esplinr (Fri, 11 Oct 2019 15:19:17 GMT):
Thanks Cam. This is an exciting PR! cc @ashcherbakov

esplinr (Fri, 11 Oct 2019 15:20:32 GMT):
Evernym's regular update on our work with Indy and Aries: https://www.youtube.com/watch?v=dh0gY1xowKg&list=PLRp0viTDxBWGLdZk0aamtahB9cpJGV7ZF

Fosol (Fri, 11 Oct 2019 18:22:49 GMT):
Is my understanding of the public ledger correct? - Does not store any private data - Is decentralized and therefore many instances - Must be able to sync with other Trusted instances - Issuers are the only ones who write to the public ledger

Fosol (Fri, 11 Oct 2019 18:22:49 GMT):
Is my understanding of the public ledger correct? - Does not store any private data - Is decentralized and therefore many instances - Must be able to sync with other Trusted instances - Issuers are the only ones who write to the public ledger - Agents read from the public ledger to update their own information which allows them to verify Credentials

sebastillar (Sun, 13 Oct 2019 14:16:44 GMT):
Has joined the channel.

sebastillar (Sun, 13 Oct 2019 14:16:46 GMT):
Hi everyone! I am starting to work on an SSI project to obtain my university degree So, knowing about Sovrin, I would like to confirm if I need to develop the blockchain under Indy architecture or just integrate to Sovrin and develop the agents/wallets

sebastillar (Sun, 13 Oct 2019 14:18:18 GMT):
I really appreciate any suggest or recommendation to start to implement :slight_smile:

liormargalit (Sun, 13 Oct 2019 19:15:11 GMT):
Hi, a question about https://github.com/hyperledger/indy-sdk/blob/master/samples/python/src/getting_started.py

liormargalit (Sun, 13 Oct 2019 19:15:19 GMT):
logger.info("\"Alice\" -> Send authcrypted \"Job-Certificate\" Credential Request to Acme") alice['job_certificate_cred_values'] = json.dumps({ "first_name": {"raw": "Alice", "encoded": "245712572474217942457235975012103335"}, "last_name": {"raw": "Garcia", "encoded": "312643218496194691632153761283356127"}, "employee_status": {"raw": "Permanent", "encoded": "2143135425425143112321314321"}, "salary": {"raw": "2400", "encoded": "2400"}, "experience": {"raw": "10", "encoded": "10"} }) acme['authcrypted_job_certificate_cred_request'] = alice['authcrypted_job_certificate_cred_request'] acme['job_certificate_cred_values'] = alice['job_certificate_cred_values']

liormargalit (Sun, 13 Oct 2019 19:15:38 GMT):
It does not make sense, why would alice provide the values to acme?

yunho.chung (Mon, 14 Oct 2019 01:10:05 GMT):
Has joined the channel.

yunho.chung (Mon, 14 Oct 2019 01:16:47 GMT):
Hi! Can I get the info regarding the difference between SSI and DID?

DibbsZA (Mon, 14 Oct 2019 07:03:38 GMT):
Perhaps this document will help you clarify your requirement and thus where you could focus your efforts. https://github.com/hyperledger/aries-cloudagent-python/blob/master/docs/GettingStartedAriesDev/IndyAriesDevOptions.md

DibbsZA (Mon, 14 Oct 2019 07:05:45 GMT):
This doc and it subsequent follow ups may help set the context. https://github.com/hyperledger/aries-cloudagent-python/blob/master/docs/GettingStartedAriesDev/IndyBasics.md

gazelle (Mon, 14 Oct 2019 08:36:56 GMT):
Has joined the channel.

manavgupta (Mon, 14 Oct 2019 10:26:52 GMT):
Hi @lynn.bendixsen Any updates on that Issue?

sebastillar (Mon, 14 Oct 2019 10:33:26 GMT):
thanks for answer!

benjamin.verhaegen (Mon, 14 Oct 2019 10:34:41 GMT):
Has joined the channel.

sebastillar (Mon, 14 Oct 2019 10:35:02 GMT):
I've read this docs but I'm not clear about in which scenario is better for build my own ledger or if a better solution goes to Sovrin ledger

DibbsZA (Mon, 14 Oct 2019 11:05:16 GMT):
Well I don't know. To me, building another blockchain before you can build solutions that people will use seems like extra work. :grin: Then again, if your university studies is all about the details of blockchain, crypto and distributed systems, then it may not be extra effort.

ThanawanSachdev (Mon, 14 Oct 2019 13:37:49 GMT):
Has joined the channel.

ThanawanSachdev (Mon, 14 Oct 2019 13:37:57 GMT):
Hi,

cam-parra (Mon, 14 Oct 2019 15:11:42 GMT):
I would highly advise to only maintain your own network if you have expertise with Indy. It's not your average blockchain and it can be quite hard to wrestle with when things go wrong (not often but nothing is perfect). @lynn.bendixsen is the Sovrin Networks maintainer and has a deeper knowledge than I do about maintaining a production network. I am sure if you ask nicely he can give you some pointers

snogueir (Mon, 14 Oct 2019 17:35:44 GMT):
Has joined the channel.

snogueir (Mon, 14 Oct 2019 17:35:45 GMT):
Hi all, does anyone know if it is possible to deploy indy nodes in a Centos/Redhat distribution? thanks,

lynn.bendixsen (Mon, 14 Oct 2019 18:39:45 GMT):
none yet, the people I pinged have not responded :(

haydarmajeed (Mon, 14 Oct 2019 18:45:01 GMT):
Has joined the channel.

pranavkirtani88 (Tue, 15 Oct 2019 11:42:44 GMT):
Has joined the channel.

pranavkirtani88 (Tue, 15 Oct 2019 11:42:46 GMT):
Hi, any good source material around credential schemas and schema definitions?

Fosol (Tue, 15 Oct 2019 15:58:04 GMT):
Where can I read up on how the ledger is distributed and remains in sync between instances?

swcurran (Tue, 15 Oct 2019 16:11:35 GMT):
That is the indy-node repo, and the #indy-node channel in rocketchat.

sergey.minaev (Tue, 15 Oct 2019 18:35:20 GMT):
I suggest to gather logs from libindy and then we can analyze it

manavgupta (Wed, 16 Oct 2019 05:39:48 GMT):
which logs are you talking about?

donqui (Wed, 16 Oct 2019 12:00:20 GMT):
not at this moment

lynn.bendixsen (Wed, 16 Oct 2019 13:12:40 GMT):
Logs needed from all nodes are `/var/log/indy/sandbox/.log` ```sudo journalctl -u indy-node sudo journalctl -u indy-node-control```

sergey.minaev (Wed, 16 Oct 2019 17:52:01 GMT):
@manavgupta https://github.com/hyperledger/indy-sdk/tree/master/wrappers/java#logging

sergey.minaev (Wed, 16 Oct 2019 17:52:01 GMT):
@manavgupta https://github.com/hyperledger/indy-sdk/tree/master/wrappers/java#logging if you are using java wrapper

snogueir (Wed, 16 Oct 2019 22:51:22 GMT):
and is it possible to deploy a docker container with a pool of 4 nodes and another container with indy-sdk working as a client of the pool and connecting using the name of the container instead of a fixed ip? I'm trying this and I facing some problems with the connectivity

manavgupta (Thu, 17 Oct 2019 07:06:20 GMT):

Gamma.log

manavgupta (Thu, 17 Oct 2019 07:06:46 GMT):

Beta.log

manavgupta (Thu, 17 Oct 2019 07:06:46 GMT):

Theta.log

manavgupta (Thu, 17 Oct 2019 07:06:52 GMT):

Alpha.log

manavgupta (Thu, 17 Oct 2019 07:06:53 GMT):
Logs of all the nodes

manavgupta (Thu, 17 Oct 2019 12:45:34 GMT):
@sergey.minaev you can check the logs

WoodenSheep (Thu, 17 Oct 2019 16:37:17 GMT):
Has joined the channel.

VithushanJegatheeswaran (Fri, 18 Oct 2019 09:57:12 GMT):
``` Validator Node3 is in unknown state Update time: Friday, October 18, 2019 9:51:04 AM +0000 Validator DID: DKVxG2fXXTU8yT5N7hGEbXB3dfdAnYv1JczDUHpmDxya Verification Key: 5LCf9oRXkDBNG61PpSsBnWDLhyCZWzYYoHAJSwusoSLZCsAre9a5BGL BLS Key: 3WFpdbg7C5cnLYZwFZevJqhubkFALBfCBBok15GdrKMUhUjGsk3jV6QKj6MZgEubF7oqCafxNdkm7eswgA4sdKTRc82tLGzZBd6vNqU8dupzup6uYUf32KTHTPQbuUM8Yk4QFXjEf2Usu2TJcNkdgpyeUSX42u5LqdDDpNSWUK5deC5 Node HA: 0.0.0.0:9705 Client HA: 0.0.0.0:9706 Metrics: Uptime: 1 hour, 6 minutes, 56 seconds Total audit Transactions: 39 Total ledger Transactions: 12 Total pool Transactions: 4 Total config Transactions: 0 Read Transactions/Seconds: 0.00 Write Transactions/Seconds: 0.00 Reachable Hosts: 4/4 Node1 (0) Node2 (1) Node4 Node3 Unreachable Hosts: 0/4 Software Versions: indy-node: 1.9.2.dev1061 sovrin: unknown Validator Node2 is in unknown state Update time: Friday, October 18, 2019 9:51:04 AM +0000 Validator DID: 8ECVSk179mjsjKRLWiQtssMLgp6EPhWXtaYyStWPSGAb Verification Key: 72XEVLaj3htbTYyNXeJWXJCj3ffeJHTuYLtV4PdU1n72YsCzo65aBZ6 BLS Key: 37rAPpXVoxzKhz7d9gkUe52XuXryuLXoM6P6LbWDB7LSbG62Lsb33sfG7zqS8TK1MXwuCHj1FKNzVpsnafmqLG1vXN88rt38mNFs9TENzm4QHdBzsvCuoBnPH7rpYYDo9DZNJePaDvRvqJKByCabubJz3XXKbEeshzpz4Ma5QYpJqjk Node HA: 0.0.0.0:9703 Client HA: 0.0.0.0:9704 Metrics: Uptime: 1 hour, 6 minutes, 56 seconds Total audit Transactions: 39 Total pool Transactions: 4 Total ledger Transactions: 12 Total config Transactions: 0 Read Transactions/Seconds: 0.00 Write Transactions/Seconds: 0.00 Reachable Hosts: 4/4 Node1 (0) Node2 (1) Node4 Node3 Unreachable Hosts: 0/4 Software Versions: indy-node: 1.9.2.dev1061 sovrin: unknown Validator Node1 is in unknown state Update time: Friday, October 18, 2019 9:51:04 AM +0000 Validator DID: Gw6pDLhcBcoQesN72qfotTgFa7cbuqZpkX3Xo6pLhPhv Verification Key: 33nHHYKnqmtGAVfZZGoP8hpeExeH45Fo8cKmd5mcnKYk7XgWNBxkkKJ BLS Key: 4N8aUNHSgjQVgkpm8nhNEfDf6txHznoYREg9kirmJrkivgL4oSEimFF6nsQ6M41QvhM2Z33nves5vfSn9n1UwNFJBYtWVnHYMATn76vLuL3zU88KyeAYcHfsih3He6UHcXDxcaecHVz6jhCYz1P2UZn2bDVruL5wXpehgBfBaLKm3Ba Node HA: 0.0.0.0:9701 Client HA: 0.0.0.0:9702 Metrics: Uptime: 1 hour, 6 minutes, 56 seconds Total audit Transactions: 39 Total ledger Transactions: 12 Total pool Transactions: 4 Total config Transactions: 0 Read Transactions/Seconds: 0.00 Write Transactions/Seconds: 0.00 Reachable Hosts: 4/4 Node1 (0) Node2 (1) Node4 Node3 Unreachable Hosts: 0/4 Software Versions: indy-node: 1.9.2.dev1061 sovrin: unknown Validator Node4 is in unknown state Update time: Friday, October 18, 2019 9:51:04 AM +0000 Validator DID: 4PS3EDQ3dW1tci1Bp6543CfuuebjFrg36kLAUcskGfaA Verification Key: 68yVKe5AeXynD5A8K91aTZFjCQEoKV4hKPtauqjHa9phgitWEGkS5TR BLS Key: 2zN3bHM1m4rLz54MJHYSwvqzPchYp8jkHswveCLAEJVcX6Mm1wHQD1SkPYMzUDTZvWvhuE6VNAkK3KxVeEmsanSmvjVkReDeBEMxeDaayjcZjFGPydyey1qxBHmTvAnBKoPydvuTAqx5f7YNNRAdeLmUi99gERUU7TD8KfAa6MpQ9bw Node HA: 0.0.0.0:9707 Client HA: 0.0.0.0:9708 Metrics: Uptime: 1 hour, 6 minutes, 56 seconds Total audit Transactions: 39 Total pool Transactions: 4 Total ledger Transactions: 12 Total config Transactions: 0 Read Transactions/Seconds: 0.00 Write Transactions/Seconds: 0.00 Reachable Hosts: 4/4 Node1 (0) Node2 (1) Node4 Node3 Unreachable Hosts: 0/4 Software Versions: indy-node: 1.9.2.dev1061 sovrin: unknown ```

VithushanJegatheeswaran (Fri, 18 Oct 2019 09:57:12 GMT):
I'm running 4 nodes inside of a docker container and exposing the ports (9701-9708) Now I tried to replace the IP addresses in the genesis file to my local IP address. some how I can't connect to the nodes ... did anyone faced the same issue ? ``` Validator Node3 is in unknown state Update time: Friday, October 18, 2019 9:51:04 AM +0000 Validator DID: DKVxG2fXXTU8yT5N7hGEbXB3dfdAnYv1JczDUHpmDxya Verification Key: 5LCf9oRXkDBNG61PpSsBnWDLhyCZWzYYoHAJSwusoSLZCsAre9a5BGL BLS Key: 3WFpdbg7C5cnLYZwFZevJqhubkFALBfCBBok15GdrKMUhUjGsk3jV6QKj6MZgEubF7oqCafxNdkm7eswgA4sdKTRc82tLGzZBd6vNqU8dupzup6uYUf32KTHTPQbuUM8Yk4QFXjEf2Usu2TJcNkdgpyeUSX42u5LqdDDpNSWUK5deC5 Node HA: 0.0.0.0:9705 Client HA: 0.0.0.0:9706 Metrics: Uptime: 1 hour, 6 minutes, 56 seconds Total audit Transactions: 39 Total ledger Transactions: 12 Total pool Transactions: 4 Total config Transactions: 0 Read Transactions/Seconds: 0.00 Write Transactions/Seconds: 0.00 Reachable Hosts: 4/4 Node1 (0) Node2 (1) Node4 Node3 Unreachable Hosts: 0/4 Software Versions: indy-node: 1.9.2.dev1061 sovrin: unknown Validator Node2 is in unknown state Update time: Friday, October 18, 2019 9:51:04 AM +0000 Validator DID: 8ECVSk179mjsjKRLWiQtssMLgp6EPhWXtaYyStWPSGAb Verification Key: 72XEVLaj3htbTYyNXeJWXJCj3ffeJHTuYLtV4PdU1n72YsCzo65aBZ6 BLS Key: 37rAPpXVoxzKhz7d9gkUe52XuXryuLXoM6P6LbWDB7LSbG62Lsb33sfG7zqS8TK1MXwuCHj1FKNzVpsnafmqLG1vXN88rt38mNFs9TENzm4QHdBzsvCuoBnPH7rpYYDo9DZNJePaDvRvqJKByCabubJz3XXKbEeshzpz4Ma5QYpJqjk Node HA: 0.0.0.0:9703 Client HA: 0.0.0.0:9704 Metrics: Uptime: 1 hour, 6 minutes, 56 seconds Total audit Transactions: 39 Total pool Transactions: 4 Total ledger Transactions: 12 Total config Transactions: 0 Read Transactions/Seconds: 0.00 Write Transactions/Seconds: 0.00 Reachable Hosts: 4/4 Node1 (0) Node2 (1) Node4 Node3 Unreachable Hosts: 0/4 Software Versions: indy-node: 1.9.2.dev1061 sovrin: unknown Validator Node1 is in unknown state Update time: Friday, October 18, 2019 9:51:04 AM +0000 Validator DID: Gw6pDLhcBcoQesN72qfotTgFa7cbuqZpkX3Xo6pLhPhv Verification Key: 33nHHYKnqmtGAVfZZGoP8hpeExeH45Fo8cKmd5mcnKYk7XgWNBxkkKJ BLS Key: 4N8aUNHSgjQVgkpm8nhNEfDf6txHznoYREg9kirmJrkivgL4oSEimFF6nsQ6M41QvhM2Z33nves5vfSn9n1UwNFJBYtWVnHYMATn76vLuL3zU88KyeAYcHfsih3He6UHcXDxcaecHVz6jhCYz1P2UZn2bDVruL5wXpehgBfBaLKm3Ba Node HA: 0.0.0.0:9701 Client HA: 0.0.0.0:9702 Metrics: Uptime: 1 hour, 6 minutes, 56 seconds Total audit Transactions: 39 Total ledger Transactions: 12 Total pool Transactions: 4 Total config Transactions: 0 Read Transactions/Seconds: 0.00 Write Transactions/Seconds: 0.00 Reachable Hosts: 4/4 Node1 (0) Node2 (1) Node4 Node3 Unreachable Hosts: 0/4 Software Versions: indy-node: 1.9.2.dev1061 sovrin: unknown Validator Node4 is in unknown state Update time: Friday, October 18, 2019 9:51:04 AM +0000 Validator DID: 4PS3EDQ3dW1tci1Bp6543CfuuebjFrg36kLAUcskGfaA Verification Key: 68yVKe5AeXynD5A8K91aTZFjCQEoKV4hKPtauqjHa9phgitWEGkS5TR BLS Key: 2zN3bHM1m4rLz54MJHYSwvqzPchYp8jkHswveCLAEJVcX6Mm1wHQD1SkPYMzUDTZvWvhuE6VNAkK3KxVeEmsanSmvjVkReDeBEMxeDaayjcZjFGPydyey1qxBHmTvAnBKoPydvuTAqx5f7YNNRAdeLmUi99gERUU7TD8KfAa6MpQ9bw Node HA: 0.0.0.0:9707 Client HA: 0.0.0.0:9708 Metrics: Uptime: 1 hour, 6 minutes, 56 seconds Total audit Transactions: 39 Total pool Transactions: 4 Total ledger Transactions: 12 Total config Transactions: 0 Read Transactions/Seconds: 0.00 Write Transactions/Seconds: 0.00 Reachable Hosts: 4/4 Node1 (0) Node2 (1) Node4 Node3 Unreachable Hosts: 0/4 Software Versions: indy-node: 1.9.2.dev1061 sovrin: unknown ```

VithushanJegatheeswaran (Fri, 18 Oct 2019 10:02:52 GMT):
Here is my genesis file which works fine ``` {"reqSignature":{},"txn":{"data":{"data":{"alias":"Node1","blskey":"4N8aUNHSgjQVgkpm8nhNEfDf6txHznoYREg9kirmJrkivgL4oSEimFF6nsQ6M41QvhM2Z33nves5vfSn9n1UwNFJBYtWVnHYMATn76vLuL3zU88KyeAYcHfsih3He6UHcXDxcaecHVz6jhCYz1P2UZn2bDVruL5wXpehgBfBaLKm3Ba","blskey_pop":"RahHYiCvoNCtPTrVtP7nMC5eTYrsUA8WjXbdhNc8debh1agE9bGiJxWBXYNFbnJXoXhWFMvyqhqhRoq737YQemH5ik9oL7R4NTTCz2LEZhkgLJzB3QRQqJyBNyv7acbdHrAT8nQ9UkLbaVL9NBpnWXBTw4LEMePaSHEw66RzPNdAX1","client_ip":"10.0.0.2","client_port":9702,"node_ip":"10.0.0.2","node_port":9701,"services":["VALIDATOR"]},"dest":"Gw6pDLhcBcoQesN72qfotTgFa7cbuqZpkX3Xo6pLhPhv"},"metadata":{"from":"Th7MpTaRZVRYnPiabds81Y"},"type":"0"},"txnMetadata":{"seqNo":1,"txnId":"fea82e10e894419fe2bea7d96296a6d46f50f93f9eeda954ec461b2ed2950b62"},"ver":"1"} {"reqSignature":{},"txn":{"data":{"data":{"alias":"Node2","blskey":"37rAPpXVoxzKhz7d9gkUe52XuXryuLXoM6P6LbWDB7LSbG62Lsb33sfG7zqS8TK1MXwuCHj1FKNzVpsnafmqLG1vXN88rt38mNFs9TENzm4QHdBzsvCuoBnPH7rpYYDo9DZNJePaDvRvqJKByCabubJz3XXKbEeshzpz4Ma5QYpJqjk","blskey_pop":"Qr658mWZ2YC8JXGXwMDQTzuZCWF7NK9EwxphGmcBvCh6ybUuLxbG65nsX4JvD4SPNtkJ2w9ug1yLTj6fgmuDg41TgECXjLCij3RMsV8CwewBVgVN67wsA45DFWvqvLtu4rjNnE9JbdFTc1Z4WCPA3Xan44K1HoHAq9EVeaRYs8zoF5","client_ip":"10.0.0.2","client_port":9704,"node_ip":"10.0.0.2","node_port":9703,"services":["VALIDATOR"]},"dest":"8ECVSk179mjsjKRLWiQtssMLgp6EPhWXtaYyStWPSGAb"},"metadata":{"from":"EbP4aYNeTHL6q385GuVpRV"},"type":"0"},"txnMetadata":{"seqNo":2,"txnId":"1ac8aece2a18ced660fef8694b61aac3af08ba875ce3026a160acbc3a3af35fc"},"ver":"1"} {"reqSignature":{},"txn":{"data":{"data":{"alias":"Node3","blskey":"3WFpdbg7C5cnLYZwFZevJqhubkFALBfCBBok15GdrKMUhUjGsk3jV6QKj6MZgEubF7oqCafxNdkm7eswgA4sdKTRc82tLGzZBd6vNqU8dupzup6uYUf32KTHTPQbuUM8Yk4QFXjEf2Usu2TJcNkdgpyeUSX42u5LqdDDpNSWUK5deC5","blskey_pop":"QwDeb2CkNSx6r8QC8vGQK3GRv7Yndn84TGNijX8YXHPiagXajyfTjoR87rXUu4G4QLk2cF8NNyqWiYMus1623dELWwx57rLCFqGh7N4ZRbGDRP4fnVcaKg1BcUxQ866Ven4gw8y4N56S5HzxXNBZtLYmhGHvDtk6PFkFwCvxYrNYjh","client_ip":"10.0.0.2","client_port":9706,"node_ip":"10.0.0.2","node_port":9705,"services":["VALIDATOR"]},"dest":"DKVxG2fXXTU8yT5N7hGEbXB3dfdAnYv1JczDUHpmDxya"},"metadata":{"from":"4cU41vWW82ArfxJxHkzXPG"},"type":"0"},"txnMetadata":{"seqNo":3,"txnId":"7e9f355dffa78ed24668f0e0e369fd8c224076571c51e2ea8be5f26479edebe4"},"ver":"1"} {"reqSignature":{},"txn":{"data":{"data":{"alias":"Node4","blskey":"2zN3bHM1m4rLz54MJHYSwvqzPchYp8jkHswveCLAEJVcX6Mm1wHQD1SkPYMzUDTZvWvhuE6VNAkK3KxVeEmsanSmvjVkReDeBEMxeDaayjcZjFGPydyey1qxBHmTvAnBKoPydvuTAqx5f7YNNRAdeLmUi99gERUU7TD8KfAa6MpQ9bw","blskey_pop":"RPLagxaR5xdimFzwmzYnz4ZhWtYQEj8iR5ZU53T2gitPCyCHQneUn2Huc4oeLd2B2HzkGnjAff4hWTJT6C7qHYB1Mv2wU5iHHGFWkhnTX9WsEAbunJCV2qcaXScKj4tTfvdDKfLiVuU2av6hbsMztirRze7LvYBkRHV3tGwyCptsrP","client_ip":"10.0.0.2","client_port":9708,"node_ip":"10.0.0.2","node_port":9707,"services":["VALIDATOR"]},"dest":"4PS3EDQ3dW1tci1Bp6543CfuuebjFrg36kLAUcskGfaA"},"metadata":{"from":"TWwCRQRZ2ZHMJFn9TzLp7W"},"type":"0"},"txnMetadata":{"seqNo":4,"txnId":"aa5e817d7cc626170eca175822029339a444eb0ee8f0bd20d3b0b76e566fb008"},"ver":"1"} ``` I would like to replace the 8 docker IPs with my own local IP address :)

eduelias (Fri, 18 Oct 2019 15:27:27 GMT):
Hi! Let's say I don't "wanna" use libindy to communicate with the node, is there any document/resource I can find to help me?

swcurran (Fri, 18 Oct 2019 17:00:40 GMT):
Use the libindy code as a model and write your own?

lynn.bendixsen (Sat, 19 Oct 2019 14:49:28 GMT):
Genesis file needs external IP addresses (/etc/indy/indy.env has local IP addrs)

kdenhartog (Sun, 20 Oct 2019 03:23:11 GMT):
You can reference the `transactions.md` file in Indy Node which can be found here https://github.com/hyperledger/indy-node/blob/master/docs/source/transactions.md. Additionally, there's a node-js implementation that can be found here. https://github.com/Picolab/node-indy-request

droconnel22 (Sun, 20 Oct 2019 21:02:52 GMT):
Has joined the channel.

droconnel22 (Sun, 20 Oct 2019 21:03:10 GMT):
Hello everyone. I was able to successfully get an Indy Pool up and running in my IBM cloud Virtual machine using this article: https://medium.com/@smaldeniya/setup-hyperledger-indy-pool-in-local-linux-environment-using-docker-304d13eb86dc. I now want to communicate to the indy pool using a Java API agent using the IndySDK: Java. My question is there a resource or How To for connecting to a remote pool? As a best practice I also set up nginx https and certbot. Does anyone know how to route Nginx to the internal docker hosted ports?

droconnel22 (Sun, 20 Oct 2019 21:03:10 GMT):
Thank you in advance for your help.

droconnel22 (Sun, 20 Oct 2019 21:05:26 GMT):
Hello everyone. I was able to successfully get an Indy Pool up and running in my IBM cloud Virtual machine using this article: https://medium.com/@smaldeniya/setup-hyperledger-indy-pool-in-local-linux-environment-using-docker-304d13eb86dc. I now want to communicate to the indy pool using a Java API agent using the IndySDK: Java. My question is there a resource or How To for connecting to a remote pool? As a best practice I also set up nginx https and certbot. Does anyone know how to route Nginx to the internal docker hosted ports? Thank you in advance. Apologies for the multi-message spam.

droconnel22 (Sun, 20 Oct 2019 21:05:26 GMT):
Hello everyone. I was able to successfully get an Indy Pool up and running in my IBM cloud Ubuntu v18 virtual machine using this article: https://medium.com/@smaldeniya/setup-hyperledger-indy-pool-in-local-linux-environment-using-docker-304d13eb86dc. I now want to communicate to the indy pool using a Java API agent using the IndySDK: Java. My question is there a resource or How To for connecting to a remote pool? As a best practice I also set up nginx https and certbot. Does anyone know how to route Nginx to the internal docker hosted ports? Thank you in advance. Apologies for the multi-message spam

droconnel22 (Sun, 20 Oct 2019 21:15:30 GMT):
This is the gensis file I used from the tutorial: `{"reqSignature":{},"txn":{"data":{"data":{"alias":"Node1","blskey":"4N8aUNHSgjQVgkpm8nhNEfDf6txHznoYREg9kirmJrkivgL4oSEimFF6nsQ6M41QvhM2Z33nves5vfSn9n1UwNFJBYtWVnHYMATn76vLuL3zU8 8KyeAYcHfsih3He6UHcXDxcaecHVz6jhCYz1P2UZn2bDVruL5wXpehgBfBaLKm3Ba","blskey_pop":"RahHYiCvoNCtPTrVtP7nMC5eTYrsUA8WjXbdhNc8debh1agE9bGiJxWBXYNFbnJXoXhWFMvyqhqhRoq737YQemH5ik9oL7R4N TTCz2LEZhkgLJzB3QRQqJyBNyv7acbdHrAT8nQ9UkLbaVL9NBpnWXBTw4LEMePaSHEw66RzPNdAX1","client_ip":"127.0.0.1","client_port":9702,"node_ip":"127.0.0.1","node_port":9701,"services":["VALI DATOR"]},"dest":"Gw6pDLhcBcoQesN72qfotTgFa7cbuqZpkX3Xo6pLhPhv"},"metadata":{"from":"Th7MpTaRZVRYnPiabds81Y"},"type":"0"},"txnMetadata":{"seqNo":1,"txnId":"fea82e10e894419fe2bea7d 96296a6d46f50f93f9eeda954ec461b2ed2950b62"},"ver":"1"} {"reqSignature":{},"txn":{"data":{"data":{"alias":"Node2","blskey":"37rAPpXVoxzKhz7d9gkUe52XuXryuLXoM6P6LbWDB7LSbG62Lsb33sfG7zqS8TK1MXwuCHj1FKNzVpsnafmqLG1vXN88rt38mNFs9TENzm4QHd BzsvCuoBnPH7rpYYDo9DZNJePaDvRvqJKByCabubJz3XXKbEeshzpz4Ma5QYpJqjk","blskey_pop":"Qr658mWZ2YC8JXGXwMDQTzuZCWF7NK9EwxphGmcBvCh6ybUuLxbG65nsX4JvD4SPNtkJ2w9ug1yLTj6fgmuDg41TgECXjLCij 3RMsV8CwewBVgVN67wsA45DFWvqvLtu4rjNnE9JbdFTc1Z4WCPA3Xan44K1HoHAq9EVeaRYs8zoF5","client_ip":"127.0.0.1","client_port":9704,"node_ip":"127.0.0.1","node_port":9703,"services":["VALI DATOR"]},"dest":"8ECVSk179mjsjKRLWiQtssMLgp6EPhWXtaYyStWPSGAb"},"metadata":{"from":"EbP4aYNeTHL6q385GuVpRV"},"type":"0"},"txnMetadata":{"seqNo":2,"txnId":"1ac8aece2a18ced660fef86 94b61aac3af08ba875ce3026a160acbc3a3af35fc"},"ver":"1"} {"reqSignature":{},"txn":{"data":{"data":{"alias":"Node3","blskey":"3WFpdbg7C5cnLYZwFZevJqhubkFALBfCBBok15GdrKMUhUjGsk3jV6QKj6MZgEubF7oqCafxNdkm7eswgA4sdKTRc82tLGzZBd6vNqU8dupzup 6uYUf32KTHTPQbuUM8Yk4QFXjEf2Usu2TJcNkdgpyeUSX42u5LqdDDpNSWUK5deC5","blskey_pop":"QwDeb2CkNSx6r8QC8vGQK3GRv7Yndn84TGNijX8YXHPiagXajyfTjoR87rXUu4G4QLk2cF8NNyqWiYMus1623dELWwx57rLCF qGh7N4ZRbGDRP4fnVcaKg1BcUxQ866Ven4gw8y4N56S5HzxXNBZtLYmhGHvDtk6PFkFwCvxYrNYjh","client_ip":"127.0.0.1","client_port":9706,"node_ip":"127.0.0.1","node_port":9705,"services":["VALI DATOR"]},"dest":"DKVxG2fXXTU8yT5N7hGEbXB3dfdAnYv1JczDUHpmDxya"},"metadata":{"from":"4cU41vWW82ArfxJxHkzXPG"},"type":"0"},"txnMetadata":{"seqNo":3,"txnId":"7e9f355dffa78ed24668f0e 0e369fd8c224076571c51e2ea8be5f26479edebe4"},"ver":"1"} {"reqSignature":{},"txn":{"data":{"data":{"alias":"Node4","blskey":"2zN3bHM1m4rLz54MJHYSwvqzPchYp8jkHswveCLAEJVcX6Mm1wHQD1SkPYMzUDTZvWvhuE6VNAkK3KxVeEmsanSmvjVkReDeBEMxeDaayjcZjF GPydyey1qxBHmTvAnBKoPydvuTAqx5f7YNNRAdeLmUi99gERUU7TD8KfAa6MpQ9bw","blskey_pop":"RPLagxaR5xdimFzwmzYnz4ZhWtYQEj8iR5ZU53T2gitPCyCHQneUn2Huc4oeLd2B2HzkGnjAff4hWTJT6C7qHYB1Mv2wU5iHH GFWkhnTX9WsEAbunJCV2qcaXScKj4tTfvdDKfLiVuU2av6hbsMztirRze7LvYBkRHV3tGwyCptsrP","client_ip":"127.0.0.1","client_port":9708,"node_ip":"127.0.0.1","node_port":9707,"services":["VALI DATOR"]},"dest":"4PS3EDQ3dW1tci1Bp6543CfuuebjFrg36kLAUcskGfaA"},"metadata":{"from":"TWwCRQRZ2ZHMJFn9TzLp7W"},"type":"0"},"txnMetadata":{"seqNo":4,"txnId":"aa5e817d7cc626170eca175 822029339a444eb0ee8f0bd20d3b0b76e566fb008"},"ver":"1"}`

droconnel22 (Sun, 20 Oct 2019 21:16:18 GMT):
The idea is I can connect to the pool hosted in the cloud from my local machine while I develop the Java Indy SDK

idowuakinde (Sun, 20 Oct 2019 22:35:55 GMT):
Has joined the channel.

manavgupta (Mon, 21 Oct 2019 03:45:09 GMT):
Hi @k=

manavgupta (Mon, 21 Oct 2019 03:48:15 GMT):
hi @lynn.bendixsen I am running the alice faber getting started python script in java(by converting it in java). In one function did.keyForDid the function is not searching the key on ledger whereas in python it is searching that on ledger. Can you please help me with that.

youngle37 (Mon, 21 Oct 2019 06:05:21 GMT):
Has joined the channel.

sklump (Mon, 21 Oct 2019 10:35:13 GMT):
Has left the channel.

lynn.bendixsen (Tue, 22 Oct 2019 01:10:25 GMT):
Sorry @manavgupta , but I am not familiar with that at all.

manavgupta (Tue, 22 Oct 2019 04:40:36 GMT):
thank you, i got that done. And any updates on custom genesis part?

heenas06 (Tue, 22 Oct 2019 05:44:54 GMT):
Has joined the channel.

chuda (Tue, 22 Oct 2019 11:23:00 GMT):
Has joined the channel.

chuda (Tue, 22 Oct 2019 11:53:59 GMT):
hi linn..iam new to indy before worked in fabric..but unfortunately i didnot get any official doc that give step by step process like fabric to install network locally or cloud and creating node pool local or diffrent machines...could please help me on this ..iam really struggling to create own network and pool as there is many repositories indivisually works...

lynn.bendixsen (Tue, 22 Oct 2019 15:35:39 GMT):
@manavgupta nothing yet, sorry

lynn.bendixsen (Tue, 22 Oct 2019 15:38:29 GMT):
I have a node that will not start and stay started. It is getting an error "keyerror txnMetadata" . any ideas as to what might cause that?

ankita.p (Tue, 22 Oct 2019 16:42:27 GMT):
Has joined the channel.

manavgupta (Wed, 23 Oct 2019 05:16:02 GMT):
what exactly you mean by Start and stay started?

lehors (Wed, 23 Oct 2019 13:58:59 GMT):
hi there

lehors (Wed, 23 Oct 2019 13:59:21 GMT):
I understand Indy has moved to Ursa so is indy-crypto still used at all?

lehors (Wed, 23 Oct 2019 14:01:49 GMT):
the repo points to the wiki for info on the relationship between the different repos but unfortunately it seems a bit out of date and doesn't even reference indy-crypto

lehors (Wed, 23 Oct 2019 14:01:56 GMT):
https://wiki.hyperledger.org/display/indy/

donqui (Wed, 23 Oct 2019 14:45:05 GMT):
`indy-crypto` is still used in `indy-plenum`. `indy-sdk` has migrated to Ursa completely, and we have a PR open for moving `indy-plenum` to it as well, so `indy-crypto` will not be used anymore after that.

lehors (Wed, 23 Oct 2019 14:47:30 GMT):
do you have a timeframe?

lehors (Wed, 23 Oct 2019 14:47:56 GMT):
I'm not asking for a commitment just trying to get an idea on when that might happen

donqui (Wed, 23 Oct 2019 14:52:58 GMT):
soonish, I would say in a couple of weeks.

lehors (Wed, 23 Oct 2019 14:53:13 GMT):
ok, thank you very much!

Philippede 2 (Thu, 24 Oct 2019 12:17:09 GMT):
Has joined the channel.

Philippede 2 (Thu, 24 Oct 2019 12:17:10 GMT):
Hello, I am new to this channel and need some help. I wrote a Backend with VCX (Java-Wrapper). It runs on my server with the Dummy Cloud Agent which listens to port 8070 and 8071. The connection between Backend and Cloud Agent works fine. I am using the open source connector-app to scan the generated QR-code to get a connection with the Cloud Agent. The problem is that, after scanning the code, it runs into a timeout (Failed to connect). I configured the agency-did, -verkey, -endpoint and poolconfig in the developer mode of the app. The Dummy Cloud Agent runs on localhost:8070. I have an nginx configuration that maps requests from .com/agency/ to localhost:8070. When I browse for ".com/agency/" in my browser it shows did and verkey from the Cloud Agent. That means that I can reach it but the app somehow can't. Does anyone know that problem?

PatrikStas (Thu, 24 Oct 2019 12:48:57 GMT):
@Philippede 2 two things come in my mind: 1. Are you running agency on HTTPS? Try without, don't, dummy cloud agent doesn't support that (I am not sure if that can be issue in case there's only 1 agency) 2. Have you set up `forward_agent.endpoint` in agency config file? It should have value of `http://.com`, where the agency resides. Check the content of the QR code to see what address is the connection invitation pointing to

PatrikStas (Thu, 24 Oct 2019 12:48:57 GMT):
@Philippede 2 two things come in my mind: 1. Are you running agency on HTTPS? Try without, don't, dummy cloud agent doesn't support that (I am not sure if that can be issue in case there's only 1 agency) 2. Have you set up `forward_agent.endpoint` in agency config file? It should have value of `http://.com`, where the agency resides. Check the content of the QR code to see what address is the connection invitation pointing to

PatrikStas (Thu, 24 Oct 2019 12:48:57 GMT):
@Philippede 2 two things come in my mind: 1. Are you running agency on HTTPS? Try without, don't, dummy cloud agency doesn't support that (I am not sure if that can be issue in case there's only 1 agency) 2. Have you set up `forward_agent.endpoint` in agency config file? It should have value of `http://.com`, where the agency resides. Check the content of the QR code to see what address is the connection invitation pointing to

BrandonWest (Thu, 24 Oct 2019 17:05:25 GMT):
Has joined the channel.

jjlyon (Fri, 25 Oct 2019 00:02:33 GMT):
Has joined the channel.

jjlyon (Fri, 25 Oct 2019 00:02:35 GMT):
I've been following some getting started guides for indy-sdk, but I've run into an issue with pool timeouts. I set up my pool using this guide: https://hyperledger-indy.readthedocs.io/projects/sdk/en/latest/docs/getting-started/run-getting-started.html And have run various examples in: https://github.com/hyperledger/indy-sdk/tree/master/samples As far as I can tell the indy_pool docker container is running and the ports are open. It seems that I can't access them with localhost (or 0.0.0.0), but I've set the TEST_POOL_ID environment variable to the correct ip address, and it's being pulled in the various apps correctly. I clear the ~/.indy_client/ before runs. I can't think of what else might be the cause. I've been having trouble debugging due to the async calls being made as it gets down into the libindy code. Any thoughts?

Philippede 2 (Fri, 25 Oct 2019 07:56:51 GMT):
The agency was running on HTTPS but with HTTP I get the same error. The endpoint is set correctly. In the react-native debugger I found out that the app cannot initialize VCX for some reason (VCX_INIT_FAIL: CS-002 Failed to init vcx: VCX Exception).

lynn.bendixsen (Fri, 25 Oct 2019 17:10:31 GMT):
Sorry, that was meant to be a separate issue question to the general group here. What I meant was that the indy-node tried to start and failed and then tried a few more times but did not stay running. The issue turned out to be that the node had a year+ old genesis file that was in an old format. When we replaced it with the current genesis files, it worked fine.

lynn.bendixsen (Fri, 25 Oct 2019 17:17:12 GMT):
Hi @chuda It looks like the best place for you to start is with https://github.com/hyperledger/indy-sdk in the "How to start local nodes pool with docker" section of the main readme. If you have already tried that and have some more specific questions or problems with running nodes on different machines, then let me know and I can help more.

esplinr (Sun, 27 Oct 2019 05:22:41 GMT):
Keeping our schedule, the Evernym team provided an update on our contributions to Indy and Aries. https://www.youtube.com/watch?v=r2dgQkjSG2o&list=PLRp0viTDxBWGLdZk0aamtahB9cpJGV7ZF

esplinr (Sun, 27 Oct 2019 05:52:34 GMT):
This Monday only, we will be taking advantage of the staggered adoption of daylight savings time to have a call with additional Indy contributors. The Indy Contributors call will be on October 28 at 11AM US Pacific / 19H Central European Time / 21H Moscow Standard Time / 7H October 29 in New Zealand.

esplinr (Sun, 27 Oct 2019 05:52:58 GMT):
Zoom link is changed too: https://zoom.us/j/715671233

droconnel22 (Sun, 27 Oct 2019 18:33:39 GMT):
Hi again

droconnel22 (Sun, 27 Oct 2019 18:33:46 GMT):
I just built my lib indy

droconnel22 (Sun, 27 Oct 2019 18:34:01 GMT):
➜ libindy git:(master) cargo build Compiling openssl-sys v0.9.49 Compiling openssl v0.10.24 Compiling ursa v0.2.0-dev-1 Compiling libindy v1.11.1 ([mypath]/indy-sdk/libindy) Finished dev [unoptimized + debuginfo] target(s) in 57.54s

droconnel22 (Sun, 27 Oct 2019 18:34:08 GMT):
were is the .so file located?

droconnel22 (Sun, 27 Oct 2019 18:34:26 GMT):
https://github.com/hyperledger/indy-sdk/blob/master/wrappers/java/README.md

donqui (Sun, 27 Oct 2019 18:35:44 GMT):
I believe that it is in `target/debug` directory

donqui (Sun, 27 Oct 2019 18:35:44 GMT):
I believe that it is in the `target/debug` directory

droconnel22 (Sun, 27 Oct 2019 18:36:16 GMT):
-rw-r--r-- 1 dennismacpro staff 0 Oct 12 14:26 .cargo-lock drwxr-xr-x 209 dennismacpro staff 6688 Oct 12 14:27 .fingerprint drwxr-xr-x 68 dennismacpro staff 2176 Oct 12 14:27 build drwxr-xr-x 429 dennismacpro staff 13728 Oct 27 14:31 deps drwxr-xr-x 2 dennismacpro staff 64 Oct 12 14:27 examples drwxr-xr-x 4 dennismacpro staff 128 Oct 12 14:28 incremental -rw-r--r-- 2 dennismacpro staff 152946584 Oct 27 14:30 libindy.a -rw-r--r-- 1 dennismacpro staff 10474 Oct 27 14:31 libindy.d -rwxr-xr-x 2 dennismacpro staff 42883732 Oct 27 14:30 libindy.dylib -rw-r--r-- 2 dennismacpro staff 135757308 Oct 27 14:30 libindy.rlib drwxr-xr-x 2 dennismacpro staff 64 Oct 12 14:27 native

droconnel22 (Sun, 27 Oct 2019 18:36:21 GMT):
dont see it

droconnel22 (Sun, 27 Oct 2019 18:36:56 GMT):
maybe the buiild?

droconnel22 (Sun, 27 Oct 2019 18:39:51 GMT):
hm no luck

droconnel22 (Sun, 27 Oct 2019 18:42:25 GMT):
@donqui do you know if the readme above is still relevant?

droconnel22 (Sun, 27 Oct 2019 18:43:46 GMT):
@donqui tried the cargo build again, looks like it finsihed clean

droconnel22 (Sun, 27 Oct 2019 18:43:51 GMT):
i have all the required packages

donqui (Sun, 27 Oct 2019 18:46:29 GMT):
perfect. I thin the README is ok, there's not much to it.

donqui (Sun, 27 Oct 2019 18:46:29 GMT):
perfect. I think the README is ok, there's not much to it.

droconnel22 (Sun, 27 Oct 2019 18:47:11 GMT):
ok

droconnel22 (Sun, 27 Oct 2019 18:47:17 GMT):
makes sense

droconnel22 (Sun, 27 Oct 2019 18:47:27 GMT):
i was able to add the dependency in my pom file

droconnel22 (Sun, 27 Oct 2019 18:47:36 GMT):
just need to find the .so flie

droconnel22 (Sun, 27 Oct 2019 18:48:51 GMT):
@donqui does the ./lib/ refer to a library folder I have to create in my Java/Maven root directory?

droconnel22 (Sun, 27 Oct 2019 18:49:07 GMT):
or is that in the indy-sdk project directory itself

droconnel22 (Sun, 27 Oct 2019 18:49:18 GMT):
I understand ill need the base c-callable library to run the java wrapper

donqui (Sun, 27 Oct 2019 18:52:12 GMT):
hm, I am confused about that line... I don't think you have to have any special dirs. If you are on Ubuntu I think you only need to copy the `.so` to `/usr/local/lib` and try the wrapper. If you get a NullPointerException I might be wrong :)

droconnel22 (Sun, 27 Oct 2019 18:52:38 GMT):
@donqui currently on macOs catelina

droconnel22 (Sun, 27 Oct 2019 18:52:57 GMT):
@donqui i was able to stand up an ubuntu node pool in my cloud vm

droconnel22 (Sun, 27 Oct 2019 18:53:03 GMT):
trying to write java agent to connect

donqui (Sun, 27 Oct 2019 18:56:03 GMT):
then I think you only need to place the `.so` on your LIBRARY_PATH

donqui (Sun, 27 Oct 2019 18:56:20 GMT):
so that the wrapper tries to load it by name, the OS can find it on the system

donqui (Sun, 27 Oct 2019 18:56:29 GMT):
not sure how that is done on MacOS

droconnel22 (Sun, 27 Oct 2019 18:56:48 GMT):
makes sense

droconnel22 (Sun, 27 Oct 2019 18:56:59 GMT):
i see the wrapper throwing a null pointer exception

droconnel22 (Sun, 27 Oct 2019 18:57:08 GMT):
so i did come at this a bit backwards

droconnel22 (Sun, 27 Oct 2019 18:57:17 GMT):
still can't find the .so file

droconnel22 (Sun, 27 Oct 2019 19:01:02 GMT):
@donqui ill have to wait to ask some one who has built on mac os

droconnel22 (Sun, 27 Oct 2019 19:01:21 GMT):
@donqui thanks for your help

donqui (Sun, 27 Oct 2019 19:04:23 GMT):
np check this out in the meanwhile if you haven't already: https://github.com/hyperledger/indy-sdk/blob/master/docs/build-guides/mac-build.md

droconnel22 (Sun, 27 Oct 2019 19:11:56 GMT):
thanks

droconnel22 (Sun, 27 Oct 2019 19:12:02 GMT):
just reran it manually

droconnel22 (Sun, 27 Oct 2019 19:12:09 GMT):
still no luck with showing the .so file

droconnel22 (Sun, 27 Oct 2019 19:12:24 GMT):
despite finishing cleanly and with sudo perms

droconnel22 (Sun, 27 Oct 2019 19:20:26 GMT):
@donqui ./lib/ might refer to ./usr/lib as i see some shared library there for libsodium etc

droconnel22 (Sun, 27 Oct 2019 19:20:33 GMT):
still no indy however

rahul766gupta (Mon, 28 Oct 2019 10:49:40 GMT):
Has joined the channel.

lynn.bendixsen (Mon, 28 Oct 2019 16:58:05 GMT):
I am on mac and I don't get a libindy.so, I get a /indy-sdk/libindy/target/debug/libindy.dylib (which is the equivalent)

lynn.bendixsen (Mon, 28 Oct 2019 16:58:05 GMT):
I am on mac and I don't get a libindy.so, I get a indy-sdk/libindy/target/debug/libindy.dylib (which is the equivalent)

esplinr (Mon, 28 Oct 2019 18:02:02 GMT):
We are getting started with the Indy Contributors call: https://zoom.us/j/715671233

kevinvanderpoll (Mon, 28 Oct 2019 20:11:15 GMT):
Anyone here have any experience dockerizing an indy environment for testing purposes in concourse?

kevinvanderpoll (Mon, 28 Oct 2019 20:11:17 GMT):
Thanks

droconnel22 (Mon, 28 Oct 2019 23:55:59 GMT):
@lynn.bendixsen thank you!

droconnel22 (Mon, 28 Oct 2019 23:56:32 GMT):
@lynn.bendixsen did you move the libindy.dylib into your /usr/lib folder ?

lynn.bendixsen (Mon, 28 Oct 2019 23:57:37 GMT):
No I have had to add several environment variables, though, that all point to the same place. I will add those here in a sec

lynn.bendixsen (Mon, 28 Oct 2019 23:57:59 GMT):
I am not sure which env variables were needed for which builds that I did...

droconnel22 (Mon, 28 Oct 2019 23:58:21 GMT):
@lynn.bendixsen ok. I am trying to run the Java wrapper if that helps

lynn.bendixsen (Mon, 28 Oct 2019 23:58:49 GMT):
Me as well: ```export LIBRARY_PATH=~/github/indy-sdk/libindy/target/debug/ export LD_LIBRARY_PATH=~/github/indy-sdk/libindy/target/debug/ export LIBINDY_DIR=~/github/indy-sdk/libindy/target/debug/```

lynn.bendixsen (Mon, 28 Oct 2019 23:58:49 GMT):
Me as well (I am using the java wrapper for some of what I do too): ```export LIBRARY_PATH=~/github/indy-sdk/libindy/target/debug/ export LD_LIBRARY_PATH=~/github/indy-sdk/libindy/target/debug/ export LIBINDY_DIR=~/github/indy-sdk/libindy/target/debug/```

droconnel22 (Mon, 28 Oct 2019 23:59:58 GMT):
ok great - thank you

droconnel22 (Tue, 29 Oct 2019 00:00:06 GMT):
ill play around with these

droconnel22 (Tue, 29 Oct 2019 00:02:51 GMT):
@lynn.bendixsen ok I updated the export links and tried to clean build my Java Spring Maven project with the dependencies in the POM, but still getting a null pointer on the first static call to the Pool version

lynn.bendixsen (Tue, 29 Oct 2019 00:10:50 GMT):
You are a bit beyond my experience level at that point, I am afraid, but I will add one more thing to try before I let the experts help you. There was a time a few months ago where I built a .dylib on Mac (a different indy related library) and because something trying to use it was looking for a .so I simply copied the x.dylib to x.so in the same directory and it worked! (Disclaimer: This is not a great long term solution because 1. you have to remember to copy the lib every time you build, and 2. its obviously not meant to work in this klugey way)

droconnel22 (Tue, 29 Oct 2019 00:12:49 GMT):
@lynn.bendixsen ok that might work - thanks x100 for the help

hyper-sunder (Tue, 29 Oct 2019 07:21:50 GMT):
Has joined the channel.

hyper-sunder (Tue, 29 Oct 2019 07:21:51 GMT):
hi guys I am new to indy. I have create docker image *indy-pool* from the `indy-sdk/ci` repo. I created a pool. when I try to connect to the pool i am thrown *Invalid library state: Ledger merkle tree doesn't acceptable for current tree.* error

hyper-sunder (Tue, 29 Oct 2019 07:21:58 GMT):
May I know what I am messing up?

manavgupta (Tue, 29 Oct 2019 10:14:18 GMT):
Hi @hyper-sunder the error which you are getting is due to pool genesis mismatch i.e. the pool genesis for client and indy pool is not same. So make sure they are same and it will work for you. Even the IP address in the pool genesis should be same.

droconnel22 (Wed, 30 Oct 2019 01:00:21 GMT):
Class not loaded : org.hyperledger.indy.sdk.LibIndy

droconnel22 (Wed, 30 Oct 2019 01:00:25 GMT):
still no luck

droconnel22 (Wed, 30 Oct 2019 01:48:35 GMT):
@lynn.bendixsen i fixed it - if you look in the NodeJs wrapper it says for macs to move the libindy.dylib to osx /usr/local/lib/libindy.dylib

droconnel22 (Wed, 30 Oct 2019 02:57:13 GMT):
@lynn.bendixsen are you able to connect to a remote indy pool?

droconnel22 (Wed, 30 Oct 2019 02:57:19 GMT):
I keep getting this error: name: 'IndyError', message: 'PoolLedgerNotCreatedError', indyCode: 300, indyName: 'PoolLedgerNotCreatedError', indyCurrentErrorJson: null

droconnel22 (Wed, 30 Oct 2019 02:58:04 GMT):
await indy.setProtocolVersion(2); let pool = await indy.openPoolLedger("name-of-deployed-pool", fs.readFileSync(process.cwd() + "/" + fileName).toJSON());

droconnel22 (Wed, 30 Oct 2019 02:58:09 GMT):
is the code block

droconnel22 (Wed, 30 Oct 2019 02:58:30 GMT):
i am trying to connect to a remote pool - I have already updated the IP in the gensis.json filie

droconnel22 (Wed, 30 Oct 2019 02:58:39 GMT):
the test_pool_ip

hyper-sunder (Wed, 30 Oct 2019 07:07:17 GMT):
@manavgupta thanks for reply.

hyper-sunder (Wed, 30 Oct 2019 07:14:08 GMT):
@manavgupta Using docker image created indy pool based on the **dockerfile** given in the **indy-sdk** repository in AWS instance. But I am not able to connect to that pool. my transaction genesis file are same.

hyper-sunder (Wed, 30 Oct 2019 07:14:08 GMT):
@manavgupta Using docker image created indy pool based on the **dockerfile** given in the **indy-sdk** repository in AWS instance. But I am not able to connect to that pool. my transaction genesis file are same. Could you please help me?

manavgupta (Wed, 30 Oct 2019 07:16:00 GMT):
Can you provide me the genesis of both?

Huid 1 (Wed, 30 Oct 2019 09:36:29 GMT):
Has joined the channel.

droconnel22 (Wed, 30 Oct 2019 11:48:15 GMT):
@manavgupta I am having the same issue as @hyper-sunder

droconnel22 (Wed, 30 Oct 2019 11:48:33 GMT):
I am using an IBM cloud instance (linux vm) same difference

droconnel22 (Wed, 30 Oct 2019 11:48:51 GMT):
@hyper-sunder what os are you on locally?

hyper-sunder (Wed, 30 Oct 2019 11:52:38 GMT):
i am using ubuntu 16.x in AWS

hyper-sunder (Wed, 30 Oct 2019 11:52:41 GMT):
@droconnel22

hyper-sunder (Wed, 30 Oct 2019 11:57:09 GMT):
@droconnel22

VimalDarji (Wed, 30 Oct 2019 12:03:10 GMT):
Has joined the channel.

VimalDarji (Wed, 30 Oct 2019 12:03:11 GMT):
I am getting error while installing indy-sdk using npm package LINK : warning LNK4044: unrecognized option '/LD:\....\node_modules\indy-sdk.lib'; ignored [D:\....\node_modules\indy-sdk\build\indynodejs.vcxproj] LINK : fatal error LNK1181: cannot open input file '.lib' [D:\....\node_modules\indy-sdk\build\indynodejs.vcxproj]

VimalDarji (Wed, 30 Oct 2019 12:03:16 GMT):
on windows

VimalDarji (Wed, 30 Oct 2019 12:04:25 GMT):
I have set the path for lib also

VimalDarji (Wed, 30 Oct 2019 12:06:11 GMT):
npm install --save indy-sdk@1.9.0

droconnel22 (Wed, 30 Oct 2019 12:10:27 GMT):
@VimalDarji try "indy-sdk": "^1.12.0-dev-1367",

droconnel22 (Wed, 30 Oct 2019 12:10:51 GMT):
also do a git pull on indy sdk and try regenerating the libindy binary

VimalDarji (Wed, 30 Oct 2019 12:11:38 GMT):
I tried as you suggested getting same error

VimalDarji (Wed, 30 Oct 2019 12:11:52 GMT):
LINK : fatal error LNK1181: cannot open input file 'indy.dll.lib'

droconnel22 (Wed, 30 Oct 2019 12:13:25 GMT):
@VimalDarji did you check your sys path env variables

VimalDarji (Wed, 30 Oct 2019 12:13:26 GMT):
you mean clone git pull of https://github.com/hyperledger/indy-sdk.git right?

droconnel22 (Wed, 30 Oct 2019 12:13:45 GMT):
yes exactly

VimalDarji (Wed, 30 Oct 2019 12:15:36 GMT):
need to download this zip - https://repo.sovrin.org/windows/libindy/stable/1.12.0/ and set that path right?

VimalDarji (Wed, 30 Oct 2019 12:22:09 GMT):
@droconnel22 are you there?

VimalDarji (Wed, 30 Oct 2019 12:24:57 GMT):
I have cloned the indy-sdk now what to do?

bansatya (Wed, 30 Oct 2019 13:41:35 GMT):
I am trying to use the libvcx Nodejs wrapper. Init VCX API is working as expected but while trying to use 'Connection.create', it's throwing an error as below: vcx will attempt Connection through SMS { Error: Wallet record not found

bansatya (Wed, 30 Oct 2019 13:41:35 GMT):
I am trying to use the libvcx Nodejs wrapper. Init VCX API is working as expected but while trying to use 'Connection.create', it's throwing an error as below: vcx will attempt Connection through SMS { Error: Wallet record not found However, wallet is created and available with same name given in configuration file.

droconnel22 (Wed, 30 Oct 2019 15:26:48 GMT):
cd into the root

droconnel22 (Wed, 30 Oct 2019 15:26:59 GMT):
run cargo build

droconnel22 (Wed, 30 Oct 2019 15:26:59 GMT):
cd into /indysdk then run cargo build

droconnel22 (Wed, 30 Oct 2019 15:27:11 GMT):
that is a rust package runner as far as I understand

droconnel22 (Wed, 30 Oct 2019 15:27:18 GMT):
it well generate the latest distributable

droconnel22 (Wed, 30 Oct 2019 19:28:58 GMT):
all has anyone been able to connect their indy-sdk to a remote indy-pool?

lynn.bendixsen (Wed, 30 Oct 2019 20:04:30 GMT):
Yes, I can connect. When I do the open_pool_ledger, I pass in the name of an existing pool that is in my local wallet. Did you create_pool_ledger locally before opening it?

droconnel22 (Wed, 30 Oct 2019 20:05:05 GMT):
i only created the pool in a remote linux instance in the ibm cloud

lynn.bendixsen (Wed, 30 Oct 2019 20:06:35 GMT):
Create_pool creates a local pool name that references the remote pool, then you use that name to open the pool (i.e. connect to it). The verbiage is a bit confusing, so keep asking if you need more explanation

lynn.bendixsen (Wed, 30 Oct 2019 20:06:57 GMT):
Create_pool creates a local pool name that references the remote pool, then you use that name to open the pool (i.e. connect to it). The verbiage is a bit confusing, so keep asking if you need more explanation

droconnel22 (Wed, 30 Oct 2019 20:07:33 GMT):
ok, i am also using the exact same gensis.txn file i used remotely

lynn.bendixsen (Wed, 30 Oct 2019 20:11:48 GMT):
here's a link to the code with some small doc explaining the create_pool_config then open_pool workflow.

lynn.bendixsen (Wed, 30 Oct 2019 20:11:58 GMT):
https://github.com/hyperledger/indy-sdk/blob/master/libindy/src/api/pool.rs

droconnel22 (Wed, 30 Oct 2019 20:12:50 GMT):
thanks, ive also made sure to replace the IP in the genesis file with the remote instance

droconnel22 (Wed, 30 Oct 2019 20:13:13 GMT):
the gensis.txn file is a mutli line string?

droconnel22 (Wed, 30 Oct 2019 20:13:17 GMT):
its not json?

lynn.bendixsen (Wed, 30 Oct 2019 20:13:46 GMT):
@droconnel22 I'll check mine...

droconnel22 (Wed, 30 Oct 2019 20:13:57 GMT):
https://github.com/hyperledger/indy-sdk/blob/master/samples/java/src/main/java/utils/PoolUtils.java#L33

lynn.bendixsen (Wed, 30 Oct 2019 20:17:14 GMT):
genesis files are multiline strings. Here are the genesis files that Sovrin uses on its networks as examples of some that are working. https://github.com/sovrin-foundation/sovrin/tree/master/sovrin

lynn.bendixsen (Wed, 30 Oct 2019 20:21:55 GMT):
And yes, as @droconnel22 states, the genesis files must match everywhere.

droconnel22 (Wed, 30 Oct 2019 20:32:52 GMT):
@lynn.bendixsen getting closer

droconnel22 (Wed, 30 Oct 2019 20:32:57 GMT):
i get the error -> c.d.D.l.i.indy.IndyPoolProxy : A non-empty string must be provided for the 'configName' parameter.

droconnel22 (Wed, 30 Oct 2019 20:34:44 GMT):
@lynn.bendixsen nevermind found the error.

droconnel22 (Wed, 30 Oct 2019 20:54:07 GMT):
2019-10-30 16:53:30.059 INFO 61762 --- [ Thread-5] o.h.i.s.L.native.indy.commands.pool : src/commands/pool.rs:75 | OpenAck handle 2, pool_id 2, result Err(IndyError { inner: Ledger merkle tree is not acceptable for current tree.

droconnel22 (Wed, 30 Oct 2019 20:54:15 GMT):
@lynn.bendixsen have you ever seen this one before?

droconnel22 (Wed, 30 Oct 2019 21:07:51 GMT):
@lynn.bendixsen are you passing in the absolute path for PoolJSONParameters.CreatePoolLedgerConfigJSONParameter(genesisFile.getAbsolutePath());

droconnel22 (Wed, 30 Oct 2019 21:08:25 GMT):
{gensis_txn:'/path/to/genesis.txn'}

droconnel22 (Wed, 30 Oct 2019 21:08:41 GMT):
i would think it would be the contents of of the file?

lynn.bendixsen (Wed, 30 Oct 2019 21:16:01 GMT):
I am pretty sure its a fully qualified path (not the contents)

lynn.bendixsen (Wed, 30 Oct 2019 21:16:42 GMT):
I have not seen the "merkle tree not acceptable"

droconnel22 (Wed, 30 Oct 2019 22:33:25 GMT):
2019-10-30 18:32:51.992 INFO 62290 --- [ Thread-5] o.h.i.s.L.native.indy.commands.pool : src/commands/pool.rs:75 | OpenAck handle 2, pool_id 2, result Err(IndyError { inner: Ledger merkle tree is not acceptable for current tree. Invalid library state })

droconnel22 (Wed, 30 Oct 2019 22:33:41 GMT):
@lynn.bendixsen i updated genesis txn to the one deployed

droconnel22 (Wed, 30 Oct 2019 22:33:51 GMT):
when i run my Java Wrapper I get this error

droconnel22 (Wed, 30 Oct 2019 22:34:01 GMT):
anyone from the dev team can help?

droconnel22 (Wed, 30 Oct 2019 22:35:14 GMT):
I also then get the following spammed in my log: | Unexpected timeout: req_id GAAEM9VUE2bzJqiVhfhPWNU8zZLvG5LhenXFEhnqQxNC, node_alias Node2

manavgupta (Thu, 31 Oct 2019 03:28:48 GMT):
ledger merkel tree error is because of genesis mismatch of pool and client

manavgupta (Thu, 31 Oct 2019 03:29:01 GMT):
check whether they are same or not

sauveergoel (Thu, 31 Oct 2019 03:56:16 GMT):
Hi, I am getting "**AnoncredsProofRejected**" Error when i try to verify the proof. any possible reasons for the same??? as the data submitted seems to all in good shape

pedreviljoen (Thu, 31 Oct 2019 08:03:37 GMT):
Has joined the channel.

pedreviljoen (Thu, 31 Oct 2019 08:03:38 GMT):
Hey guys, I am having issues with the iOS SDK, it relates to the following: https://forum.sovrin.org/t/ios-indy-sdk-image-not-found/1306 It may be an issue with the newer versions of XCode. A logic next step would be to downgrade XCode to an older version, but that may not be the best approach for a long term solution. Is there any chance a new version of the libindy-objc will be released? That is compatible with newer versions of XCode?

pedreviljoen (Thu, 31 Oct 2019 08:03:38 GMT):
Hey guys, We are trying to develop a React Native package, wrapping around the LibIndy SDK's available (iOS and Java) I am having issues with the iOS SDK, it relates to the following: https://forum.sovrin.org/t/ios-indy-sdk-image-not-found/1306 It may be an issue with the newer versions of XCode. A logic next step would be to downgrade XCode to an older version, but that may not be the best approach for a long term solution. Is there any chance a new version of the libindy-objc will be released? That is compatible with newer versions of XCode?

tomislav (Thu, 31 Oct 2019 12:31:02 GMT):
@pedreviljoen This has been a lingering issue for a while. Your best bet right is to build Indy.Framework yourself, by manually replacing the libindy-objc pod files. This means you would have to `cargo build` libindy with iOS targets, which actually isn't very complex.

tomislav (Thu, 31 Oct 2019 12:31:29 GMT):
https://github.com/hyperledger/indy-sdk/tree/master/wrappers/ios

WoodenSheep (Thu, 31 Oct 2019 12:33:00 GMT):
Can anyone link me to some starting docs on connect.wallet? My google fu is failing me.

PHilArc (Thu, 31 Oct 2019 17:46:48 GMT):
Has joined the channel.

cstoecker (Fri, 01 Nov 2019 15:05:35 GMT):
Has joined the channel.

esplinr (Fri, 01 Nov 2019 20:41:37 GMT):
I'm sorry that I don't know the answer to your question. But I am curious about your use case. Are you using LibVCX in a new project? Why are you using LibVCX instead of one of the Aries libraries?

esplinr (Fri, 01 Nov 2019 20:43:51 GMT):
My guess is that you are referring to Connect.Me, which is Evernym's commercial identity wallet. See: https://www.evernym.com/products/ And: https://try.connect.me/

donqui (Sat, 02 Nov 2019 10:23:49 GMT):
openssl

abityildiz (Mon, 04 Nov 2019 07:33:24 GMT):
Has joined the channel.

Philippede 2 (Mon, 04 Nov 2019 09:01:03 GMT):
I still have the same issue. Does somebody know how to fix it?

PatrikStas (Mon, 04 Nov 2019 09:08:16 GMT):
@Philippede 2 one more wild guess. You say you can access the agency verkey/did from localhost in browser. What's the URL you can see this data? is it `http://.com/agency` or `http://.com/agency/` ?

PatrikStas (Mon, 04 Nov 2019 09:08:47 GMT):
It must be `http://.com/agency`

Philippede 2 (Mon, 04 Nov 2019 09:09:03 GMT):
The URL I see is http://.com/agency

PatrikStas (Mon, 04 Nov 2019 09:09:25 GMT):
damn...that's not, I thought it might be misconfiguration in nginx mapping

PatrikStas (Mon, 04 Nov 2019 09:09:42 GMT):
is this agency on internet?

PatrikStas (Mon, 04 Nov 2019 09:10:19 GMT):
or is it some sort of local network setup?

Philippede 2 (Mon, 04 Nov 2019 09:10:52 GMT):
It is on internet.

PatrikStas (Mon, 04 Nov 2019 09:11:42 GMT):
do you mind to share url?

PatrikStas (Mon, 04 Nov 2019 09:13:51 GMT):
Here's a thing you can try. In VCX repo there's vcx faber-alice demos implemented in multiple languages. They typically run against local agency

PatrikStas (Mon, 04 Nov 2019 09:14:07 GMT):
try to run these demos, but instead of using local agency, used the one you have deployed on interrrnet

hyper-sunder (Mon, 04 Nov 2019 09:14:16 GMT):
Hi Guys, I am trying to use node `indy-sdk`. I am getting the the below error when i start running my node application. Could anyone help me? error is `Library not loaded: /usr/local/opt/libsodium/lib/libsodium.18.dylib Referenced from: /usr/local/lib/libindy.dylib`

PatrikStas (Mon, 04 Nov 2019 09:14:39 GMT):
if the demos work, you can be sure it's mobile issue (or mobile networking) and not agency issue

vineethkrishnan12 (Mon, 04 Nov 2019 09:16:47 GMT):
Has joined the channel.

PatrikStas (Mon, 04 Nov 2019 09:17:48 GMT):
seem like you are running on osx and missing libsodium on yourr system, that's runtime binary dependency of libindy. Try `brew install libsodium` and then try agaian

hyper-sunder (Mon, 04 Nov 2019 09:18:52 GMT):
sure thanks

hyper-sunder (Mon, 04 Nov 2019 09:20:27 GMT):
still facing the same

hyper-sunder (Mon, 04 Nov 2019 09:21:01 GMT):
using libindy-1.12.0

PatrikStas (Mon, 04 Nov 2019 09:21:24 GMT):
does this path exist on your system? `/usr/local/opt/libsodium/` ?

hyper-sunder (Mon, 04 Nov 2019 09:23:22 GMT):
yes it does exist

PatrikStas (Mon, 04 Nov 2019 09:24:01 GMT):
what's here? `ls /usr/local/opt/libsodium/lib`

hyper-sunder (Mon, 04 Nov 2019 09:24:56 GMT):

Screenshot 2019-11-04 at 2.54.39 PM.png

hyper-sunder (Mon, 04 Nov 2019 09:26:33 GMT):
do I still miss anything?

hyper-sunder (Mon, 04 Nov 2019 09:29:17 GMT):
actually its looking `libsodium.18.dylib` seems but its available 23

PatrikStas (Mon, 04 Nov 2019 09:29:18 GMT):
Well seems like you have too recent version of libsodium. Try `brew uninstall libsodium`, then if there's no errors `brew install libsodium@1.0.12`

hyper-sunder (Mon, 04 Nov 2019 09:32:51 GMT):
sorry to bother more

hyper-sunder (Mon, 04 Nov 2019 09:32:55 GMT):
getting `No available formula with the name "libsodium@1.0.12" `

PatrikStas (Mon, 04 Nov 2019 09:36:37 GMT):
no problem, just gotta find out how to list available versions on brew and figure out which to install. Before that, can you verify that `ls /usr/local/opt/libsodium/lib` no longer contains the dylib file after you run the uninstall?

hyper-sunder (Mon, 04 Nov 2019 09:38:05 GMT):
thanks.... let me try. brew search libsodium lists only 1.0.18 :cry:

PatrikStas (Mon, 04 Nov 2019 09:40:02 GMT):
https://github.com/hyperledger/indy-sdk/blob/master/docs/build-guides/mac-build.md

PatrikStas (Mon, 04 Nov 2019 09:40:32 GMT):
...according to that you should run `brew install https://raw.githubusercontent.com/Homebrew/homebrew-core/65effd2b617bade68a8a2c5b39e1c3089cc0e945/Formula/libsodium.rb `

hyper-sunder (Mon, 04 Nov 2019 09:42:13 GMT):
thank you :) let me try now

akshay.sood (Mon, 04 Nov 2019 10:56:47 GMT):
Has joined the channel.

johnfranklin (Mon, 04 Nov 2019 12:09:12 GMT):
Has joined the channel.

esplinr (Mon, 04 Nov 2019 16:03:15 GMT):
Starting the Indy Maintainers c all: https://zoom.us/j/615818107

VithushanJegatheeswaran (Mon, 04 Nov 2019 16:30:55 GMT):
loglogger

VithushanJegatheeswaran (Mon, 04 Nov 2019 16:50:50 GMT):
are there debugging outputs in the Node.JS wrapper ?

mattatkiva (Mon, 04 Nov 2019 17:11:11 GMT):
the wrapper is literally that...a wrapper and no additional logic. to get debugging logging, you need to follow the instructions for turning on rust logging and implementing your own logger to intercept the debug messages

mattatkiva (Mon, 04 Nov 2019 17:11:11 GMT):
the wrapper is literally that...a wrapper and no additional logic. to get debugging logging any kind of logging you will want to turn on rust logging (which indy sdk is built on). You need to follow the instructions for turning on rust logging and implementing your own logger to intercept the debug messages

donqui (Mon, 04 Nov 2019 17:16:24 GMT):
https://github.com/hyperledger/indy-sdk/tree/master/wrappers/nodejs#logger

donqui (Mon, 04 Nov 2019 17:16:24 GMT):
Maybe this will help: https://github.com/hyperledger/indy-sdk/tree/master/wrappers/nodejs#logger

VithushanJegatheeswaran (Mon, 04 Nov 2019 17:22:17 GMT):
@donqui Thank you :) I searched for the following: ``` indy.setLogger(function (level, target, message, modulePath, file, line) { console.log('libindy said:', level, target, message, modulePath, file, line) }) ```

nehalshah50 (Mon, 04 Nov 2019 23:30:37 GMT):
Does anyone know the sovrin testnet/sandbox support which version on libindy and libvcx?

nehalshah50 (Mon, 04 Nov 2019 23:30:37 GMT):
Does anyone know the version being supported for libindy and libvcx in sovrin testnet/sandbox?

huxd (Tue, 05 Nov 2019 03:43:59 GMT):
Has joined the channel.

vardan10 (Tue, 05 Nov 2019 09:26:34 GMT):
As per proof-of-non-revocation (https://github.com/hyperledger/indy-hipe/tree/master/text/0011-cred-revocation#presenting-proof-of-non-revocation): The prover can calculate an updated witness u by finding the product of all of the non-revoked entries in the tails file and excluding its own entry. The Prover also knows its own factor, b, because it was given its entry's index in the tails file, and can calculate u*b = e, where e is the accumulator. The prover then uses zero knowledge proof to prove to the issuer that he knows u and b using product proofs. Now let’s say the prover has 2 credential definitions, one which is revoked and one which is not. My question is what stops the prover to use u and b from the credential definition that was not revoked while sharing a credential definition that was revoked. Example: Let’s say a prover has a driving license and a university degree from a same issuer. Then the issuer revokes the driving license but not the university degree. In this case while sharing driving license with the verifier the prover also shares u and b values of university degree and not driving license (since driving licence was revoked).

nitika-goel (Tue, 05 Nov 2019 13:00:57 GMT):
Hi, as per the official demo of Indy, we have a separate port for each agent of Indy. However, this does not look scalable. Can someone help me with a strategy to allow unlimited users in Indy.

esplinr (Tue, 05 Nov 2019 14:51:37 GMT):
I don't understand your question. You want to run a separate agent for each of your users, but do it on the same machine?

esplinr (Tue, 05 Nov 2019 14:51:52 GMT):
There is no inherent limit to the number of users in Indy.

esplinr (Tue, 05 Nov 2019 14:52:23 GMT):
But scaling up your agent infrastructure can be a challenge. However it's not inherently different then scaling up multiple instances of any other software.

esplinr (Tue, 05 Nov 2019 14:55:16 GMT):
I don't believe that we have had a backwards incompatible release since before 1.8 (start of the year), so any recent release should work. In general, I recommend using the latest release of LibIndy and LibVCX. Once the Sovrin Network starts enforcing the Transaction Author Agreement, you will need to be on a release that supports it, which is 1.9.0 or above (May 2019).

nehalshah50 (Tue, 05 Nov 2019 15:16:24 GMT):
My current setup is VCX Java Wrapper->Libvcx->LibIndy on enterprise side and Using Connect.Me on client side. So all should work if I stick with release tag 1.8.0? I am currently using libraries using `sudo add-apt-repository "deb https://repo.sovrin.org/sdk/deb bionic stable"`

nehalshah50 (Tue, 05 Nov 2019 15:17:00 GMT):
but when i ran test i keep getting "Error Generating Proof" on Connect.Me which doesn't tell what's wrong

nehalshah50 (Tue, 05 Nov 2019 15:34:15 GMT):
Today Connect.Me app is not able to setup connections using SMS. Anyone who can help?

WoodenSheep (Tue, 05 Nov 2019 16:07:23 GMT):
Thanks. A bunch.

esplinr (Tue, 05 Nov 2019 16:11:59 GMT):
Yes, that should work. I'm not personally familiar with that error.

donqui (Tue, 05 Nov 2019 16:17:31 GMT):
there were some changes over the weekend, are you running the newest version? If not could you do a update and see if it works then.

nehalshah50 (Tue, 05 Nov 2019 16:17:57 GMT):
yes i am running latest...installed it today

nehalshah50 (Tue, 05 Nov 2019 16:18:12 GMT):
i am on Android Demo v 1.0.78.96512

donqui (Tue, 05 Nov 2019 16:22:29 GMT):
are you not getting the SMS, or is the Invitation link in the SMS invalid?

nehalshah50 (Tue, 05 Nov 2019 16:22:55 GMT):
i am getting SMS but when I click on the link, it opens up the app but then nothing happens

donqui (Tue, 05 Nov 2019 16:25:31 GMT):
is the app pointing to the right environment. How are you sending the invite?

esplinr (Tue, 05 Nov 2019 16:25:55 GMT):
This is a known issue. I'll work with you directly @nehalshah50

kevinvanderpoll (Tue, 05 Nov 2019 21:41:32 GMT):
Hello hello! Does anyone have any resource/tutorial on how to host indy with kubernetes?

kevinvanderpoll (Tue, 05 Nov 2019 21:44:11 GMT):
So far, i've been trying to get this(https://github.com/kevvdp/hyperledger-indy-kubernetes)working but am all but bleeding from my ears from bashing my head against the desk.

kevinvanderpoll (Tue, 05 Nov 2019 21:44:11 GMT):
So far, i've been trying to get this https://github.com/kevvdp/hyperledger-indy-kubernetes working but am all but bleeding from my ears from bashing my head against the desk.

scottsen (Wed, 06 Nov 2019 03:10:39 GMT):
Has joined the channel.

nitika-goel (Wed, 06 Nov 2019 06:00:14 GMT):
Is there an alternate to running a separate agent for each user? The demo I used https://github.com/hyperledger/education/tree/master/LFS171x/indy-material/nodejs runs a separate port for each user. Please help. Apologies if this is basic/stupid. I'm new to Hyperledger Indy.

DibbsZA (Wed, 06 Nov 2019 06:28:39 GMT):
@nitika-goel The current agent architectures (which manage their own wallets) mostly require seperate instances. I have just put an API gateway in front of all of mine to limit public port (I am using hostname or path routing on the gateway). Spaceman.id have a custom multi-tennant wallet/agent solution, which I gather uses url/post parameters to open/close access to different wallet stores.

nitika-goel (Wed, 06 Nov 2019 07:41:10 GMT):
Thanks @DibbsZA, I have already made a custom application for 1 of my customers hence would like to go for the API gateway approach only. Is it possible for you to share some code with me?

DibbsZA (Wed, 06 Nov 2019 09:29:27 GMT):
There is no code as such. I have my agents running as docker containers and then I used the free *Kong* API gateway ( https://konghq.com/kong/ also a container) to set up the reverse proxy services and routes.

DibbsZA (Wed, 06 Nov 2019 09:29:27 GMT):
There is no code as such. I have my agents running as docker containers and then I used the free *Kong* API gateway https://konghq.com/kong/ (also a container) to set up the reverse proxy services and routes. I used the free Konga UI to manage that. External to my solution I use Cloudflare's DNS to configure all the virtual hostnames

DibbsZA (Wed, 06 Nov 2019 09:29:27 GMT):
There is no code as such. I have my agents running as docker containers and then I used the free *Kong* API gateway https://konghq.com/kong/ (also a container) to set up the reverse proxy services and routes. I used the free Konga UI to manage that. External to my solution I use Cloudflare's DNS to configure all the virtual hostnames (CNAME records)

DibbsZA (Wed, 06 Nov 2019 09:29:27 GMT):
There is no code as such. I have my agents running as docker containers, each using the same internal port but with different external ports which are set by environment variables. Then I used the free *Kong* API gateway https://konghq.com/kong/ (also a container) to set up the reverse proxy services and routes. I used the free Konga UI to manage that. External to my solution I use Cloudflare's DNS to configure all the virtual hostnames (CNAME records)

DibbsZA (Wed, 06 Nov 2019 09:29:27 GMT):
There is no code as such. I have my agents running as docker containers, each using the same internal port but with different names & datastores which are set by environment variables. ^^^ Then I used the free *Kong* API gateway https://konghq.com/kong/ (also a container) to set up the reverse proxy services and routes. I used the free Konga UI to manage that. External to my solution I use Cloudflare's DNS to configure all the virtual hostnames (CNAME records) ^^^ updated - Using Kong means I don't even *have * to expose the external port of a container as it connects via docker's internal networking to each container instead.

DibbsZA (Wed, 06 Nov 2019 09:29:27 GMT):
There is no code as such. I have my agents running as docker containers, each using the same internal port but with different names & datastores which are set by environment variables. ^^^ Then I used the free *Kong* API gateway https://konghq.com/kong/ (also a container) to set up the reverse proxy services and routes. I used the free Konga UI to manage that. External to my solution I use Cloudflare's DNS to configure all the virtual hostnames (CNAME records) ^^^ updated - Using Kong means I don't even *have * to expose the external port of a container as Kong can route traffic to it via docker's internal networking instead.

VithushanJegatheeswaran (Wed, 06 Nov 2019 11:58:32 GMT):
Hey guys, what is the NodeJS equivalent for this python code: source: https://github.com/hyperledger/indy-sdk/blob/master/docs/how-tos/issue-credential/python/issue_credential.py ``` (cred_json, _, _) = \ await anoncreds.issuer_create_credential(issuer_wallet_handle, cred_offer_json, cred_req_json, cred_values_json, None, None) ```

bansatya (Wed, 06 Nov 2019 12:06:50 GMT):
HI, I am exploring LibVCX and able to run Alice-Faber demo by running dummy agent separately. If I build my own custom application how can we handle the agent communication. Does Indy provides any agent or library to be consumed separately? Can you guys please clarify. I am not getting much about agent and agent2agent communication in wiki.

donqui (Wed, 06 Nov 2019 12:18:24 GMT):
Indy does not provide any agent libraries, you can take a look at #aries if you are interested in this. It is still work in progress but there are some frameworks and consumable agents (https://github.com/hyperledger/aries-cloudagent-python)

bansatya (Wed, 06 Nov 2019 13:02:58 GMT):
@donqui , Thanks. So, implementing custom applications , if we need to establish communication, we need to use an agent separately? Apologies, if it's not making any sense, I am very new to Llibvcx/Indy.

donqui (Wed, 06 Nov 2019 13:04:19 GMT):
If you need to build an app, you need to either: 1. take libvcx and use its methods to build what you need 2. take an already built agent and just make it behave in the way you want

donqui (Wed, 06 Nov 2019 13:05:34 GMT):
So if you build a custom app, you would manage all of the communication throught libvcx, or some other framework

donqui (Wed, 06 Nov 2019 13:05:55 GMT):
or you would consume an agent lib that you will just use to manage that communication for you

VithushanJegatheeswaran (Wed, 06 Nov 2019 13:07:08 GMT):
I found this function ``` let [credJson, ] = await indy.issuerCreateCredential(walletHandle, credOfferJson, credReqJson, credValuesJson,undefined,undefined) ``` but I cant set the last two parameters as `undefined`

donqui (Wed, 06 Nov 2019 13:07:13 GMT):
#aries is the project aimed at building interoperable agents, so you might look at that

donqui (Wed, 06 Nov 2019 13:07:42 GMT):
libvcx works now, and you can build things with it, but aries will be the future of this kind of development

bansatya (Wed, 06 Nov 2019 13:10:19 GMT):
@donqui , Thanks a lot. So, As I understood , through libvcx, I can create inter operable agents. But, then another question, why Alice-faber demo needs a dummy agents run separately though the python script use libvcx.

VithushanJegatheeswaran (Wed, 06 Nov 2019 13:12:45 GMT):
Hey guys, I would like to use this function ``` let [credJson, ] = await indy.issuerCreateCredential(walletHandle, credOfferJson, credReqJson, credValuesJson,undefined,undefined) ``` but I can't set the last two parameters as `undefined` What should I do ?

donqui (Wed, 06 Nov 2019 13:14:01 GMT):
the way libvcx is written it depends on another type of agent, a cloud agent in order to deliver messages to recipients. It dos not support direct Edge 2 Edge communication. Dummy agent acts as that Cloud Agent

bansatya (Wed, 06 Nov 2019 13:14:41 GMT):
was it build earlier when libvcx didn't support this agent 2 agent communication?

VithushanJegatheeswaran (Wed, 06 Nov 2019 13:15:12 GMT):
This is the funtion definition: ``` indy.issuerCreateCredential = function issuerCreateCredential (wh, credOffer, credReq, credValues, revRegId, blobStorageReaderHandle, cb) { cb = wrapIndyCallback(cb, function (data) { return [fromJson(data[0]), data[1], fromJson(data[2])] }) capi.issuerCreateCredential(wh, toJson(credOffer), toJson(credReq), toJson(credValues), revRegId, blobStorageReaderHandle, cb) return cb.promise } ```

donqui (Wed, 06 Nov 2019 13:16:33 GMT):
not sure about that, there are efforts to make libvcx interoperable and to speak the Aries Agent 2 Agent protocol as specified by the RFCs, but I honesty do not know in which builds that is/will be

bansatya (Wed, 06 Nov 2019 13:17:34 GMT):
Got it .Thanks

gaila (Wed, 06 Nov 2019 13:48:05 GMT):
Hello, I am currently working with Indy and I am trying to implement for my use case. I have different people from different organizations that exchange documents that have to be signed and approved by other people and i would like to store these claims in wallets of different organizations. My idea is that each organization has a wallet where it can see all the approved and signed documents from other organizations. But i would like to keep the signatures and hash data also in a project but it is not accessible to everyone. Should i include it in a schema of a proof? Or is it not the common practice in Indy?

gaila (Wed, 06 Nov 2019 13:49:02 GMT):
Hello, I am currently working with Indy and I am trying to implement for my use case. I have different people from different organizations that exchange documents that have to be signed and approved by other people and i would like to store these claims in wallets of different organizations. My idea is that each organization has a wallet where it can see all the approved and signed documents from other organizations. But i would like to keep the signatures and hash data also in a project but it is not accessible to everyone. Should i include it in a schema of a proof? Or is it not the common practice in Indy?

akshay.sood (Wed, 06 Nov 2019 14:47:25 GMT):
Hey guys, I am very new to Hyperledger Indy. Can anyone tell me from where to start? I have read the use-case and example of Alice. I was reading the official documentation "read the docs". Most of the links are broken. Can anyone tell me from where I can start?

swcurran (Wed, 06 Nov 2019 14:59:46 GMT):
I'm a little bias, since I wrote a lot of this :-), but suggest you start here: https://github.com/hyperledger/aries-cloudagent-python/tree/master/docs/GettingStartedAriesDev It begins with the content from an edX course on Indy that was available last year, covering the identity challenges on the Internet and then how Indy addresses those issues. We've expanded that content and will be launching a new edX course coming Real Soon (Nov. 21 launch is currently planned), so that will be another good resource once that starts. We're starting on an Aries Developer edX course that we hope launch in early 2020.

vbhagg (Wed, 06 Nov 2019 15:55:08 GMT):
Is DID generation similar to bip32 where we have a seed phrase and can create multiple keys from that seed?? Please tell the process of how they are generated. Thanks

akshay.sood (Wed, 06 Nov 2019 15:55:56 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=GG3a2ZBLYqRiHux9R) Thanks @swcurran

huxd (Thu, 07 Nov 2019 04:33:37 GMT):
my node just keep reconnecting

huxd (Thu, 07 Nov 2019 04:33:43 GMT):
```2019-11-07 04:13:34,447|INFO|motor.py|Alpha changing status from stopped to starting 2019-11-07 04:13:34,448|INFO|zstack.py|Alpha will bind its listener at 9.47.152.179:9701 2019-11-07 04:13:34,457|INFO|stacks.py|CONNECTION: Alpha listening for other nodes at 9.47.152.179:9701 2019-11-07 04:13:34,457|INFO|zstack.py|AlphaC will bind its listener at 9.47.152.179:9702 2019-11-07 04:13:34,483|INFO|node.py|Alpha first time running... 2019-11-07 04:13:34,483|INFO|node.py|Alpha processed 0 Ordered batches for instance 0 before starting catch up 2019-11-07 04:13:34,483|INFO|ordering_service.py|Alpha:0 reverted 0 batches before starting catch up 2019-11-07 04:13:34,483|INFO|node_leecher_service.py|Alpha:NodeLeecherService starting catchup (is_initial=True) 2019-11-07 04:13:34,483|INFO|node_leecher_service.py|Alpha:NodeLeecherService transitioning from Idle to PreSyncingPool 2019-11-07 04:13:34,483|INFO|cons_proof_service.py|Alpha:ConsProofService:0 starts 2019-11-07 04:13:34,484|INFO|kit_zstack.py|CONNECTION: Alpha found the following missing connections: Node1 2019-11-07 04:13:34,485|INFO|zstack.py|CONNECTION: Alpha looking for Node1 at 127.0.0.1:9701 2019-11-07 04:13:42,500|INFO|zstack.py|Alpha reconnecting to Node1:HA(host='127.0.0.1', port=9701) 2019-11-07 04:13:50,510|INFO|zstack.py|Alpha reconnecting to Node1:HA(host='127.0.0.1', port=9701) 2019-11-07 04:13:58,536|INFO|zstack.py|Alpha reconnecting to Node1:HA(host='127.0.0.1', port=9701) 2019-11-07 04:14:06,550|INFO|zstack.py|Alpha reconnecting to Node1:HA(host='127.0.0.1', port=9701) 2019-11-07 04:14:14,556|INFO|zstack.py|Alpha reconnecting to Node1:HA(host='127.0.0.1', port=9701) 2019-11-07 04:14:22,575|INFO|zstack.py|Alpha reconnecting to Node1:HA(host='127.0.0.1', port=9701) 2019-11-07 04:14:30,599|INFO|zstack.py|Alpha reconnecting to Node1:HA(host='127.0.0.1', port=9701) 2019-11-07 04:14:34,478|NOTIFICATION|node.py|Alpha primary has been disconnected for too long 2019-11-07 04:14:34,478|INFO|node.py|Alpha The node is not ready yet so view change will not be proposed now, but re-scheduled. 2019-11-07 04:14:34,479|INFO|node.py|Alpha scheduling a view change in 60 sec```

huxd (Thu, 07 Nov 2019 04:35:49 GMT):
any idea what's wrong? from previous message I think the node1 (only node in the network) is listening at port 9701...

pwalimbe (Thu, 07 Nov 2019 04:36:48 GMT):
Has left the channel.

PraedipC (Thu, 07 Nov 2019 04:39:25 GMT):
Has joined the channel.

PraedipC (Thu, 07 Nov 2019 04:40:11 GMT):
I tried to run Indy from The git Hub

PraedipC (Thu, 07 Nov 2019 04:40:38 GMT):
I am in the middle of the running set up

PraedipC (Thu, 07 Nov 2019 04:40:59 GMT):
libindy i have gone through

PraedipC (Thu, 07 Nov 2019 04:41:37 GMT):
the steps I followed are mentioned in the Git hub Link only

PraedipC (Thu, 07 Nov 2019 04:46:23 GMT):

Screenshot from 2019-11-06 17-18-43.png

PraedipC (Thu, 07 Nov 2019 04:58:28 GMT):

Indy 32.png

PraedipC (Thu, 07 Nov 2019 04:58:40 GMT):

Indy35.png

PraedipC (Thu, 07 Nov 2019 04:58:40 GMT):

Indy31.png

PraedipC (Thu, 07 Nov 2019 04:58:41 GMT):

Indy2.png

PraedipC (Thu, 07 Nov 2019 04:58:41 GMT):

Indy33.png

PraedipC (Thu, 07 Nov 2019 04:58:42 GMT):

Indy1.png

conanoc (Thu, 07 Nov 2019 06:18:33 GMT):
Has joined the channel.

esplinr (Thu, 07 Nov 2019 15:27:01 GMT):
@PraedipC It is much easier for us to answer a question, then to analyze your screen shots.

esplinr (Thu, 07 Nov 2019 15:27:39 GMT):
I suggest that you explain what you are trying to do, what you tried, and how it behaved. Then ask your question.

sauveergoel (Fri, 08 Nov 2019 04:10:57 GMT):
Hi Gaila,

sauveergoel (Fri, 08 Nov 2019 04:10:57 GMT):
Hi Gaila, you can use any encrypted storage or you can also store these details in an Indy/Identity Wallet without mentioning them in any schema using the non-secrets API. you can have a look here: https://hyperledger-indy.readthedocs.io/projects/sdk/en/latest/docs/design/003-wallet-storage/README.html

sauveergoel (Fri, 08 Nov 2019 10:15:18 GMT):
Hi @nikita-goel, I am guessing that you want to make multiple wallets not agents. every user will be having one unique wallet which then will be used to support various Indy operations. you can create unlimited number of wallets using a single agent. In case you want to know more about agents, wallets etc. feel free to drop me a DM.

sauveergoel (Fri, 08 Nov 2019 10:15:18 GMT):
@nikita-goel, I am guessing that you want to make multiple wallets not agents. every user will be having one unique wallet which then will be used to support various Indy operations. you can create unlimited number of wallets using a single agent. In case you want to know more about agents, wallets etc. feel free to drop me a DM.

sauveergoel (Fri, 08 Nov 2019 10:15:18 GMT):
@nitika-goel , I am guessing that you want to make multiple wallets not agents. every user will be having one unique wallet which then will be used to support various Indy operations. you can create unlimited number of wallets using a single agent. In case you want to know more about agents, wallets etc. feel free to drop me a DM.

MikaLindela (Fri, 08 Nov 2019 11:39:35 GMT):
Has joined the channel.

sam.morrison (Fri, 08 Nov 2019 12:35:31 GMT):
Has joined the channel.

sam.morrison (Fri, 08 Nov 2019 12:35:32 GMT):
Hi all, I have recently started looking at Indy and I was wondering if someone might be able to help me understand some points about proof negotiation. I have been looking at this RFC (https://github.com/hyperledger/aries-rfcs/tree/master/features/0037-present-proof), and have a few queries about the _cred_def_id_. From my current understanding, a credentialDefinition is required for every issuer who wishes to issue a credential for a particular schema. When a verifier requests proof does this mean that they have to ask for a specific definition of the credential in the proof? Is there any way to request proof that the holder owns a credential for a particular schema without specifying the actual definition of the schema?

sukalpomitra (Fri, 08 Nov 2019 15:13:24 GMT):
Hi, how I understand this is as follows:- a credential definition is unique per issuer and the definition id also holds the public did if the issuer. An issuer creates a credential definition from a schema. A verifier introduces a human trust element by adding a restriction in its proof request by adding a list of cred_def_id. Which generally translates that a verifier is trusting a list of issuers. A credential from the same schema but issued by a different issuer not in list in the cred def id provided by the verifier is not acceptable by the verifier. A verifier can also add a restriction of schema name and schema version

sam.morrison (Fri, 08 Nov 2019 16:47:59 GMT):
@sukalpomitra thanks, that helps a lot. From the spec I was reading that as a single cred_def_id...but having a list makes sense. Just to confirm, in the indy implementation I could use something like this when creating the proof: ``` .... 'attr3_referent': { 'name': 'degree', 'restrictions': [ {'cred_def_id': faberTranscriptCredDefId} {'cred_def_id': someOtherTranscriptCredDefId} ] }, .... ``` As an aside, are there any indy docs that provide more details of the format of the proof request other than the spec linked above?

sukalpomitra (Fri, 08 Nov 2019 16:52:01 GMT):
Your understanding is correct but your json structure is wrong It will look something like this ```{\"name\": \"Age Proof\", \"version\": \"1.0\", \"nonce\": \"161176543312099780664559294393629608879\", \"requested_attributes\": {\"390f4a3e-30ab-430f-a5cc-7b0aebd2c9a6\": {\"name\": \"fname\", \"restrictions\": [{\"cred_def_id\": \"8eTQrWPnXgk8iCGtC54xPC:3:CL:635:default\"}, {\"cred_def_id\": \"8eTQrWPnXgk8iCGtC54xPC:3:CL:705:default\"}]}}, \"requested_predicates\": {\"42b0837b-87a5-4a36-8cfe-2e524c31492a\": {\"name\": \"age\", \"restrictions\": [{\"cred_def_id\": \"8eTQrWPnXgk8iCGtC54xPC:3:CL:705:default\"}], \"p_type\": \">=\", \"p_value\": \"21\"}}}"```

sukalpomitra (Fri, 08 Nov 2019 16:52:12 GMT):
I donot know any other docs

sam.morrison (Fri, 08 Nov 2019 16:53:54 GMT):
:thumbsup:

esplinr (Fri, 08 Nov 2019 20:10:35 GMT):
Evernym's latest demo of our contributions to Hyperledger includes a demo of LibVCX working with ACA-Py, and a discussion of performance and stability testing of Indy Node with the view change enhancements: https://www.youtube.com/watch?v=OSpssHQXlOI&list=PLRp0viTDxBWGLdZk0aamtahB9cpJGV7ZF

bansatya (Sat, 09 Nov 2019 03:54:45 GMT):
Hi Guys, I am facing issue while using libvcx nodejs wrapper(Dynamic Symbol Retrieval Error: /usr/lib/libvcx.so: undefined symbol: vcx_update_webhook_url) - libvcx.so r in place and the function seems to be there. not sure whether version issues. Anyone faced this short of issues?

PatrikStas (Sat, 09 Nov 2019 08:32:12 GMT):
The symbol represents function added on latest (or very recent) master branch of libvcx. You either recompile libvcx library on your system, or use slightly older version of libvcx wrapper (a revision where the function is not yet present). In conclusion, you need to make sure inysdk binaries on your system are compatible with wrapper you are using).

bansatya (Sat, 09 Nov 2019 09:00:37 GMT):
Thanks @PatrikStas . Let me check

bansatya (Sat, 09 Nov 2019 09:00:37 GMT):
Thanks @PatrikStas . Let me check.Libvcx issue resolved with build form source. but below issues because of libindy even though build from same source

bansatya (Sat, 09 Nov 2019 13:44:41 GMT):

UnhandledPromiseRejectionWarning: Error: Object (json, config, key, credential and etc...) passed to libindy has invalid structure
    at Function.<anonymous> (/home/satya/indy/education/docs/vcx-web-app/resources/node_modules/node-vcx-wrapper/dist/src/api/schema.js:71:23)
    at Generator.throw (<anonymous>)
    at rejected (/home/satya/indy/education/docs/vcx-web-app/resources/node_modules/node-vcx-wrapper/dist/src/api/schema.js:6:65)
    at <anonymous>

bansatya (Sat, 09 Nov 2019 13:44:41 GMT):
UnhandledPromiseRejectionWarning: Error: Object (json, config, key, credential and etc...) passed to libindy has invalid structure

bansatya (Sat, 09 Nov 2019 14:48:27 GMT):
Follow the steps mentioned in

bansatya (Sat, 09 Nov 2019 14:48:29 GMT):
https://hyperledger-indy.readthedocs.io/projects/sdk/en/latest/docs/build-guides/ubuntu-build.html

vishnupradeepcm (Sun, 10 Nov 2019 04:58:49 GMT):
Has joined the channel.

bansatya (Sun, 10 Nov 2019 17:56:21 GMT):
@PraedipC , Seems like u r trying to start the node pool. Remember, it uses 9701 to 9708 port. Screenshot shows, port r already in use.

bansatya (Sun, 10 Nov 2019 17:56:59 GMT):
Before running the pool, please make sure those ports r not in use or modify the docker file accordingly

SuperSeiyan (Mon, 11 Nov 2019 05:12:01 GMT):
Has joined the channel.

gordon_hkpkiforum (Mon, 11 Nov 2019 10:28:23 GMT):
Has joined the channel.

kunju_shihas (Mon, 11 Nov 2019 12:51:44 GMT):
Has joined the channel.

Chem (Mon, 11 Nov 2019 12:53:25 GMT):
Has joined the channel.

bansatya (Mon, 11 Nov 2019 14:51:23 GMT):
@donqui , I am facing issues while try to create schema using node js wraper of libvcx.The same I can do using python wrapper. I have ensured that all (vcx,libindy) are with latest version. ERROR: (node:32273) UnhandledPromiseRejectionWarning: Error: Object (json, config, key, credential and etc...) passed to libindy has invalid structure

bansatya (Mon, 11 Nov 2019 14:51:30 GMT):
Any thoughts

donqui (Mon, 11 Nov 2019 15:24:07 GMT):
can you post in #indy-sdk, we have indy-sdk experts there that could help you

droconnel22 (Mon, 11 Nov 2019 20:26:35 GMT):
For the Java SDK when connecting to an up and running deploy indy pool in a cloud vm. When I call openPoolLedger I get the exception: Ledger merkle tree is not acceptable for current tree.

droconnel22 (Mon, 11 Nov 2019 20:26:38 GMT):
any thoughts

droconnel22 (Mon, 11 Nov 2019 20:26:53 GMT):
im using the exact same gensis file, saved to my local machine as .txt file

droconnel22 (Mon, 11 Nov 2019 20:48:06 GMT):
note this happens for my remote and local docker hosted versoin

droconnel22 (Mon, 11 Nov 2019 20:48:44 GMT):
i am using the genesis file here: https://github.com/hyperledger/indy-sdk/blob/master/samples/java/src/main/java/utils/PoolUtils.java

droconnel22 (Mon, 11 Nov 2019 20:49:19 GMT):
i have done everything provided out of the box and nothing i have is custom

droconnel22 (Mon, 11 Nov 2019 21:00:46 GMT):
ok so i was able to connect to my local docker pool on 0.0.0.0

droconnel22 (Mon, 11 Nov 2019 21:01:00 GMT):
however there needs to be a guide for connectting to a remote pool

droconnel22 (Mon, 11 Nov 2019 21:28:06 GMT):
https://github.com/hyperledger/indy-sdk/blob/master/samples/java/src/main/java/Ledger.java

droconnel22 (Mon, 11 Nov 2019 21:28:11 GMT):
I keep getting my wallet to be null

droconnel22 (Mon, 11 Nov 2019 21:28:18 GMT):
in the above example

droconnel22 (Mon, 11 Nov 2019 21:28:33 GMT):
anyone encounter this error?

esplinr (Mon, 11 Nov 2019 22:52:44 GMT):
Indy Contributors meeting in 10 minutes: https://zoom.us/j/615818107

sanjaysb (Tue, 12 Nov 2019 07:33:00 GMT):
Has joined the channel.

sanjaysb (Tue, 12 Nov 2019 07:33:06 GMT):
Hi

sanjaysb (Tue, 12 Nov 2019 07:34:46 GMT):
I am facing issues while trying to use the indy-crypto javascript wrapper. I followed the steps to build it. the build script is however failing with "Failed at the indycrypto@0.4.3 build script."

sauveergoel (Tue, 12 Nov 2019 10:13:26 GMT):
Hi @sanjaysb, could you be more specific..maybe pasting the traceback can give a better idea

donqui (Tue, 12 Nov 2019 11:28:03 GMT):
Is there a specific reason why you are trying to use indy-crypto. #ursa project is a new project containing all of the crypto related code, and indy-crypto is goig to be deprecated

VithushanJegatheeswaran (Tue, 12 Nov 2019 12:39:32 GMT):
Hey guys, I have some question about this code: https://github.com/hyperledger/indy-sdk/blob/master/docs/how-tos/negotiate-proof/nodejs/negotiateProof.js 1) Why is the pool not mentioned here ? 2) Where are `proverDid` and `issuerDid` coming from ? I tried to repace them with some random strings but got an error. 3) apparently nothing is stored on the ledger. Do I even need the ledger then / what is the purpose of it? Thank you in advance :)

eduelias (Tue, 12 Nov 2019 15:46:30 GMT):
Hi, I have a strong Ethereum background and kinda like the crypto side of blockchain and I've been hunting the private keys from did's and how to access them from my wallet. My goal is to create a PoC that communicates directly to the node without indy-sdk (just for curiosity). I was forwarded to this (amazing) repo https://github.com/Picolab/node-indy-request but I need some real private keys to start using this approach. How can I access the private keys inside my wallet? (through indy-cli) Thanks guys, sorry if it's a silly/inappropriate question.

lynn.bendixsen (Tue, 12 Nov 2019 19:53:59 GMT):
You can use `wallet export` to export the entire contents of the wallet to a file. You would then need to decrypt the contents of the exported file, and I don't know how to do that.

sanjaysb (Wed, 13 Nov 2019 04:43:42 GMT):
I am looking for something in javascript or with a javascript wrapper that implements the cl signature

vineethkrishnan12 (Wed, 13 Nov 2019 05:31:45 GMT):
helloi..If i want to start Indy from the very beginning, Where should i start? I mean from setting up the network to developing an application..Can u provide few links or applications did on indy?

sauveergoel (Wed, 13 Nov 2019 06:41:44 GMT):
edx

sauveergoel (Wed, 13 Nov 2019 06:43:39 GMT):
you can have a look at this msg from curran: https://chat.hyperledger.org/channel/indy?msg=GG3a2ZBLYqRiHux9R

Philippede 2 (Wed, 13 Nov 2019 08:13:15 GMT):
Hi. When I try to connect to my Cloud Agent that runs on a public IP address I run into a Pool timeout. I captured the traffic with wireshark and it seems like it connects to the right Ports (9701-9708). The client sends a "Hello" and gets back a "Welcome", then sends a "INITIATE" and gets back a "Ready", then sends a "MESSAGE". After this the client keeps sending a TCP retransmission until it resets the TCP connection. Locally, everything works fine. Can someone help me fixing this problem? Thanks!

Philippede 2 (Wed, 13 Nov 2019 08:13:22 GMT):

Clipboard - November 13, 2019 9:13 AM

VenkateshSorapalli (Wed, 13 Nov 2019 08:24:34 GMT):
How to generate multiple wallets for single agent ? Is this possible or not?

sukalpomitra (Wed, 13 Nov 2019 08:37:08 GMT):
I think one agent can have only one wallet.

eduelias (Wed, 13 Nov 2019 09:44:59 GMT):
Can someone tell me how to decrypt an exported wallet file? Thanks! :D

ruairiq (Wed, 13 Nov 2019 10:30:02 GMT):
Has joined the channel.

donqui (Wed, 13 Nov 2019 12:12:34 GMT):
When you export it you provide a key. I think you just need to use that same key for importing it.

donqui (Wed, 13 Nov 2019 12:19:08 GMT):
If you just want to decrypt it without importing it so you have a export file in clear text, please take a look at how that is done in the wallet export/import logic of indy-sdk code.

eduelias (Wed, 13 Nov 2019 14:54:27 GMT):
When I try to use the RAW method, I get the error "Caused by: The base58 input contained a character not part of the base58 alphabet"

ayushmanss (Wed, 13 Nov 2019 17:16:26 GMT):
Has joined the channel.

ayushmanss (Wed, 13 Nov 2019 17:16:27 GMT):
Hi all, I am new to hyperledger projects and wanted to use indy. I have a mac and tried to install the sdk but couldn't start the cli. It says └─> ./indy-cli dyld: Library not loaded: /Users/jenkins/workspace/indy-sdk_indy-sdk-cd_rc/libindy/target/release/deps/libindy.dylib I followed the steps to download all stable builds and updated the LIBRARY_PATH accordingly but it seems the cli points to some other location for libindy binary can someone help as to how to run the given alice example I dont know if this is the right platform to asks such question. If not, can someone point me in the right direction

VithushanJegatheeswaran (Thu, 14 Nov 2019 00:05:07 GMT):
Is it possible to get a credential from an issuer without revealing information.

VithushanJegatheeswaran (Thu, 14 Nov 2019 00:05:07 GMT):
Is it possible to get a credential from an issuer without revealing any information.

VithushanJegatheeswaran (Thu, 14 Nov 2019 00:05:07 GMT):
Is it possible to get a credential from an issuer without revealing any information. For example if Alice want a verifiable credential from Bob that she is over 18. Bob needs to know her date of birth and name or some kind of knowledge that the claim is true right ? Is there a way to get a credential anonymously ?

swcurran (Thu, 14 Nov 2019 00:50:57 GMT):
No - the issuer is supposed to be someone that knows something about - is an authority. Alice doesn't ask a random person (Bob) to say she is over 18, she asks an authority (e.g. a gov't) that is trusted to issue her a credential that has her DoBirth as a claim in the credential. Then, when a verifier asks her to prove she is over 18, she can present a proof that includes a "true" for an "over 18 predicate" based on her DoBirth in the credential she has, without revealing to the verifier her DoBirth (or anything else). The verifier will know: 1. Who issued the credential that has the DoBirth claim in it. 2. That it was issued to Alice (the person being asked). 3. That the data has not been tampered with. 4. That the credential from which the claim was taken has not been revoked. 5. That she is over 18.

rajisp (Thu, 14 Nov 2019 08:34:50 GMT):
Has joined the channel.

VithushanJegatheeswaran (Thu, 14 Nov 2019 08:53:39 GMT):
I want to create a new usecase

VithushanJegatheeswaran (Thu, 14 Nov 2019 08:57:50 GMT):
For my bachelor's thesis I would like to create an anonymous voting system based on DID's. For example, employees of a company can vote their new works works council.

VithushanJegatheeswaran (Thu, 14 Nov 2019 09:03:13 GMT):
Each employee would get a verifiable credential that they are actually working for the company. They can use that credential to authenticate themselves and get redirect to the voting screen. To prevent double voting, the did of the voter is stored In a database. My questions would be: who issues the claim ? How can I make sure that everything is anonymous? Are DID's the right way or is this impossible?

JeromeK (Thu, 14 Nov 2019 14:59:54 GMT):
Has joined the channel.

JeromeK (Thu, 14 Nov 2019 15:00:43 GMT):
Hey Guys Does anybody have the current Node-Versions that sovrin is using in there Builder/Staging/MainNet

swcurran (Thu, 14 Nov 2019 15:51:33 GMT):
I don't think in Indy there is a way to do this. There is a mechanism that has been discussed for how this could be done, but it requires a feature not yet in Indy.

donqui (Thu, 14 Nov 2019 15:54:45 GMT):
one way is to use this https://indyscan.io/home/sovmain, checkout the config ledger and look at the latest NODE_UPGRADE TXN

donqui (Thu, 14 Nov 2019 15:57:25 GMT):
the cross reference with this: https://github.com/sovrin-foundation/sovrin/releases

donqui (Thu, 14 Nov 2019 15:57:25 GMT):
then cross reference with this: https://github.com/sovrin-foundation/sovrin/releases

sukalpomitra (Thu, 14 Nov 2019 15:59:02 GMT):
@swcurran if the voting is done through the established connection, wont it be achieved? the backend of the voting system can be developed in a way so that the voting results are kept separate and who has voted is checked through a flag. That way the flag helps in preventing double voting, while a fact table just shows the voting counts anonymously

donqui (Thu, 14 Nov 2019 16:01:52 GMT):
@lynn.bendixsen maybe you know an easier way to get insight into the current version of packages deployed on Sovrin?

JeromeK (Thu, 14 Nov 2019 16:18:57 GMT):
Thanks for the answer, i actually did this and came up with the following if somebody is interested: (Take no responsibility :P) MainNet: 1.1.58 -> IndyNode: 1.10.0 StagingNet: 1.1.60 -> IndyNode: 1.11.0 BuilderNet: 1.1.56 -> IndyNode: 1.9.2

swcurran (Thu, 14 Nov 2019 16:21:00 GMT):
There are ways if you have an identifier per person, but of course that leaves it open to collusion. The tricky part is being able to do it without that, so there is no chance of being identified.

lynn.bendixsen (Thu, 14 Nov 2019 22:22:14 GMT):
The Sovrin BuilderNet info is not up-to-date on indyscan.io. We did a Reset on that Network a few months ago, but indyscan did not do a similar reset. The current versions for BuilderNet are the same as StagingNet. Sovrin : 1.1.60 -> IndyNode 1.11.0 I know of an easier way for Stewards and Trustees to get the current versions of packages deployed, but nothing better than what was suggested for other users.

lynn.bendixsen (Thu, 14 Nov 2019 22:40:56 GMT):
@channel Does anyone know who is the maintainer of indyscan.io? I need to get them a message about our somewhat recent reset of the Sovrin BuilderNet so that they can also do a reset.

lynn.bendixsen (Thu, 14 Nov 2019 23:14:21 GMT):
hmmm, looks like it is @PatrikStas https://github.com/Patrik-Stas/indyscan

VithushanJegatheeswaran (Fri, 15 Nov 2019 10:10:06 GMT):
my idea would be to get a trusted third party involed as an issuer. He can issue the verifiable credential. The DID's and the vote would get logged. This way I can make sure that a user only votes once but I won't be able to associate the DID with the real Identity. Do you guys see any problems with this ?

PatrikStas (Fri, 15 Nov 2019 15:07:49 GMT):
Hi @lynn.bendixsen what shall be done?

lynn.bendixsen (Fri, 15 Nov 2019 17:58:06 GMT):
I am not 100% familiar, but I think that the old "database" containing the old BuilderNet ledger data needs removed and then start over with rereading all of the current data from the buildernet.

brrussell (Fri, 15 Nov 2019 18:52:43 GMT):
Has joined the channel.

brrussell (Fri, 15 Nov 2019 18:52:44 GMT):
Is there someone that has used/is familiar with the Indy SDK/libindy-objc? I can create a wallet, open wallet, create a DID, and create a pool, but when I try to openLedger with that pool, I just get log output of panic "Fixme: Operation not supported". I can't find anywhere about what part of the operation isn't supported or what I may have setup incorrectly. I used the libindy from 6 November

sairanjit (Sat, 16 Nov 2019 11:53:58 GMT):
Has joined the channel.

acaldas18 (Sat, 16 Nov 2019 19:17:13 GMT):
Hi

acaldas18 (Sat, 16 Nov 2019 19:18:14 GMT):
Can someone explain me the difference between trustee and steward role?

foravneet (Sun, 17 Nov 2019 02:09:54 GMT):
Has joined the channel.

foravneet (Sun, 17 Nov 2019 03:12:50 GMT):
Hi All, struggling with getting some basic getting started example for sovrin/indy to work on my machine.. There are so many different 'getting started' pages, broken links etc that I'm not sure which document should i assume to be the working resource for getting started for a developer. Can someone pl point me to a right resource which i can use today to try out basic functionalities & learn about the system hands on? Thanks !

BalakrishnaChandrasekaran (Mon, 18 Nov 2019 05:49:24 GMT):
Has joined the channel.

BalakrishnaChandrasekaran (Mon, 18 Nov 2019 05:49:41 GMT):
genesis

JeromeK (Mon, 18 Nov 2019 08:37:44 GMT):
I can verify that, something seems outdated

manavgupta (Mon, 18 Nov 2019 10:08:59 GMT):
https://github.com/hyperledger/indy-sdk/tree/master/docs/getting-started just write docker-compose up this script will help you understand how to setup Indy node and key function for Indy

manavgupta (Mon, 18 Nov 2019 10:10:33 GMT):
It is because the Indy Sdk is not able to find the Indy pool network

manavgupta (Mon, 18 Nov 2019 10:10:51 GMT):
i.e.. Indy nodes

manavgupta (Mon, 18 Nov 2019 10:11:11 GMT):
i.e. Indy node

manavgupta (Mon, 18 Nov 2019 10:15:47 GMT):
change the testpoolIP to your remote server IP in both the genesis

awais.ahmad (Mon, 18 Nov 2019 11:01:42 GMT):
Has joined the channel.

awais.ahmad (Mon, 18 Nov 2019 11:01:44 GMT):
Hi all, Is there any support for creating a mobile app using indy ? like I want to go about doing something similar to what the faber college example is doing.

awais.ahmad (Mon, 18 Nov 2019 11:02:44 GMT):
Hi all, which is the favoured wrapper for developing the backend based on indy-sdk ie which one is the most active ?

ekubilay (Mon, 18 Nov 2019 12:46:25 GMT):
Has joined the channel.

ekubilay (Mon, 18 Nov 2019 12:46:26 GMT):
Hi all, could any of you please elaborate on why Indy SDK for Java is at the moment risky to use? As the information provided here: https://github.com/hyperledger/indy-sdk/tree/master/wrappers/java is highly limited. I'd also like to know which SDK would be recommended for production use. Many thanks in advance.

donqui (Mon, 18 Nov 2019 14:43:24 GMT):
I think that the Java wrapper is mature, and that that comment is just something that hasn't been deleted and is not valid. I also see people using Python and NodeJS wrappers quiter regularly so I would say that those should be a good choice.

awais.ahmad (Mon, 18 Nov 2019 15:09:19 GMT):
@donqui which is the most active wrapper?

donqui (Mon, 18 Nov 2019 15:25:23 GMT):
Rust/Java/Python/NodeJS/Objective C are all equally active. They are maintained by the people working on the SDK and are tested aainst CI on every PR.

donqui (Mon, 18 Nov 2019 15:25:23 GMT):
Rust/Java/Python/NodeJS/Objective C are all equally active. They are maintained by the people working on the SDK and are tested by CI on every PR.

awais.ahmad (Mon, 18 Nov 2019 15:25:58 GMT):
thanks.

ekubilay (Mon, 18 Nov 2019 15:27:50 GMT):
Thank you.

esplinr (Mon, 18 Nov 2019 15:54:45 GMT):
Will start the Indy Contributors call in ~5 minutes. https://wiki.hyperledger.org/display/indy/Indy+Contributors+Meeting

esplinr (Mon, 18 Nov 2019 15:55:23 GMT):
Note that most of the wrapper development has moved to Aries.

esplinr (Mon, 18 Nov 2019 16:01:49 GMT):
We are starting the Indy Contributors call: https://zoom.us/j/615818107

foravneet (Mon, 18 Nov 2019 23:54:46 GMT):
Thanks @manavgupta ! , that folder finally had stuff that worked for me.. I was able to start network using 'run-getting-started.md' and execute test transactions from 'getting-started.ipynb'

ankita.p (Tue, 19 Nov 2019 06:49:04 GMT):
revocation

PatrikStas (Tue, 19 Nov 2019 09:15:22 GMT):
@lynn.bendixsen I am using following as genesis pool file for buildernet https://github.com/sovrin-foundation/sovrin/blob/master/sovrin/pool_transactions_builder_genesis

PatrikStas (Tue, 19 Nov 2019 09:16:57 GMT):
@lynn.bendixsen Is that correct one? Also, I haven't detected any domain transactions on mainnet for past 11 days and even more for its other subledgers. Am I having some issue there too? Or is there really no new transactions?

Henrycoffin (Tue, 19 Nov 2019 09:27:33 GMT):
Hi. Just wondering how others are handling the initial trust before a credential is issued. How are you verifying that the person making the connection request is who they say they are? Would be really interesting to hear how people are dealing with this.

JeromeK (Tue, 19 Nov 2019 12:11:36 GMT):
Hey Guys Where can i get the information about Ledger-Resets for the Sovrin Network(Lifecycle etc) since they only mention it partly in the Glossary https://docs.google.com/document/d/1gfIz5TT0cNp2kxGMLFXr19x1uoZsruUe_0glHst2fZ8/edit#heading=h.qbgronng23rl

JeromeK (Tue, 19 Nov 2019 12:11:36 GMT):
Hey Guys Where can i get the information about Ledger-Resets for the Sovrin Network(Lifecycle etc) since they only mention it partly ("need to be frequently reset. ")in the Glossary https://docs.google.com/document/d/1gfIz5TT0cNp2kxGMLFXr19x1uoZsruUe_0glHst2fZ8/edit#heading=h.qbgronng23rl

JeromeK (Tue, 19 Nov 2019 12:12:44 GMT):
And also why there is no Author-Agreement on Staging but one in the BuilderNet

DibbsZA (Tue, 19 Nov 2019 12:42:56 GMT):
I think its important to clarify the language and the context before moving to deeper subject. A *connection *is just a channel between two agents. At this level there is *no awareness of who/what exists beyond* the endpoints of that connection. Thus connection requests can be safely anonymous. Now we are back where every authenticated website has to ask the user to provide something unique that they know that can be verified independantly by the website. In my KYC scenario, the user provides their national ID number and a Selfie which is used in a call to our National Identity database for biometric comparison. On success I can Issue a NationalID credential to the user which they can use another time to do authentication .

DibbsZA (Tue, 19 Nov 2019 12:42:56 GMT):
I think its important to clarify the language and the context before moving to deeper subject. A *connection *is just a channel between two agents. At this level there is *no awareness of who/what exists beyond* the endpoints of that connection. Thus connection requests can be safely anonymous. Now we are back where every authenticated website has to *ask the user to provide something unique* that they know that can be verified independantly by the website. In my KYC scenario, the user provides their national ID number and a Selfie which is used in a call to our National Identity database for biometric comparison. Only on success can I Issue a NationalID credential to the user which they can use another time to proove who they are.

ayushmanss (Tue, 19 Nov 2019 13:15:01 GMT):
Can anybody tell me the difference between steward and trustee role. From the auth map it seems both of them can perform the same transactions

lynn.bendixsen (Tue, 19 Nov 2019 15:00:12 GMT):
@PatrikStas Yes that is the correct genesis pool file for BuilderNet. I added some txns to MainNet just now...I will go check. Thanks for the heads-up.

lynn.bendixsen (Tue, 19 Nov 2019 15:00:12 GMT):
@PatrikStas Yes that is the correct genesis pool file for BuilderNet. I added some txns to MainNet just now...I will go check. Thanks for noticing and checking in.

lynn.bendixsen (Tue, 19 Nov 2019 15:02:10 GMT):
Just checked and the txn's I added showed up under the config ledger, so it seems to still be working fine. thx!

PatrikStas (Tue, 19 Nov 2019 15:03:04 GMT):
cool :) What's the issue with Buildernet though? There was actually some new domain transactions 6 hours ago

lynn.bendixsen (Tue, 19 Nov 2019 15:04:51 GMT):
I am not sure I understand what you mean by "issue". We created a tool where someone can add themselves as an ENDORSER to the BuilderNet or StagingNet any time they want... maybe it was that?

PatrikStas (Tue, 19 Nov 2019 15:06:25 GMT):
oh well, I mean ... you mentioned that something seem off with data dsiplaye for Buildernet on indyscan.io. But from what I can see the latest are correctly displayed

lynn.bendixsen (Tue, 19 Nov 2019 15:06:46 GMT):
oh, I see, lemme check.

lynn.bendixsen (Tue, 19 Nov 2019 15:08:14 GMT):
It could be that it finally got as many txns as what it used to have and has been overwriting the entries as it goes and finally caught up...

lynn.bendixsen (Tue, 19 Nov 2019 15:09:05 GMT):
(Domain has todays date, but config and pool are definitely still old)

PatrikStas (Tue, 19 Nov 2019 15:10:23 GMT):
I see...do you know by a chance how many pool and config txs is on buildernet really?

lynn.bendixsen (Tue, 19 Nov 2019 15:13:11 GMT):
from my node: ```Total config Transactions: 140 Total ledger Transactions: 310 Total audit Transactions: 77943 Total pool Transactions: 32 Total 1001 Transactions: 43 ```

lynn.bendixsen (Tue, 19 Nov 2019 15:15:59 GMT):
Hmm my theory is wrong, I looked at all of the pool txns in the tool and none have a recent timestamp.

lynn.bendixsen (Tue, 19 Nov 2019 15:15:59 GMT):
Hmm my theory seems wrong, I looked at all of the pool txns in the tool and none have a recent timestamp.

esplinr (Tue, 19 Nov 2019 22:25:17 GMT):
A steward hosts a node. A trustee can run operations across the entire network.

esplinr (Tue, 19 Nov 2019 22:25:47 GMT):
The Sovrin Glossary can help: do you know why the field names in schemas and cred defs have different canonicalization rules?

esplinr (Tue, 19 Nov 2019 22:25:47 GMT):
The Sovrin Glossary can help: https://sovrin.org/library/glossary/

esplinr (Tue, 19 Nov 2019 22:26:58 GMT):
They haven't yet established a policy for ledger resets of Builder Net and Staging Net, but they want everyone to know it could happen. The data there is not expected to be persistent over the long term.

esplinr (Tue, 19 Nov 2019 22:27:55 GMT):
The TAA on Builder Net was to test the system. They are working on the real TAA for Builder Net and expect to have one on Staging Net as well. The text hasn't been drafted yet.

esplinr (Tue, 19 Nov 2019 22:29:22 GMT):
I don't think they have ever reset Staging Net, but Builder Net has been reset when things got out-of-wack.

smohan (Wed, 20 Nov 2019 04:57:40 GMT):
Has joined the channel.

muddahany (Wed, 20 Nov 2019 09:09:05 GMT):
Has joined the channel.

gaila (Wed, 20 Nov 2019 09:58:28 GMT):
Hello, I am checking the use cases of hyperledger indy and I am interested in this one "Software Development Pipeline — Verify development builds were completed properly and officially signed by the developers for consumers of the software. This could also be useful because build credentials could be revoked if the software has a vulnerability informing consumers of the software, leading to very powerful decentralized security alert system" Can someone please explain how indy would help with it? And wouldn't be better to use, for example, fabric, to keep information about each build?

esplinr (Wed, 20 Nov 2019 16:07:14 GMT):
The information about each build isn't kept on the ledger. The developer issues a credential that is distributed along with the build, and users of the build can verify it by referencing the keys on the blockchain.

sebastillar (Wed, 20 Nov 2019 22:18:39 GMT):
set up

sebastillar (Wed, 20 Nov 2019 22:20:27 GMT):
indy-node

VithushanJegatheeswaran (Thu, 21 Nov 2019 13:44:33 GMT):
Hey guys can someone explain to me why the pool is not opened in this code: https://github.com/hyperledger/indy-sdk/blob/master/docs/how-tos/negotiate-proof/nodejs/negotiateProof.js

VithushanJegatheeswaran (Thu, 21 Nov 2019 13:44:33 GMT):
Hey guys can someone explain to me why the pool is not opened in this sample: https://github.com/hyperledger/indy-sdk/blob/master/docs/how-tos/negotiate-proof/nodejs/negotiateProof.js Is the pool necessary ?

esplinr (Thu, 21 Nov 2019 19:14:59 GMT):
The Evernym team's fortnightly demo explains some fixes to Indy Node and shows the Aries Test Suite running against LibVCX. https://www.youtube.com/watch?v=lH5tjrb1uu0&list=PLRp0viTDxBWGLdZk0aamtahB9cpJGV7ZF

manavgupta (Fri, 22 Nov 2019 04:47:26 GMT):
Hi @lynn.bendixsen Can uou tell me how can i see the ledger transaction on indy node? like if i want to see schemaId, credential definition, did's and all

manavgupta (Fri, 22 Nov 2019 04:47:26 GMT):
Hi @lynn.bendixsen Can you tell me how can i see the ledger transaction on indy node? like if i want to see schemaId, credential definition, did's and all

HLFPOC (Fri, 22 Nov 2019 06:05:15 GMT):
Hi Team, Can anyone please differentiate between Steward and a trust anchor role in Sovrin? As per my understanding, steward is the one who takes responsibility of running a node of Sovrin, and trust anchor issues/verifies credentials. In order to do that, trust anchor will also have to run a node in Sovrin? So indirectly a trust anchor is also a steward ? Can someone please give some inputs on this ?

swcurran (Fri, 22 Nov 2019 16:10:11 GMT):
For public instance of indy, you can use this - https://indyscan.io/home/sovmain If you want to run your own network and have a browser for it, you can use this - https://github.com/bcgov/von-network

lynn.bendixsen (Fri, 22 Nov 2019 17:56:50 GMT):
From a node there is a tool named `read_ledger`

lynn.bendixsen (Fri, 22 Nov 2019 17:57:29 GMT):
start with `read_ledger help` to see what its all about. And here is an example:

lynn.bendixsen (Fri, 22 Nov 2019 17:57:29 GMT):
start with `read_ledger help` to see what its all about. And here is an example of its capabilities:

lynn.bendixsen (Fri, 22 Nov 2019 17:57:41 GMT):
```sudo read_ledger --type main --frm 0 --to 3000 |vim -```

sebastillar (Sat, 23 Nov 2019 15:26:22 GMT):
Hi everyone. I am trying to run a local network on Ubuntu 16.04. Which are the commands? Do I need to use IndyCLI?

sebastillar (Sat, 23 Nov 2019 15:26:38 GMT):
Thanks in advance

ayushmanss (Sun, 24 Nov 2019 19:33:31 GMT):
@sebastillar - you dont need indycli to run a network but to interact with it. Easiest way is to use a docker image packaged with all the requirements. You can use an image from von. Here are the steps to get everything running https://github.com/bcgov/von-network#running-the-network-locally

manavgupta (Mon, 25 Nov 2019 03:41:53 GMT):
thank you @swcurran and @lynn.bendixsen for your help

liormargalit (Mon, 25 Nov 2019 13:00:14 GMT):
I have encountered the following error when trying to connect to the test pool docker: "Ledger merkle tree is not acceptable for current tree". What may cause this error?

JeromeK (Mon, 25 Nov 2019 13:00:58 GMT):
Did you set the Protocol-Version = 2 ?

liormargalit (Mon, 25 Nov 2019 13:01:49 GMT):
Yes, i have set it

JeromeK (Mon, 25 Nov 2019 13:06:41 GMT):
Could you send some code-snippets - would help

liormargalit (Mon, 25 Nov 2019 13:09:39 GMT):
i am using the getting started code. It worked for me when I ran the test pool docker "manually" but I encounter this error when I try to run it inside minikube

liormargalit (Mon, 25 Nov 2019 13:10:37 GMT):
I am trying to figure out how to configure the deployment to allow the connection and meanwhile encountered this error

liormargalit (Mon, 25 Nov 2019 13:11:19 GMT):
do you have an example of deploying the test pool in kubernetes?

liormargalit (Mon, 25 Nov 2019 14:15:30 GMT):
for now as a workaround I am running the test pool in the same pod with the client. In case anyone else encounters this issue...

JeromeK (Mon, 25 Nov 2019 14:19:18 GMT):
did you update the genesis-file, exposed the ports and so on?

liormargalit (Mon, 25 Nov 2019 14:25:51 GMT):
yes, I have set the TEST_POOL_IP env variable with the clusterIP (this env variable is used in pool_genesis_txn_data() in https://github.com/hyperledger/indy-sdk/blob/master/samples/python/src/utils.py

liormargalit (Mon, 25 Nov 2019 14:26:00 GMT):
Also exposed the ports

liormargalit (Mon, 25 Nov 2019 14:27:01 GMT):
the ports 9701-9708 in the service

liormargalit (Mon, 25 Nov 2019 14:28:00 GMT):
Something interesting, in https://github.com/hyperledger/indy-sdk/#docker-port-mapping-on-macos it mentions ports 9701-9709

liormargalit (Mon, 25 Nov 2019 14:28:22 GMT):
should I have exposed 9709 either?

JeromeK (Mon, 25 Nov 2019 16:10:25 GMT):
Hmm strange i dont think you have to expose 9709

JeromeK (Mon, 25 Nov 2019 16:10:41 GMT):
did you check if the Env works?

JeromeK (Mon, 25 Nov 2019 16:10:55 GMT):
and else just update the Docker-File and add the modified genesis

JeromeK (Mon, 25 Nov 2019 16:10:57 GMT):
could help

JeromeK (Mon, 25 Nov 2019 16:11:19 GMT):
or maybe it could also be a Host-File error on Mac

athira (Mon, 25 Nov 2019 19:01:00 GMT):
Has joined the channel.

athira (Mon, 25 Nov 2019 19:01:17 GMT):
#indy-agent

athira (Mon, 25 Nov 2019 19:02:12 GMT):
#aries

athira (Mon, 25 Nov 2019 19:03:12 GMT):
#indy-outreach

esplinr (Mon, 25 Nov 2019 23:02:40 GMT):
Indy Contributors call is starting: https://zoom.us/j/615818107

manavgupta (Tue, 26 Nov 2019 11:22:31 GMT):
Hi @lynn.bendixsen Any update on custom genesis file problem which i had back in october?

Philippede 2 (Tue, 26 Nov 2019 12:47:15 GMT):
Hi, I have an issue with the connector-app. When I try to send a proof to the backend I get the following error in the connector-app: "Error generating proof". Does anyone know this problem?

VithushanJegatheeswaran (Tue, 26 Nov 2019 14:12:15 GMT):
How can the issuer be sure that you are the person youre telling to be ? for example: The government wants to issue a credential that you are "Alice". But how can they be sure that you are really Alice ?

sukalpomitra (Tue, 26 Nov 2019 14:39:28 GMT):
The same they are sure today and issue Alice her passport

sukalpomitra (Tue, 26 Nov 2019 14:39:28 GMT):
The same way they are sure today and issue Alice her passport.

JeromeK (Wed, 27 Nov 2019 08:42:55 GMT):
Hey Guys Does the sovrin have some kind of bottlenecks? Because i want to get a cred-def from the ledger with a schema of 88 Attributes but i always get an Timeout. Any known "Bugs" or fixes for that Problem? Already tried to fix it with the indy-cli or over the python-wrapper either got the cred-def. But its for sure on the Ledger (looked it up in indyscan) Greets Jerome

PatrikStas (Wed, 27 Nov 2019 08:46:11 GMT):
Hi @JeromeK which network, which creddef?

JeromeK (Wed, 27 Nov 2019 08:50:53 GMT):
Hey Its on Staging

JeromeK (Wed, 27 Nov 2019 08:51:05 GMT):
and im now able to pull it with my python-script

JeromeK (Wed, 27 Nov 2019 08:51:15 GMT):
but not with the CLI

JeromeK (Wed, 27 Nov 2019 08:53:12 GMT):
https://indyscan.io/tx/sovstaging/domain/79462

JeromeK (Wed, 27 Nov 2019 08:53:12 GMT):
https://indyscan.io/tx/sovstaging/domain/79461

PatrikStas (Wed, 27 Nov 2019 08:57:02 GMT):
`ledger custom {"reqId":1,"identifier":"V4SGRU86Z58d6TV7PBUe6f","operation":{"type":"3","ledgerId":1,"data":79461},"protocolVersion":2}`

PatrikStas (Wed, 27 Nov 2019 08:57:04 GMT):
Works for me

PatrikStas (Wed, 27 Nov 2019 08:57:36 GMT):
Maybe check you cli version? I am running `1.11.1`

JeromeK (Wed, 27 Nov 2019 08:58:46 GMT):
Really Strange... it works with your "custom-command"

JeromeK (Wed, 27 Nov 2019 08:58:57 GMT):
but if i try it like that : `ledger get-cred-def schema_id=79461 signature_type=CL tag=tprm-tag1 origin=VmcAeAQ7D5eRyLWLMKD9oq`

JeromeK (Wed, 27 Nov 2019 08:58:59 GMT):
it does not

PatrikStas (Wed, 27 Nov 2019 08:59:05 GMT):
try `ledger get-schema did=VmcAeAQ7D5eRyLWLMKD9oq name=tprm-demo version=1.0`

PatrikStas (Wed, 27 Nov 2019 08:59:20 GMT):
ahha... yeah you want cred def, right

JeromeK (Wed, 27 Nov 2019 08:59:27 GMT):
correct

JeromeK (Wed, 27 Nov 2019 08:59:41 GMT):
and ya it works in my python script sometimes

JeromeK (Wed, 27 Nov 2019 08:59:57 GMT):
thats why i asked for some known bottlenecks

JeromeK (Wed, 27 Nov 2019 09:01:14 GMT):
yesterday we had no chance getting this

PatrikStas (Wed, 27 Nov 2019 09:05:43 GMT):
What version of cli you are using? Cause on 1.11 the `get-cred-def` actually requires different parameters than what you have there

PatrikStas (Wed, 27 Nov 2019 09:06:02 GMT):
fro `1.11 cli`: ``` Examples: ledger cred-def schema_id=1 signature_type=CL tag=1 primary={"n":"1","s":"2","rms":"3","r":{"age":"4","name":"5"},"rctxt":"6","z":"7"} ```

JeromeK (Wed, 27 Nov 2019 09:08:52 GMT):
the same with this command: `ledger custom {"reqId":1,"identifier":"LibindyDid111111111111","operation":{"type":"108","ref":79461,"signature_type":"CL","origin":"VmcAeAQ7D5eRyLWLMKD9oq","tag":"tprm-tag1"},"protocolVersion":2}` I mean there is a limit of 125 Attributes when publishing(this works just fine), but could it be that you can only get cred-def(s) with around 50 Attributes because of some server-performance?

JeromeK (Wed, 27 Nov 2019 09:08:52 GMT):
the same with this command: `ledger custom {"reqId":1,"identifier":"LibindyDid111111111111","operation":{"type":"108","ref":79461,"signature_type":"CL","origin":"VmcAeAQ7D5eRyLWLMKD9oq","tag":"tprm-tag1"},"protocolVersion":2}` I mean there is a limit of 125 Attributes when publishing schemas(this works just fine), but could it be that you can only get cred-def(s) with around 50 Attributes because of some server-performance?

PatrikStas (Wed, 27 Nov 2019 09:10:51 GMT):
Yeah seems like timing out for me as well. Maybe post this issue on `indy-node` channel as well.

JeromeK (Wed, 27 Nov 2019 09:11:25 GMT):
I will, thanks trying :)

eduelias (Wed, 27 Nov 2019 09:50:41 GMT):
Guys, sorry if it's not the right place to ask this, but I didn't find anywhere better. Here it goes: We have a proof request that looks like this: ``` { } ```

Henrycoffin (Wed, 27 Nov 2019 09:53:22 GMT):
Hi everyone. Can anyone point me at any docs explaining how wallet security works? Specifically around storage. I am really looking for something i can share with business people. Thanks

eduelias (Wed, 27 Nov 2019 09:55:56 GMT):
Hi guys, sorry if it's not the right place to ask this, but I didn't find anywhere better. We have a proof request like this: ``` { [...] "requested_attributes": { "attr0_referent": { "name": "first_name", "restrictions": [{"cred_def_id": credDefIdShareGlobal}] } } [...] } ``` Is it possible to have multiple "cred_def_id" restrictions on the same attribute, like this: `"restrictions": [{"cred_def_id": CredDefId1}, {"cred_def_id": CredDefId2}]` ? Thanks in advance!

lsl88 (Wed, 27 Nov 2019 10:36:23 GMT):
Has joined the channel.

utkmaheshwari245 (Wed, 27 Nov 2019 10:36:39 GMT):
Has joined the channel.

utkmaheshwari245 (Wed, 27 Nov 2019 10:36:41 GMT):
hi

utkmaheshwari245 (Wed, 27 Nov 2019 10:37:01 GMT):
Is there a way to generate a credential request without a credential offer ?

utkmaheshwari245 (Wed, 27 Nov 2019 10:38:18 GMT):
As per the documentation, it looks like it is possible - https://hyperledger-indy.readthedocs.io/projects/sdk/en/latest/docs/how-tos/issue-credential/README.html#step-4

utkmaheshwari245 (Wed, 27 Nov 2019 10:38:31 GMT):
But I can't find a way to do it ?

utkmaheshwari245 (Wed, 27 Nov 2019 10:39:04 GMT):
Is there a way to do so ?

Philippede 2 (Wed, 27 Nov 2019 11:49:37 GMT):
Hi Guys. How can you get the attribute values from a requested proof in VCX? When I use "ProofApi.getProof(...)", the attribute values are encrypted. Is there a way to decrypt the attribute values?

JeromeK (Wed, 27 Nov 2019 11:51:21 GMT):
I think the important things are Cred-Def-ID and Schema-ID to get everything, so you dont a "ready to use" credential offer since you can genrate that on the fly

JeromeK (Wed, 27 Nov 2019 11:51:21 GMT):
I think the important things are Cred-Def-ID and Schema-ID to get all the information's you need, so you don't a "ready to use" credential offer since you can generate that on the fly

sarkoi (Wed, 27 Nov 2019 15:06:26 GMT):
Has joined the channel.

sukalpomitra (Wed, 27 Nov 2019 16:11:44 GMT):
yes its possible

eduelias (Wed, 27 Nov 2019 16:12:10 GMT):
Can you tell me how? The way I did is the correct one?

sukalpomitra (Wed, 27 Nov 2019 16:12:29 GMT):
yes its correct, the way you have done

eduelias (Wed, 27 Nov 2019 16:12:36 GMT):
Thanks! :D

swcurran (Wed, 27 Nov 2019 16:49:56 GMT):
I believe a credentialOffer is required first in Indy. It's part of the ZKP crypto - the issuer is commiting to what it is going to issue.

swcurran (Wed, 27 Nov 2019 16:50:26 GMT):
Actually, I'm sure it's needed, and think that is why it is needed :-)

bansatya (Thu, 28 Nov 2019 07:03:55 GMT):
@sukalpomitra , you mean while mobile agent create connection with passport issuer, it will ask for some short of details? Can you please elaborate. How the trust will be built?

sukalpomitra (Thu, 28 Nov 2019 07:09:57 GMT):
The trust is in several layers. First there is trust in cryptography and blockchain. You hold the private key to prove that you own the public key that is in the did doc. When the credential is issued to the holder. It is signed by the issuer's public key. When the holder sends this credential to the verifier, it counter signs with its own public key also. Next trust comes from human level, where the verifier in the proof request can add restriction that they only accept credentrial from issuer A or Issuer B or both issuer A and B

bansatya (Thu, 28 Nov 2019 07:12:47 GMT):
@sukalpomitra , I understand the cryptographical trust. But question is on human trust. How passport issuer will be sure that Alice who is asking for credential is Alice and not other people?

sukalpomitra (Thu, 28 Nov 2019 07:13:35 GMT):
because he trusts the issuer

sukalpomitra (Thu, 28 Nov 2019 07:15:36 GMT):
thats what I said, how in today's world we do it. For Indians, we have VFS office where we go and submit all the relevant documents. They can also may be able to submit through an e-app. once VFS does their verification, the same offline process that they do today, they now can either issue a physical passport or issue a verifiable credential

sukalpomitra (Thu, 28 Nov 2019 07:15:36 GMT):
thats what I said, how in today's world we do it. For Indians, we have VFS office where we go and submit all the relevant documents. They can also may be able to submit through an e-app. once VFS does their verification, the same offline process that they do today (police checks, biometric scan etc), they now can either issue a physical passport or issue a verifiable credential

utkmaheshwari245 (Thu, 28 Nov 2019 07:35:52 GMT):
i see - I actually tried creating a cred offer json on the fly - with just shemaId and defId - I am getting CommonInvalidStructure exception !

utkmaheshwari245 (Thu, 28 Nov 2019 07:36:06 GMT):
looks like cred offer has to be created

iampukar (Thu, 28 Nov 2019 09:59:06 GMT):
Has joined the channel.

Vincent_S (Thu, 28 Nov 2019 11:10:31 GMT):
Has joined the channel.

Vincent_S (Thu, 28 Nov 2019 11:10:35 GMT):
Hi everyone! Is it correct that currently only _larger, smaller than_ ZKP s for integers are possible in indy? Thanks very much for any responses in advance!

brentzundel (Thu, 28 Nov 2019 12:06:46 GMT):
Yes, this is currently true. Work is underway to greatly expand the scope of predicate proofs, but it will not be in place until someone next year.

Vincent_S (Thu, 28 Nov 2019 12:29:00 GMT):
Thank you very much for the helpful reply. :)

JeromeK (Thu, 28 Nov 2019 15:18:43 GMT):
Hey Guys Are they any other Payment-Methods beside sov-token?

JeromeK (Thu, 28 Nov 2019 15:18:43 GMT):
Hey Guys Are there any other Payment-Methods beside sov-token?

donqui (Thu, 28 Nov 2019 16:03:01 GMT):
there is null (libnullpay - https://github.com/hyperledger/indy-sdk/tree/master/libnullpay), like a mock for testing. Also there is a plugin API so you can write your own

sebastillar (Thu, 28 Nov 2019 20:58:32 GMT):
Thanks for your answer! Is there any other way to set up a network without docker?

gnarula (Fri, 29 Nov 2019 13:45:15 GMT):
Has joined the channel.

gnarula (Fri, 29 Nov 2019 13:54:34 GMT):
Hi! I'm looking for a way to do authorization and delegation and stumbled upon https://github.com/w3c/vc-data-model/issues/204 which led me to https://w3c-ccg.github.io/zcap-ld. Is there a plan to publish capabilities and their invocations in the indy ledger?

Yoroitchi (Fri, 29 Nov 2019 14:14:34 GMT):
Has joined the channel.

SigmaS 1 (Fri, 29 Nov 2019 14:18:25 GMT):
Has joined the channel.

Yoroitchi (Fri, 29 Nov 2019 14:23:18 GMT):
Hi everyone! Can someone explain to me how a verifier can interact with the ledger if it doesn't have a wallet?

swcurran (Fri, 29 Nov 2019 16:05:27 GMT):
Initial work has started on delegation in Indy as described here: https://github.com/hyperledger/aries-rfcs/tree/master/concepts/0104-delegatable-credentials

swcurran (Fri, 29 Nov 2019 16:08:58 GMT):
Not sure how you are using the term "wallet". A verifier has to have an agent that can receive a proof and execute the crypto to verify the correctness of the proof. As well, it should have as part of the agent the ability to connect to and read transactions from the ledger related to the proof - the credential definition to know the DID of the issuer, for example, and if the credential is revocable, the revocation registry entries to confirm the "proof of non-revocation" that the prover would have included.

AbhinavChawla (Fri, 29 Nov 2019 18:54:31 GMT):
Has joined the channel.

AbhinavChawla (Fri, 29 Nov 2019 18:54:32 GMT):
Hello, I'm looking to install indy-sdk on MacOS and hitting issues, can someone help me with it? ``` Reason: Incompatible library version: libindy.dylib requires version 21.0.0 or later, but libsodium.18.dylib provides version 20.0.0 ```

lepar (Fri, 29 Nov 2019 19:10:50 GMT):
Yeah update the libsodium library

AbhinavChawla (Fri, 29 Nov 2019 19:13:54 GMT):
But the indy-sdk requires `libsodium.18.dylib`

AbhinavChawla (Fri, 29 Nov 2019 19:15:24 GMT):
I get error like: ``` /indy-sdk/build/Release/indynodejs.node, 1): Library not loaded: /usr/local/opt/libsodium/lib/libsodium.18.dylib Referenced from: /usr/local/lib/libindy.dylib Reason: Incompatible library version: libindy.dylib requires version 21.0.0 or later, but libsodium.18.dylib provides version 20.0.0 ```

lepar (Fri, 29 Nov 2019 19:16:13 GMT):
Oh, not sure

AbhinavChawla (Fri, 29 Nov 2019 19:23:52 GMT):
Ah, okay. Anyone else you know who might be able to help me out?

tomislav (Fri, 29 Nov 2019 19:51:41 GMT):
Are you running the install script for MacOS or doing it step by step? @AbhinavChawla

AbhinavChawla (Fri, 29 Nov 2019 20:12:58 GMT):
I used `npm install indy-sdk` which installed fine (after I set up the pre-reqs, and downloaded lib-indy from https://repo.sovrin.org/macos/libindy/)

AbhinavChawla (Fri, 29 Nov 2019 20:13:46 GMT):
I get the error when I try to do `require(indy-sdk)` in a .js file, and run it as `node file.js`

tomislav (Fri, 29 Nov 2019 21:30:40 GMT):
We're seeing some issues with the pre-built binaries. Can you try running the script `mac.build.sh` from `libindy` folder? Once complete, just copy `target/debug/libindy.dylib` to /usr/local/lib`

tomislav (Fri, 29 Nov 2019 21:30:40 GMT):
We're seeing some issues with the pre-built binaries. Can you try running the script `mac.build.sh` from `libindy` folder? Once complete, just copy `target/debug/libindy.dylib` to `/usr/local/lib`

RohitShitre (Sat, 30 Nov 2019 16:05:49 GMT):
anki

ItsOmerShafiq (Sun, 01 Dec 2019 22:28:01 GMT):
Has joined the channel.

AbhinavChawla (Mon, 02 Dec 2019 05:54:36 GMT):
Thanks, I was able to build using the script. When I run the test python code, I get errors: ``` Error: dlopen(/Users/abhinav/Documents/Work/libindy/libindy_1.12.0/node_modules/indy-sdk/build/Release/indynodejs.node, 1): Library not loaded: /usr/local/Cellar/openssl@1.1/1.1.1d/lib/libcrypto.1.1.dylib Referenced from: /usr/local/Cellar/openssl@1.1/1.1.1d/lib/libssl.1.0.0.dylib Reason: image not found ``` It's stuck on a fixed version of openssl for some reason, is there a guide how to download the correct version?

DilshodRakhmanov (Mon, 02 Dec 2019 07:21:00 GMT):
Has joined the channel.

DilshodRakhmanov (Mon, 02 Dec 2019 07:21:27 GMT):
Hi Everyone, What is the error: "_load_cdll: Can't load libindy: dlopen(libindy.dylib, 6): image not found"

gnarula (Mon, 02 Dec 2019 09:10:39 GMT):
Thanks!

VithushanJegatheeswaran (Mon, 02 Dec 2019 13:33:16 GMT):
What role is the blockchain/pool playing with DID's ?

gnarula (Mon, 02 Dec 2019 13:40:28 GMT):
@swcurran to follow up, are there plans to log the _use_ of a delegated credential in the ledger for audibility?

VithushanJegatheeswaran (Mon, 02 Dec 2019 13:46:56 GMT):
a user can create multiple DID's, but can he use the same verifiable crendetial across all his DID's ?

VithushanJegatheeswaran (Mon, 02 Dec 2019 13:48:26 GMT):
can a VC get stolen/ can someone else use your VC in combination with their DID ?

gnarula (Mon, 02 Dec 2019 13:55:15 GMT):
yes, and the secret sauce that makes it possible is the user being able to prove it knows a secret that the issuer blindly signed. idk how up to date this is but it explains the details pretty well https://sovrin.org/wp-content/uploads/2019/01/How-DIDs-Keys-Credentials-and-Agents-Work-Together-in-Sovrin-131118.pdf

VithushanJegatheeswaran (Mon, 02 Dec 2019 13:55:50 GMT):
thank you :)

lijiachuan (Mon, 02 Dec 2019 14:42:39 GMT):
#Aries Hi all, I am a new one for Aries, and currently I am trying to setup one cloud agent with Aries Cloud Agent Python, from this document https://github.com/hyperledger/aries-cloudagent-python/blob/master/DevReadMe.md#docker, am trying to provision a new agent with "scripts/run_docker provision" command, but as there is no sample command script, so I could not figure out how should I format a correct command for this. Can anyone give me one suggestion for this sample provision command? And whether we need to provision a new agent then we can use start command to start it? Kindly advise. Thanks.

swcurran (Mon, 02 Dec 2019 16:58:57 GMT):
^^^ @andrew.whitehead @ianco Can you help?

heenas06 (Tue, 03 Dec 2019 06:57:11 GMT):
Hi All I am trying to do indy-sdk setup & while doing "npm run start" command i am getting this issue....will attach the snap....

heenas06 (Tue, 03 Dec 2019 06:57:34 GMT):

Screenshot .png

Chem (Tue, 03 Dec 2019 06:59:16 GMT):
Hi, How credential verification is done in indy?

manavgupta (Tue, 03 Dec 2019 12:05:13 GMT):
hi @lynn.bendixsen How can we add a new TRUSTEE role on the ledger? we are using default pool genesis.

kumar89 (Tue, 03 Dec 2019 14:11:50 GMT):
Has joined the channel.

lynn.bendixsen (Tue, 03 Dec 2019 14:44:46 GMT):
1. Open your CLI, and connect to your pool 2. Open your wallet and use the default Trustee DID

lynn.bendixsen (Tue, 03 Dec 2019 14:44:46 GMT):
1. Open your CLI, and connect to your pool 2. Open your wallet and use the default Trustee DID 3. Did new seed= 4. ledger nym did= verkey= role=TRUSTEE

lynn.bendixsen (Tue, 03 Dec 2019 14:44:46 GMT):
1. Open your CLI, and connect to your pool 2. Open your wallet and use the default Trustee DID 3. indy-cli> did new seed= 4. indy-cli> ledger nym did= verkey= role=TRUSTEE

lynn.bendixsen (Tue, 03 Dec 2019 14:52:15 GMT):
If you get an error, post it here and I can help resolve it

dbluhm (Tue, 03 Dec 2019 19:21:54 GMT):
responded on #indy-sdk

sebastillar (Tue, 03 Dec 2019 22:40:29 GMT):
Hi everyone

sebastillar (Tue, 03 Dec 2019 22:41:00 GMT):
I'm trying with indy-cli with docker indy-pool image

sebastillar (Tue, 03 Dec 2019 22:41:27 GMT):
and, when I try to connect to sandbox pool, I get this response: Pool "sandbox" has not been connected.

sebastillar (Tue, 03 Dec 2019 22:41:27 GMT):
and, when I try to connect to sandbox pool, I get this response: Pool "sandbox" has not been connected.

sebastillar (Tue, 03 Dec 2019 22:41:27 GMT):
and, when I try to connect to sandbox pool, I get this response: _Pool "sandbox" has not been connected._

sebastillar (Tue, 03 Dec 2019 22:48:25 GMT):
I've already been tried with another "test" pool but got it same response

sebastillar (Tue, 03 Dec 2019 22:48:25 GMT):
I've already tried with another "test" pool but got it same response

lynn.bendixsen (Tue, 03 Dec 2019 23:49:26 GMT):
I suspect that the genesis file that you are using for your CLI does not match the genesis file of the docker indy-pool. They must be the same for you to be able to connect.

lynn.bendixsen (Tue, 03 Dec 2019 23:50:21 GMT):
If the genesis files match, then I would look for a network issue next.

manavgupta (Wed, 04 Dec 2019 03:34:55 GMT):
where will the deafult TRUSTEE did be used?

eduelias (Wed, 04 Dec 2019 09:14:19 GMT):
Hi. sorry to bug you with that again, but is there anyway I can do an "or" between the two restriction conditions?

sukalpomitra (Wed, 04 Dec 2019 09:16:20 GMT):
the list means its an or

sukalpomitra (Wed, 04 Dec 2019 09:16:39 GMT):
so u can choose the values from either cred A or cred B

eduelias (Wed, 04 Dec 2019 09:17:28 GMT):
Is there any documents with this information so I can learn more about it? (Thanks for your attention so far, you're being so helpful! Thanks!)

sukalpomitra (Wed, 04 Dec 2019 09:17:52 GMT):
I dont know of such maybe @swcurran can help

Philippede 2 (Wed, 04 Dec 2019 09:26:12 GMT):
Is it possible to get attribute values from a proof in vcx? For example when I need to know the name of a person. The attribute values in a proof are encrypted and I didn't find a way to decrypt them.

PatrikStas (Wed, 04 Dec 2019 10:19:50 GMT):
I think you are just overlooking where they are =) It's a bit mess, but here's how you'd extract proof data in JS. The parameter to this function is serialized vcx proof. ``` function extractProofData (serProof) { if ([1, 2].includes(serProof.data.proof_state) && serProof.data.state === 4) { const proof = JSON.parse(serProof.data.proof.libindy_proof) const attrs = proof.requested_proof.revealed_attrs const data = Object.keys(attrs).map(attr => { const result = {} result[attr] = attrs[attr].raw return result }) return data } return null } ```

sebastillar (Wed, 04 Dec 2019 11:20:39 GMT):
the genesis file is the same indy-pool from docker

paul.sitoh (Wed, 04 Dec 2019 11:43:34 GMT):
Is there a sawtooth SDK channel?

JeromeK (Wed, 04 Dec 2019 11:59:45 GMT):
https://chat.hyperledger.org/channel/sawtooth-sdk-dev

Philippede 2 (Wed, 04 Dec 2019 12:52:25 GMT):
Thank you! I can't believe I didn't see this...

eduelias (Wed, 04 Dec 2019 14:53:49 GMT):
Can someone tell which are the causes of a "IndyError: AnoncredsProofRejected"? I'm adding a new attribute to an existing credential definition, updating it's version and "verifierVerifyProof" it ... it's returning this error.

eduelias (Wed, 04 Dec 2019 15:20:24 GMT):
The changes Iv

eduelias (Wed, 04 Dec 2019 15:20:24 GMT):
The changes I've made are here: https://github.com/eduelias/indy-sdk/commit/6f7b6807a42ed77b82c4a09a3deabf150ea6f169

lynn.bendixsen (Wed, 04 Dec 2019 16:11:58 GMT):
in the CLI you must "activate" the DID you will use to send transactions to the ledger. Whichever DID is activated will be the DID that signs all of the transactions. Like this: > did use

sebastillar (Wed, 04 Dec 2019 17:13:10 GMT):
Which are the steps one by one to install and running python wrapper? Thanjs!:pray:

VithushanJegatheeswaran (Wed, 04 Dec 2019 17:45:00 GMT):
Is Hyperledger Indy using "JSON" or "JSON-LD" ? https://www.w3.org/TR/did-core/#json-ld

cam-parra (Wed, 04 Dec 2019 17:47:03 GMT):
JSON currently but there are groups working supporting on JSON-LD

cam-parra (Wed, 04 Dec 2019 17:47:27 GMT):
@burdettadam ^

VithushanJegatheeswaran (Wed, 04 Dec 2019 17:49:16 GMT):
@cam-parra ok nice. Do you knwo if Indy uses something like a DID-Document ? https://www.w3.org/TR/did-core/#example-2-minimal-self-managed-did-document

sebastillar (Wed, 04 Dec 2019 19:35:43 GMT):
docker container it is not render ok in browser

sebastillar (Wed, 04 Dec 2019 19:36:07 GMT):

any suggestions about this issue?

dbluhm (Wed, 04 Dec 2019 22:41:35 GMT):
@sebastillar The docker pool in indy-sdk are speak ZMQ, not HTTP. There is nothing to point your browser at from the docker pool. What were you expecting? You might be looking for a different Indy network docker setup such as BC Gov's images which do host a simple web interface that shows activity and the contents of the domain ledger.

dbluhm (Wed, 04 Dec 2019 22:41:35 GMT):
@sebastillar The docker pool in indy-sdk speaks ZMQ, not HTTP. There is nothing to point your browser at from the docker pool. What were you expecting? You might be looking for a different Indy network docker setup such as BC Gov's images which do host a simple web interface that shows activity and the contents of the domain ledger.

swcurran (Wed, 04 Dec 2019 22:43:35 GMT):
If so, you wouldn't be looking at port 9701, which is the a node in the pool. For reference, the BC Gov von-network has a web server ledger browser at http://localhost:9000.

sebastillar (Wed, 04 Dec 2019 22:57:41 GMT):

localhost:9000

swcurran (Wed, 04 Dec 2019 22:57:59 GMT):
Are you using von-network?

sebastillar (Wed, 04 Dec 2019 22:58:04 GMT):
@dbluhm @swcurran Thanks for your anwsers!

sebastillar (Wed, 04 Dec 2019 22:58:53 GMT):
not, I'm trying with indy-sdk

sebastillar (Wed, 04 Dec 2019 22:58:57 GMT):
1) Starting the test pool on localhost Start the pool of local nodes on 127.0.0.1:9701-9708 with Docker by running: docker build -f ci/indy-pool.dockerfile -t indy_pool . docker run -itd -p 9701-9708:9701-9708 indy_pool

sebastillar (Wed, 04 Dec 2019 22:59:30 GMT):
I'm following indy-sdk doc

swcurran (Wed, 04 Dec 2019 23:00:09 GMT):
OK - so there is nothing to look at through a browser.

sebastillar (Wed, 04 Dec 2019 23:00:23 GMT):
ahhhhhhhh

sebastillar (Wed, 04 Dec 2019 23:00:37 GMT):
thanks Stephen!

swcurran (Wed, 04 Dec 2019 23:01:21 GMT):
There are nodes running, given you got something. But nothing useful will come out except through talking to the nodes with a cli.

sebastillar (Wed, 04 Dec 2019 23:01:46 GMT):
How can I do interact with pool? Only with cli?

swcurran (Wed, 04 Dec 2019 23:02:20 GMT):
The cli (which has libindy in it) or an application with libindy embedded.

sebastillar (Wed, 04 Dec 2019 23:03:24 GMT):
so, for set up our dev environment, do you recommend to use a Von Network?

swcurran (Wed, 04 Dec 2019 23:05:37 GMT):
We built von-network because it makes our lives dead simple :-). - It spins up both the nodes with one command, - has a ledger browser so I can inspect things, - allows me to make DIDs, - allows me to have a url for the genesis file for the ledger.

swcurran (Wed, 04 Dec 2019 23:06:08 GMT):
We also have a ledger here that works the same: http://test.bcovrin.vonx.io/

swcurran (Wed, 04 Dec 2019 23:07:34 GMT):
We use that with devs by having them: - use the genesis file for it to connect - generate a random DID on each startup, so that the DID and related entities (schema, cred def, etc.) are unique. That saves each dev having to run their own ledger at all.

swcurran (Wed, 04 Dec 2019 23:08:26 GMT):
It's a 4 node ledger running on Digital Ocean. Not decentralized, gets updated to new versions from time to time.

swcurran (Wed, 04 Dec 2019 23:09:08 GMT):
But it has a known Trustee that can write to the ledger so no need to worry about that until you need.

sebastillar (Wed, 04 Dec 2019 23:10:35 GMT):
and how can I to communicate with this Trustee?

sebastillar (Wed, 04 Dec 2019 23:13:18 GMT):
and How can I connect this test network to an Aries Cloud Agent?

sebastillar (Wed, 04 Dec 2019 23:16:24 GMT):
I really appreciate your help. We are developing an ssi project to obtain our degree but we are stuck in the integration with indy

swcurran (Wed, 04 Dec 2019 23:18:08 GMT):
To start an indy agent, you have to pass in the genesis file that is specific to the network to which you want to connect.

ianco (Wed, 04 Dec 2019 23:18:19 GMT):
FYI you can use the ledger browser in `von-network` to browse the local indy nodes running from `indy-sdk`

swcurran (Wed, 04 Dec 2019 23:18:47 GMT):
If you run von-network, or use BCovrin test, you can just download the genesis file.

ianco (Wed, 04 Dec 2019 23:19:04 GMT):
In `von-network`: ```GENESIS_FILE=/pool_transactions_local_genesis.txt REGISTER_NEW_DIDS=true PORT=9000 python -m server.server ```

acaldas18 (Wed, 04 Dec 2019 23:21:56 GMT):
and how can I to communicate with this Trustee?

sebastillar (Wed, 04 Dec 2019 23:28:07 GMT):
@ianco thanks a lot!

sebastillar (Wed, 04 Dec 2019 23:44:45 GMT):
What should be the workflow to implement all requests and responses that are not implemented in the cli? That is, what would be the best way forward to implement a system that communicates with the ledger using Aries cloud agents?

swcurran (Thu, 05 Dec 2019 00:19:57 GMT):
Depends on your use case. If your app is going to run on other than a mobile device, you can look at ACA-Py - Aries Cloud Agent - Python.

manavgupta (Thu, 05 Dec 2019 03:35:10 GMT):
ok thank you so much, we tried that with java client and it worked and will try with Indy cli too

esplinr (Thu, 05 Dec 2019 16:06:26 GMT):
We are getting started with the identity implementers working group call: https://zoom.us/j/232861185

shemnon (Thu, 05 Dec 2019 18:06:26 GMT):
Has joined the channel.

pengyuc (Thu, 05 Dec 2019 21:12:49 GMT):
Has joined the channel.

esplinr (Thu, 05 Dec 2019 21:27:10 GMT):
Hyperledger has a new process for managing calendars. Administators of the mailing list group have the right to create calendar appointments. I requested to become an admin so that I could migrate the Contributors calls to the new calendar. Hopefully my efforts are helpful. It would be great if others are admins and can manage this process.

mtfk (Thu, 05 Dec 2019 22:54:27 GMT):
hi guys could anyone simple explain what for was meant non-secret records in the indy wallet? Is it for general data or some specific object related with identity?

tomislav (Thu, 05 Dec 2019 23:42:57 GMT):
general record data, that's not a secret

tomislav (Thu, 05 Dec 2019 23:43:51 GMT):
most aries implementation frameworks, utilize this API to build application business layer on top of it

donqui (Fri, 06 Dec 2019 07:58:09 GMT):
It can be used for general data if if it makes sense for that data to live in the wallet. The reason non-secrets API was introduces was to allow storage of application specific data that had a need to be safely stored and searchable. https://github.com/hyperledger/indy-sdk/blob/5f9144feab5b9e969389e325da47d7218f87081e/docs/design/003-wallet-storage/README.md

donqui (Fri, 06 Dec 2019 07:58:09 GMT):
It can be used for general data if if it makes sense for that data to live in the wallet. The reason non-secrets API was introduces was to allow storage of application specific data that had a need to be safely stored and searchable. https://github.com/hyperledger/indy-sdk/blob/5f9144feab5b9e969389e325da47d7218f87081e/docs/design/003-wallet-storage/README.md#non-secrets-api

donqui (Fri, 06 Dec 2019 07:58:09 GMT):
It can be used for general data if if it makes sense for that data to live in the wallet. The reason non-secrets API was introduces was to allow storage of application specific data (usually identity related) that had a need to be safely stored and searchable. https://github.com/hyperledger/indy-sdk/blob/5f9144feab5b9e969389e325da47d7218f87081e/docs/design/003-wallet-storage/README.md#non-secrets-api

mtfk (Fri, 06 Dec 2019 08:21:52 GMT):
does it mean that it is a simple version of the data vault? So I can store there what ever I like? For example my health data record?

donqui (Fri, 06 Dec 2019 09:43:28 GMT):
whatever you wish if it fits the model

lijiachuan (Sat, 07 Dec 2019 14:33:04 GMT):
hi, I am trying to register my new created Aries Cloud Agent to the VON network, I can see this agent's DID in agent start command line window, but when I try to register this agent in VON web page, verkey is required, but I could not find where can I know the verkey for this agent's DID. Does anyone can provide a guidance? Thanks a lot.

JeromeK (Mon, 09 Dec 2019 11:47:01 GMT):
Hey Here a nice overview over the whole Wallet Storage Design

JeromeK (Mon, 09 Dec 2019 11:47:01 GMT):
Hey Here a nice overview of the whole Wallet Storage Design

JeromeK (Mon, 09 Dec 2019 11:47:02 GMT):
https://hyperledger-indy.readthedocs.io/projects/sdk/en/latest/docs/design/003-wallet-storage/README.html#wallet-components

JeromeK (Mon, 09 Dec 2019 11:59:53 GMT):
if they are any more questions feel free to ask. :thumbsup:

moulay (Mon, 09 Dec 2019 22:52:22 GMT):
Has joined the channel.

moulay (Mon, 09 Dec 2019 22:52:24 GMT):
Hi guys i am using hyperledger indy, and i didn't found any library for react native to connect it with the network, would you please help me.

esplinr (Mon, 09 Dec 2019 23:01:30 GMT):
@moulay Most people would start with the JavaScript wrapper.

esplinr (Mon, 09 Dec 2019 23:01:50 GMT):
Starting the Indy Contributors call: https://zoom.us/j/615818107

moulay (Mon, 09 Dec 2019 23:05:34 GMT):
@esplinr thanks, there any starter project which may help me.

esplinr (Tue, 10 Dec 2019 00:05:19 GMT):
The Sovrin Connector App is React Native, but it's rather out-of-date and not very maintained: https://github.com/sovrin-foundation/connector-app/ You might get better advice in #aries-javascript

kaelJ (Tue, 10 Dec 2019 07:20:35 GMT):
Has joined the channel.

kaelJ (Tue, 10 Dec 2019 07:20:37 GMT):
have anyone faced this failed to get default targets for cc proposal: failed to discover peers: access denied ,i do need help

palra (Tue, 10 Dec 2019 09:33:36 GMT):
Has joined the channel.

JeromeK (Tue, 10 Dec 2019 10:24:05 GMT):
Hey Guys Does somebody has a sample on how to connect the libsovtoken(library) to the Wrapper (to create a payment-address)

lmtriet (Tue, 10 Dec 2019 13:14:14 GMT):
Has joined the channel.

lsl88 (Tue, 10 Dec 2019 13:51:35 GMT):
Hi! I am trying to follow the walthrough. Is there a similar guide but using indy-cli? Otherwhise, where shall I write the code of the examples? Thank you in advance :)

lsl88 (Tue, 10 Dec 2019 13:51:35 GMT):
Hi! I am trying to follow the walthrough. Is there a similar guide but using indy-cli in interactive mode? Otherwhise, where shall I write the code of the examples? Thank you in advance :)

cam-parra (Tue, 10 Dec 2019 15:56:12 GMT):
It was in the readme when I wrote it :)

esplinr (Wed, 11 Dec 2019 01:40:44 GMT):
Two announcements: 1. I posted a new demo video for Indy: https://www.youtube.com/watch?v=HapWIQdtMOA&list=PLRp0viTDxBWGLdZk0aamtahB9cpJGV7ZF&index=1 Key topics include a discussion of simulation tests we used to fix the instance change process, fixing zeroMQ auto-reconnection between validator nodes, a fix for using two attributes from the same credential, and an update on the most recent releases.

esplinr (Wed, 11 Dec 2019 01:43:05 GMT):
2. Next Tuesday, December 17, @ashcherbakov will be hosting a webinar about Indy Plenum: https://ssimeetup.org/hyperledger-indy-public-blockchain-node-alexander-shcherbakov-webinar-43/ This is a great opportunity to learn about the details of Indy Plenum from the project architect.

manavgupta (Wed, 11 Dec 2019 06:07:57 GMT):
Hi @lynn.bendixsen i want to pass two restrictions against one attribute while creating proof request. How can i create such proof request? Also tell me the way to create the proof accordingly

lsl88 (Wed, 11 Dec 2019 08:21:10 GMT):
I am trying to open this site but get 404: https://github.com/hyperledger/indy-node/blob/master/getting-started.md

JeromeK (Wed, 11 Dec 2019 08:48:58 GMT):
https://github.com/sovrin-foundation/libsovtoken#installing-indy-sdk

JeromeK (Wed, 11 Dec 2019 08:48:59 GMT):
here ?

JeromeK (Wed, 11 Dec 2019 08:49:48 GMT):
https://github.com/sovrin-foundation/libsovtoken/tree/master/samples

JeromeK (Wed, 11 Dec 2019 08:49:55 GMT):
no samples there :/

donqui (Wed, 11 Dec 2019 10:05:58 GMT):
it should be the same as with `nullpay`, a plugin is inited, the payment method is registered, and then you work with that

donqui (Wed, 11 Dec 2019 10:05:58 GMT):
it should be the same as with `nullpay: a plugin is loaded, than inited, the payment method is registered, and then you work with that

donqui (Wed, 11 Dec 2019 10:06:04 GMT):
maybethis can explain beter

donqui (Wed, 11 Dec 2019 10:06:04 GMT):
maybe this can explain better

donqui (Wed, 11 Dec 2019 10:06:05 GMT):
https://github.com/hyperledger/indy-sdk/blob/bc14c5b514dc1c76ce62dd7f6bf804120bf69f5e/libnullpay/tests/payments.rs#L487

JeromeK (Wed, 11 Dec 2019 12:39:59 GMT):
Okay Thanks I will have a look

JeromeK (Wed, 11 Dec 2019 12:40:32 GMT):
Indyscan down or is it some sort of maintainance?

JeromeK (Wed, 11 Dec 2019 12:40:32 GMT):
Indyscan down or is it some sort of maintenance?

mtfk (Wed, 11 Dec 2019 21:33:52 GMT):
is there any API exposing that storage to the outside. Let say I would like to put there some information which later on would like to share with someone. Is there any API allowing for that?

mtfk (Wed, 11 Dec 2019 21:34:14 GMT):
or exchange the data more less same way as I exchange VC

Silona (Wed, 11 Dec 2019 21:44:57 GMT):
Hello everyone - Do you know of a developer event that we would like to get some Hyperledger representation at? Please submit it here. The marketing committee will review them all as we go thru our budget for 2020 Thank you! - Silona

carl.diclementi (Wed, 11 Dec 2019 21:46:35 GMT):
Hi all. Any chance someone can point me to an example DID Document for a DID on Indy?

swcurran (Wed, 11 Dec 2019 21:51:15 GMT):
No there isn't. Right now, we're using it (as are others that use non-secrets) for connection records, and for protocol state objects only. AFAIK, no one is using it for storing something they want to share via a protocol. The general architecture is that there is a controller related to an agent instance that handles the business rules that control the agent. Any non-agent-specific data would be stored by the controller - which is likely what you are talking about. That separation will likely be further emphasized when we got to KMS (keys/ key like objects only, with metadata) and separate "protected storage" for other agent accessed data.

Silona (Wed, 11 Dec 2019 22:01:57 GMT):
https://wiki.hyperledger.org/display/Marketing/Developer+Events

manavgupta (Thu, 12 Dec 2019 05:56:09 GMT):
How can i retrieve all the credentials stored in wallet?

eduelias (Thu, 12 Dec 2019 10:45:39 GMT):
Is there anyway I can relate a taaAcceptance to the TAA transaction itself?

JeromeK (Thu, 12 Dec 2019 11:54:33 GMT):
await Anoncreds.get_credentials(wallet_handle, {})

JeromeK (Thu, 12 Dec 2019 11:54:33 GMT):
https://github.com/hyperledger/indy-sdk/blob/7bce833f9720e7a0e534ea4c34c0376c961dded5/wrappers/python/indy/anoncreds.py#L889

Vincent_S (Thu, 12 Dec 2019 13:54:11 GMT):
Hello everyone! I have three major questions left after intensively trying out the How-To's, which I could not find an answer to. Any help on this would be much appreciated! 1. What exactly is a pool? It seems to be different from the network itself. 2. In the "credential" tutorial, what is the encoding for the encoded values in the credential? I could not find any pattern that makes sense to me, so could somebody please give me some information on this? 3. When creating Steward, what is its seed for? And why does it have to have a certain form? I tried different variations, and only the seeds as in the tutorial seem to run (with a little variation). Thank you!

carl.diclementi (Thu, 12 Dec 2019 15:25:24 GMT):
For anyone else who may be looking for this, you can find one here: https://uniresolver.io/

lynn.bendixsen (Thu, 12 Dec 2019 21:27:26 GMT):
@manavgupta Last I heard there were issues related to this making it so that you can only pass 1 restriction. @ashcherbakov Do you know if this is working now?

lynn.bendixsen (Thu, 12 Dec 2019 21:53:51 GMT):
1. A "pool" is a set of nodes that participate in consensus with each other. The word "network" sometimes refers to the same thing, but in our definition also includes the agents and clients that connect to the pool. 2. I think the encoding is configurable and we suggest to use SHA256 (which I think is the default...). 3. The seed is a 32 character string used to generate your Steward DID and keys. It is a value that you can save in case your wallet is lost, or in case you want to allow someone else in your organization to act as the Steward. Using the seed, you can create the exact same DID and keys in multiple wallets. I am not sure why the original developers chose 32 characters as the requirement, but it probably has a lot to do with security and consistency for the ledger.

jethrojones (Thu, 12 Dec 2019 23:11:28 GMT):
Has joined the channel.

manavgupta (Fri, 13 Dec 2019 04:30:08 GMT):
well i tried passing multiple restriction it does not allow in 1.10 version of indy

Yoroitchi (Fri, 13 Dec 2019 10:24:58 GMT):
Thank you for your answer it was exactly what i wanted to know

lijiachuan (Fri, 13 Dec 2019 15:02:43 GMT):
hi All, I am trying to setup a basic Aries dotnet framework in my machine, I followed this document https://github.com/hyperledger/aries-framework-dotnet, installed .NET Core, created a new webapp, added the framework dependencies, registered the agent middleware, but when I start the app, below error message is shown: InvalidOperationException: No service for type 'Hyperledger.Aries.AspNetCore.AgentMiddleware' has been registered. Does anyone encounter this error before? Kindly advise .Thanks

lynn.bendixsen (Fri, 13 Dec 2019 17:17:12 GMT):
I am assuming that when you say "restrictions" you are referring to the "predicates". For now I think you can send 2 calls. For example: Call 1: age > 18 Call 2: age < 65 Be careful with taking this method to the extreme, though, because you can end up defeating the purpose by making enough calls to narrow it down to where you will know the exact age. Here is more information. It doesn't answer your question or solve your problem but it might be a useful reference. https://github.com/hyperledger/aries-rfcs/tree/master/features/0037-present-proof#presentation-preview

lynn.bendixsen (Fri, 13 Dec 2019 17:18:09 GMT):
https://github.com/hyperledger/indy-sdk/tree/master/docs/design/002-anoncreds

barrysieg (Sun, 15 Dec 2019 03:32:13 GMT):
Has joined the channel.

barrysieg (Sun, 15 Dec 2019 13:31:29 GMT):
@swcurran @ianco Hi all, I was able to go through the How-To steps and got it working. There seem to be no clarity deploying TOB. Is there any go to guide for deploying TOB for production?

swcurran (Sun, 15 Dec 2019 17:24:44 GMT):
We have lots of experience in doing that. Notably, there is a lot of configuration and code for deploying to an OpenShift (Kubernetes) production environment. However, a lot of the information depends on the production environment to which you are deploying. Let us know some more about what you are doing we can likely point you to or help you with the task.

manavgupta (Mon, 16 Dec 2019 04:42:05 GMT):
hi @lynn.bendixsen Let say indy network is Up and now i am adding a new steward who will have their own node. My understanding is that the pool_transaction_genesis would have to be updated to include this steward. How does this file get updated now on each and every client, who presently has the old pool_transaction_genesis, just wondering...

manavgupta (Mon, 16 Dec 2019 04:42:05 GMT):
hi @lynn.bendixsen Let say indy network is Up and now i am adding a new steward who will have their own node. My understanding is that the pool_transaction_genesis would have to be updated to include this steward. How does this file get updated now on each and every client, who presently has the old pool_transaction_genesis, just wondering... Or is there even the need for this updated file to be available to all or they can continue with old file.

steveLiuu (Mon, 16 Dec 2019 06:12:37 GMT):
Has joined the channel.

manavgupta (Mon, 16 Dec 2019 06:43:28 GMT):
I change the logging level in indy-node/indy_common/config.py but this does not change the logging level when I run the Indy nodes. I want to see all logs from all the different log levels for Indy Nodes. Please advise what to do...

manavgupta (Mon, 16 Dec 2019 06:44:37 GMT):
No i am not referring to predicates in that. I am actually talking about the restrictions given in attributes in the Job Request

Deepalisaraswat (Mon, 16 Dec 2019 10:45:08 GMT):
Has joined the channel.

Deepalisaraswat (Mon, 16 Dec 2019 10:45:11 GMT):
Hey, Is there any direct integration possible between hyperledger fabric and hyperledger indy?

ozbek (Mon, 16 Dec 2019 12:06:35 GMT):
Has joined the channel.

easeev (Mon, 16 Dec 2019 12:15:27 GMT):
Has left the channel.

tomislav (Mon, 16 Dec 2019 13:25:40 GMT):
Not yet. There are design conversations and plans to do this, but no specific work yet, AFAIK

jrallen (Mon, 16 Dec 2019 13:31:37 GMT):
Has joined the channel.

esplinr (Mon, 16 Dec 2019 16:01:55 GMT):
Starting the Indy Contributors call: https://zoom.us/j/615818107

tsmca (Mon, 16 Dec 2019 16:36:09 GMT):
Has joined the channel.

tsmca (Mon, 16 Dec 2019 16:36:13 GMT):
Hi guys, does someone know what the issue could be: I'm calling the Anoncreds.store_credential and I receive an error with "Wallet item not found"

ianco (Mon, 16 Dec 2019 17:30:41 GMT):
The credential definition info in your wallet is out of sync with the credential definition on the ledger

esplinr (Tue, 17 Dec 2019 17:03:50 GMT):
Alex Shcherbakov is starting a deep dive into Indy's Plenum ledger now: https://ssimeetup.org/hyperledger-indy-public-blockchain-node-alexander-shcherbakov-webinar-43/

esplinr (Tue, 17 Dec 2019 18:00:19 GMT):
Did you figure this out? The taaAcceptance date is in the signature of transaction you are trying to author, which also contains a hash to the TAA. There is a ledger transaction to look up the TAA text from the hash. If you store the taaAcceptance date somewhere for future use, you need to also store the TAA hash with it.

ketankumar.patil (Wed, 18 Dec 2019 07:39:06 GMT):
Has joined the channel.

ketankumar.patil (Wed, 18 Dec 2019 07:39:10 GMT):
Hi All, I am trying to setup indy wallet with external storage using postgres plugin by following the steps mentioned in https://github.com/hyperledger/indy-sdk/tree/v1.13.0/experimental/plugins/postgres_storage *While performing cargo build for 'indy-sdk/experimental/plugins/postgres_storage', I am getting following errors.* 1) /home/ubuntu/.cargo/....../ursa-0.1.1/src/signatures/ed25519.rs:27:21 Keypair::generate(&mut rng) - the trait `rand_core::CryptoRng` is not implemented for `rand_chacha::ChaChaRng` 2) /home/ubuntu/.cargo/....../ursa-0.1.1/src/signatures/ed25519.rs:27:21 Keypair::generate(&mut rng) - the trait `rand_core::RngCore` is not implemented for `rand_chacha::ChaChaRng` 3) /home/ubuntu/.cargo/...../ursa-0.1.1/src/signatures/ed25519.rs:35:17 Keypair::generate(&mut rng) - the trait `rand_core::CryptoRng` is not implemented for `rand::rngs::OsRng 4) /home/ubuntu/.cargo/...../ursa-0.1.1/src/signatures/ed25519.rs:35:17 Keypair::generate(&mut rng) - the trait `rand_core::RngCore` is not implemented for `rand::rngs::OsRng` error: aborting due to 4 previous errors error: Could not compile `ursa` Does anyone encounter this error before? Kindly advise .Thanks

ianco (Wed, 18 Dec 2019 14:32:44 GMT):
I believe there's a PR that addresses this: https://github.com/hyperledger/indy-sdk/pull/1986

JHamilton (Wed, 18 Dec 2019 15:10:10 GMT):
Has joined the channel.

JHamilton (Wed, 18 Dec 2019 15:10:10 GMT):
Greetings Indy community! The new Blockchain Automation Framework lab https://github.com/hyperledger-labs/blockchain-automation-framework is working to provide production grade automated install of Indy. We plan to demo our progress on January 9th and we would appreciate any feedback, as well as any opportunities to collaborate and make BAF useful for the Indy community. https://lists.hyperledger.org/g/labs/viewevent?eventid=654211&calstart=2020-01-09

swcurran (Wed, 18 Dec 2019 15:16:19 GMT):
Sounds interesting. Is there a definition of what the Indy installer is for? A single node? A node on the Sovrin network? The SDK? Thanks

ianco (Wed, 18 Dec 2019 15:33:30 GMT):
Based on a quick glance it looks like it is deploying a set of Indy nodes.

ianco (Wed, 18 Dec 2019 15:33:54 GMT):
It would be interesting to add Aries agents to the mix!

JHamilton (Wed, 18 Dec 2019 15:39:01 GMT):
sorry in meetings and not a good multitasker :) yes this is for deploying a new multi-node Indy network

swcurran (Wed, 18 Dec 2019 15:51:25 GMT):
Have you looked at what we have in von-network?

aaronr (Wed, 18 Dec 2019 15:53:46 GMT):
Not sure if this is the right place to ask this, would appreciate pointers in the right direction if it is not. Consider a scenario where I have mutliple verifiable credentials that are all based on the same credential schema, possibly all from the same issuer (so cred def for all could be the same). Perhaps like Microsoft skills certifications or inspections issued by a city government. And I want to submit all of them or a subset of them in a proof response. Is that possible? All of the proof requests that I've seen examples of seem to only allow for the user to respond with one trait value from one qualifying credential for each requested trait in the proof request. Am I missing something? Or is this a scenario that is currently unsupported?

JHamilton (Wed, 18 Dec 2019 15:54:31 GMT):
i'm not familiar and full disclosure i'm very new to the blockchain space. will follow up with the team to take a look!

swcurran (Wed, 18 Dec 2019 16:00:35 GMT):
OK - here is the link. https://github.com/bcgov/von-network

swcurran (Wed, 18 Dec 2019 16:06:34 GMT):
Basically, with von-network, we release container images with versions of indy-node code, and have docker compose/bash scripts that make it easy to spin up dev instances that include a ledger browser web UI, and the images and ledger browser can be used when deploying to a cloud platform (e.g. AWS). The core idea is that you are running the entire network - all of the nodes. Decentralized production deployments are quite a different beast, because you are likely joining an existing network and running just one node of the network. Sovrin has the only real production instance of an Indy network.

harrywright (Wed, 18 Dec 2019 17:35:41 GMT):
Has joined the channel.

esplinr (Wed, 18 Dec 2019 19:54:34 GMT):
The recording from Alex's presentation at the SSI Meetup is posted. I highly recommend that everyone who is interested in Indy spends some time with this material: https://ssimeetup.org/hyperledger-indy-public-blockchain-node-alexander-shcherbakov-webinar-43/

Vincent_S (Thu, 19 Dec 2019 09:04:47 GMT):
Thank you very much for your answer!

Vincent_S (Thu, 19 Dec 2019 09:08:54 GMT):
I think that there are still some questions left on my side. As for: 2. I do not think that this is the encoding as used in the tutorials. It does not have a fixed length and the values are equal for numerical values. 3. Not all seeds work in the tutorial, but only those with specific properties - do you know why this is?

SigmaS 1 (Thu, 19 Dec 2019 11:04:58 GMT):
Hello @Vincent_S according to this(https://forum.sovrin.org/t/claim-values-encode/620/4) the tutorials have dummy values in encoding. About the seed you can find a cleary explanation here: https://stackoverflow.com/questions/59089178/hypelerdger-indy-node-seed-value Hope it helps!

Vincent_S (Thu, 19 Dec 2019 13:35:27 GMT):
Great, thanks.

eduelias (Thu, 19 Dec 2019 14:30:45 GMT):
Is there anyway I can get the last "block" number ... Like, when I'm building a GET_TX transaction, when I NEED to send a number, how can I know the most recent tx number? ... Maybe it should have a way to send "latest" and retrieve the last "block" or something like that.

esplinr (Thu, 19 Dec 2019 15:22:32 GMT):
The Identity Implementers WG call starts in 45 minutes. Please note that the Zoom link has changed. The correct link is in the community calendar: https://zoom.us/j/244779296

DavoB (Thu, 19 Dec 2019 20:44:56 GMT):
Has joined the channel.

DavoB (Thu, 19 Dec 2019 20:44:56 GMT):
Anyone know of any working Android example code that connects to an Indy pool? I have an Indy pool set up and can connect to it via the Indy-CLI, the Python example, the NodeJS example, but can't connect to it via Android. I've started based on the example at https://github.com/ap050492/Hyperledger-Indy-Wallet---Android . That works for creating a local wallet but I can't connect it to that same Indy pool, even though I'm using the same genesis file. I can't get past "java.util.concurrent.ExecutionException: org.hyperledger.indy.sdk.ledger.TimeoutException: Timeout happens for ledger operation."

esplinr (Thu, 19 Dec 2019 22:44:48 GMT):
Evernym's last demo video for 2019 shows the new documentation for the changes to the Transaction Author Agreement, PBFT view change, empty attributes in credentials, and credentials with lots of attributes: https://www.youtube.com/watch?v=Y6BEWk-quBM&list=PLRp0viTDxBWGLdZk0aamtahB9cpJGV7ZF

FIM-BC (Thu, 19 Dec 2019 22:51:07 GMT):
Has joined the channel.

FIM-BC (Thu, 19 Dec 2019 22:51:09 GMT):
Hello there, I am experimenting with Indy and trying to understand how things work. Unfortunately I came across some questions where I need help to move along. 1) I have found C header files and plenty .rs files, what seems to be Rust. But where can I find the C source code of the libraries implementing all functions of the CLI? 2) I want to analyse which functions do execute transactions on the ledger and what has been written into it. I have a manually created network (without Docker) and wrapped them together into a pool in the CLI. I have heard that there is a command to just show the 'plain' content of the 3 ledgers? 3) There are three messages which appear in the log of each node on a regular basis and I would like to know what happens there a) Every 5 minutes there is a "Ledger [0|1|2] is not updated for 301 seconds, so its freshness state is going to be updated now" and right after that an "ordered batch request" b) I suppose the following message is also connected to the first ones, but I'm note sure: "Node1:0 set last ordered as (0, 7535)" I look forward for your reply and I have to thank you for your excellent job done so far!

Nikhilck (Fri, 20 Dec 2019 05:34:05 GMT):
Has joined the channel.

Nikhilck (Fri, 20 Dec 2019 05:34:10 GMT):
Hii All,

Nikhilck (Fri, 20 Dec 2019 05:35:47 GMT):
Hii prendsxcrvftgbyhnujmik,ol.;' ``` ```

Nikhilck (Fri, 20 Dec 2019 05:41:37 GMT):
Hii all,``` i am very new to Indy ``` when trying to build the network from the indy-ssivc-tutorial (https://github.com/IBM-Blockchain-Identity/indy-ssivc-tutorial) is throwing error``` `E: Version '1.2.237' for 'indy-plenum' was not found E: Version '1.2.297' for 'indy-node' was not found ERROR: Service 'client' failed to build: The command '/bin/sh -c apt-get update -y && apt-get install -y indy-plenum=${indy_plenum_ver} indy-anoncreds=${indy_anoncreds_ver} indy-node=${indy_node_ver} python3-indy-crypto=${python3_indy_crypto_ver} libindy-crypto=${indy_crypto_ver} vim' returned a non-zero code: 100` `````` I look forward for your reply and thanks ``` ``` ```

Nikhilck (Fri, 20 Dec 2019 05:41:37 GMT):
Hii all, i am very new to Indy ``` when trying to build the network from the indy-ssivc-tutorial (https://github.com/IBM-Blockchain-Identity/indy-ssivc-tutorial) is throwing error``` `E: Version '1.2.237' for 'indy-plenum' was not found E: Version '1.2.297' for 'indy-node' was not found ERROR: Service 'client' failed to build: The command '/bin/sh -c apt-get update -y && apt-get install -y indy-plenum=${indy_plenum_ver} indy-anoncreds=${indy_anoncreds_ver} indy-node=${indy_node_ver} python3-indy-crypto=${python3_indy_crypto_ver} libindy-crypto=${indy_crypto_ver} vim' returned a non-zero code: 100` `````` I look forward for your reply and thanks ``` ``` ```

Nikhilck (Fri, 20 Dec 2019 05:41:37 GMT):
Hii all, i am very new to Indy when trying to build the network from the indy-ssivc-tutorial (https://github.com/IBM-Blockchain-Identity/indy-ssivc-tutorial) is throwing error``` `E: Version '1.2.237' for 'indy-plenum' was not found E: Version '1.2.297' for 'indy-node' was not found ERROR: Service 'client' failed to build: The command '/bin/sh -c apt-get update -y && apt-get install -y indy-plenum=${indy_plenum_ver} indy-anoncreds=${indy_anoncreds_ver} indy-node=${indy_node_ver} python3-indy-crypto=${python3_indy_crypto_ver} libindy-crypto=${indy_crypto_ver} vim' returned a non-zero code: 100` `````` I look forward for your reply and thanks ``` ``` ```

eduelias (Fri, 20 Dec 2019 11:34:37 GMT):
Man, I cannot help you with all the questions but I'll try some: 1) Rust can compile to C "readable" libs. There's no C code, only wrappers to the DLL. 2) indyscan.io. I'm creating an explorer that you can download and read the block directly, but is still WIP. You can use this repo (https://github.com/eduelias/node-indy-request) to see how to send transactions directly to the block. Changing the ledgerId and "data" will return different transactions from different blocks.

FIM-BC (Fri, 20 Dec 2019 14:13:03 GMT):
Thank you, this did already help me I will try to implement your explorer at the beginning of next year.

eduelias (Fri, 20 Dec 2019 14:16:17 GMT):
As soon as I deploy it I'll let you know! :D

lijiachuan (Fri, 20 Dec 2019 15:01:34 GMT):
Hello Team, may I know where can I find the tutorials about how to use Aries DotNet Framework? From below website, only the initial setup of the environment seems like with latest Aries framework information, but in the sample folder's code is still using AgentFramework's version of code, so may I have anyone's instruction about after setup one local environment, how can I make this agent work with the indy test network? Thanks a lot. https://github.com/hyperledger/aries-framework-dotnet

jljordan_bcgov (Sun, 22 Dec 2019 19:12:00 GMT):
Go to the #aries channel for those types of question I suggest

ultimo2020 (Thu, 26 Dec 2019 07:31:37 GMT):
Has joined the channel.

manavgupta (Thu, 26 Dec 2019 10:29:29 GMT):
Hi @lynn.bendixsen In the file verifier.rs the following equality is checked "c_hver == proof.aggregated_proof.c_hash" . Where (which file) can we find the code of how is c_hash created. The issue we are facing is: When we send combined proof of multiple credentials from android (Alice on Android) to unix (Acme on Ubuntu) the above equality fails. It does not fail if both are on Ubuntu.

ultimo2020 (Fri, 27 Dec 2019 06:58:31 GMT):
Hi everyone. I am looking for a use case online or a test where Indy can protect identity of the holder with for example Anonimity? How can we be sure that the issuer does not compromise the identity or knows all the details. Can we use the DID paradigm to totally protect the identity holder in the future?

biligunb (Fri, 27 Dec 2019 09:56:14 GMT):
Has joined the channel.

biligunb (Fri, 27 Dec 2019 10:20:10 GMT):
Hi guys I am a Fabric developer. Looking to integrate Fabric with Indy. Is there any tutorial or example projects that I can look over and try?

ultimo2020 (Sat, 28 Dec 2019 13:15:24 GMT):
Greetings. Does anyone have a working online start demo for a VPS , to test Indy on a VPS server? I would like to test and see if the identity can be protected (like anonymity) from everyone? Thanks in advance.

biligunb (Mon, 30 Dec 2019 07:47:47 GMT):
@swcurran Hi. I got 1 sdk related question. Can I use indy-sdk without indy-agent to connect with indy-node?

VithushanJegatheeswaran (Mon, 30 Dec 2019 13:34:34 GMT):
Hey guys, how can I make sure that a user uses his verifiable claim with only one of his DIDs. If a user has

VithushanJegatheeswaran (Mon, 30 Dec 2019 13:34:34 GMT):
Hey guys, how can I make sure that a user uses his verifiable claim with only one of his DIDs. For example: If a user has three DIDs and a verifiable claim. He is able to use the same VC with all of his DIDs.

ultimo2020 (Mon, 30 Dec 2019 14:00:57 GMT):
Hi everyone. I have downloaded the https://github.com/hyperledger/indy-node/blob/master/environment/docker/pool/README.md and tried to start the pool, but I get the error User with UID 0 Problem.

ultimo2020 (Mon, 30 Dec 2019 14:00:57 GMT):
Hi everyone. I have downloaded the https://github.com/hyperledger/indy-node/blob/master/environment/docker/pool/README.md and tried to start the pool, but I get the error User with UID 0 Problem. Did anyone encounte the same issue on Ubuntu 16.04 ?

ultimo2020 (Mon, 30 Dec 2019 14:00:57 GMT):
Hi everyone. I have downloaded the https://github.com/hyperledger/indy-node/blob/master/environment/docker/pool/README.md and tried to start the pool, but I get the error User with UID 0 Problem. Did anyone encounte the same issue on Ubuntu 16.04 ? Does anyone has a working demo or something to start with on a VPS? What are you guys using here for the infrastructure?

swcurran (Mon, 30 Dec 2019 23:06:21 GMT):
Yes, you can do that. Not sure the use case you would have for that, but it can be done.

swcurran (Mon, 30 Dec 2019 23:12:34 GMT):
That is expected behaviour. A user is expected to have a DID per connection, so that his connections are unable to correlate his DIDs. You should look at the Indy "anoncreds" (Anonymous Credentials) code and docs to learn about the "link secret" that is used by an identity when they are issued credentials, and that they use when proving the claims from the credentials. A verifier can see that the same link secret is used across the claims within a proof. Basically - DIDs are used for two things. First, a credential issuer must have a DID on the ledger so that a verifier can determine who issued the credential from which came the claims in a proof. Second, DIDs are used between entities to communicate using a secure channel. Beyond the issuer's DID there are no other DIDs involved in a verifiable credential exchange. That's possible through the use of ZKPs in Indy - the "anonymous" part.

swcurran (Mon, 30 Dec 2019 23:13:52 GMT):
We're running dev networks on Digital Ocean using the von-network (https://github.com/bcgov/von-network). Using that makes it pretty easy to spin up a shared network on a VPS.

ultimo2020 (Tue, 31 Dec 2019 06:30:02 GMT):
Hi @swcurran Thanks for the reply. I will try the von network. Do you have maybe an example app with the demonstration of the indy-sdk? Or does it come included with the von github clone? What is the difference between the von github and the official Indy hyperledger education github?

lynn.bendixsen (Tue, 31 Dec 2019 15:23:01 GMT):
I don't know. @MALodder might know...

swcurran (Tue, 31 Dec 2019 16:11:00 GMT):
von-network is based on indy-node. We have built container images that embedded indy-node, indy-sdk and some utilities (notably, a web ledger browser), and then von-network uses that image in spinning up a network. You can see an example of one running at Digital Ocean here - http://dev.bcovrin.vonx.io/

swcurran (Tue, 31 Dec 2019 16:12:35 GMT):
Not sure which official Indy github you are talking about. If you specifically mean the hyperledger/education repo, the indy components in that are snapshots of von-network from a long time ago.

LedgerXYZ (Tue, 31 Dec 2019 16:18:34 GMT):
Has joined the channel.

MALodder (Tue, 31 Dec 2019 17:37:12 GMT):
C_hash is created in prover.rs

MALodder (Tue, 31 Dec 2019 17:39:14 GMT):
https://github.com/hyperledger/ursa/blob/master/libursa/src/cl/prover.rs#L1101

ultimo2020 (Tue, 31 Dec 2019 19:11:08 GMT):
Hi. Thanks for the info. Yes I meant the hyperledger education github: https://github.com/hyperledger/education/blob/master/LFS171x/docs/introduction-to-hyperledger-indy.md

ultimo2020 (Tue, 31 Dec 2019 19:13:13 GMT):
Is the von-network the same like Sovrin foundation. Where do deveopers start? I will try to spin the von-network with nodes. Which indy-sdk or app samples are you using to build a PoC ? Any links would be appreciated. Thanks for your help and I wish you a good start in 2020.

swcurran (Wed, 01 Jan 2020 00:19:21 GMT):
von-network is using Indy - same as the Sovrin Foundation. For developers to start, we think a great starting spot is with Aries - https://github.com/hyperledger/aries-cloudagent-python/blob/master/docs/GettingStartedAriesDev/README.md It covers pretty much everything in indy-sdk (because that is embedded in Aries) and is geared to Devs wanting to build apps on Indy/Aries.

ultimo2020 (Wed, 01 Jan 2020 07:34:29 GMT):
Thanks for the update. Many useful Infos. I will certainly check it out. I wanted also to difference, Von network is maintained by the Gov of British Columbia and other, and Sovrin is maintained from Hyperledger native itself. Why don't you use Sovrin and do you have some demos and examples there? Thanks in advance.

jljordan_bcgov (Wed, 01 Jan 2020 08:09:01 GMT):
The networks we run on Digital ocean are only 4 node test and dev networks

jljordan_bcgov (Wed, 01 Jan 2020 08:09:28 GMT):
we do have transactions on the sovrin mainnet ledger for our production systems ... for example https://www.orgbook.gov.bc.ca/en/issuer/1

ultimo2020 (Wed, 01 Jan 2020 08:45:06 GMT):
Thanks. Will check it out. How can one develop / access the Sovrin testnet ledger?

ultimo2020 (Wed, 01 Jan 2020 13:36:13 GMT):
Hi. I have managed to get the VPS setup running from the https://github.com/bcgov/von-network instruction. I was wondering , is there some demo app for this setup, since this is only infrastructure of indy-node powered by bcgov and von ?

jljordan_bcgov (Wed, 01 Jan 2020 15:47:19 GMT):
https://github.com/hyperledger/aries-cloudagent-python/blob/master/docs/GettingStartedAriesDev/README.md

swcurran (Wed, 01 Jan 2020 20:30:57 GMT):
For Sovrin, I know there is a place to get the genesis file and places where you can create your own DID, so you can write to the ledger. Not exactly sure, but you can ask at Sovrin. They have a forum and Rocketchat.

VithushanJegatheeswaran (Thu, 02 Jan 2020 10:44:21 GMT):
Thank you so much for the detailed answer. I learned that the holder has a different DID for every connection. Could you help me with these question to understand DID's better: Can the holder generate more that one DID to connect to the same person ? Is it possible for the verifier to track all the For example: The holder connects to the issuer using different DIDs

VithushanJegatheeswaran (Thu, 02 Jan 2020 10:44:21 GMT):
Thank you so much for the detailed answer. I learned that the holder has a different DID for every connection. Could you help me with these question to understand DID's better: Can the holder generate more that one DID to connect to the same person ? For example: The holder connects to the issuer using different DIDs. And is it possible to revoke the crendential after the fist use ? (single use verifiable credential) Is it possible for the verifier to keep track of all verifiable claims that he has verified in order to prevent double use of a credential?

swcurran (Thu, 02 Jan 2020 15:30:51 GMT):
Good questions:

swcurran (Thu, 02 Jan 2020 15:30:51 GMT):
Good questions - some answers: - Yes, a holder could generate more than one DID per connection. Not normal, but there is no technology way to prevent that. - Single verification. No automatic way to do that. The issuer must be the one to revoke the credential. Thus, in such a scenario, the verifier would have to contact the issuer to let them know and request the revocation. The verifier must also get a claim that uniquely identifies the holder to tell the issuer. - Double use. Maybe, but not likely the way you are thinking. The verifier would have to get a claim that uniquely identifies the holder and retain that to check for double use. However, there is some work going on within Ursa that I've heard of that might support an anonymous version of that, which could be used for elections. I don't have details on that, or the timing of when it might be available.

swcurran (Thu, 02 Jan 2020 15:30:51 GMT):
Multiple DIDs: Yes. Not a typical practice, but nothing to prevent that in the technology. Single Use: Not an automatic way. A revocation must be done by the issuer, so for that to work, the verifier must notify the issuer to do the revocation. Further, he verifier would have to have received in the proof a claim that uniquely identifies the credential so the issuer would know the credential to revoke. Not easy. Double Use: Yes, the verifier can track the claims received, and if there is a claim in the proven data, they can prevent double use. I suspect you would like that prevention to be anonymous, and that is not yet supported. It might be possible with some coming enhancements in Ursa and Rich Schema updates, but no timed yet.

swcurran (Thu, 02 Jan 2020 15:44:11 GMT):
Multiple DIDs: Yes. Not a typical practice, but nothing to prevent that in the technology. Single Use: Not an automatic way. A revocation must be done by the issuer, so for that to work, the verifier must notify the issuer to do the revocation. Further, he verifier would have to have received in the proof a claim that uniquely identifies the credential so the issuer would know the credential to revoke. Not easy. Double Use: Yes, the verifier can track the claims received, and if there is a claim in the proven data, they can prevent double use. I suspect you would like that prevention to be anonymous, and that is not yet supported. It might be possible with some coming enhancements in Ursa and Rich Schema updates, but no timed yet.

VithushanJegatheeswaran (Thu, 02 Jan 2020 16:02:39 GMT):
"Yes, the verifier can track the claims received, and if there is a claim in the proven data, they can prevent double use" That sounds interesing to me For my bachelor thesis, I want to build a voting system using DID for authentication purposes. For that I need to make sure that a user ony uses his claim once. I can track the DIDs, who voted. But after voting, the user could make a new connection to the verifier (=> new DID) and use the same credential again. How should the verifyer track the claim ? Must the claim be uniquely identifiable ?

swcurran (Thu, 02 Jan 2020 16:22:40 GMT):
With current technology, yes, which means it is not anonymous, which is not good for a voting system. That likely means that some "trusted third party" is needed somewhere. As I mentioned, there is a plan to implement that in the future, but no timeline.

swcurran (Thu, 02 Jan 2020 16:24:03 GMT):
"that" - meaning a way to use a ZKP to detect the use of a credential so that double voting is detected and handled - e.g. only the last vote counts.

biligunb (Thu, 02 Jan 2020 23:06:19 GMT):
Tnx. I see that indy agents are in kind of a ? state. Due to Aries project. So until Aries is ready I thought maybe just use SDK directly

swcurran (Thu, 02 Jan 2020 23:08:35 GMT):
Aries agents are more mature versions of indy agents.

swcurran (Thu, 02 Jan 2020 23:10:37 GMT):
Wherever you would use Indy Agents, use Aries instead.

swcurran (Thu, 02 Jan 2020 23:11:23 GMT):
Using the indy-sdk directly is pretty difficult and that work is encapsulated in Aries agents.

biligunb (Fri, 03 Jan 2020 05:29:09 GMT):
But aries agent is kinda in DEVELOPMENT stage right? Doc says NOT MATURE ENOUGH until further notice.

ultimo2020 (Fri, 03 Jan 2020 09:05:22 GMT):
@swcurran Hi. I wanted to test and run the complete VON network on a VPS. I am reading the tutorial here.https://github.com/bcgov/greenlight/blob/master/docker/VONQuickStartGuide.md but I seem not to find the REPO that has all of the folders to start all of three instances. Which repo should I clone to start the: Open shell windows (Git Bash for instance) to your working copies of .../von-network, .../TheOrgBook/docker, and .../greenlight/docker.

lijiachuan (Sat, 04 Jan 2020 05:18:43 GMT):
hi all, I have one question that, if some of target users are not using mobile phone in their daily work, but we still need these individuals be participants for the Indy network, what is the best way for them to securely manage their wallets, public/private keys, DIDs, and VCs, etc? Kindly advise. Thanks.

jljordan_bcgov (Sat, 04 Jan 2020 22:04:32 GMT):
From an open source point of view Aries Cloud Agent - Python and Aries Framework DotNet are the most mature agents available. Work is progressing on Aries Framework Go as well. The first two agents mentioned (Python and DotNet) support Indy / ZKP based verifiable credential exchange protocols.

swcurran (Sat, 04 Jan 2020 22:07:16 GMT):
I'm interested in where it says that - I'll look. That would be bad advice, and worse would be to use raw indy-sdk.

swcurran (Sat, 04 Jan 2020 22:07:16 GMT):
I'm interested in where it says "Not mature enough" - I'll look. That would be bad advice, and worse would be to use raw indy-sdk.

swcurran (Sat, 04 Jan 2020 22:07:16 GMT):
I'm interested in where it says "Not mature enough" - I'll look. That would be bad advice, and much worse advice to suggest using raw indy-sdk.

PaulA (Mon, 06 Jan 2020 03:18:05 GMT):
@ultimo2020 Have you accessed Greenlight & TOB repos ? from there you should start the instances as stated on the "Quick Start Guide" section. Start the instance of each repo on a separate terminals.

daisuke1983 (Mon, 06 Jan 2020 05:07:50 GMT):
Has joined the channel.

ultimo2020 (Mon, 06 Jan 2020 06:32:39 GMT):
Hi. I have managed to start this one. https://github.com/hyperledger/education/blob/master/LFS171x/indy-material/nodejs/README.md

ketankumar.patil (Mon, 06 Jan 2020 07:29:40 GMT):
agent

ajayjadhav (Mon, 06 Jan 2020 10:45:26 GMT):
@swcurran , @esplinr - Can a holder herself verify her own credential on ledger to see if it is currently revoked by the issuer or not?

PaulA (Mon, 06 Jan 2020 13:06:53 GMT):
Sorry for the late response. I'm trying to deploy on openshift using both from catalogue and source deployment strategies. It uploads but failed to sync with staging net.

PaulA (Mon, 06 Jan 2020 13:08:53 GMT):
I am facing doubts regarding INDY's wallet. Where are credentials stored? are they naturally stored in a database e.g postgresql, and are they encrypted? On TOB where is the credential stored?

barrysieg (Mon, 06 Jan 2020 13:12:57 GMT):
Sorry for the late response. I'm trying to deploy on openshift using both from catalogue and source deployment strategies. It uploads but failed to sync with staging net. I am facing doubts regarding INDY's wallet. Where are credentials stored? are they naturally stored in a database e.g postgresql, and are they encrypted? On TOB where is the credential stored?

barrysieg (Mon, 06 Jan 2020 13:12:57 GMT):
Sorry for the late response. I'm trying to deploy on openshift by catalogue and source deployment strategies. It uploads but failed to sync with staging net. I am facing doubts regarding INDY's wallet. Where are credentials stored? are they naturally stored in a database e.g postgresql, and are they encrypted? On TOB where is the credential stored?

swcurran (Mon, 06 Jan 2020 15:56:13 GMT):
Yes. The holder constructs a "Proof of non-revocation" when they construct a proof, and can do so at any time. They can monitor what entries are put into the Rev. Reg and tell if their credential has been revoked. BTW - the verifier cannot do that - only the holder can tell that. Same is true when a proof is created.

RettPop (Mon, 06 Jan 2020 20:50:21 GMT):
Has joined the channel.

mattatkiva (Mon, 06 Jan 2020 23:14:47 GMT):
whats the reason a nonce is not valid: ``` console.log node_modules/winston/lib/winston/transports/console.js:79 [error] 102 - 01-06 20:11:59.950 : CommonInvalidStructure - { reqid: undefined, service: 'identity_wallet_service', url: 'POST /v1/credential/proof', exception: IndyError { name: 'IndyError', message: 'CommonInvalidStructure', indyCode: 113, indyName: 'CommonInvalidStructure', indyCurrentErrorJson: '{"backtrace":"","message":"Error: Invalid structure\\n Caused by: Invalid ProofRequest json has been passed\\n Caused by: Invalid nonce provided: 0271050207\\n"}', indyMessage: 'Error: Invalid structure\n Caused by: Invalid ProofRequest json has been passed\n Caused by: Invalid nonce provided: 0271050207\n', indyBacktrace: '' }, stack: 'IndyError: CommonInvalidStructure\n at Object.callback (/www/node_modules/indy-sdk/src/wrapIndyCallback.js:15:10)' }```

tplooker (Mon, 06 Jan 2020 23:55:14 GMT):
@mattatkiva I have encountered this issue before and I believe it is caused when the nonce starts with a `0` @tomislav do you remember us encountering this issue with agent framework?

mattatkiva (Tue, 07 Jan 2020 00:15:32 GMT):
@tplooker we are using indysdk nodejs wrapper. no agent code at this time

tplooker (Tue, 07 Jan 2020 00:20:39 GMT):
@mattatkiva yeap I understand, I think the problem is because your nonce starts with a -

tplooker (Tue, 07 Jan 2020 00:20:39 GMT):
@mattatkiva yeap I understand, I think the problem is because your nonce starts with a 0

tplooker (Tue, 07 Jan 2020 00:20:48 GMT):
Is this you nonce? 0271050207

tplooker (Tue, 07 Jan 2020 00:20:48 GMT):
Is this you nonce? `0271050207`

manavgupta (Tue, 07 Jan 2020 09:05:17 GMT):
we are trying to present proof by combining two credentials (i.e. two attributes from first credential and one attribute from second credential), but when the verifier verifies the proof it return false. I want to confirm that is that possible or not? we are using indy v1.10.1. @lynn.bendixsen check if you could help in that

PatrikStas (Tue, 07 Jan 2020 10:09:24 GMT):
Hi all, announcing Indyscan 3.x, now with fulltext search, sorting and improved UI. Check out [indyscan.io](https://indyscan.io) Also much easier to run on localhost!

ajayjadhav (Tue, 07 Jan 2020 10:42:51 GMT):
Thanks @swcurran. This is really useful.

SigmaS 1 (Tue, 07 Jan 2020 11:31:00 GMT):
Hi all, are all Indy nodes automatically stewards or observers?

mattatkiva (Tue, 07 Jan 2020 14:24:47 GMT):
yes

tomislav (Tue, 07 Jan 2020 15:02:41 GMT):
Use the `anoncreds.generate_nonce` method

cam-parra (Tue, 07 Jan 2020 15:40:02 GMT):
Yeah it looks like this is broken in the NodeJS wrapper as it returns a CString instead of a byte array.

cam-parra (Tue, 07 Jan 2020 15:41:09 GMT):
https://github.com/hyperledger/indy-sdk/blob/f708f496dd23ada341303db7a4e6a0a7e29a549f/wrappers/nodejs/src/index.js#L212

swcurran (Tue, 07 Jan 2020 16:53:17 GMT):
That does work. We've used that for many versions of Indy. There is something else going on.

tomislav (Tue, 07 Jan 2020 17:08:19 GMT):
What environment you're testing this against and what does the proof request look like? How is the nonce generated?

mattatkiva (Tue, 07 Jan 2020 17:19:33 GMT):
is this problem specific to latest indy node?

jljordan_bcgov (Tue, 07 Jan 2020 18:36:53 GMT):
indyscan

jljordan_bcgov (Tue, 07 Jan 2020 18:38:11 GMT):
Nice improvements @PatrikStas ! Thanks for your contribution.

manavgupta (Wed, 08 Jan 2020 03:58:15 GMT):
@tomislav verifier is on java and presenter is in android. The nonce is generated by the indy funtion Anoncreds.generateNonce The proof is same as in the Alice faber example with one new attribute which have the restriction with different credDef id

manavgupta (Wed, 08 Jan 2020 06:55:26 GMT):
@swcurran following the proof request format send to alice(android) from Acme company(java) ```{"name":"Job - Application","requested_attributes":{"attr5_referent":{"name":"status","restrictions":{"cred_def_id":"ARH7aVdfsUN62mTzDPxfHY:3:CL:196:TAG2"}},"attr2_referent":{"name":"last_name"},"attr6_referent":{"name":"ssn","restrictions":{"cred_def_id":"ARH7aVdfsUN62mTzDPxfHY:3:CL:195:TAG1"}},"attr1_referent":{"name":"first_name"},"attr3_referent":{"name":"degree","restrictions":{"cred_def_id":"ARH7aVdfsUN62mTzDPxfHY:3:CL:195:TAG1"}},"attr7_referent":{"name":"year"}},"nonce":"854031919543470864959106","version":"0.1","requested_predicates":{"predicate1_referent":{"p_value":4,"name":"average","p_type":">=","restrictions":{"cred_def_id":"ARH7aVdfsUN62mTzDPxfHY:3:CL:195:TAG1"}}}}``` Is the format of proof is right??

manavgupta (Wed, 08 Jan 2020 06:55:26 GMT):
@swcurran following the proof request format send to alice(android) from Acme company(java) ```{"name":"Job - Application","requested_attributes":{"attr4_referent":{"name":"status","restrictions":{"cred_def_id":"ARH7aVdfsUN62mTzDPxfHY:3:CL:196:TAG2"}},"attr2_referent":{"name":"last_name"},"attr5_referent":{"name":"ssn","restrictions":{"cred_def_id":"ARH7aVdfsUN62mTzDPxfHY:3:CL:195:TAG1"}},"attr1_referent":{"name":"first_name"},"attr3_referent":{"name":"degree","restrictions":{"cred_def_id":"ARH7aVdfsUN62mTzDPxfHY:3:CL:195:TAG1"}},"attr6_referent":{"name":"year"}},"nonce":"854031919543470864959106","version":"0.1","requested_predicates":{"predicate1_referent":{"p_value":4,"name":"average","p_type":">=","restrictions":{"cred_def_id":"ARH7aVdfsUN62mTzDPxfHY:3:CL:195:TAG1"}}}}``` Is the format of proof is right??

manavgupta (Wed, 08 Jan 2020 06:58:05 GMT):
The attribute4 is restricted by different credDef id

manavgupta (Wed, 08 Jan 2020 06:58:49 GMT):
@tomislav can you please check the format of proof?

manavgupta (Wed, 08 Jan 2020 06:58:49 GMT):
@tomislav can you please check the format of proof request ?

Diiaablo95 (Wed, 08 Jan 2020 08:25:07 GMT):
Hello guys! Does Indy support verification of a credential validity that was used in the past? I mean, assuming there is an event requiring a user to prove that the credential he is using is not revoked, and that such proof must then be used by another party to collect money from the credential issuer, for instance. If a week passes between the generation of the proof and the money collection by the party offering the service, is he able to proof that, at the time of the transaction, the credential was valid? How would that work, under the woods? Thanks a lot! :)

Diiaablo95 (Wed, 08 Jan 2020 08:25:07 GMT):
Hello guys! Does Indy support *verification of a credential validity that was used in the past?* I mean, assuming there is an event requiring a user to prove that the credential he is using is not revoked, and that such proof must then be used by another party to collect money from the credential issuer, for instance. If a week passes between the generation of the proof and the money collection by the party offering the service, is he able to proof that, at the time of the transaction, the credential was valid? How would that work, under the woods? Thanks a lot! :)

Diiaablo95 (Wed, 08 Jan 2020 08:25:07 GMT):
Hello guys! Does Indy support *verification of a credential validity that was used in the past?* I mean, assuming there is an event requiring a user to prove that the credential he is using is not revoked, and that such proof must then be used by another party to collect money from the credential issuer, for instance. If a week passes between the generation of the proof and the money collection by the party offering the service, is he able to prove that, at the time of the transaction, the credential was valid? How would that work, under the woods? Thanks a lot! :)

Vincent_S (Wed, 08 Jan 2020 10:04:13 GMT):
Hello everyone! We encountered an issue regarding credential verification while implementing a process where the workflow is as follows: 1. An issuer creates _Credential 1_ and adds it to the accumulator. 2. A proof verification using _Credential 1_ is conducted. 3. The same issuer creates another credential _Credential 2_ under the same claim definition, adding it to the same accumulator. 4. A proof verification using _Credential 1_ is conducted. 5. A proof verification using _Credential 2_ is conducted. We observe that all proofs related to _Credential 1_ are accepted, while those for _Credential 2_ aren't. Even if we issue and revoke several credentials in step 3, a proof verification in step 4 is still succesful. In our opinion, this indicates that we use the correct witness deltas in the proof of non-revocation. Everything for _Credential 1_ (the first credential in the accumulator) works as it should, while it does not for _Credential 2_. Does anybody have an idea why this occurs? Any help would be much appreciated. Thanks!

wip-abramson (Wed, 08 Jan 2020 12:31:45 GMT):
Hi, is there a document somewhere that describes the different roles that nyms can take and the associated authority these roles have within the network?

swcurran (Wed, 08 Jan 2020 14:15:30 GMT):
The holder/prover must re-prove the non-revocation of the credential. Only the prover is able to prove non-revocation at any point in time.

tomislav (Wed, 08 Jan 2020 14:48:23 GMT):
All of this looks good. I would try this request by removing the predicate referent and see if that causes it to pass

tomislav (Wed, 08 Jan 2020 14:48:40 GMT):
But first try to use version "1.0" instead one starting with 0

tomislav (Wed, 08 Jan 2020 14:48:48 GMT):
At least in case of schemas, this is not allowed

tomislav (Wed, 08 Jan 2020 14:48:52 GMT):
Could be the same problem

swcurran (Wed, 08 Jan 2020 15:20:42 GMT):
I believe that has been changed recently (last 6 months). Suggest looking through recently completed JIRAs. I know that roles have been made configurable, so that a config transaction can be executed to set the capabilities associated with a role. ADAIK, Sovrin is the only group that has applied those config capabilities to their ledger instances.

Vincent_S (Wed, 08 Jan 2020 15:39:50 GMT):
Hello everyone, yet another question: we encountered that we cannot issue credentials where the values are too long (e.g. a long URL and such). Does anybody have an explanation for this? Thanks very much!!

Vincent_S (Wed, 08 Jan 2020 15:43:30 GMT):
This applies to the "raw" values by the way.

MALodder (Wed, 08 Jan 2020 16:16:28 GMT):
The values need to be mapped to 256-bit integers. The easiest method to do this, is to use SHA256 for string values

lynn.bendixsen (Wed, 08 Jan 2020 16:40:02 GMT):
Hi @SigmaS 1 I believe that all Indy nodes are currently Stewards as Observers are not yet implemented.

VithushanJegatheeswaran (Wed, 08 Jan 2020 18:15:53 GMT):
Nice! I'll try this out

manavgupta (Thu, 09 Jan 2020 05:32:48 GMT):
@tomislav I have tried both of the suggested solutions but i am still getting the same error ( i.e when verifier verifies the proof, it returns false) Can you please suggest further?

DeepakMaurya (Thu, 09 Jan 2020 10:23:15 GMT):
Has joined the channel.

DeepakMaurya (Thu, 09 Jan 2020 10:27:32 GMT):
Hi team , When I start example sample in java . I got an error "Caused by: org.hyperledger.indy.sdk.pool.PoolLedgerConfigExistsException: A pool ledger configuration already exists with the specified name."

JeromeK (Thu, 09 Jan 2020 10:39:31 GMT):
Just comment out this section, where the Pool gets created (you cant create multiple Pools with the same name) Or you just delete the PoolName-Folder in .indy_client/pool

Vincent_S (Thu, 09 Jan 2020 11:12:54 GMT):
Resolved!

ketankumar.patil (Thu, 09 Jan 2020 13:03:10 GMT):
Hello All, I am trying to run the Indy demo example present at https://github.com/hyperledger/education/tree/master/LFS171x/indy-material/nodejs After running the credential definition step, I was going through the in Indy Domain ledger details. Below are the transaction details. "txn":{ "data":{ "data":{ "primary":{ "n":"118....153", "r":{ "email":"484....845", "master_secret":"101....217", "name":"195....914", "tax_id":"603....277" }, "rctxt":"132....857", "s":"977....802", "z":"970....009" } }, "ref":8, "signature_type":"CL", "tag":"GOVID" } *I am not able to find the meaning of primary.n, primary.rctxt, primary.s, primary.z* Does anybody have an explanation for this? Kindly advise.Thanks

ketankumar.patil (Thu, 09 Jan 2020 13:04:15 GMT):
Hello All, I am trying to run the Indy demo example present at https://github.com/hyperledger/education/tree/master/LFS171x/indy-material/nodejs After running the credential definition step, I was going through the in Indy Domain ledger details. Below are the transaction details. `"txn":{ "data":{ "data":{ "primary":{ "n":"118....153", "r":{ "email":"484....845", "master_secret":"101....217", "name":"195....914", "tax_id":"603....277" }, "rctxt":"132....857", "s":"977....802", "z":"970....009" } }, "ref":8, "signature_type":"CL", "tag":"GOVID" }` *I am not able to find the meaning of primary.n, primary.rctxt, primary.s, primary.z* Does anybody have an explanation for this? Kindly advise.Thanks

DeepakMaurya (Thu, 09 Jan 2020 13:17:02 GMT):
Hi Jeromek , I am new in indy. can you explain step up process and and running node for sample demo java

DeepakMaurya (Thu, 09 Jan 2020 13:17:16 GMT):
please reply

DeepakMaurya (Thu, 09 Jan 2020 13:18:06 GMT):
you have any url please share with me

tomislav (Thu, 09 Jan 2020 13:23:34 GMT):
I'm out of ideas, I'm sorry. We had similar issue and it was related to Ubuntu 16.04. Upgrading to 18.04 fixed the problem.

snogueir (Thu, 09 Jan 2020 14:05:11 GMT):
disaster

JeromeK (Thu, 09 Jan 2020 14:49:29 GMT):
Hey Guys Is it possible to decrypt the Data I exported from the Wallet with the export Key? Exp: `await Wallet.export_wallet(wh, { 'path': EXPORT_PATH, 'key': '123', 'key_derivation_method': 'ARGON2I_MOD' # Could also be "ARGON2I_INT" or "RAW" })` If i call this function I get a file in my Export_Path where the key '123' is used for encription, right?

JeromeK (Thu, 09 Jan 2020 14:50:07 GMT):
And just out of curiosity I would like to have a look at this file

JeromeK (Thu, 09 Jan 2020 14:51:11 GMT):
https://github.com/hyperledger/indy-sdk/blob/5f9144feab5b9e969389e325da47d7218f87081e/docs/design/009-wallet-export-import/README.md

JeromeK (Thu, 09 Jan 2020 14:51:25 GMT):
here i saw that there could be a "plain stream"?

clemenswickboldt (Thu, 09 Jan 2020 14:56:03 GMT):
Has joined the channel.

JeromeK (Thu, 09 Jan 2020 15:16:27 GMT):
here is the Link to a Docker file

JeromeK (Thu, 09 Jan 2020 15:16:46 GMT):
in my opinion the easiest way to setup the Network

JeromeK (Thu, 09 Jan 2020 15:16:47 GMT):
https://github.com/hyperledger/indy-sdk/blob/master/ci/indy-pool.dockerfile

JeromeK (Thu, 09 Jan 2020 15:17:34 GMT):
also here the link to a sample on how to connect to a pool: https://github.com/hyperledger/indy-sdk/blob/master/samples/java/src/main/java/Ledger.java

DeepakMaurya (Thu, 09 Jan 2020 17:37:23 GMT):
Exception in thread "main" java.util.concurrent.ExecutionException: org.hyperledger.indy.sdk.pool.PoolConfigNotCreatedException: The requested pool cannot be opened because it does not have an existing configuration.

DeepakMaurya (Thu, 09 Jan 2020 17:38:16 GMT):
when I run the application . I got this error

DeepakMaurya (Thu, 09 Jan 2020 17:38:34 GMT):
how to resolve this error

pfeairheller (Thu, 09 Jan 2020 17:52:36 GMT):
Has joined the channel.

pfeairheller (Thu, 09 Jan 2020 17:52:37 GMT):
Hello Indy Folks- Ramping up on indy and have a question. I've been through all walk-through code and all the how to code running against a local docker network and a network of nodes created on GCE using the instructions from https://github.com/hyperledger/indy-node/blob/master/docs/source/start-nodes.md. My question is that I want to automate the creation of GCE nodes using terraform and don't want to use `generate_indy_pool_transactions` since that seems specific for create a test network with a well known seed. Is there any code or documentation for creating the genesis txns and files needed to create the nodes and link them together?

LedgerXYZ (Thu, 09 Jan 2020 19:37:21 GMT):
Is there any writing or video that explains the security model of the wallet. Especially: where are they cryptographic keys stored

LedgerXYZ (Thu, 09 Jan 2020 19:42:38 GMT):
What is the justification for their security in relation to storage? In the demos when you create a new wallet, the returned value is assigned to a variable, isn't this is a security fall (especially for js applications). Finally for applications that want to use secure storage enclaves like android keystore, are they any example implementations that can be referenced?

pfeairheller (Thu, 09 Jan 2020 19:47:20 GMT):
Ok, left to my own devices I think I've cobbled together the answer. Based on the start-nodes doc I have to create the pool txn genesis file and the domain txn genesis file with the first Trustee (domain) and then and additional Stewards (for the nodes?) and then the Node txns for each node. I see most of the code I need in or referenced by test_network_setup.py in indy-plenum (minus the stuff that is specific to the test set up). I'm going to put this together into a script and try to add it to my terraform automation. I feel like I might need to submit each node txn separately after I create the first node... tbd.

swcurran (Thu, 09 Jan 2020 20:56:06 GMT):
A question for @kenebert and @brentzundel. We've had a couple of reasonable requests about when a proof is "verified" with respect to the proof request used. These are not part of the current process, and I wonder if they were planned to be supported in the Rich Schema work? 1) Recently add to Indy is the idea of a group of claims that have common restrictions. Will it be possible with Rich Schema to express in the Rich Schema work that the credentials in the group MUST come from the same credential? I believe that currently, all claims are handled independently in the proof/verification process. So far example, I can say "must use a Driver's License Schema" and such that the claims can be from any jurisdiction that issued Driver's License schema credentials, but claims returned must be from the same credential. 2) We have found that in Indy, it appears that all claims are optional. That is, I can have a proof request asking for some claims, get back a proof with some of the claims provided and the proof will be "verified" (which is cryptographically correct). Will there be in Rich Schema to indicate required/optional claims such that the proof would only be verified if all the required claims were provided. Note that the existing behaviour is also very useful, so we don't want to lose it. If that is not planned, at what level would you expect such logic to be added? If not in Indy, I'm assuming the app would have to do that business checking and presumably send a Problem Report message back to the Prover and redo the process. It would be much better if the prover knew those requirements ahead of time and only constructed a proof meeting those guidelines.

Whatzhub (Fri, 10 Jan 2020 06:00:03 GMT):
Has joined the channel.

Whatzhub (Fri, 10 Jan 2020 06:00:03 GMT):
Hello everyone, may I ask HL Indy has integrations with uPort? Or just Sovrin?

shubhamchaudhari (Fri, 10 Jan 2020 06:10:11 GMT):
Has joined the channel.

shubhamchaudhari (Fri, 10 Jan 2020 06:10:13 GMT):
Hi all, I have a question that in indy document walkthrough in connection establishment section of onboarding part its written that the steward sends the NYM transaction for did created for feber. Is that mean the Pseudonym-did is stored on the ledger ? Because in the "White Paper: Sovrin: What Goes On The Ledger" mentioned that private-dids are not stored on the ledger.

DeepakMaurya (Fri, 10 Jan 2020 10:21:27 GMT):
Hi Team I got this error when I try to run sample java Exception in thread "main" java.util.concurrent.ExecutionException: org.hyperledger.indy.sdk.pool.PoolConfigNotCreatedException: The requested pool cannot be opened because it does not have an existing configuration.

DeepakMaurya (Fri, 10 Jan 2020 10:21:27 GMT):
Hi Team I got this error when I try to run sample java Exception in thread "main" java.util.concurrent.ExecutionException: org.hyperledger.indy.sdk.pool.PoolConfigNotCreatedException: The requested pool cannot be opened because it does not have an existing configuration.

DeepakMaurya (Fri, 10 Jan 2020 10:22:56 GMT):
please reply, how to resolve this error?

DavoB (Fri, 10 Jan 2020 13:30:01 GMT):
If you are using one the Java examples in the How-Tos and followed JeromeK's advice of commenting out the Pool.createPoolLedgerConfig line to get past your previous error, note that, you'd eventually need to re-enable that line. When the How-To samples run to completion, they execute a Pool.deletePoolLedgerConfig command.

DavoB (Fri, 10 Jan 2020 13:30:01 GMT):
If you are using one the Java examples in the How-Tos and followed JeromeK's advice of commenting out the Pool.createPoolLedgerConfig line to get past your previous error, note that, you'd eventually need to re-enable that line. When the How-To samples run to completion, they execute a Pool.deletePoolLedgerConfig command at the end of the scripts.

DavoB (Fri, 10 Jan 2020 13:54:56 GMT):
I found some useful discussion of roles from an indy ledger browser devleoper at https://stackoverflow.com/questions/59089178/hypelerdger-indy-node-seed-value

DavoB (Fri, 10 Jan 2020 13:54:56 GMT):
I found some useful discussion of roles from PatrikStas at https://stackoverflow.com/questions/59089178/hypelerdger-indy-node-seed-value

lynn.bendixsen (Sat, 11 Jan 2020 17:08:17 GMT):
Thanks Patrik! Many people at the Sovrin Foundation have used and enjoyed the great new features!

VithushanJegatheeswaran (Sun, 12 Jan 2020 11:22:46 GMT):
OK I tought abot this. is it somehow possible to request a hashed version of the full verifiable claim from the holder ? with that hash I can verify if a VC is used twice right ?

VithushanJegatheeswaran (Sun, 12 Jan 2020 11:22:46 GMT):
@swcurran OK I thought abot this. is it somehow possible to request a hashed version of the full verifiable claim from the holder ? with that hash I can verify if a VC is used twice right ?

Sukalpo (Sun, 12 Jan 2020 13:43:18 GMT):
Has joined the channel.

MALodder (Sun, 12 Jan 2020 17:52:48 GMT):
We can do this with Pedersen commitments and equality proofs which very lightweight. The disadvantage of hashing is you get link ability which in most voting systems is not allowed. The link ability comes from the presentation. Do you want to know if someone has voted already but not who they are necessarily

MALodder (Sun, 12 Jan 2020 17:52:48 GMT):
We can do this with Pedersen commitments and equality proofs which are very lightweight. The disadvantage of hashing is you get link ability which in most voting systems is not allowed. The link ability comes from the presentation. you want to know if someone has voted already but not who they are necessarily

amarts (Mon, 13 Jan 2020 06:43:40 GMT):
Has joined the channel.

VithushanJegatheeswaran (Mon, 13 Jan 2020 12:46:00 GMT):
test

VithushanJegatheeswaran (Mon, 13 Jan 2020 12:47:07 GMT):
@MALodder I can prove that a certain DID has voted but a user can create multiple connections (=> multiple DIDs) to the verifier, I can't make sure that someone votes twice ... I really want this to work, but it seems that I am too early with this technology.

MALodder (Mon, 13 Jan 2020 15:26:34 GMT):
@VithushanJegatheeswaran you can make it DID agnostic.

MALodder (Mon, 13 Jan 2020 15:33:26 GMT):
You can bind the vote to an identity value and require them to present certain information to prevent duplication

VithushanJegatheeswaran (Mon, 13 Jan 2020 15:34:18 GMT):
what do you mean by identity value ?

Yoroitchi (Mon, 13 Jan 2020 15:34:33 GMT):
Hello team. I'm trying to connect to a pool i created but i have the following error ```Error: Invalid library state Caused by: Ledger merkle tree is not acceptable for current tree. ``` What does it mean ?

VithushanJegatheeswaran (Mon, 13 Jan 2020 15:36:04 GMT):
I already thought about adding a random number for each credential but then the issuer and verifier can communicate and track down the voter

MALodder (Mon, 13 Jan 2020 15:38:03 GMT):
Each credential in indy is bound to a link secret. You can bind the vote to the link secret without revealing the link secret using a pedersen commitment. When someone casts a vote, you require they present a commitment to their link secret with their vote. A proof of other attributes from a credential signed by an issuer should be used as well to prove they are allowed to vote like age, registration, and other meaningful values if needed. The point is, if they try to vote again, then the duplication will be detected. The other option, is to issue then a one time spendable credential. The credential can still be issued to the link secret which is only known to the holder. When the holder votes, the credential is spent and cannot be used again.

MALodder (Mon, 13 Jan 2020 15:38:39 GMT):
What is the purpose of the random number?

VithushanJegatheeswaran (Mon, 13 Jan 2020 15:46:21 GMT):
i wanted to add a random number to each claim. So the holder can send it to the verifier, which logs that number. if the same random number appears on a vote, the verifier can be sure that the person already voted. BUT on the other hand the verifier can also ask the issuer to which person the number belongs ... which breaks the anonymity

MALodder (Mon, 13 Jan 2020 15:48:03 GMT):
Right, I'm saying its not needed because the link secret can be used in its place. Instead you ask the holder to present a commitment to their link secret and add the commitment to the voted set. If the set detects a duplicate, then the person has already voted. If not, then add the commitment to the set

MALodder (Mon, 13 Jan 2020 15:50:25 GMT):
The commitment serves three purposes, one it hides the actual link secret from the verifier and protects the holder's identity, two, it is created at random so even if the holder tries to vote multiple times, the presentations are unlinkable, three, it is used to detect if the same value has already been used, double spend proofing

MALodder (Mon, 13 Jan 2020 15:50:39 GMT):
All of this can be done in zero-knowledge

Yoroitchi (Mon, 13 Jan 2020 15:51:59 GMT):
docker ps

VithushanJegatheeswaran (Mon, 13 Jan 2020 15:52:22 GMT):
Thank you so much for taking your time :) I'll have to research what a commitment is What you describe sounds very good and gives me hope.

MALodder (Mon, 13 Jan 2020 15:53:06 GMT):
This link provides a pretty good description and how they work - https://crypto.stackexchange.com/questions/64437/what-is-a-pedersen-commitment

VithushanJegatheeswaran (Mon, 13 Jan 2020 15:53:29 GMT):
nice THANK YOU

esplinr (Mon, 13 Jan 2020 16:00:01 GMT):
Starting the Indy Contributors call: https://zoom.us/j/615818107

esplinr (Mon, 13 Jan 2020 17:00:44 GMT):
I'm not clear on why the call died. But pay attention online and I'll circulate whether we have a meeting next Monday. Thank you everyone for your participation!

swcurran (Mon, 13 Jan 2020 19:04:28 GMT):
Not sure if this came clear, but this is my understanding of how this would work in a practical sense. Each voter would have a credential that entitled them to vote. That could be a a gov't ID, an organization membership credential. If the election was for a city, it could be a national ID and the person would prove that there address had that city (but nothing else). The election organizer (e.g. the election commission) would send out a proof request that included the claims to be proven to show eligibility to vote, and a unique ID for the election. Voters would prove the claim, their election choices and use the unique election ID with their Indy link secret to handle the commitment. The election organizer would receive the proofs and be able to detect duplicates, applying some policy on handling duplicate votes (most likely - last one counts). Note that the commitment functionality as described here is NOT currently implemented in Indy. It is possible and will very likely get done, but is not in place today. It's one of the key features that needs to be added to the technology.

swcurran (Mon, 13 Jan 2020 19:04:28 GMT):
Not sure if this came clear, but this is my understanding of how this would work in a practical sense. Each voter would have a credential that entitled them to vote. That could be a a gov't ID, an organization membership credential. If the election was for a city, it could be a national ID and the person would prove that their address had that city (but nothing else). The election organizer (e.g. the election commission) would send out a proof request that included the claims to be proven to show eligibility to vote, and a unique ID for the election. Voters would prove the claim, their election choices and use the unique election ID with their Indy link secret to handle the commitment. The election organizer would receive the proofs and be able to detect duplicates, applying some policy on handling duplicate votes (most likely - last one counts). Note that the commitment functionality as described here is NOT currently implemented in Indy. It is possible and will very likely get done, but is not in place today. It's one of the key features that needs to be added to the technology.

swcurran (Mon, 13 Jan 2020 19:04:28 GMT):
Not sure if this came clear, but this is my understanding of how this would work in a practical sense. Each voter would have a credential that entitled them to vote. That could be a a gov't ID, an organization membership credential. If the election was for a city, it could be a national ID and the person would prove that their address had that city (but nothing else). The election organizer (e.g. the election commission) would send out a proof request that included the claims to be proven to show eligibility to vote, and a unique ID for the election. Voters would prove the claim, their vote(s) and use the unique election ID with their Indy link secret to handle the commitment. The election organizer would receive the proofs and be able to detect duplicates, applying some policy on handling duplicate votes (most likely - last one counts). Note that the commitment functionality as described here is NOT currently implemented in Indy. It is possible and will very likely get done, but is not in place today. It's one of the key features that needs to be added to the technology.

swcurran (Mon, 13 Jan 2020 19:07:39 GMT):
Note that in such a system, voting multiple times is not a bad thing, as long as an eligible voter only gets one vote counted, and there is a clear policy on how multiple votes are handled.

swcurran (Mon, 13 Jan 2020 19:09:26 GMT):
BTW - the hash idea doesn't work because either the issuer provides it and then it's just a unique ID that can be correlated with the voter, or the voter self-issues it, in which case they can just make up a different value each time (since no one can check the hash).

VithushanJegatheeswaran (Mon, 13 Jan 2020 19:22:07 GMT):
Yes you're right. The hash idea was stupid :D "Note that in such a system, voting multiple times is not a bad thing..." I had no problem with voting multiple times but with voting with multiple DIDs. As you already said, I'll count the vote only "Note that the commitment functionality as described here is NOT currently implemented in Indy." I will (somehow) try to implement this into my app

MALodder (Mon, 13 Jan 2020 19:25:11 GMT):
It is available in Ursa though. I’m working to implement this in Aries next

VithushanJegatheeswaran (Mon, 13 Jan 2020 19:28:21 GMT):
ohh nice I'll check that out I only worked with indy so far .. let's see what ursa is all about

MALodder (Mon, 13 Jan 2020 19:40:16 GMT):
Ursa is the crypto behind Indy

swcurran (Mon, 13 Jan 2020 19:44:33 GMT):
A rocketchat...such fun. I didn't see this sub-thread. So my note can be summarized as ... what Mike said :-)

swcurran (Mon, 13 Jan 2020 19:44:33 GMT):
Ah rocketchat...such fun. I didn't see this sub-thread. So my note can be summarized as ... what Mike said :-)

VithushanJegatheeswaran (Mon, 13 Jan 2020 19:50:16 GMT):
earlier i had problems replying to the old thread (HTTP ERROR 400) Thats why I started this thread :sweat_smile:

Silona (Tue, 14 Jan 2020 16:50:11 GMT):
Help Us Help you! Attend the Developer Relationship Meeting with Myself and our Marketing Dept. tomorrow at 9:00am Pacific Time. For the agenda and Dial in info https://wiki.hyperledger.org/display/Marketing/2020-01-15+Meeting+notes

Silona (Tue, 14 Jan 2020 17:01:26 GMT):
Calling all Projects! We will have a special Kiosk setup at Hyperledger Global Forum for Projects. We are asking that all projects sign up to do 10 minute presentations. https://wiki.hyperledger.org/display/HGF/Projects+Kiosk We will close this page on Feb 28 for printing reasons.

Silona (Tue, 14 Jan 2020 17:05:04 GMT):
Calling all Projects, SIG, and WG!!! We will have a Video recording Studio setup at HGF (Hyperledger Global Forum). We are asking that all projects and groups help us create a 5 minute video about your group so that we can promote it afterward. Sign up Here! https://wiki.hyperledger.org/display/HGF/Video+Recording+Schedule

KyleHuang (Wed, 15 Jan 2020 06:43:21 GMT):
Has joined the channel.

KyleHuang (Wed, 15 Jan 2020 06:43:22 GMT):
Hi, guys, I have some questions about the Hyperledger Indy focusing on the credential revocation. Thank you for your help. The credential revocation of Indy refers to https://hyperledger-indy.readthedocs.io/projects/hipe/en/latest/text/0109-anoncreds-protocol/README.html The CKS accumulator refers to this paper https://eprint.iacr.org/2008/539.pdf Questions 1. In the Non-revocation Credential Cryptographic Setup phase, a type-III pairing is chosen. Is there any suggestion while choosing all three groups G1, G2 and GT? For example, how many bits is the order r suggested to be? By the way, is there some special purpose to use a type-III pairing? The referred CKS accumulator uses a type-I pairing which G1 = G2. The CKS accumulator could be referred here. 2. The architecture includes an accumulator system that the issuer side stores the states (V and {g'_i} for all i in [1, 2L]) and the blockchain stores the indices V and accumulator acc. By the definition of accumulator on wiki: A cryptographic accumulator is a one way membership function. It answers a query as to whether a potential candidate is a member of a set without revealing the individual members of the set. We have the following questions: It seems not a one-way function if everyone can observe V and acc together from the blockchain. The mapping of indices i and parameters g'_i could be gradually found after several accumulator updates. Following the above question, after the mapping relationship is totally observed, I believe the membership of the current accumulator is entirely revealed. If V is stored on the ledger, the size of V keeps growing alone with more holders get involved in. In my opinion, the size of a cryptographic accumulator should be independent from the number of members. In short, we estimate that the indices V may not be suitable to be published and stored on the ledger. Please feel free to indicate if we misunderstood something. Thank you very much.

VithushanJegatheeswaran (Wed, 15 Jan 2020 16:11:56 GMT):
Is there an APP to store (indy) credentials for Android ? @swcurran used the streetcred APP (https://streetcred.id/) in this video (https://www.youtube.com/watch?v=bZrWAsD42-I) but thats for IOS only.

gnarula (Wed, 15 Jan 2020 16:39:38 GMT):
Hi everyone! Is it possible for a DID Doc to embed or refer to another DID doc? Say Alice and Bob want to use a common service S but using the service S requires Alice to know Bob's DID pair with S and Bob to know Alice's DID pair with S.

swcurran (Wed, 15 Jan 2020 16:39:54 GMT):
Coming soon! The folks at Streetcred are busy on that. Not quite ready yet, but soon!

VithushanJegatheeswaran (Wed, 15 Jan 2020 16:48:19 GMT):
ok Thank you :)

kdenhartog (Thu, 16 Jan 2020 00:41:31 GMT):
THese questions will likely be better asked in the #ursa channel because they're very specific to the cryptography chosen and the people who decided these types of things are most active on those channels. @MALodder

KyleHuang (Thu, 16 Jan 2020 02:48:59 GMT):
Thank you very much, I'll post there.

parthibanselvaraj (Thu, 16 Jan 2020 06:10:33 GMT):
Has joined the channel.

aravindavk (Thu, 16 Jan 2020 09:27:34 GMT):
Has joined the channel.

sankarshanm (Thu, 16 Jan 2020 10:30:18 GMT):
Has joined the channel.

gaila (Thu, 16 Jan 2020 13:59:40 GMT):
Hello, I have a question about indy wallets. How can I get the indy wallet private key?

awais.ahmad (Thu, 16 Jan 2020 14:05:17 GMT):
Hi Guys, how would I go about getting the credentials from an indy node to a mobile phone.

esplinr (Thu, 16 Jan 2020 16:07:34 GMT):
"the network isn't ready for prime time" or "you can't really use this because..."

esplinr (Thu, 16 Jan 2020 16:07:34 GMT):
Identety WG call starting now.

esplinr (Thu, 16 Jan 2020 16:07:34 GMT):
Identity Implementors WG call starting now. https://zoom.us/j/244779296

esplinr (Thu, 16 Jan 2020 16:08:21 GMT):
Agenda here: https://wiki.hyperledger.org/display/IWG/2020-01-16+Identity+WG+Implementers+Call

esplinr (Thu, 16 Jan 2020 20:36:06 GMT):
The latest Evernym Sprint Demo video covers recent releases, bug fixes, and support for VC-Authn-OIDC in LibVCX. https://www.youtube.com/watch?v=ymEcmc2As7w&list=PLRp0viTDxBWGLdZk0aamtahB9cpJGV7ZF

geonnave (Fri, 17 Jan 2020 07:33:26 GMT):
Has joined the channel.

awais.ahmad (Fri, 17 Jan 2020 10:24:52 GMT):
Just a small question regarding Indy, Is it possible to track how many verifications a verifier has done?

swcurran (Fri, 17 Jan 2020 14:52:40 GMT):
The verifier could do that, but no other party. There has in the past been discussions of the notion of "Premium Credentials" that would enable some sort of mechanism for enable payment to another party (most likely the issuer, but could be the prover) prior to verification, but I don't recall seeing details of how that would work. A key to the approach would be to ensure that the issuer knew an issued credential was verified, but not which credential, so the issuer could not monitor prover activity.

swcurran (Fri, 17 Jan 2020 14:52:40 GMT):
The verifier could do that, but no other party. There has in the past been discussions of the notion of "Premium Credentials" that would enable some sort of mechanism to enable payment to another party (most likely the issuer, but could be the prover) prior to verification, but I don't recall seeing details of how that would work. A key to the approach would be to ensure that the issuer knew an issued credential was verified, but not which credential, so the issuer could not monitor prover activity.

awais.ahmad (Fri, 17 Jan 2020 16:38:07 GMT):
Thank you for a great reply. This premium credentials mechanism is exactly what I am looking for. I will try and read up if I can find something about it.

manavgupta (Mon, 20 Jan 2020 04:46:24 GMT):
That the pool_transactions_genesis of Indy-node pool and client is not same

Yoroitchi (Mon, 20 Jan 2020 07:37:45 GMT):
Ok make sense, thank you for your answer

ShamGir (Mon, 20 Jan 2020 08:40:48 GMT):
Has joined the channel.

ShamGir (Mon, 20 Jan 2020 09:53:48 GMT):
Hello guys, I am facing this issue when create a wallet in new indy-sdk version 1.14.1 Attempt to invoke interface method 'int org.hyperledger.indy.sdk.LibIndy$API.indy_create_wallet(int, java.lang.String, java.lang.String, com.sun.jna.Callback)' on a null object reference

esplinr (Mon, 20 Jan 2020 23:00:36 GMT):
We are starting the Indy Contributors call: https://zoom.us/j/615818107

esplinr (Tue, 21 Jan 2020 00:40:10 GMT):
Call recording posted: https://wiki.hyperledger.org/display/indy/2020-01-20+Indy+Contributors+Call

ShamGir (Tue, 21 Jan 2020 07:56:51 GMT):
Hello guys, anyone can tell me, what data is stored in "setDidMetadata" method of indy-sdk ??

Diiaablo95 (Tue, 21 Jan 2020 08:43:56 GMT):
So it is not possible, tomorrow, to verify if a proof presented today was valid today? I do not need to prove that it is valid tomorrow, but tomorrow I need to prove that it was valid today.

ekubilay (Tue, 21 Jan 2020 11:22:47 GMT):
Hello everyone, could any of you tell me where the value of revRegDefTag(Revocation registry definition tag) is stored? Many thanks in advance!

dgunseli (Tue, 21 Jan 2020 16:03:05 GMT):
Hello people, we are using prover_create_master_secret function of Indy-SDK Java, we are getting key correctness proof error, is there anybody who faced before with this error?

dgunseli (Tue, 21 Jan 2020 16:03:05 GMT):
Hello people, we are using prover_create_master_secret function of Indy-SDK Java, we are getting key correctness proof error, is there anybody who faced before with this error? Full error message : Error: Invalid structure Caused by: UrsaCryptoError: Invalid Credential key correctness proof

ShamGir (Wed, 22 Jan 2020 06:54:25 GMT):
Yes, I got same error, but not remember the cause. Can you share what are you passing to this method of indy ?

dgunseli (Wed, 22 Jan 2020 08:37:32 GMT):
wallet handle, my did for this connection, credential offer, credential definition and a randomly generated master secret key

ShamGir (Wed, 22 Jan 2020 08:42:28 GMT):
It takes only two parameters, wallet-handler and master-secret. pass null in master-secret then try it

ShamGir (Wed, 22 Jan 2020 08:42:58 GMT):
Anoncreds.proverCreateMasterSecret(wallet, null)

dbukrek (Wed, 22 Jan 2020 11:15:18 GMT):
Has joined the channel.

hyper-sunder (Thu, 23 Jan 2020 04:43:18 GMT):
Hi Guys, I downloaded the latest release of `indy-sdk(1.14.1)` and tried to run sample in `vcx/wrapper/node` along with `dummy-agent`. When run `npm run demo:faber` I do get the following error saying ...

hyper-sunder (Thu, 23 Jan 2020 04:43:44 GMT):
``` ERROR|vcx::api::utils | src/api/utils.rs:84 | vcx_agent_provision_async_cb(command_handle: 0, rc: Error: Invalid HTTP response. Caused by: Invalid HTTP response. Caused by: POST failed with: IndyError { error_code: CommonInvalidStructure, message: "Error: Invalid structure\n Caused by: Unable to open sodium sealedbox\n", indy_backtrace: Some("") } Can't unbundle message. , config: NULL (node:28585) UnhandledPromiseRejectionWarning: Error: Invalid HTTP response. ```

hyper-sunder (Thu, 23 Jan 2020 04:44:05 GMT):
can anyone help me to fix this???

mirgee (Thu, 23 Jan 2020 08:20:14 GMT):
Has joined the channel.

hyper-sunder (Thu, 23 Jan 2020 09:56:40 GMT):
As well as, am looking for another guidance too. I have started `indy-pool` using the following steps (https://github.com/hyperledger/indy-sdk/blob/master/README.md#how-to-start-local-nodes-pool-with-docker). I would like to know how to setup explorer for my local indy network? As like (https://indyscan.io/home/sovmain)

JeromeK (Thu, 23 Jan 2020 15:08:30 GMT):
https://github.com/Patrik-Stas/indyscan/tree/master/start

brrussell (Thu, 23 Jan 2020 16:22:59 GMT):
That comes from the genesis file, correct? My machine can get to that IP address and a python demo is able to work with that IP as well. Any other info I can use to progress? I've tried several genesis files and get the same result

hyper-sunder (Fri, 24 Jan 2020 03:45:59 GMT):
thanks @JeromeK for the link

VithushanJegatheeswaran (Fri, 24 Jan 2020 19:06:52 GMT):
Hey guys, whats the password for the backend ? https://github.com/bcgov/iiwbook I'm trying to approve the attendee, to get a credential.

biligunb (Mon, 27 Jan 2020 07:50:58 GMT):
Can anyone point me to Indy performance report? or benchmark?

VithushanJegatheeswaran (Mon, 27 Jan 2020 15:25:05 GMT):
What is a "Tails" ? // What purpose does it serve? Is it important ? http://www.plantuml.com/plantuml/svg/fPNFQkCm4CRlUegXfnRQFi12ba8BYo6mPXjydzgZHY1Rtj4uPTz-PKLkx4JPXEsbXz5l_imtdvidByWoj5OfhCKnt1JU3yGtW1wAALQoswLOESKgOgDDYbPAzSXYQjDZ9t2pfkOzFQsNBniqriU7ZLAT4uAnD1JteSO3_EofWzuv2rli6ai-n9wPC60VQ2t9BF1Il7X5NGDZ5QLMAtYLa8JSWEcCWDFWwosrU0yrKqET6BJGa1vVZUiWmE2to3IYiZKU6XI4rKgf0Xw-Gm7VKYy_gv2Axf5rG4hd9GE0EB1H1hMUglsWBAdPjOdDTCxDdk-UXJO1YhBAg3nyi95S6wQTgp7gmlHklF2VuoOPPDRjiEyrUTI5luUK4oliNPZEqmzlrjJxwzmcpOgNCsUzXyV7nxkPCSeqJLG_9tbr0OZUUZxqOleza9Ve7pFEUT4oJkkzB04f0x9Xx7m8dKAZ93sxtUIQCesLttKnPJdnqm0jUXarf90hfy_ryKQHIKja9qpOqbS-6Oi9ct_xzpe_MS6bbSvBDfSeK-yz5BmergiuNamPhrsoSpgUDPgZcktJnwT9YYfpqoLHdZcZ1-qun-Q5FTcrThkSTf_zCX-sbn19mF56wVRLwIRxpjNrHfppb_vhN07lq0vqk4FwtPbalXKUw2uUxTSXUVv1flZNlNWmnYxEa7ZKGWoyQFxrjJHQJtTdUmg8QSaBjZsyaNPC_pa1V1ZPpeev_Ud9clenuQBqJzGrGsl_0W00

gnarula (Mon, 27 Jan 2020 15:29:25 GMT):
This should help: https://hyperledger-indy.readthedocs.io/projects/sdk/en/latest/docs/concepts/revocation/cred-revocation.html

VithushanJegatheeswaran (Mon, 27 Jan 2020 15:46:38 GMT):
Thank you

drew-streetcred (Mon, 27 Jan 2020 16:05:30 GMT):
Has joined the channel.

mattatkiva (Mon, 27 Jan 2020 16:24:22 GMT):
I was wondering what the best practices are for deleting a credential. Is revocation working and should the credential be revoked first? and last question, if I delete a credential without revoking it, what happens if I try creating it again? (I asked this in indy-sdk too)

Dan (Mon, 27 Jan 2020 17:14:18 GMT):
Hi Everyone, We are hoping to hear from everyone as we assess the health of our open source community. Please take 2 minutes to respond here: https://www.surveymonkey.com/r/DCIWGsurvey

GuillermoDLCO (Mon, 27 Jan 2020 19:22:04 GMT):
Has joined the channel.

Silona (Mon, 27 Jan 2020 22:26:50 GMT):
The Linux Foundation’s CommunityBridge engineers are working on a tool to measure the health of critical open source projects and one of the key areas identified is QA Testing. They request that our communities provide honest and detailed information on testing tools and methodologies you use in your projects for us to come up with a detailed analysis, which they will share with all respondents and projects. https://www.surveymonkey.com/r/9H5G2GV. It’s only 5 questions long.

VithushanJegatheeswaran (Tue, 28 Jan 2020 08:58:35 GMT):
The credential is stored in the wallet of the prover. you (I guess you're the issuer) can only revoke the credential. If the Prover deletes the credential, there is no way to get it back(correct me if I'm wrong).

VithushanJegatheeswaran (Tue, 28 Jan 2020 12:26:36 GMT):
done

daveek (Tue, 28 Jan 2020 12:45:33 GMT):
Has joined the channel.

daveek (Tue, 28 Jan 2020 12:45:38 GMT):
I'm working with Hyperledger Indy-SDK Library. I want to connect indy-sdk with web server through REST Api and perform operations, how to do so? Forexample, Here is a tutorial of How To Write a DID and Query Its Verkey. So, for this, one has to run indy_pool in docker and paste this code in Python IDE and run it inorder to perform the task. Now, I want to write a Did or Issue a Did or perform any task by sending a request through API. Is this Possible and yes, then how? Thanks

VithushanJegatheeswaran (Tue, 28 Jan 2020 12:56:12 GMT):
check out this service: https://developer.streetcred.id/ streetcred.id/ they have an API or These two repos: https://github.com/bcgov/indy-email-verification (lhttp://localhost:5000/api/doc) https://github.com/bcgov/iiwbook (http://localhost:4000/api/doc)

VithushanJegatheeswaran (Tue, 28 Jan 2020 12:56:12 GMT):
check out this service: https://developer.streetcred.id/ streetcred.id/ they have an API or These two repos: https://github.com/bcgov/indy-email-verification (lhttp://localhost:5000/api/doc) https://github.com/bcgov/iiwbook (http://localhost:4000/api/doc) interesting read: https://vonx.io/how_to/iiwbook

daveek (Tue, 28 Jan 2020 13:14:59 GMT):
What if want to write simple script to issue a DID through a API Request?

VithushanJegatheeswaran (Tue, 28 Jan 2020 13:18:37 GMT):
This Demo shows how to issue a credential, that proves, that the holder has acces to an email address. https://github.com/bcgov/indy-email-verification mybe you can build on top of that or modify the code ....

VithushanJegatheeswaran (Tue, 28 Jan 2020 15:53:19 GMT):
@MALodder "...the link secret can be used in its place. Instead you ask the holder to present a commitment to their link secret..." How can I be sure that the holder uses his link secret and not some other random value ? Will the holder reveal his link secret after the end of the voting ?

JeromeK (Tue, 28 Jan 2020 16:20:15 GMT):
You just have to design an API on top, have a look at the aries-cloud-agent https://github.com/hyperledger/aries-cloudagent-python to get some inspiration

MALodder (Tue, 28 Jan 2020 16:23:57 GMT):
Because the link secret was signed by the issuer and if it’s changed the proof is invalid

swcurran (Tue, 28 Jan 2020 20:20:21 GMT):
Check out the /demo folder in that repo for some examples and steps to use.

daveek (Wed, 29 Jan 2020 03:56:55 GMT):
Thanks to y'all :)

lsl88 (Wed, 29 Jan 2020 13:59:39 GMT):
Hi! I have a question. Which P2P protocol does Indy use?

lauravuo (Wed, 29 Jan 2020 17:37:04 GMT):
Has joined the channel.

tomislav (Wed, 29 Jan 2020 18:54:06 GMT):
Indy uses ZMQ for communication with the ledger nodes.

lauravuo-techlab (Thu, 30 Jan 2020 06:42:07 GMT):
Has joined the channel.

lauravuo (Thu, 30 Jan 2020 06:49:31 GMT):
Has left the channel.

sanjaysb (Thu, 30 Jan 2020 07:25:23 GMT):
Hi

sanjaysb (Thu, 30 Jan 2020 07:25:48 GMT):
I am unable to load the indy java library

sanjaysb (Thu, 30 Jan 2020 07:26:04 GMT):
Could not find org.hyperledger:indy:1.14.1 when I do gradle build

sanjaysb (Thu, 30 Jan 2020 07:33:33 GMT):
resolved. had to add sovrin repo in gradle.

ShamGir (Thu, 30 Jan 2020 07:42:46 GMT):
You can use java wrapper

ShamGir (Thu, 30 Jan 2020 07:43:31 GMT):
can you share what exactly you add in gradle file ??

sanjaysb (Thu, 30 Jan 2020 07:44:46 GMT):
inside the repositories block, ```maven { url "https://repo.sovrin.org/repository/maven-public" }```

mirgee (Fri, 31 Jan 2020 09:09:10 GMT):
Hi guys, I am playing with libvcx and want to rotate an issuers verkey, which AFAIK, is best done using libindy using `replaceKeys*` methods, which require wallet handle. I initialized a wallet using its config through `provisionAgent` in vcx nodejs wrapper (which calls `vcx_agent_provision_async`), which does not return the wallet handle. Is there a way to the obtain wallet handle using its config and key?

gnarula (Fri, 31 Jan 2020 09:10:37 GMT):
Yes, there you go https://github.com/hyperledger/indy-sdk/blob/154b4d31c0dd3fb01a03b38ef3267e1bba2a6fe5/libindy/src/api/wallet.rs#L217

mirgee (Fri, 31 Jan 2020 09:12:08 GMT):
@gnarula Sure, I know, but vcx is keeping it opened.

mirgee (Fri, 31 Jan 2020 09:12:08 GMT):
@gnarula Thanks for such a quick feedback. Sure, I know, but vcx is keeping it opened.

mirgee (Fri, 31 Jan 2020 09:12:08 GMT):
@gnarula Thanks for such a quick feedback. Sure, I know, but vcx is keeping it opened (and one can't close it without its handle).

mirgee (Fri, 31 Jan 2020 09:12:08 GMT):
@gnarula Thanks for such a quick feedback. Sure, I know, but vcx is keeping it opened (and one can't close it without its handle) https://github.com/hyperledger/indy-sdk/blob/154b4d31c0dd3fb01a03b38ef3267e1bba2a6fe5/vcx/libvcx/src/api/vcx.rs#L144.

mirgee (Fri, 31 Jan 2020 09:12:08 GMT):
@gnarula Thanks for such a quick feedback. Sure, I know, but vcx is keeping it opened (and one can't close it without its handle) https://github.com/hyperledger/indy-sdk/blob/154b4d31c0dd3fb01a03b38ef3267e1bba2a6fe5/vcx/libvcx/src/api/vcx.rs#L144, so one gets `WalletAlreadyOpenedError`

PaulA (Fri, 31 Jan 2020 10:52:00 GMT):
@ianco @swcurran going through this tutorial : Writing Transactions to a Ledger for an Un-privileged Author. ./manage \ indy-cli export-wallet \ walletName=myorg_issuer \ storageType=postgres_storage \ storageConfig='{"url":"192.168.65.3:5435"}' \ storageCredentials='{"account":"DB_USER","password":"DB_PASSWORD","admin_account":"postgres","admin_password":"mysecretpassword"}' \ exportPath=/tmp/myorg_issuer_wallet_initialized_with_did.export

PaulA (Fri, 31 Jan 2020 10:52:00 GMT):
@ianco @swcurran going through this tutorial : Writing Transactions to a Ledger for an Un-privileged Author. ./manage \ indy-cli export-wallet \ walletName=myorg_issuer \ storageType=postgres_storage \ storageConfig='{"url":"192.168.65.3:5435"}' \ storageCredentials='{"account":"DB_USER","password":"DB_PASSWORD","admin_account":"postgres","admin_password":"mysecretpassword"}' \ exportPath=/tmp/myorg_issuer_wallet_initialized_with_did.export gives error: Caused by: Permission denied (os error 13) Batch execution failed at line #4

PaulA (Fri, 31 Jan 2020 10:54:17 GMT):
The error appears to be on this line: exportPath=/tmp/myorg_issuer_wallet_initialized_with_did.export when i add <~> at ~/tmp/... it works fine. Now, i can't find the path to where it is saved and can't write schema to ledger due to file not found.

PaulA (Fri, 31 Jan 2020 10:54:22 GMT):
Please any help on this?

swcurran (Fri, 31 Jan 2020 15:08:11 GMT):
The error is a permission denied, so it's likely the path doesn't exist somewhere - the parent directory perhaps. I'm not familiar with the tutorial, so if you could point me to it that would be appreciated. Regardless, Ian probably knows the answer :-)

ianco (Fri, 31 Jan 2020 17:15:48 GMT):
https://github.com/bcgov/von-network/blob/master/docs/Writing%20Transactions%20to%20a%20Ledger%20for%20an%20Un-privileged%20Author.md

ianco (Fri, 31 Jan 2020 17:16:05 GMT):
Yes I think it's an issue with permissions on the local `./tmp` directory

ianco (Fri, 31 Jan 2020 17:16:39 GMT):
indy CLI is running in the docker container and it's writing to a mounted local `./tmp` directory

swcurran (Fri, 31 Jan 2020 17:30:28 GMT):
A quick workaround is to precreate the necessary folders, give them world rw permissions, including the "s" parameter in chmod so that items created within the folder inherit the permissions of the parent folder. The real answer is in docker handling that is pretty tricky. The docker daemon is creating the volume on startup and so it is created with root privileges (since the docker daemon runs as root), and so when the non-root use running in the container tries to write to the folder, it is not allowed.

esplinr (Sat, 01 Feb 2020 00:47:27 GMT):
Video by Evernym on the latest release of Hyperledger Indy, including a brief project update, description of how to test changes to the pool, and fixes to lots of issues reported by SDK users: https://www.youtube.com/watch?v=G-qlQkRJVWo&list=PLRp0viTDxBWGLdZk0aamtahB9cpJGV7ZF

PaulA (Sat, 01 Feb 2020 14:11:13 GMT):
Thanks! changing the privileges fixed it.

PaulA (Sat, 01 Feb 2020 17:17:27 GMT):
@swcurran @ianco I'm having error with syncing von-agent with local indy. Followed the steps, got all correct until this: curl -X POST \ http://localhost:5001/ian-co/issue-credential \ -H 'content-type: application/json' \ -d '[ { "attributes": { "corp_num": "1234567890", "legal_name": "My Test Corp", "permit_id": "123834234234999", "permit_type": "Unlimited Use Authorization", "permit_issued_date": "2017-11-01T08:00:00+00:00", "permit_status": "ACT", "effective_date": "2017-11-01T08:00:00+00:00" }, "schema": "ian-permit.ian-co", "version": "1.0.0" } ] ' result: [{"success": false, "result": "Unknown connection id: ianco"}]

PaulA (Sat, 01 Feb 2020 17:20:05 GMT):
Changed the connection id to match connection_id in route.yml which is bctob, result: [{"success": false, "result": "Connection is not yet synchronized: bctob"}]

PaulA (Sat, 01 Feb 2020 17:26:25 GMT):
And, on starting the agent, I got the followng: myorg-agent_1 | 2020-02-01 17:07:24,763 ERROR [vonx.common.service]: Error during indy sync: Any idea on the configuration i might be missing before starting up the von-agent-? I've tried changing some details and rebuilding the container yet these errors persist

ianco (Sat, 01 Feb 2020 17:40:52 GMT):
@PaulA can you confirm which github repo's you are using?

PaulA (Sun, 02 Feb 2020 03:35:08 GMT):
https://github.com/bcgov/von-agent-template

PaulA (Sun, 02 Feb 2020 03:35:08 GMT):
https://github.com/PaulXchange/von-agent-template https://github.com/PaulXchange/TheOrgBook https://github.com/PaulXchange/von-network https://github.com/bcgov/von-network/blob/master/docs/Writing%20Transactions%20to%20a%20Ledger%20for%20an%20Un-privileged%20Author.md

PaulA (Sun, 02 Feb 2020 04:50:32 GMT):
https://github.com/PaulXchange/von-agent-template https://github.com/PaulXchange/TheOrgBook https://github.com/PaulXchange/von-network https://github.com/bcgov/von-network/blob/master/docs/Writing%20Transactions%20to%20a%20Ledger%20for%20an%20Un-privileged%20Author.md

PaulA (Sun, 02 Feb 2020 04:50:32 GMT):
These: https://github.com/PaulXchange/von-agent-template https://github.com/PaulXchange/TheOrgBook https://github.com/PaulXchange/von-network https://github.com/bcgov/von-network/blob/master/docs/Writing%20Transactions%20to%20a%20Ledger%20for%20an%20Un-privileged%20Author.md

PaulA (Sun, 02 Feb 2020 04:50:32 GMT):
@ianco These: https://github.com/PaulXchange/von-agent-template https://github.com/PaulXchange/TheOrgBook https://github.com/PaulXchange/von-network https://github.com/bcgov/von-network/blob/master/docs/Writing%20Transactions%20to%20a%20Ledger%20for%20an%20Un-privileged%20Author.md

ianco (Sun, 02 Feb 2020 18:22:13 GMT):
Hi @PaulA , in the `von-agent-template` repository, when you run the configuration script `. init.sh` it should update the configuration files in a consistent state. If you have done any manual edits, the configuration files (`config/*.yml`) may be in an inconsistent state and that could be the cause of the error. I suggest re-setting your environment (stop all docker containers and run `./manage rm` in each of the 3 shells), `git stash` any changes you've made in `von-agent-template`, and then re-run the `. init.sh` configuration script (and rebuild your docker containers).

ianco (Sun, 02 Feb 2020 18:23:09 GMT):
A second recommendation - `OrgBook` is the "old" version of the code, the new version - which has been migrated to the new Aries agent - is in `https://github.com/bcgov/indy-catalyst`

ianco (Sun, 02 Feb 2020 18:24:07 GMT):
.. and there is a corresponding sample agent which is in `https://github.com/bcgov/indy-catalyst-issuer-controller`

Logi (Mon, 03 Feb 2020 11:22:29 GMT):
Has joined the channel.

Logi (Mon, 03 Feb 2020 11:22:31 GMT):
Hi All - I am setting up a fabri network & planning to use Hyperledger INdy for identity management. Do we have any existing repo for sample setup review or any tutorials which outlined these integrations?.. thanks much in advance..

esplinr (Mon, 03 Feb 2020 22:50:39 GMT):
We'll be starting the Indy Contributors call in 10 minutes.

esplinr (Mon, 03 Feb 2020 22:52:56 GMT):
https://wiki.hyperledger.org/display/indy/Indy+Contributors+Meeting

esplinr (Mon, 03 Feb 2020 23:02:45 GMT):
https://zoom.us/j/615818107

sanjaysb (Tue, 04 Feb 2020 04:43:43 GMT):
Hi, I am facing an error "org.hyperledger.indy.sdk.InvalidStateException: The SDK library experienced an unexpected internal error." When trying to call openPoolLedger. I am running a react native App on an android device. I have written a Module in Java where I have used the indy java sdk v1.14.1. The loading of the sodium and zmq libraries happen succesfully in 'onCreate'. 'createPoolLedgerConfig' is also working. I am using a node pool on a remote host using docker.

PaulA (Tue, 04 Feb 2020 07:11:42 GMT):
Thanks. I've fixed it by ./manage rm and redoing everything. I will also use the new repo.

sanjaysb (Tue, 04 Feb 2020 07:20:59 GMT):
I have fixed this. It was an issue with how I built my image and run the container. I deleted the containers and built and run using following commands and it started working. ``` ```

sanjaysb (Tue, 04 Feb 2020 07:20:59 GMT):
I have fixed this. It was an issue with how I built my image and run the container. I deleted the containers and built and run using following commands and it started working. ``` docker build --build-arg pool_ip= -f ci/indy-pool.dockerfile -t indy_pool . docker run -itd -p :9701-9708:9701-9708 indy_pool ```

gnarula (Tue, 04 Feb 2020 09:39:12 GMT):
Hi everyone! I've a question about how the verkey in a get_nym transaction response differs from the public keys mentioned in the DID Doc. Does the verkey refer to one of the keys in the did doc or is it something totally different?

gnarula (Tue, 04 Feb 2020 09:46:03 GMT):
also, for a message received from Alice, I reckon Bob should verify the signature using the verkey (as indy_key_for_did suggests). In that case, shouldn't the verkey also exist in the publicKey block of a DID Doc? What happens if it's revoked/rotated? Would get_nym transaction return the new key?

gnarula (Tue, 04 Feb 2020 09:55:20 GMT):
ah, verkey is base58(last 16 bytes) of one of the keys in publicKeys block and did is base58(first 16 bytes). nvm

mirgee (Tue, 04 Feb 2020 12:16:13 GMT):
Regarding my question above (https://chat.hyperledger.org/channel/indy?msg=CQXXvzJ67h95zn7rY), would it make sense to expose `vcx_get_handle` (https://github.com/hyperledger/indy-sdk/blob/master/vcx/libvcx/src/utils/libindy/wallet.rs#L18) as part of VCX API so that applications (like ours), which want to use libindy functions which need to access the wallet used by VCX?

mirgee (Tue, 04 Feb 2020 12:16:13 GMT):
Regarding my question above (https://chat.hyperledger.org/channel/indy?msg=CQXXvzJ67h95zn7rY), would it make sense to expose `vcx_get_wallet_handle` (https://github.com/hyperledger/indy-sdk/blob/master/vcx/libvcx/src/utils/libindy/wallet.rs#L18) as part of VCX API so that applications (like ours), which want to use libindy functions which need to access the wallet used by VCX?

droconnel22 (Tue, 04 Feb 2020 13:09:58 GMT):
Hi All ihave a question on linking using the Java wrapper

droconnel22 (Tue, 04 Feb 2020 13:10:13 GMT):
I am building a nym request and when I sign and accept the request I get this error:

droconnel22 (Tue, 04 Feb 2020 13:10:16 GMT):
{"reqId":1580821692382922000,"identifier":"BwrCbGRF6hfxWYzHkoRCbx","reason":"client request invalid: CouldNotAuthenticate('Can not find verkey for BwrCbGRF6hfxWYzHkoRCbx',)","op":"REQNACK"}

droconnel22 (Tue, 04 Feb 2020 13:10:37 GMT):
is there something I am doing when my new account is requesting a link to a trustee/steward account

droconnel22 (Tue, 04 Feb 2020 13:10:48 GMT):
for both i open/create wallets

droconnel22 (Tue, 04 Feb 2020 13:10:54 GMT):
create new did/verkey pairs

droconnel22 (Tue, 04 Feb 2020 13:11:00 GMT):
construct the num request

droconnel22 (Tue, 04 Feb 2020 13:11:12 GMT):
I don't get an error, just a successful failure

droconnel22 (Tue, 04 Feb 2020 13:11:24 GMT):
has anyone encountered this and if so what do you recommend?

droconnel22 (Tue, 04 Feb 2020 13:11:25 GMT):
thank you all

PaulA (Tue, 04 Feb 2020 15:25:11 GMT):
@ianco Trying to write to the staging net but i get this error. Transaction has been rejected: client request invalid: insufficient number of valid signatures, 1 is required but 0 valid and 1 invalid have been provided. The following signatures are invalid: did=3KnkCS2qrqjk9TMRFX2BwM, signature=56En8qSuZtHrtUV8aV9wH5gmbmNBJaBJjbyuuczBux8TDKxovE6crpAaWCpzm4Faxv47qfTKVDDo74bSfcDHXvCe

PaulA (Tue, 04 Feb 2020 15:25:30 GMT):
Despite the fact that i created a schema and signed the transaction.

PaulA (Tue, 04 Feb 2020 15:26:09 GMT):
What could i have missed out?

PaulA (Tue, 04 Feb 2020 15:32:36 GMT):
Transaction has been rejected: validation error [SafeRequest]: Endorser must sign the request

PaulA (Tue, 04 Feb 2020 15:32:41 GMT):
Another error.

gnarula (Tue, 04 Feb 2020 15:54:38 GMT):
what's the format for the `endpoint` attribute in sovrin? ACA-Py sort of assumes `endpoint: { 'endpoint': 'url' }` where the Service block in the DIDDoc gets a service of type "IndyAgent" and the given url. How would that work for multiple endpoints?

ianco (Tue, 04 Feb 2020 16:10:46 GMT):
It's missing a signature, are you sure you followed all the steps?

PaulA (Tue, 04 Feb 2020 16:13:02 GMT):
Hi @ianco I'm following the steps from this link: https://github.com/bcgov/von-network/blob/master/docs/Writing%20Transactions%20to%20a%20Ledger%20for%20an%20Un-privileged%20Author.md I hope it's up to date. I'm using same did as author and endorser.

PaulA (Tue, 04 Feb 2020 17:56:15 GMT):
@ianco on an open wallet, do i need to use the exact endorser's did registered on staging net to write or read from the staging net? also, must endorsement be done using the same did ?

ianco (Tue, 04 Feb 2020 17:57:03 GMT):
The endorsement needs to be done by a DID with write permissions to staging net

ianco (Tue, 04 Feb 2020 17:57:36 GMT):
In this example (https://github.com/bcgov/von-network/blob/master/docs/Writing%20Transactions%20to%20a%20Ledger%20for%20an%20Un-privileged%20Author.md) there are two wallets and two did's involved

ianco (Tue, 04 Feb 2020 17:57:52 GMT):
One is the transaction author, and one is the endorser

ianco (Tue, 04 Feb 2020 17:58:24 GMT):
I think the transaction author can write the transaction once it's endorsed (?) ... I don't recall all the details offhand

PaulA (Tue, 04 Feb 2020 18:02:16 GMT):
@ianco I have a did and verkey with endorsement rights on the staging net although i've lost the keys for the original wallet i used in creating that did. Would I still be able to use the did with a new wallet provided i am still in possession of the seeds?

ianco (Tue, 04 Feb 2020 18:02:43 GMT):
Yes you can recreate the DID from the seed

PaulA (Tue, 04 Feb 2020 18:02:57 GMT):
when i recreate, the Verkey changes.

PaulA (Tue, 04 Feb 2020 18:03:36 GMT):
it's not the same as the original, and i'm thinking that could be affecting something (?)

ianco (Tue, 04 Feb 2020 18:03:38 GMT):
Is it a different format of verkey? (i.e. with or without a leading `~`?)

ianco (Tue, 04 Feb 2020 18:04:06 GMT):
I'm not sure what is the impact of having a different verkey

PaulA (Tue, 04 Feb 2020 18:05:18 GMT):
yes, the verkey starts with ~

ianco (Tue, 04 Feb 2020 18:06:22 GMT):
AFAIK when you generate a DID from a seed the verkey should be the same, HOWEVER there are two formats of verkey, and that's about the limit of my knowledge

ianco (Tue, 04 Feb 2020 18:06:37 GMT):
Try creating a DID using the CLI and see what the verkey looks like

ianco (Tue, 04 Feb 2020 18:06:43 GMT):
(indy CLI)

PaulA (Tue, 04 Feb 2020 18:08:17 GMT):
I've created many did and they all start with ~

PaulA (Tue, 04 Feb 2020 18:11:40 GMT):
?

PaulA (Tue, 04 Feb 2020 18:13:35 GMT):
Do i keep using this indy-cli or aca-py cli?

ianco (Tue, 04 Feb 2020 18:14:42 GMT):
It should work either way, I don't know offhand if there is any difference

PaulA (Tue, 04 Feb 2020 18:18:18 GMT):
Ok thanks. Will keep trying.

MALodder (Tue, 04 Feb 2020 22:30:29 GMT):
I’m going to be upgrading the Jenkins CI/CD tonight.

nath1014 (Wed, 05 Feb 2020 08:50:24 GMT):
Has joined the channel.

iamShashvat (Wed, 05 Feb 2020 11:10:58 GMT):
how can I install libindy in ubuntu 18? "sudo apt-get install -y libindy" unable to find libindy

SigmaS 1 (Wed, 05 Feb 2020 11:33:46 GMT):
https://github.com/hyperledger/indy-sdk#ubuntu-based-distributions-ubuntu-1604-and-1804

Logi (Wed, 05 Feb 2020 11:49:50 GMT):
Try this -> https://medium.com/akeo-tech/step-by-step-guide-to-set-up-hyperledger-indy-64eeb524f558 and also see if you have the 'pip' installed already

ayushmanss (Wed, 05 Feb 2020 12:38:19 GMT):
Hi guys, I am developing an application based on indy. I have already written an agent using aries, but up until now was using test indy pools for the same. Need help running a production pool of indy. Was following the docs but need some clarity. 1 - Is generate_indy_pool_transactions the only way to generate domain and pool txn. Cause it feels this script was created to be only used in test environments 2 - If not generate_indy_pool_transactions then how to generate the same after initializing nodes using init_indy_node 3 - What is the purpose of seed in init_indy_node? 4 - I was able to run the network using generate_indy_pool_transactions, but how can i get the possesion of trustee did, so I can configure the network more 5 - My current implementation of production network is as follows. Let me know if this is the best way to go about it - I spin up a gcp node. But I am using a docker container to run the indy-node with proper volume mounting. If you guys can point me to some guide on how to run, manage your own production indy network that will be very helpful. Thank you in advance

lynn.bendixsen (Wed, 05 Feb 2020 19:44:40 GMT):
Hi Ayushman, I think what you are looking for is https://github.com/sovrin-foundation/steward-tools/tree/master/create_genesis That answers questions 1, 2, and 4 for you, but I can clarify more if needed. (All Trustee DIDs are created by you and put into the file as part of the script) 3- the seed returned by init-indy-node is the seed used to generate the keys for that node. If your node needs re-installed, add the seed as the last parameter to the same init-indy-node command that you ran the first time, and it will generate the same seeds as when you ran it the first time. 5- Your implementation should work, but be careful to make sure you use sufficient Memory (32G/node is recommended and tests have shown issues when using 8G or less) and make sure you don't run out of disk space. I have seen cases where out-of-disc-space problems give unrelated error messages that are hard to track down. The Sovrin production Nodes that I coordinate are each on a different server and I know of other successful production networks that do the same, so that is what I would recommend : 1 node per server and avoid using Docker if possible (it adds unnecessary overhead?).

iamShashvat (Thu, 06 Feb 2020 06:51:23 GMT):
thank you SigmaS and Logi for reply, I resolve the issue it was a public key issue, firewall blocked the connection, So, there were some issue, I will update/PR the document soon regarding this.

anillewis (Thu, 06 Feb 2020 14:50:40 GMT):
Team question on when an identity receives a connection request e.g How does an identity Jim know that the connection request is sent by an employer like Google... I am assuming the pairwise did and verkey sent by Google is published in the ledger so that Jim can verify that in the ledger and not Google's trust anchor did and verkey

SigmaS 1 (Thu, 06 Feb 2020 14:54:44 GMT):
isn't a bad choice to sore the pairwise did (even from the Issuer side) in the ledger? I think Jim verifies Google from its public did through the ledger

SigmaS 1 (Thu, 06 Feb 2020 14:54:44 GMT):
isn't a bad choice to store the pairwise did (even from the Issuer side) in the ledger? I think Jim verifies Google from its public did through the ledger

anillewis (Thu, 06 Feb 2020 14:55:56 GMT):
if Jim verifies Google from the public DID, what prevents Oracle to send Google's public DID and the send their DID and verkey in the pairwise object?

anillewis (Thu, 06 Feb 2020 14:56:40 GMT):
now Jim will be exchanging messages with Oracle thinking it is exchanging messages with Google

SigmaS 1 (Thu, 06 Feb 2020 15:00:29 GMT):
I am new to Indy, however I believe that Jim will send he's pairwise to Google's did endpoint and he will cryptoAnonCrypt first with Google's pairwise and second with Google's public. So even Oracle has Google dids, it can go any further and establish a relationship with Jim

anillewis (Thu, 06 Feb 2020 15:02:05 GMT):
this is good when the connection request is issued from Jim to Google but what about the other way ?

sukalpomitra (Thu, 06 Feb 2020 15:22:43 GMT):
@anillewis if Oracle passes Google's public did, the did auth protocol will fail...afaik

anillewis (Thu, 06 Feb 2020 15:23:23 GMT):
how?

sukalpomitra (Thu, 06 Feb 2020 15:26:20 GMT):
Did auth if I am not wrong is kind of a challenge response thing, where Jim will encrypt a small text with public key found in the public did and will expect to decrypt with its pvt key. Which Oracle will fail

anillewis (Thu, 06 Feb 2020 15:29:39 GMT):
see https://github.com/hyperledger/aries-rfcs/issues/144#issuecomment-567216211 .. i think i got my answer and has nothing to do with did auth protocol...see the solution recommended by talltree

sheldon.regular (Thu, 06 Feb 2020 21:03:01 GMT):
Has joined the channel.

ayushmanss (Fri, 07 Feb 2020 08:23:24 GMT):
Thank you so much for the info. Will it be better if we link these tools in some documentation of indy-node.

ayushmanss (Fri, 07 Feb 2020 08:25:54 GMT):
Also what is the best way to generate Dids using seeds? One way to go is using the sdk, create a wallet and create and store the did in the wallet. But it seems like an overkill for such a simple task. Or will it be better if I use plenum's DidSigner class to generate did?

ayushmanss (Fri, 07 Feb 2020 08:46:01 GMT):
plenum did the job :)

swcurran (Fri, 07 Feb 2020 17:40:27 GMT):
Should there be an option to create a Credential Definition with a seed? We do that for DIDs, but not for Cred Defs. It does make some things easier to manage.

tomislav (Fri, 07 Feb 2020 21:01:55 GMT):
It'll definitely add more portability for issuers. I'm not sure the crypto used for cred def supports determinism.

tomislav (Fri, 07 Feb 2020 21:01:55 GMT):
It'll definitely add more portability for issuers. I'm not sure the crypto used for cred def supports deterministic generation, as it used different method.

swcurran (Fri, 07 Feb 2020 21:06:44 GMT):
It starts from a random seed vs. a parameter. As such, pretty sure it would work.

andrew.whitehead (Fri, 07 Feb 2020 21:07:36 GMT):
Would be useful for testing

MALodder (Sat, 08 Feb 2020 16:35:55 GMT):
For anoncreds 1.0 this wouldn't work

MALodder (Sat, 08 Feb 2020 16:36:03 GMT):
for anoncreds 2.0 this would work

MALodder (Sat, 08 Feb 2020 16:36:11 GMT):
the hard part is generating safe primes

MALodder (Sat, 08 Feb 2020 16:36:20 GMT):
which is what anoncreds 1.0 depends

MALodder (Sat, 08 Feb 2020 16:44:08 GMT):
anoncreds 2.0 should make this easier

MALodder (Sat, 08 Feb 2020 16:44:33 GMT):
because all the parameters are based on the curve so all you have to do is seed the CSPRNG to generate the keys

MALodder (Sat, 08 Feb 2020 16:44:54 GMT):
this is not the first time I've been asked about this

iamShashvat (Mon, 10 Feb 2020 05:58:46 GMT):
payment

Logi (Mon, 10 Feb 2020 08:30:02 GMT):
Hi - have anyone tried to use Indy as custom msp in hyperledger fabric?..

sukalpomitra (Mon, 10 Feb 2020 11:18:02 GMT):
that would be really cool

geocontrol (Mon, 10 Feb 2020 12:21:18 GMT):
Has joined the channel.

gnarula (Mon, 10 Feb 2020 13:24:27 GMT):
Hi everyone! I'm not able to access indy-sdk maven artifacts from https://repo.sovrin.org/repository/maven-public. Is the repository down?

droconnel22 (Mon, 10 Feb 2020 14:36:29 GMT):
Hi everyone !! Does there exist any talks, presentations, or wikis on how to Indy Agents and wallets and what happens if they get hacked, best practices for storage, or why they are safer then traditional central authorities?

gnarula (Mon, 10 Feb 2020 14:41:43 GMT):
@swcurran I see some verkeys begin with a `~` and some don't (and they're usually 32 bytes long). Is the convention that keys beginning with ~ represent the shorthand version?

lynn.bendixsen (Mon, 10 Feb 2020 14:57:55 GMT):
yes, the ~ represents the shorthand version

gnarula (Mon, 10 Feb 2020 14:58:16 GMT):
Thanks!

esplinr (Mon, 10 Feb 2020 16:03:05 GMT):
Starting the Indy Contributors call: https://zoom.us/j/615818107 https://wiki.hyperledger.org/display/indy/2020-02-10+Indy+Contributors+Call

esplinr (Mon, 10 Feb 2020 18:25:58 GMT):
Recording is posted to the agenda page.

esplinr (Mon, 10 Feb 2020 18:41:03 GMT):
https://sovrin.org/library/lost-phone/

esplinr (Mon, 10 Feb 2020 18:41:46 GMT):
And lots of good stuff here: https://ssimeetup.org/

droconnel22 (Mon, 10 Feb 2020 18:42:02 GMT):
@esplinr this is great thank you so much : D

esplinr (Mon, 10 Feb 2020 18:42:03 GMT):
I was told that they are doing maintenance today.

esplinr (Mon, 10 Feb 2020 23:09:17 GMT):
Per our discussion in the Indy Contributors call this morning, we will be having two special calls this month: * Tuesday, February 11 @ 15H UTC to collaborate on W3C Verifiable Credentials / Indy Rich Schemas (*Note this is Tomorrow*) * Tuesday, February 25 @ 15H UTC to discuss how best to support an Indy Network

esplinr (Mon, 10 Feb 2020 23:09:17 GMT):
Per our discussion in the Indy Contributors call this morning, we will be having two special calls this month: * Tuesday, February 11 @ 15H UTC to collaborate on W3C Verifiable Credentials / Indy Rich Schemas * Tuesday, February 25 @ 15H UTC to discuss how best to support an Indy Network

esplinr (Mon, 10 Feb 2020 23:24:31 GMT):
I updated the agendas in the wiki, and the Hyperledger calendar.

DeepakMaurya (Tue, 11 Feb 2020 07:54:00 GMT):
Hi team , I'm new in hyperledger indy. please explain how to start . any git repo(JAVA) for beginner , please share with me.

JeromeK (Tue, 11 Feb 2020 10:55:29 GMT):
I recently looked into elixir: https://elixir-lang.org/ Which looks to me like a language that Indy could benefit of. What are your thoughts about Elixir together with Indy, would it makes sence to write a new wrapper? Looking forward for some discussen :) Greets Jerome

JeromeK (Tue, 11 Feb 2020 10:55:29 GMT):
I recently looked into elixir: https://elixir-lang.org/ Which looks to me like a language that Indy could benefit of. What are your thoughts about Elixir together with Indy and maybe Aries, would it makes sence to write a new wrapper? Looking forward for some discussen :) Greets Jerome

JeromeK (Tue, 11 Feb 2020 10:55:29 GMT):
I recently looked into elixir: https://elixir-lang.org/ Which looks to me like a language that Indy could benefit of. What are your thoughts about Elixir together with Indy and maybe Aries, would it makes sence to write a new wrapper? Looking forward to talk about this topic :) Greets Jerome

thanga_troondx (Tue, 11 Feb 2020 11:19:38 GMT):
Has joined the channel.

thanga_troondx (Tue, 11 Feb 2020 11:19:43 GMT):
can we store indy wallet credentials in other formats rather than storing it as sqlite.db & sqlite.db-wal. can i store it in .pem or .cert format ? @JeromeK @esplinr @swcurran @VithushanJegatheeswaran

aravindavk (Tue, 11 Feb 2020 11:36:09 GMT):
Yes, possible by creating custom storage methods https://docs.rs/indy-sys/1.14.2/indy_sys/wallet/fn.indy_register_wallet_storage.html

aravindavk (Tue, 11 Feb 2020 11:38:40 GMT):
example in aca-py https://github.com/hyperledger/aries-cloudagent-python/blob/master/aries_cloudagent/storage/provider.py#L15-L19

thanga_troondx (Tue, 11 Feb 2020 11:51:32 GMT):
but in .cert or .pem formats ?

JeromeK (Tue, 11 Feb 2020 12:18:29 GMT):
I think you can simply create a Schema with exp: `cert` and `pem` attributes

JeromeK (Tue, 11 Feb 2020 12:19:14 GMT):
and then fill in the values as needed

thanga_troondx (Tue, 11 Feb 2020 12:41:22 GMT):
@aravindavk can you explain a bit please ?

sukalpomitra (Tue, 11 Feb 2020 13:24:22 GMT):
I agree with Jerome. Best way is to screate schema with cert and pem attri butes

aravindavk (Tue, 11 Feb 2020 16:21:51 GMT):
@thanga_troondx as per my understanding add_record(or similar) function can save as cert and pem instead of storing the values in db

JeromeK (Tue, 11 Feb 2020 16:58:08 GMT):
add_record also stores the values into the wallet as non_secrets

NikhilPrakash (Wed, 12 Feb 2020 00:44:20 GMT):
Has joined the channel.

iamShashvat (Wed, 12 Feb 2020 03:56:03 GMT):
I have a doubt, is it necessary to have any specific numbers of validator/observer nodes to start the network? If no, let's say in a new pool no nodes are available and I have to make a steward then who will going to valid the transaction?(In a real life scenario 6 players want to start the network all wanted to be steward, what will be the flow?)

thanga_troondx (Wed, 12 Feb 2020 04:53:58 GMT):
I couldn't find add_record in indy-sdk..

amarts (Wed, 12 Feb 2020 05:15:02 GMT):
Thanks for the update @esplinr . It is still not up and running.

sanjaysb (Wed, 12 Feb 2020 06:03:47 GMT):
is the https://repo.sovrin.org/repository/maven-public repo down?

sanjaysb (Wed, 12 Feb 2020 06:04:04 GMT):
I am unable to complete gradle build

pranav_kirtani (Wed, 12 Feb 2020 07:25:02 GMT):
Has joined the channel.

pranav_kirtani (Wed, 12 Feb 2020 07:27:29 GMT):
Dont know

droconnel22 (Wed, 12 Feb 2020 11:51:48 GMT):
Hi All, are there guidelines for wallet id and key best practices? Im assuming they should be base64 encoding if ever persisted?

dgunseli (Wed, 12 Feb 2020 12:16:48 GMT):
Hello all, As I saw a proof can be verified with only the proof request which contains same nonce value, Is there any way to create proof without any request for disclosuring some credentials that I've own already to publicly?

NoushadMohamed (Wed, 12 Feb 2020 13:26:13 GMT):
Has joined the channel.

NoushadMohamed (Wed, 12 Feb 2020 13:26:14 GMT):
What are the variables that determine the number of nodes required in Hyperledger Indy Setup?

NoushadMohamed (Wed, 12 Feb 2020 13:26:32 GMT):
As part of one of my project implementation, I have to use Hyperledger Indy for providing digital identity to users. The estimated number of users on the platform is 20k. I am stuck on determining what must be the number of nodes required to run Hyperledger Indy for efficient use on production.

gut (Wed, 12 Feb 2020 15:10:49 GMT):
hi all, im having a segmentation fault in indy-sdk when trying to create a pool with a random name a valid example configuration data. Is this issue already been reported or am I missing something?

dgunseli (Wed, 12 Feb 2020 15:41:47 GMT):
I guess I have to explain a little bit more

dgunseli (Wed, 12 Feb 2020 15:48:23 GMT):
Let’s consider this, users can create proofs with proof request and proof requester can verify the proof with the specific request that they provided

dgunseli (Wed, 12 Feb 2020 15:49:43 GMT):
But is there any way to accomplish this scenario, user want to create a proof and stream this proof

esplinr (Wed, 12 Feb 2020 16:12:21 GMT):
@MALodder said that he is aware of the problem, but hasn't been able to solve it yet.

MALodder (Wed, 12 Feb 2020 16:12:36 GMT):
I'm looking at it right now if that helps

MALodder (Wed, 12 Feb 2020 16:12:41 GMT):
haven't solved it yet

swcurran (Wed, 12 Feb 2020 16:12:49 GMT):
I think if I understand the question, the answer is yes. The prover can create their own proof request and proof if they want. The nonce is to prevent replays of a proof, so if the prover is producing the request and proof, that protection is lost.

swcurran (Wed, 12 Feb 2020 16:13:55 GMT):
Even then, if a verifier would to just give the nonce and assume the prover knows the rest of the proof request, I think even that could be accomplished.

swcurran (Wed, 12 Feb 2020 16:14:08 GMT):
Let me know if I have misunderstood the question.

VithushanJegatheeswaran (Wed, 12 Feb 2020 16:15:13 GMT):
Hey there is a java wrapper for indy-sdk. check out the how-tos: https://github.com/hyperledger/indy-sdk/tree/master/docs/how-tos

swcurran (Wed, 12 Feb 2020 16:17:09 GMT):
The number of nodes depends on the number of faulty nodes you want to permit. To support f faults in the system, you need 3f+1 total nodes. So if you are OK with only 2 nodes failing at any one time, you need f=2 or 7 total nodes. The practical limit has found to be f=8 or 25 total nodes. That balances a robust network with the speed of consensus.

swcurran (Wed, 12 Feb 2020 16:18:31 GMT):
How that relates to the number of users on the platform is a different question. It depends on (a) how many issuers, and (b) how much revocation is happening.

swcurran (Wed, 12 Feb 2020 16:19:42 GMT):
If there is only one issuer and no revocation, there would only be 3 extra transactions beyond the genesis transactions. Not many...

amarts (Wed, 12 Feb 2020 16:39:17 GMT):
Thanks for the update @esplinr @MALodder :-) It helps

MALodder (Wed, 12 Feb 2020 17:37:10 GMT):
That server is unavailable right now and I’m trying to diagnose why it went down and why I can’t get it back up

iamShashvat (Wed, 12 Feb 2020 18:53:45 GMT):
Let's say in a new private network no nodes are available and I have to make a steward then who will going to valid the transaction? Also in private network can anybody become a steward? In a real life scenario 6 players want to start the network all wanted to be steward, what will be the flow?

MALodder (Wed, 12 Feb 2020 19:22:27 GMT):
Aha the server did go down and our auto fix solution has repairs it 90%. Just trying to fix the final parts

dgunseli (Wed, 12 Feb 2020 19:42:02 GMT):
Thanks @swcurran , It's about exactly my question, we implemented this solution and we were looking for a native way, but it's seems like it's a native solution

esplinr (Wed, 12 Feb 2020 19:48:09 GMT):
Thank you Mike.

esplinr (Wed, 12 Feb 2020 19:49:27 GMT):
How do you have a network with no nodes?

esplinr (Wed, 12 Feb 2020 19:49:45 GMT):
When you create the network, you establish a couple of trustee DIDs. They can make stewards.

TimoGlastra (Wed, 12 Feb 2020 19:59:56 GMT):
Has joined the channel.

MALodder (Wed, 12 Feb 2020 20:14:06 GMT):
It is back up now

MALodder (Wed, 12 Feb 2020 20:14:14 GMT):
sorry that took so long

mccown (Wed, 12 Feb 2020 22:50:42 GMT):
Just a reminder about the Identity Implementer's WG Call tomorrow. This is a chance to get updates on the WG calls you may have missed this week. Our guest, John Callahan (Veridium, @johncallahan ), will be giving an overview of the Aries Ruby SDK. Here is the calendar link with times and Zoom details. https://tinyurl.com/s3detbc

iamShashvat (Thu, 13 Feb 2020 03:40:59 GMT):
Thank you!

Yoroitchi (Thu, 13 Feb 2020 08:21:24 GMT):
Hello, One of indy's dependencies is ZeroMQ, does this library is only used for connection with the ledger in client-server scenario or is it also used for connection between 2 clients ? i-e Is it only used to connect an actor (holder, verifier or issuer) with a node or is it also used for connection between 2 actors ?

jrallen (Thu, 13 Feb 2020 09:45:03 GMT):
See also: https://github.com/sovrin-foundation/steward-tools/tree/master/create_genesis

ashcherbakov (Thu, 13 Feb 2020 13:40:51 GMT):
ZeroMQ is used to client-to-node (client-server) and node-to-node (server-to-server). ZeroMQ is not used for connections between 2 actors. In principle, any transport can be used for such connections.

Yoroitchi (Thu, 13 Feb 2020 14:50:06 GMT):
Thank you !

pschwarz (Thu, 13 Feb 2020 15:50:43 GMT):
Has left the channel.

Silona (Thu, 13 Feb 2020 18:24:51 GMT):
Howdy Contributors and Maintainers! Are you wondering about tapping into Developer marketing for your group or project? Do you have a blog post idea? An awesome announcement? Please attend our Contributor/marketing meeting! https://wiki.hyperledger.org/display/Marketing/2020-02-19+Meeting+notes

lmtriet (Thu, 13 Feb 2020 21:26:22 GMT):
genesis

hyper-sunder (Fri, 14 Feb 2020 06:48:20 GMT):
Hey Guys, I am trying to issue two different credentials to one user. Is possible to do?

hyper-sunder (Fri, 14 Feb 2020 06:49:10 GMT):
I am able to do for singe schema, but i am not able to issue credential for another schema. I dont know why

hyper-sunder (Fri, 14 Feb 2020 06:49:19 GMT):
Can anyone help me on this?

NoushadMohamed (Fri, 14 Feb 2020 08:57:52 GMT):
Ok Thank you

shamzarmy (Sat, 15 Feb 2020 22:52:31 GMT):
Has joined the channel.

hyper-sunder (Mon, 17 Feb 2020 13:27:41 GMT):
Hi Guys, Having a very basic doubt.... Should we need to create `schema-def` for every issue? or single `schema-def` and use the `handler` to issue multiple time? The reason for this question is, getting the following error `Invalid Credential Definition handle` when try to issue. This def has been created 3 days back

lauravuo-techlab (Mon, 17 Feb 2020 14:23:45 GMT):
Issuer defines credential definition (for a single schema). Then the issuer can issue multiple different credentials for this single credential definition. Note that you have to issue the credentials with the same wallet you have created the credential definition with.

hyper-sunder (Mon, 17 Feb 2020 14:24:37 GMT):
thanks a lot @lauravuo-techlab

futurama92 (Mon, 17 Feb 2020 20:17:57 GMT):
Has joined the channel.

futurama92 (Mon, 17 Feb 2020 20:20:12 GMT):
Hi guys, i'm trying to send a message between trust anchor and user with packMessage() method. Is it possibile that someone can steal this message and does a replay attack?

Silona (Mon, 17 Feb 2020 22:18:36 GMT):
Are you wondering about tapping into Developer marketing for your group or project? Do you have a blog post idea? An awesome announcement? Please attend TOMORROW! https://wiki.hyperledger.org/display/Marketing/2020-02-19+Meeting+notes

iamShashvat (Tue, 18 Feb 2020 04:45:07 GMT):
how can I use libindy apis using c only? Do I have to make wrapper for it(I think its' not required!)? if you have some example please share! Thanks!

sanjaysb (Tue, 18 Feb 2020 05:17:51 GMT):
what is the encoding used to encode a credential value. In the example I can see ``` let transcriptCredValues = { "first_name": {"raw": "Alice", "encoded": "1139481716457488690172217916278103335"}, ... ```

DearKen (Tue, 18 Feb 2020 06:47:08 GMT):
Has joined the channel.

DearKen (Tue, 18 Feb 2020 06:47:08 GMT):
Hi Hyperledger community!! I have some questions! I'm researching indy-crypto and ursa now. I heard that the zero-knowledge proof (in indy and ursa) verify the committed value (attributes) over the point (do not verify under value) using range proof. Is it true? Is these cryptosystem based on integer? (How can verify non-integer value? - trasnform?) How can I find relative documents?

lovesh (Tue, 18 Feb 2020 11:30:41 GMT):
The range proof works over elements of finite field, so others like fractional or negative values have to be transformed appropriately. eg if we are working in a 256 bit finite field and we want to do range proofs over 32 bit negative values like prove [2^{-32} < x < 2^{32}, there would be 2 range proofs, 0 <= x < 2^{32} and 2^{224} < x < 2^256 (-ve values wrap around and map to the larger field elements). We don't have such examples in Ursa as of now. We use Bulletproofs and here is the range proof code https://github.com/hyperledger/ursa/blob/master/libzmix/bulletproofs_amcl/src/r1cs/gadgets/bound_check.rs

lovesh (Tue, 18 Feb 2020 11:30:41 GMT):
The range proof works over elements of finite field, so others like fractional or negative values have to be transformed appropriately. eg if we are working in a 256 bit finite field and we want to do range proofs over 32 bit negative values like prove `2^{-32} < x < 2^{32}`, there would be 2 range proofs, 0 <= x < 2^{32} and 2^{224} < x < 2^256 (-ve values wrap around and map to the larger field elements). We don't have such examples in Ursa as of now. We use Bulletproofs and here is the range proof code https://github.com/hyperledger/ursa/blob/master/libzmix/bulletproofs_amcl/src/r1cs/gadgets/bound_check.rs

lovesh (Tue, 18 Feb 2020 11:30:41 GMT):
The range proof works over elements of finite field, so others like fractional or negative values have to be transformed appropriately. eg if we are working in a 256 bit finite field and we want to do range proofs over 32 bit negative values like prove `2^{-32} < x < 2^{32}`, there would be 2 range proofs, 0 <= x < 2^{32} and 2^{224} < x < 2^{256} (-ve values wrap around and map to the larger field elements). We don't have such examples in Ursa as of now. We use Bulletproofs and here is the range proof code https://github.com/hyperledger/ursa/blob/master/libzmix/bulletproofs_amcl/src/r1cs/gadgets/bound_check.rs

puria (Tue, 18 Feb 2020 14:35:50 GMT):
Has joined the channel.

esplinr (Tue, 18 Feb 2020 18:12:03 GMT):
My team posted a new video discussing our recent contributions to Indy: * A tool for testing ZMQ connectivity * Documentation for troubleshooting Indy networks * Overview RFC and HIPE for Rich Schemas that explains how to proceed with the implementation. https://www.youtube.com/watch?v=X1JAvhUDfNM&list=PLRp0viTDxBWGLdZk0aamtahB9cpJGV7ZF

sownak (Wed, 19 Feb 2020 05:34:58 GMT):
Has joined the channel.

iamShashvat (Wed, 19 Feb 2020 06:13:27 GMT):
how can I use libindy apis using c only? Do I have to make wrapper for it(I think its' not required!)? if you have some example please share

JHamilton (Wed, 19 Feb 2020 06:20:30 GMT):
pardon reviving an old thread :) Here is a demo of Hyperledger Labs Blockchain Automation Framework deploying an Indy network, which we understand to be different from von-network deploying an application to an existing Indy network. https://wiki.hyperledger.org/download/attachments/24781016/zoom_0.mp4?version=1&modificationDate=1578964535000&api=v2

JHamilton (Wed, 19 Feb 2020 06:20:55 GMT):
Please let us know of feedback on how this could become more useful for the Indy community! We are also reachable on #blockchain-automation-framework

iamShashvat (Wed, 19 Feb 2020 10:57:34 GMT):
I am writing a program in C, using libindy include files (.h header files and libindy.so). I stuck at command_handle parameter asked in indy_create_pool_ledger_config(indy_handle_t command_handle, const char * config_name, const char * config,void (*cb)(indy_handle_t command_handle_, indy_error_t err); What should I pass? Is there any example out there?

tomislav (Wed, 19 Feb 2020 14:05:05 GMT):
Yes, there are menu examples. Check the indy-sdk repo for unit tests and code documentation. The parameter `config_name` is looking for a name of configuration (of your choice)

VithushanJegatheeswaran (Wed, 19 Feb 2020 14:26:56 GMT):

Docker logs

VithushanJegatheeswaran (Wed, 19 Feb 2020 14:29:19 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=JmQ4Da5pTzkRoQbLH) Hey guys, I'm having trouble running the verification service of the iiw-book example `iiw-book-agent_1 | aries_cloudagent.ledger.error.LedgerError: Credential definition 4QxzWk3ajdnEA37NdNU5Kt:3:CL:1068:default is on ledger default but not in wallet default` `iiw-book-service_1 | INFO 2020-02-19 14:24:03,473 apps 30 140098540341056 500 Internal Server Error` I also included a log file, but I think that these two lines above describe the problem very good.

swcurran (Wed, 19 Feb 2020 15:53:27 GMT):
That likely means that you ran the test previously, deleted your wallet but not your ledger (I'm guessing you are running von-network?). Reset everything and it should work again. Unlike DIDs, you can't use a seed to create a Cred Def, so once created, they can't be recreated, so you must keep your wallet and ledger in sync,

swcurran (Wed, 19 Feb 2020 15:53:27 GMT):
That likely means that you ran the test previously, deleted your wallet but not your ledger (I'm guessing you are running von-network?). Reset everything and it should work again. Unlike DIDs, you can't use a seed to create a Cred Def, so once created, they can't be recreated, so you must keep your wallet and ledger in sync.

iamShashvat (Thu, 20 Feb 2020 05:04:23 GMT):
Thank you for reply, I'll check that, but I am still confused what should I put in command_handle (int32_t)?

tomislav (Thu, 20 Feb 2020 13:31:57 GMT):
Incremental int32. The library will pass that handle back as a reference, in the callback method.

VithushanJegatheeswaran (Thu, 20 Feb 2020 17:54:32 GMT):
Hey, I'm using the BCovrin Test network (http://138.197.138.255/) What do you mean by "reset everything" should I delete all my docker containers and images ? I cloned both repos (email-verification-service and iiw-book) and followed the readme file.

swcurran (Thu, 20 Feb 2020 18:03:42 GMT):
Probably should take this to #aries-cloudagent-python as the devs there can help better. My guess is that it is starting with an explicit seed for the DID, and that DID is then creating the other objects. I would guess if you start with a different seed, all would work, but that's a guess.

swcurran (Thu, 20 Feb 2020 18:03:42 GMT):
Probably should take this to #aries-cloudagent-python as the devs there can help better. My guess is that it is starting with an explicit seed for the DID, and that DID is then creating the other objects. I would guess if you start with a different seed, all would work. Not certain, but worth a try. If not - jump over to the other channel.

VithushanJegatheeswaran (Thu, 20 Feb 2020 18:05:26 GMT):
I already tried that I changed the seed, registered the seed on the BCovrin Test network. bust still same error Thank you anyways. I will ask over there

VithushanJegatheeswaran (Thu, 20 Feb 2020 18:18:59 GMT):
Thank you. It worked. I just tried again with a new seed.

nacerix (Sun, 23 Feb 2020 14:32:47 GMT):
Has joined the channel.

iamShashvat (Sun, 23 Feb 2020 18:49:40 GMT):
I have multiple questions at concept level: 1. I have check out the APIs I didn't find any API to add more public keys in DID doc. Is it possible? it should be according to the standards! 2. Is there any way by which I can revoke the DID itself? 3. In a situation, A device creating DID for itself (off ledger) and this device can’t do the NYM‌ request/transaction itself, It required some third service to do the transaction on behalf. So, what should I do? Do I require Agents or Endorsers to do the work?

iamShashvat (Sun, 23 Feb 2020 18:49:40 GMT):
I have multiple questions at concept level: 1. I have check out the APIs I didn't find any API to add more public keys in DID doc. Is it possible? it should be according to the standards! 2. Is there any way by which I can revoke the DID itself? 3. In a situation, A service creating DID for itself (off ledger) and this service can’t do the NYM‌ request/transaction itself, It required some third service to do the transaction on behalf. So, what should I do? Do I require Agents or Endorsers to do the work?

swcurran (Sun, 23 Feb 2020 20:24:14 GMT):
1. It depends what the DID method you are talking about. Indy does not yet support that. With did:peer you could send to the other side anything, but the other agent is unlikely to know what to do. 2. Not really, but arguable. If the DID is on a public ledger (immutable), what does "revoke" mean? You could null out the public key(s)/endpoint in the DIDDoc as an indicator to say "abandoned!". The previous public keys are still there. As well, you can just stop responding to interactions with that DID. It's still there, but you unilaterally ignore it. You could even delete your private key for it. 3. If you are creating a public DID, you create it yourself, and then you send it to be published on a ledger. That would be the "third party" you are talking about. If it is a private, pairwise DID (did:peer), you send it to the other party that you want to use the DID.

DearKen (Mon, 24 Feb 2020 02:08:38 GMT):
Hi, Guys.

DearKen (Mon, 24 Feb 2020 02:09:23 GMT):
I want to find the further works of hyperledger indy or unsolved works. How can I find these documents?

DearKen (Mon, 24 Feb 2020 02:25:02 GMT):
@lovesh Thank you for reply, I checked the code. However, I cannot find the negative value in this code. I think that the cryptosystems works based on positivie integer. eg. if we are working in 256 bit finite field,

DearKen (Mon, 24 Feb 2020 02:31:57 GMT):
@lovesh Thank you for reply, I checked the code. However, I cannot find the negative value in this code. I think that the cryptosystems works based on positivie integer. eg. if we are working in 256 bit finite field, there would be 2 range proofs, 0 <= x < 2^{32} and 2^{224} < x < 2^{256} - Is ( 2^{224} < x < 2^{256} ) represented in range of negative values? Why this range cannot check equality statement?

DearKen (Mon, 24 Feb 2020 02:31:57 GMT):
@lovesh Thank you for reply, I checked the code. However, I cannot find the negative value in this code. I think that the cryptosystems works based on positivie integer. eg. if we are working in 256 bit finite field, there would be 2 range proofs, 0 <= x < 2^{32} and 2^{224} < x < 2^{256} - Is ( 2^{224} < x < 2^{256} ) represented in range of negative values? Why this range cannot check equals sign?

gpaliwal (Mon, 24 Feb 2020 07:02:24 GMT):
Has joined the channel.

thomas_kim (Mon, 24 Feb 2020 11:20:13 GMT):
Has joined the channel.

thomas_kim (Mon, 24 Feb 2020 11:20:13 GMT):
@sergey.minaev Hi, Sergey. Can you please tell me the purpose of the option vcx_init_minimal? Can I connect with the issuer and exchange credentials without dummy cloud agent?

thomas_kim (Mon, 24 Feb 2020 11:20:13 GMT):
@sergey.minaev Hi, Sergey. Can you please tell me the purpose of the option vcx_init_minimal for LibVCX? Can I connect with the issuer and exchange credentials without dummy cloud agent?

thomas_kim (Mon, 24 Feb 2020 11:20:13 GMT):
@sergey.minaev @esplinr Hi, Guys. Can you please tell me the purpose of the option vcx_init_minimal for LibVCX? Can I connect with the issuer and exchange credentials without dummy cloud agent?

thomas_kim (Mon, 24 Feb 2020 11:20:13 GMT):
@sergey.minaev @esplinr Hi, Guys. Can you please tell me the purpose of the option vcx_init_minimal for LibVCX? Can I connect with the issuer and exchange credentials without dummy cloud agent if I use this option?

iamShashvat (Mon, 24 Feb 2020 11:26:12 GMT):
Thanks @swcurran! Can you please give more insight for 3rd point? If I create public DID (in device) and send it to a service for doing the (some application which interact with ledger), service which is not on the device. For this reason

iamShashvat (Mon, 24 Feb 2020 11:29:59 GMT):
Thanks swcurran! Can you please give more insight for 3rd point? If I create public DID (in device) and send it to a service (some application which interact with ledger) which do the transaction on behalf of device, also this service is hosted elsewhere. For this reason do I have to use agents or it can be done without agents?

iamShashvat (Mon, 24 Feb 2020 11:29:59 GMT):
Thanks swcurran! Can you please give more insight for 3rd point? If I create public DID (in some service) and send it to an another service (some application which interact with ledger) which do the transaction on behalf of service, also this service is hosted elsewhere. For this reason do I have to use agents or it can be done without agents?

sairanjit (Mon, 24 Feb 2020 12:42:24 GMT):
I am getting the following error while building mobile app

sairanjit (Mon, 24 Feb 2020 12:43:14 GMT):
I am getting the following error while building mobile app com.sun.jna.Native: JNA: Callback org.hyperledger.indy.sdk.wallet.Wallet$1@f549fed threw the following exception com.sun.jna.Native: java.lang.UnsatisfiedLinkError: Error looking up function 'indy_get_current_error': undefined symbol: indy_get_current_error

omerfaruk 1 (Mon, 24 Feb 2020 12:49:46 GMT):
Has joined the channel.

omerfaruk 1 (Mon, 24 Feb 2020 12:49:48 GMT):
hey, i have 4 validator nodes(these are initial nodes) in docker, each has its own container. I go into Node1 container, i saw 'bls_sk' file in var/lib/indy/sandbox/keys/Node2/bls_keys folder. And the value in bls_sk is the same as in Node2 container var/lib/indy/sandbox/keys/Node2/bls_keys/bls_sk file. To sum up, nodes' bls secret keys known by each other. Why?

iamShashvat (Mon, 24 Feb 2020 13:10:11 GMT):
My concern is for production level, I have to pay for making DID but my device can't pay itself so I have to pay it using third party (meta-transaction) that is why I am asking you about this!

iamShashvat (Mon, 24 Feb 2020 13:10:11 GMT):
My concern is for production level, I have to pay for making DID but my service can't pay itself so I have to pay it using third party (meta-transaction) that is why I am asking you about this!

esplinr (Mon, 24 Feb 2020 15:56:00 GMT):
Starting the Indy Contributors call in 5 minutes: https://wiki.hyperledger.org/display/indy/2020-02-24+Indy+Contributors+Call https://zoom.us/j/615818107

iamShashvat (Mon, 24 Feb 2020 19:14:49 GMT):
@swcurran can you please help me with this? I'm stuck!

esplinr (Mon, 24 Feb 2020 21:24:38 GMT):
Reminder that we will be having an additional Indy Contributors call tomorrow at 7AM Pacific / 3PM UK / 4PM Europe to discuss how to support an Indy Network.

esplinr (Mon, 24 Feb 2020 21:25:39 GMT):
https://wiki.hyperledger.org/display/indy/2020-02-25+Special%3A+Troubleshooting+an+Indy+Network

esplinr (Mon, 24 Feb 2020 21:26:19 GMT):
We recommend that attendees review this document before joining the call: https://github.com/hyperledger/indy-node/blob/master/docs/source/troubleshooting.md

swcurran (Mon, 24 Feb 2020 22:27:44 GMT):
I don't know for sure, but I'm guessing because you are in a sandbox for dev, and there is no protectio from that visibility. It might be that they are using the same volume for the storage in docker and so have visibility to each others data., but I'm not sure.

swcurran (Mon, 24 Feb 2020 22:32:18 GMT):
You can have the DID created using a seed you save using an agent/process that can pay for it to be published on the ledger. Later when you use the device that *can't* pay for something, it will use the same *seed* to create the DID in its own wallet, and verify that it is on the ledger. That device doesn't have to pay for the write operation. Does that help?

RahulSonar87 (Tue, 25 Feb 2020 13:28:34 GMT):
Has joined the channel.

VithushanJegatheeswaran (Tue, 25 Feb 2020 18:04:49 GMT):
Hey guys, for my Bachelor thesis I'm building a voting system based on DIDs. I must write why I chose this technology. Could you guys help me with that, please? :grimacing: Why should someone choose DID over some other technology?

sairanjit (Tue, 25 Feb 2020 18:15:44 GMT):
1) Your identity is easily verified

sairanjit (Tue, 25 Feb 2020 18:22:49 GMT):
1)Your identity is easily verified 2)Lowerrisk of mishandled data 3) Prove your identity once to a trusted third party, then reuse that verified identity over and over again 4)Decentralizing data to store it on devices rather than on centralized databases, users get the final say over where, when, and how their information is shared

VithushanJegatheeswaran (Tue, 25 Feb 2020 18:29:29 GMT):
Thank you for your help

iamShashvat (Wed, 26 Feb 2020 08:28:16 GMT):

Screenshot (2).png

Yoroitchi (Wed, 26 Feb 2020 08:49:35 GMT):
have you tried to not entering the key ?

Yoroitchi (Wed, 26 Feb 2020 08:50:14 GMT):
`wallet create mywallet key`

iamShashvat (Wed, 26 Feb 2020 09:33:22 GMT):
Yes.

iamShashvat (Wed, 26 Feb 2020 09:33:42 GMT):
Same error

sanjaysb (Wed, 26 Feb 2020 10:11:51 GMT):
Thi

sanjaysb (Wed, 26 Feb 2020 10:14:45 GMT):
When two applications using idy sdk share the same wallet home (usually ```~/.indy_client``` if you follow the samples) I have faced similar issue. I have not used Indy CLI. Try using a different home.

sanjaysb (Wed, 26 Feb 2020 10:14:45 GMT):
When two applications using indy sdk share the same wallet home (usually ```~/.indy_client``` if you follow the samples) I have faced similar issue. I have not used Indy CLI. Try using a different home.

hyper-sunder (Wed, 26 Feb 2020 10:41:38 GMT):
@iamShashvat you should attach the wallet before you open it. By default, wallets that are created via 'cli` will be listed in command `wallet list`. Wallets created by any other program, should be attached and open in `cli`

hyper-sunder (Wed, 26 Feb 2020 10:41:38 GMT):
@iamShashvat you should attach the wallet before you open it. By default, wallets that are created via `cli` will be listed in command `wallet list`. Wallets created by any other program, should be attached and open in `cli`

doomguy (Wed, 26 Feb 2020 11:56:33 GMT):
Has joined the channel.

doomguy (Wed, 26 Feb 2020 11:56:34 GMT):
Hello all ! Sorry if I write to wrong channel, but I've found Cordenity project on github [https://github.com/hyperledger-labs/cordentity] to integrate Indy capabilities into the Corda platform. Is there something similar, but for HL Fabric ? I mean a seamless way to integrate Indy features into Fabric blockchain.

hyper-sunder (Wed, 26 Feb 2020 13:33:31 GMT):
Hello Guys, I would like to know how to view actual credentials information shared by the `alice` when `faber` requested for proof? Could you please any one guide me on this?

VithushanJegatheeswaran (Wed, 26 Feb 2020 14:15:38 GMT):
voting

davidd (Wed, 26 Feb 2020 14:30:02 GMT):
Has joined the channel.

jljordan_bcgov (Wed, 26 Feb 2020 14:31:22 GMT):
If you run this OpenAPI demo you will see all the messages involved in inter agent communication including credentials. So if you are looking to see the low level data this should help ... https://github.com/hyperledger/aries-cloudagent-python/tree/master/demo

msaidm (Wed, 26 Feb 2020 20:23:43 GMT):
Has joined the channel.

msaidm (Wed, 26 Feb 2020 20:23:44 GMT):
Hi everyone I am a Computer engineer student and I working on a graduation project that will verify College certificates using Indy, but to be honest I am lost and I don't know where to begin. What I know that I need a mobile app that will the wallet that stores the certificate and I will need to make a website for my college that will act as Faber in the example in the documentation. I need to know how to begin the app and the website.

Koushik (Wed, 26 Feb 2020 22:02:10 GMT):
Has joined the channel.

msaidm (Wed, 26 Feb 2020 22:09:39 GMT):
@esplinr how can I create the network sir ?

iamShashvat (Thu, 27 Feb 2020 04:42:30 GMT):
This works! Thank you @hyper-sunder!

notsteward (Thu, 27 Feb 2020 08:37:03 GMT):
Has joined the channel.

notsteward (Thu, 27 Feb 2020 08:37:04 GMT):
Hi! I'm new with indy, I wants to know can I use Hyperledger Indy as a private network? Private network is only for test or it can be used for a real project like in ethereum?

notsteward (Thu, 27 Feb 2020 10:50:11 GMT):
Is it required to start network with at least 4 Validator nodes? or it can be combination of 4 (validator and observer) nodes?

sergey.khoroshavin (Thu, 27 Feb 2020 12:15:55 GMT):
Hi, nothing can stop you from creating and using private Indy network, if you wish to do so. Of course, if all nodes will belong just to one organization it loses decentralization advantage, because entity controlling more than 2/3 of all nodes can rewrite ledger at its whim, but for internal corporate use cases it might be okay. Btw, is there any particular reason why you don't want to use Sovrin public network?

notsteward (Thu, 27 Feb 2020 12:22:58 GMT):
Thank you for reply @sergey.khoroshavin No No... I am just comparing it with the functionalities of ethereum, no particular reason!

notsteward (Thu, 27 Feb 2020 12:24:14 GMT):
@sergey.khoroshavin Can you help me with this?

notsteward (Thu, 27 Feb 2020 12:44:09 GMT):
and also to use public network we have to pay for transactions/ or for becoming to steward(which is understood!)

sergey.khoroshavin (Thu, 27 Feb 2020 13:04:11 GMT):
Yes, in order to get some fault tolerance you need at least 4 validator nodes (in this case pool can survive 1 offline or otherwise misbehaving node), and this is a recommended minimal configuration. As far as I remember it is possible to run smaller pools, but in this case any node going down will render pool nonfunctional, and such configuration were not tested, so there might be other problems if you try to run less than 4 nodes.

sergey.khoroshavin (Thu, 27 Feb 2020 13:05:07 GMT):
As for observer nodes - as of now this feature is only partly implemented, so in effect all nodes are either validators or not participating at all

sergey.khoroshavin (Thu, 27 Feb 2020 13:05:55 GMT):
So, quick recap - minimal safe configuration is pool of 4 validator nodes

sergey.khoroshavin (Thu, 27 Feb 2020 13:06:58 GMT):
I see your point, thanks!

notsteward (Thu, 27 Feb 2020 13:09:13 GMT):
:thumbsup: Thanks for the knowledge @sergey.khoroshavin I didn't knew that observer nodes are partly implemented!

swcurran (Thu, 27 Feb 2020 16:48:10 GMT):
You can look at von-network for a quick way to spin up a network and experiment. https://github.com/bcgov/von-network

MarinaELias (Thu, 27 Feb 2020 19:14:54 GMT):
Has joined the channel.

MarinaELias (Thu, 27 Feb 2020 19:14:54 GMT):
HI there, I am new to Hyperledger Indy and I am trying to develop the demo provided by hyperledger indy about college certificates all from scratch by myself but I don't know where to start or what I will need so I need some help. I already read the Indy documentation and understood the details. Would I be working with hyperldger Aries frameworks or would I need to use the indy-sdk and von network? ANy help would be much appreciated.

MarinaELias (Thu, 27 Feb 2020 19:14:54 GMT):
HI there, I am new to Hyperledger Indy and I am trying to develop the demo provided by hyperledger indy about college certificates all from scratch by myself but I don't know where to start or what I will need so I need some help. I already read the Indy documentation and understood the details. Would I be working with hyperldger Aries frameworks or would I need to use the indy-sdk and von network? Any help would be much appreciated.

echo.harker (Thu, 27 Feb 2020 19:52:17 GMT):
Has joined the channel.

swcurran (Thu, 27 Feb 2020 20:44:44 GMT):
We've created a course on edX aimed at developers asking that question. It launches next week. In the meantime, a quick and dirty mini version of the course is here that you could start with. https://github.com/hyperledger/aries-cloudagent-python/blob/master/docs/GettingStartedAriesDev/README.md We definitely recommend starting and building on Aries, which embeds Indy implementations vs. trying to start at the pure Indy level.

swcurran (Thu, 27 Feb 2020 21:36:46 GMT):
Question from a Slack user for @andrew.whitehead or @ianco I'd like to establish a DIDComm connection using aries-cloudagent-python. I got it working with creating/receiving invitations using the REST API. Now I'd like to establish a connection using DIDComm only with public DIDs (implicit invitation). Receiving such a connection request for an implicit invitation seems to be supported by aries-cloudagent-python (https://github.com/hyperledger/aries-cloudagent-python/issues/358#issuecomment-581503697 and https://github.com/hyperledger/aries-cloudagent-python/blob/master/aries_cloudagent/protocols/connections/manager.py#L380). But I couldn't find how to send this connection request using aries-cloudagent-python - is this already possible, are there any examples available? Comment on #358 Use of public DID in connection hyperledger/aries-cloudagent-python | Feb 3rd | Added by GitHub aries_cloudagent/protocols/connections/manager.py:380 hyperledger/aries-cloudagent-python | Added by GitHub Stephen Curran 1:35 PM Hi Michael - I'm not sure. I'll check with the devs on this. I imagine what you are seeing is that a connection ID is required on the API call, but since it is supposed to be implicit, it shouldn't be required. I can confirm, or I can just raise an issue.

swcurran (Thu, 27 Feb 2020 21:36:46 GMT):
Question from a Slack user for @andrew.whitehead or @ianco I'd like to establish a DIDComm connection using aries-cloudagent-python. I got it working with creating/receiving invitations using the REST API. Now I'd like to establish a connection using DIDComm only with public DIDs (implicit invitation). Receiving such a connection request for an implicit invitation seems to be supported by aries-cloudagent-python (https://github.com/hyperledger/aries-cloudagent-python/issues/358#issuecomment-581503697 and https://github.com/hyperledger/aries-cloudagent-python/blob/master/aries_cloudagent/protocols/connections/manager.py#L380). But I couldn't find how to send this connection request using aries-cloudagent-python - is this already possible, are there any examples available? Comment on #358 Use of public DID in connection hyperledger/aries-cloudagent-python | Feb 3rd | Added by GitHub aries_cloudagent/protocols/connections/manager.py:380 hyperledger/aries-cloudagent-python | Added by GitHub Stephen Curran 1:35 PM Hi Michael - I'm not sure. I'll check with the devs on this. I imagine what you are seeing is that a connection ID is required on the API call, but since it is supposed to be implicit, it shouldn't be required.

sergiomello (Fri, 28 Feb 2020 00:30:29 GMT):
Has joined the channel.

sergiomello (Fri, 28 Feb 2020 00:30:36 GMT):
Hi

notsteward (Fri, 28 Feb 2020 05:25:48 GMT):
I have read this in docs,"A Steward can add one and only one Validator Node". Is this rule implemented?

sergey.khoroshavin (Fri, 28 Feb 2020 10:07:28 GMT):
Yes, for one DID with Steward role there can be only one Validator node. However if you're in full control of network (for example - creating test pool inside one organization) you can create and control as many Steward DIDs as you need for running as many nodes as you need. For an example you can look through scripts for creating test pool in docker, they are located in https://github.com/hyperledger/indy-node/tree/master/environment/docker/pool

notsteward (Fri, 28 Feb 2020 10:08:52 GMT):
Thank you @sergey.khoroshavin You have cleared my doubt.

notsteward (Fri, 28 Feb 2020 10:26:12 GMT):
I saw this on stack overflow today! and I also try to do the same as follows: Is this a bug or I have done something wrong? { reqId: 1582624387016905500, identifier: 'Th7MpTaRZVRYnPiabds81Y', //STEWARD operation: { type: '1', dest: 'KSCHYzEVGcvAoccpMFcBQR', verkey: 'B3kMCkxuiwz4hEHLpkdvqZScVCdudEYpPpHZChBY2UBn', role: null }, //////// ROLE IS NULL protocolVersion: 2 } Sending NYM request to the ledger Generating and storing test DID and verkey Test DID: DQp4x5rpjY4MUf4i8TgWoC Test Verkey: 7mKfM5T5PhLjcecnnwmht91nACYyHaY3efSiqYZshQrC Building NYM request to add Test to the ledger { reqId: 1582624387185413000, identifier: 'KSCHYzEVGcvAoccpMFcBQR', //////// ROLE NULL DID is doing transaction to add "test Role:101 Trust_Anchor" operation: { type: '1', dest: 'DQp4x5rpjY4MUf4i8TgWoC', verkey: '7mKfM5T5PhLjcecnnwmht91nACYyHaY3efSiqYZshQrC', role: '101' }, protocolVersion: 2 } This is the successful transaction! How is this possible?

MarinaELias (Fri, 28 Feb 2020 12:27:55 GMT):
Thank you so much for the reply. I will be waiting for the course.

lynn.bendixsen (Fri, 28 Feb 2020 16:39:29 GMT):
This is possible if your auth-rules are setup to allow it. On the Sovrin MainNet not even the first transaction would be allowed since only a Transaction Endorser role (name was changed from Trust_anchor) can add a Transaction Author (NULL Role) to that ledger. To check your auth-rules from the indy-cli run `ledger get-auth-rule`

CavitOzbay (Sat, 29 Feb 2020 19:39:24 GMT):
Has joined the channel.

CavitOzbay (Sat, 29 Feb 2020 19:39:26 GMT):
Hi, I am working on implementing an additional functionality to Indy in the scope of my graduation project. I have theoretic design to implement and I need to add a new transaction type to the domain ledger in order to simulate it. Is there anybody has tried a similar thing or anybody who can suggest a general guidance?

hyper-sunder (Tue, 03 Mar 2020 07:27:00 GMT):
Hi Guys, Can anyone guide me how to revoke the issued credentials in indy? Have been searching for the same for long time but found no clue.

rtorres (Tue, 03 Mar 2020 09:52:49 GMT):
Has joined the channel.

swcurran (Tue, 03 Mar 2020 14:02:39 GMT):
Best place to start is with this document - https://github.com/hyperledger/indy-hipe/tree/master/text/0011-cred-revocation

swcurran (Tue, 03 Mar 2020 14:03:13 GMT):
There are some examples in the community. There is a PR in aries-cloudagent-python that is nearing completion.

swcurran (Tue, 03 Mar 2020 14:03:51 GMT):
In Ursa, there is a new version that instead of a tails file there is a merkle tree. However, that has not been added to Indy as yet.

hyper-sunder (Tue, 03 Mar 2020 16:24:52 GMT):
thanks @swcurran Let me try

hyper-sunder (Tue, 03 Mar 2020 16:30:01 GMT):
@swcurran the given link tells/explains about the theoretical idea of revocation. I am looking for kind sdk reference so that I can make use of it in my POC

hyper-sunder (Tue, 03 Mar 2020 16:30:19 GMT):
Could you please share link? Thanks in advance

swcurran (Tue, 03 Mar 2020 18:19:11 GMT):
If you go to ACA-Py - look for the revocation PR - https://github.com/hyperledger/aries-cloudagent-python

swcurran (Tue, 03 Mar 2020 18:20:23 GMT):
There is another example that is older code (don't use) but has good test cases and code. https://von-anchor.readthedocs.io/en/latest/

DearKen (Wed, 04 Mar 2020 02:20:30 GMT):
Hi, Indy community! I have a one question. In hyperledger Indy, how can resolve the credential ID correlation? (https://www.w3.org/TR/vc-data-model/#dfn-verifiable-credentials, Section 7.3)

vladyslavmunin (Wed, 04 Mar 2020 15:42:13 GMT):
Has joined the channel.

vladyslavmunin (Wed, 04 Mar 2020 15:42:14 GMT):
Hi guys, what is best approach to resolve / register DID in existing/ running application with golang backend. Does Indy SDK have some golang wrappers ( at least that will cover this purpose ) or should we use some separate microservise written in other language for this?

vladyslavmunin (Wed, 04 Mar 2020 15:42:14 GMT):
Hi guys, what is best approach to resolve / register DID from von-network / Indy in existing application with golang backend. Does Indy SDK have some golang wrappers ( at least that will cover this purpose ) or should we use some separate microservise written in other language for this?

swcurran (Wed, 04 Mar 2020 21:48:16 GMT):
The new indy-vdr component can be configured to run like that. It has a configuration that can be used as a web service that you can call to interact with the ledger - read or write. It is not the full indy-sdk and includes just what is necessary to interact with the indy ledger. https://github.com/hyperledger/indy-vdr

syngin (Thu, 05 Mar 2020 03:57:49 GMT):
Has joined the channel.

syngin (Thu, 05 Mar 2020 03:57:50 GMT):
Hi. Would it be possible to attach a binary such as an image to a credential? Is there a size limit to an attribute? could it be base64 encoded?

Vritra (Thu, 05 Mar 2020 04:46:10 GMT):
Has joined the channel.

ravitejasriram (Thu, 05 Mar 2020 09:36:10 GMT):
Has joined the channel.

sanjaysb (Thu, 05 Mar 2020 10:09:45 GMT):
down

sanjaysb (Thu, 05 Mar 2020 10:11:14 GMT):
Is the sovrin maven repo down?

lynn.bendixsen (Thu, 05 Mar 2020 18:38:39 GMT):
@MALodder ^^

MALodder (Thu, 05 Mar 2020 18:49:45 GMT):
Nexus is up and running as far as I know

pknowles (Fri, 06 Mar 2020 04:02:36 GMT):
From one of the developers in our project team ... "I did an investigation of Indy SDK issues that we have been facing over the past two days. INDY-SDK function `verifierVerifyProof` throws an error - _AnoncredsProofRejected_. We have seen this failure in the past when trying to use dashes (-) inside credential attribute values. Unfortunately, the INDY SDK doesn't want to handle it despite the fact that they claim support of integers and strings as valid Credential Attributes (see _A Note About Encoding Attributes_ excerpt at https://hyperledger-indy.readthedocs.io/projects/hipe/en/latest/text/0109-anoncreds-protocol/README.html )." Has anyone come across this issue and, if so, is there a workaround?

pknowles (Fri, 06 Mar 2020 04:02:36 GMT):
From one of the developers in our project team ... "I did an investigation of Indy SDK issues that we have been facing over the past two days. INDY-SDK function `verifierVerifyProof` throws an error *AnoncredsProofRejected*. We have seen this failure in the past when trying to use dashes (-) inside credential attribute values. Unfortunately, the INDY SDK doesn't want to handle it despite the fact that they claim support of integers and strings as valid Credential Attributes (see _A Note About Encoding Attributes_ excerpt at https://hyperledger-indy.readthedocs.io/projects/hipe/en/latest/text/0109-anoncreds-protocol/README.html )." Has anyone come across this issue and, if so, is there a workaround?

pknowles (Fri, 06 Mar 2020 04:02:36 GMT):
From one of the developers in our project team ... "I did an investigation of Indy SDK issues that we have been facing over the past two days. INDY-SDK function `verifierVerifyProof` throws an error "AnoncredsProofRejected". We have seen this failure in the past when trying to use dashes (-) inside credential attribute values. Unfortunately, the INDY SDK doesn't want to handle it despite the fact that they claim support of integers and strings as valid Credential Attributes (see _A Note About Encoding Attributes_ excerpt at https://hyperledger-indy.readthedocs.io/projects/hipe/en/latest/text/0109-anoncreds-protocol/README.html )." Has anyone come across this issue and, if so, is there a workaround?

shailendrShrivastav (Fri, 06 Mar 2020 07:02:11 GMT):
Has joined the channel.

syngin (Fri, 06 Mar 2020 09:54:10 GMT):
I managed to include a base64 encoded image into a credential offer.

JasSingh (Fri, 06 Mar 2020 15:57:13 GMT):
Has joined the channel.

swcurran (Fri, 06 Mar 2020 16:55:57 GMT):
There are practical limits - who is getting the credential? If going to a mobile phone, that has to be considered.

jameshiester (Sat, 07 Mar 2020 02:26:12 GMT):
Has joined the channel.

jameshiester (Sat, 07 Mar 2020 02:26:13 GMT):
Is there an example csv file for using the genesis tx creator tool?

SantosCastro (Sun, 08 Mar 2020 11:08:00 GMT):
Has joined the channel.

SantosCastro (Sun, 08 Mar 2020 11:08:25 GMT):
Hi, I'm new to indy.

SantosCastro (Sun, 08 Mar 2020 11:09:22 GMT):
I created a basic workflow (issuer, prover, verifier) and found something weird so I'm probably doing something wrong. :)

SantosCastro (Sun, 08 Mar 2020 11:10:56 GMT):
If the prover modify the requested proof section of the proof before sending it to the verifier the proof pasees the verification.

SantosCastro (Sun, 08 Mar 2020 11:11:22 GMT):
So the prover can send fake data. Any idea about what I'm doing wrong?

SantosCastro (Sun, 08 Mar 2020 11:11:22 GMT):
So the prover can send fake data. Any idea about what am I doing wrong?

SantosCastro (Sun, 08 Mar 2020 11:11:22 GMT):
So the prover can send fake data. Any idea about what I'm doing wrong?

iamShashvat (Sun, 08 Mar 2020 19:18:56 GMT):
setEndpointForDid ( wh, did, address, transportKey ) in this API what does transportKey means? I have search in all the Docs I can't find use of it!

sethjback (Sun, 08 Mar 2020 19:47:28 GMT):
Has joined the channel.

sethjback (Sun, 08 Mar 2020 19:47:29 GMT):
I've put together a Postman collection/environment for interacting with `indy-vdr` if anyone is interested

sethjback (Sun, 08 Mar 2020 19:50:23 GMT):
major kudos for putting `indy-vdr` together - we are primarily using golang and that really accelerated our working with the ledger

andrew.whitehead (Sun, 08 Mar 2020 19:57:28 GMT):
Awesome, I’m glad it’s working for you. You’re using the proxy server app sounds like?

sethjback (Sun, 08 Mar 2020 20:17:53 GMT):
Yep, we are running the proxy for interacting with the ledger

sethjback (Sun, 08 Mar 2020 20:20:04 GMT):
one request, and hopefully we'll get a chance to PR it in the future, but having a consistent well formatted JSON reply would be helpful. The replies have two JSON objects, One with the data payload and one with the node information. If we could combine that into a single JSON body and return an "application/json" content type header that would make parsing the response easier

andrew.whitehead (Sun, 08 Mar 2020 20:22:06 GMT):
Yep, that's definitely in the queue. The node information will become a header response

sethjback (Sun, 08 Mar 2020 20:22:33 GMT):
perfect!

sethjback (Sun, 08 Mar 2020 20:22:56 GMT):
that would be great

alecfokapu (Sun, 08 Mar 2020 23:25:47 GMT):
Has joined the channel.

pknowles (Mon, 09 Mar 2020 05:20:16 GMT):
https://chat.hyperledger.org/channel/aries?msg=nCASyrCPXaZys7isS

pknowles (Mon, 09 Mar 2020 05:40:59 GMT):
https://chat.hyperledger.org/channel/aries?msg=A7fjfXEGbhi5LyQDt

daveek (Mon, 09 Mar 2020 09:24:36 GMT):
Hi.. What Cryptographic Primitives are used in Indy & Aries?

MarinaELias (Mon, 09 Mar 2020 13:26:47 GMT):
Hello, I tried starting with ACA-py but I found it too complicated to begin with, so I tried using Streetcred but even the tutorial seems to be incomplete. Whenever I try to begin with something I find myself lost inside all the repos out there. Is there a definite way or link to begin from? We want to make a mobile application to act as a wallet for credentials, but we're overwhelmed and can't find helpful steps. So can you suggest how to begin with Streetcred if the Javascript/C# code is unavailable? Any help would be appreciated. Thanks in advance.

esplinr (Mon, 09 Mar 2020 15:00:18 GMT):
Starting the Indy Contributors call: https://zoom.us/j/615818107

sethjback (Mon, 09 Mar 2020 15:58:01 GMT):
Hi Marina - I just went through this exact phase and am somewhat out the other side. If you want to chat briefly sometime let me know and I can give you a 10000 ft. view as much as I am able

swcurran (Mon, 09 Mar 2020 17:12:16 GMT):
We're always trying to make ACA-Py easier to use, so would appreciate what you think would have helped you. There is now a course on edX that might help. I'm a co-author of the course, so glad to take questions: https://www.edx.org/course/becoming-a-hyperledger-aries-developer

hamidm (Mon, 09 Mar 2020 21:41:28 GMT):
Hello, as I'm back into testing against testnet, I'm refused the creation of schema, because not TAA info in the transaction. Any documentation on how to set it up ? what to do with AML and TAA ? thanks

sanjaysb (Wed, 11 Mar 2020 04:54:40 GMT):
Hi, I am using react-native to build a mobile indy wallet application. I am facing performance issues in android. I am using indy java sdk to write java modules that do indy specific functions which I can call using native modules. Even the calls that do not use pool connection like storing credentials or reading them, creating a pseudonymous did etc. seem to take considerable amount of time. Has anyone built a similar wallet. Is there any documentation I can refer to.

iamShashvat (Wed, 11 Mar 2020 04:56:19 GMT):
@esplinr @swcurran Please can you help?

lauravuo-techlab (Wed, 11 Mar 2020 07:22:37 GMT):
@sethjback sounds interesting approach! I'm curious, did you just want to avoid indy-sdk dependency or what was you reason for using the proxy? my team has developed "an agency" that interacts with the ledger using golang. we ended up writing a wrapper for indy-sdk.

VithushanJegatheeswaran (Wed, 11 Mar 2020 10:44:38 GMT):
Hey guys, Do I need to create a new credential definition for every connection ?

VithushanJegatheeswaran (Wed, 11 Mar 2020 10:45:27 GMT):
I think one is enough, am I right ?

swcurran (Wed, 11 Mar 2020 13:31:23 GMT):
Yes. One Cred Def per credential type, generating as many credential as you want.

Abhishekkishor (Wed, 11 Mar 2020 15:41:49 GMT):
Hello Guys, I have a question regarding Agent to agent communication in hyperledger Q. Can Agents of one organization can communicate with other agents of other organization? If yes then how? Also which protocol they use to communicate? @swcuswcurran sir can you please help me with this? Thanks.

Abhishekkishor (Wed, 11 Mar 2020 15:57:05 GMT):
* @swcurran

swcurran (Wed, 11 Mar 2020 16:00:34 GMT):
Yes, they can - using Aries agents and DIDComm.

swcurran (Wed, 11 Mar 2020 16:03:04 GMT):
To be more specific than that, I suspect that you need to do some learning about Aries, and defining your use case. There is an edX course for learning about the general topic and about agents - https://www.edx.org/course/introduction-to-hyperledger-sovereign-identity-blockchain-solutions-indy-aries-and-ursa If that is helpful, there is also an edX course about developing apps on top of Aries.

Abhishekkishor (Wed, 11 Mar 2020 16:04:07 GMT):
@swcurran thanks a lot sir.

esplinr (Wed, 11 Mar 2020 20:14:20 GMT):
I'm a bit late posting the Evernym team's most recent discussion of the Indy and Aries projects. The key topic is a discussion of options for CI / CD. https://www.youtube.com/watch?v=EEHpINC7nqQ&list=PLRp0viTDxBWGLdZk0aamtahB9cpJGV7ZF&index=1

iamShashvat (Thu, 12 Mar 2020 02:49:50 GMT):
Can we add multiple endpoints in anyway?

iamShashvat (Thu, 12 Mar 2020 05:47:41 GMT):
Hi, I have set endpoint/metadata for a did, now how can I make transaction for that particular change?

sanjaysb (Thu, 12 Mar 2020 07:22:42 GMT):
I seem to have solved the issue. I am using key_derivation_method specified here : https://github.com/hyperledger/indy-sdk/blob/master/wrappers/java/src/main/java/org/hyperledger/indy/sdk/wallet/Wallet.java#L124

sanjaysb (Thu, 12 Mar 2020 07:22:42 GMT):
This seems to have solved the issue. I am using key_derivation_method specified here : https://github.com/hyperledger/indy-sdk/blob/master/wrappers/java/src/main/java/org/hyperledger/indy/sdk/wallet/Wallet.java#L124

iamShashvat (Thu, 12 Mar 2020 09:21:09 GMT):
I want to add service endpoints and some metadata, I did that, but its not on a blockchain. How can I record it on Indy?

iamShashvat (Thu, 12 Mar 2020 09:21:09 GMT):
Is there any way by which we can store service endpoints or metadata on blockchain? Can anyone please help me with this?

sanjaysb (Thu, 12 Mar 2020 09:37:59 GMT):
As per this example, https://github.com/hyperledger/indy-sdk/blob/master/README.md#2-starting-the-test-pool-on-a-specific-ip-address we can provide an ip address that the mobile app can resolve. How can I run the containers if I know just the fully qualified domain name of the vm I wish to run the containers on

ShamGir (Thu, 12 Mar 2020 12:53:18 GMT):
I am also using indy java sdk, in Anroid app, but there is no any performance issue. But some it takes few seconds even minutes in creating credentialsDefinations using sdk. If you are facing issues then try to use indy-functionalities in background thread instead of UI thread.

JeromeK (Thu, 12 Mar 2020 13:21:59 GMT):
You can put it into a nym-transaction

iamShashvat (Thu, 12 Mar 2020 13:31:11 GMT):
So is there any API by which I can create the request?

JeromeK (Thu, 12 Mar 2020 14:18:47 GMT):
Those functions should be in every Wrapper, yes

iamShashvat (Thu, 12 Mar 2020 14:34:48 GMT):
Ok, I got it.. my mistake! Thank you

sanjaysb (Fri, 13 Mar 2020 04:52:27 GMT):
I am not using the app for issuer and verifier. I am just using it for holder. For me creating did, creating master secret, storing a credential received etc. was taking around 8s each. generating proof was taking more that 30s. This includes fetching the definition from the pool. However once I changed the key_derivation_method it all came to 3-4s. except for proof generation which took around 10s.

ShamGir (Fri, 13 Mar 2020 09:14:28 GMT):
It seems interesting, I never faced this issue, in my Android app, I have all type of users (issuer, verifier and prover) in my sample app. But never face performance issue.

vladyslavmunin (Fri, 13 Mar 2020 13:11:20 GMT):
Hi,how to get handle of already opened wallet? Use case, two apps that use should use the same wallet keys. After first app opened a wallet and got wallet handle, it's not possible to do it from second app. And second app doesn't have an access to wallet handle , but has access to wallet creds.

vladyslavmunin (Fri, 13 Mar 2020 13:11:20 GMT):
Hi,how to get handle of already opened wallet? Use case, two apps that use the same wallet keys. After first app opened a wallet and got wallet handle, it's not possible to do it from second app. And second app doesn't have an access to wallet handle , but has access to wallet creds.

nage (Fri, 13 Mar 2020 20:27:12 GMT):
Though it isn't strictly on-topic here, many of you may have seen my email post to the mailing lists, but if you are looking for more information about Sovrin Foundation Staff status, these links might be of interest, or feel free to msg me here https://chat.sovrin.org/channel/jobs?msg=EHTdQJCLpLNDRKBGP https://sovrin.org/the-status-of-the-sovrin-foundation-2/

cmonkeydo (Sat, 14 Mar 2020 18:01:07 GMT):
Has joined the channel.

ssinew (Mon, 16 Mar 2020 01:49:21 GMT):
Has joined the channel.

ssinew (Mon, 16 Mar 2020 01:49:22 GMT):
Hi, im new in SSI but i need to do a PoC and I cant do the example about Alice in my virtual machine. I saw the tutorial with indy but I couldn't do it either. the links is it: https://github.com/hyperledger/indy-sdk/blob/master/docs/getting-started/indy-walkthrough.md I have to do a job for the university and I would like to show that this works and if possible explain the content of did documnentos did or SSI issues, as much as possible through a local network through a virtual machine. What do you think? Thank you!!

sanjaysb (Mon, 16 Mar 2020 12:20:08 GMT):
Hi, I want to deploy the indy pool on a vm which does not have a static IP. I tried running the container as per https://github.com/hyperledger/indy-sdk/blob/master/README.md#2-starting-the-test-pool-on-a-specific-ip-address using the fqdn instead of ip. it says Invalid ip address. how can I deploy it correctly?

lynn.bendixsen (Mon, 16 Mar 2020 14:42:11 GMT):
I think it only works with static IP addrs.

japidei (Mon, 16 Mar 2020 17:07:18 GMT):
Has joined the channel.

esplinr (Mon, 16 Mar 2020 21:57:40 GMT):
Instead of a sprint demo video, our team decided to share a brief project update: https://www.youtube.com/watch?v=efOwa6KOXEc&list=PLRp0viTDxBWGLdZk0aamtahB9cpJGV7ZF Includes a summary of our work on W3C Verifiable Credentials (a.k.a. Rich Schemas), key sessions from the Hyperledger Global Forum, and plans for upcoming releases.

AnnaJ (Tue, 17 Mar 2020 01:41:28 GMT):
From the Sovrin Foundation Board of Trustees: Next Steps for the Sovrin Foundation https://sovrin.org/next-steps-for-the-sovrin-foundation/

shiseki (Tue, 17 Mar 2020 05:24:09 GMT):
Has joined the channel.

HLFPOC (Tue, 17 Mar 2020 07:10:17 GMT):
Hi Team, not sure if this is a valid question, but wanted to check if it is possible to record a custom transaction/record in Indy ledger apart from schema or cred-defs? I am running a local indy pool having 4 nodes using von-network, and want to record some more transactions as part of use-case. Is is possible to do so ? If yes, can anyone guide what should be the approach?

mirgee (Tue, 17 Mar 2020 14:25:20 GMT):
Hi, what is the signature algorithm used in Indy for cryptographic signatures?

mirgee (Tue, 17 Mar 2020 14:25:20 GMT):
Hi, what is the signature algorithm used in Indy for cryptographic signatures? I am guessing it would be ECDSA, but I couldn't find any reference or documentation of it.

MALodder (Tue, 17 Mar 2020 14:40:59 GMT):
@mirgee for ledger txns, its Ed25519

mirgee (Tue, 17 Mar 2020 14:57:46 GMT):
Thank you

swcurran (Tue, 17 Mar 2020 15:25:28 GMT):
Not 100% certain on this, but pretty sure. Transactions are checked at the client and ledger and only the specific set that are explicitly defined are allowed. The closest you can get to being "free format" -- and maybe this would work -- is the ATTRIB transaction that you can associated with a DID.

ashcherbakov (Tue, 17 Mar 2020 15:56:59 GMT):
yes, you can store arbitrary data (either raw, encoded or hashed) in ATTRIB txn.

ashcherbakov (Tue, 17 Mar 2020 15:58:05 GMT):
Actually supporting a custom transaction family on a custom Indy Pool instance is a relatively easy task due to Indy Ledger's Pluggable Request Handlers

swcurran (Tue, 17 Mar 2020 16:00:16 GMT):
Agreed -- that's been demonstrated a lot :-). But not on a standard deployment. E.g. the out-of-the-box Indy, or on Sovrin Mainnet. I was steering @HLFPOC away from a custom ledger.

andrew.whitehead (Tue, 17 Mar 2020 16:00:50 GMT):
I think the ledger rejects any attrib that isn’t named ‘endpoint’

swcurran (Tue, 17 Mar 2020 16:01:39 GMT):
:-(

HLFPOC (Tue, 17 Mar 2020 16:03:46 GMT):
@swcurran @andrew.whitehead Does ACA-Py supports ATTRIB txns ?

ashcherbakov (Tue, 17 Mar 2020 16:05:55 GMT):
@andrew.whitehead > I think the ledger rejects any attrib that isn’t named ‘endpoint’ That should not be the case, and I'm pretty sure we have tests for non-endpoint attributes. Have you seen such behaviour? If so, this is a bug, and it's worth a ticket.

andrew.whitehead (Tue, 17 Mar 2020 16:06:01 GMT):
It can automatically publish the endpoint when it detects a change, but no custom attribs

ashcherbakov (Tue, 17 Mar 2020 16:07:25 GMT):
FYI: https://github.com/hyperledger/indy-node/blob/master/indy_node/test/api/test_attrib_reply.py#L25

swcurran (Tue, 17 Mar 2020 16:10:29 GMT):
So @andrew.whitehead - we need to add an issue for ACA-Py to enable that?

andrew.whitehead (Tue, 17 Mar 2020 16:15:11 GMT):
Hm, looks like it might only have additional restrictions when the attrib is named endpoint. The value does have to be a dict for raw attribs though

andrew.whitehead (Tue, 17 Mar 2020 16:15:11 GMT):
Hm, looks like it might only have additional restrictions when the attrib is named endpoint. The value does have to be a dict with one entry for raw attribs though

swcurran (Tue, 17 Mar 2020 16:19:19 GMT):
"it" in that case is the ledger, right? ACA-Py has the "endpoint only" restrictions? So we need to loosen those?

andrew.whitehead (Tue, 17 Mar 2020 16:21:04 GMT):
We could try adding other methods to fetch or publish an attrib

swcurran (Tue, 17 Mar 2020 16:23:23 GMT):
OK - I'll add the basic issue to the repo and we'll see where that fits. I'll tag it with help wanted for now. Does indy-vdr need the issue as well, or does it just take what it is given?

davidd (Tue, 17 Mar 2020 17:02:12 GMT):
Hi, Is it Possible to Connect OSMA Agent with Cloud Agent (ACA-PY), any other web agent except Aries .Net?

domwoe (Wed, 18 Mar 2020 06:39:21 GMT):
@danielhardman I'm reading your latest article on the evernym blog. There you write "ZKP-oriented credentials are fully capable of supporting public non-repudiation, like their simpler counterpart. If the verifier and prover agree that a proof will be non-repudiable (e.g., in the case of the mortgage), then the proof makes non-repudiation enforceable through the cryptography. This is, and always has been, the default behavior of the ZKP-oriented credentials in Hyperledger Indy." The latest sentence was surprising for me. Can you explain how this works or point me to a resource?

ajayjadhav (Wed, 18 Mar 2020 13:08:43 GMT):
Just curious, can a verifiable credential be tranferable like value tokens - e.g. transferable movie tickets Is there any RFC which relates with this idea ?

danielhardman (Wed, 18 Mar 2020 13:49:20 GMT):
@ajayjadhav : yes, if you don't bind the credential to a specific holder.

danielhardman (Wed, 18 Mar 2020 13:49:47 GMT):
@domwoe: excellent question. I'll respond, but it will take me a few hours; I need some time after my morning meeting.

domwoe (Wed, 18 Mar 2020 13:51:59 GMT):
thx a lot!

ajayjadhav (Wed, 18 Mar 2020 13:54:10 GMT):
That's great to know @danielhardman . Do you think it can be achieved using the delegatable credential approach ? (RFC 104)

danielhardman (Wed, 18 Mar 2020 14:51:58 GMT):
Yes, that would be one way to do it. It could also be achieved more simply, by just giving the digital file to a new person (no protocol, just file copy); whoever has/uses the file first gets to watch the movie. There are ways you could combine this with exclusivity so it can only be held by one person at a time (e.g., by embedding in the cred a reference to something that's harder to transfer than a digital file, that must accompany the transfer and use).

danielhardman (Thu, 19 Mar 2020 01:35:57 GMT):
@domwoe: please see this doc for an answer to your question: https://docs.google.com/document/d/1t9JfZOdvapbby83i_zZlLziw7FkB9bLIXwvYRaO5lRk/edit#

domwoe (Thu, 19 Mar 2020 07:12:45 GMT):
Thanks Daniel. There's still a gap in my understanding. As far as I know, the default DIDcomm mode in Aries is authenticated encryption. Thereby, messages are authenticated but repudiable.

domwoe (Thu, 19 Mar 2020 07:12:45 GMT):
Thanks @danielhardman . There's still a gap in my understanding. As far as I know, the default DIDcomm mode in Aries is authenticated encryption. Thereby, messages are authenticated but repudiable.

gokulalex (Fri, 20 Mar 2020 05:27:07 GMT):
Has joined the channel.

iamShashvat (Mon, 23 Mar 2020 04:46:19 GMT):
Is it possible to create ATTRIB request by A's DID(owner) and send it to B for submit it to ledger? constraint: 'A' don't know the DID of 'B' means 'A' don't know the submitter DID or endorser DID, It only knows the address(URL) and APIs of B service for doing transaction. @danielhardman

andrew.whitehead (Mon, 23 Mar 2020 04:58:20 GMT):
Not unless B has A's private keys. The request has to be signed by A after the 'endorser' property is set to B's DID

iamShashvat (Mon, 23 Mar 2020 06:04:43 GMT):
Okay. Thanks!

vladyslavmunin (Mon, 23 Mar 2020 13:27:40 GMT):
Hi all, is it possible to register rich schema using indy-sdk ?

vladyslavmunin (Mon, 23 Mar 2020 13:27:40 GMT):
Hi all, is it possible to register rich schema using indy-sdk, e.g. with w3c credentials schema ?

vladyslavmunin (Mon, 23 Mar 2020 13:27:40 GMT):
Hi all, is it possible to register rich schema using indy-sdk, e.g. w3c credentials schema ?

khalifa (Mon, 23 Mar 2020 14:14:23 GMT):
Has joined the channel.

ashcherbakov (Mon, 23 Mar 2020 14:15:21 GMT):
Not yet. but we've been working on this feature right now. It's already possible to store rich schema on the Ledger (indy-node) side. The corresponding indy-sdk's API is coming

vladyslavmunin (Mon, 23 Mar 2020 14:16:55 GMT):
thanks for the answer

esplinr (Mon, 23 Mar 2020 15:02:51 GMT):
Starting the Indy Contributors call: https://zoom.us/j/244779296

zossimov (Tue, 24 Mar 2020 06:35:52 GMT):
Has joined the channel.

zossimov (Tue, 24 Mar 2020 06:43:48 GMT):
Hi, I'm defining custom Schemas and creating Credentials in Aca-Py and Aries .Net Agent. How to see if they successfully defined and created on indy ledger? How to retrieve the existing one? Thanks

zossimov (Tue, 24 Mar 2020 06:43:48 GMT):
Hi, I'm defining custom Schemas and creating Credentials in Aca-Py and Aries .Net Framework. How to see if they successfully defined and created on indy ledger? How to retrieve the existing one? Thanks

zossimov (Tue, 24 Mar 2020 06:43:48 GMT):
Hi, I'm defining custom Schemas and creating Credentials in Aca-Py and Aries .Net Framework. How to see if they are successfully defined and created on indy ledger? How to retrieve the existing one? Thanks

zossimov (Tue, 24 Mar 2020 06:43:48 GMT):
Hi, I'm defining custom Schemas and creating Credentials in Aca-Py and Aries .Net Framework. How to see if they are successfully defined and created on indy ledger? How to retrieve the existing one from indy ledger? Thanks

haydarK (Tue, 24 Mar 2020 15:30:29 GMT):
Has joined the channel.

lynn.bendixsen (Tue, 24 Mar 2020 17:01:21 GMT):
I use the Indy CLI to get the information from the ledger for both of your questions. `ledger get-schema ...` and `ledger get-cred-def ...` If you need a coding way to do it, you can look in the CLI code for an example.

domwoe (Tue, 24 Mar 2020 19:51:29 GMT):
@danielhardman You've put so much effort in writing this doc. It'd be great if you'd take another second to fill my gap here :)

haardikkk (Tue, 24 Mar 2020 21:45:24 GMT):
Has joined the channel.

haardikkk (Tue, 24 Mar 2020 21:45:25 GMT):
Im experimenting with the Indy SDK's, and using the js sdk to build a schema request, submit it, then build a credential definition request, and submit it. The schema goes through just fine, but the credential definition gives me this error: ``client request invalid: InvalidClientRequest("Mentioned seqNo (1) isn't seqNo of the schema.",)`

haardikkk (Tue, 24 Mar 2020 21:45:25 GMT):
Im experimenting with the Indy SDK's, and using the js sdk to build a schema request, submit it, then build a credential definition request, and submit it. The schema goes through just fine, but the credential definition gives me this error: `client request invalid: InvalidClientRequest('validation error [ClientClaimDefSubmitOperation]: cannot be smaller than 1 (ref=0)',)`

haardikkk (Tue, 24 Mar 2020 21:46:41 GMT):
But im creating the request using the SDK. The property its failing on is set by the SDK, not by me. Here's the code im using: ``` const [ credDefId, credDef ] = await indy.issuerCreateAndStoreCredentialDef( walletHandle, trustAnchorDid, schema, "tag_1", "CL", { support_revocation: true } ); console.log(credDef); const credDefRequest = await indy.buildCredDefRequest( trustAnchorDid, credDef ); console.log("Cred Def Request: ", credDefRequest); console.log("Credential Definition ID: ", credDefId); const credDefResponse = await indy.signAndSubmitRequest( poolHandle, walletHandle, trustAnchorDid, credDefRequest ); console.log("Cred Def Response: ", credDefResponse); ```

andrew.whitehead (Tue, 24 Mar 2020 21:55:21 GMT):
It means that the schema you're plugging into issuerCreateAndStoreCredentialDef doesn't have a sequence number. You might need to fetch it from the ledger rather than using the original

haardikkk (Tue, 24 Mar 2020 21:55:52 GMT):
Alright, thank you!

swcurran (Wed, 25 Mar 2020 00:06:14 GMT):
Just read the doc and have a similar question. What exactly does (a) the verifier track and (b) what do the present to show non-repudiation to a third party? This is a pretty important question given some of the use cases that I've seen for Indy - notably the consent receipt mechanism that relies on the creation and presentation of a proof as a demonstration of consent.

zossimov (Wed, 25 Mar 2020 01:07:17 GMT):
@lynn.bendixsen Can I use indy-cli to see what's on the ledger, being put on by Aries .Net or Aca-py? I tried on Aries .Net Web Agent, and it returned everything empty. indy> did list There is no opened wallet now indy> wallet list There are no wallets

lynn.bendixsen (Wed, 25 Mar 2020 01:15:59 GMT):
@zossimov yes, indy-cli can show anything thats on the ledger. You just need to create a pool and connect to it. You do not need a wallet or a did to get info from the ledger.

zossimov (Wed, 25 Mar 2020 01:18:01 GMT):
Wouldn't Aries .Net create a pool by its own?

zossimov (Wed, 25 Mar 2020 01:18:12 GMT):
here it is

zossimov (Wed, 25 Mar 2020 01:18:14 GMT):
indy> pool list +--------------+ | Pool | +--------------+ | local-pool | +--------------+ | INDYSCANPOOL | +--------------+ | default | +--------------+

zossimov (Wed, 25 Mar 2020 01:20:05 GMT):
to get the schemas, did is required

zossimov (Wed, 25 Mar 2020 01:20:05 GMT):
to get the schema did is required

lynn.bendixsen (Wed, 25 Mar 2020 01:29:39 GMT):
I am not too familiar with aries .net i guess, but you should just be able to create a wallet, open it, then > did new then > did use Then try the get-schema command again.

zossimov (Wed, 25 Mar 2020 01:35:45 GMT):
ok. Thanks

danielhardman (Wed, 25 Mar 2020 05:11:04 GMT):
It turns out that what I wrote in that doc is not completely accurate. The assertion that we are non-repudiable by default IS accurate; I just explained it wrong. (In my defense, I was repeating what a crypto engineer told me. I'm now asking for a deeper answer that I can share. Don't give up on me...)

haardikkk (Wed, 25 Mar 2020 15:52:21 GMT):
In the indy js issue-credential example, what is the encoding used for `credValues`? https://github.com/hyperledger/indy-sdk/blob/master/docs/how-tos/negotiate-proof/nodejs/step2.js#L42

haardikkk (Wed, 25 Mar 2020 15:52:21 GMT):
In the indy js issue-credential example, what is the encoding used for `credValues`? https://github.com/hyperledger/indy-sdk/blob/c34cbb411a972fc6c302bd931dad36393c202790/docs/how-tos/negotiate-proof/nodejs/step2.js#L48

haardikkk (Wed, 25 Mar 2020 15:52:21 GMT):
In the indy js negotiate proof example, what is the encoding used for `credValues`? https://github.com/hyperledger/indy-sdk/blob/c34cbb411a972fc6c302bd931dad36393c202790/docs/how-tos/negotiate-proof/nodejs/step2.js#L48

gnarula (Wed, 25 Mar 2020 23:59:31 GMT):
Hi everyone! I'm trying to build the state trie for the Pool state and I'm not getting the right root hash. My hunch is the key value pairs I'm adding to the state are probably different from what's there in the actual network. Can anyone out there help me by sharing the pool state (i.e. all k-v pairs in it) for buildernet please? That would help me reconcile the differences

gnarula (Thu, 26 Mar 2020 00:02:27 GMT):
So far, I've been using the logic in `plenum/server/request_handlers/node_handler.py: NodeHandler::update_state()` to update the state by parsing all pool txns. Am I missing something?

TharinduSandaruwan1 (Thu, 26 Mar 2020 04:10:37 GMT):
Has joined the channel.

TharinduSandaruwan1 (Thu, 26 Mar 2020 04:10:38 GMT):
I am studying Hyperledger Indy in there I tried getting started in nodejs wrapper but when I running it is failing on the following point. ` assert(await indy.verifierVerifyProof(jobApplicationProofRequestJson, decryptedJobApplicationProofJson, schemasJson, credDefsJson, revocRefDefsJson, revocRegsJson));` The error which is I am getting this point is, `IndyError: AnoncredsProofRejected` Can someone help me to figure out where the issue is?. I tried a lot to find out it but I was unable to do so that is why I am looking for your help. All other commands up to this point in getting started.js. Thank you very much.

vladyslavmunin (Thu, 26 Mar 2020 10:30:38 GMT):
verify

notsteward (Thu, 26 Mar 2020 12:52:24 GMT):
Hi! required help during implementation, My code meant to create wallet and store DID for A, if I run code for the first time, it will create it but when I run it second time how can I put check which sees if the DID already present in wallet, it won't create?

notsteward (Thu, 26 Mar 2020 12:54:44 GMT):
Hi! Required help during implementation! My code meant to create wallet and store DID for A, and if I run code for the first time, it will create it successfully but when I run it second time how can I insure that it won't create new keys again, how can I put check which sees if the DID already present in wallet or not?

hyper-sunder (Thu, 26 Mar 2020 13:56:17 GMT):
Hi All, Is there any way to recovery my wallet if I forgot the `wallet_key`? Like in bitcoin wallet we may get `six` word for recovery purpose. Is there anything like that?

esplinr (Thu, 26 Mar 2020 15:53:40 GMT):
The Indy wallet has a recovery seed. Some applications wrap that seed with a six word recovery, but that isn't something that Indy standardizes.

hyper-sunder (Fri, 27 Mar 2020 04:16:53 GMT):
@esplinr thanks for the reply. Could you please tell me how to generate the `recovery seed` for my indy wallet?

esplinr (Fri, 27 Mar 2020 05:02:20 GMT):
I was wrong. The seed isn't for the entire wallet, but for a specific DID key.

esplinr (Fri, 27 Mar 2020 05:02:42 GMT):
https://hyperledger-indy.readthedocs.io/projects/sdk/en/latest/docs/design/001-cli/README.html

hyper-sunder (Fri, 27 Mar 2020 05:57:05 GMT):
@esplinr thanks for the reply. Let me go through the link

notsteward (Fri, 27 Mar 2020 07:00:28 GMT):
@esplinr @danielhardman

notsteward (Fri, 27 Mar 2020 07:02:33 GMT):
How to check if there are keys present in wallet or not, and what are that keys using libindy?

iamShashvat (Fri, 27 Mar 2020 09:05:26 GMT):
you can open your wallet and use api like 'listmydidwithmeta' which return all the keys.

esplinr (Fri, 27 Mar 2020 15:59:13 GMT):
This week's Evernym demo video shows: * The rich schema implementation in Indy Node and Indy VDR * Aries interoperability between LibVCX and Streetcred https://www.youtube.com/watch?v=lZ84bCxEKWo&list=PLRp0viTDxBWGLdZk0aamtahB9cpJGV7ZF

Abhishekkishor (Fri, 27 Mar 2020 17:53:16 GMT):
Hello Guys, Hope you are doing I have a query. How to do performance testing of hyperledger indy? Is there any tool for that? @swcswcurran sir @esplinr sir can you please help me with this? Thanks

Abhishekkishor (Fri, 27 Mar 2020 17:53:16 GMT):
Hello Guys, Hope you are doing I have a query. How to do performance testing of hyperledger indy? Is there any tool for that? @swcurran sir @esplinr sir can you please help me with this? Thanks

swcurran (Fri, 27 Mar 2020 17:59:32 GMT):
What are you interested in testing? For example, some are interested in ledger performance, but you can issue 1M credentials with just 3 transactions on the ledger and reads by the agents receiving them. Do you have a use case in mind that you want to understand?

Abhishekkishor (Fri, 27 Mar 2020 18:03:09 GMT):
Actually we are building one application on hyperledger indy. So, we are instructed to test it's performance also. But not getting any resource for this.

Abhishekkishor (Fri, 27 Mar 2020 19:52:27 GMT):
Guys, I'm try to build 2 factor authentication using hyperledger indy. But not getting correct guidance and resource on it. So, can anyone help me with this by providing me the guidance and resource on this? I'll be grateful to you guys. Thanks.

Abhishekkishor (Fri, 27 Mar 2020 19:52:27 GMT):
Guys, I'm trying to build 2 factor authentication using hyperledger indy. But not getting correct guidance and resource on it. So, can anyone help me with this by providing me the guidance and resource on this? I'll be grateful to you guys. Thanks.

esplinr (Fri, 27 Mar 2020 20:35:29 GMT):
You need to be more specific.

esplinr (Fri, 27 Mar 2020 20:36:55 GMT):
See the notes and recording for this meeting: https://wiki.hyperledger.org/display/IWG/2019-08-15+Identity+WG+Implementers+Call

esplinr (Fri, 27 Mar 2020 20:38:12 GMT):
And this one: https://wiki.hyperledger.org/pages/viewpage.action?pageId=24781339

GKumar (Sat, 28 Mar 2020 10:42:11 GMT):
Has joined the channel.

mohammadhossein73 (Sun, 29 Mar 2020 05:45:40 GMT):
Has joined the channel.

hamidm (Sun, 29 Mar 2020 21:14:35 GMT):
Hello guys, what would the best practice to store pictures, like a picture as an attachment to a VerCred ? As an embedded attribute(strongest in my opinion), using some king of encoding (Base64) ? Thanks

swcurran (Sun, 29 Mar 2020 21:36:44 GMT):
That can and has been done. It will get better/easier when we move to Rich Schema with typing of claims. An alternative that can be used in some cases is to put a hash of the image data in a claim and then have the image elsewhere. Depending on the data, it could be stored in a public location. In that case, you might use a hashlink to indicate both the location of the data and its hash to prove it has not been tampered with. You could also store it on IPFS in that case. If the data is private, it's harder to handle. In that case, the verifier needs to get the data from somewhere - the issuer or the holder in most cases. It's technically OK if it comes from the holder, as the hash of the image has been committed to the credential. The logistics of the holder getting it, keeping it and making it available to a verifier is NOT so easy though. With the issuer holding it, there needs to be some sort of authorization (ideally from the holder) for the issuer to release the image to the verifier. This also results in a "call home problem" (the issuer knows the holder used their credential).

swcurran (Sun, 29 Mar 2020 21:36:44 GMT):
That can and has been done. It will get better/easier when we move to Rich Schema with typing of claims. An alternative that can be used in some cases is to put a hash of the image data in a claim and then have the image elsewhere. Depending on the data, it could be stored in a public location. In that case, you might use a hashlink to indicate both the location of the data and its hash to prove it has not been tampered with. You could also store it on IPFS in that case. If the data is private, it's harder to handle. In that case, the verifier needs to get the data from somewhere - the issuer or the holder in most cases. It's technically OK if it comes from the holder, as the hash of the image has been committed to the credential. The logistics of the holder getting it, keeping it and making it available to a verifier is NOT so easy though. With the issuer holding it, there needs to be some sort of authorization (ideally from the holder) for the issuer to release the image to the verifier. This also results in a "call home problem" (the issuer knows the holder used their credential).

hamidm (Sun, 29 Mar 2020 21:44:59 GMT):
Thanks

haardikkk (Mon, 30 Mar 2020 00:35:46 GMT):
has anyone successfully built the js indy sdk inside a docker container?

haardikkk (Mon, 30 Mar 2020 02:20:56 GMT):
Im trying to run the indy js sdk inside a docker container, with the indy pool running in another docker container. When I try to use `openPoolLedger` it gives me the `CommonInvalidState` error - but it works if the indy sdk js app is not inside docker

haardikkk (Mon, 30 Mar 2020 02:21:00 GMT):
anyone know why this might be happening?

haardikkk (Mon, 30 Mar 2020 02:51:44 GMT):
i feel like it has something to do with my pool ip, but im not sure

haardikkk (Mon, 30 Mar 2020 02:51:54 GMT):
@swcurran do you know anything about this?

haardikkk (Mon, 30 Mar 2020 02:59:57 GMT):
Nevermind, I was able to get it to work by changing the `pool_ip` ARG in the `indy-pool.dockerfile` to the name of my docker container (which I link to from my app)

andrew.whitehead (Mon, 30 Mar 2020 03:30:50 GMT):
Makes sense, I thought only IP addresses were allowed in there

TharinduSandaruwan1 (Mon, 30 Mar 2020 08:12:05 GMT):
can someone guide me to correctly setup Hyperledger Indy getting started in Nodejs wrapper. Becuase when I run it it failed at this point. `assert(await indy.verifierVerifyProof(jobApplicationProofRequestJson, decryptedJobApplicationProofJson, schemasJson, credDefsJson, revocRefDefsJson, revocRegsJson));` by giving an eror. If someone can help me it is really really great-full for me

hamidm (Mon, 30 Mar 2020 11:08:26 GMT):
Hello, is the the BlobStorageApi separate from the WalletStorageApi ? What I ask here, is: is there a blob storage capability in the Wallet ?

haardikkk (Mon, 30 Mar 2020 14:44:54 GMT):
Yeah, I was trying with your python indy-cli dockerfile example for a while, which I think uses 10.0.0.2, but couldn't get it to work. Tried `indy-pool` on a whim and it worked :D

haardikkk (Mon, 30 Mar 2020 17:02:13 GMT):
what error?

TharinduSandaruwan1 (Tue, 31 Mar 2020 11:20:25 GMT):
@haardikkk the error was

TharinduSandaruwan1 (Tue, 31 Mar 2020 11:20:29 GMT):
IndyError: AnoncredsProofRejected

iamShashvat (Tue, 31 Mar 2020 11:55:02 GMT):
how can I update Attribute (JSON)? In JSON, I have multiple key value pairs and I want to update the only one field!

iamShashvat (Tue, 31 Mar 2020 11:59:52 GMT):
{ "Attribute": {"keyA": {"keyAA": "valueA"},"keyB": {"keyBA": "valueBA","keyBB":"valueBB"}}}, here if I want to update only one value, how can I do that? buildAttribRequest ( submitterDid, targetDid, hash, raw, enc )

iamShashvat (Tue, 31 Mar 2020 12:00:01 GMT):
Attrib

domwoe (Tue, 31 Mar 2020 13:07:19 GMT):
you have to send the entire json with the updated field.

konda.kalyan (Wed, 01 Apr 2020 05:32:59 GMT):
Has joined the channel.

iamShashvat (Wed, 01 Apr 2020 05:42:39 GMT):
hmm okay! but don't you think there should be a way to update particular value, because there will be the case where I don't want to store some data persistent locally, every time if I have to change some data, I have to fetch it first from the ledger and overwrite json again.

iamShashvat (Wed, 01 Apr 2020 05:43:38 GMT):
just a point, we can have discussion over it

YaswanthSorapalli (Wed, 01 Apr 2020 09:21:30 GMT):
Has joined the channel.

YaswanthSorapalli (Wed, 01 Apr 2020 09:21:35 GMT):
what is the difference among indy, aries and sovrin ?

zossimov (Wed, 01 Apr 2020 09:26:38 GMT):
ledger get-schema did=Th7MpTaRZVRYnPiabds81Y name=fictional-passport-7f19495b8e3649e1965d20ee2f4740c9 version=1.1 There is no opened pool now

zossimov (Wed, 01 Apr 2020 09:27:18 GMT):
how to open a pool?

TharinduSandaruwan1 (Wed, 01 Apr 2020 11:44:15 GMT):
@YaswanthSorapalli Everynm is an organization who build the self sovrin identity and given it to the sovrin non-profit organization which is available as sovrin global network and then sovrin gave their sovrin code base to the Linux foundation and at the end which become as Hyperleger Indy project. Sovrin is one of the contributors of Hyperledger Indy. Hyperledger Aries is a protocol used for peer to peer communication, wallet, messaging, key management

TharinduSandaruwan1 (Wed, 01 Apr 2020 11:45:07 GMT):
Everynm is an organization who build the self sovrin identity and given it to the sovrin non-profit organization which is available as sovrin global network and then sovrin gave their sovrin code base to the Linux foundation and at the end which become as Hyperleger Indy project. Sovrin is one of the contributors of Hyperledger Indy. Hyperledger Aries is a protocol used for peer to peer communication, wallet, messaging, key management

l.m (Wed, 01 Apr 2020 12:22:16 GMT):
Has joined the channel.

l.m (Wed, 01 Apr 2020 12:22:24 GMT):
proxy

YaswanthSorapalli (Wed, 01 Apr 2020 12:45:57 GMT):
Thax man

gnarula (Wed, 01 Apr 2020 18:46:03 GMT):
Hi! I'm working on something that involves validating an indy txn and as a result I need to validate the BLS signatures. Does anyone know which curve is used by Indy? I think it's BN254 but are https://tools.ietf.org/id/draft-kasamatsu-bncurves-01.html#curve254a the parameters?

Abhishekkishor (Wed, 01 Apr 2020 20:35:56 GMT):
Hello Guys, I have a question. Can we integrate hyperledger Iny with caliper? If yes then can you please provide me some resources where I can get an idea about @swcurran sir @esplinr sir can you please guide me through this? Thanks & regards, Abhishek.

Abhishekkishor (Wed, 01 Apr 2020 20:45:19 GMT):
*Hyperledger Indy

TharinduSandaruwan1 (Thu, 02 Apr 2020 06:51:18 GMT):
When I am running Hyperledger Indy samples nodejs sample it gives me this error.``` "Acme" -> Create "Job-Application" Proof Request "Acme" -> Get key for Alice did "Acme" -> Authcrypt "Job-Application" Proof Request for Alice "Acme" -> Send authcrypted "Job-Application" Proof Request to Alice "Alice" -> Authdecrypt "Job-Application" Proof Request from Acme "Alice" -> Get credentials for "Job-Application" Proof Request "Alice" -> Get Schema from Ledger (node:122396) UnhandledPromiseRejectionWarning: IndyError: LedgerNotFound at Object.callback (/home/hyperledger_indy_sample/samples/mynodejs/node_modules/indy-sdk/src/wrapIndyCallback.js:15:10) (node:122396) UnhandledPromiseRejectionWarning: Unhandled promise rejection. This error originated either by throwing inside of an async function without a catch block, or by rejecting a promise which was not handled with .catch(). (rejection id: 1) (node:122396) [DEP0018] DeprecationWarning: Unhandled promise rejections are deprecated. In the future, promise rejections that are not handled will terminate the Node.js process with a non-zero exit code. azureadmin@mithfc5 ``` Can someone help me to find out this error?

hyper-sunder (Thu, 02 Apr 2020 09:32:13 GMT):
Hi Guys,

hyper-sunder (Thu, 02 Apr 2020 09:33:48 GMT):
I am using `vcx` wrapper library in my java project. When try to create `schema`, getting `Error: Unknown libindy error Caused by: Error: Unknown payment method type Caused by: Unknown payment method null`. Can anyone guide me how to configure `libnullpay` plugin? to overcome this issue

PatrikStas (Thu, 02 Apr 2020 09:39:07 GMT):
You don't have to interact with libnullpay, vcx loads it up during its initialization. Just make sure the compiled binary is on your system ( `/usr/lib/` on linux, `/usr/local/lib/`)

PatrikStas (Thu, 02 Apr 2020 09:39:07 GMT):
You don't have to interact with libnullpay, vcx loads it up during its initialization. Just make sure the compiled binary is on your system ( `/usr/lib/` on linux, `/usr/local/lib/` on osx)

hyper-sunder (Thu, 02 Apr 2020 09:49:40 GMT):
But `readme` file says, to use this plugin we should call `nullpay_init()` so that it allows null value while creating `schema`.

hyper-sunder (Thu, 02 Apr 2020 09:49:49 GMT):
@PatrikStas Thanks for the reply :)

hyper-sunder (Thu, 02 Apr 2020 09:53:14 GMT):
https://github.com/hyperledger/indy-sdk/tree/master/libnullpay#null-payment-plugin-for-indy-sdk

hyper-sunder (Thu, 02 Apr 2020 10:01:07 GMT):
in node sample also, its being initialised https://github.com/hyperledger/indy-sdk/blob/master/vcx/wrappers/node/demo/common.js#L22

PatrikStas (Thu, 02 Apr 2020 10:02:34 GMT):
Ohhh... yeah you are right, the VCX is actually not performing the libnullpay init, you have to do it, that's right. Just like in the NodeJS VCX demo. But once you do it, you are all set

hyper-sunder (Thu, 02 Apr 2020 10:05:07 GMT):
Thanks :)

notsteward (Thu, 02 Apr 2020 11:13:55 GMT):
Hey, I have to basic question, currently Public Hyperledger Indy network is Sovrin Mainnet, is this correct? And Is read operation in Indy/Sovrin is free?

MALodder (Thu, 02 Apr 2020 12:26:19 GMT):
Reads are free yes

hyper-sunder (Thu, 02 Apr 2020 12:29:10 GMT):
@PatrikStas Have completed with help of library `jnr-ffi`.

swcurran (Thu, 02 Apr 2020 13:46:45 GMT):
I've not heard of anyone doing anything with Caliper and Indy. Since Indy doesn't use (for the most part) the blockchain for large volumes of transactions, performance has not been as big an issue, so it hasn't been a big focus. Some performance testing has been done on Indy and for the time being, performance was deemed sufficient and not a big focus. What are you thinking about with that?

swcurran (Thu, 02 Apr 2020 13:47:27 GMT):
What code base are you running?

TharinduSandaruwan1 (Thu, 02 Apr 2020 13:47:52 GMT):
nodejs

swcurran (Thu, 02 Apr 2020 13:48:25 GMT):
I'm not sure, but @MALodder would know.

MALodder (Thu, 02 Apr 2020 13:48:54 GMT):
Yes its BN254

swcurran (Thu, 02 Apr 2020 17:47:56 GMT):
Which one - e.g. path to the repo you are using.

swcurran (Thu, 02 Apr 2020 17:48:06 GMT):
There are several :-(

TharinduSandaruwan1 (Thu, 02 Apr 2020 21:57:22 GMT):
I am using this one

TharinduSandaruwan1 (Thu, 02 Apr 2020 21:57:26 GMT):
https://github.com/hyperledger/indy-sdk/tree/master/samples/nodejs

swcurran (Thu, 02 Apr 2020 23:05:19 GMT):
OK - sorry I'm not familiar. The error seems to be that trying to connect to the ledger didn't work. Is the genesis file correct? Is it being read? Are the nodes of the ledger you are trying to use publically accessible - or at least from the JS code that you are running? Those are general thoughts. Sorry I can't help further... :-(

esplinr (Fri, 03 Apr 2020 00:36:09 GMT):
Anyone interested in the CI / CD for Indy should see this thread in #cicd : https://chat.hyperledger.org/channel/cicd?msg=ZwnJ8qPkEoDuFy347

vikramsharma13 (Fri, 03 Apr 2020 09:58:13 GMT):
Has joined the channel.

domwoe (Fri, 03 Apr 2020 10:01:04 GMT):
@danielhardman any updates on this

domwoe (Fri, 03 Apr 2020 10:01:04 GMT):
@danielhardman any updates on this?

zossimov (Fri, 03 Apr 2020 15:19:31 GMT):
WalletAgent:indy> pool connect Aries Pool "Aries" has not been connected.

zossimov (Fri, 03 Apr 2020 15:20:18 GMT):
can't connect with Pool

zossimov (Fri, 03 Apr 2020 15:43:24 GMT):
WebAgent:indy> pool connect Aries Pool "Aries" has not been connected.

zossimov (Fri, 03 Apr 2020 15:43:55 GMT):
Error: Invalid library state Caused by: Ledger merkle tree is not acceptable for current tree.

zossimov (Fri, 03 Apr 2020 15:44:10 GMT):
can't connect with pool

ultimo2020 (Fri, 03 Apr 2020 19:45:15 GMT):
Hi everyone. Do you know if Hyperledger Ursa is FIPS compliant?

lynn.bendixsen (Fri, 03 Apr 2020 21:34:41 GMT):
It is not, but theres a larger discussion around that issue that @MALodder is quite conversant about :)

MALodder (Fri, 03 Apr 2020 21:35:01 GMT):
yes I answered it in the #ursa channel

zossimov (Sat, 04 Apr 2020 01:24:16 GMT):
@lynn.bendixsen any idea?

zossimov (Sat, 04 Apr 2020 01:24:16 GMT):
@lynn.bendixsen

zossimov (Sat, 04 Apr 2020 01:24:56 GMT):
@lynn.bendixsen any idea why it return this?

ultimo2020 (Sat, 04 Apr 2020 05:56:06 GMT):
Thanks

thhkmgl (Sat, 04 Apr 2020 11:27:15 GMT):
Has joined the channel.

lynn.bendixsen (Sun, 05 Apr 2020 02:24:34 GMT):
It looks to me like your pool has been corrupted somehow. Is it your own test pool? If so, one way to fix it is that you can start over by deleting all of the ledger data (unless you need the data for some reason). Stop indy-node, delete the contents of the `/var/lib/indy//data` directory from each node and then restart indy-node on all of the nodes. This suggestion is more of a quick-and-dirty "workaround" than a fix, but it's quite difficult to determine the real fix through this chat channel and if your merkle tree is actually corrupted, there might be no other choice anyway. :)

ultimo2020 (Sun, 05 Apr 2020 08:18:01 GMT):
Hi everyone. Are there any official tests of the Indy scalability?

zossimov (Sun, 05 Apr 2020 10:54:53 GMT):
I created a new Pool name Aries and the gen_txt_file is https://github.com/Yoroitchi/AriesWebApp/blob/master/AriesWebApp/AriesTest.txn

zossimov (Sun, 05 Apr 2020 10:55:20 GMT):
I'm using Windows 10, where will I find this directory `/var/lib/indy//data`

zossimov (Sun, 05 Apr 2020 10:55:20 GMT):
I'm using Windows 10, where will I find this directory /var/lib/indy//data

zossimov (Sun, 05 Apr 2020 10:56:39 GMT):
`.indy_client\pool` directory for pool in windows, is this the one u told to remove?

lynn.bendixsen (Sun, 05 Apr 2020 14:03:01 GMT):
No, the error appears to be on the node side, not in your windows client. Are you usung someone else's pool?

zossimov (Sun, 05 Apr 2020 14:05:16 GMT):
https://github.com/Yoroitchi/AriesWebApp/blob/master/AriesWebApp/AriesTest.txn

lynn.bendixsen (Sun, 05 Apr 2020 14:13:53 GMT):
The directory i suggested to remove is on the nodes, not in your windows client. If you don't control the nodes, then I don't think you can fix a merkle tree problem. About all you can do from the client side is `pool delete`, then `pool create` to try again...

zossimov (Sun, 05 Apr 2020 14:15:52 GMT):
aah, now got it. Is there any way to know, of whom Node i'm using?

lynn.bendixsen (Sun, 05 Apr 2020 14:30:13 GMT):
The genesis file you are using appears to have local IP addresses in it, so that is not likely to be publicly available. I am not sure what you are trying to test, but it looks like you are trying to use someone else's pool. You could try the Sovrin BuilderNet if you are looking for a public test pool. That pool requires you to add your own schemas/cred_defs/etc. https://raw.githubusercontent.com/sovrin-foundation/sovrin/master/sovrin/pool_transactions_builder_genesis

zossimov (Sun, 05 Apr 2020 14:31:20 GMT):
ok. thanks alot :)

sanjay_sharma87 (Sun, 05 Apr 2020 18:06:59 GMT):
Has joined the channel.

vikramsharma13 (Mon, 06 Apr 2020 06:46:17 GMT):
git

vikramsharma13 (Mon, 06 Apr 2020 06:50:21 GMT):
Hey Guys, I do not see many sample/opensource implementations of Indy. Can anyone point me to the right resources? I found one KYC example using Indy but that too is 2 years old.

letsblockchain (Mon, 06 Apr 2020 07:56:02 GMT):
Has joined the channel.

letsblockchain (Mon, 06 Apr 2020 07:58:29 GMT):
Should Indy be used in any implementations? given that we its in active development and going for complete revamp?

swcurran (Mon, 06 Apr 2020 13:20:26 GMT):
You should be looking at Aries. Aries embeds in Indy and enables interfaces with other ledger and verifiable credential implementations. It is a very active set of projects and in regular use. Depending on your use case, you might look at different repos.

swcurran (Mon, 06 Apr 2020 13:20:26 GMT):
You should be looking at Aries. Aries embeds in Indy and enables interfaces with other ledger and verifiable credential implementations. It is a very active set of projects and in regular use. Depending on your use case, you might look at different repos in Aries. We put together this edX course on Aries - https://www.edx.org/course/becoming-a-hyperledger-aries-developer

Diiaablo95 (Mon, 06 Apr 2020 13:37:55 GMT):
Hello people! I have tried so hard, but I still did not manage to solve the permission error that I get every time I try to save some tails files in an Android application. I updated to libindy 1.15.0, but still I get the same error (https://postimg.cc/mcyzL50q). I also tried to pre-create the needed folder and set world-wide read/write permissions on it, with no luck. Has anyone had the same problem, and been able to find a solution? The error says: V/indy::api::anoncreds: prepare_result_3: >>> Err(IndyError { inner: Os { code: 13, kind: PermissionDenied, message: "Permission denied" }

haydarmajeed (Mon, 06 Apr 2020 22:30:50 GMT):
My team and I have developed an [SSI digital wallet] that works on the top of Hyperledger Indy to issuer Verifiable Credentials. Currently, we are using an internal ledger that is hosted on Google Cloud Platform and we would like to move to the Sovrin Network. We made this decision because we are trying to assist the country of Iraq with the COVID-19 crisis by issuing digital credentials for patients, they’re one of the countries that do not have any means to identity people who have been tested/vaccinated. We have developed the digital wallet (mobile app), the web-portals for (issuers & verifiers). We would like to have the credentials issued on the Sovrin Network. So I was looking for any advices or recommendations you may have based on your experience with Sovrin. We are paying out of pocket for this project as NGO and non-profit organization so your assistance will be a tremendous here. Please see the diagram below for our setup. I would like to know things like, Are we making the right decision by considering Sovrin? What is there a way for Sovrin or someone else to give us a rough estimate for the cost, I know the pricing are available on the Sovrin’s website but wanted to see if I can get more details? Thank you kindly, Haydar

haydarmajeed (Mon, 06 Apr 2020 22:30:50 GMT):
@all My team and I have developed an [SSI digital wallet] that works on the top of Hyperledger Indy to issuer Verifiable Credentials. Currently, we are using an internal ledger that is hosted on Google Cloud Platform and we would like to move to the Sovrin Network. We made this decision because we are trying to assist the country of Iraq with the COVID-19 crisis by issuing digital credentials for patients, they’re one of the countries that do not have any means to identify people who have been tested/vaccinated. We have developed the digital wallet (mobile app), the web-portals for (issuers & verifiers). We would like to have the credentials issued on the Sovrin Network. So I was looking for any advices or recommendations you may have based on your experience with Sovrin. We are paying out of pocket for this project as NGO and non-profit organization so your assistance will be a tremendous here. Please see the diagram below for our setup. I would like to know things like, Are we making the right decision by considering Sovrin? What is there a way for Sovrin or someone else to give us a rough estimate for the cost, I know the pricing is available on the Sovrin’s website but wanted to see if I can get more details? Thank you kindly, Haydar

swcurran (Mon, 06 Apr 2020 22:53:39 GMT):
It's relatively easy to determine the cost based on: - the number of issuers you have, - the number of unique schemas you will be supporting, - the number of credential definitions you will need (number of issuers of each schema) - the revocation model you will use. The next challenge you will have is actually writing to the ledger. That has to be done by having you and an endorser (usually a steward) sign the transactions to be written. That is quite a painful sequence right now. We have an example of the steps if you are interested.

swcurran (Mon, 06 Apr 2020 23:44:28 GMT):
If you can provide those numbers, we can figure out the costs.

haydarmajeed (Tue, 07 Apr 2020 04:20:34 GMT):
Currently I am thinking about piloting this application to about 18 provinces in Iraq, each province will have (2) issuers. We are still debating the number of schemas but think we will need (3) schemas initially: A. schema for people tested negative, B. people tested positive. not sure about the credential definition and the revocation model.

haydarmajeed (Tue, 07 Apr 2020 04:20:34 GMT):
@swcurran Currently I am thinking about piloting this application to about 18 provinces in Iraq, each province will have (2) issuers. We are still debating the number of schemas but think we will need (3) schemas initially: A. schema for people tested negative, B. people tested positive. not sure about the credential definition and the revocation model.

ultimo2020 (Tue, 07 Apr 2020 07:17:18 GMT):
Hi all. I am trying to setup Selfserve.sovrin.org and the download of libindy is really slow. Do you have any alternatives to setup on Ubuntu 19 server the Indy cli and wallet?

newbiestat (Tue, 07 Apr 2020 09:41:20 GMT):
Has joined the channel.

notsteward (Tue, 07 Apr 2020 09:48:27 GMT):
I am trying to understand the architecture of Indy, I want to know why each node have 2 open ports(client and server)?

notsteward (Tue, 07 Apr 2020 11:28:29 GMT):
Hi! I want to discuss a case: What if, I did the transaction and it took time to get commit on ledger. Do I have to wait for transaction to get complete or I can do other upcoming processes? In short: Any pending stages can be possible during transaction?

AnushkaIbm (Tue, 07 Apr 2020 14:17:29 GMT):
Has joined the channel.

AnushkaIbm (Tue, 07 Apr 2020 14:17:31 GMT):
Hi I am facing issue when m trying to run the script of NegotiateProof of Indy-sdk . Can anyone plz help with the following error: { IndyError: CommonInvalidState at Object.callback (C:\Users\AnushkaAgarwal\Desktop\Digital Identity Framework\indy-sdk\samples\nodejs\node_modules\indy-sdk\src\wrapIndyCallback.js:15:10) name: 'IndyError', message: 'CommonInvalidState', indyCode: 112, indyName: 'CommonInvalidState', indyCurrentErrorJson: null }

AnushkaIbm (Tue, 07 Apr 2020 14:19:18 GMT):
I have one more question regarding the *seed* creation on Indy-pool , is there any way to generalize it or its always specific as its needed for the DID creation further

VithushanJegatheeswaran (Tue, 07 Apr 2020 17:06:55 GMT):
Hey guys, Is it possible to check with ZKP ? or does ZKP only work with numbers at the moment ? ``` "requested_predicates": { "predicate1_referent": { "name": "age", "p_type": ">=", "p_value": 18, "restrictions": [{"schema_id": gvt_schema_id}] } } ``` instead of the age I'd like to check if a date is older than 90 days for example.

swcurran (Tue, 07 Apr 2020 17:11:17 GMT):
For reasons we're trying to clarify, predicates other than ">=" have not been implemented. We're not sure if there is a technical reason for that and we're trying to determine that.

swcurran (Tue, 07 Apr 2020 17:11:58 GMT):
Without additional predicates (notably "<") you can't do the type of test you are wanting to do as a predicate.

VithushanJegatheeswaran (Tue, 07 Apr 2020 17:12:19 GMT):
thank you :)

danielhardman (Tue, 07 Apr 2020 17:18:05 GMT):
@esplinr ^^

notsteward (Tue, 07 Apr 2020 18:59:26 GMT):
@lynn.bendixsen @danielhardman @swcurran Please tell me, is there any pending stage possible?

danielhardman (Tue, 07 Apr 2020 19:00:57 GMT):
Transactions typically have a latency of ~1 second. Sometimes under heavy load it is as much as a handful of seconds. While that time is elapsing, I suppose you could consider the status as "pending". However, from the ledger's perspective, the transaction is atomic -- either committed or not.

ultimo2020 (Tue, 07 Apr 2020 19:01:46 GMT):
Hi. Managed to compile under Ubuntu 19.04 the @lynn.bendixsen Indy sdk and Indy cli version latest 1.15 but I am getting errors when compiling the libsovtoken, that it depends on libindy 1.6.8 although I have the latest, any advice here maybe?

ultimo2020 (Tue, 07 Apr 2020 19:01:46 GMT):
Hi. Managed to compile under Ubuntu 19.04 the @lynn.bendixsen Indy sdk and Indy cli version latest 1.15 but I am getting errors when compiling the libsovtoken, that it depends on libindy 1.6.8 although I have the latest, any advice here maybe? @everyone I have successfully built the Indy sdk and started the pool on Ubuntu 19.04 but I get those libsovtoken errors on trying to install https://github.com/sovrin-foundation/libsovtoken I would like to connect and further develop on the Buildernet of Sovrin.

swcurran (Tue, 07 Apr 2020 19:12:39 GMT):
Further, ledger writes are generally not time sensitive. When you are an issuer (the only participant that needs to write), the writes should be done during a provisioning stage, not operational. Even revocations are likely to be a batched vs. real time process. However, revocation is the closest to a real time operation.

notsteward (Tue, 07 Apr 2020 19:16:04 GMT):
Pardon me @swcurran "the writes should be done during a provisioning stage, not operational" can you please tell what do you mean by that?

danielhardman (Tue, 07 Apr 2020 19:16:32 GMT):
@notsteward: they're part of issuer setup. They don't need to happen on an ongoing basis.

danielhardman (Tue, 07 Apr 2020 19:16:47 GMT):
They only happen once (or a handful of times) at the beginning.

notsteward (Tue, 07 Apr 2020 19:29:58 GMT):
Oh! I guess we are not on same page, let me put it in different way. In other consensus, like POS we have miners, they mine batch of transactions and commit it on blockchain. If mining is getting delayed, it simply give the status of "pending",with a (hash/Uid) by which in future.. client application can look to status for(successful or failure). Basically Non-blocking system. But in Indy if submitter submit some transaction in reply it directly get success or failure (basically have to wait it transaction complete)(Blocking system)

notsteward (Tue, 07 Apr 2020 19:30:43 GMT):
Let me know if you get it?

danielhardman (Tue, 07 Apr 2020 19:41:25 GMT):
@notsteward: Indy does not use a proof-of-work mechanism for consensus. Instead, it uses a byzantine fault tolerant consensus algorithm. This means there is no such thing as a pending state like there is in bitcoin. A transaction gets submitted. The nodes talk amongst themselves to come to consensus on it (which takes a second or two). Then the transaction is committed, once, permanently, globally, instantly, forever. You don't have to wait until more blocks get committed to achieve high confidence that it can't be undone. Therefore there is nothing like Bitcoin's pending status.

notsteward (Tue, 07 Apr 2020 19:48:34 GMT):
Thank you daniel and swcurran, My main confusion lies with RBFT then, So basically every transaction done quickly no long waiting period like POS/POW consensus! Thanks for resolving my confusion!

danielhardman (Tue, 07 Apr 2020 19:49:24 GMT):
(This is a major reason why Indy was created. It's purpose-built for identity, and its consensus uses this alternate mechanism instead of proof of work so there is no significant latency and no pending status on transactions.)

notsteward (Tue, 07 Apr 2020 19:50:33 GMT):
This is true USP! Thanks..

swcurran (Tue, 07 Apr 2020 20:29:44 GMT):
And with all that - back to the, ledger transactions are rare. To be a first time credential issuer, you have between 2 and 4 transactions and then you can issue 10 or 1 million credentials - no further writes needed. If you revoke credentials, there is a write, but that is likely to periodic - e.g. daily. What's important is you are not doing a ledger write per app level transaction.

iamShashvat (Wed, 08 Apr 2020 05:27:07 GMT):
Agreed! I was asking mostly for NYM Transaction(1000-2000 transactions probably)

notsteward (Wed, 08 Apr 2020 05:29:05 GMT):
Okay Agreed! Thanks, Mainly I was asking for NYM transactions(2000-3000) at a time.

danielhardman (Wed, 08 Apr 2020 05:29:58 GMT):
You only need to do NYM transactions for issuers. Ordinary DIDs can be peer DIDs that don't require a write to the ledger.

notsteward (Wed, 08 Apr 2020 05:34:53 GMT):
True.. but I was looking for the public DID which should be on a ledger for the issuers. And Ordinary DIDs means pairwise DID correct?

notsteward (Wed, 08 Apr 2020 05:57:01 GMT):
True.. but I was looking for the public DID which should be on a ledger, daniel, here use case might differ. And Ordinary DIDs means pairwise DID correct?

ultimo2020 (Wed, 08 Apr 2020 12:38:28 GMT):
Hi. Managed to get it all working with Ubuntu 16.04 and sovrin repo. I have created wallets and sucessfully wrote some dids

wizAmit (Wed, 08 Apr 2020 14:11:53 GMT):
Has joined the channel.

YaswanthSorapalli (Wed, 08 Apr 2020 14:48:56 GMT):
How to connect to indy mainnet? So that I can do transactions on mainnet.

lynn.bendixsen (Wed, 08 Apr 2020 14:59:39 GMT):
You can connect to the Sovrin MainNet (Sovrin's production Indy network) using this genesis file: https://raw.githubusercontent.com/sovrin-foundation/sovrin/stable/sovrin/pool_transactions_live_genesis MainNet write transactions require an Endorser and there is a small fee for each write. To become an Endorser yourself, you must sign the Endorser agreement, then the Trustees will add your Endorser DID to the ledger.

lynn.bendixsen (Wed, 08 Apr 2020 15:48:17 GMT):
One reason for 2 ports is to avoid DDOS attack from client side causing problems for Node/consensus communications. The Node port is for Node communications only and is recommended to be firewalled such that only other nodes can use it. The Client port is open to the world for access to read from and write to the ledger.

mccown (Wed, 08 Apr 2020 19:20:20 GMT):
Just a reminder about the Identity Implementer's WG Call tomorrow morning (April 9th @ 9am MT). This is a chance to get updates on the WG calls you may have missed this week. Our guest, Riley Hughes (Streetcred), will be giving an overview and insights into the Sovrin Ecosystem transition process. Here is the calendar invite with times and Zoom details. https://tinyurl.com/toykyqp

danielhardman (Wed, 08 Apr 2020 23:42:19 GMT):
By "ordinary DIDs" I meant DIDs that aren't used to issue credentials. All ordinary DIDs can be off ledger. Putting ordinary DIDs on the ledger is not allowed on the Sovrin MainNet (because they are PII which can cause legal problems in some jurisdictions), but might be allowed in other instances of an Indy network.

danielhardman (Wed, 08 Apr 2020 23:53:06 GMT):
I have rewritten my doc to answer your question more accurately and clearly. Please let me know if it helps. https://docs.google.com/document/d/1tB_TxD6-0tF9ggc4k7RWwx8QQw2cYirQJGGrDdUB5Hs/edit

esplinr (Thu, 09 Apr 2020 00:07:15 GMT):
And in my opinion you should consider Sovrin. It's designed exactly for your use case.

letsblockchain (Thu, 09 Apr 2020 05:02:19 GMT):
But Aries is still in incubation?

letsblockchain (Thu, 09 Apr 2020 05:03:15 GMT):
But Aries is still in Incubation?

letsblockchain (Thu, 09 Apr 2020 05:04:36 GMT):
Thanks for the course. It is definitely helpful to go through the course.

ultimo2020 (Thu, 09 Apr 2020 08:26:27 GMT):
Greetings everyon. I am going trough the Python samples in this url: https://github.com/hyperledger/indy-sdk/blob/master/docs/how-tos/write-did-and-query-verkey/python/utils.py I have seen that for the DID creation and all other ledger actions, the ledger interraction is bein local with DOCKER containers, the 4 ones. I have now got test tokens for the Sovrin Builder Net. Does anyone knows similar Python samples from Sovrin so that one can play around with the "pool" of Sovrin builder net or staging and to do writes there and not on the local indy Docker nodes? Thanks in advance,

swcurran (Thu, 09 Apr 2020 20:28:17 GMT):
An update to this. I've looked through the code in the indy-sdk and it looks like there might be more predicates supported - per this line: https://github.com/hyperledger/indy-sdk/blob/ebdf1b62b4b744b94155cb6f032367540b33556c/libindy/src/services/anoncreds/prover.rs#L324 There are no test cases in the sdk for the other predicates (other than GE). We're going to do some experimenting to see if it actually works.

danielhardman (Thu, 09 Apr 2020 20:29:50 GMT):
It is often the case that if you implement some of the comparison operators, others can be implemented by reversing the operands. That may be why tests only cover one? (Still, we should fix this.)

PatrikStas (Thu, 09 Apr 2020 21:05:50 GMT):
Hello fellow indy devs, I've released and just redeployed new version of Indyscan https://indyscan.io/ Brings some nice UI/UX improvments, most notably live updates on homepage. No need to refresh page anymore. Enjoy.

MumtahinHabib (Fri, 10 Apr 2020 16:02:08 GMT):
Has joined the channel.

rakaar (Fri, 10 Apr 2020 16:35:42 GMT):
Has joined the channel.

rickkce (Fri, 10 Apr 2020 18:10:19 GMT):
Has joined the channel.

rickkce (Fri, 10 Apr 2020 18:10:20 GMT):
Hello guys, the German government is supporting open-source projects similar to the summer of code from google. For that, I would like to work on Hyperledger. The start will be July. I need a mentor and an idea for content I can contribute to the open-source project. I'm an experienced fullstack dev, worked with Etherum and have a knowledge of other Blockchains. Is there anyone that can help me find a topic I could work on? Highly appreciate you help

swcurran (Fri, 10 Apr 2020 18:33:10 GMT):
We should be able to find something in Indy and Aries. Could you provide some more information on what you would like to do? Are you familiar with Indy/Aries? There are multiple levels to the Indy/Aries/Ursa stack, and so where you might fit would depend on your interests and background.

rickkce (Fri, 10 Apr 2020 18:38:14 GMT):
Indy is a very interesting project for me. I have already dealt a lot with the topic of digital identity. Among others the Cosmos SDK with Tendermint and other SDKs that try to make a ledger for KYC an industry standard. The idea would be to find a topic here, so that I can apply for the scholarship with it and then to get familiar with Hyperledger Indy until July, so that I can start directly in and work on it until September.

rickkce (Fri, 10 Apr 2020 18:38:14 GMT):
Indy is a very interesting project for me. I have already dealt a lot with the topic of digital identity. Among others the Cosmos SDK with Tendermint and other SDKs that try to make a ledger for KYC an industry standard. The idea would be to find a topic here, so that I can apply for the scholarship with it and then to get familiar with Hyperledger Indy until July, so that I can start directly and work on it until September.

rickkce (Fri, 10 Apr 2020 18:39:07 GMT):
I have previous experience with other Blockchains so it should be no problem to get into Hyperledger until July

rickkce (Fri, 10 Apr 2020 18:40:34 GMT):
Its a three months program

rickkce (Fri, 10 Apr 2020 18:41:03 GMT):
but I'd like to keep working on it afterwards

rickkce (Fri, 10 Apr 2020 18:41:14 GMT):
So the program is good to kickstart it

swcurran (Fri, 10 Apr 2020 20:08:05 GMT):
The thing with Indy is that the blockchain layer is relatively light - it is not used as much as other ledgers. With Indy, the exchange of credentials is the high volume interaction and that is all off chain. So if you specifically want to work at the blockchain level, there is not a lot going on. We could find something, but right now, not a lot. There is more at the crypto level if you are cryptographer, or at the ledger client layer.

swcurran (Fri, 10 Apr 2020 20:09:23 GMT):
So, if you are wanting to do Blockchain with Indy, I would think about the performance tradeoffs of the current implementation (Plenum) with other comparable approaches (e.g. Tendermint).

swcurran (Fri, 10 Apr 2020 20:09:52 GMT):
There has been some investigation into that, but it has slowed down - no one to work it.

rickkce (Fri, 10 Apr 2020 20:10:54 GMT):
To work on cryptography is also interesting to me. However I see the potential here to work with Tendermint

swcurran (Fri, 10 Apr 2020 20:10:55 GMT):
The volume of writes will go up in proportion to the number of issuers*credential types (not credentials). That will eventually pressure the ledger.

swcurran (Fri, 10 Apr 2020 20:11:04 GMT):
I'

swcurran (Fri, 10 Apr 2020 20:11:04 GMT):
I'll try to find the conversations about that...will take me a bit to do that, but let's follow up on that.

rickkce (Fri, 10 Apr 2020 20:11:22 GMT):
Thats seems to be a very interesting topic to me

rickkce (Fri, 10 Apr 2020 20:12:15 GMT):
I just found out about the scholarship and have to apply in 7 days

rickkce (Fri, 10 Apr 2020 20:12:28 GMT):
Do you think that is enough time to define the topic and find a mentor?

rickkce (Fri, 10 Apr 2020 20:12:46 GMT):
The program then starts in July like I said so there is enough time to discuss more in detail about it later

rickkce (Fri, 10 Apr 2020 20:14:54 GMT):
I just need to write a 2 page proposal which is easy for me since I'm experienced with Blockchain

swcurran (Fri, 10 Apr 2020 20:15:03 GMT):
Maybe. I would not be best if the focus is at that layer. But let's see. Take a look at this: https://docs.google.com/presentation/d/1vs_wWhSCRlEMA9RdS4bGUn8XIcsHUKK8t7Nw2qf6w8A/edit#slide=id.g45e53a55e3_0_0 And scan the #indy-ledger-next channel for the discussion. Focus on the "plenum" vs. other options. There is some talk about DNS and things that are not very relevant - skip that.

rickkce (Fri, 10 Apr 2020 20:15:04 GMT):
And a statement from the mentor

swcurran (Fri, 10 Apr 2020 20:15:59 GMT):
Suggest adding a question on that channel. I'll check with others as well.

rickkce (Fri, 10 Apr 2020 20:16:30 GMT):
Okay I'll have a better overall look on the topic tomorrow and will go through the google doc so we can follow up

rickkce (Fri, 10 Apr 2020 20:16:39 GMT):
Thank you very much for the help!

rickkce (Fri, 10 Apr 2020 20:17:21 GMT):
Would you liked to follow up in Direct Messages or what is the best way to state in touch with you?

rickkce (Fri, 10 Apr 2020 20:17:21 GMT):
Would you liked to follow up in Direct Messages or what is the best way to stay in touch with you?

rickkce (Fri, 10 Apr 2020 20:18:46 GMT):
Would you liked to follow up in Direct Messages or what is the best way to stay in touch with you?

swcurran (Fri, 10 Apr 2020 20:21:22 GMT):
DM would work. I would definitely ask questions and for a mentor on that other channel if the project or some angle on the project is interesting.

rickkce (Fri, 10 Apr 2020 20:22:31 GMT):
Okay thanks that helped a lot

AshutoshKumar7 (Sat, 11 Apr 2020 02:56:36 GMT):
Has joined the channel.

swcurran (Sat, 11 Apr 2020 21:45:57 GMT):
Maintainers: Now that security advisories are supported in github, can we switch from Jira to GitHub issues for Indy issues? https://help.github.com/en/github/managing-security-vulnerabilities/managing-security-vulnerabilities-in-your-project

iamShashvat (Sun, 12 Apr 2020 17:19:49 GMT):
`client request invalid: InvalidClientRequest(\'validation error [ClientAttribOperation]: not a valid hash (needs to be in hex too) (hash=0x00)` I am setting 'raw' value in buildAttribRequest(DID, DID, hash, raw, enc) (my golang wrapper), hash and enc are (optional), so I have to set it with null, if I set it null(0x00) I got this error! any solution please?

andrew.whitehead (Sun, 12 Apr 2020 17:29:13 GMT):
I think it needs to leave the key out of the request entirely

iamShashvat (Sun, 12 Apr 2020 17:51:17 GMT):
I can't put an empty parameter BuildAttribRequest(did, did,"", {"hello":"world"}, "") --> this giving an error

andrew.whitehead (Sun, 12 Apr 2020 17:55:46 GMT):
It needs to be null in the encoded JSON (not 0), or not present

iamShashvat (Sun, 12 Apr 2020 17:59:11 GMT):
yes, agreed! and I am trying for the same... like JS, java, etc have null! in golang does not have predefined null or None.

andrew.whitehead (Sun, 12 Apr 2020 18:01:12 GMT):
Looks like it might be 'nil'?

iamShashvat (Sun, 12 Apr 2020 18:01:39 GMT):
cannot use nil as type string in argument to indysdk.BuildAttribRequest

iamShashvat (Sun, 12 Apr 2020 18:01:49 GMT):
comile time error

iamShashvat (Sun, 12 Apr 2020 18:02:07 GMT):
compile time error :(

andrew.whitehead (Sun, 12 Apr 2020 18:02:47 GMT):
What SDK wrapper is it using?

iamShashvat (Sun, 12 Apr 2020 18:03:34 GMT):
I make my own golang wrapper!

andrew.whitehead (Sun, 12 Apr 2020 18:08:25 GMT):
I don't really know Go but it looks like you can use a pointer to accept optional values

notsteward (Sun, 12 Apr 2020 18:54:54 GMT):
cannot use strPtr("foo") (type *string) as type string in argument to indysdk.BuildAttribRequest

iamShashvat (Sun, 12 Apr 2020 18:55:51 GMT):
cannot use strPtr("foo") (type *string) as type string in argument to indysdk.BuildAttribRequest

KGiou (Sun, 12 Apr 2020 19:34:38 GMT):
Has joined the channel.

notsteward (Sun, 12 Apr 2020 20:30:36 GMT):
@andrew.whitehead that works, I guess!

notsteward (Sun, 12 Apr 2020 20:35:36 GMT):
`{ identifier: 'QRxM3YDB9HVUegdji39v4P', operation: { dest: 'QRxM3YDB9HVUegdji39v4P', raw: '{"endpoint":{"ha":"127.0.0.1:5555"}}', type: '100' }, protocolVersion: 2, reqId: 1586723364273894400, signature: '3J85UdtHZADRv7ZrnUNqoEnyV6vxsXXqo4rsyze6RWfbqJdKE41XrMEhDehJxcUFNip4xP3nroi4si2SBQ4dJHKi' } { identifier: 'QRxM3YDB9HVUegdji39v4P', reqId: 1586723364273894400, reason: 'client request invalid: insufficient number of valid signatures, 1 is required but 0 valid and 1 invalid have been provided. The following signatures are invalid: did=QRxM3YDB9HVUegdji39v4P, signature=3J85UdtHZADRv7ZrnUNqoEnyV6vxsXXqo4rsyze6RWfbqJdKE41XrMEhDehJxcUFNip4xP3nroi4si2SBQ4dJHKi', op: 'REQNACK' }` Functions: request = BuildAttribRequest(did, did, `{"endpoint":{"ha":"127.0.0.1:5555"}}`) SignRequest(wh, did, request) submitRequest(ph,signedRequest) What is the problem here? @andrew.whitehead

iamShashvat (Sun, 12 Apr 2020 20:48:37 GMT):
`{ identifier: 'QRxM3YDB9HVUegdji39v4P', operation: { dest: 'QRxM3YDB9HVUegdji39v4P', raw: '{"endpoint":{"ha":"127.0.0.1:5555"}}', type: '100' }, protocolVersion: 2, reqId: 1586723364273894400, signature: '3J85UdtHZADRv7ZrnUNqoEnyV6vxsXXqo4rsyze6RWfbqJdKE41XrMEhDehJxcUFNip4xP3nroi4si2SBQ4dJHKi' } { identifier: 'QRxM3YDB9HVUegdji39v4P', reqId: 1586723364273894400, reason: 'client request invalid: insufficient number of valid signatures, 1 is required but 0 valid and 1 invalid have been provided. The following signatures are invalid: did=QRxM3YDB9HVUegdji39v4P, signature=3J85UdtHZADRv7ZrnUNqoEnyV6vxsXXqo4rsyze6RWfbqJdKE41XrMEhDehJxcUFNip4xP3nroi4si2SBQ4dJHKi', op: 'REQNACK' } ` Functions: request = BuildAttribRequest(did, did,nil, {"endpoint":{"ha":"127.0.0.1:5555"}},nil) SignRequest(wh, did, request) submitRequest(ph,signedRequest) What is the problem here?

andrew.whitehead (Sun, 12 Apr 2020 20:53:16 GMT):
Not sure. Seems like it's not generating the signature correctly

andrew.whitehead (Sun, 12 Apr 2020 20:56:09 GMT):
Or somehow the public key on the ledger doesn't match the key in the wallet

iamShashvat (Sun, 12 Apr 2020 21:17:32 GMT):
I have checked the public key, it's correct and matched with the ledger! I guess signature is not correct. How can it be possible, any more possible cases where signature can get invalid?

andrew.whitehead (Sun, 12 Apr 2020 21:27:14 GMT):
Does the output match what you're passing to SignRequest?

andrew.whitehead (Sun, 12 Apr 2020 21:31:11 GMT):
Signatures are sensitive to the order of the keys but it should handle that automatically

MrPhenom (Mon, 13 Apr 2020 08:51:36 GMT):
Has joined the channel.

aditya520 (Mon, 13 Apr 2020 13:57:12 GMT):
Has joined the channel.

danielhardman (Mon, 13 Apr 2020 22:39:20 GMT):
+1 from me

rmscott (Tue, 14 Apr 2020 05:36:07 GMT):
Has joined the channel.

cam-parra (Tue, 14 Apr 2020 14:40:01 GMT):
@andrew.whitehead @swcurran Do you have maintainer privalages on the indy-node/indy-plenum repos? If @esplinr team is out for the month we should not halt progress in these repos

esplinr (Tue, 14 Apr 2020 14:46:13 GMT):
I'm completely comfortable using GitHub issues for the new repos (i.e. indy-vdr). I'm a bit nervous with indy-node and indy-plenum because there are lots of good information in Jira that it would be sad to lose.

esplinr (Tue, 14 Apr 2020 14:46:51 GMT):
But if moving to GitHub will assist new contributors, I won't oppose it.

esplinr (Tue, 14 Apr 2020 14:48:10 GMT):
@sergey.minaev can help give maintainer privileges to those who need them.

swcurran (Tue, 14 Apr 2020 15:32:00 GMT):
@sergey.minaev - please add @andrew.whitehead at least to the maintainers. Not sure who else, but I think Andrew should have it.

cam-parra (Tue, 14 Apr 2020 15:33:47 GMT):
Agreed. and if @esplinr could have alex comment quickly on any PRs that he might be concerned about or that should just not be merged

swcurran (Tue, 14 Apr 2020 15:36:00 GMT):
The information in the wiki will not be lost and we can direct people to it that need it. But definitely having it in github with the code makes it for more accessible to other contributors. Does anyone else need to be involved, other than the contributors?

swcurran (Tue, 14 Apr 2020 15:36:00 GMT):
The information in the wiki will not be lost and we can direct people to it that need it. But definitely having it in github with the code makes it for more accessible to other contributors. Does anyone else need to be involved, other than the maintainers? Anyone at HL? Ry?

andrew.whitehead (Tue, 14 Apr 2020 15:36:54 GMT):
I can review PRs but I'm not sure about merging them if I don't know what the plan is for releases

cam-parra (Tue, 14 Apr 2020 15:38:12 GMT):
I am more concerned about these updates that won't be merged soon like https://github.com/hyperledger/indy-plenum/pull/1481

cam-parra (Tue, 14 Apr 2020 15:38:48 GMT):
and https://github.com/hyperledger/indy-plenum/pull/1480

andrew.whitehead (Tue, 14 Apr 2020 15:43:20 GMT):
Looks like I have a big merge button

andrew.whitehead (Tue, 14 Apr 2020 16:00:08 GMT):
Merged 1480 because it shouldn't cause any issues apart from node needing the same orderedset dependency. I'll look into running tests on 1481

SethiSaab (Wed, 15 Apr 2020 08:45:15 GMT):
Is there anyone who has done KeyCloak SSo integration with DIDs

SethiSaab (Wed, 15 Apr 2020 08:45:15 GMT):
Is there anyone who has done KeyCloak SSo integration with DIDs ?

ayushmanss (Wed, 15 Apr 2020 09:29:32 GMT):
do diddocs resides on the ledger?

PatrikStas (Wed, 15 Apr 2020 12:28:24 GMT):
@ayushmanss no, but diddoc can be constructed from data which reside on the ledger (sequence of nym and attrib transactions)

iamShashvat (Wed, 15 Apr 2020 12:35:35 GMT):
Earlier I asked the question about signature mismatch while submitting the request. I found out that the signature is invalid because of reqID in the response of BuildAttribRequest. Basically, I hardcoded the request and just change the reqID value with previous transaction's reqID and submit it. It's a successful transaction. How is this possible? whenever we build a request for NYM/ATTRIB etc we get reqID(Unique ID number of the request with transaction).

iamShashvat (Wed, 15 Apr 2020 12:38:24 GMT):
Earlier I asked the question about signature mismatch while submitting the request. I found out that the signature is invalid because of reqID in the response of BuildAttribRequest. whenever we build a request for NYM/ATTRIB etc we get reqID(Unique ID number of the request with the transaction) with same reqID I can do multiple transactions. Basically, I hardcoded the request and just change the reqID value with the previous transaction's reqID and submit it. It's a successful transaction. How is this possible?

iamShashvat (Wed, 15 Apr 2020 19:08:29 GMT):
Suggestion: In NodeJS wrapper there should be bigInt handling for reqID(transaction), currently it makes the last two digits zero. example: Generated by library: 1586764515258869234, In NodeJS: 1586764515258869200. It creates a problem when your server in node js and your client in some other language. possible problem: Signature Mismatch! (I faced it)

YaswanthSorapalli (Thu, 16 Apr 2020 11:25:06 GMT):
Can anyone tell me that where the credentials uploaded in server (which was hosted somewhere i.e. aws). Custom blockchain was deployed on that aws server. All the credentials uploaded by agents will be stored in server or not? If stored in which format (encoded or not)?

andrew.whitehead (Thu, 16 Apr 2020 16:21:42 GMT):
Looks like my review doesn't count as an approving one

cam-parra (Thu, 16 Apr 2020 18:26:42 GMT):
Is this on plenum?

andrew.whitehead (Thu, 16 Apr 2020 18:26:56 GMT):
Yep

cam-parra (Thu, 16 Apr 2020 18:30:45 GMT):
to me it shows as approved

cam-parra (Thu, 16 Apr 2020 18:30:47 GMT):
https://github.com/hyperledger/indy-plenum/pull/1481

cam-parra (Thu, 16 Apr 2020 18:30:53 GMT):
this can be merged

andrew.whitehead (Thu, 16 Apr 2020 18:31:34 GMT):

Screen Shot 2020-04-16 at 11.31.17.png

cam-parra (Thu, 16 Apr 2020 18:34:16 GMT):
I wonder if it's because you're not on the code owners

cam-parra (Thu, 16 Apr 2020 18:34:39 GMT):
@rjones maybe you can chime in

swcurran (Thu, 16 Apr 2020 19:28:08 GMT):
@danielhardman @domwoe - I tried to add this to the thread we had going, but got an error. To continue the thread from the second document that Daniel posted: I don't think the document answers the original question. The only statement in the second document about non-repudiation is this assertion `NIZK proofs are considered non-repudiable, because the verifier can share the proof in a convincing way; sharing the mathematical construct provided by the prover is enough` Why is that assertion true about NIZK Proofs? What do I have to save to be able to show a non-repudiable statement from the prover? In practical terms, what do I have to show in court? Thanks

danielhardman (Thu, 16 Apr 2020 21:14:47 GMT):
@swcurran and @domwoe: I've tried yet again. See if this explanation (entirely from scratch) makes more sense: https://docs.google.com/document/d/1t9JfZOdvapbby83i_zZlLziw7FkB9bLIXwvYRaO5lRk/edit#heading=h.ri90qdc4riam

hynese (Fri, 17 Apr 2020 09:30:31 GMT):
Has joined the channel.

championshuttler (Fri, 17 Apr 2020 20:40:04 GMT):
Has joined the channel.

karburator (Fri, 17 Apr 2020 22:30:54 GMT):
Has joined the channel.

jljordan_bcgov (Sat, 18 Apr 2020 19:23:36 GMT):
Can the Jira issue become some sort of repository ?

jljordan_bcgov (Sat, 18 Apr 2020 19:23:36 GMT):
Can the Jira issue become some sort of archived repository ?

jljordan_bcgov (Sat, 18 Apr 2020 19:26:03 GMT):
Interesting thread on that ... https://stackoverflow.com/questions/31125655/is-there-a-way-to-import-jira-issues-to-github

jljordan_bcgov (Sat, 18 Apr 2020 19:26:46 GMT):
and maybe a tool within that SO post ... https://github.com/susinda/github-client

daveek (Sat, 18 Apr 2020 20:33:29 GMT):
Hi, Can anyone please explain or share some docs about Indy/Aries Threat Model? a) What's the Threat Model for Hyperledger Indy/Aries? b) Where is Indy/Aries most vulnerable to attack? c) What are the most relevant threats? d) What do I need to do to safeguard against these threats?

NewAlexandria (Sun, 19 Apr 2020 01:32:39 GMT):
Has joined the channel.

swcurran (Sun, 19 Apr 2020 21:33:54 GMT):
The Indy Contributors call is tomorrow morning (US) at 7AM Pacific, 8AM Mountain. We want to talk about migrations to GitHub Issues and GitHub Actions, along with Revocation 2.0 and status on Indy-SDK breakup efforts. Great if some folks from DSR could join us ( @sergey.minaev and others), @cam-parra, @WadeBarnes and @andrew.whitehead --- and anyone else that is interested. Wiki Page: https://wiki.hyperledger.org/display/indy/2020-04-20+Indy+Contributors+Call Zoom: https://zoom.us/j/244779296

swcurran (Sun, 19 Apr 2020 21:33:54 GMT):
The Indy Contributors call is tomorrow morning (US) at ~7~ 8AM Pacific, ~8~ 9AM Mountain. We want to talk about migrations to GitHub Issues and GitHub Actions, along with Revocation 2.0 and status on Indy-SDK breakup efforts. Great if some folks from DSR could join us ( @sergey.minaev and others), @cam-parra, @WadeBarnes and @andrew.whitehead --- and anyone else that is interested. Wiki Page: https://wiki.hyperledger.org/display/indy/2020-04-20+Indy+Contributors+Call Zoom: https://zoom.us/j/244779296 Note the corrected time for the call above.

swcurran (Sun, 19 Apr 2020 21:33:54 GMT):
The Indy Contributors call is tomorrow morning (US) at ~7~ 8AM Pacific, ~8~ 9AM Mountain. We want to talk about migrations to GitHub Issues and GitHub Actions, along with Revocation 2.0 and status on Indy-SDK breakup efforts. Great if some folks from DSR could join us ( @sergey.minaev and others), @cam-parra, @WadeBarnes and @andrew.whitehead --- and anyone else that is interested. Wiki Page: https://wiki.hyperledger.org/display/indy/2020-04-20+Indy+Contributors+Call Zoom: https://zoom.us/j/244779296 Note the corrected time for the call.

andrew.whitehead (Sun, 19 Apr 2020 21:36:34 GMT):
It's usually 8am :thinking:

DhirajSharma8 (Mon, 20 Apr 2020 05:23:44 GMT):
Has joined the channel.

DhirajSharma8 (Mon, 20 Apr 2020 05:23:44 GMT):
For mentorship i am interested in this project https://wiki.hyperledger.org/pages/viewpage.action?pageId=31195277 I have applied on community bridge website using my gmail id dhiraj.8.sharma@gmail.com .May i know what all next steps I need to do?

RicardoPeixoto (Mon, 20 Apr 2020 14:42:36 GMT):
Has joined the channel.

RicardoPeixoto (Mon, 20 Apr 2020 14:57:51 GMT):
Hi, when i was looking into Indy Node i had a doubt. If only trustees, stewards or trust anchors can add a new Identity Owner, how does a new user establish a peer-to-peer connection given that his DID is not yet in the ledger?

swcurran (Mon, 20 Apr 2020 16:10:43 GMT):
Aries agents do that using Peer DIDs - a DID method that is exchanged between participants, but not stored on the ledger. I did a video of that awhile back that explains it: https://www.youtube.com/watch?v=EKH7vjp_P9o

RicardoPeixoto (Mon, 20 Apr 2020 17:04:44 GMT):
@swcurran thank you for your response. From the video i got one more question: 1- How to send the DID Doc safely to the other entity? More specifically, how to prevent someone from pretending to be Bob or even man-in-the-middle attacks?

swcurran (Mon, 20 Apr 2020 17:13:17 GMT):
The first message is sent plaintext, the rest are encrypted, so it is the first message that is at risk. The approach that must be taken in that scenario is that you assume you don't know who you are talking to once the connection is established. You have a new, secure channel to talk to someone. After that, you need to find out who you are talking to via verifiable credentials. In other words, when there is a plaintext channel used to establish the connection, find out after the connection has been made to whom you are talking.

swcurran (Mon, 20 Apr 2020 17:18:08 GMT):
We had a good Indy Contributors call today where we talked about: - updates on the latest progress on the various sub-projects, including indy-vdr, indy-shared-rs, deprecating indy-crypto and Aries secure storage - migration from Jira to Github Issues (we will, with lazy copying of open issues from Jira), - migration from Jenkins to GitHub Actions (we're getting started...) - initial discussions about the design for handling revocation 2.0 merkle trees on the ledger The recording for the call and tracking of the discussion/decisions can be found here: https://wiki.hyperledger.org/display/indy/2020-04-20+Indy+Contributors+Call

swcurran (Mon, 20 Apr 2020 17:20:08 GMT):
@rjones - per the decisions made in the call today - can you please do some updates on GitHub: - activate "issues" for indy-sdk, indy-node and indy-plenum - add Andrew Whitehead as a maintainer on indy-plenum Thanks!

rjones (Mon, 20 Apr 2020 17:20:08 GMT):
Has joined the channel.

rjones (Mon, 20 Apr 2020 18:54:56 GMT):
@swcurran done!

swcurran (Mon, 20 Apr 2020 18:55:55 GMT):
Awesome - thanks!

tplooker (Tue, 21 Apr 2020 02:44:55 GMT):
Hi all for those who are not connected over at DIF where I placed a similar message in the claims and credentials working group, we (Mattr) will be demoing a new digital signature approach to ZKP capable verifiable credentials at IIW that have the following properties. - Do not require credential definitions so hence are ledger agnostic i.e only require a did on a ledger - Assertion format agnostic (i.e you could render them as a JWS or linked data proofs) we will be showing how to use with linked data proofs (i.e JSON-LD) - More performant due to the underlying crypto used (ECC based rather than RSA) - Smaller in size also due to the change in crypto primitive used e.g (ECC based rather than RSA) - Verifiable credential spec compliant Following this we will be open sourcing a reference implementation along side a specification for the scheme

Elaina (Tue, 21 Apr 2020 08:09:05 GMT):
Has joined the channel.

Elaina (Tue, 21 Apr 2020 08:09:06 GMT):
Hello everyone! I had a couple of questions to inquire the team so if anyone can help out, that would be highly appreciated 🙂 1. Request for information/documentations on Pluggable Crypto Service - Is there a set schedule for the release of the service? - If possible, please do share with us more details (interface, architecture, etc.) on the pluggable crypto service 2. Request for information on Aries-KMS - Please kindly share with us the release schedule for Aries-KMS - Please kindly share with us the architecture of Aries-KMS (documents for developers to review please)

Elaina (Tue, 21 Apr 2020 08:09:06 GMT):
Hello everyone! I had a couple of questions to inquire the team so if anyone can help out, that would be highly appreciated 🙂 1. Request for information/documentations on Pluggable Crypto Service - Is there a set schedule for the release of the service? - If possible, please do share with us more details (interface, architecture, etc.) on the pluggable crypto service 2. Request for information on Aries-KMS - Please kindly share with us the release schedule for Aries-KMS - Please kindly share with us the architecture of Aries-KMS (documents for developers to review please) Reference : https://github.com/mac-arrap/aries-rfcs/tree/master/concepts/0276-key-management-service

Elaina (Tue, 21 Apr 2020 08:09:06 GMT):
Hello everyone! I had a couple of questions to inquire the team so if anyone can help out, that would be highly appreciated 🙂 1. Request for information/documentations on Pluggable Crypto Service - Is there a set schedule for the release of the service? - If possible, please do share with us more details (interface, architecture, etc.) on the pluggable crypto service 2. Request for information on Aries-KMS - Please kindly share with us the release schedule for Aries-KMS - Please kindly share with us the architecture of Aries-KMS (documents for developers to review please) Reference : - https://github.com/mac-arrap/aries-rfcs/tree/master/concepts/0276-key-management-service - https://wiki.hyperledger.org/display/indy/2020-01-13+Indy+Contributors+Call

kh_touati (Tue, 21 Apr 2020 12:16:38 GMT):
Has joined the channel.

Hanu7743 (Tue, 21 Apr 2020 14:59:39 GMT):
Has joined the channel.

Hanu7743 (Tue, 21 Apr 2020 15:08:16 GMT):
Hi All, how to start the indy with java?please can you guide me

GKumar (Tue, 21 Apr 2020 15:16:43 GMT):
how can i customize did schema part ex:did:*sov*:BzCbsNYhMrjHiqZDTUASHg;spec to did:*xyz*:BzCbsNYhMrjHiqZDTUASHg;spec

GKumar (Tue, 21 Apr 2020 15:16:43 GMT):
how can i customize did schema part ex:did:*sov*:BzCbsNYhMrjHiqZDTUASHg;spec to did:*xyz*:BzCbsNYhMrjHiqZDTUASHg;spec

GKumar (Tue, 21 Apr 2020 15:19:02 GMT):
schema

tomislav (Tue, 21 Apr 2020 15:25:14 GMT):
you can pass `method_name` parameter when creating a did. You can also unqualify and qualify a did using the same parameters.

GKumar (Tue, 21 Apr 2020 15:27:32 GMT):
tnx

GKumar (Tue, 21 Apr 2020 15:29:23 GMT):
So it mean schema could be multiple name for a given trust-anchor even though ledger is same

swcurran (Tue, 21 Apr 2020 16:17:13 GMT):
The short answer is that you can't do that arbitrarily in an Aries agent unless you have full support for the DID method "xyz" (whatever that is). You need to understand the DID Spec. and DID Methods to do that. I'm wondering about what you are trying to do -- why you want to do that. My guesses: 1. You want to use a different DID method. For that, you have to write an Aries interface to that ledger -- a process that is not well understood. You would be pioneering. 2. You want to reference a different instance of an Indy network other than Sovrin. That is currently only possible at start times of agents, when you load the genesis txns of the ledger to which you want to connect. This is something we need, but not currently standardized. Am I right with my guesses? Something else?

GKumar (Tue, 21 Apr 2020 16:21:10 GMT):
Hi @swcurran Thanks a lot for detailed answer. I was just playing around indy and aries. I completed the deployment and connections. Finally I thought of why don't we start something with my own *method *name. :pray:

swcurran (Tue, 21 Apr 2020 17:51:10 GMT):
No!!!! :-)

swcurran (Tue, 21 Apr 2020 17:51:18 GMT):
Not the thing you want to do.

Shubham-koli (Wed, 22 Apr 2020 06:33:34 GMT):
Hi Good Afternoon Guys. I was playing with *Aries OpenAPI Demo*. I ran into an error while starting the agent for Faber. ``` [node1] (local) root@192.168.0.48 ~/aries-cloudagent-python/demo $ LEDGER_URL=http://dev.greenlight.bcovrin.vonx.io ./run_demo faber --events --no-auto --bg Running in faber in the background. Note that you cannot use the command line console in this mode. To see the logs use: "docker logs faber". While viewing logs, hit CTRL-C to return to the command line. To stop the agent, use: "docker stop faber". The docker environment will be removed on stop. Preparing agent image... sha256:528ea823655c8b1cb230b978988be3c0561a233e1cb4f12ec174c5b91900bda1 ip172-18-0-75-bqfsrvlim9m000fdt4vg-{PORT}.direct.labs.play-with-docker.com Starting faber... 04c7d0221815cc6891179f076441f2cd29999021df46bbcf388bb0602a0b14ed [node1] (local) root@192.168.0.48 ~/aries-cloudagent-python/demo $ docker logs -f faber #1 Provision an agent and wallet, get back configuration details Faber | Registering Faber.Agent with seed d_000000000000000000000000671507 Traceback (most recent call last): File "/home/indy/.pyenv/versions/3.6.9/lib/python3.6/runpy.py", line 193, in _run_module_as_main "__main__", mod_spec) File "/home/indy/.pyenv/versions/3.6.9/lib/python3.6/runpy.py", line 85, in _run_code exec(code, run_globals) File "/home/indy/demo/runners/faber.py", line 440, in main(args.port, args.no_auto, args.revocation, args.timing) File "/home/indy/.pyenv/versions/3.6.9/lib/python3.6/asyncio/base_events.py", line 484, in run_until_complete return future.result() File "/home/indy/demo/runners/faber.py", line 161, in main await agent.register_did() File "/home/indy/demo/runners/support/agent.py", line 348, in register_did raise Exception(f"Error registering DID, response code {resp.status}") Exception: Error registering DID, response code 500 ``` I tried this solution on a similar issue but it did not work. *https://github.com/hyperledger/aries-cloudagent-python/issues/139#issuecomment-521341580*

Shubham-koli (Wed, 22 Apr 2020 06:33:34 GMT):
Hi Good Afternoon Guys. I was playing with *Aries OpenAPI Demo*. I ran into an error while starting the agent for Faber. ``` [node1] (local) root@192.168.0.48 ~/aries-cloudagent-python/demo $ LEDGER_URL=http://dev.greenlight.bcovrin.vonx.io ./run_demo faber --events --no-auto --bg Running in faber in the background. Note that you cannot use the command line console in this mode. To see the logs use: "docker logs faber". While viewing logs, hit CTRL-C to return to the command line. To stop the agent, use: "docker stop faber". The docker environment will be removed on stop. Preparing agent image... sha256:528ea823655c8b1cb230b978988be3c0561a233e1cb4f12ec174c5b91900bda1 ip172-18-0-75-bqfsrvlim9m000fdt4vg-{PORT}.direct.labs.play-with-docker.com Starting faber... 04c7d0221815cc6891179f076441f2cd29999021df46bbcf388bb0602a0b14ed [node1] (local) root@192.168.0.48 ~/aries-cloudagent-python/demo $ docker logs -f faber #1 Provision an agent and wallet, get back configuration details Faber | Registering Faber.Agent with seed d_000000000000000000000000671507 Traceback (most recent call last): File "/home/indy/.pyenv/versions/3.6.9/lib/python3.6/runpy.py", line 193, in _run_module_as_main "__main__", mod_spec) File "/home/indy/.pyenv/versions/3.6.9/lib/python3.6/runpy.py", line 85, in _run_code exec(code, run_globals) File "/home/indy/demo/runners/faber.py", line 440, in main(args.port, args.no_auto, args.revocation, args.timing) File "/home/indy/.pyenv/versions/3.6.9/lib/python3.6/asyncio/base_events.py", line 484, in run_until_complete return future.result() File "/home/indy/demo/runners/faber.py", line 161, in main await agent.register_did() File "/home/indy/demo/runners/support/agent.py", line 348, in register_did raise Exception(f"Error registering DID, response code {resp.status}") Exception: Error registering DID, response code 500 ``` I tried this solution on a similar issue but it did not work. *https://github.com/hyperledger/aries-cloudagent-python/issues/139#issuecomment-521341580* I'm using let's play with docker for experimenting.

Shubham-koli (Wed, 22 Apr 2020 06:35:57 GMT):
any help would be greatly appreciated.

SethiSaab (Wed, 22 Apr 2020 06:36:47 GMT):
Hi Team , I want to integrate Cloud HSM with Indy for key management . Is there anyone who has done that before ?

SethiSaab (Wed, 22 Apr 2020 06:38:19 GMT):
or if anyone can provide some reference

swcurran (Wed, 22 Apr 2020 14:41:05 GMT):
Please ask this on ACA-Py and we'll get it answered.

swcurran (Wed, 22 Apr 2020 14:55:05 GMT):
I think the greenlight ledger is having trouble.

rjones (Wed, 22 Apr 2020 15:05:40 GMT):
Has left the channel.

domwoe (Wed, 22 Apr 2020 15:45:37 GMT):
Given Rich Schemas, would it be possible to define an object or even an array of objects in a schema?

brentzundel (Wed, 22 Apr 2020 17:53:41 GMT):
yes to both questions. In order to sign the array an issuer will need to specify a maximum number of elements.

domwoe (Wed, 22 Apr 2020 17:58:50 GMT):
Thanks a lot!

brentzundel (Wed, 22 Apr 2020 18:14:11 GMT):
:thumbsup:

mccown (Wed, 22 Apr 2020 20:50:05 GMT):
Implementer’s Call Invite Just a reminder about the Identity Implementer's WG Call tomorrow morning (April 23th @ 9am MT). This is a chance to get updates on the WG calls you may have missed this week. Our guest will be Pico Labs (@sanbachs from BYU) who will present their research into Verifiable Credentials. They will be giving a system demonstration for how they can be used to conduct commerce. Here is the calendar invite with times and Zoom details. https://tinyurl.com/y7ntqd4n

Shubham-koli (Thu, 23 Apr 2020 05:59:59 GMT):
@swcurran thank you very much! I will do that!

gnom1gnom (Thu, 23 Apr 2020 08:04:52 GMT):
Hi All, I was wondering what would be the consequences if in a SSI network, only DIDs of credential issuers were stored in the ledger, while personal DIDs would remain only in wallets of individuals. Relationships are established using pairwise DIDs anyway, so I think the only drawback would be that a private DID owner would not be able to designate a service endpoint, thus any interaction with a persona would require that wallet owner is on-line. Am I missing something? Can you please explain or point me to a place that elaborates in detail about this topic?

theblockstalk (Thu, 23 Apr 2020 13:26:42 GMT):
Has joined the channel.

swcurran (Thu, 23 Apr 2020 17:19:42 GMT):
That is reasonably correct. However, an owner can have a mediator agent that can queue messages for when the wallet owner is online. Those messages are encrypted by the sender, so the mediator agent just knows to whom they are going. This is necessary for all mobile agents, and can be used by any edge agents. See the RFC for Mediators and Agents here: https://github.com/hyperledger/aries-rfcs/tree/master/concepts/0046-mediators-and-relays There are several related RFCs about queuing, routing, etc.

swcurran (Thu, 23 Apr 2020 21:10:53 GMT):
Hey folks - interested in contributing to Indy Revocation 2? We have posted a couple of Help Wanted issues for tech spikes into our revocation plans. These are for one or two standalone programs (any language!) to test out the ideas we have for revocation - think of them as assignments you would get in a Computer Science class on algorithms and data structures. If you can spare a 1/2 to 1 day to help out, that would probably do it! Issues are here: https://github.com/hyperledger/indy-node/issues, with the ones I'm talking about tagged with "Help Wanted". Questions welcome, contributions even more so!

swcurran (Thu, 23 Apr 2020 21:34:16 GMT):
I think the summary of the document is that a ZKP (and likely any) proof is non-repudiable if the claims within it are sufficient to be trusted by a third party. The verifier passes the proof to the third party, the third party verifies it and they see both enough information in the claims and they trust the issuers of those claims. So it is not that the ZKP is non-repudiable it's the content of the proof. As you say, nothing magical about the ZKP at all other than that it makes the claims within authoritative from the issuers through to the 3rd party. That makes sense. It has the ramification that a verifier wanting non-repudiation has to collect more information from the prover, and must preserve that information. Thanks. BTW - I've seen another product recently that aligns really nicely with the LOA 3 that you describe in the doc. It would be a great generic addition to wallets. I'll share more about it one of these days...

danielhardman (Fri, 24 Apr 2020 01:31:40 GMT):
> think the summary of the document is that a ZKP (and likely any) proof is non-repudiable if the claims within it are sufficient to be trusted by a third party. I'd summarize it almost this way, but with this nuance: it's not just general strength of claims. What makes a proof non-repudiable is whether the claims *bind the proof to the prover* strongly enough that the prover can't deny having made it. With interactive ZKPs, this is never true, because the verifier of the proof could game the binding. With non-interactive ones, this problem doesn't exist.

GKumar (Fri, 24 Apr 2020 09:44:41 GMT):
I am using following command to run agent, But i could not find any db artifacts created in my pgsql admin console. Can some one help me to see the pgsql-agent connections really happening

GKumar (Fri, 24 Apr 2020 09:44:41 GMT):
I am using following command to run agent, But i could not find any db artifacts created in my pgsql admin console. Can some one help me to see the pgsql-agent connections really happening *GENESIS_FILE=./indy_genensis POSTGRES=true DEFAULT_POSTGRES=true sudo python -m runners.faber --port 8020*

swcurran (Sat, 25 Apr 2020 20:53:48 GMT):
For those going to IIW, I went to the orientation session today and it was worth the time. The platform is an IIW-style Open Space running on top of Zoom. If you have been to IIW and know Zoom it will make sense, but worth going to the orientation. The platform is pretty cool!

iamShashvat (Mon, 27 Apr 2020 10:11:27 GMT):
Is it possible to create DID and attrib functionalities from external created ED25199 key pair?

iamShashvat (Mon, 27 Apr 2020 10:11:27 GMT):
Is it possible to create DID and do attrib functionalities from externally created(not from indy-sdk) ED25519 key pair?

adamjlemmon (Mon, 27 Apr 2020 10:41:35 GMT):
Has joined the channel.

pranav_kirtani (Mon, 27 Apr 2020 12:09:06 GMT):
Can I create a custom DID (did:myorg:unique_id) via Indy? any wrappers that allow me to do it?

Siva_Kannan (Mon, 27 Apr 2020 14:47:22 GMT):
Has joined the channel.

pranav_kirtani (Tue, 28 Apr 2020 09:05:29 GMT):
how do I add a node to indy pool running as docker containers?

lynn.bendixsen (Tue, 28 Apr 2020 14:52:18 GMT):
It should be the same process as for adding a new node to any network. Similar to the instructions found here: https://docs.google.com/document/d/18MNB7nEKerlcyZKof5AvGMy0GP9T82c4SWaxZkPzya4/edit#heading=h.9a4ud1gh7a6x Was there a specific issue or problem that you were running into that we can help with? The new node does not need to become part of the genesis file.

arner (Tue, 28 Apr 2020 20:16:50 GMT):
Hi, when I create keys on Sovrin using the universal registrar, it returns the private and public key in Base58 format. Any guidance on how to use these in NodeJS? The crypto.sign function expects them in pem or der format... Or would it be better to use something like tweetnacl?

arner (Tue, 28 Apr 2020 20:16:50 GMT):
Hi, when I create keys on Sovrin using the universal registrar, it returns the private and public key (ED25519) in Base58 format. Any guidance on how to use these in NodeJS? The crypto.sign function expects them in pem or der format... Or would it be better to use something like tweetnacl?

SUSHOBHAN (Wed, 29 Apr 2020 06:26:21 GMT):
Has joined the channel.

jcourt (Wed, 29 Apr 2020 09:05:49 GMT):
Has joined the channel.

jcourt (Wed, 29 Apr 2020 09:09:22 GMT):
I have been trying to follow the SelfServe instructions at https://docs.google.com/document/d/1sXZoN18lpFoAF075QoptofwDV_1otUylPGFKRQnA56E/edit so that I can issue some builder net tokens but I come unstuck just after creating the DID using indy-cli. The step completes but when I try and verify it is on the ledger I get "NYM not found"

jcourt (Wed, 29 Apr 2020 09:09:22 GMT):
I have been trying to follow the SelfServe instructions at https://docs.google.com/document/d/1sXZoN18lpFoAF075QoptofwDV_1otUylPGFKRQnA56E/edit so that I can issue some builder net tokens but I come unstuck just after creating the DID using indy-cli. The step completes but when I try and verify it is on the ledger I get "NYM not found" pool(buildernet):jctest1:indy> did list +------------------------+-------------------------+----------+ | Did | Verkey | Metadata | +------------------------+-------------------------+----------+ | 6B5fcN8D1DPqQ7sWzazAaC | ~K9aioHqZQ9mZvrHSMuVZm3 | - | +------------------------+-------------------------+----------+ pool(buildernet):jctest1:indy> ledger get-nym did=6B5fcN8D1DPqQ7sWzazAaC NYM not found

jcourt (Wed, 29 Apr 2020 09:17:52 GMT):
Is there some way of verifying the indy-cli is able to talk to the ledger ? I thought about using "ledger get-fees payment_method=sov" since I figure that will need to talk to the ledger but that is returning Unknown payment method.

jcourt (Wed, 29 Apr 2020 09:17:52 GMT):
Is there some way of verifying the indy-cli is able to talk to the ledger ? I thought about using "ledger get-fees payment_method=sov" since I figure that will need to talk to the ledger but that is returning Unknown payment method. pool(buildernet):jctest1:indy> ledger get-fees payment_method=sov Unknown payment method sov pool(buildernet):jctest1:indy>

jcourt (Wed, 29 Apr 2020 09:22:03 GMT):
The little bit that niggles me is there is a step that talks about building libsovtoken but I can't find anywhere in the indy-sdk where that lives anymore ? So I am suspicious that the payment method fails because that library is missing but it still doesn't explain why indy-cli has connected to a pool happily and indicated it created a did yet I can't fetch it and there are no errors.

jcourt (Wed, 29 Apr 2020 11:43:55 GMT):
So I was able to verify that I could perform a ledger get-nym on someone else's existing DID so it must be something to do with the "did new" that I am not understanding rather than a pool connection issue.

jcourt (Wed, 29 Apr 2020 11:43:55 GMT):
And I think the answer is I am mis-interpreting step V.B of the self serve instructions. Clearly the DID isn't written at that point to the ledger.

jcourt (Wed, 29 Apr 2020 11:53:00 GMT):
And once I found where I could get a libsovtoken to build its all happy again :-). Sorry for the fire drill

lynn.bendixsen (Wed, 29 Apr 2020 14:00:03 GMT):
Yes, the DID is only in your wallet at that point. In fact, that is where most DID's in the world will stay. DID's rarely need to be written to the ledger under the Indy way of thinking. In most use cases, the only DIDs that have to be on the ledger are those that are used to write to the ledger.

iamShashvat (Wed, 29 Apr 2020 17:16:16 GMT):
Is it possible to import externally created ED25519 keys in our wallet(indy-sdk)?

iamShashvat (Wed, 29 Apr 2020 18:10:29 GMT):
@andrew.whitehead @lynn.bendixsen @swcurran?

ayushmanss (Wed, 29 Apr 2020 22:19:08 GMT):
Hi all, I have run into a blocker and been banging my head since. REQUIREMENT - I want to run postgres with the aries-dotnet framework SOLUTION - I figured it can be the same as it is done with python-cloudagent. Basically load the library before calling any wallet functions and indy-sdk will take care of the rest PROBLEM - I didnt find any dll for indystrgpostgres and decided to build from source. I followed the steps exactly for building libindy for windows and the image build out without any errors but I get "Unhandled exception. System.BadImageFormatException: Bad IL format." when I try to load the library using "AssemblyLoadContext.Default.LoadFromAssemblyPath". REASON - I want to run mediator for large number of users and hence need HA and thus need a better common wallet storage for multiple instances of dotnet mediator containers running Any solution to the problem will be really helpful

esplinr (Thu, 30 Apr 2020 03:08:01 GMT):
Both of those ideas are slowly evolving. There aren't many documents and there is no release schedule yet.

davidd (Thu, 30 Apr 2020 15:34:08 GMT):
I cloned the von-network

davidd (Thu, 30 Apr 2020 15:34:19 GMT):
``` ~/von-network$ docker-compose up WARNING: The IP variable is not set. Defaulting to a blank string. WARNING: The IPS variable is not set. Defaulting to a blank string. WARNING: The DOCKERHOST variable is not set. Defaulting to a blank string. WARNING: The LOG_LEVEL variable is not set. Defaulting to a blank string. WARNING: The RUST_LOG variable is not set. Defaulting to a blank string. WARNING: The GENESIS_URL variable is not set. Defaulting to a blank string. WARNING: The ANONYMOUS variable is not set. Defaulting to a blank string. WARNING: The LEDGER_SEED variable is not set. Defaulting to a blank string. WARNING: The LEDGER_CACHE_PATH variable is not set. Defaulting to a blank string. WARNING: The WEB_ANALYTICS_SCRIPT variable is not set. Defaulting to a blank string. WARNING: The INFO_SITE_TEXT variable is not set. Defaulting to a blank string. WARNING: The INFO_SITE_URL variable is not set. Defaulting to a blank string. Creating network "vonnetwork_von" with the default driver Creating vonnetwork_webserver_1 ... Creating vonnetwork_synctest_1 ... Creating vonnetwork_nodes_1 ... Creating vonnetwork_client_1 ... Creating vonnetwork_node3_1 ... Creating vonnetwork_node4_1 ... Creating vonnetwork_webserver_1 Creating vonnetwork_nodes_1 Creating vonnetwork_synctest_1 Creating vonnetwork_node2_1 ... Creating vonnetwork_node1_1 ... Creating vonnetwork_client_1 Creating vonnetwork_node4_1 Creating vonnetwork_node3_1 Creating vonnetwork_node2_1 Creating vonnetwork_node2_1 ... error ERROR: for vonnetwork_node2_1 Cannot start service node2: driver failed programming external connectivity on endpoint vonnetwork_node2_1 (1a5ec19261603273cff841b6d4ff3fba222dc1d13f578c444dc9ae20741af975): Bind for 0.0.0.0:9704 failed: portCreating vonnetwork_synctest_1 ... error ERROR: for vonnetwork_synctest_1 Cannot start service synctest: driver failed programming external connectivity on endpoint vonnetwork_synctest_1 (c65be5409adbd2eb90eb6194655511ac058996411f55d495cab5b6a132708a54): Bind for 0.0.0.0:9000 faiCreating vonnetwork_nodes_1 ... error ERROR: for vonnetwork_nodes_1 Cannot start service nodes: driver failed programming external connectivity on endpoint vonnetwork_nodes_1 (2559934776d34e94fa296d6ecd9c5264c9b995a73f2158b3af6c07a0e8112303): Bind for 0.0.0.0:9702 failed: portCreating vonnetwork_node3_1 ... done ERROR: for nodes Cannot start service nodes: driver failed programming external connectivity on endpoint vonnetwork_nodes_1 (2559934776d34e94fa296d6ecd9c5264c9b995a73f2158b3af6c07a0e8112303): Bind for 0.0.0.0:9702 failed: port is already allocated ERROR: for node2 Cannot start service node2: driver failed programming external connectivity on endpoint vonnetwork_node2_1 (1a5ec19261603273cff841b6d4ff3fba222dc1d13f578c444dc9ae20741af975): Bind for 0.0.0.0:9704 failed: port is already allocated ERROR: for synctest Cannot start service synctest: driver failed programming external connectivity on endpoint vonnetwork_synctest_1 (c65be5409adbd2eb90eb6194655511ac058996411f55d495cab5b6a132708a54): Bind for 0.0.0.0:9000 failed: port is already allocated ERROR: Encountered errors while bringing up the project. ```

davidd (Thu, 30 Apr 2020 15:40:30 GMT):
```~/von-network$ GENESIS_URL=https://raw.githubusercontent.com/sovrin-foundation/sovrin/master/sovrin/pool_transactions_sandbox_genesis LEDGER_SEED=000000000000000000000IanCostanzo PORT=9000 python -m server.server Traceback (most recent call last): File "/usr/lib/python3.6/runpy.py", line 193, in _run_module_as_main "__main__", mod_spec) File "/usr/lib/python3.6/runpy.py", line 85, in _run_code exec(code, run_globals) File "/home/mujeeb/von-network/server/server.py", line 14, in from .anchor import ( File "/home/mujeeb/von-network/server/anchor.py", line 17, in from indy import ledger, pool ModuleNotFoundError: No module named 'indy'```

davidd (Thu, 30 Apr 2020 15:40:30 GMT):
```~/von-network$ GENESIS_URL=https://raw.githubusercontent.com/sovrin-foundation/sovrin/master/sovrin/pool_transactions_sandbox_genesis LEDGER_SEED=000000000000000000000IanCostanzo PORT=9000 python -m server.server Traceback (most recent call last): File "/usr/lib/python3.6/runpy.py", line 193, in _run_module_as_main "__main__", mod_spec) File "/usr/lib/python3.6/runpy.py", line 85, in _run_code exec(code, run_globals) File "/home/ubuntu/von-network/server/server.py", line 14, in from .anchor import ( File "/home/ubuntu/von-network/server/anchor.py", line 17, in from indy import ledger, pool ModuleNotFoundError: No module named 'indy'```

andrew.whitehead (Thu, 30 Apr 2020 15:46:33 GMT):
Looks like you have other docker containers running on those ports

davidd (Thu, 30 Apr 2020 15:49:22 GMT):
Stopped the all docker containers, Restart the docker too.

davidd (Thu, 30 Apr 2020 15:49:37 GMT):
Try everything, still getting the same error

davidd (Thu, 30 Apr 2020 15:50:07 GMT):
and any idea why `No module name 'indy'` error

andrew.whitehead (Thu, 30 Apr 2020 15:51:54 GMT):
Because you don't have the requirements installed, particularly the python3-indy module but you would need indy-sdk for that to work. It's designed to run inside docker though where the webserver container can easily find the verifier nodes

andrew.whitehead (Thu, 30 Apr 2020 15:52:24 GMT):
Something is bound to those ports. I can't tell you what that is

davidd (Thu, 30 Apr 2020 15:53:15 GMT):
okay, I will check it again

davidd (Thu, 30 Apr 2020 15:53:42 GMT):
https://github.com/hyperledger/indy-sdk#ubuntu-based-distributions-ubuntu-1604-and-1804

davidd (Thu, 30 Apr 2020 15:53:54 GMT):
If follow these steps and install the indy-sdk

davidd (Thu, 30 Apr 2020 15:54:23 GMT):
and then start following these steps -> https://github.com/bcgov/von-network#running-the-the-web-server-on-your-local-machine

davidd (Thu, 30 Apr 2020 15:54:38 GMT):
and on the last like, where to start it, get this error

andrew.whitehead (Thu, 30 Apr 2020 15:55:12 GMT):
Try pip install python3-indy

davidd (Thu, 30 Apr 2020 16:09:23 GMT):
```~/von-network$ GENESIS_URL=https://raw.githubusercontent.com/sovrin-foundation/sovrin/master/sovrin/pool_transactions_sandbox_genesis LEDGER_SEED=000000000000000000000IanCostanzo PORT=9000 python -m server.server Traceback (most recent call last): File "/usr/lib/python2.7/runpy.py", line 163, in _run_module_as_main mod_name, _Error) File "/usr/lib/python2.7/runpy.py", line 119, in _get_module_details code = loader.get_code(mod_name) File "/usr/lib/python2.7/pkgutil.py", line 281, in get_code self.code = compile(source, self.filename, 'exec') File "/home/pc/von-network/server/server.py", line 53 async def index(request): ^ SyntaxError: invalid syntax```

andrew.whitehead (Thu, 30 Apr 2020 16:10:14 GMT):
Requires python 3.6+

davidd (Thu, 30 Apr 2020 16:11:06 GMT):
```python3 --version Python 3.6.9```

andrew.whitehead (Thu, 30 Apr 2020 16:11:53 GMT):
Run with `python3` instead of `python`. You'll also need to install packages with `pip3`

davidd (Thu, 30 Apr 2020 16:12:08 GMT):
did that

davidd (Thu, 30 Apr 2020 16:12:10 GMT):
```GENESIS_URL=https://raw.githubusercontent.com/sovrin-foundation/sovrin/master/sovrin/pool_transactions_sandbox_genesis LEDGER_SEED=000000000000000000000IanCostanzo PORT=9000 python3 -m server.server Traceback (most recent call last): File "/usr/lib/python3.6/runpy.py", line 193, in _run_module_as_main "__main__", mod_spec) File "/usr/lib/python3.6/runpy.py", line 85, in _run_code exec(code, run_globals) File "/home/mujeeb/von-network/server/server.py", line 9, in import aiohttp_jinja2 ModuleNotFoundError: No module named 'aiohttp_jinja2'```

davidd (Thu, 30 Apr 2020 16:12:10 GMT):
```GENESIS_URL=https://raw.githubusercontent.com/sovrin-foundation/sovrin/master/sovrin/pool_transactions_sandbox_genesis LEDGER_SEED=000000000000000000000IanCostanzo PORT=9000 python3 -m server.server Traceback (most recent call last): File "/usr/lib/python3.6/runpy.py", line 193, in _run_module_as_main "__main__", mod_spec) File "/usr/lib/python3.6/runpy.py", line 85, in _run_code exec(code, run_globals) File "/home/ubuntu/von-network/server/server.py", line 9, in import aiohttp_jinja2 ModuleNotFoundError: No module named 'aiohttp_jinja2'```

davidd (Thu, 30 Apr 2020 16:12:16 GMT):
ok

davidd (Thu, 30 Apr 2020 16:12:28 GMT):
which packages i need?

davidd (Thu, 30 Apr 2020 16:15:29 GMT):
`pip3 install -r server/requirements.txt`

davidd (Thu, 30 Apr 2020 16:16:45 GMT):
```~/von-network$ pip3 install Layers Collecting Layers Collecting bashutils (from Layers) Collecting PyYaml (from Layers) Installing collected packages: bashutils, PyYaml, Layers Successfully installed Layers-0.1.5 PyYaml-5.3.1 bashutils-0.0.4 ```

davidd (Thu, 30 Apr 2020 16:16:45 GMT):
```~/von-network$ pip3 install Layers Collecting Layers Collecting bashutils (from Layers) Collecting PyYaml (from Layers) Installing collected packages: bashutils, PyYaml, Layers Successfully installed Layers-0.1.5 PyYaml-5.3.1 bashutils-0.0.4 mujeeb@ip-172-31-18-44:~/von-network$ GENESIS_URL=https://raw.githubusercontent.com/sovrin-foundation/sovrin/master/sovrin/pool_transactions_sandbox_genesis LEDGER_SEED=000000000000000000000IanCostanzo PORT=9000 python3 -m server.server Traceback (most recent call last): File "/usr/lib/python3.6/runpy.py", line 193, in _run_module_as_main "__main__", mod_spec) File "/usr/lib/python3.6/runpy.py", line 85, in _run_code exec(code, run_globals) File "/home/mujeeb/von-network/server/server.py", line 14, in from .anchor import ( File "/home/mujeeb/von-network/server/anchor.py", line 19, in from plenum.common.signer_simple import SimpleSigner File "/home/mujeeb/.local/lib/python3.6/site-packages/plenum/__init__.py", line 2, in from plenum.Network import Network File "/home/mujeeb/.local/lib/python3.6/site-packages/plenum/Network.py", line 3, in from Layers import InputLayer, Sequential ModuleNotFoundError: No module named 'Layers'```

davidd (Thu, 30 Apr 2020 16:18:00 GMT):
```~/von-network$ GENESIS_URL=https://raw.githubusercontent.com/sovrin-foundation/sovrin/master/sovrin/pool_transactions_sandbox_genesis LEDGER_SEED=000000000000000000000IanCostanzo PORT=9000 python3 -m server.server Traceback (most recent call last): File "/usr/lib/python3.6/runpy.py", line 193, in _run_module_as_main "__main__", mod_spec) File "/usr/lib/python3.6/runpy.py", line 85, in _run_code exec(code, run_globals) File "/home/ubuntu/von-network/server/server.py", line 14, in from .anchor import ( File "/home/ubuntu/von-network/server/anchor.py", line 19, in from plenum.common.signer_simple import SimpleSigner File "/home/ubuntu/.local/lib/python3.6/site-packages/plenum/__init__.py", line 2, in from plenum.Network import Network File "/home/ubuntu/.local/lib/python3.6/site-packages/plenum/Network.py", line 3, in from Layers import InputLayer, Sequential ModuleNotFoundError: No module named 'Layers'```

davidd (Thu, 30 Apr 2020 16:18:35 GMT):
install the Layers module but still says `No module named 'Layers'`

davidd (Thu, 30 Apr 2020 16:18:35 GMT):
install the Layers module but still says 'No module found'

davidd (Thu, 30 Apr 2020 16:23:06 GMT):
```pip3 install -r server/requirements.txt Installing collected packages: pyyaml, pycparser, cffi, pycares, typing, aiodns, attrs, multidict, async-timeout, chardet, idna, yarl, idna-ssl, typing-extensions, aiohttp, MarkupSafe, jinja2, aiohttp-jinja2, aiosqlite, base58, cchardet, rlp, meld3, supervisor Segmentation fault (core dumped)```

andrew.whitehead (Thu, 30 Apr 2020 16:24:29 GMT):
I think the 'Layers' its referencing there is a part of indy-plenum. I'm not going to debug running the webserver this way right now, it works in docker.

davidd (Thu, 30 Apr 2020 16:25:57 GMT):
to start docker, I need to run this `~/von-network$ docker-compose up` ?

andrew.whitehead (Thu, 30 Apr 2020 16:26:19 GMT):
Use the manage script

davidd (Thu, 30 Apr 2020 16:27:24 GMT):
`./manage build` runs successfully

davidd (Thu, 30 Apr 2020 16:28:24 GMT):
````./manage start` returns this (docker detail) Define and run multi-container applications with Docker. Usage: docker-compose [-f ...] [options] [COMMAND] [ARGS...] docker-compose -h|--help .... ....```

davidd (Thu, 30 Apr 2020 16:28:24 GMT):
`./manage start` returns this (docker detail) ```Define and run multi-container applications with Docker. Usage: docker-compose [-f ...] [options] [COMMAND] [ARGS...] docker-compose -h|--help .... ....```

davidd (Thu, 30 Apr 2020 16:28:24 GMT):
```./manage start` returns this (docker detail) Define and run multi-container applications with Docker. Usage: docker-compose [-f ...] [options] [COMMAND] [ARGS...] docker-compose -h|--help .... ....```

andrew.whitehead (Thu, 30 Apr 2020 16:29:34 GMT):
That's odd. Running docker desktop maybe?

davidd (Thu, 30 Apr 2020 16:29:44 GMT):
i'm on ubuntu

davidd (Thu, 30 Apr 2020 16:29:44 GMT):
i'm using ubuntu in AWS

andrew.whitehead (Thu, 30 Apr 2020 16:30:31 GMT):
Hm, not sure what's going on thre

andrew.whitehead (Thu, 30 Apr 2020 16:30:31 GMT):
Hm, not sure what's going on there

davidd (Thu, 30 Apr 2020 16:36:13 GMT):
https://github.com/hyperledger/indy-sdk#how-to-start-local-nodes-pool-with-docker do I need to run this aswell?

andrew.whitehead (Thu, 30 Apr 2020 16:38:46 GMT):
That's another way to run a local indy-node ledger instance, without the browser container that von-network provides

davidd (Thu, 30 Apr 2020 16:39:23 GMT):
i want the von-network (the one with browser)

davidd (Thu, 30 Apr 2020 16:40:51 GMT):
and how will I able to access the browser UI in host machine, right now I'm using ubuntu through SSH

davidd (Thu, 30 Apr 2020 16:41:17 GMT):
I will have a look at docker, maybe I need some help, If i failed to start

davidd (Thu, 30 Apr 2020 16:41:36 GMT):
thanks for ur time and @swcurran any idea what am I missing?

davidd (Thu, 30 Apr 2020 16:41:36 GMT):
thanks for ur time and @swcurran any idea?

ianco (Thu, 30 Apr 2020 20:47:18 GMT):
Jumping on this thread a bit late, but ...

ianco (Thu, 30 Apr 2020 20:47:31 GMT):
(1) if you want to run the webserver local you need to install indy-plenum

ianco (Thu, 30 Apr 2020 20:48:03 GMT):
(2) you can run the docker version of von-network and have it available from localhost as follows:

ianco (Thu, 30 Apr 2020 20:48:53 GMT):
`./manage start 192.168.0.11` (substitute your own local IP)

ianco (Thu, 30 Apr 2020 20:50:03 GMT):
(note that if you do this then you cannot connect to your Indy nodes from within another docker container)

ianco (Thu, 30 Apr 2020 20:51:00 GMT):
If you try this, you can connect to your local `http://localhost:9000/genesis` and you should see your local IP's in the genesis transactions

davidd (Thu, 30 Apr 2020 21:04:28 GMT):
I'm using Ubuntu Image on AWS

davidd (Thu, 30 Apr 2020 21:05:07 GMT):
and want to run von-network (don't really care how), and want to make it accessible publicly

davidd (Thu, 30 Apr 2020 21:05:54 GMT):
the reason why I'm installing von-network is to make a private network in which I can run my Aries Agents

ianco (Thu, 30 Apr 2020 21:06:28 GMT):
so a private network available publicly?

ianco (Thu, 30 Apr 2020 21:06:36 GMT):
(private as in for you personally)

davidd (Thu, 30 Apr 2020 21:07:15 GMT):
for private I mean we don't want for now a sovrin network but want to create our own

ianco (Thu, 30 Apr 2020 21:07:35 GMT):
the trick is with the IP addresses of the nodes ...

ianco (Thu, 30 Apr 2020 21:07:57 GMT):
you have to be able to connect (from your agent) using the same IP address that the node itself is using

ianco (Thu, 30 Apr 2020 21:08:10 GMT):
... because the IP address is in the genesis transactions

ianco (Thu, 30 Apr 2020 21:08:37 GMT):
with AWS can you specify the exposed IP address?

davidd (Thu, 30 Apr 2020 21:08:42 GMT):
I got it (i actually know that I need to change the IP in genisis file)

davidd (Thu, 30 Apr 2020 21:08:59 GMT):
yes (for AWS)

andrew.whitehead (Thu, 30 Apr 2020 21:09:24 GMT):
Changing the IP in the genesis file will just make it not match the merkle tree

davidd (Thu, 30 Apr 2020 21:10:00 GMT):
ok here is my environment (let me explain my problem)

davidd (Thu, 30 Apr 2020 21:12:38 GMT):
We have Web Agent and Mobile Agent based on Aries dotnet. We want to create our own private network (node, ledger etc) and for now don't want to use Sovrin Network. We are going to host that on AWS machine in Ubuntu environment. What we want is to connect these Web/Mobile Agent with our private network. Questions are how to create our own Private Network and to connect the Aries Agents with this network?

davidd (Thu, 30 Apr 2020 21:12:38 GMT):
We have Web Agent and Mobile Agent based on Aries dotnet. We want to create our own private network (node, ledger etc) and for now don't want to use Sovrin Network. We are going to host that on AWS machine in Ubuntu enviornment. What we want is to connect these Web/Mobile Agent with our network. Questions are how to create our own Private Network and to connect the Aries Agents with this network?

davidd (Thu, 30 Apr 2020 21:13:51 GMT):
Agents will offcourse be public and we want the network to be private!

ianco (Thu, 30 Apr 2020 21:19:30 GMT):
We (BC Gov) have some test networks that we run on Digital Ocean - one example is http://test.bcovrin.vonx.io

davidd (Thu, 30 Apr 2020 21:19:50 GMT):
I already checked this out

ianco (Thu, 30 Apr 2020 21:19:55 GMT):
... I've sent a note to the devops guy (Wade) to find out where he keeps his configs

davidd (Thu, 30 Apr 2020 21:20:07 GMT):
and we actually want something like this

davidd (Thu, 30 Apr 2020 21:20:14 GMT):
@ianco thanks

ianco (Thu, 30 Apr 2020 21:20:25 GMT):
My limited understanding of this is you just have to make sure the Indy nodes are running on the same IP's as they are published on

davidd (Thu, 30 Apr 2020 21:21:23 GMT):
okay and in genesis file there is a specific IP which is for Node

davidd (Thu, 30 Apr 2020 21:21:38 GMT):
what am I going to put there?

davidd (Thu, 30 Apr 2020 21:22:36 GMT):
and what is the best practice to run a private network (docker or webserver)?

davidd (Thu, 30 Apr 2020 21:22:36 GMT):
and what is the best practice to run a private network (docker or webserver)

ianco (Thu, 30 Apr 2020 21:23:26 GMT):
Maybe this is what you want: https://github.com/bcgov/von-network#running-the-network-on-a-vps

ianco (Thu, 30 Apr 2020 21:25:12 GMT):
... ok this is what Wade just told me: Example; `./manage start 138.197.138.255 WEB_SERVER_HOST_PORT=80 "LEDGER_INSTANCE_NAME=BCovrin Test" "INFO_SITE_TEXT=vonx.io" "INFO_SITE_URL=https://vonx.io/"`

ianco (Thu, 30 Apr 2020 21:25:25 GMT):
the IP is the public IP of the machine

ianco (Thu, 30 Apr 2020 21:26:12 GMT):
... this is the end result of the VPS docs

davidd (Thu, 30 Apr 2020 21:26:26 GMT):
thanks, can u ask him some questions regarding the scalability also

Moshe7 (Thu, 30 Apr 2020 21:26:29 GMT):
Has joined the channel.

davidd (Thu, 30 Apr 2020 21:26:56 GMT):
like, what If I have a thousands of issuers/holder?

ianco (Thu, 30 Apr 2020 21:27:06 GMT):
I'll go one better ... @WadeBarnes

ianco (Thu, 30 Apr 2020 21:27:38 GMT):
We haven't tested with thousands of issuers, but we've issued millions of credentials on a single issuer

davidd (Thu, 30 Apr 2020 21:27:54 GMT):
for aries agents we connect them with pool

davidd (Thu, 30 Apr 2020 21:28:10 GMT):
which is through genesis file

davidd (Thu, 30 Apr 2020 21:28:30 GMT):
which genesis file would I need?

ianco (Thu, 30 Apr 2020 21:28:39 GMT):
There are only a small number of ledger writes (public DIDs, schemas and cred defs)

ianco (Thu, 30 Apr 2020 21:29:06 GMT):
Once you start the network you can get the genesis file from `http:///genesis`

davidd (Thu, 30 Apr 2020 21:29:14 GMT):
so for every Issuer we need to create a new ledger?

davidd (Thu, 30 Apr 2020 21:29:25 GMT):
okay got it

ianco (Thu, 30 Apr 2020 21:29:51 GMT):
No, for each issuer you need to write a new DID to the ledger, plus any schemas and cred defs for credentials that issuer will issue

ianco (Thu, 30 Apr 2020 21:30:04 GMT):
When you issue a credential nothing is written to the ledger

davidd (Thu, 30 Apr 2020 21:31:21 GMT):
'we've issued millions of credentials on a single issuer' to reference ur quote, I mean that if I want multiple Issuers?

davidd (Thu, 30 Apr 2020 21:32:18 GMT):
can I have multiple Issuers and everyone can Issue unlimited credentials

davidd (Thu, 30 Apr 2020 21:32:25 GMT):
on this 1 network?

davidd (Thu, 30 Apr 2020 21:32:32 GMT):
will that be enough

davidd (Thu, 30 Apr 2020 21:32:32 GMT):
will that be enough?

ianco (Thu, 30 Apr 2020 21:32:51 GMT):
yes issuing credentials is not limited by the network

davidd (Thu, 30 Apr 2020 21:33:21 GMT):
okay, because Scalability is our concern (and it should be)

ianco (Thu, 30 Apr 2020 21:34:15 GMT):
We run our applications on openshift, and each component scales independently ...

ianco (Thu, 30 Apr 2020 21:34:41 GMT):
So obviously when you are issuing millions of credentials there are scalability concerns and bottlenecks, but nothing related to the Indy network

davidd (Thu, 30 Apr 2020 21:35:10 GMT):
great :) thanks for ur time

davidd (Thu, 30 Apr 2020 21:36:10 GMT):
If i have some questions can I ask u or @WadeBarnes (for DevOps qs)?

ianco (Thu, 30 Apr 2020 21:36:16 GMT):
:+1:

ianco (Thu, 30 Apr 2020 21:36:35 GMT):
just ask the questions on this channel and `@` us

davidd (Thu, 30 Apr 2020 21:36:46 GMT):
okay

RicardoPeixoto (Fri, 01 May 2020 13:46:30 GMT):
Hi, does indy sdk implement peer dids?

swcurran (Fri, 01 May 2020 14:51:42 GMT):
Indy SDK supports the crypto to use peer dids (creation, storage, pack and unpack of messages). How those DIDs are used is a higher level abstraction - an Aries thing.

RicardoPeixoto (Fri, 01 May 2020 16:37:46 GMT):
Is it possible to create off ledger DID Docs and interact with them with indy sdk?

swcurran (Fri, 01 May 2020 16:45:03 GMT):
Yes

qtle (Sat, 02 May 2020 13:28:21 GMT):
Has joined the channel.

iamShashvat (Sun, 03 May 2020 19:12:59 GMT):
Is it possible to import externally created ED25519 keys in our wallet(indy-sdk)?

andrew.whitehead (Sun, 03 May 2020 19:23:21 GMT):
Yes, you can pass the seed into `create_and_store_my_did` or `generate_wallet_key`

andrew.whitehead (Sun, 03 May 2020 19:23:21 GMT):
Yes, you can pass the seed into `create_and_store_my_did` or `create_key`

iamShashvat (Sun, 03 May 2020 20:13:58 GMT):
@andrew.whitehead but what if I dont have seed? let's say according to an architecture, a old service creates a crypto keypair(ed25519) using crypto.js library(without seed) and if I want to use that keys for Indy requests to sign, I have to make a copy/import of keys in indy wallet. Is that possible? Can I use import_wallet() in anyway to store those keys?

andrew.whitehead (Sun, 03 May 2020 20:15:21 GMT):
If you have the private key, the seed is the first 32 bytes. If you don’t, then I don’t see how you can import it

iamShashvat (Sun, 03 May 2020 20:52:05 GMT):
@andrew.whitehead size of the private key(indy/ED25519) is of 32 bytes so if I put whole private key in `create_and_store_did` as seed I must get the same keypair, is it?

andrew.whitehead (Sun, 03 May 2020 21:03:09 GMT):
Give it a try

kdenhartog (Sun, 03 May 2020 21:58:26 GMT):
Looks like the secret key is the first 32 bytes of a sha_512 hash where the seed is the input to the hash function. I don't think that key will be recoverable to be passed into indy_sdk.

kdenhartog (Sun, 03 May 2020 22:00:37 GMT):
With that in mind, if you really need a key to be put in IndySDK you can sideload it, but it's a painful process that you'll have to look through the codebase to understand and be able to perform, but it's not impossible.

kdenhartog (Sun, 03 May 2020 22:00:37 GMT):
With that in mind, if you really need a key to be put in IndySDK and you have the public and private key you can sideload it, but it's a painful process that you'll have to look through the codebase to understand and be able to perform, but it's not impossible. I'd only recommend going that route if it's for a production instance.

pierluigiriti (Sun, 03 May 2020 22:17:37 GMT):
Has joined the channel.

pierluigiriti (Sun, 03 May 2020 22:31:35 GMT):
Hello, I still try to install indy-sdk and indy-node on Ubuntu 19.10, but look like is not supported

iamShashvat (Mon, 04 May 2020 03:07:21 GMT):
Oh okay! Thank you for replying!

anurkalem (Mon, 04 May 2020 22:40:45 GMT):
Has joined the channel.

swcurran (Mon, 04 May 2020 23:18:33 GMT):
There are parallel Presentation Request "standards" efforts with Digital Bazaar (W3C CCG) and DIF. Both look on the surface like they have fewer features than the Indy Presentation Request format - particularly the Digital Bazaar one. It also looks to me that the differ significantly from one another. Digital Bazaar Proposal: https://digitalbazaar.github.io/vp-request-spec/ DIF: https://identity.foundation/presentation-exchange/ Questions: 1. Is there a concise document that describes the Indy format as it exists today that can be used as a measuring stick against the others? - Some of the "Indy-isms" may need to be translated, but it would be interesting to see if the concepts in the Indy format are present in the other two. 2. Is anyone from the Indy community involved in either of those efforts?

kdenhartog (Tue, 05 May 2020 00:47:26 GMT):
As for question 1, I don't believe so. One major thing worth noting is the RFCs defined are describing an entire protocol where as those are more just about the request format and are combined with other pieces of the puzzle (such as CHAPI) to go through the verification flow. As for 2, I've been keeping an eye on both of those. I believe @brentzundel was helping out with the DIF one initially, but unsure of his level of contribution now. The DIF description is further along as far as a spec stands, and offers methods that will be more obvious how selective disclosure is done. With regards to the DB approach, this can also be achieved, but the object would need to be defined. It's entirely possible that the DIF object is defined as a JSON LD object that defines a selective disclosure query that could be put in the attachment of RFC 37 [Request Presentation object](https://github.com/hyperledger/aries-rfcs/tree/master/features/0037-present-proof#request-presentation) or in CHAPI as has been done so far in the DHS interop work.

kdenhartog (Tue, 05 May 2020 00:47:44 GMT):
As for question 1, I don't believe so. One major thing worth noting is the RFCs defined are describing an entire protocol where as those are more just about the request format and are combined with other pieces of the puzzle (such as CHAPI) to go through the verification flow. As for 2, I've been keeping an eye on both of those. I believe @brentzundel was helping out with the DIF one initially, but unsure of his level of contribution now. The DIF description is further along as far as a spec stands, and offers methods that will be more obvious how selective disclosure is done. With regards to the DB approach, this can also be achieved, but the object would need to be defined. It's entirely possible that the DIF object is defined as a JSON LD object that defines a selective disclosure query that could be put in the attachment of RFC 37 [Request Presentation object](https://github.com/hyperledger/aries-rfcs/tree/master/features/0037-present-proof#request-presentation) or in CHAPI as has been done so far in the DHS interop work.

kdenhartog (Tue, 05 May 2020 00:47:44 GMT):
As for question 1, I don't believe so. One major thing worth noting is the RFCs defined are describing an entire protocol where as those are more just about the request format and are combined with other pieces of the puzzle (such as CHAPI) to go through the verification flow. As for 2, I've been keeping an eye on both of those. I believe @brentzundel was helping out with the DIF one initially, but unsure of his level of contribution now. The DIF description is further along as far as a spec stands, and offers methods that will be more obvious how selective disclosure is done. With regards to the DB approach, this can also be achieved, but the object would need to be defined and the spec is more at a strawman proposal stage at this point with few to no normative statements yet. It's entirely possible that the DIF object is defined as a JSON LD object that defines a selective disclosure query that could be put in the attachment of RFC 37 [Request Presentation object](https://github.com/hyperledger/aries-rfcs/tree/master/features/0037-present-proof#request-presentation) or in CHAPI as has been done so far in the DHS interop work.

swcurran (Tue, 05 May 2020 02:27:24 GMT):
Agree that the deliberately RFC does not define the Indy presentation request format. The Indy format is well defined, I'm just looking for a document that we can share with the other communities. We've used it a lot in many different ways, so it's extensive. What I want to be sure is that we list the concepts that are covered in the Indy one, including a mapping to JSON-LD concepts. The two specs jump straight into examples without describing what should be supported.

brentzundel (Tue, 05 May 2020 02:41:07 GMT):
I was pretty heavily involved in the initial architecture of the DIF presentation definition, but am not able to continue participating. I'm not sure where things stand now, but at the time it looked like what they were specifying would probably be something that we could use with Indy and Aries.

swcurran (Tue, 05 May 2020 02:45:09 GMT):
Good to hear - thanks. The DIF one definitely look better.

notsteward (Tue, 05 May 2020 18:12:24 GMT):
Hey! I am writing program in C using libindy.so, but I can't understand why the result from libindy.so is intermittent. I have to use sleep() to wait for the result. Please check my code and please correct me. `char did[50]={0}; char verkey[100]={0}; indy_handle_t wallet_handle=0; indy_handle_t ht = 0; void callback(indy_handle_t command_handle, indy_error_t err,const char _did[30],const char _verkey[50]){ sleep(2); memcpy(did,_did,strlen(_did)); memcpy(verkey,_verkey,strlen(_verkey)); } err = indy_create_and_store_my_did(ht,wallet_handle,"{}",&callback); sleep(3); printf("\nError did %d",err); sleep(1); printf("\n Did: %s",did); printf("\n Verkey: %s\n",verkey); ` Please let me know how can I avoid sleep()?

andrew.whitehead (Tue, 05 May 2020 18:22:40 GMT):
It's an asynchronous call so the result takes as long as it takes. Is there a reason you're using C? You could await the result much more naturally in javascript or python

notsteward (Tue, 05 May 2020 18:27:59 GMT):
I have to use it in C due to some limitations, So how can I handle these asynchronous call in C?

andrew.whitehead (Tue, 05 May 2020 18:35:10 GMT):
This is a general programming question not really an indy question. There's a ton of ways you might implement it depending on what you're writing. You might want to create separate threads for each call and join the thread when it completes, for instance. In C++ you might be able to use std::future

notsteward (Tue, 05 May 2020 18:37:40 GMT):
Yes, sorry! I understand this is not indy based question I just wanted to know if someone had implemented before using C.

andrew.whitehead (Tue, 05 May 2020 18:42:34 GMT):
It would be good if there were examples in C to look at, but I'm not aware of any

notsteward (Tue, 05 May 2020 18:47:28 GMT):
I have a doubt regarding keys in indy, please correct me if I am wrong: 1.) Indy creates the ED25519 keypair then it encode it in base58, out of which the first 16 bits of the publicKey/Verkey is DID 2.) signature: Ed25519 digital signature! Is it encode into base64 after signature generation?

RicardoPeixoto (Tue, 05 May 2020 19:06:46 GMT):
Hi, is there some way to query all records of a given type in "openWalletSearch"? I tried to pass an empty query "{}" but it did not return any records. If i do the simple query by id it works just fine.

notsteward (Tue, 05 May 2020 19:11:48 GMT):
You can do listMyDidsWithMeta ( wh ) where wh is wallet handle! it will returns all the DID present in your wallet.

notsteward (Tue, 05 May 2020 19:12:23 GMT):
@andrew.whitehead please correct my understanding.

RicardoPeixoto (Tue, 05 May 2020 19:13:06 GMT):
But here i'm in the non secrects section

andrew.whitehead (Tue, 05 May 2020 19:13:33 GMT):
I don't think you can find anything without at least the record type identifier

RicardoPeixoto (Tue, 05 May 2020 19:14:32 GMT):
There are options to query like mongoDB but i want all records from a type without filters

andrew.whitehead (Tue, 05 May 2020 19:14:33 GMT):
Are you talking about indy_crypto_sign?

andrew.whitehead (Tue, 05 May 2020 19:15:37 GMT):
This is the nodejs wrapper I assume

RicardoPeixoto (Tue, 05 May 2020 19:16:14 GMT):
Yes you're right.

notsteward (Tue, 05 May 2020 19:16:22 GMT):
yes create_and_store_did, create_keys, crypto_sign yes!

andrew.whitehead (Tue, 05 May 2020 19:18:50 GMT):
I think that should work. Not sure if it expects the query as a string or as inline JSON

andrew.whitehead (Tue, 05 May 2020 19:19:46 GMT):
Then I believe it returns the raw signature in bytes, not encoded as base64

RicardoPeixoto (Tue, 05 May 2020 19:22:06 GMT):
I passed as inline JSON like they have in the example: {}. The quotes were just to represent the exact thing i passed.

RicardoPeixoto (Tue, 05 May 2020 19:23:07 GMT):
I can probably tag all records of that type with a specific value and query by tag but i was looking for a cleaner way.

notsteward (Tue, 05 May 2020 19:23:59 GMT):
sign_request encode in base64? and what about first point, is that correct?

andrew.whitehead (Tue, 05 May 2020 19:25:56 GMT):
Yes that's how it derives the DID, although you can specify a custom one instead. sign_request sets the signature property of the message. I believe that one is encoded in base58

mccown (Tue, 05 May 2020 19:27:35 GMT):
aws

RicardoPeixoto (Tue, 05 May 2020 19:45:11 GMT):
It works with inline JSON {}. I was passing the wrong wallet handle.

notsteward (Tue, 05 May 2020 19:45:29 GMT):
Okay Okay, Thanks

mccown (Tue, 05 May 2020 21:10:20 GMT):
Has anyone ever tested with the indy-sdk docker (ledger) running on an AWS ec2 instance? Normally, I run the test ledger docker locally. However, I wanted to do some network tests and so I built a docker on AWS. For testing simplicity (on the client), I'm using the indy-sdk Java sample. While connecting to a local docker works fine, when I try to open to the AWS-based ledger pool (Pool.openPoolLedger(poolName, "{}")), I'm getting timeout errors. When building / running the docker, I set the IP to be our local AWS instance's IP. In the client, I set the testPoolIp to our AWS instance's external address. If any of you have tried this, is this the right setup?

swcurran (Tue, 05 May 2020 21:38:08 GMT):
Not sure if you are asking about indy-sdk or indy-node (ledger), but assuming indy-node. We run von-network on Digital Ocean so quite similar. There is some docs on doing that - https://github.com/bcgov/von-network#running-the-network-on-a-vps @WadeBarnes has a lot of experience in doing that, so may be able to help you debug a bit -- especially if you are using von-network as a basis. It's the easiest way to run a ledger.

mccown (Tue, 05 May 2020 21:57:36 GMT):
You're correct, my question is really an indy-node question masquerading as an indy-sdk question. :-) Many thanks for the doc, it looks like my answer is in the "Running the web server in Docker against another ledger" section. I'll do some reading and then likely have some questions for Wade.

WadeBarnes (Tue, 05 May 2020 21:59:57 GMT):
I'll be happy to answer any questions you have.

kdenhartog (Tue, 05 May 2020 23:26:35 GMT):
Agree, both those specs are in flight which has been the hardest part to work against. If you write an initial comparison doc, I can review it and help get it to a point where we can share it.

kdenhartog (Tue, 05 May 2020 23:27:21 GMT):
Or even just write up the table for Indy and I can add what DIF and VP-request-spec support within the table

mccown (Wed, 06 May 2020 03:16:42 GMT):
@WadeBarnes I'm not a docker developer and I think my question is more of a docker question. I'd be grateful for any insights you might have. Within the Indy SDK, there are instructions for building a docker that runs 4 indy-node instances. I've been using this locally to test some client apps and now I'd like to move it to an AWS instance to test some mobile clients. I was planning to move the docker to AWS and forward my client traffic there. However, it appears that dockers (internally) control the network traffic and seem to be blocking non-local traffic. Do you know of a straightforward way to enable a docker accept non-local traffic?

mccown (Wed, 06 May 2020 03:16:42 GMT):
@WadeBarnes I'm not a docker developer and I think my question is more of a docker question. I'd be grateful for any insights you might have. Within the Indy SDK, there are instructions for building a docker that runs 4 indy-node instances. I've been using this locally to test some client apps and now I'd like to move it to an AWS instance to test some mobile clients. I was planning to move the docker to AWS and forward my client traffic there. However, it appears that dockers (internally) control the network traffic and seem to be blocking non-local traffic. Do you know of a straightforward way to enable a docker to accept non-local traffic?

shailaja.mahara (Wed, 06 May 2020 11:27:47 GMT):
Has joined the channel.

WadeBarnes (Wed, 06 May 2020 12:32:18 GMT):
To confirm, you're wanting to set up an Indy Node ledger, for test purposes, that is publicly available? Like this; http://test.bcovrin.vonx.io/

WadeBarnes (Wed, 06 May 2020 12:35:15 GMT):
If that's correct, it's super easy using `von-network`. It creates a network of 4 Indy Node instance plus a ledger browser. `test.bcovrin.vonx.io` is one such instance we maintain.

WadeBarnes (Wed, 06 May 2020 12:35:15 GMT):
If that's correct, it's super easy using `von-network`. It creates a network of 4 Indy Node instances plus a ledger browser. `test.bcovrin.vonx.io` is one such instance we maintain.

WadeBarnes (Wed, 06 May 2020 12:38:08 GMT):
Once you have a docker environment installed on the machine it's a simple matter of pulling down a copy of the `von-network` repository and running; ``` ./manage build ./manage start ```

WadeBarnes (Wed, 06 May 2020 12:38:08 GMT):
Once you have a docker environment installed on the machine (instructions here [Running the Network on a VPS](https://github.com/bcgov/von-network#running-the-network-on-a-vps)) it's a simple matter of pulling down a copy of the `von-network` repository and running; ``` ./manage build ./manage start ```

WadeBarnes (Wed, 06 May 2020 12:38:08 GMT):
Once you have a docker environment installed on the machine it's a simple matter of pulling down a copy of the `von-network` repository and running; ``` ./manage build ./manage start ``` Instructions for all that can be found here; [Running the Network on a VPS](https://github.com/bcgov/von-network#running-the-network-on-a-vps))

WadeBarnes (Wed, 06 May 2020 12:38:08 GMT):
Once you have a docker environment installed on the machine it's a simple matter of pulling down a copy of the `von-network` repository and running; ``` ./manage build ./manage start ``` Instructions for all that can be found here; [Running the Network on a VPS](https://github.com/bcgov/von-network#running-the-network-on-a-vps)

WadeBarnes (Wed, 06 May 2020 12:39:19 GMT):
Here is the command used to start `test.bcovrin.vonx.io` and make it publicly available, as an example: ``` ./manage start 138.197.138.255 WEB_SERVER_HOST_PORT=80 "LEDGER_INSTANCE_NAME=BCovrin Test" "INFO_SITE_TEXT=vonx.io" "INFO_SITE_URL=https://vonx.io/" ```

WadeBarnes (Wed, 06 May 2020 12:41:10 GMT):
That command is constructed based on the [Running the Network on a VPS](https://github.com/bcgov/von-network#running-the-network-on-a-vps) and [Customize your Ledger Browser Deployment](https://github.com/bcgov/von-network#customize-your-ledger-browser-deployment)

WadeBarnes (Wed, 06 May 2020 12:41:10 GMT):
That command is constructed based on the [Running the Network on a VPS](https://github.com/bcgov/von-network#running-the-network-on-a-vps) and [Customize your Ledger Browser Deployment](https://github.com/bcgov/von-network#customize-your-ledger-browser-deployment) documentation.

WadeBarnes (Wed, 06 May 2020 12:43:31 GMT):
In the example above `138.197.138.255` is the public IP address of the host machine. That's the part that tells the network to listen on the public address and not the internal one.

WadeBarnes (Wed, 06 May 2020 12:45:39 GMT):
You will also need to ensure ports `9701-9708` are open on the machine. Indy Node and it's client's use ports 9701 through 9799 for communication.

WadeBarnes (Wed, 06 May 2020 12:47:35 GMT):
In this case `von-network is only using `9701-9708` since there are only 4 nodes. Each node, as you may know, uses two ports; one for admin traffic, and the other for client traffic.

WadeBarnes (Wed, 06 May 2020 12:47:35 GMT):
In this case `von-network` is only using `9701-9708` since there are only 4 nodes. Each node, as you may know, uses two ports; one for admin traffic, and the other for client traffic.

ken5scal (Wed, 06 May 2020 13:18:19 GMT):
Has joined the channel.

rcaldeira (Wed, 06 May 2020 14:43:59 GMT):
Has joined the channel.

jameshiester (Wed, 06 May 2020 20:21:03 GMT):
Can someone please expand on the notion of endpoints in DID documents? If what I'm understanding is correct, these should be the domain of the agent? I've been looking through the demo docs/code, but haven't found where/how these are defined?

swcurran (Wed, 06 May 2020 22:06:15 GMT):
The endpoint is a URI where you would send a message to the agent. That might be an endpoint controlled by the agent itself, or it might be the endpoint of another agent that serves as a mediator, passing on messages received to the intended agent. Think of it as the entry point to the domain of the agent. The endpoint can be any addressable point using supported transports - a URL, a websocket, an email address (if an agent supports that). The endpoint combines the address and the protocol - e.g. a URL. In Indy, an endpoint for a public DID is set via an "ATTRIB" transaction on the ledger.

jameshiester (Wed, 06 May 2020 22:32:40 GMT):
Do the original nodes from the genesis block have to remain online, or can new nodes be bootstrapped from nodes that are added later

swcurran (Wed, 06 May 2020 22:54:26 GMT):
AFAIK - at least some of them need to be online - not all. Technically, a new genesis file can be created with some or all of the current nodes and used just as well. However, as part of the governance structure, you do want to keep that static so people don't have to determine if they trust a new version of the genesis file all the time.

jameshiester (Wed, 06 May 2020 22:57:56 GMT):
is this done anywhere in the samples currently?

swcurran (Wed, 06 May 2020 22:58:40 GMT):
ACA-Py does that, but I'm not sure where. I'll take a look. I don't think the demos do that.

swcurran (Wed, 06 May 2020 23:09:52 GMT):
Hmm...weird. Just spun up an agent to look at the ACA-Py API and there does not seem to be a request the controller can make to do that. However, there is in the code a way to set the endpoint. @andrew.whitehead -- could you fill in the blank on that? Is it because the the endpoint is picked up with the start up parameters?

swcurran (Wed, 06 May 2020 23:10:21 GMT):
I suspect it is automatically set based on this start parameter: ``` -e [ ...], --endpoint [ ...] Specifies the endpoints to put into DIDDocs to inform other agents of where they should send messages destined for this agent. Each endpoint could be one of the specified inbound transports for this agent, or the endpoint could be that of another agent (e.g. 'https://example.com/agent-endpoint') if the routing of messages to this agent by a mediator is configured. The first endpoint specified will be used in invitations. The endpoints are used in the formation of a connection with another agent. ```

andrew.whitehead (Wed, 06 May 2020 23:14:18 GMT):
The agent publishes its endpoint automatically when starting up or changing the public DID using /wallet/did/public

jameshiester (Wed, 06 May 2020 23:22:50 GMT):
Cool, another question on the indy-sdk, I'm assuming any storage plugin must be defined in rust?

andrew.whitehead (Wed, 06 May 2020 23:24:43 GMT):
I believe that's true for the SDK

andrew.whitehead (Wed, 06 May 2020 23:29:39 GMT):
Are you looking at adding another backend?

jameshiester (Wed, 06 May 2020 23:40:56 GMT):
yeah, I'm interested in a cloud backend implementation

jameshiester (Wed, 06 May 2020 23:41:52 GMT):
most likely dynamodb

mccown (Thu, 07 May 2020 01:59:30 GMT):
@WadeBarnes thank you, this is exactly what I needed. I'm downloading the modules now. I may have a few other questions about your system in general, but this will help me setup my test environment.

tsekwena (Thu, 07 May 2020 06:59:30 GMT):
Has joined the channel.

tsekwena (Thu, 07 May 2020 06:59:32 GMT):
Hello Guys , I am new here , can you please assist me with documents or links on: 1. how to revoke your details(digital wallet) should there be any abuse by user or vendor ? 2. how information gets shared between user and vendor or relier in hyperleder ? Thank you

tsekwena (Thu, 07 May 2020 10:34:46 GMT):
My question above is basically trying to find different ways in which revocation takes places between Issuer ,Prover and Verifier ?

swcurran (Thu, 07 May 2020 14:48:22 GMT):
Revocation is done unilaterally by the issuer by publishing a transaction to the ledger that revokes one or more issued credentials. When an issuer use that transaction or later, they are unable to create a "proof of non-revocation", and thus, the verifier knows that the credential has been revoked. An overview of the current process is here - https://github.com/hyperledger/indy-hipe/tree/master/text/0011-cred-revocation.

jameshiester (Thu, 07 May 2020 22:35:33 GMT):
"How the DIDs for the author and endorser formally get created on a production ledger is not covered" is there a document anywhere covering production setup?

swcurran (Thu, 07 May 2020 22:52:56 GMT):
Could you add some more context about that question?

swcurran (Thu, 07 May 2020 22:53:32 GMT):
Are you trying to see how to add a DID (and other objects) if you are not an endorser?

jameshiester (Thu, 07 May 2020 22:54:44 GMT):
I'd like to understand how I could initialize the network with the steward keys so that I can lock down DIDs. I understand that's not really the use case for the repo, but even the start using docker documentation doesn't really address key generation currently

swcurran (Thu, 07 May 2020 22:55:27 GMT):
Ah...I found the quote.

swcurran (Thu, 07 May 2020 22:56:26 GMT):
OK - in both cases, the DID is created by someone with authority and created with the appropriate permissions.

swcurran (Thu, 07 May 2020 22:57:55 GMT):
You want to be an author. You create a seed and keep it secret - it must be treated the same as private key. You use that to create a DID in your wallet - possibly using a UI, or just using Indy CLI. In doing that, you get the nym (the DID) and the verkey (public key).

swcurran (Thu, 07 May 2020 22:59:26 GMT):
You give those two public pieces of info to someone who already has rights to the ledger. They use a UI or the CLI to execute the transaction to write the DID to the ledger. Done - you have your DID. It's easy because they don't have to write the DID to the ledger, etc.

swcurran (Thu, 07 May 2020 23:00:30 GMT):
A UI like the "Authenticate a New DID" on the http://dev.bcovrin.vonx.io/ site can be used. Note that on that you can set the permissions you want.

jameshiester (Thu, 07 May 2020 23:02:12 GMT):
I guess what I'm trying to accomplish is hosting the network in the cloud like the bcovin is doing, but controlling the keys such that there is not an option for others to use the ui or api to create a did with the endorser role without going through me

jameshiester (Thu, 07 May 2020 23:03:33 GMT):
I mean what's protecting the prod network from ddos attacks?

jameshiester (Thu, 07 May 2020 23:04:58 GMT):
maybe my mentality is wrong, do you see problems with a bunch of issuers being able to join the network?

jcourt (Thu, 07 May 2020 23:59:07 GMT):
@WadeBarnes Running through all this instructions gets the VON network started on my EC2 node but it is stuck in a loop seemingly not synchronising nodes ? Have you seen this before ? I would think it is a traffic blocking problem but wondered if you had hit this before ? node2_1 | 2020-05-07 23:56:30,728|NOTIFICATION|primary_connection_monitor_service.py|Node2:0 primary has been disconnected for too long node2_1 | 2020-05-07 23:56:30,729|INFO|primary_connection_monitor_service.py|Node2:0 The node is not ready yet so view change will not be proposed now, but re-scheduled. node2_1 | 2020-05-07 23:56:30,730|INFO|primary_connection_monitor_service.py|Node2:0 scheduling primary connection check in 60 sec node2_1 | 2020-05-07 23:56:30,730|NOTIFICATION|primary_connection_monitor_service.py|Node2:0 primary has been disconnected for too long node2_1 | 2020-05-07 23:56:30,730|INFO|primary_connection_monitor_service.py|Node2:0 The node is not ready yet so view change will not be proposed now, but re-scheduled.

jcourt (Fri, 08 May 2020 00:13:47 GMT):
My understanding is that you have to create that chain of trust from a Steward down to an Endorser. Only if you provide an open service that "Allows" people to become an Endorser is there a risk of DID Write attacks. Some networks set up for testing like the Sovrin Staging allow this as a public service for developers but your own Indy Ledger instance wouldn't have to.

jameshiester (Fri, 08 May 2020 00:37:57 GMT):
My understanding is the same. I like the convenience of the VON setup, but there's steps missing from making it a controlled network. It would be nice if the steward seeds could be configured via args in the setup script (for both VON and the indy node) so you could easily create your own secret keys. I know there's helper scripts in plenum, but with the lack of documentation it seems like quite a slog.

jameshiester (Fri, 08 May 2020 00:38:59 GMT):
is this the correct seed for the genesis transaction steward? 'seed': '000000000000000000000000Steward1' (taken from the getting started story docs)

jcourt (Fri, 08 May 2020 00:47:51 GMT):
I believe so

jameshiester (Fri, 08 May 2020 00:51:54 GMT):
did you check your security group rules?

jcourt (Fri, 08 May 2020 00:53:19 GMT):
Yeah so I think the issue is that they are listening on a 0.0.0.0:Port but trying to connect to their peers via the public IP ?

jcourt (Fri, 08 May 2020 00:56:35 GMT):
Its pretty likely security related

jcourt (Fri, 08 May 2020 02:02:59 GMT):
@jameshiester it was the EC2 security group wasn't allowing the nodes to access the public IP. Thanks !

shemnon (Fri, 08 May 2020 05:13:13 GMT):
Has left the channel.

WadeBarnes (Fri, 08 May 2020 12:37:13 GMT):
Glad you got that figured out.

WadeBarnes (Fri, 08 May 2020 12:39:06 GMT):
@jcourt, would you be able to submit a PR to the `von-network` documentation with some "Pro-Tips" for what you had to do to get it to work?

swcurran (Fri, 08 May 2020 14:59:25 GMT):
Good idea. We deliberately left the BCovrin sandboxes open, but it's bitten us a few times when people have decided to play with adding nodes to the ledger and breaking consensus. It's not a huge problem for us, but we'll be locking down the steward key at some point to prevent that on the open networks. If you do get that done, a PR or issue to von-network to document what you did would be appreciated.

swcurran (Fri, 08 May 2020 15:02:24 GMT):
A prod network is locked down such that the issuer has to go through an endorser to get their DID. There are other forms of DDOS attacks (e.g. bogus write requests, read requests), and those would be handled as in any enterprise. The network uses two network interfaces with one exposed to the dangerous world, and then other one only open to the other nodes on the network for internal communications. That should remain open and useable in the event of a DDOS.

conanoc (Mon, 11 May 2020 03:41:06 GMT):
Is there any scenario using Indy network to implement KYC(or Know Your Customer)? What I want specifically is a simple scenario like this. An identity owner (usually a person not an organization) signs a document with his key and then others can verify that the document is signed by that person. It seems hard to implement because we usually do not know who the identity owner is in real life and the anonymity of the SSI avoids correlation between the identity and the real person.

jcourt (Mon, 11 May 2020 03:55:48 GMT):
Submitted an PR at https://github.com/bcgov/von-network/issues/112

andrew.whitehead (Mon, 11 May 2020 05:04:31 GMT):
It sounds like you would need a common schema issued by trusted issuers (probably a white list of supported credential definitions), and at least one of the issuers would have to issue a credential confirming that the holder has signed the document. That issuer would know who they are, but presumably wouldn't share the information

jcourt (Mon, 11 May 2020 22:02:25 GMT):
Sorry my mind automatically converted PR to "Product Request" should have realised you would have use the term "Issue". I need to get my fork back into a proper state using branches and will look at updating that doc.

swcurran (Mon, 11 May 2020 22:19:58 GMT):
Help Wanted request: The BC Gov team had been planning on doing the initial work for converting the Indy CI/CD pipelines from Jenkins to GitHub Actions. However, priorities have shifted in the past few weeks and the resources expected to be available for that work are going to be working on other Trust over IP priorities for the foreseeable future. As such, that work is back to being unassigned. If anyone has the expertise and time to do that work, the community would greatly appreciate it.

conanoc (Tue, 12 May 2020 01:45:50 GMT):
Is there any reference document or discussion thread related to what you mentioned? When the issuer issues a credential confirming that the holder has signed the document, it could not verify that the holder really signed the document. The signature should be the one owned by the holder not the issuer. But, the idea of introducing a third party who knows the personal information but keeps it secret sounds good. Thanks.

andrew.whitehead (Tue, 12 May 2020 01:52:59 GMT):
I can't think of any relevant reading, I'm afraid.

priyashankar 1 (Tue, 12 May 2020 05:08:41 GMT):
Has joined the channel.

Audrius (Tue, 12 May 2020 06:43:12 GMT):
@conanoc I am trying to understand your mentioned KYC use case. Why do not get verifiable credential (VC) directly from trusted issuer, like a "gov passport authority" and use it to prove identity in KYC process?

conanoc (Tue, 12 May 2020 10:20:00 GMT):
Let's assume "gov passport authority" issued a passport to Alice. I am an insurance company and want to make a travel insurance contract with Alice digitally. It would be nice if Alice can use the digital passport or something from her SSI wallet to sign the contract. I can verify it's Alice who sent me the signed document by verifying her passport before receiving the document. But I cannot verify that the document has signed by her private key because Alice might have signed it using Bob's key.

conanoc (Tue, 12 May 2020 10:20:00 GMT):
@Audrius Let's assume "gov passport authority" issued a passport to Alice. I am an insurance company and want to make a travel insurance contract with Alice digitally. It would be nice if Alice can use the digital passport or something from her SSI wallet to sign the contract. I can verify it's Alice who sent me the signed document by verifying her passport before receiving the document. But I cannot verify that the document has signed by her private key because Alice might have signed it using Bob's key.

Audrius (Tue, 12 May 2020 11:13:25 GMT):
@conanoc I guess for a use case clarity we need to separate technical and legal parts of "sign the contract" meaning. From a technology and my point of view that could be: Alice is identified by "passport VC" and Insurance company could organize confirmation/signature of the contract (or just record in their DB) in a way they are happy with that. They know, that was Alice, which proved her identity. As example Alice’s “signature” could be issued VC by her. She is ok with the contract No123 from Insurance company xyz. That is technology part. The legal part to "sign the contract" could be more complicated, mainly it depends on a particular country law. I am not sure if Indy will cover all possible use cases from a law point of view. In this particular Alice case if law is requiring "contract signature" between supplier and customer, first of all, VC should be a part of legal regulation. Based on that any VC issued by Alice could be treated as a legal action. Without VC legislation signature by VC will be not legal. As per my opinion legal part is most challenging in “digital” KYC implementation. That is one of the reasons why EU with already legal “electronic signature” solution (eIDAS) proposing gateway approach for a quick solution between SSI and existing eIDAS. From a legal point of view If you sign VC with already legal e-signature your VC should be treated as a part of legal regulation. Sorry for the long text…

WadeBarnes (Tue, 12 May 2020 14:12:50 GMT):
@jcourt, Thanks for the contribution!

Audrius (Tue, 12 May 2020 15:59:16 GMT):
@swcurran BC gov is already using Verifiable Credentials in few cases. What is their legal status?

Audrius (Tue, 12 May 2020 15:59:16 GMT):
@swcurran BC gov is already using Verifiable Credentials in few cases. What are their legal status?

conanoc (Wed, 13 May 2020 05:28:35 GMT):
@Audrius I agree that the legal part is more complicated and vary from country to country. I am concerning the technical part for now. You mentioned that "Alice’s signature could be issued VC by her". But the problem is that Alice cannot issue a VC because she is not an endorser or a trustee in the network.

conanoc (Wed, 13 May 2020 05:28:35 GMT):
@Audrius I agree that the legal part is more complicated and varies from country to country. I am concerning the technical part for now. You mentioned that "Alice’s signature could be issued VC by her". But the problem is that Alice cannot issue a VC because she is not an endorser or a trustee in the network.

conanoc (Wed, 13 May 2020 05:40:05 GMT):
Without signing, what we can make sure is that Alice has visited the web site and read the document. She may have agreed or disagreed and marked it in the web site. But the DB record could verify nothing. This is a simple digital signing scenario and I might be wrong to title it as "KYC".

Audrius (Wed, 13 May 2020 07:19:08 GMT):
@conanoc Alice could issue "contract signature VC" without trustee and network. Maybe in this case use VC naming is confusing, however based on verifiable ID she could sign/prove her decision, for Insurance company.

PiyushSarwal (Wed, 13 May 2020 14:12:04 GMT):
Has joined the channel.

swcurran (Wed, 13 May 2020 14:14:41 GMT):
NOT A LAWYER :-) In most places I know of, a sufficient "digital signature" is a declaration of agreement by a known entity. More simply put, having a web session where the person "logged in" and agrees to sign (acknowledge by some act) is sufficient. So, if the session with the user can be verified by the provision of a proof from a trusted authority, that will be good enough, and much better than the existing techniques. Currently, what you do is login and then over share data such that the other party (insurance company) is "pretty certain" that you are the person you say you are. They mitigate their risk. With VCs in the game, they can collect far less data if they can trust from whence the credential was provided.

tsekwena (Wed, 13 May 2020 17:54:25 GMT):
Is there any method/tutorial i can follow to do a POC for a revoked credential and proof of non revocation ?

conanoc (Thu, 14 May 2020 01:51:22 GMT):
I agree that most of the web services work like that. Users login and do some action and type password once more if some action is sensitive, type secondary password or do fingerprint verification on their phones and so on. I'm just thinking about some other process using SSI, including digital signing if possible. @Audrius I didn't know that anyone can issue credentials. It will be great if we can just correlate the proof record of the passport VC and the credential issued by Alice. I suppose the two artifacts both signed, or can be signed, by the same key.

Audrius (Thu, 14 May 2020 08:16:59 GMT):
@conanoc I would recommend to check Aries RFC 0257: Private Credential Issuance. In relation to the "agreement signature" scenario could be worth to check 0234: Signature Decorator or RFC 0066: Non-Repudiable Signature for Cryptographic Envelope

filfwt (Thu, 14 May 2020 15:06:19 GMT):
Has joined the channel.

conanoc (Fri, 15 May 2020 05:04:38 GMT):
@ㅁ

conanoc (Fri, 15 May 2020 05:04:38 GMT):
@Audrius Thank you for the references.

Audrius (Fri, 15 May 2020 06:55:28 GMT):
@conanoc you are welcome :-)

swcurran (Fri, 15 May 2020 13:27:09 GMT):
Good references and a much better answer. I would add that attachments can also be signed, per https://github.com/hyperledger/aries-rfcs/tree/master/concepts/0017-attachments#signing-attachments A document signed with the same key as used to convey a proof should be a strong mechanism.

shantsum (Sat, 16 May 2020 13:47:26 GMT):
Has joined the channel.

anurkalem (Sat, 16 May 2020 15:56:37 GMT):
Hi everyone, I am senior student and I select performance testing of indy hyperledger. I running Getting Started code. I want to test performance of indy. First, I want to add some node to pool and try a lot getting schemas or verifying process. How can I add some nodes to pool? txn_genesis is only have 4 node configurations. Second, can you advise me any other scenerio for performance testing? Thanks for your information...

anurkalem (Sat, 16 May 2020 15:56:37 GMT):
Hi everyone, I am senior student and I select performance testing of indy hyperledger as my senior project. I running Getting Started code. I want to test performance of indy. First, I want to add some node to pool and try a lot getting schemas or verifying process. How can I add some nodes to pool? txn_genesis is only have 4 node configurations. Second, can you advise me any other scenerio for performance testing? Thanks for your information...

swcurran (Sun, 17 May 2020 23:55:41 GMT):
Join us tomorrow (Monday) morning (in the US and Canada) at 8AM Pacific/11AM Eastern for the Indy Contributors call. Topic for tomorrow is Revocation 2.0 and a couple of tech spikes that are happening that provide a couple of ways of doing bit array-based revocation. Zoom info: https://zoom.us/j/244779296

HarishKumarG (Mon, 18 May 2020 10:48:28 GMT):
Has joined the channel.

hamidm (Tue, 19 May 2020 18:02:05 GMT):
Hi, I have some issue with the setup of Libindy. I'm using both Ubuntu and MacOs and in both case the Libindy doesn't work, no matter, if I copy the library in the bin folder of the app. When installing in linux using apt, there is an issue with a package not existing any more linsodium18. Any way to get it to work? Thanks

Alex.Del (Thu, 21 May 2020 09:56:08 GMT):
Has joined the channel.

Alex.Del (Thu, 21 May 2020 09:56:09 GMT):
Hello everybody, After some trials of nym request, I found out that the role of the DID which signs the transactions is only checked after the COMMIT messages (and fails if it does not have the privilege to write on the ledger). Can anyone explain me why the check of roles is not done at the same moment than the signature check, it means before a node sends a PROPAGATE message? Thanks

lynn.bendixsen (Thu, 21 May 2020 16:42:41 GMT):
Hi @anurkalem, There is a guide that we use to add Nodes to Sovrin Networks. It is comprehensive, so you can probably skip to step 3.1.4 to start with and adjust the instructions to meet your needs. https://docs.google.com/document/d/18MNB7nEKerlcyZKof5AvGMy0GP9T82c4SWaxZkPzya4/edit#heading=h.9a4ud1gh7a6x

anurkalem (Thu, 21 May 2020 17:13:36 GMT):
Thank you a lot :-)

hyper-sunder (Fri, 22 May 2020 11:36:02 GMT):
Hi Guys, I am trying to create `Schema` and `SchemaDef` using **JAVA-VCX** wrapper library. I do get `InvalidSchemahandleException` exception when I try to get `SchemaId` from `schemaHandle`. Could anyone help me what I am messing up???

esplinr (Fri, 22 May 2020 17:42:42 GMT):
I put together a project update video to help people track the progress of Aries, Indy, and Ursa: https://www.youtube.com/watch?v=CQ2PP4qmEL4&feature=youtu.be This mostly covers in-flight projects and why I am less visible in community meetings than previously.

jameshiester (Fri, 22 May 2020 18:59:41 GMT):
is there documentation around production deployment/creating the genesis tx from scratch so we can control the keys?

jameshiester (Fri, 22 May 2020 19:12:01 GMT):
also, is there a separate chat for Aries, or does channel cover the full umbrella?

jameshiester (Fri, 22 May 2020 20:54:48 GMT):
also, looking at the email service implementation, can someone point me to how the api-key gets defined/injected into both the controller and the cloud agent?

swcurran (Fri, 22 May 2020 21:21:35 GMT):
Sounds like you are using ACA-Py - there is a channel #aries-cloudagent-python for that with the devs there (try @esune about email verification). There is also an #aries channel.

esune (Fri, 22 May 2020 21:28:50 GMT):
Responded in the #aries-cloudagent-python channel: https://chat.hyperledger.org/channel/aries-cloudagent-python?msg=mLbofp864nTZJxq2o

coolhand812 (Sun, 24 May 2020 03:55:44 GMT):
Has joined the channel.

coolhand812 (Sun, 24 May 2020 03:55:47 GMT):
Hello! Can I use Indy in a private network, like that of a university?

redongjun (Sun, 24 May 2020 06:18:31 GMT):
Has joined the channel.

rc004 (Tue, 26 May 2020 04:47:13 GMT):
Has joined the channel.

rc004 (Tue, 26 May 2020 04:48:10 GMT):
Hi! I was trying to use indy sdk with angular but getting this error I was working on a project where hyperledger indy sdk was used with angular, but I am getting this error bindings.js:203 Uncaught TypeError: exists is not a function at Function.getRoot (bindings.js:203) at bindings (bindings.js:82) at Object../node_modules/indy-sdk/src/indyBinding.js (indyBinding.js:1) at webpackrequire__ (bootstrap:78) at Object../node_modules/indy-sdk/src/index.js (index.js:1) at webpackrequire__ (bootstrap:78) at Module../src/app/_services/indy-new.service.ts (indy-new.service.ts:12) at webpackrequire__ (bootstrap:78) at Module../src/app/_services/index.ts (index.ts:3) at webpackrequire__ (bootstrap:78)

rc004 (Tue, 26 May 2020 04:57:36 GMT):
Hi! Is Indy sdk compatible with angular because I was trying to use indy sdk 1.15.0 with angular 7 by importing node module in angular but getting error bindings.js:203 Uncaught TypeError: exists is not a function at Function.getRoot (bindings.js:203) or is this due to version issue?

pknowles (Tue, 26 May 2020 07:10:51 GMT):
NEWSFLASH: After nearly 18 months in existence, the *Hyperledger Indy Semantics WG* will be closing shop for good on *June 1st*. We are currently in the process of setting up a new *Decentralized Semantics WG* at the *Trust over IP Foundation*. The mission and scope of the new group will be to define a data capture architecture consisting of immutable schema bases and interoperable overlays for Internet-scale deployment. For more information, check out the new wiki page at https://wiki.trustoverip.org/pages/viewpage.action?pageId=65746

fmalarflower (Tue, 26 May 2020 08:04:42 GMT):
Has joined the channel.

fmalarflower (Tue, 26 May 2020 08:04:43 GMT):
hi

fmalarflower (Tue, 26 May 2020 08:04:51 GMT):
Is Indy works on RHEL 7 ?

fmalarflower (Tue, 26 May 2020 08:44:47 GMT):
im getting error in my ubuntu

fmalarflower (Tue, 26 May 2020 08:44:55 GMT):
Get https://registry-1.docker.io/v2/: net/http: request canceled while waiting for connection (Client.Timeout exceeded while awaiting headers)

MartenMeijboom (Tue, 26 May 2020 10:43:35 GMT):
Has joined the channel.

MartenMeijboom (Tue, 26 May 2020 10:43:36 GMT):
This might be a dumb question, I'm doing a project for uni using hyperledger indy. We are using the NodeJs wrapper. While trying to get things up and running, we're following along with the to-do documents. First we create a pool ledger using createPoolLedgerConfig(poolName, poolConfig). Then we try to open the pool ledger using openPoolLedger(poolName, undefined). This results in a PoolLedgerTimeout (307) error. Anyone know why this happens? I've already checked, and the pool does exists.

SamuelKitavi (Tue, 26 May 2020 12:25:54 GMT):
Has joined the channel.

jakubkoci (Tue, 26 May 2020 15:36:01 GMT):
Hi, I see you mentioned Angular, so I guess you're trying to npm install indy-sdk and run an Angular app in the browser right? I'm afrad it won't be possible because indy-sdk is only for Node.js runtime because it relies on native libindy binary to be available in your operating system.

jakubkoci (Tue, 26 May 2020 15:40:06 GMT):
If you want to interact with indy from you Angular web app you would need to communicate via REST API with server-side Node.js application.

jakubkoci (Tue, 26 May 2020 15:40:06 GMT):
If you want to interact with indy from you Angular web app you would need to communicate via REST API with server-side Node.js application with indy-sdk dependency

Jintolus (Tue, 26 May 2020 18:15:29 GMT):
Has joined the channel.

rc004 (Wed, 27 May 2020 04:44:43 GMT):
Thanks a lot! Will try it via REST APIs

pknowles (Wed, 27 May 2020 05:20:49 GMT):
If you are interested in being included in the kick-off plans for the newly proposed *Decentralized Semantics WG* at the *Trust over IP Foundation*, please add your name and email address to the following list. We will then add your name to the wiki and send a calendar invite to your email address. Please note that your email address will not be added to the wiki for privacy reasons. https://drive.google.com/file/d/1XiE4IzeVke-tm0oCHC_GsKp2JgGuXIkZ/view?usp=sharing

tsekwena (Wed, 27 May 2020 12:05:16 GMT):

Clipboard - May 27, 2020 2:03 PM

tsekwena (Wed, 27 May 2020 12:05:29 GMT):
Hi Guys is there anyone who has worked with http://play-with-von.vonx.io/ It just freezes and does allow me to write anything after I add a new node {Image attached above}

shakeib98 (Wed, 27 May 2020 13:33:14 GMT):
Has joined the channel.

shakeib98 (Wed, 27 May 2020 13:33:15 GMT):
I have came from public blockchain eco-system and I find it hard to visualize a block in indy. Like is there a block or it's just sequence of transaction? So the word blockchain isn't really implemented. It's just a DLT. Please need comments over this. :)

antonio.jack (Wed, 27 May 2020 14:02:42 GMT):
Has joined the channel.

shakeib98 (Wed, 27 May 2020 15:08:42 GMT):
Also what is root hash in indy?

swcurran (Wed, 27 May 2020 15:31:15 GMT):
@WadeBarnes -^^^ You can also use https://labs.play-with-docker.com/ for the same purpose. Not sure what's up with Play with VON.

WadeBarnes (Wed, 27 May 2020 15:31:59 GMT):
Looking

WadeBarnes (Wed, 27 May 2020 15:44:16 GMT):
@tsekwena, It's back up and running now.

WadeBarnes (Wed, 27 May 2020 15:44:54 GMT):
I had to flush a bunch of orphaned containers.

tsekwena (Wed, 27 May 2020 15:45:01 GMT):
Thanks for Assistance guys

rowanshedden (Thu, 28 May 2020 02:46:34 GMT):
Has joined the channel.

rowanshedden (Thu, 28 May 2020 02:50:43 GMT):
Has anyone installed libindy on Mac OS Catalina? I upgraded last week from Sierra and now the libindy dylib fails with a missing dll. I suspect libsodium has been quarantined by Catalina and this is causing my issue. I have tried various things, such s re-compiling libsodium (v1.0.12) and re-building libindy. But still no go. Any advice gratefully accepted ....

randyshu (Thu, 28 May 2020 06:54:00 GMT):
Has joined the channel.

OllieBourneKC (Thu, 28 May 2020 07:50:10 GMT):
Has joined the channel.

OllieBourneKC (Thu, 28 May 2020 07:55:59 GMT):
Hi, I managed to build and run the indy-sdk on catalina last week following the Mac OS instructions. Maybe upgrading broke some of the packages installed with homebrew. I'm new to this so don't have any detailed knowledge about exactly what the issue might be but I would try going through the MacOS build instructions and ensuring the dylib is in your path if you haven't already.

OllieBourneKC (Thu, 28 May 2020 08:01:15 GMT):
Hey, I've been running through some of the "how tos" in node and noticed that there were a few errors and the node examples were missing for a couple of them. I've ported the missing ones from python to node and got some fixes for other examples. I'm happy to do a PR for these if it would be helpful. The contributing guide says to create a ticket but I didn't want to do that before confirming if it would be a welcome contribution. Let me know and I'll contribute when I have a moment.

RimAbdallah (Thu, 28 May 2020 13:53:00 GMT):
Has joined the channel.

RimAbdallah (Thu, 28 May 2020 13:53:01 GMT):
When a holder receive an issued credential than can be later revoked, To present a prrof it must construct the revocation state and access the tail file, as a result when i attempt to present a proof i get "Error when constructing revocation state -Invalid len of bytes representation for PoingG2". how can the probe access the tail file and what are the steps required for a credential check of non-revocation

ianco (Thu, 28 May 2020 17:53:47 GMT):
I had to re-install xcode after I upgraded to Catalina

aguel (Fri, 29 May 2020 08:55:14 GMT):
Has left the channel.

swcurran (Fri, 29 May 2020 18:49:50 GMT):
This week's Indy Contributors call will mostly be about how we can enable dynamic access to multiple Indy ledgers --- allowing an agent (for example, a Mobile Wallet app) to access multiple Indy ledgers in a single session with minimal effort from the agent owner. An example for those that have used the Streetcred and esatus wallets: using both Sovrin Staging and Sovrin Mainnet without having to switch the settings over and over. We'll also cover the latest progress on Revocation 2.0 work. The call is Monday, June 1 at 8AM US Pacific, dial-in information: https://zoom.us/j/244779296 Agenda can be found here: https://wiki.hyperledger.org/display/indy/2020-06-01+Indy+Contributors+Call Join us!

crector (Fri, 29 May 2020 20:03:29 GMT):
Has joined the channel.

esplinr (Fri, 29 May 2020 23:09:02 GMT):
The reason the contributing guide asks for an issue is to help evaluate if a contribution will be welcome before writing code. What you briefly describe here sounds useful, and if you've already got a PR ready you can just submit it. Also, the contributing guide is a bit out-of-date. We used to use a Jira-centric workflow, but now are now keeping our issues in GitHub. Thank you for your interest in improving Indy!

XKBHX (Mon, 01 Jun 2020 16:41:13 GMT):
Has joined the channel.

XKBHX (Mon, 01 Jun 2020 16:41:13 GMT):
Hey

tomislav (Mon, 01 Jun 2020 17:29:09 GMT):
I missed the call this morning. I would love to hear the recording if someone can update the wiki with a link. Thanks

swcurran (Mon, 01 Jun 2020 17:32:28 GMT):
It's there - the link is at the top of the page.

tomislav (Mon, 01 Jun 2020 17:33:40 GMT):
Whoops, I had to refresh. Thanks @swcurran

pknowles (Mon, 01 Jun 2020 19:26:32 GMT):
UPDATE: The newly proposed *Decentralized Semantics WG* at the *Trust over IP Foundation* will be reviewed by the *ToIP Steering Committee* on *Wednesday, June 10th*. To be kept in the loop, please add your name and email address to the following distribution list. We will then add your name to the wiki and send a calendar invite to your email address. Please note that your email address will not be added to the wiki for privacy reasons. https://drive.google.com/file/d/1XiE4IzeVke-tm0oCHC_GsKp2JgGuXIkZ/view?usp=sharing

StevenTCramer (Tue, 02 Jun 2020 14:55:15 GMT):
Has joined the channel.

CarloTravaglio (Tue, 02 Jun 2020 16:01:23 GMT):
Has joined the channel.

CarloTravaglio (Tue, 02 Jun 2020 16:01:23 GMT):
Hi! is SAP using indy? where can i read about it? thanks!

shubhamparasher (Wed, 03 Jun 2020 11:22:24 GMT):
Has joined the channel.

shubhamparasher (Wed, 03 Jun 2020 11:22:25 GMT):
hi,

shubhamparasher (Wed, 03 Jun 2020 11:24:53 GMT):
I was trying to call - indy.issuerCreateCredential - which requires transcriptCredValues as one of the parameters. I am confused or not sure about how the raw values are encoded as in this -- let transcriptCredValues = { "first_name": {"raw": "Alice", "encoded": "1139481716457488690172217916278103335"}, "last_name": {"raw": "Garcia", "encoded": "5321642780241790123587902456789123452"}, "degree": {"raw": "Bachelor of Science, Marketing", "encoded": "12434523576212321"}, "status": {"raw": "graduated", "encoded": "2213454313412354"}, "ssn": {"raw": "123-45-6789", "encoded": "3124141231422543541"}, "year": {"raw": "2015", "encoded": "2015"}, "average": {"raw": "5", "encoded": "5"} };

hyper-sunder (Wed, 03 Jun 2020 15:35:38 GMT):
Hi All,

hyper-sunder (Wed, 03 Jun 2020 15:37:26 GMT):
Hi All, I am trying to use `JAVA-VCX Wrapper` library to imitate the sample given for `nodejs` in the repo. My program is getting crashed at the line `IssuerApi.issuerSendCredential(credentialHandle, connectionHandle).get()`. Could anyone help me why it happening???

hyper-sunder (Wed, 03 Jun 2020 15:38:29 GMT):
The error is ::: ``` # A fatal error has been detected by the Java Runtime Environment: # # SIGSEGV (0xb) at pc=0x00007fff64f71c92, pid=33239, tid=0x0000000000009d33 # # JRE version: Java(TM) SE Runtime Environment (8.0_131-b11) (build 1.8.0_131-b11) # Java VM: Java HotSpot(TM) 64-Bit Server VM (25.131-b11 mixed mode bsd-amd64 compressed oops) # Problematic frame: # C [libsystem_platform.dylib+0x1c92] _platform_strlen+0x12 # # Failed to write core dump. Core dumps have been disabled. To enable core dumping, try "ulimit -c unlimited" before starting Java again # # An error report file with more information is saved as: # /path/hs_err_pid33239.log Compiled method (nm) 228533 618 n 0 com.sun.jna.Native::getStringBytes (native) total in heap [0x000000010be10bd0,0x000000010be10f40] = 880 relocation [0x000000010be10cf8,0x000000010be10d38] = 64 main code [0x000000010be10d40,0x000000010be10f38] = 504 oops [0x000000010be10f38,0x000000010be10f40] = 8 # # If you would like to submit a bug report, please visit: # http://bugreport.java.com/bugreport/crash.jsp # The crash happened outside the Java Virtual Machine in native code. # See problematic frame for where to report the bug. ```

hyper-sunder (Thu, 04 Jun 2020 05:13:17 GMT):
Could anyone help me on this ^^^^ ?

shubhamparasher (Thu, 04 Jun 2020 06:28:15 GMT):
Please help me out wit this one if anybody hava any idea about this? Are there mutiple ways for doing the same(raw values to coded number)? If yes, which is the most commonly used? Also where can i find additional info on this thing is why we are encoding the raw values.I am doing the development in Nodejs using indy-sdk.

thomas_kim (Thu, 04 Jun 2020 08:55:12 GMT):
https://github.com/hyperledger/indy-sdk/pull/2160 You would like to apply this PR that fixes an issue in the Issuer side.

syngin (Fri, 05 Jun 2020 01:03:05 GMT):
Not an expert but, are the IPs and ports in the genesis file accessible from your program.

jameshiester (Fri, 05 Jun 2020 03:00:12 GMT):
it looks like the link for the docker directions is broken for starting a pool?

shubhamparasher (Fri, 05 Jun 2020 06:32:59 GMT):
Hi All, I was trying to call - indy.issuerCreateCredential - which requires transcriptCredValues as one of the parameters. I am confused or not sure about how the raw values are encoded as in this -- let transcriptCredValues = { "first_name": {"raw": "Alice", "encoded": "1139481716457488690172217916278103335"}, "last_name": {"raw": "Garcia", "encoded": "5321642780241790123587902456789123452"}, "degree": {"raw": "Bachelor of Science, Marketing", "encoded": "12434523576212321"}, "status": {"raw": "graduated", "encoded": "2213454313412354"}, "ssn": {"raw": "123-45-6789", "encoded": "3124141231422543541"}, "year": {"raw": "2015", "encoded": "2015"}, "average": {"raw": "5", "encoded": "5"} }; shubhamparasher shubhamparasher 11:58 AM Please help me out wit this one if anybody hava any idea about this? Are there mutiple ways for doing the same(raw values to coded number)? If yes, which is the most commonly used? Also where can i find additional info on this thing is why we are encoding the raw values.I am doing the development in Nodejs using indy-sdk.

shubhamparasher (Fri, 05 Jun 2020 06:40:29 GMT):
Hi All, I am building nodejs application and trying to call - indy.issuerCreateCredential for issuing credentials - which requires transcriptCredValues as one of the parameters. Could anyone help me on how the raw values corresponding the credentials are encoded as in this* -- let transcriptCredValues = { "first_name": {"raw": "Alice", "encoded": "1139481716457488690172217916278103335"}, "last_name": {"raw": "Garcia", "encoded": "5321642780241790123587902456789123452"}, "degree": {"raw": "Bachelor of Science, Marketing", "encoded": "12434523576212321"}, "status": {"raw": "graduated", "encoded": "2213454313412354"}, "ssn": {"raw": "123-45-6789", "encoded": "3124141231422543541"}, "year": {"raw": "2015", "encoded": "2015"}, "average": {"raw": "5", "encoded": "5"} }; Are there multiple ways for doing the same(raw values to encoded number)? If yes, which is the most commonly used? Also, is there somewhere, where I can find additional info on this thing i.e. why encoding is required for the raw values. I am doing the development in Nodejs using indy-SDK.

SytseOegema (Fri, 05 Jun 2020 13:57:11 GMT):
Has joined the channel.

SytseOegema (Fri, 05 Jun 2020 14:04:17 GMT):
I don't know how the values in your example are created. But acording to the indy sdk how to issue a credential it is the decimal representation of the SHA256 hash value. See: https://github.com/hyperledger/indy-sdk/blob/master/docs/how-tos/issue-credential/java/step4.java In that example there is "male". The SHA256 hex value of "male" is 0d248e82c62c9386878327d491c762a002152d42ab2c391a31c44d9f62675ddf Converting the hex value to decimal results in 5944657099558967239210949258394887428692050081607692519917050011144233115103.

jameshiester (Sat, 06 Jun 2020 00:52:18 GMT):
LINK : fatal error LNK1181: cannot open input file 'indy.dll.lib' [C:\apps\indy-network\app\client\node_modules\indy-sdk\build\indynodejs.vcxproj] i'm getting this error when trying to install the node.js wrapper?

jameshiester (Sat, 06 Jun 2020 01:08:58 GMT):
I'm trying to install on windows

jameshiester (Sat, 06 Jun 2020 01:28:56 GMT):
How do I get indy.dll.lib?

jameshiester (Sat, 06 Jun 2020 01:39:58 GMT):
nevermind, looks like nodejs wrapper doesn't work with v1.15 yet because the indy.dll.lib file has been removed?

jameshiester (Sat, 06 Jun 2020 01:40:22 GMT):
I was able to install with 1.14.3

jameshiester (Sat, 06 Jun 2020 01:40:45 GMT):
is there a PR to update the nodejs wrapper?

ap (Sun, 07 Jun 2020 08:37:34 GMT):
I am building the react-native arise SDK and facing one issue.The issue is, In java whenever we perform any wallet related transaction it's taking too much time to execute that transaction while in iOS it's very less time to execute the same transaction. Is anyone facing same issue? Or may I am doing something wrong while integrating indy SDK in android

jameshiester (Sun, 07 Jun 2020 16:54:21 GMT):
are there utils for converting seed (string) => did/verkey? (looking specifically for node.js implementation)

hyper-sunder (Sun, 07 Jun 2020 17:01:13 GMT):
Thanks a lot @thomas_kim . Let me try

jameshiester (Sun, 07 Jun 2020 17:19:14 GMT):
would it be correct to use: createKey(seed) and then the first 16 bits would be the did?

MaxSor (Mon, 08 Jun 2020 14:46:50 GMT):
Has joined the channel.

Primordium (Mon, 08 Jun 2020 15:24:20 GMT):
Has joined the channel.

rowanshedden (Tue, 09 Jun 2020 02:01:14 GMT):
thanks for the replies. I ended up rebuilding libindy and creating a symlink to a newly installed libsodium. The fixed the problem for me. The libsodium was installed under the covers by the mac.build.sh. I had to remove all traces of the old libsodium first.

CarloTravaglio (Tue, 09 Jun 2020 19:46:28 GMT):
Hi! is SAP using indy? where can i read about it? thanks!

Mopack (Wed, 10 Jun 2020 03:18:52 GMT):
Has joined the channel.

nc-crtr_linx (Wed, 10 Jun 2020 13:50:31 GMT):
Has joined the channel.

nc-crtr_linx (Wed, 10 Jun 2020 13:50:32 GMT):
Hi, is there any documentation on how much space an indy installation can take up?

jameshiester (Thu, 11 Jun 2020 13:37:07 GMT):
Does the request to get validator info have to be signed? I'm noticing it is getting signed in the green light app

sheru (Thu, 11 Jun 2020 13:43:08 GMT):
Good Morning, Hope everyone doing well. I am trying to play around with aries-mobileagent-xamarin I have setup the mediator and successfully make a connection with cloud agent by scan the qr code. the cloud agent offer a credential to the mobile agent. When I am trying to request for accept the credential from the mobile agent I am getting the following error... ```**Hyperledger.Indy.PoolApi.PoolConfigNotCreatedException:** 'The requested pool cannot be opened because it does not have an existing configuration.'``` Can anyone please help me what I am missing , Thank you...

swcurran (Thu, 11 Jun 2020 14:49:14 GMT):
Not sure what you mean, so I'll give two answers. If you are looking for detailed status info on a validator node, yes, that has to be signed (permissioned). If you are talking about doing a read operation from the ledger, no, that does not have to be signed (public).

andrew.whitehead (Thu, 11 Jun 2020 15:35:44 GMT):
Validator info includes some sensitive information about the nodes like what package versions they're running, so the request can only be submitted by (I think) a steward

jameshiester (Thu, 11 Jun 2020 21:55:50 GMT):
do the same wallet credentials always open the same wallet handle? If so, does the wallet have to be open to call signAndSubmitRequest?

shubhamparasher (Fri, 12 Jun 2020 08:07:19 GMT):
yes, the same credentials are required to open a wallet which were used while creating it. The wallet handle returned after open wallet is required to sign and submit any transaction or request, hence the wallet needs to be open at the time of submitting any request that requires signing with private key.

swcurran (Fri, 12 Jun 2020 19:56:38 GMT):
This week's Indy Contributors call will be about the progress in Ursa on revocation 2.0 and BBS+ and what we can do in Indy to prepare. Mike Lodder will be joining us to cover that. I've also been doing lots lately with Sovrin and the Sovrin Mainnet and I'll give a draft overview of the important things I'd like to see happening Indy for Sovrin and other Indy deployments. The call is Monday, June 15 at 8AM US Pacific, dial-in information: https://zoom.us/j/244779296 Agenda can be found here: https://wiki.hyperledger.org/display/indy/2020-06-15+Indy+Contributors+Call Join us!

jameshiester (Fri, 12 Jun 2020 20:38:27 GMT):
To clarify, I meant if the handle number is always the same? I'm assuming that this is not the case

jameshiester (Fri, 12 Jun 2020 20:43:15 GMT):
Has there been any interest in a network visualizer? I've been working on a node.js/typescript/react implementation similar to BC Gov's work. I was thinking it would be cool to see a visual representation of the transactions, something like a force directed chart: https://www.amcharts.com/demos/collapsible-force-directed-tree/. The idea would be to show the nyms, schemas, and cred_defs as nodes, with each type as a different color, and then link them together using the transaction history. Might be an interesting way to illustrate the network of trust

HichamTAHIRI (Sun, 14 Jun 2020 15:54:03 GMT):
Has joined the channel.

PatrikStas (Tue, 16 Jun 2020 12:29:25 GMT):
There's also http://indyscan.io/ which I've made, but it's similar to BC Gov's explorer. No graphs unfortunately. I've started working on adding tracking state of DIDs, but didn't have much time lately to finish that up.

Primordium (Tue, 16 Jun 2020 13:08:25 GMT):
when creating a DID document, how can one add a "parameter type message" ? example curl -X POST 'http://localhost:9080/1.0/register?driverId=driver-universalregistrar%2Fdriver-did-sov' -d '{"options":{"network":"danube"}, "didDocument":{"message":"something something"}}'

semosemo941 (Thu, 18 Jun 2020 09:45:41 GMT):
Has joined the channel.

x0axz (Thu, 18 Jun 2020 10:54:31 GMT):
Has joined the channel.

MALodder (Thu, 18 Jun 2020 15:40:50 GMT):
Interesting read https://eprint.iacr.org/2020/724.pdf

jameshiester (Thu, 18 Jun 2020 16:03:52 GMT):

Clipboard - June 18, 2020 11:02 AM

jameshiester (Thu, 18 Jun 2020 16:04:50 GMT):
the idea is you can hover over the circle to learn more about the details

sheru (Thu, 18 Jun 2020 22:37:03 GMT):
Hey all, I am getting this error while tried to accept credential offer from the ACA-PY cloudagent to xamrin mobile agent. ```bcgovimages/aries-cloudagent:py36-1.14-0_0.4.2``` Can anyone please suggest me how to solve this. Thanks for your valuable time. Please share if there is any doc available regarding this. Thank you again.

sheru (Thu, 18 Jun 2020 22:37:03 GMT):
Hey all, I am getting this error while tried to accept credential offer from the ACA-PY cloudagent (bcgovimages/aries-cloudagent:py36-1.14-0_0.4.2) to xamrin mobile agent. ```Hyperledger.Indy.IndyException Message=An unmapped error with the code '309' was returned by the SDK. Source=mscorlib StackTrace: at Hyperledger.Aries.Ledger.DefaultLedgerService+<>c__DisplayClass2_0.g__LookupDefinition|0 () [0x001ad] in :0 at Polly.NoOp.NoOpEngine.ImplementationAsync[TResult] (System.Func`3[T1,T2,TResult] action, Polly.Context context, System.Threading.CancellationToken cancellationToken, System.Boolean continueOnCapturedContext) [0x0007a] in <49153d3d67fd4ef18b1c5d2be572af7a>:0 at Polly.AsyncPolicy`1[TResult].ExecuteAsync (System.Func`3[T1,T2,TResult] action, Polly.Context context, System.Threading.CancellationToken cancellationToken, System.Boolean continueOnCapturedContext) [0x000b8] in <49153d3d67fd4ef18b1c5d2be572af7a>:0 at Hyperledger.Aries.Ledger.DefaultLedgerService.LookupDefinitionAsync (Hyperledger.Aries.Agents.IAgentContext agentContext, System.String definitionId) [0x000a1] in :0 at Hyperledger.Aries.Features.IssueCredential.DefaultCredentialService.CreateRequestAsync (Hyperledger.Aries.Agents.IAgentContext agentContext, System.String credentialId) [0x00261] in :0 at Osma.Mobile.App.ViewModels.Credentials.CredentialViewModel.b__10_0 () [0x0035f] in <52755e517d5e4de5b59ebea70b11124f>:0 at Osma.Mobile.App.ViewModels.Credentials.CredentialViewModel.b__10_0 () [0x00430] in <52755e517d5e4de5b59ebea70b11124f>:0 at System.Runtime.CompilerServices.AsyncMethodBuilderCore+<>c.b__7_0 (System.Object state) [0x00000] in :0 at Android.App.SyncContext+<>c__DisplayClass2_0.b__0 () [0x00000] in :0 at Java.Lang.Thread+RunnableImplementor.Run () [0x00008] in :0 at Java.Lang.IRunnableInvoker.n_Run (System.IntPtr jnienv, System.IntPtr native__this) [0x00008] in :0 at (wrapper dynamic-method) Android.Runtime.DynamicMethodNameCounter.47(intptr,intptr)``` Can anyone please suggest me how to solve this. Thanks for your valuable time. Please share if there is any doc available regarding this. Thank you again.

lijiachuan (Sat, 20 Jun 2020 01:54:39 GMT):
Hi all, I am using Aries Framework to build the agent, and I always encounter the first time failed issue, below are two common happened scenarios, does anyone also encounter this before? Thanks. 1. The first time when I create a new schema, then it returned Ledger operation rejected 500 error, then when I create again, then it is success. 2. The first time when I retrieve all schemas, then it returned The ledger message is unknown or malformed. 500 error, then when I refresh the page, the schemas were returned successfully

andrew.whitehead (Sat, 20 Jun 2020 16:42:25 GMT):
Which aries framework?

pouya (Sun, 21 Jun 2020 09:59:10 GMT):
Has joined the channel.

lijiachuan (Sun, 21 Jun 2020 15:56:46 GMT):
hi Andrew, the dot net framework.

x0axz (Mon, 22 Jun 2020 19:07:16 GMT):
hi @swcurran @tomislav, here are few questions: 1) Where does private and public key stores? in wallet? not on ledger? 2) What exactly stores in Wallets? 3) @swcurran , You once said "No data is sent to the Blockchain. Ever. It is only used for DIDs, schema/credential metadata and revocation information." What do u mean by metadata? 4) Can you please explain the credential verification flow, issuer-verifer-holder? 4a) Without having information about credentials (it isn't stored on ledger, only store in holder wallet), how will verifer confirms (verify) that these credentials are same as they were issued by issuer? I mean what if holder received credentials, changes later, how will verifer/issuer know that?

swcurran (Tue, 23 Jun 2020 00:26:12 GMT):
Answered on the #aries channel.

swcurran (Tue, 23 Jun 2020 04:35:36 GMT):
This week's Indy Contributors call will be about things needed for operational support of an Indy Network like Sovrin -- monitoring, metrics, etc -- with the goal of getting some collaboration to go in building out those tools. The call is Tuesday, June 23 at 8AM US Pacific. 15:00 UTC, dial-in information: https://zoom.us/j/244779296 Agenda can be found here: https://wiki.hyperledger.org/display/indy/2020-06-23+Indy+Contributors+Call Join us!

Primordium (Tue, 23 Jun 2020 13:55:21 GMT):
I'm using the von-network to test communication, is there a way to have von-network running in two diferent machines while using the same blockchain? So both machines have the same record

andrew.whitehead (Tue, 23 Jun 2020 16:39:53 GMT):
@brentzundel It looks like the latest indy-node brings in indy-plenum 1.12.3 which still depends on indy-crypto

andrew.whitehead (Tue, 23 Jun 2020 16:50:35 GMT):
I added the current master to our requirements.txt as `git+https://github.com/hyperledger/indy-plenum.git@a5941cb9d9faea4601ba2ad4e8c5fc070d5075df#egg=indy-plenum`, and with ursa-0.3.2 preinstalled it seems happy

brentzundel (Tue, 23 Jun 2020 20:09:40 GMT):
the plenum version I bump to in my indy-node PR removing indy-crypto is 1.13.0.dev1024. The latest stable plenum release still uses indy-crypto

MrWoody (Tue, 23 Jun 2020 23:29:21 GMT):
Has joined the channel.

MrWoody (Tue, 23 Jun 2020 23:29:21 GMT):
Hello all

MrWoody (Tue, 23 Jun 2020 23:30:02 GMT):
I'm playing with indi_pool and I'm wondering if I can get a nice web viewer for the ledger as we do in other places like https://github.com/cloudcompass/ToIPLabs/blob/master/src/indy-material/nodejs/docker-compose.yml

MrWoody (Tue, 23 Jun 2020 23:30:24 GMT):
Sorry if the question is inappropriate for this channel. If such a channel exists, please do share! :)

MikeRichardson (Wed, 24 Jun 2020 06:32:30 GMT):
I am trying to create a payment address using indy-cli but I keep getting unknown method. >payment-address create payment_method=sov >Unknown payment method sov

MikeRichardson (Wed, 24 Jun 2020 06:32:30 GMT):
I am trying to create a payment address for the Sovrin Staging Net using indy-cli but I keep getting unknown method. >payment-address create payment_method=sov >Unknown payment method sov

swcurran (Wed, 24 Jun 2020 16:46:20 GMT):
If you are talking about the the Sovrin ledger - indyscan.io. If you are talking about using it on your own ledger, the von-network browser can be deployed to point at any ledger. I think there instructions in the repo for that -- let us know if you can't find that.

MrWoody (Wed, 24 Jun 2020 17:11:24 GMT):
The von-network is just what I needed. Thank you!!

MrWoody (Wed, 24 Jun 2020 17:14:32 GMT):
If a startup wishes to use SSI using Hyperledger Indy, what is the recommended approach? a) Create it's own ledger using a pool of Indy nodes (on company hardware or somewhere in the cloud), or b) Use a central standard provided ledger (such as Sovrin or other)?

lynn.bendixsen (Wed, 24 Jun 2020 17:44:08 GMT):
Hi Mike, It looks like the plugin is not loaded for you. If that is the case try adding something like this to your cli config file `"plugins": "/Users/lynnbendixsen/github/libsovtoken/libsovtoken/target/debug/libsovtoken.so:sovtoken_init"` or run the load-plugin command from the CLI to load it each time you start up.

swcurran (Wed, 24 Jun 2020 20:48:24 GMT):
The answer depends on how interoperable you want to be. If you will have a closed systems of agents and users, then you can run your own. You can definitely do that for developers and for limited proof of concepts. If you want a wider ecosystem, and want to (for example) allow users to bring their own mobile wallet (e.g. Trinsic, esatus or Evernym), you have to use a ledger that they support (or convince them to support your network -- which might be tough :-) ). In that case, Sovrin is the place to go. Oh, and Sovrin would work for the "closed ecosystem" as well.

MrWoody (Thu, 25 Jun 2020 00:28:25 GMT):
Thank you again for sharing your knowledge.

raish (Thu, 25 Jun 2020 07:39:55 GMT):
Has joined the channel.

raish (Thu, 25 Jun 2020 07:43:58 GMT):
Hi - some newbie questions - is there an implementation (library/framework etc) - for using a wallet in a mobile phone? The intent is that the user is truly in-charge of the identity... thus, has the wallet in their mobile app (rather that being stored in some backend). We want to associate users biometeric idendity (face scan) to use the wallet. this leads to another question - are there ideas/implementations in this case (wallet in mobile) for identity (wallet) recovery?

raish (Thu, 25 Jun 2020 07:45:03 GMT):
another set of questions - re identity consumption ... we want a website to be able to 'consume' a SSI created on Indy .... is there some material to read on how to implement this (protocols, any ref implementations) etc.

romeu (Thu, 25 Jun 2020 07:53:12 GMT):
Has joined the channel.

martinezg (Thu, 25 Jun 2020 07:59:27 GMT):
Has joined the channel.

MikeRichardson (Thu, 25 Jun 2020 08:00:18 GMT):
thanks. I'll give it a try.

vela (Thu, 25 Jun 2020 08:22:58 GMT):
Has joined the channel.

rileyphughes (Thu, 25 Jun 2020 13:05:06 GMT):
To add to Stephen's answer, esatus and Trinsic wallets both support adding your own network via genesis txn file. So as long as the users in your closed ecosystem are highly technical, that is an approach as well 😊

rileyphughes (Thu, 25 Jun 2020 13:05:40 GMT):
Frankly I'd recommend just using Sovrin because it's already there, and there's no need to do any additional work to get it up and running and stable.

rileyphughes (Thu, 25 Jun 2020 13:07:49 GMT):
Yes, the Trinsic wallet is an edge wallet with backup. We built the framework used to make the app and contributed it HL as #aries-framework-dotnet. An open source reference mobile app using that framework is found at https://github.com/hyperledger/aries-mobileagent-xamarin

jameshiester (Fri, 26 Jun 2020 02:44:44 GMT):
are these fees still accurate for sovrin? https://sovrin.org/issue-credentials/

jameshiester (Fri, 26 Jun 2020 02:46:22 GMT):
besides spam and possibly less confidence in the identity of the issuers, what are the cons of allowing open did registration like in the bc gov application?

lijiachuan (Fri, 26 Jun 2020 15:12:46 GMT):
hi @andrew.whitehead , do you have any suggestion about this issue? Thanks.

andrew.whitehead (Fri, 26 Jun 2020 15:24:15 GMT):
Sorry, I don't have experience developing with the .net framework, seems like it could be signing ledger requests before the DID is registered perhaps

jameshiester (Fri, 26 Jun 2020 16:15:04 GMT):
This would be very helpful: https://jira.hyperledger.org/browse/IS-903

andrew.whitehead (Fri, 26 Jun 2020 16:18:20 GMT):
There's a version of that in aca-py using pynacl: https://github.com/hyperledger/aries-cloudagent-python/blob/7bdae139195561449fcd636d8a4934ee47b51a9d/aries_cloudagent/wallet/crypto.py#L77

jameshiester (Fri, 26 Jun 2020 16:22:15 GMT):
yeah, I would like to see it in the base sdk so it can be used with the other wrappers too, like nodejs. Makes me nervous as DID methods evolve there would be multiple places to maintain the same basic function.

jameshiester (Fri, 26 Jun 2020 16:22:56 GMT):
I think the same function was also implemented in von-network, so there's already a lot of duplication happening

andrew.whitehead (Fri, 26 Jun 2020 16:23:47 GMT):
indy-shared-rs also exposes a rust version

andrew.whitehead (Fri, 26 Jun 2020 16:23:47 GMT):
indy-shared-rs also exposes a rust version using ursa, not libsodium

vela (Mon, 29 Jun 2020 13:44:10 GMT):
Hi, according to the documentation Indy can reach consensus with 1/3 +1 of the nodes available, but what I cannot find is if this also true when all the nodes are in the genesis file. In our tests with 6 Indy nodes, being all of them in the genesis file, if I stop 2 of them I can read from the blockchain but when I try to write (for example creating an attrib transaction or updating a DID endpoint) a timeout is raised using indy-sdk. Shall all the nodes in the gensis file be available, so only additional nodes follow the 1/3 +1 rule or we just made a mistake in our tests?

swcurran (Mon, 29 Jun 2020 14:26:07 GMT):
To handle 2 faults, you need 7 nodes (f=2, 3f+1 = 7). So your 6 node network can only handle 1 fault. The genesis file is not relevant.

Alexi (Mon, 29 Jun 2020 14:33:41 GMT):
does anyone know what happened to the IOS wrappers? https://jenkins.hyperledger.org/ is not available anymore and the last build on https://repo.sovrin.org/ios/libindy/master/indy-objc/ is two years old

tomislav (Tue, 30 Jun 2020 14:19:27 GMT):
I don't think it's been maintained actively :(

romeu (Wed, 01 Jul 2020 07:12:44 GMT):
Mmmm so there is no native implementation of the wrappers on iOS / Android actively developed?

shubhamparasher (Wed, 01 Jul 2020 07:35:12 GMT):
Hi All,

shubhamparasher (Wed, 01 Jul 2020 07:50:47 GMT):
Hi All, I am facing an issue while calling indy.verifierVerifyProof -- error is -- AnoncredsProofRejected .. I am using nodejs indy SDK for development. Also can some help in understanding how the credentials are maintained in the wallet. For instance, if a user gets credentials from two trusted anchors corresponding to their credentials definition, where both credentials defn. schema includes a common field or attribute, how are then that common credentials stored in wallet? for example, if grade attribute is common in both the credentials defn. will both be stored in a wallet with mapping with the corresponding trusted anchor. Now if a third party (verifier) asks for the proof of grade will the grade will be picked for credentials defn required in the proof request by verifier or will indy.verifierVerifyProof -- throws -- AnoncredsProofRejected kind of error? Thanks.

x0axz (Wed, 01 Jul 2020 16:26:48 GMT):
Hi, on my Issuer Web Agent, which is running on Amazon EC2 (e.g. IP is 42.56.822.342), I'm using these commands to build & run pool in docker: ```docker build --build-arg pool_ip=127.0.0.1 -f indy-pool.dockerfile -t indy_pool . docker run -itd -p 127.0.0.1:9701-9708:9701-9708 indy_pool``` This Genesis file is being used on Issuer Web Agent: https://github.com/hyperledger/aries-framework-dotnet/blob/master/samples/aspnetcore/pool_genesis.txn I'm able to Define Schema/Credentials, Issue Credentials e.t.c but when I try to connect from Holder (Osma) with Web Agent, I get this error on Osma Mobile App. ```Hyperledger.Indy.PoolApi.PoolConfigNotCreatedException: 'The requested pool cannot be opened because it does not have an existing configuration.'``` and the reason is that the configuration is not proper done. What will be the right configuration needed to run a pool & nodes on server (Issuer Web Agent) & phone (osma/holder)? Kindly guide me!!!

x0axz (Wed, 01 Jul 2020 16:26:48 GMT):
Hi all, @swcurran @andrew.whitehead @WadeBarnes, on my Issuer Web Agent, which is running on Amazon EC2 (e.g. IP is 42.56.822.342), I'm using these commands to build & run pool in docker: ```docker build --build-arg pool_ip=127.0.0.1 -f indy-pool.dockerfile -t indy_pool . docker run -itd -p 127.0.0.1:9701-9708:9701-9708 indy_pool``` This Genesis file is being used on Issuer Web Agent: https://github.com/hyperledger/aries-framework-dotnet/blob/master/samples/aspnetcore/pool_genesis.txn I'm able to Define Schema/Credentials, Issue Credentials e.t.c but when I try to connect from Holder (Osma) with Web Agent, I get this error on Osma Mobile App. ```Hyperledger.Indy.PoolApi.PoolConfigNotCreatedException: 'The requested pool cannot be opened because it does not have an existing configuration.'``` and the reason is that the configuration is not proper done. What will be the right configuration needed to run a pool & nodes on server (Issuer Web Agent) & phone (osma/holder)? Kindly guide me!!!

x0axz (Wed, 01 Jul 2020 16:26:48 GMT):
Hi all, @swcurran @andrew.whitehead @WadeBarnes, on my Issuer Web Agent, which is running on Amazon EC2 (e.g. IP is 42.56.822.342), I'm using these commands to build & run pool in docker: ```docker build --build-arg pool_ip=127.0.0.1 -f indy-pool.dockerfile -t indy_pool . docker run -itd -p 127.0.0.1:9701-9708:9701-9708 indy_pool``` This Genesis file being used on Issuer Web Agent: https://github.com/hyperledger/aries-framework-dotnet/blob/master/samples/aspnetcore/pool_genesis.txn I'm able to Define Schema/Credentials, Issue Credentials e.t.c but when I try to connect from Holder (Osma) with Web Agent, I get this error on Osma Mobile App. ```Hyperledger.Indy.PoolApi.PoolConfigNotCreatedException: 'The requested pool cannot be opened because it does not have an existing configuration.'``` and the reason is that the configuration is not proper done. What will be the right configuration needed to run a pool & nodes on server (Issuer Web Agent) & phone (osma/holder)? Kindly guide me!!!

x0axz (Wed, 01 Jul 2020 16:26:48 GMT):
Hi all, @swcurran @andrew.whitehead @WadeBarnes, on my Issuer Web Agent, which is running on Amazon EC2 (e.g. IP is 42.56.822.342), I'm using these commands to build & run pool in docker: ```docker build --build-arg pool_ip=127.0.0.1 -f indy-pool.dockerfile -t indy_pool . docker run -itd -p 127.0.0.1:9701-9708:9701-9708 indy_pool``` This Genesis file being used in Issuer Web Agent: https://github.com/hyperledger/aries-framework-dotnet/blob/master/samples/aspnetcore/pool_genesis.txn I'm able to Define Schema/Credentials, Issue Credentials e.t.c but when I try to connect from Holder (Osma) with Web Agent, I get this error on Osma Mobile App. ```Hyperledger.Indy.PoolApi.PoolConfigNotCreatedException: 'The requested pool cannot be opened because it does not have an existing configuration.'``` and the reason is that the configuration is not proper done. What will be the right configuration needed to run a pool & nodes on server (Issuer Web Agent) & phone (osma/holder) and I want same configuration for both issuer & holder? Kindly guide me!!!

x0axz (Wed, 01 Jul 2020 16:26:48 GMT):
Hi all, @swcurran @andrew.whitehead @WadeBarnes @ianco , on my Issuer Web Agent, which is running on Amazon EC2 (e.g. IP is 42.56.822.342), I'm using these commands to build & run pool in docker: ```docker build --build-arg pool_ip=127.0.0.1 -f indy-pool.dockerfile -t indy_pool . docker run -itd -p 127.0.0.1:9701-9708:9701-9708 indy_pool``` This Docker files being used in Issuer Web Agent: https://github.com/hyperledger/indy-sdk/blob/master/ci/indy-pool.dockerfile https://github.com/Yoroitchi/AriesWebApp/blob/Issuer7000/AriesWebApp/docker/dockerfile This Genesis file being used in Issuer Web Agent: https://github.com/hyperledger/aries-framework-dotnet/blob/master/samples/aspnetcore/pool_genesis.txn I'm able to Define Schema/Credentials, Issue Credentials e.t.c but when I try to connect from Holder (Osma) with Web Agent, I get this error on Osma Mobile App. ```Hyperledger.Indy.PoolApi.PoolConfigNotCreatedException: 'The requested pool cannot be opened because it does not have an existing configuration.'``` and the reason is that the configuration is not proper done. What will be the right configuration needed to run a pool & nodes on server (Issuer Web Agent) & phone (osma/holder) and I want same configuration for both issuer & holder? Kindly guide me!!!

x0axz (Wed, 01 Jul 2020 16:26:48 GMT):
Hi all, @swcurran @andrew.whitehead @WadeBarnes @ianco , on my Issuer Web Agent, which is running on Amazon EC2 (e.g. IP is 42.56.822.342), I'm using these commands to build & run pool in docker: ```docker build --build-arg pool_ip=127.0.0.1 -f indy-pool.dockerfile -t indy_pool . docker run -itd -p 127.0.0.1:9701-9708:9701-9708 indy_pool``` This Docker files being used in Issuer Web Agent: https://github.com/hyperledger/indy-sdk/blob/master/ci/indy-pool.dockerfile https://github.com/Yoroitchi/AriesWebApp/blob/Issuer7000/AriesWebApp/docker/dockerfile This Genesis file being used in Issuer Web Agent: https://github.com/hyperledger/aries-framework-dotnet/blob/master/samples/aspnetcore/pool_genesis.txn with pool running on 127.0.0.1 on my cloud instance, i'm able Define Schema/Credentials, Issue Credentials e.t.c but when I try to connect from Holder (Osma) with Web Agent, I get this error on Osma Mobile App. ```Hyperledger.Indy.PoolApi.PoolConfigNotCreatedException: 'The requested pool cannot be opened because it does not have an existing configuration.'``` and the reason is that the configuration is not proper done. What will be the right configuration needed to run a pool & nodes on server (Issuer Web Agent) & phone (osma/holder) and I want same configuration for both issuer & holder? Kindly guide me!!!

x0axz (Wed, 01 Jul 2020 16:26:48 GMT):
Hi all, @swcurran @andrew.whitehead @WadeBarnes @ianco , on my Issuer Web Agent, which is running on Amazon EC2 (e.g. IP is 42.56.822.342), I'm using these commands to build & run pool in docker: ```docker build --build-arg pool_ip=127.0.0.1 -f indy-pool.dockerfile -t indy_pool . docker run -itd -p 127.0.0.1:9701-9708:9701-9708 indy_pool``` This Docker files being used in Issuer Web Agent: https://github.com/hyperledger/indy-sdk/blob/master/ci/indy-pool.dockerfile https://github.com/Yoroitchi/AriesWebApp/blob/Issuer7000/AriesWebApp/docker/dockerfile This Genesis file being used in Issuer Web Agent: https://github.com/hyperledger/aries-framework-dotnet/blob/master/samples/aspnetcore/pool_genesis.txn with pool running on 127.0.0.1 on my cloud instance, i'm able Define Schema/Credentials, Issue Credentials e.t.c but when I try to connect from Holder (Osma) with Web Agent, I get this error on Osma Mobile App. ```Hyperledger.Indy.PoolApi.PoolConfigNotCreatedException: 'The requested pool cannot be opened because it does not have an existing configuration.'``` and the reason is that the configuration is not proper done. What will be the right configuration needed to run a pool & nodes on server (Issuer Web Agent) & phone (osma/holder) and I want same configuration for both issuer & holder? Kindly guide me!!!

tomislav (Wed, 01 Jul 2020 16:33:05 GMT):
See my answer above on running the PoolConfigurator

tomislav (Wed, 01 Jul 2020 16:33:05 GMT):
See my answer in #aries-framework-dotnet on running the PoolConfigurator

x0axz (Wed, 01 Jul 2020 16:34:54 GMT):
yes, i read that.. the problem is that the pool on server (web agent) isn't configure properly, means can't access from phone (osma)

x0axz (Wed, 01 Jul 2020 16:35:58 GMT):
@tomislav configuration on xamarin is propelry created.. i followed this commit (https://github.com/hyperledger/aries-mobileagent-xamarin/commit/cb56d53f4e8a3da45cefa65503960b9e52e71099)

MrWoodyz (Thu, 02 Jul 2020 03:35:18 GMT):
Has joined the channel.

sheru (Thu, 02 Jul 2020 08:20:35 GMT):
Hey everyone Good Morning. I am able to send basic message and receive the message in the cloud-agent via web-hook but trying to receive the basic message in the AMAX Mobile APP. But still fighting with this. I haven't found any test cases for this as of now. Can anyone please guide me on this. Thanks

AnneGoellnitz (Thu, 02 Jul 2020 08:34:25 GMT):
Has joined the channel.

x0axz (Thu, 02 Jul 2020 11:05:40 GMT):
I followed these steps: https://github.com/hyperledger/indy-sdk#2-starting-the-test-pool-on-a-specific-ip-address In this case the client_ip and pool_ip in Genesis file was 42.56.822.342 Build Docker on 42.56.822.342 IP (cloud instance public IP) `docker build --build-arg pool_ip=42.56.822.342 -f ci/indy-pool.dockerfile -t indy_pool .` ```Successfully built fac50e6e708f Successfully tagged indy_pool:latest``` Run Docker on 42.56.822.342 IP (cloud instance public IP) `docker run -itd -p 42.56.822.342:9701-9708:9701-9708 indy_pool` ```59fba0839a4e5fb2aa6d1ac3bac4838551d0d089c797e8b1f3a62a07d2e500b6 docker: Error response from daemon: driver failed programming external connectivity on endpoint magical_tereshkova (6590de0360c45de3c92d81e534b4ebbde7d232093f0698689fea73858f4e628e): Error starting userland proxy: listen tcp 52.77.228.234:9708: bind: cannot assign requested address.```

x0axz (Thu, 02 Jul 2020 11:05:40 GMT):
I followed these steps: https://github.com/hyperledger/indy-sdk#2-starting-the-test-pool-on-a-specific-ip-address Build Docker on 42.56.822.342 `docker build --build-arg pool_ip=42.56.822.342 -f ci/indy-pool.dockerfile -t indy_pool .` ```Successfully built fac50e6e708f Successfully tagged indy_pool:latest``` Run Docker on 42.56.822.342 `docker run -itd -p 42.56.822.342:9701-9708:9701-9708 indy_pool` ```59fba0839a4e5fb2aa6d1ac3bac4838551d0d089c797e8b1f3a62a07d2e500b6 docker: Error response from daemon: driver failed programming external connectivity on endpoint magical_tereshkova (6590de0360c45de3c92d81e534b4ebbde7d232093f0698689fea73858f4e628e): Error starting userland proxy: listen tcp 52.77.228.234:9708: bind: cannot assign requested address.```

x0axz (Thu, 02 Jul 2020 11:05:40 GMT):
I followed these steps: https://github.com/hyperledger/indy-sdk#2-starting-the-test-pool-on-a-specific-ip-address Build Docker on 42.56.822.342 IP (cloud instance public IP) `docker build --build-arg pool_ip=42.56.822.342 -f ci/indy-pool.dockerfile -t indy_pool .` ```Successfully built fac50e6e708f Successfully tagged indy_pool:latest``` Run Docker on 42.56.822.342 IP (cloud instance public IP) `docker run -itd -p 42.56.822.342:9701-9708:9701-9708 indy_pool` ```59fba0839a4e5fb2aa6d1ac3bac4838551d0d089c797e8b1f3a62a07d2e500b6 docker: Error response from daemon: driver failed programming external connectivity on endpoint magical_tereshkova (6590de0360c45de3c92d81e534b4ebbde7d232093f0698689fea73858f4e628e): Error starting userland proxy: listen tcp 52.77.228.234:9708: bind: cannot assign requested address.```

x0axz (Thu, 02 Jul 2020 11:05:40 GMT):
I followed these steps: https://github.com/hyperledger/indy-sdk#2-starting-the-test-pool-on-a-specific-ip-address In this case the client_ip and pool_ip in Genesis file was 42.56.822.342 Build Docker on 42.56.822.342 IP (cloud instance public IP) `docker build --build-arg pool_ip=42.56.822.342 -f ci/indy-pool.dockerfile -t indy_pool .` ```Successfully built fac50e6e708f Successfully tagged indy_pool:latest``` Run Docker on 42.56.822.342 IP (cloud instance public IP) `docker run -itd -p 42.56.822.342:9701-9708:9701-9708 indy_pool` ```59fba0839a4e5fb2aa6d1ac3bac4838551d0d089c797e8b1f3a62a07d2e500b6 docker: Error response from daemon: driver failed programming external connectivity on endpoint magical_tereshkova (6590de0360c45de3c92d81e534b4ebbde7d232093f0698689fea73858f4e628e): Error starting userland proxy: listen tcp 52.77.228.234:9708: bind: cannot assign requested address.```

x0axz (Thu, 02 Jul 2020 11:10:19 GMT):
I also try to Build & Run docker on 0.0.0.0 and put the 0.0.0.0 in client_ip and pool_ip in Genesis file Build Docker on 42.56.822.342 IP (cloud instance public IP) `docker build --build-arg pool_ip=0.0.0.0 -f ci/indy-pool.dockerfile -t indy_pool .` `docker run -itd -p 0.0.0.0:9701-9708:9701-9708 indy_pool` Succesfully build and run but when I run `dotnet run`, to run the .Net Web Agent, it returned this error: ```crit: Hyperledger.Aries.Configuration.PoolConfigurationService[0] Couldn't create ledger configuration Hyperledger.Indy.InvalidStructureException: A value being processed is not valid. at Hyperledger.Aries.Ledger.DefaultPoolService.CreatePoolAsync(String poolName, String genesisFile) at Hyperledger.Aries.Configuration.PoolConfigurationService.StartAsync(CancellationToken cancellationToken) Unhandled exception. Hyperledger.Indy.InvalidStructureException: A value being processed is not valid. at Hyperledger.Aries.Ledger.DefaultPoolService.CreatePoolAsync(String poolName, String genesisFile) at Hyperledger.Aries.Configuration.PoolConfigurationService.StartAsync(CancellationToken cancellationToken) at Microsoft.Extensions.Hosting.Internal.Host.StartAsync(CancellationToken cancellationToken) at Microsoft.Extensions.Hosting.HostingAbstractionsHostExtensions.RunAsync(IHost host, CancellationToken token) at Microsoft.Extensions.Hosting.HostingAbstractionsHostExtensions.RunAsync(IHost host, CancellationToken token) at WebAgent.Program.

(String[] args)```

x0axz (Thu, 02 Jul 2020 11:10:19 GMT):
I also try to Build & Run docker on 0.0.0.0 and put the 0.0.0.0 in client_ip and pool_ip in Genesis file Build Docker on 42.56.822.342 IP (cloud instance public IP) `docker build --build-arg pool_ip=0.0.0.0 -f ci/indy-pool.dockerfile -t indy_pool .` `docker run -itd -p 0.0.0.0:9701-9708:9701-9708 indy_pool` Succesfully build and run but when I run `dotnet run`, to run the .Net Web Agent, it returned this error: ```crit: Hyperledger.Aries.Configuration.PoolConfigurationService[0] Couldn't create ledger configuration Hyperledger.Indy.InvalidStructureException: A value being processed is not valid. at Hyperledger.Aries.Ledger.DefaultPoolService.CreatePoolAsync(String poolName, String genesisFile) at Hyperledger.Aries.Configuration.PoolConfigurationService.StartAsync(CancellationToken cancellationToken) Unhandled exception. Hyperledger.Indy.InvalidStructureException: A value being processed is not valid. at Hyperledger.Aries.Ledger.DefaultPoolService.CreatePoolAsync(String poolName, String genesisFile) at Hyperledger.Aries.Configuration.PoolConfigurationService.StartAsync(CancellationToken cancellationToken) at Microsoft.Extensions.Hosting.Internal.Host.StartAsync(CancellationToken cancellationToken) at Microsoft.Extensions.Hosting.HostingAbstractionsHostExtensions.RunAsync(IHost host, CancellationToken token) at Microsoft.Extensions.Hosting.HostingAbstractionsHostExtensions.RunAsync(IHost host, CancellationToken token) at WebAgent.Program.

(String[] args)```

x0axz (Thu, 02 Jul 2020 11:10:19 GMT):
I also try to Build & Run docker on 0.0.0.0 and put the 0.0.0.0 in client_ip and pool_ip in Genesis file Build Docker on 42.56.822.342 IP (cloud instance public IP) `docker build --build-arg pool_ip=0.0.0.0 -f ci/indy-pool.dockerfile -t indy_pool .` `docker run -itd -p 0.0.0.0:9701-9708:9701-9708 indy_pool` Succesfully build and run but when I run `dotnet run`, to run the .Net Web Agent, it returned this error: ```crit: Hyperledger.Aries.Configuration.PoolConfigurationService[0] Couldn't create ledger configuration Hyperledger.Indy.InvalidStructureException: A value being processed is not valid. at Hyperledger.Aries.Ledger.DefaultPoolService.CreatePoolAsync(String poolName, String genesisFile) at Hyperledger.Aries.Configuration.PoolConfigurationService.StartAsync(CancellationToken cancellationToken) Unhandled exception. Hyperledger.Indy.InvalidStructureException: A value being processed is not valid. at Hyperledger.Aries.Ledger.DefaultPoolService.CreatePoolAsync(String poolName, String genesisFile) at Hyperledger.Aries.Configuration.PoolConfigurationService.StartAsync(CancellationToken cancellationToken) at Microsoft.Extensions.Hosting.Internal.Host.StartAsync(CancellationToken cancellationToken) at Microsoft.Extensions.Hosting.HostingAbstractionsHostExtensions.RunAsync(IHost host, CancellationToken token) at Microsoft.Extensions.Hosting.HostingAbstractionsHostExtensions.RunAsync(IHost host, CancellationToken token) at WebAgent.Program.

(String[] args)```

x0axz (Thu, 02 Jul 2020 11:10:19 GMT):
I also try to Build & Run docker on 0.0.0.0 and put the 0.0.0.0 in client_ip and pool_ip in Genesis file Build Docker on 42.56.822.342 IP (cloud instance public IP) `docker build --build-arg pool_ip=0.0.0.0 -f ci/indy-pool.dockerfile -t indy_pool .` `docker run -itd -p 0.0.0.0:9701-9708:9701-9708 indy_pool` Succesfully build and run but when I run `dotnet run`, to run the .Net Web Agent, it returned this error: ```crit: Hyperledger.Aries.Configuration.PoolConfigurationService[0] Couldn't create ledger configuration Hyperledger.Indy.InvalidStructureException: A value being processed is not valid. at Hyperledger.Aries.Ledger.DefaultPoolService.CreatePoolAsync(String poolName, String genesisFile) at Hyperledger.Aries.Configuration.PoolConfigurationService.StartAsync(CancellationToken cancellationToken) Unhandled exception. Hyperledger.Indy.InvalidStructureException: A value being processed is not valid. at Hyperledger.Aries.Ledger.DefaultPoolService.CreatePoolAsync(String poolName, String genesisFile) at Hyperledger.Aries.Configuration.PoolConfigurationService.StartAsync(CancellationToken cancellationToken) at Microsoft.Extensions.Hosting.Internal.Host.StartAsync(CancellationToken cancellationToken) at Microsoft.Extensions.Hosting.HostingAbstractionsHostExtensions.RunAsync(IHost host, CancellationToken token) at Microsoft.Extensions.Hosting.HostingAbstractionsHostExtensions.RunAsync(IHost host, CancellationToken token) at WebAgent.Program.

(String[] args)```

x0axz (Thu, 02 Jul 2020 11:10:19 GMT):
I also try to Build & Run docker on 0.0.0.0 and put the 0.0.0.0 in client_ip and pool_ip in Genesis file Build Docker on 42.56.822.342 IP (cloud instance public IP) `docker build --build-arg pool_ip=0.0.0.0 -f ci/indy-pool.dockerfile -t indy_pool .` `docker run -itd -p 0.0.0.0:9701-9708:9701-9708 indy_pool` Succesfully build and run but when I run `dotnet run`, to run the .Net Web Agent, it returned this error: ```crit: Hyperledger.Aries.Configuration.PoolConfigurationService[0] Couldn't create ledger configuration Hyperledger.Indy.InvalidStructureException: A value being processed is not valid. at Hyperledger.Aries.Ledger.DefaultPoolService.CreatePoolAsync(String poolName, String genesisFile) at Hyperledger.Aries.Configuration.PoolConfigurationService.StartAsync(CancellationToken cancellationToken) Unhandled exception. Hyperledger.Indy.InvalidStructureException: A value being processed is not valid. at Hyperledger.Aries.Ledger.DefaultPoolService.CreatePoolAsync(String poolName, String genesisFile) at Hyperledger.Aries.Configuration.PoolConfigurationService.StartAsync(CancellationToken cancellationToken) at Microsoft.Extensions.Hosting.Internal.Host.StartAsync(CancellationToken cancellationToken) at Microsoft.Extensions.Hosting.HostingAbstractionsHostExtensions.RunAsync(IHost host, CancellationToken token) at Microsoft.Extensions.Hosting.HostingAbstractionsHostExtensions.RunAsync(IHost host, CancellationToken token) at WebAgent.Program.

(String[] args)```

droconnel22 (Thu, 02 Jul 2020 13:40:08 GMT):
Hi All, I am debugging my Java SDK implementation and my LibIndy.api is null when I step into the source code. I've built libindy sdk and updated my paths.

droconnel22 (Thu, 02 Jul 2020 13:40:35 GMT):

Clipboard - July 2, 2020 9:40 AM

droconnel22 (Thu, 02 Jul 2020 13:41:17 GMT):
export DYLD_LIBRARY_PATH=/Users/*/*/indy-sdk/libindy/target/debug export LD_LIBRARY_PATH=/Users/*/*/indy-sdk/libindy/target/debug

droconnel22 (Thu, 02 Jul 2020 13:41:53 GMT):
Users/*/*/indy-sdk/libindy/target/debug (base) ➜ debug git:(master) ✗ ls build deps examples incremental libindy.a libindy.d libindy.dylib libindy.dylib.dSYM libindy.rlib libindy.so native

droconnel22 (Thu, 02 Jul 2020 13:42:03 GMT):
im running in Java 14 env

droconnel22 (Thu, 02 Jul 2020 13:42:39 GMT):
I am trying to connect to a von network node which ive successfully deployed remotely

debumonty (Thu, 02 Jul 2020 15:37:27 GMT):
Has joined the channel.

debumonty (Thu, 02 Jul 2020 15:37:28 GMT):
Hi team, I got a query on Hyperledger Indy and Aries. I have worked with Indy for about 6 months. My question is that, every holder/user on the network is assigned a service endpoint which the user gets connected to. The service endpoint is a different port number for each user on the network. So would this architecture be scalable as we might be needing to assign identity to thousands, millions or billions of users at some later point of time. How Sovrin would be handling the IP address ranges to be assigned to the users/holders in future?

bomlinh (Fri, 03 Jul 2020 01:19:00 GMT):
Has joined the channel.

AnneliPedersen (Mon, 06 Jul 2020 07:34:06 GMT):
Has joined the channel.

sachinmurali (Mon, 06 Jul 2020 08:09:32 GMT):
Has joined the channel.

sachinmurali (Mon, 06 Jul 2020 08:10:55 GMT):

Screenshot from 2020-07-06 13-40-13.png

sachinmurali (Mon, 06 Jul 2020 08:28:23 GMT):
HELP!!

x0axz (Mon, 06 Jul 2020 08:28:29 GMT):
is docker running?

sachinmurali (Mon, 06 Jul 2020 08:29:11 GMT):
yes

sachinmurali (Mon, 06 Jul 2020 08:30:00 GMT):

Screenshot from 2020-07-06 13-59-38.png

x0axz (Mon, 06 Jul 2020 08:39:32 GMT):
are u running that on server or localhost?

sachinmurali (Mon, 06 Jul 2020 08:57:16 GMT):
aws ec2 instace

x0axz (Mon, 06 Jul 2020 12:23:56 GMT):
First, clone Indy-sdk ```git clone https://github.com/x0axz/indy-sdk.git cd indy-sdk``` Start the pool with the IP address of Server so that mobile apps can reach the pool. ```# replace 192.168.179.90 with Server IP address docker build --build-arg pool_ip=192.168.179.90 -f ci/indy-pool.dockerfile -t indy_pool . docker run -itd -p 0.0.0.0:9701-9708:9701-9708 indy_pool``` To connect to the pool, the IP addresses in `/var/lib/indy/sandbox/pool_transactions_genesis` (in docker) and the pool configuration in mobile app must match.

x0axz (Mon, 06 Jul 2020 12:23:56 GMT):
First, clone Indy-sdk ```git clone https://github.com/hyperledger/indy-sdk.git``` Start the pool with the IP address of Server so that mobile apps can reach the pool. ```# replace 192.168.179.90 with Server IP address docker build --build-arg pool_ip=192.168.179.90 -f ci/indy-pool.dockerfile -t indy_pool . docker run -itd -p 0.0.0.0:9701-9708:9701-9708 indy_pool``` To connect to the pool, the IP addresses in `/var/lib/indy/sandbox/pool_transactions_genesis` (in docker) and the pool configuration in mobile app must match.

x0axz (Mon, 06 Jul 2020 12:23:56 GMT):
First, clone Indy-sdk ```git clone https://github.com/hyperledger/indy-sdk.git cd indy-sdk``` Start the pool with the IP address of Server so that mobile apps can reach the pool. ```# replace 192.168.179.90 with Server IP address docker build --build-arg pool_ip=192.168.179.90 -f ci/indy-pool.dockerfile -t indy_pool . docker run -itd -p 0.0.0.0:9701-9708:9701-9708 indy_pool``` To connect to the pool, the IP addresses in `/var/lib/indy/sandbox/pool_transactions_genesis` (in docker) and the pool configuration in mobile app must match.

x0axz (Mon, 06 Jul 2020 12:23:56 GMT):
First, clone Indy-sdk ```git clone https://github.com/hyperledger/indy-sdk.git cd indy-sdk``` Start the pool with the IP address of Server so that mobile apps can reach the pool. ```# replace 192.168.179.90 with Server IP address docker build --build-arg pool_ip=192.168.179.90 -f ci/indy-pool.dockerfile -t indy_pool . docker run -itd -p 0.0.0.0:9701-9708:9701-9708 indy_pool``` To connect to the pool, the IP addresses in `/var/lib/indy/sandbox/pool_transactions_genesis` (in docker) and the pool configuration in mobile app must match.

x0axz (Mon, 06 Jul 2020 12:23:56 GMT):
First, clone Indy-sdk ```git clone https://github.com/hyperledger/indy-sdk.git cd indy-sdk``` Start the pool with the IP address of Server so that mobile apps can reach the pool. ```# replace 192.168.179.90 with Server IP address docker build --build-arg pool_ip=192.168.179.90 -f ci/indy-pool.dockerfile -t indy_pool . ``` `docker run -itd -p 0.0.0.0:9701-9708:9701-9708 indy_pool` To connect to the pool, the IP addresses in `/var/lib/indy/sandbox/pool_transactions_genesis` (in docker) and the pool configuration in mobile app must match.

x0axz (Mon, 06 Jul 2020 12:23:56 GMT):
First, clone Indy-sdk ```git clone https://github.com/hyperledger/indy-sdk.git cd indy-sdk``` Start the pool with the IP address of Server so that mobile apps can reach the pool. ```# replace 192.168.179.90 with Server IP address docker build --build-arg pool_ip=192.168.179.90 -f ci/indy-pool.dockerfile -t indy_pool . ``` ```docker run -itd -p 0.0.0.0:9701-9708:9701-9708 indy_pool``` To connect to the pool, the IP addresses in `/var/lib/indy/sandbox/pool_transactions_genesis` (in docker) and the pool configuration in mobile app must match.

x0axz (Mon, 06 Jul 2020 12:23:56 GMT):
First, clone Indy-sdk ```git clone https://github.com/hyperledger/indy-sdk.git cd indy-sdk``` Start the pool with the IP address of Server so that mobile apps can reach the pool. ```# replace 192.168.179.90 with Server IP address docker build --build-arg pool_ip=192.168.179.90 -f ci/indy-pool.dockerfile -t indy_pool . ``` ```docker run -itd -p 0.0.0.0:9701-9708:9701-9708 indy_pool``` To connect to the pool, the IP addresses in `/var/lib/indy/sandbox/pool_transactions_genesis` (in docker) and the pool configuration in mobile app must match.

x0axz (Mon, 06 Jul 2020 12:29:46 GMT):
Also, look into the Genesis file sandbox pool in .indy_client ```cd .indy_client/pool/sandbox vi sandbox.txn``` The client_ip and pool_ip should be the server IP of AWS Instance (the one u entered at time of building docker pool e.g. 192.168.179.90)

x0axz (Mon, 06 Jul 2020 12:29:46 GMT):
Also, look into the Genesis file sandbox pool in .indy_client ```cd .indy_client/pool/sandbox vi sandbox.txn``` The client_ip and pool_ip should be the server IP of AWS Instance (the one u entered at time of building docker pool e.g. 192.168.179.90)

x0axz (Mon, 06 Jul 2020 12:29:46 GMT):
Also, look into the Genesis file of sandbox pool in .indy_client ```cd .indy_client/pool/sandbox vi sandbox.txn``` The client_ip and pool_ip should be the server IP of AWS Instance (the one u entered at time of building docker pool e.g. 192.168.179.90)

x0axz (Mon, 06 Jul 2020 12:29:46 GMT):
Also, look into the Genesis file of sandbox pool in .indy_client ```cd .indy_client/pool/sandbox vi sandbox.txn``` The `client_ip` and `pool_ip` should be the server IP of AWS Instance (the one u entered at the time of building docker pool e.g. `192.168.179.90`)

x0axz (Mon, 06 Jul 2020 12:29:46 GMT):
Also, look into the Genesis file of sandbox pool in .indy_client ```cd .indy_client/pool/sandbox vi sandbox.txn``` The `client_ip` and `pool_ip` should be the server IP of AWS EC2 Instance (the one u entered at the time of building docker pool e.g. `192.168.179.90`)

cmgabriel (Mon, 06 Jul 2020 13:12:43 GMT):
Has joined the channel.

san1ker (Tue, 07 Jul 2020 01:13:34 GMT):
Has joined the channel.

san1ker (Tue, 07 Jul 2020 01:13:34 GMT):
Hi team :)

san1ker (Tue, 07 Jul 2020 01:13:39 GMT):

Clipboard - 2020년 7월 7일 오전 10시 13분

san1ker (Tue, 07 Jul 2020 01:14:30 GMT):
Is observer node not supported yet?? I saw readthedocs says "services (array of strings; optional): the service of the Node. VALIDATOR is the only supported one now."

san1ker (Tue, 07 Jul 2020 01:16:37 GMT):
My Indy network pool has 4 validator nodes. If Indy supporting observer node, Do I have to change or set pool config?

san1ker (Tue, 07 Jul 2020 01:23:08 GMT):
ADD) Other Node transactions with service [] or ["VALIDATOR"] are successfully being recorded to pool ledger. I cannot send only Node txn with service ["OBSERVER"]

swcurran (Tue, 07 Jul 2020 01:48:41 GMT):
Observer nodes are not yet supported. The validator nodes serve as the handlers of reads from the ledger.

san1ker (Tue, 07 Jul 2020 04:31:31 GMT):
Got it. Thanks:thumbsup:

davidd (Fri, 10 Jul 2020 17:38:33 GMT):
Hi all, how to decrypt/see the content in Wallet?

davidd (Fri, 10 Jul 2020 17:38:33 GMT):
Hi all, how to decrypt/read the content of Wallet?

swcurran (Fri, 10 Jul 2020 20:16:06 GMT):
That's deliberately not easy -- there is nothing in the libraries to do that.

tangelo1 (Fri, 10 Jul 2020 20:34:38 GMT):
Has joined the channel.

hamidm (Sat, 11 Jul 2020 18:40:56 GMT):
Hello, is there an api or a way to get the full DIDDoc for a specific DID from the ledger ? Thanks.

san1ker (Mon, 13 Jul 2020 01:25:22 GMT):
Hi team, I have a question about primary field parameters of cred-def.

san1ker (Mon, 13 Jul 2020 01:26:41 GMT):
Indy cli example: ledger cred-def schema_id=1 signature_type=CL tag=1 primary={"n":"1","s":"2","rms":"3","r":{"age":"4","name":"5"},"rctxt":"6","z":"7"} Can I ask about definition and meaning of "n", "r", "s", "rms", "rctxt" and "z"?? Thanks:)

san1ker (Mon, 13 Jul 2020 01:26:41 GMT):
Indy cli example: ledger cred-def schema_id=1 signature_type=CL tag=1 primary={"n":"1","s":"2","rms":"3","r":{"age":"4","name":"5"},"rctxt":"6","z":"7"} Can I ask about definition and meaning of "n", "s", "rms", "rctxt" and "z"?? (maybe "r" means attr and value of target schema) Thanks:)

berserkr (Mon, 13 Jul 2020 05:35:15 GMT):
Hi Team, are there end-to-end examples like the ones for hyperledger fabric that would help me get started? The walkthrough docs have too many assumptions to follow up

san1ker (Mon, 13 Jul 2020 05:51:52 GMT):
https://github.com/bcgov/von-network

san1ker (Mon, 13 Jul 2020 05:52:50 GMT):
Samples with 4 nodes and nodes always makes their keys with same seed like 00000...Node1

berserkr (Mon, 13 Jul 2020 05:53:39 GMT):
I'll take a look, thank you

san1ker (Mon, 13 Jul 2020 05:53:45 GMT):
:)

san1ker (Mon, 13 Jul 2020 05:53:58 GMT):
:)

berserkr (Mon, 13 Jul 2020 08:14:49 GMT):
Another question... I see a lot of the same examples on the acme/alice/thrift samples... I read that PII is not to be stored in the ledger, but many examples show Alice's SSN... how are entities viewing Alice's SSN without having access to the ledger?

domwoe (Mon, 13 Jul 2020 08:24:30 GMT):
There's no API for that in indy-sdk. You can have a look at the implementation of the did:sov driver for the universal resolver. How the indy-sdk can be used to generate a DIDdoc. https://github.com/decentralized-identity/uni-resolver-driver-did-sov

ridaitxi (Mon, 13 Jul 2020 09:33:05 GMT):
Has joined the channel.

ridaitxi (Mon, 13 Jul 2020 09:33:05 GMT):
Hello good people, Is anybody here familiar with the process for building libvcx for iOS? I am using this article as a reference and I am stuck at the last paragraph: https://dev.to/jakubkoci/how-to-build-a-vcx-ios-library-f32 "This still not over. To call output libvcxall.a library from our code we need to wrap it with Objective-C and c-callable headers. First, we copy the libvcxall.a to indy-sdk/vcx/wrappers/ios/vcx which contains these headers and then we use xcodebuild and lipo utility to create our final vcx.framework." I'd appreciate any help,

ridaitxi (Mon, 13 Jul 2020 09:33:05 GMT):
Hello good people, Is anybody here familiar with the process for building libvcx for iOS? I am using this article as a reference and I am stuck at the last paragraph: https://dev.to/jakubkoci/how-to-build-a-vcx-ios-library-f32 "This still not over. To call output libvcxall.a library from our code we need to wrap it with Objective-C and c-callable headers. First, we copy the libvcxall.a to indy-sdk/vcx/wrappers/ios/vcx which contains these headers and then we use xcodebuild and lipo utility to create our final vcx.framework." I'd appreciate any help, Rida

swcurran (Mon, 13 Jul 2020 13:50:27 GMT):
The credential goes from the Issuer to the Holder, e.g. from the gov't (including SSN) to Alice. Later Alice presents proof of the claims in the credential (some or all, possibly combined with claims from other credentials) to a verifier (say, Acme). None of the data in the credential goes to the ledger. What goes to the ledger are metadata about the credentials (e.g. names of the claims - data elements) and cryptographic public keys needed in constructing the proof (by Alice) and to verify the proof (executed by Acme).

berserkr (Mon, 13 Jul 2020 17:26:31 GMT):
Team, trying to get started with plenum but the main getting started link in the repo is broken: https://github.com/hyperledger/indy-node/blob/master/docs/setup-dev.md --- gives a 404... do you guys have the actual updated link? Thank you

berserkr (Mon, 13 Jul 2020 17:29:47 GMT):
The `most up to date l;ink` is also broken - https://indy.readthedocs.io/projects/plenum/en/latest/index.html

berserkr (Mon, 13 Jul 2020 17:29:47 GMT):
The `most up to date link` is also broken - https://indy.readthedocs.io/projects/plenum/en/latest/index.html

PatrikStas (Mon, 13 Jul 2020 17:34:58 GMT):
@jakubkoci is your man :-)

swcurran (Mon, 13 Jul 2020 20:24:15 GMT):
@esplinr ^^^^^

berserkr (Mon, 13 Jul 2020 21:01:55 GMT):
Team, what is the reasoning behind having 4+ ledgers for Indy? Just makes things more complicated than they should be imho

esplinr (Mon, 13 Jul 2020 23:03:10 GMT):
Each ledger fulfills a specific purpose.

esplinr (Mon, 13 Jul 2020 23:04:18 GMT):
It looks like when someone experimented with the readthedocs integration, they forgot to update that link. It should point here: https://github.com/hyperledger/indy-node/blob/master/docs/source/setup-dev.md

swcurran (Mon, 13 Jul 2020 23:04:22 GMT):
I know about three of the ledgers and have a guess on the fourth: - Domain is the identity transactions -- the DIDs (NYMs), Attributes of nyms, schema, credential definitions, revocation registries and revocation entries. That's the "data" of the ledger. - Config are the transactions that change the network. For example, to trigger updates to the software. I think (not certain) the authorization data is there as well. - Node lists the nodes (current and past) that are participating in the validation. - Fourth is optional (I believe) based on the configuration for a token that can be added to the network.

lynn.bendixsen (Mon, 13 Jul 2020 23:05:01 GMT):
Maybe this is the doc you are needing? https://github.com/hyperledger/indy-node/blob/master/docs/source/setup-dev.md

swcurran (Mon, 13 Jul 2020 23:13:47 GMT):
To see the transactions on an indy ledger, you can look at indyscan.io which has a ledger browser for each of the Sovrin networks (Main, Staging and Builder).

berserkr (Mon, 13 Jul 2020 23:40:46 GMT):
Thank you

berserkr (Mon, 13 Jul 2020 23:44:03 GMT):
I understand that each one has a separate function, but it does complicate things. Like each ledger needs to have its own sync logic rather than having one ledger that has all the necessary logic. Are there documents/papers that describe the benefit of such scheme? The reason is that I am responsible for building a network for decentralized identity and need to understand the reasoning behind indy... I've played and delivered end-to-end solutions with Bitcoin, ethereum, hlf, and a few other ledgers. Indy is a bit new and still grasping the architecture.

JamesEbert (Tue, 14 Jul 2020 00:12:43 GMT):
Has joined the channel.

JamesEbert (Tue, 14 Jul 2020 00:12:43 GMT):
Hello folks! I'm using Libindy from React-Native using the Java wrapper within Android (Indy 1.14.2). Upon trying to open a pool handle (openPoolLedger()), the app crashes with a signal 6 SIGABRT, with no other error messages. It crashes on the socket.connect() call, and since that uses zmq, that leads it to be my most suspected issue. I have enabled the internet and networking permissions for the app (as well as enabling cleartext messages in the manifest), and have ensured the call is being run in a separate thread from the main thread. I am open to any ideas anyone might have. Thanks!

esplinr (Tue, 14 Jul 2020 01:03:06 GMT):
https://www.youtube.com/watch?v=ncdvaJrOm_Q&list=PLRp0viTDxBWH2hRisVD6Wij43afXt0FHN&index=3

san1ker (Tue, 14 Jul 2020 04:17:55 GMT):
Hi team :) Are fully qualified DIDs not recorded to ledger?? from field is tx only shows the main part of Identifier. If so, fully qualified DID is only used in wallet?? Thanks.

san1ker (Tue, 14 Jul 2020 04:17:55 GMT):
Hi team :) Are fully qualified DIDs not recorded to ledger?? "from" field in tx only shows the main part of Identifier. If so, fully qualified DID is only used in wallet?? Thanks.

san1ker (Tue, 14 Jul 2020 04:19:22 GMT):

Clipboard - 2020년 7월 14일 오후 1시 19분

san1ker (Tue, 14 Jul 2020 04:31:19 GMT):

Clipboard - 2020년 7월 14일 오후 1시 31분

san1ker (Tue, 14 Jul 2020 04:31:25 GMT):
Hi Team :) Can fully-qualified DID not send schema tx?? other txs (nym and cred-def to existing schema) work but schema tx occurs error like parse error thanks

ridaitxi (Tue, 14 Jul 2020 06:28:26 GMT):
@PatrikStas indeed he is, thanks for mentioning him I appreciate it. I hope he gets to see this. Cheers.

CHempel (Tue, 14 Jul 2020 12:20:40 GMT):
Hi, I have a question about the Verkeys. When I use the Indy-CLI to create a DID based on a seed I get a short verkey (~VERKEY). The Full-Verkey on the ledger seems to be DID+VERKEY. But when I create a DID using Indy-SDK with the CreateAndStoreMyDidAsync method I get a DID and a FULLVERKEY. Now I would like to know how this looks like when using this DID based on the seed. If I want to use a DID created with the Indy-CLI later in an application that uses the Indy-SDK (or Aries) or vice versa to sign transactions for the ledger, this might not work, because the verkey created in the application does not match the verkey on the ledger.

CHempel (Tue, 14 Jul 2020 12:20:40 GMT):
Hi, I have a question about the Verkeys. When I use the Indy-CLI to create a DID based on a seed I get a short verkey (~VERKEY). The Full-Verkey on the ledger seems to be DID+VERKEY. But when I create a DID using Indy-SDK with the CreateAndStoreMyDidAsync method I get a DID and a FULLVERKEY. Now I would like to know how this looks like when using this DID based on the seed. If I want to use a DID created with the Indy-CLI later in an application that uses the Indy-SDK (or Aries) or vice versa to sign transactions for the ledger, this might not work, because the verkey created in the application does not match the verkey on the ledger. As it looks the CLI only shows the abbreviated verkey, but the question is if I have to use it for transactions using the Indy-SDK.

tomislav (Tue, 14 Jul 2020 17:57:30 GMT):
A note on key abbreviation and DID relationship. The DID is simply the first 16 bytes of the verkey, represented in base58. The abbreviated key, is the other lefotver bytes with a ~ prefix, so you can easily construct the full verkey by concatenating the the DID and abbreviated verkey. When a user rotates their key for a DID and write that to the ledger, that key can no longer be represented in abbreviated form.

tomislav (Tue, 14 Jul 2020 17:57:30 GMT):
A note on key abbreviation and DID relationship. The DID is simply the first 16 bytes of the verkey, represented in base58. The abbreviated key, is the other lefotver bytes with a ~ prefix, so you can easily construct the full verkey by concatenating the the DID and abbreviated verkey (concat the bytes, not the base58 strings). When a user rotates their key for a DID and write that to the ledger, that key can no longer be represented in abbreviated form.

hamidm (Wed, 15 Jul 2020 00:49:06 GMT):
Thanks. I’ll check it out.

CHempel (Wed, 15 Jul 2020 08:57:30 GMT):
Thanks for the explanation! Then there might be an issue in Indy-Scan when creating the enriched data. There the verkeyFull attribute is shown as a concatenation of the pure strings, which caused this confusion on my side.

CHempel (Wed, 15 Jul 2020 08:57:30 GMT):
Thanks for the explanation! Then there might be an issue in Indyscan when creating the enriched data. There the verkeyFull attribute is shown as a concatenation of the pure strings, which caused this confusion on my side.

tomislav (Wed, 15 Jul 2020 14:11:43 GMT):
Ah. That can actually work in some cases, but is definitely not a rule

kangme (Wed, 15 Jul 2020 19:02:01 GMT):
Has joined the channel.

mccown (Wed, 15 Jul 2020 19:26:57 GMT):
Just a reminder about the Identity Implementer's WG Call tomorrow morning (July 16th @ 9am MT). This is a chance to get updates on the WG calls you may have missed this week. Our guest presenter will be Horacio Nunez from Kiva. Horacio will be presenting on a new project and open source repo. Here are links to the Wiki and the Zoom meeting. Wiki: https://wiki.hyperledger.org/display/IWG/2020-07-16+Identity+WG+Implementers+Call Zoom: https://zoom.us/j/244779296

berserkr (Fri, 17 Jul 2020 18:33:17 GMT):
Hi all, is there a reason why indy nodes are running at 25-40% cpu when idle? Meaning no interaction with the network yet running a mid CPU util?

berserkr (Fri, 17 Jul 2020 18:33:44 GMT):
I have 4 nodes in a pool, all running at: `31.8 0.5 1527:16 start_indy_node`

pknowles (Fri, 17 Jul 2020 20:50:35 GMT):
https://sovrin.org/advance-the-mission-of-identity-for-all-as-a-trustee-of-the-worlds-leading-ssi-community/

pknowles (Fri, 17 Jul 2020 20:50:35 GMT):
Sovrin Foundation is looking to expand their Board by electing a few more Trustees. https://sovrin.org/advance-the-mission-of-identity-for-all-as-a-trustee-of-the-worlds-leading-ssi-community/

berserkr (Tue, 21 Jul 2020 00:22:06 GMT):
Hi All, is there a good document that describes predicates?

berserkr (Tue, 21 Jul 2020 00:22:06 GMT):
Hi All, is there a good document that describes predicates? And how to use them or what is supported. I am explicitly talking about `requested_predicates`

esplinr (Tue, 21 Jul 2020 00:53:55 GMT):
https://github.com/hyperledger/indy-sdk/blob/544bac74942667fc0a512f68f16b9bcad5ba66ab/docs/design/002-anoncreds/README.md

esplinr (Tue, 21 Jul 2020 00:54:36 GMT):
https://github.com/hyperledger/indy-sdk/blob/eb32486c4660b392db7b64643ae7a310eb158337/docs/getting-started/indy-walkthrough.md

esplinr (Tue, 21 Jul 2020 00:54:54 GMT):
I probably should have shared those in the opposite order, as the second is a better introduction.

berserkr (Tue, 21 Jul 2020 00:56:55 GMT):
Thank you. I saw that after doing some searching. Looks like predicate support is limited as of now. Will suffice though.

berserkr (Tue, 21 Jul 2020 01:01:19 GMT):
@esplinr Do you know what part of libindy performs these checks or generates the proofs for predicates or values in general?

berserkr (Tue, 21 Jul 2020 01:01:30 GMT):
Credential values I mean

san1ker (Tue, 21 Jul 2020 04:50:12 GMT):
Hi team I have a question about anoncreds 1.0 VS 2.0 Using tails, revoc entry contains only index number (matched prime number is not exposed) so revoked information can be hidden. But using merkle tree, as described in https://jira.hyperledger.org/browse/INDY-2365, (I understood like 0010000.. -> cred with index number 3 revoked) index number is the only information (there is no matched hidden factor like prime number in tails) so I think anyone who has merkle root can derive revoked information of credential. Is there any way to hide revoked information of cred with merkle tree(anoncreds2.0)?? thanks.

kangme (Tue, 21 Jul 2020 07:06:02 GMT):
Hello, I am new to indy,when I run the how-to Tutorials,there is one error:from utils import get_pool_genesis_txn_path, PROTOCOL_VERSION ImportError: cannot import name 'get_pool_genesis_txn_path' from 'utils',I have installed all dependencies and with a running indy pool,could anyone help me to figure out the problem. Thanks

berserkr (Tue, 21 Jul 2020 07:29:20 GMT):
did you run `pip install .` in the local dir?

berserkr (Tue, 21 Jul 2020 07:34:42 GMT):
My advice is to go ahead and install the pool as described in the indy-sdk example

berserkr (Tue, 21 Jul 2020 07:34:54 GMT):
then you run install inside of the wrapper of your choice

berserkr (Tue, 21 Jul 2020 07:36:41 GMT):
https://github.com/hyperledger/indy-sdk#how-to-start-local-nodes-pool-with-docker

berserkr (Tue, 21 Jul 2020 07:36:48 GMT):
Use taht to run your pool

FarhanShafiq (Wed, 22 Jul 2020 08:32:01 GMT):
Has left the channel.

MurugesanRathinam (Wed, 22 Jul 2020 12:00:03 GMT):
Has joined the channel.

MurugesanRathinam (Wed, 22 Jul 2020 12:00:03 GMT):
I am setting up hyper ledger indy network able to create cli and need some help on creating trasactions any document will be helpfull

WadeBarnes (Wed, 22 Jul 2020 12:41:06 GMT):
Some setup docs here: https://chat.hyperledger.org/channel/indy-node?msg=6AHLjgssHtEAjcuz5

rangeshsripathi (Thu, 23 Jul 2020 03:16:48 GMT):
Has joined the channel.

rangeshsripathi (Thu, 23 Jul 2020 03:16:49 GMT):
Hello, I am trying to Browse indy ledger using VON custom set up

rangeshsripathi (Thu, 23 Jul 2020 03:17:05 GMT):

Clipboard - July 23, 2020 8:47 AM

rangeshsripathi (Thu, 23 Jul 2020 03:20:00 GMT):
I am pointing to genesis file(local indy pool) and trying to run the server at port 9000 ... But unable to run the same , I did install the packages . Any clue why I am not able to run the VON server ?

Mahadevan 3 (Thu, 23 Jul 2020 04:43:17 GMT):
Has joined the channel.

MansH (Thu, 23 Jul 2020 07:20:50 GMT):
Has joined the channel.

MansH (Thu, 23 Jul 2020 07:20:50 GMT):
Why does credential_def only store transaction seqNo.? I think it is better to store schema_id. Is there a reason for this implementation?

MansH (Thu, 23 Jul 2020 07:20:50 GMT):
Why does credential_def only store transaction seqNo.? I think it is better to store schema_id. Is there a reason for this implementation? I use von-network server and ACA-py agent.

jameshiester (Thu, 23 Jul 2020 12:52:00 GMT):
Devs can correct me if I'm wrong, but I think it's because the SDK allows retrieving by seqNo, but not by schema_id. If you wanted to retrieve the schema given only the credential_def, there would be no efficient way to do so using only schema_id without having all the transaction history cached

RafaelAPB (Sun, 26 Jul 2020 13:27:32 GMT):
Has joined the channel.

RafaelAPB (Sun, 26 Jul 2020 13:27:33 GMT):
Hello all. Can someone tell me what is recorded on an Indy blockchain ,except for the hash of DID documents, definitions and schemas of credentials, and revocation registries?

Primordium (Sun, 26 Jul 2020 14:00:29 GMT):
maybe this document can help you https://www.evernym.com/wp-content/uploads/2017/07/What-Goes-On-The-Ledger.pdf

JakeZ2020 (Mon, 27 Jul 2020 14:57:35 GMT):
Has joined the channel.

rangeshsripathi (Mon, 27 Jul 2020 16:09:49 GMT):
Hi , I am trying to connect to local indy node and I get "indy.error.PoolLedgerTimeout" , Any clue where I am going wrong ? Attached logs below.. `"INFO:server.anchor:Genesis file already exists: /gen/genesis-pool-config-local DEBUG:indy.pool:create_pool_ledger_config: >>> config_name: 'local-new-pool-ver2', config: '{"genesis_txn": "/gen/genesis-pool-config-local"}' DEBUG:indy.pool:create_pool_ledger_config: Creating callback DEBUG:indy.pool:create_pool_ledger_config: <<< res: None {} DEBUG:indy.pool:open_pool_ledger: >>> config_name: 'local-new-pool-ver2', config: '{}' DEBUG:indy.pool:open_pool_ledger: Creating callback WARNING:indy.libindy:_indy_loop_callback: Function returned error ERROR:server.anchor:Initialization error: Traceback (most recent call last): File "/home/indy/server/anchor.py", line 216, in _open_pool self._pool = await pool.open_pool_ledger(pool_name, json.dumps(pool_cfg)) File "/home/indy/.pyenv/versions/3.5.7/lib/python3.5/site-packages/indy/pool.py", line 83, in open_pool_ledger open_pool_ledger.cb) File "/home/indy/.pyenv/versions/3.5.7/lib/python3.5/asyncio/futures.py", line 381, in __iter__ yield self # This tells Task to wait for completion. File "/home/indy/.pyenv/versions/3.5.7/lib/python3.5/asyncio/tasks.py", line 310, in _wakeup future.result() File "/home/indy/.pyenv/versions/3.5.7/lib/python3.5/asyncio/futures.py", line 294, in result raise self._exception indy.error.PoolLedgerTimeout The above exception was the direct cause of the following exception: Traceback (most recent call last): File "/home/indy/server/anchor.py", line 373, in open await self._open_pool() File "/home/indy/server/anchor.py", line 218, in _open_pool raise AnchorException("Error opening pool ledger connection") from e server.anchor.AnchorException: Error opening pool ledger connection"`

rangeshsripathi (Mon, 27 Jul 2020 16:09:49 GMT):
Hi , I am trying to connect to local indy node and I get "indy.error.PoolLedgerTimeout" , Any clue where I am going wrong ? Attached logs below.. ``` `"INFO:server.anchor:Genesis file already exists: /gen/genesis-pool-config-local DEBUG:indy.pool:create_pool_ledger_config: >>> config_name: 'local-new-pool-ver2', config: '{"genesis_txn": "/gen/genesis-pool-config-local"}' DEBUG:indy.pool:create_pool_ledger_config: Creating callback DEBUG:indy.pool:create_pool_ledger_config: <<< res: None {} DEBUG:indy.pool:open_pool_ledger: >>> config_name: 'local-new-pool-ver2', config: '{}' DEBUG:indy.pool:open_pool_ledger: Creating callback WARNING:indy.libindy:_indy_loop_callback: Function returned error ERROR:server.anchor:Initialization error: Traceback (most recent call last): File "/home/indy/server/anchor.py", line 216, in _open_pool self._pool = await pool.open_pool_ledger(pool_name, json.dumps(pool_cfg)) File "/home/indy/.pyenv/versions/3.5.7/lib/python3.5/site-packages/indy/pool.py", line 83, in open_pool_ledger open_pool_ledger.cb) File "/home/indy/.pyenv/versions/3.5.7/lib/python3.5/asyncio/futures.py", line 381, in __iter__ yield self # This tells Task to wait for completion. File "/home/indy/.pyenv/versions/3.5.7/lib/python3.5/asyncio/tasks.py", line 310, in _wakeup future.result() File "/home/indy/.pyenv/versions/3.5.7/lib/python3.5/asyncio/futures.py", line 294, in result raise self._exception indy.error.PoolLedgerTimeout The above exception was the direct cause of the following exception: Traceback (most recent call last): File "/home/indy/server/anchor.py", line 373, in open await self._open_pool() File "/home/indy/server/anchor.py", line 218, in _open_pool raise AnchorException("Error opening pool ledger connection") from e server.anchor.AnchorException: Error opening pool ledger connection"` ```

lynn.bendixsen (Mon, 27 Jul 2020 18:29:50 GMT):
Regularly a Timeout error is due to something with your IP addresses or networking configuration. Check your pool Genesis file first and then track it down from there.

jameshiester (Mon, 27 Jul 2020 21:24:20 GMT):
Does the pool ledger connection timeout?

Taaanos (Tue, 28 Jul 2020 14:31:24 GMT):
Has joined the channel.

RafaelAPB (Tue, 28 Jul 2020 18:58:38 GMT):
That's very useful, thank you!

mccown (Wed, 29 Jul 2020 20:18:15 GMT):
Just a reminder about the Identity Implementer's WG Call tomorrow morning (July 30th @ 9am MT). This is a chance to get updates on the WG calls you may have missed this week. Our guest presenter will be Kyle Den Hartog from Mattr. Kyle will be presenting a new proposal to add DIDDoc support to indy-node and how that will benefit supporting applications. Here are links to the Wiki and the Zoom meeting. Wiki: https://wiki.hyperledger.org/display/IWG/2020-07-30+Identity+WG+Implementers+Call Zoom: https://zoom.us/j/244779296

m00sey (Thu, 30 Jul 2020 16:20:54 GMT):
Has joined the channel.

Primordium (Thu, 30 Jul 2020 18:32:59 GMT):
I had 4 nodes running on different machines, we had a to restart all the VMs and now when I start the nodes the following error is showing ``` ubuntu@vm-blk-node1:~$ sudo start_indy_node Node1 0.0.0.0 9701 0.0.0.0 9702 & [1] 30214 ubuntu@vm-blk-node1:~$ Traceback (most recent call last): File "/usr/local/bin/start_indy_node", line 19, in client_ip=sys.argv[4], client_port=int(sys.argv[5])) File "/usr/local/lib/python3.5/dist-packages/indy_node/utils/node_runner.py", line 50, in run_node ha=node_ha, cliha=client_ha) File "/usr/local/lib/python3.5/dist-packages/indy_node/server/node.py", line 87, in __init__ bootstrap_cls=bootstrap_cls) File "/usr/local/lib/python3.5/dist-packages/plenum/server/node.py", line 188, in __init__ self.bootstrapper = self._bootstrap_node(bootstrap_cls, storage) File "/usr/local/lib/python3.5/dist-packages/plenum/server/node.py", line 3183, in _bootstrap_node bootstrapper.init(domain_storage=storage) File "/usr/local/lib/python3.5/dist-packages/plenum/server/ledgers_bootstrap.py", line 78, in init self._init_storages(domain_storage=domain_storage) File "/usr/local/lib/python3.5/dist-packages/indy_node/server/node_bootstrap.py", line 84, in _init_storages super()._init_storages(domain_storage) File "/usr/local/lib/python3.5/dist-packages/plenum/server/node_bootstrap.py", line 48, in _init_storages super()._init_storages(domain_storage) File "/usr/local/lib/python3.5/dist-packages/plenum/server/ledgers_bootstrap.py", line 100, in _init_storages self._create_ledger('config'), File "/usr/local/lib/python3.5/dist-packages/plenum/server/ledgers_bootstrap.py", line 210, in _create_ledger hash_store = initHashStore(self.data_location, name, self.config, hs_type=hs_type) File "/usr/local/lib/python3.5/dist-packages/storage/helper.py", line 86, in initHashStore config=config) File "/usr/local/lib/python3.5/dist-packages/plenum/persistence/db_hash_store.py", line 28, in __init__ self.open() File "/usr/local/lib/python3.5/dist-packages/plenum/persistence/db_hash_store.py", line 97, in open read_only=self._read_only, db_config=self.config.db_merkle_nodes_config) File "/usr/local/lib/python3.5/dist-packages/storage/helper.py", line 31, in initKeyValueStorage read_only, db_config) File "/usr/local/lib/python3.5/dist-packages/storage/kv_store_rocksdb.py", line 24, in __init__ self.open() File "/usr/local/lib/python3.5/dist-packages/storage/kv_store_rocksdb.py", line 84, in open self._db = rocksdb.DB(self._db_path, opts, read_only=self._read_only) File "rocksdb/_rocksdb.pyx", line 1437, in rocksdb._rocksdb.DB.__cinit__ File "rocksdb/_rocksdb.pyx", line 84, in rocksdb._rocksdb.check_status rocksdb.errors.RocksIOError: b'IO error: While lock file: /var/lib/indy/ledgertest/data/Node1/config_merkleNodes/LOCK: Resource temporarily unavailable' ```

WadeBarnes (Thu, 30 Jul 2020 19:16:51 GMT):
You may need to delete the lock file.

MrWoody (Thu, 30 Jul 2020 22:32:30 GMT):
Hi everyone... How can I dump the whole ledger transactions via de indy-cli?

MrWoody (Thu, 30 Jul 2020 22:32:53 GMT):
I need it to validate the ledger transactions after operations for testing

MrWoody (Thu, 30 Jul 2020 22:33:04 GMT):
and document the whole ledger transactions

MrWoody (Thu, 30 Jul 2020 22:33:10 GMT):
for now, I can visually do it with Greenlight

MrWoody (Thu, 30 Jul 2020 22:33:20 GMT):
but if indy-cli could do it, it would help

andrew.whitehead (Thu, 30 Jul 2020 22:52:00 GMT):
You could use the indy-vdr-proxy tool at https://github.com/hyperledger/indy-vdr

MrWoody (Thu, 30 Jul 2020 23:03:09 GMT):
Oh, another tool... ok, I'll take a look at it. No way to do it via simple indy-cli right?

andrew.whitehead (Thu, 30 Jul 2020 23:04:05 GMT):
Not sure if it has a method to fetch transactions

MrWoody (Thu, 30 Jul 2020 23:04:46 GMT):
Ok.. I'll try VDR in the meantime. Thanks

WadeBarnes (Fri, 31 Jul 2020 13:17:04 GMT):
The `indy-cli` does not support dumping the transactions.

WadeBarnes (Fri, 31 Jul 2020 13:18:35 GMT):
As @andrew.whitehead has recommended `hyperledger/indy-vdr` is the right direction.

MrWoody (Fri, 31 Jul 2020 15:29:40 GMT):
thank you, guys!

jameshiester (Mon, 03 Aug 2020 19:22:39 GMT):
How is node naming/identification handled? Is the name guaranteed unique? Or would it be better to use the did/verkey as a primary key if storing in a db (though can't a node rotate their keys?)

jameshiester (Tue, 04 Aug 2020 15:49:04 GMT):
was anonymous mode for validatorStatus released in 1.12.3?

swcurran (Tue, 04 Aug 2020 15:51:52 GMT):
I believe it has been there for a while. It's just checking public information about all the nodes in the network. @andrew.whitehead might know more.

MurugesanRathinam (Wed, 05 Aug 2020 05:16:32 GMT):
Hi All

MurugesanRathinam (Wed, 05 Aug 2020 05:17:30 GMT):
i am trying to understand indy workflow

MurugesanRathinam (Wed, 05 Aug 2020 05:18:44 GMT):
it would be helpfull for me if some some one gives me environment setup document

andrew.whitehead (Wed, 05 Aug 2020 15:51:41 GMT):
It's not an indy-node feature, we just perform a GET_TXN request against the individual nodes instead of VALIDATOR_INFO

jameshiester (Wed, 05 Aug 2020 15:55:10 GMT):
ah ok, and then an error response is equivalent to a node being offline? for vdr, are there plans to add a status endpoint that shows nodes online/offline? @swcurran alluded to as much

swcurran (Wed, 05 Aug 2020 15:58:19 GMT):
Not in indy-vdr, but look at indy-node-mointor, which is based on indy-vdr

swcurran (Wed, 05 Aug 2020 15:58:32 GMT):
https://github.com/bcgov/indy-node-monitor

jameshiester (Wed, 05 Aug 2020 16:02:45 GMT):
ahh ok. Personally I would say it's a little misleading that the '/' endpoint in vdr returns all nodes regardless of whether they're online or not.

andrew.whitehead (Wed, 05 Aug 2020 16:05:25 GMT):
That's meant as more of a 'pool refresh status' page to see if it has the most recent pool transactions

MurugesanRathinam (Thu, 06 Aug 2020 07:56:30 GMT):
indy-node : Depends: indy-plenum (= 1.13.0~dev1021) but it is not going to be installed E: Unable to correct problems, you have held broken packages.

MurugesanRathinam (Thu, 06 Aug 2020 07:56:54 GMT):
Hi All,

MurugesanRathinam (Thu, 06 Aug 2020 07:57:07 GMT):
how t resolve above error

MurugesanRathinam (Thu, 06 Aug 2020 07:57:15 GMT):
Thank you

tomsoft1 (Thu, 06 Aug 2020 13:23:12 GMT):
Has joined the channel.

tomsoft1 (Thu, 06 Aug 2020 13:23:13 GMT):
Does Indy users, is there an existing network up and running that could be used to do some tests?

Primordium (Thu, 06 Aug 2020 14:01:48 GMT):
what linux distro are you using? You can try to install the version you need with `apt-get install indy-plenum=1.13.0~dev1021`

swcurran (Thu, 06 Aug 2020 16:55:58 GMT):
Easiest way to run a local network is using von-network, which also gives you guidance on how you can run your own on bare-metal (which is not very useful IMHO - depends on your goals). https://github.com/bcgov/von-network BC Gov operates some open instances of von-network -- listed here (BCovrin links - https://vonx.io/clickythings/) Closer to production are the three Sovrin networks - MainNet, Staging and BuilderNet -- https://sovrin.org

esplinr (Thu, 06 Aug 2020 17:35:04 GMT):
I don't know what you mean by "indy workflow".

esplinr (Thu, 06 Aug 2020 17:35:04 GMT):
I don't know what you mean by "indy workflow". Perhaps if you are more specific people can provide more help.

esplinr (Thu, 06 Aug 2020 17:36:07 GMT):
Check out the Sovrin BuilderNet: https://forum.sovrin.org/t/how-to-access-the-public-sovrin-network/1035

lynn.bendixsen (Thu, 06 Aug 2020 20:32:43 GMT):
Is Indy susceptible to the "51% attack"? why or why not? Perhaps instead it is susceptible to a "3f+1 attack"? I am answering this question to a lawyer, so some official looking references would be nice if there are any. Thx

ianco (Thu, 06 Aug 2020 22:32:19 GMT):
I think the Plenum protocol is what you're looking for: https://github.com/hyperledger/indy-plenum/wiki#:~:text=The%20Plenum%20protocol,Ben%20Mokhtar%2C%20and%20Vivien%20Qu%C3%A9ma.

esplinr (Thu, 06 Aug 2020 23:05:57 GMT):
You are correct that an Indy attack would have to take over 3f +1 machines to produce the BSL signature necessary to hijack the ledger history. This is because it is a permissioned ledger, versus a permissionless ledger like the Blockchain.

esplinr (Thu, 06 Aug 2020 23:05:57 GMT):
@lynn.bendixsen You are correct that an Indy attack would have to take over 3f +1 machines to produce the BSL signature necessary to hijack the ledger history. This is because it is a permissioned ledger, versus a permissionless ledger like the Blockchain.

esplinr (Thu, 06 Aug 2020 23:06:30 GMT):
@ianco is correct that the Plenum protocol is what provides this protection.

kclinux (Thu, 06 Aug 2020 23:59:18 GMT):
Has left the channel.

lynn.bendixsen (Fri, 07 Aug 2020 13:11:52 GMT):
Thanks @ianco and @esplinr , that is just what I needed!

jmason900 (Sat, 08 Aug 2020 01:57:59 GMT):
I'm looking to standup an Indy test network. I realize the Indy readthedocs site doesn't have current info for SDK ( even Python scripts ) on the sample scripts. Can someone point me to better documentation resources? Is there a good end-to-end tutorial recommended that shows the Indy setup, creating a steward, trust anchor and issuing identities and credentials for individuals? Thanks !

swcurran (Sat, 08 Aug 2020 20:12:39 GMT):
A good place to start to get a local network running is here: https://github.com/bcgov/von-network It spins up a 4-node network for you, with a web interface that lets you create DIDs that can be used to create other objects.

swcurran (Sat, 08 Aug 2020 20:15:35 GMT):
Unless you are planning to be an Indy network operator (which seems unlikely) manually setup an Indy network is not something you really need to know. Interacting with a network is what's interesting --- either in building Agents (e.g. developing at the Aries level), or building agents for specific business uses -- issuers, holders and verifiers.

swcurran (Sat, 08 Aug 2020 21:29:33 GMT):
Indy devs ( @esplinr ) -- How flexible are the configuration rules for writing transactions to the ledger -- e.g. rules like when an endorser/trustee/steward needs to sign a transaction to write (for example) a DID? Notably, can the rules for writing a RevReg reference the owner of the CredDef to which the RevReg is to be written? Likewise, can a RevEntry be constrained to only be written by the owner of the RevReg to which the RevEntry is associated?

esplinr (Sun, 09 Aug 2020 06:27:12 GMT):
The documentation is here: https://github.com/hyperledger/indy-node/blob/master/docs/source/transactions.md#auth_rule

esplinr (Sun, 09 Aug 2020 06:27:56 GMT):
It was intended to be flexible, but we didn't specifically address the Revocation Registry use case.

swcurran (Sun, 09 Aug 2020 20:23:35 GMT):
FYI - upcoming virtual conference: *Indy Interop-athon - Making "Network of Networks" Real*. This is a collaborative technical design event to bring Indy/Aries open source identity collaborators together for two days of focused effort to enable the following user story: > *As a prover, I want to be able to create a proof that includes claims from credentials rooted in different Indy networks that a verifier can verify so that I can easily interact with issuers using different Indy networks.* If you are deep into the DID Specification, Indy and Aries, with ideas about how to accomplish the user story, please join us. For more information and to register, please go here: https://bit.ly/indyinterop20

swcurran (Sun, 09 Aug 2020 20:23:35 GMT):
FYI - upcoming virtual conference: *Indy Interop-athon - Making "Network of Networks" Real* (Sept. 1st and 2nd). This is a collaborative technical design event to bring Indy/Aries open source identity collaborators together for two days of focused effort to enable the following user story: > *As a prover, I want to be able to create a proof that includes claims from credentials rooted in different Indy networks that a verifier can verify so that I can easily interact with issuers using different Indy networks.* If you are deep into the DID Specification, Indy and Aries, with ideas about how to accomplish the user story, please join us. For more information and to register, please go here: https://bit.ly/indyinterop20

swcurran (Sun, 09 Aug 2020 20:23:35 GMT):
FYI - upcoming virtual conference: *Indy Interop-athon - Making "Network of Networks" Real* (Sept. 1st and 2nd). This is a collaborative technical design event to bring Indy/Aries open source identity collaborators together for two days of focused effort to enable the following user story: > > *As a prover, I want to be able to create a proof that includes claims from credentials rooted in different Indy networks that a verifier can verify so that I can easily interact with issuers using different Indy networks.* > If you are deep into the DID Specification, Indy and Aries, with ideas about how to accomplish the user story, please join us. For more information and to register, please go here: https://bit.ly/indyinterop20

jameshiester (Mon, 10 Aug 2020 00:50:56 GMT):
in a proof, how do the signatures show that the holder was issued the credential if the holder/creator of the proof is using a different did than the one to which the credential was issued (as in the case of two pairwise did connections)?

calvin.heo (Mon, 10 Aug 2020 06:08:56 GMT):
Has joined the channel.

MurugesanRathinam (Mon, 10 Aug 2020 09:07:19 GMT):
While lock file: /var/lib/indy/sandbox/data/Node1/config_merkleNodes/LOCK: Resource temporarily unavailable'

MurugesanRathinam (Mon, 10 Aug 2020 09:10:57 GMT):
while am starting node i am facing below issue

swcurran (Mon, 10 Aug 2020 13:29:59 GMT):
In Indy anoncreds, the holder provides the issuer a blinded link secret that is inserted into the credential before it is issued. To prove the credential, the holder must prove (in zero knowledge) that they possess the link secret to unblind the the blinded link secret. That proves that the entity that constructed the proof was the one to which the credential was issued. When there are multiple credentials sourced into a single proof, all must use the same link secret. The pairwise DIDs are only for secure communication. They are not used as part of the credential exchange process.

jameshiester (Mon, 10 Aug 2020 13:35:49 GMT):
also just to make sure, rotating keys would not affect that ability to unblind correct?

swcurran (Mon, 10 Aug 2020 13:39:16 GMT):
In anoncreds, you don't change the link secret --- it's held in the wallet and never exposed. Rotating keys of the DIDs used for DIDComm has no impact on the link secret or the unblinding process.

cam-parra (Mon, 10 Aug 2020 15:08:59 GMT):
@swcurran is the meeting for indy contributors cancelled this morning?

swcurran (Mon, 10 Aug 2020 15:15:02 GMT):
The indy contributors meetings are on Tuesdays at 8AM Pacific -- next one is next week. The Hyperledger calendar is broken and the update we submitted about the change is not displayed.

cam-parra (Mon, 10 Aug 2020 15:15:25 GMT):
thanks! @nage ^^

esplinr (Mon, 10 Aug 2020 15:34:52 GMT):
Also, see the note in #indy-contributors . We can collaborate there, and don't need to wait for the next meeting.

lynn.bendixsen (Mon, 10 Aug 2020 18:21:47 GMT):
In looking at the auth rules, and reviewing the ledger, I am pretty sure that a rev_reg_def is created at the same time as a credential definition by the same owner and is directly associated with that cred_def. The auth_rules already indicate that any rev_reg "add"s or "edit"s must be done by the owner. `"need_to_be_owner": true` So, Yes, the RevReg entries's can be and are constrained to only be written to, by the owner of the cred_def/rev_reg_def.

WadeBarnes (Mon, 10 Aug 2020 19:23:09 GMT):
The rules appear to be different for each ledger.

WadeBarnes (Mon, 10 Aug 2020 19:25:51 GMT):
For MainNet it looks like you need a privileged DID in order to *ADD* a *REVOC_REG_DEF* or *REVOC_REG_ENTRY*

WadeBarnes (Mon, 10 Aug 2020 19:27:18 GMT):
What is the `owner` context for *REVOC_REG_DEF* txns? I assume it would be the `owner` of the *CRED_DEF*. Is that correct?

WadeBarnes (Mon, 10 Aug 2020 19:30:28 GMT):
My thoughts: - The owner of a *CRED_DEF* should be able to freely ADD/EDIT *REVOC_REG_DEF* txns for their *CRED_DEF*(s), and freely ADD/EDIT *REVOC_REG_ENTRY* txns for their owned *REVOC_REG_DEF*(s).

WadeBarnes (Mon, 10 Aug 2020 19:30:28 GMT):
My thoughts: - The owner of a *CRED_DEF* should be able to freely ADD/EDIT *REVOC_REG_DEF* txns for their *CRED_DEF*(s), and freely ADD/EDIT *REVOC_REG_ENTRY* txns for their owned *REVOC_REG_DEF*(s). - You should not need a privileged DID in order to do this. - I would expect there would be some fee associated to writing these txns, similar to the way there is a fee for writing attributes.

WadeBarnes (Mon, 10 Aug 2020 19:51:22 GMT):
The [Who Is Owner](https://github.com/hyperledger/indy-node/blob/master/docs/source/auth_rules.md#who-is-owner) section of the *Default AUTH_MAP Rules* document does not list the owner in the context of the *REVOC_REG_DEF*. How do we confirm the owner in this context?

lynn.bendixsen (Mon, 10 Aug 2020 19:51:40 GMT):
I can help explain the rules part of this. The rules I was referencing in my previous response were for the MainNet and they are mostly the same for the other networks. The main differences are that on MainNet only an endorser can ADD a credDef/RegDef. This means that they are also the only ones that can ADD/EDIT entries or other related actions. A Transaction Author must find an Endorser to sign these actions for them. This was done because Endorser was the role assigned to do all of the credential issuance type txn's while the ledger remains permissioned. Allof those txn's are fee based and are charged to the Endorser monthly. I am almost certain that the owner context for RegDef txns is the owner of the RegDef. Since the reg def is created in conjunction with a credDef, the owner of the RegDef is the same as the associated credDef. This appears to be a 1:1 relationship and I am uncertain if you can make multiple RegDef's per CredDef.

lynn.bendixsen (Mon, 10 Aug 2020 19:54:15 GMT):
You can confirm "owner" type issues by trying things out on the BuilderNet, but I am pretty certain that "owner" always refers to the original signer of the TXN.

WadeBarnes (Mon, 10 Aug 2020 19:54:33 GMT):
As I understand it you can have more than one *REVOC_REG_DEF* for a given *CRED_DEF* since the *REVOC_REG_DEF* only has limited use.

WadeBarnes (Mon, 10 Aug 2020 19:55:25 GMT):
This is also the basis of my thoughts.

WadeBarnes (Mon, 10 Aug 2020 19:55:25 GMT):
This is also the basis of my thoughts of the *CRED_DEF* owner being able to freely write a new *REVOC_REG_DEF* without the need to have the txn signed by a second or third party.

lynn.bendixsen (Mon, 10 Aug 2020 19:56:57 GMT):
That muddies the waters a bit then, but creating a RegDef for a CredDef that you do not have ownership of makes no sense.

WadeBarnes (Mon, 10 Aug 2020 19:57:33 GMT):
Agreed. That's why I'd like to confirm the `owner` context.

WadeBarnes (Mon, 10 Aug 2020 19:57:33 GMT):
Agreed. That's why I'd like to confirm the `owner` context for the *REVOC_REG_DEF*

WadeBarnes (Mon, 10 Aug 2020 19:57:33 GMT):
Agreed. That's why I'd like to confirm the `owner` context for the *REVOC_REG_DEF*. To confirm `"need_to_be_owner": true` on a *REVOC_REG_DEF* means you need to be the owner of the associated *CRED_DEF*.

lynn.bendixsen (Mon, 10 Aug 2020 19:59:42 GMT):
I think I was wrong when I said "I am pretty certain that "owner" always refers to the original signer of the TXN." I believe that the "owner" is actually the txn author, or writer of the txn (not the signer or endorser, unless the endorser is also the author).

WadeBarnes (Mon, 10 Aug 2020 20:02:33 GMT):
If the above is true, `"need_to_be_owner": true` on a *REVOC_REG_DEF* means you need to be the owner of the associated *CRED_DEF*, then we can adjust the rules on ADD/EDIT for *REVOC_REG_DEF* and *REVOC_REG_ENTRY* so they make sense for the real world use cases we're working though right now.

lynn.bendixsen (Mon, 10 Aug 2020 20:03:29 GMT):
Sorry, but I don't know for sure on the RegDef "ownership" question. In all of the ownership cases that I have tested "owner" means writer of the txn, not "writer of some other assoctiated txn"

WadeBarnes (Mon, 10 Aug 2020 20:04:59 GMT):
Example: *REVOC_REG_ENTRY*, owner means the owner of the DID used to create the corresponding *REVOC_REG_DEF*.

WadeBarnes (Mon, 10 Aug 2020 20:04:59 GMT):
Example: *REVOC_REG_ENTRY* txns, owner means the owner of the DID used to create the corresponding *REVOC_REG_DEF*.

lynn.bendixsen (Mon, 10 Aug 2020 20:07:25 GMT):
You could test on builderNet by adding a new credDef as an Endorser, then attempting to add a new RegDef for it as a Trustee. Trustees can create RegDef's on BuilderNet but "owner:true" is set appropriately...

lynn.bendixsen (Mon, 10 Aug 2020 20:09:46 GMT):
I am not sure yet why we would need to change the rules if what you say is true. Perhaps a zoom chat will help me understand what needs changed?

lynn.bendixsen (Mon, 10 Aug 2020 20:09:46 GMT):
I am not sure yet why we would need to change the rules if what you say is true. Perhaps a zoom chat will help me understand what needs changed and why?

WadeBarnes (Mon, 10 Aug 2020 20:10:30 GMT):
Sure, zoom chat works

lynn.bendixsen (Mon, 10 Aug 2020 20:11:02 GMT):
us02web.zoom.us/j/5977134207

WadeBarnes (Mon, 10 Aug 2020 20:12:09 GMT):
In the waiting room

WadeBarnes (Mon, 10 Aug 2020 20:27:43 GMT):
Thanks @lynn.bendixsen, the existing rules make far more sense now.

WadeBarnes (Mon, 10 Aug 2020 20:55:33 GMT):
I think I found the code, and it does not look like the validation for the *REVOC_REG_DEF* (https://github.com/hyperledger/indy-node/blob/master/indy_node/server/request_handlers/domain_req_handlers/revoc_reg_def_handler.py#L43) validates an owner (for the *CRED_DEF*) like the validation for the *REVOC_REG_ENTRY* (https://github.com/hyperledger/indy-node/blob/master/indy_node/server/request_handlers/domain_req_handlers/revoc_reg_entry_handler.py#L31) does (for the *REVOC_REG_DEF*).

WadeBarnes (Mon, 10 Aug 2020 20:56:46 GMT):
Does anyone know the history of why such a check is not performed?

jameshiester (Tue, 11 Aug 2020 02:21:55 GMT):
has anyone tried running the indy-sdk inside an aws lambda function?

Vidyashankar (Tue, 11 Aug 2020 05:24:57 GMT):
Has joined the channel.

Vidyashankar (Tue, 11 Aug 2020 05:24:57 GMT):
Hi team, i am trying to implement Ursa in Hyperledger Indy, could you please assist me with the basics of configuring the same ?

anchit (Tue, 11 Aug 2020 07:59:43 GMT):
Has joined the channel.

blaz (Tue, 11 Aug 2020 13:51:11 GMT):
Has joined the channel.

tomsoft1 (Tue, 11 Aug 2020 13:55:02 GMT):
I totally agree, but : 1) the von-network demo does not deploy anymore (have setup an ubuntu box for this, issue with lib version that seems to be know) 2) what is the network that you recommand to connect to to play with indy SDK and tools?

WadeBarnes (Tue, 11 Aug 2020 13:56:36 GMT):
@tomsoft1, can you provide more details on your first issue? I was using the latest version of `von-network` yesterday and it's working just fine.

tomsoft1 (Tue, 11 Aug 2020 13:57:47 GMT):
ok, let me try again

WadeBarnes (Tue, 11 Aug 2020 13:59:36 GMT):

Clipboard - August 11, 2020 6:59 AM

WadeBarnes (Tue, 11 Aug 2020 13:59:53 GMT):
Running locally just now. ^

WadeBarnes (Tue, 11 Aug 2020 14:01:02 GMT):
Make sure you get the latest and rebuild (``./manage build`).

WadeBarnes (Tue, 11 Aug 2020 14:01:22 GMT):
Please report any issues.

tomsoft1 (Tue, 11 Aug 2020 14:01:52 GMT):
thanks, let me recheck

tomsoft1 (Tue, 11 Aug 2020 14:33:25 GMT):
Sorry Wade, my bad I've mixed the Von network with another sample. It started, however I have an "Error initializing pool ledger" in the interface (as a result of the "status" call), even if it seems that all the nodes seems to be running well

WadeBarnes (Tue, 11 Aug 2020 14:35:06 GMT):
Try resetting things: `./manage down` wait for command to complete `./manage start`

WadeBarnes (Tue, 11 Aug 2020 14:36:00 GMT):
That error is typically a startup timing issue with the ledger browser connecting to the nodes.

WadeBarnes (Tue, 11 Aug 2020 14:37:56 GMT):
The startup timing can be adjusted here; https://github.com/bcgov/von-network/blob/master/docker-compose.yml#L25

WadeBarnes (Tue, 11 Aug 2020 14:38:45 GMT):
By changing the delay before startup.

tomsoft1 (Tue, 11 Aug 2020 14:41:37 GMT):
Good shot, a down and a restart have done the job! Thanks again

mccown (Wed, 12 Aug 2020 19:39:21 GMT):
Just a reminder about the Identity Implementer's WG Call tomorrow morning (Aug 13th @ 9am MT). This is a chance to get updates on the WG calls you may have missed this week. Our guest presenter will be Sam Smith of ProSapien. Sam will be presenting an overview of the new KERI (Key Event Receipt Infrastructure) methodology. KERI creates a self-certifying identifier that provides the basis for a universal DID method (DID:uni). Here are links to the Wiki and the Zoom meeting. Wiki: https://wiki.hyperledger.org/display/IWG/2020-08-13+Identity+WG+Implementers+Call Zoom: https://zoom.us/j/244779296

AmanAgrawal (Fri, 14 Aug 2020 06:26:54 GMT):
Has left the channel.

saanvijay (Sun, 16 Aug 2020 16:35:01 GMT):
Has joined the channel.

swcurran (Sun, 16 Aug 2020 17:40:44 GMT):
In talking with the Indy team some time ago I was told that "the sweet point" for the size of a production Indy ledger was 25 nodes --- large enough to be resilient (allowing up to 8 node failures), but small enough to have reasonable consensus performance. I was asked about whether there is documentation about the testing done to come up with that metric and I don't have any. @esplinr -- do you know of such information? Anyone else? Thanks

esplinr (Mon, 17 Aug 2020 14:48:09 GMT):
I don't understand your question. Indy includes Ursa as a dependency at compile time. You don't have to do anything to "implement It".

esplinr (Mon, 17 Aug 2020 14:52:17 GMT):
The work was done in a series of Jira issues: https://jira.hyperledger.org/browse/INDY-1343

esplinr (Mon, 17 Aug 2020 14:52:17 GMT):
The work was done in a series of Jira issues: https://jira.hyperledger.org/browse/INDY-1343 https://jira.hyperledger.org/browse/INDY-1717 https://jira.hyperledger.org/browse/INDY-2214

esplinr (Mon, 17 Aug 2020 15:01:13 GMT):
And was discussed in this sprint review: https://www.youtube.com/watch?time_continue=1&v=OSpssHQXlOI

esplinr (Mon, 17 Aug 2020 15:01:13 GMT):
And was discussed in a few sprint reviews: https://www.youtube.com/watch?time_continue=1&v=OSpssHQXlOI https://www.youtube.com/watch?v=g5NMhxTlkDI&list=PLRp0viTDxBWGLdZk0aamtahB9cpJGV7ZF&index=44 https://www.youtube.com/watch?time_continue=1&v=OSpssHQXlOI

esplinr (Mon, 17 Aug 2020 15:01:33 GMT):
Looking through those old tickets, it doesn't look like we ever did the test with 30 nodes. We just did the theoretical analysis.

JamesSchulte (Mon, 17 Aug 2020 18:20:11 GMT):
Has joined the channel.

sj1 4 (Wed, 19 Aug 2020 07:27:06 GMT):
Has joined the channel.

sj1 4 (Wed, 19 Aug 2020 07:27:06 GMT):
timeout

Lucampelli (Wed, 19 Aug 2020 19:41:25 GMT):
Has joined the channel.

Lucampelli (Wed, 19 Aug 2020 19:41:26 GMT):
Good Evening/ Night, I was wondering how Indy handles credential updating. For example someone's identity document credential, or the registry of their residence when they move.

Lucampelli (Wed, 19 Aug 2020 19:41:54 GMT):
Is it the most recent credential of the same type that counts?

avi23 (Wed, 19 Aug 2020 19:43:54 GMT):
Has joined the channel.

avi23 (Wed, 19 Aug 2020 19:45:14 GMT):
Hello, Does each validator holds a separate node in Indy?

lynn.bendixsen (Wed, 19 Aug 2020 19:51:31 GMT):
Yes, if I understand your question correctly. In Indy we regularly use the terms "Validator" and "Node" to mean the same thing.

avi23 (Wed, 19 Aug 2020 19:54:16 GMT):
okay, what i mean to ask is how many minimum and maximum validators can be there in any indy network? and does each validator need to assign a separate node in system?

lynn.bendixsen (Wed, 19 Aug 2020 19:59:02 GMT):
Minimum is 4, maximum is unknown, but we recommend maxing out at 25 validator nodes because after that there is a significant performance reduction. Yes, each Validator is a single node in our system (a separate server is required for each Validator Node in Sovrin's networks, for example)

avi23 (Wed, 19 Aug 2020 20:00:22 GMT):
thanks @lynn.bendixsen for the reply..

avi23 (Wed, 19 Aug 2020 20:02:20 GMT):
@lynn.bendixsen could you also please confirm that is it mandatory that Each holder and Issuer also holds a separate node?

avi23 (Wed, 19 Aug 2020 20:03:09 GMT):
i mean who all parties in any Indy network should be assign a node?

lynn.bendixsen (Wed, 19 Aug 2020 20:11:56 GMT):
I am not sure what you mean by 'holder' and 'issuer' in this context. Perhaps you are using the terms differently? In "Indy/Aries" terminology an "Issuer" and a "Holder" are terms used in delivering and having credentials, and there are an indeterminately large number of those that can exist on each Indy Network, regardless of how many nodes are in the network. In the Sovrin Indy Networks, there are Stewards who have volunteered to run the nodes (based on their governance documents). Each Steward is only allowed to run one Node. In a theoretical generic Indy network, this boils down to "each STEWARD DID on the ledger can only instantiate one Indy node".

avi23 (Wed, 19 Aug 2020 20:16:26 GMT):
thanks for the insights @lynn.bendixsen , so if i ask in general that how many nodes other than validators as you mentioned above is required to form a indy network? i mean i am just curios to know in perspective of a network setup

lynn.bendixsen (Wed, 19 Aug 2020 20:18:49 GMT):
None, there are currently no Indy nodes that are not validator nodes in an Indy Network.

avi23 (Wed, 19 Aug 2020 20:19:38 GMT):
Thanks:slight_smile:

avi23 (Wed, 19 Aug 2020 20:20:42 GMT):
Also , it would be great if you can share any github repo which i can use for sample network setup of Indy

lynn.bendixsen (Wed, 19 Aug 2020 20:22:53 GMT):
There is a "testnetwork" setup section in the readme of this repo: https://github.com/hyperledger/indy-node

avi23 (Wed, 19 Aug 2020 20:24:40 GMT):
Thanks much @lynn.bendixsen for all support

avi23 (Wed, 19 Aug 2020 20:24:47 GMT):
will go through it

TimoGlastra (Wed, 19 Aug 2020 20:41:46 GMT):
Hi, could anyone tell me the difference between full length verkeys and abbreviated verkeys? Why would I want to use one over the other? Looking at the indy code whether an abbreviated verkey is used is dependant on the `cid` parameter. What does `cid` mean? I tried to find it somewhere in the docs, code, etc.. but no luck.

andrew.whitehead (Wed, 19 Aug 2020 21:25:56 GMT):
The verkey is 32 bytes, and by default in Indy the DID is chosen as the first 16 bytes of that key, before base58 encoding. The abbreviated verkey just replaces the first 16 bytes with ~ when it matches the DID.

lynn.bendixsen (Wed, 19 Aug 2020 22:03:06 GMT):
I am not an expert, but because I was curious... Everything I see when googling "cid hyperledger indy" seems to indicate the following: 1. cid means cryptonym ID 2. cid is deprecated 3. the cid parameter you refer to indicates the length of the DID/CID and not the length of the verkey. (or at least it used to, https://docs.rs/crate/indy/0.1.1-70/source/src/api/signus.rs)

TimoGlastra (Wed, 19 Aug 2020 22:34:40 GMT):
Thanks @andrew.whitehead and @lynn.bendixsen . That makes sense. I now see the `cid` indeed indicates the length of the did, not the verkey.

AlexanderHoughton (Wed, 19 Aug 2020 22:50:17 GMT):
Has joined the channel.

swcurran (Wed, 19 Aug 2020 23:25:34 GMT):
The expected model is that the there would be a revocation of the current credential and a reissue of a new one.

swcurran (Wed, 19 Aug 2020 23:28:49 GMT):
Just to be very clear in this conversation -- if your are talking about a issuer and verifier in the Verifiable Credentials sense of the terms, neither of them needs run a node on the Indy Network. An issuer needs to be able to write certain transactions to the ledger and so needs permission to do that, and a verifier just needs to read from the network and that is public.

swcurran (Wed, 19 Aug 2020 23:30:39 GMT):
There is no "issuer" in the context of an Indy network node and the term "validator" is used to mean that the nodes validate what gets written to the ledger. All of that is well below the layer dealing with verifiable credentials.

thearhaam (Thu, 20 Aug 2020 10:30:29 GMT):
Has joined the channel.

HeMBaD (Thu, 20 Aug 2020 10:38:36 GMT):
Has joined the channel.

SamB (Thu, 20 Aug 2020 14:18:18 GMT):
Has joined the channel.

Lucampelli (Thu, 20 Aug 2020 15:17:16 GMT):
I see, thank you!

Lucampelli (Thu, 20 Aug 2020 15:20:41 GMT):
Hello again! I was wondering how and where is the wallet object stored? Locally? The walkthrough and examples make it look like a serializable object. So my guess would be to store it in a cryptographically secure file in the user's machine, or file storage.

swcurran (Thu, 20 Aug 2020 15:33:02 GMT):
Yes. It's stored in a database which can be local or on another server. The database is pluggable and there are SQLite and Postgres plugins available.

Lucampelli (Thu, 20 Aug 2020 18:08:21 GMT):
Nice! Could you point me out to one of the plugins? Would python-mysqlconnector work?

swcurran (Thu, 20 Aug 2020 18:40:08 GMT):
Postgres plug in is here. It's in the samples for historical reasons. It's in production in many places. https://github.com/hyperledger/indy-sdk/tree/master/samples/postgres-plugin/nodejs

swcurran (Thu, 20 Aug 2020 18:42:16 GMT):
Postgres is here. It's in experimental for historical reasons -- it's been in production for a long time. https://github.com/hyperledger/indy-sdk/tree/master/experimental/plugins/postgres_storage

swcurran (Thu, 20 Aug 2020 18:43:07 GMT):
MySQL could be done, but it's non-trivial to create a plugin. Not too much work, but more than just a config.

swcurran (Thu, 20 Aug 2020 18:44:38 GMT):
What is your interest. We are currently working on Aries Storage as the next generation of what is the Indy Wallet. Based on some of the concepts in Indy Wallet, but adjusted based on learnings from using Indy Wallet and new non-Indy requirements.

Lucampelli (Thu, 20 Aug 2020 18:56:07 GMT):
I'm writing my master's degree based on Indy and SSI for the issuance of Diplomas. Right now I'm studying Indy to understand how it works, and how SSI works in general. And these questions popped up while running the SDK examples.

SamB (Fri, 21 Aug 2020 05:28:51 GMT):
Hi Folks, I am wondering how exactly transaction flow works in Indy? Could you please help me to understand? Or provide some links to get more clear idea about it. Thank you

mapa 5 (Fri, 21 Aug 2020 12:48:19 GMT):
hi falks im wrapping a dart with libindy.so

mapa 5 (Fri, 21 Aug 2020 12:48:19 GMT):
hi folks im wrapping a dart with libindy.so

mapa 5 (Fri, 21 Aug 2020 12:48:19 GMT):
hi folks im wrapping a dart with ffi interaction libindy.so

mapa 5 (Fri, 21 Aug 2020 12:48:46 GMT):
need help setting standards

avi23 (Sat, 22 Aug 2020 09:19:19 GMT):
thanks @swcurran for summarizing!

avi23 (Sat, 22 Aug 2020 12:26:14 GMT):
@swcurran @lynn.bendixsen is there any concepts of smart contracts in Indy? i mean how and where we write business logics to perform operations?

jcldnatv (Sat, 22 Aug 2020 17:14:09 GMT):
Has joined the channel.

swcurran (Sat, 22 Aug 2020 19:45:56 GMT):
There are none. Indy is designed for identity -- DIDs and objects related to Verifiable Credentials. It is not used for asset transfer or other general purpose applications.

MarinaELias (Mon, 24 Aug 2020 14:25:15 GMT):
Hi all, I am currently working on a project but I am confused in some concepts. First, where is the wallet that stores the credentials saved? Is it saved on a blockchain or just a local database? Secondly, I still don't get how the verifier entity verifies the correctness of the presented values? From what I understood, the personal data of the student or the holder of a transcript are not saved on the ledger, so what does the verifier compare against the ledger if the data is not there? I would appreciate your help, thank you in advance.

jameshiester (Mon, 24 Aug 2020 16:11:18 GMT):
The wallet, including credentials, are stored in database. This can either be the default local database (I believe it's rocksDB or sqlite), or use a storage plugin such as the postgresql plugin, which would store the wallet in any postgres database (which can be either local or in the cloud). With respect to verifying the "correctness" of the presented values, it is more correct to say that the verifier is verifying the source of the presented values, which is the authority that issued the credential. The issuer authority has a public verkey that is published to the ledger, along with a key associated with the credential definition. These keys are used to "sign" the credential. When generating this signature, the private keys associated with the authority issuer and credential definition are used, as well as the actual value of the credential. This ensures that only the issuer authority can generate a valid signature, and the properties of the credential cannot be modified without invalidating the signature. Therefore, by validating the signature, the verifier is confirming the source and veracity of the credential data.

jameshiester (Mon, 24 Aug 2020 16:13:13 GMT):
Note: I'm not 100% certain that's accurate about the private key of the credential definition, but I am confident about the general gist of the signature and it's use

MarinaELias (Mon, 24 Aug 2020 16:19:26 GMT):
Thank you so much for your reply! I understand now, but my last question is from where does the verifier get the signature on the credential, is it send with the data that is presented to the verifier from the chosen credential?

MarinaELias (Mon, 24 Aug 2020 16:19:26 GMT):
@jameshiester Thank you so much for your reply! I understand now, but my last question is from where does the verifier get the signature on the credential, is it send with the data that is presented to the verifier from the chosen credential?

jameshiester (Mon, 24 Aug 2020 16:22:32 GMT):
yes, credentials are not send directly from the holder to the verifier, but rather as part of a presentation that the holder creates. This presentation includes the values (if the holder chooses to reveal them), and the signatures. This allows the holder to use multiple credentials in a single proof, as well as the ability to use zero-knowledge proofs (e.g. proving they are over 18 without including their actual birthdate in the proof).

jameshiester (Mon, 24 Aug 2020 16:22:32 GMT):
yes, credentials are not sent directly from the holder to the verifier, but rather as part of a presentation that the holder creates. This presentation includes the values (if the holder chooses to reveal them), and the signatures. This allows the holder to use multiple credentials in a single proof, as well as the ability to use zero-knowledge proofs (e.g. proving they are over 18 without including their actual birthdate in the proof).

MarinaELias (Mon, 24 Aug 2020 16:41:18 GMT):
Okay so in the end what proves the correctness of the credential is its digital signature. So what is the role of the blockchain besides storing the public keys of each entity, as the public keys could be stored anywhere else without the need for a blockchain. What am I missing?

MarinaELias (Mon, 24 Aug 2020 16:41:18 GMT):
@jameshiester Okay so in the end what proves the correctness of the credential is its digital signature. So what is the role of the blockchain besides storing the public keys of each entity, as the public keys could be stored anywhere else without the need for a blockchain. What am I missing?

jameshiester (Mon, 24 Aug 2020 17:01:35 GMT):
The blockchain acts as a decentralized authority for the keys, schemas, and revocation registries. While alternative systems are also feasible, the ledger provides an excellent way to achieve consensus around what format those keys and schemas should be published in, and a consistent and fault tolerant way of reading them. Additionally, the blockchain creates an immutable record, so it is easy to audit the entire trail of a schema or key back to who created it, and even who onboarded that identity and all the ancestor identities going back to the original genesis transaction. In short, it's a good way of allowing peer-to-peer credential sharing, while still creating an anchor of trust.

MarinaELias (Mon, 24 Aug 2020 18:51:18 GMT):
Okay, thank you so much for your help!

jbtwist (Tue, 25 Aug 2020 09:35:10 GMT):
Has joined the channel.

jbtwist (Tue, 25 Aug 2020 09:35:11 GMT):
Hi guys, I would like to ask if you know some real use cases for HL Indy

avi23 (Tue, 25 Aug 2020 10:11:53 GMT):
Hi All , can anyone please share some inputs and reference repo links on how to implement ZK-SNARK in Indy? just wondering if this has been achieved before in any Indy project?

avi23 (Tue, 25 Aug 2020 10:12:35 GMT):
@swcurran @lynn.bendixsen requesting your inputs please

avi23 (Tue, 25 Aug 2020 10:36:04 GMT):
@thearhaam

Optimmus (Tue, 25 Aug 2020 11:32:35 GMT):
Has joined the channel.

Optimmus (Tue, 25 Aug 2020 11:32:35 GMT):
Hello guys, i'm looking to know is there any official tutorial on how to import keys to the indy and use them

MarinaELias (Tue, 25 Aug 2020 12:05:20 GMT):
Hi again, I am sorry but there is something I still don't get, how does the Proof verify that the claims were issued to the presenter? What makes it impossible for someone to steal the student's identity?

jameshiester (Tue, 25 Aug 2020 12:19:06 GMT):
The answer lies in a secret passwed between the student and original issuer. The holder (student) provides the issuer a blinded link secret that is inserted into the credential before it is issued. To prove the credential, the holder must prove (in zero knowledge) that they possess the link secret to unblind the the blinded link secret. That proves that the entity that constructed the proof was the one to which the credential was issued. When there are multiple credentials sourced into a single proof, all must use the same link secret. That being said, no digital system is 100% effective in preventing identity theft. The secret is still stored in the users wallet, so if an attacker is able to get access to the wallet and decrypt it, they would be able to steal the student's identity. This is why wallets need to be protected with strong authentication, which very likely includes biometrics

lynn.bendixsen (Tue, 25 Aug 2020 13:21:49 GMT):
I don't know, sorry.

swcurran (Tue, 25 Aug 2020 14:33:40 GMT):
Indy does not use ZK-SNARK. I believe that Ursa has some support, but that Indy does not use them. Better asked on the Ursa channel.

thearhaam (Wed, 26 Aug 2020 04:48:46 GMT):
Hey :D so I was going through this repo: https://github.com/cloudcompass/ToIPLabs/tree/master/src/indy-material/nodejs the README mentions "This folder contains a demonstration of basic Hyperledger Indy and Aries Agents, which are built on the Ursa project." However the code is the same as the old indy alice demo. Is there another repo that I can look into instead?

swcurran (Wed, 26 Aug 2020 13:41:02 GMT):
The ACA-Py contains a far more advanced version that continues to evolve. Check it out - https://github.com/hyperledger/aries-cloudagent-python

thearhaam (Wed, 26 Aug 2020 17:37:47 GMT):
@swcurran Thank you, I've gone through the ACA-Py demo already. Currently I'm exploring Indy & Aries but will be looking into Ursa as well so wanted to know if there was an example or demo with Indy, Aries and Ursa all working together.

swcurran (Wed, 26 Aug 2020 17:51:27 GMT):
Indy is dependent on Ursa (includes Ursa components in the build). ACA-Py's support for Indy by embedding the Indy SDK. That's how Ursa gets into Indy and Aries.

swcurran (Wed, 26 Aug 2020 17:51:48 GMT):
So whenever you use ACA-Py, you are using Ursa.

mccown (Wed, 26 Aug 2020 19:58:23 GMT):
Just a reminder about the Identity Implementer's WG Call tomorrow morning (Aug 27th @ 9am MT). This is a chance to get updates on a variety of WG calls that you may have missed this week. Our guest presenter will be Sam Curren (Indicio). Sam will be presenting an overview of the DIDComm standards and technologies that enable agents to authenticate and communicate securely. Here are links to the Wiki and the Zoom meeting. Wiki: https://wiki.hyperledger.org/display/IWG/2020-08-27+Identity+Implementers+WG+Call Zoom: https://zoom.us/j/244779296

thearhaam (Thu, 27 Aug 2020 04:17:49 GMT):
@swcurran Oh I see, thank you so much for the info :grin:

SamB (Thu, 27 Aug 2020 07:47:57 GMT):
Hi Team, I am trying to set von-network on my Ubuntu 18.04, but getting error while starting server. :~/SSI/von-network$ GENESIS_URL=http://foo.bar/genesis.txt LEDGER_SEED=00000000000000000000000000000012 PORT=9000 python -m server.server Traceback (most recent call last): File "/usr/lib/python3.6/runpy.py", line 193, in _run_module_as_main "__main__", mod_spec) File "/usr/lib/python3.6/runpy.py", line 85, in _run_code exec(code, run_globals) File "/home/samruddhi/SSI/von-network/server/server.py", line 14, in from .anchor import ( File "/home/samruddhi/SSI/von-network/server/anchor.py", line 17, in * from indy import ledger, pool ModuleNotFoundError: No module named 'indy' *

avi23 (Thu, 27 Aug 2020 08:14:21 GMT):
@swcurran can we use Trinsic for any Indy usecase developement? is it a recommended tool?

rangeshsripathi (Thu, 27 Aug 2020 12:29:16 GMT):
"ModuleNotFoundError: No module named 'indy'" - Looks like you did not install packages.. please install them using pip3 install.

SamB (Thu, 27 Aug 2020 12:45:16 GMT):
Thank you. Can you please provide list of all packages required?

rangeshsripathi (Thu, 27 Aug 2020 12:46:02 GMT):
It should be in requirements.txt ... you could just do pip3 install -r requirements.txt

SamB (Thu, 27 Aug 2020 12:48:21 GMT):
After running requirements.txt, it is showing Requirement already satisfied... But getting the errors

rangeshsripathi (Thu, 27 Aug 2020 12:49:44 GMT):
Can you try to install pip3 install python3-indy and see if it works.

SamB (Thu, 27 Aug 2020 12:57:40 GMT):
I tried : le "/home/samruddhi/SSI/von-network/server/anchor.py", line 19, in from plenum.common.signer_simple import SimpleSigner ModuleNotFoundError: No module named 'plenum' this is popping now

SamB (Thu, 27 Aug 2020 12:57:40 GMT):
I tried : le "/home/SSI/von-network/server/anchor.py", line 19, in from plenum.common.signer_simple import SimpleSigner ModuleNotFoundError: No module named 'plenum' this is popping now

swcurran (Thu, 27 Aug 2020 15:42:17 GMT):
You bet! @RileyHughes might be able to help you with that.

avi23 (Thu, 27 Aug 2020 16:29:36 GMT):
@RileyHughes hello, could you please provide some inputs?

valesken (Thu, 27 Aug 2020 17:01:50 GMT):
Has joined the channel.

da3v21 (Thu, 27 Aug 2020 19:57:33 GMT):
Has joined the channel.

swcurran (Thu, 27 Aug 2020 21:42:40 GMT):
Let's try @rileyphughes --- he seems to have two accounts on RC.

RafaelAPB (Fri, 28 Aug 2020 12:31:42 GMT):
Has left the channel.

rileyphughes (Fri, 28 Aug 2020 13:28:18 GMT):
I can't get in to the old account to delete it... anyway, yeah @avi23 would love for you to try out the Trinsic platform and let us know what you think. We currently have a couple hundred developers actively using the platform. I recommend getting started here: https://docs.trinsic.id/

avi23 (Fri, 28 Aug 2020 13:55:28 GMT):
@rileyphughes Thanks for reply, does it also provide some support for credential revocation and implementing zkp in Indy?

pixelschnitzel (Fri, 28 Aug 2020 14:15:59 GMT):
Has joined the channel.

danielhardman (Sat, 29 Aug 2020 18:24:25 GMT):
I want to make sure the Indy community is aware of this conversation happening in the W3C, in which it has been proposed to eliminate serviceEndpoint from the DID core spec. This would be disruptive to DIDComm. It appears that the proposal has been rejected (DIDComm is safe), but there are still some voices claiming that serviceEndpoint is useless nonsense. You might want to read the ticket or weigh in. HEre's a link: https://github.com/w3c/did-core/issues/382#issuecomment-683325040

rileyphughes (Sat, 29 Aug 2020 22:47:25 GMT):
Yes to both

avi23 (Sun, 30 Aug 2020 09:13:14 GMT):
@swcurran i was referring to below repos for indy demo setup ..but couldn't find what is the difference between two ? are they both refer to the same or any difference in functionality?

avi23 (Sun, 30 Aug 2020 09:13:46 GMT):
sorry forgot to paste link i was referring ...here it is https://github.com/cloudcompass/indy-ssivc-tutorial https://github.com/cloudcompass/ToIPLabs/blob/master/src/indy-material/nodejs/RunningLocally.md

SamB (Mon, 31 Aug 2020 12:44:22 GMT):
Hello Team, I'm tring to set up for Indy-node and indy-plenum setup on Ubuntu 18.04.. But getting many errors regardring libsodium package.. Can anyone please help me to get cprrect steps to follow

rashmi3255 (Mon, 31 Aug 2020 18:14:55 GMT):
Has joined the channel.

rashmi3255 (Mon, 31 Aug 2020 18:14:56 GMT):
"Running the the web server on your local machine" for https://github.com/bcgov/von-network not working as per the steps given, I am trying to run the browser to point to my indy pool with below steps

rashmi3255 (Mon, 31 Aug 2020 18:14:56 GMT):
"Running the the web server on your local machine" for https://github.com/bcgov/von-network not working as per the steps given, I am trying to run the browser to point to my indy pool with below steps.

rashmi3255 (Mon, 31 Aug 2020 18:14:56 GMT):
"Running the the web server on your local machine" for https://github.com/bcgov/von-network not working as per the steps given, I am trying to run the browser to point to my indy pool with below steps. git clone https://github.com/bcgov/von-network.git cd von-network virtualenv --python=python3.6 venv source venv/bin/activate pip install -r server/requirements.txt GENESIS_FILE=/tmp/some-genesis.txt PORT=9000 python -m server.server

WadeBarnes (Mon, 31 Aug 2020 18:49:17 GMT):
You shouldn't have to run anything in python, it all runs in docker.

WadeBarnes (Mon, 31 Aug 2020 18:53:43 GMT):
For example, connecting the ledger browser to Sovrin BuilderNet: ``` ./manage build ./manage start GENESIS_URL=https://raw.githubusercontent.com/sovrin-foundation/sovrin/master/sovrin/pool_transactions_sandbox_genesis ```

WadeBarnes (Mon, 31 Aug 2020 18:53:53 GMT):
Wait for a bit.

WadeBarnes (Mon, 31 Aug 2020 18:54:17 GMT):
Browser to http://localhost:9000/

WadeBarnes (Mon, 31 Aug 2020 18:54:54 GMT):

Clipboard - August 31, 2020 11:54 AM

WadeBarnes (Mon, 31 Aug 2020 18:55:42 GMT):
You can now browse the transactions on that ledger;

WadeBarnes (Mon, 31 Aug 2020 18:55:44 GMT):

Clipboard - August 31, 2020 11:55 AM

danielhardman (Tue, 01 Sep 2020 02:21:23 GMT):
Slides I plan to use to frame the discussion in the interopathon session about "how to get network metadata" tomorrow: https://j.mp/3bcvZax

thearhaam (Tue, 01 Sep 2020 05:38:39 GMT):
@WadeBarnes I've been getting the following error after "./manage build" ./manage: line 321: pushd: too many arguments

WadeBarnes (Tue, 01 Sep 2020 12:41:14 GMT):
@thearhaam What OS are you running?

rashmi3255 (Tue, 01 Sep 2020 12:42:01 GMT):
ubuntu18.4

rashmi3255 (Tue, 01 Sep 2020 12:42:08 GMT):
oh sorry

thearhaam (Tue, 01 Sep 2020 12:48:06 GMT):
@WadeBarnes Ubuntu 20.04.1 LTS

WadeBarnes (Tue, 01 Sep 2020 13:01:34 GMT):
@thearhaam Moving our conversation out of this thread.

Xand (Tue, 01 Sep 2020 15:51:12 GMT):
Has joined the channel.

VictorSyntez (Tue, 01 Sep 2020 18:30:39 GMT):
Has joined the channel.

PrabhatKSINGH (Tue, 01 Sep 2020 18:35:25 GMT):
Has joined the channel.

PrabhatKSINGH (Tue, 01 Sep 2020 18:35:25 GMT):
I am developer of SOLYD platform which is blockchain agnostic. And you can develop any project on blockchain using frontend BPMN diagram drag & drop. We have developed a pilot of Baseline protocol Radish34 in 21 minutes. Here is the link for anyone interested. Now you can create any blockchain app or dapp using BPMN drag & drop elements. It is low code framework for designers, thinkers and business people. https://www.linkedin.com/feed/update/urn:li:activity:6706375114113527809

PrabhatKSINGH (Tue, 01 Sep 2020 18:35:41 GMT):
Can I develop something for you using hyoerledger blockchain? I am looking for some pilots

PrabhatKSINGH (Tue, 01 Sep 2020 18:35:41 GMT):
Can I develop something for you using hyperledger blockchain? I am looking for some pilots

PrabhatKSINGH (Tue, 01 Sep 2020 18:36:07 GMT):
Please share with your network or CTOs/CIOs

VictorSyntez (Tue, 01 Sep 2020 21:21:04 GMT):
Hello, I'm interested in volunteering for Hyperledger Indy projects as a developer-beginner. If there are some tasks that don't require extensive coding knowledge or this knowledge might be gained using free available and known resources then I am willing to help. You can contact me using private message. Cheers, Victor.

pikvik (Wed, 02 Sep 2020 04:56:34 GMT):
Has joined the channel.

SamB (Wed, 02 Sep 2020 08:37:49 GMT):
Hello Team, For Indy-Aries-ursa setup which Ubuntu version is recommended?

Xand (Wed, 02 Sep 2020 10:55:09 GMT):
@SamB I think it's 16.04 support atm: https://github.com/hyperledger/indy-node

c2bo (Wed, 02 Sep 2020 12:28:03 GMT):
Has joined the channel.

larabisch (Wed, 02 Sep 2020 16:24:05 GMT):
Has joined the channel.

swcurran (Wed, 02 Sep 2020 18:14:51 GMT):
@SamB - Indy node specifically is 16.04. Aries, indy-sdk, indy-vdr, etc. (everything on the client/agent side) is not tied to a specific Ubuntu release.

SamB (Thu, 03 Sep 2020 04:33:26 GMT):
@swcurran Thank you for your response and information:blush:

SamB (Thu, 03 Sep 2020 04:36:07 GMT):
@Xand Thank You:blush:

kukgini (Thu, 03 Sep 2020 06:13:05 GMT):
Has joined the channel.

jameshiester (Thu, 03 Sep 2020 13:18:39 GMT):
what script should be used to restart a node that goes down?

SamB (Fri, 04 Sep 2020 13:28:45 GMT):
Hi Team, Can anyone one please help me with code snippet for Adding a new TRUSTEE to the network?

chakshujain (Sun, 06 Sep 2020 14:37:08 GMT):
Has joined the channel.

gnarula (Tue, 08 Sep 2020 10:16:04 GMT):
Is there a convention for storing the routing key in Indy for a Public DID? I can store it with an attrib txn but I wonder if there's a key that's used conventionally. Like 'endpoint' is for ServiceEndpoint

paul.bastian (Tue, 08 Sep 2020 11:39:10 GMT):
Has joined the channel.

pfeairheller (Tue, 08 Sep 2020 12:36:01 GMT):
For now there is no convention used for routing keys in an ATTRIB. We have the same concern and have settled on using 'routingkeys'. However, out of the Indy Interop-athon the decision was made to use an ATTRIB transaction to store full DIDDocs for DIDs. That work will be ongoing over the next couple of months I imagine. More information can be found here: https://wiki.hyperledger.org/display/II/Agenda. In the meantime, if you are thinking of using a custom ATTRIB for the routing keys it might make sense to just go ahead and store the entire DIDDoc in a custom ATTRIB and prepare yourself for the future changes.

gnarula (Tue, 08 Sep 2020 13:00:32 GMT):
Thanks for sharing that! That's a great step! Looks like I missed the recent developments ;) Got something to review over the weekend now :)

AryamaI (Tue, 08 Sep 2020 15:40:58 GMT):
Has joined the channel.

berserkr (Wed, 09 Sep 2020 01:27:16 GMT):
Hi All, how do we register STEWARDs in Indy? For some reason I cannot use anything other than the standard stweard in the samples

berserkr (Wed, 09 Sep 2020 01:27:16 GMT):
Hi All, how do we register STEWARDs in Indy? For some reason I cannot use anything other than the standard stweard in the samples. I am getting the following error when trying to create a nym tx: ```b'{"identifier":"XYJAkVzsrqsDQgu5wN5wJE","reqId":1599615437529934931,"reason":"client request invalid: could not authenticate, verkey for XYJAkVzsrqsDQgu5wN5wJE cannot be found","op":"REQNACK"}',)```

berserkr (Wed, 09 Sep 2020 01:40:38 GMT):
I am following the steps in `getting_started.py` step by step

thwjanssen (Wed, 09 Sep 2020 10:13:25 GMT):
Has joined the channel.

Xand (Thu, 10 Sep 2020 13:04:58 GMT):
I believe you'd have to apply through a form somewhere on this page: https://sovrin.org/issue-credentials

jameshiester (Thu, 10 Sep 2020 16:46:02 GMT):
Is the intent for ATTRIB txns to allow partial updates, or should they be treated as PUTs?

SamB (Thu, 10 Sep 2020 17:05:04 GMT):
Hi All, Quick question: Can we set individual validator nodes to different clouds? Is each node maintain its own Ledger? And How can we achieve the same?

berserkr (Thu, 10 Sep 2020 23:54:31 GMT):
@Xand Thank you for the pointer. Even for the dev network / pool built through docker/indy ci?

miguelmesquita (Fri, 11 Sep 2020 09:48:30 GMT):
Has joined the channel.

miguelmesquita (Fri, 11 Sep 2020 09:48:31 GMT):
I have a question. I am building an aries cloud agent into a docker image. Unfortunately it needs libindy to build. It seems that "apt-key adv --keyserver hkp://keyserver.ubuntu.com:80 --recv-keys CE7709D068DB5E88" is at a fault. The repository seems to be missing. Does anyone knows where i can find a replacement? i get gpg: key CE7709D068DB5E88: public key "Sovrin-Repo-Master (Master key for repo.sovring.org) " imported gpg: Total number processed: 1 gpg: imported: 1 /bin/sh: 1: add-apt-repository: not found

karimStekelenburg1 (Fri, 11 Sep 2020 09:55:44 GMT):
Has joined the channel.

karimStekelenburg1 (Fri, 11 Sep 2020 09:55:44 GMT):
@miguelmesquita do you need to build your own image? The von-image has Indy preinstalled, maybe you could that image as your basis (see: https://hub.docker.com/r/bcgovimages/von-image/)

karimStekelenburg1 (Fri, 11 Sep 2020 09:56:54 GMT):
The aca-py repo also contains multiple Dockerfile's you might want to check out https://github.com/hyperledger/aries-cloudagent-python

miguelmesquita (Fri, 11 Sep 2020 10:03:39 GMT):
thanks karin i will check it out

karimStekelenburg1 (Fri, 11 Sep 2020 10:06:08 GMT):
Oh by the way: there also is a ready made cloudagent container :) https://github.com/bcgov/aries-cloudagent-container

miguelmesquita (Fri, 11 Sep 2020 10:08:19 GMT):
thanks again

pchainho (Fri, 11 Sep 2020 10:58:43 GMT):
Has joined the channel.

WadeBarnes (Fri, 11 Sep 2020 12:50:19 GMT):
There is also no need to build either container. The pre-built images are available in DockerHub: - https://hub.docker.com/r/bcgovimages/von-image/ - https://hub.docker.com/r/bcgovimages/aries-cloudagent Example use of the `aries-cloudagent` to build and agent; https://github.com/bcgov/aries-vcr/blob/master/docker/vcr-agent/Dockerfile

WadeBarnes (Fri, 11 Sep 2020 12:53:08 GMT):
You should take the conversation over to the #aries-cloudagent-python channel if you need additional help.

Xand (Fri, 11 Sep 2020 13:30:32 GMT):
Not sure about a vanilla install in a dev environment (I'd imagine not, but I'm still learning)

miguelmesquita (Fri, 11 Sep 2020 13:37:15 GMT):
thanks, i've already initiated both containers

lynn.bendixsen (Fri, 11 Sep 2020 14:10:57 GMT):
Yes for diversity it is actually best if different nodes are in different clouds. Each node has a copy of the ledger that it uses to maintain consensus, but there is just one "ledger" per network. "How can we achieve the same?" I don't think I understand the questions, but to set up individual nodes in separate clouds to build a brand new network you could use this to get started https://docs.google.com/document/d/1XE2QOiGWuRzWdlxiI9LrG9Am9dCfPXBXnv52wGHorNE

lynn.bendixsen (Fri, 11 Sep 2020 14:10:57 GMT):
Yes for diversity it is actually best if different nodes are in different clouds. Each node has a copy of the ledger that it uses to maintain consensus, but there is just one "ledger" per network. "How can we achieve the same?" I don't think I understand the question, but to set up individual nodes in separate clouds to build a brand new network you could use this to get started https://docs.google.com/document/d/1XE2QOiGWuRzWdlxiI9LrG9Am9dCfPXBXnv52wGHorNE

lynn.bendixsen (Fri, 11 Sep 2020 14:12:39 GMT):
There are also instructions for installing an Indy node in each of AWS Azure and GCP if thats what you were asking.

SamB (Fri, 11 Sep 2020 14:27:38 GMT):
Thank you so much for your response @lynn.bendixsen I question was how can we achieve to syncing of the ledger if we set validator nodes to different cloud instances?

lynn.bendixsen (Fri, 11 Sep 2020 14:30:20 GMT):
Each node has a public IP address and port that it communicates on for syncing.

SamB (Fri, 11 Sep 2020 14:52:30 GMT):
Oh okay.. Thank you

danielhardman (Fri, 11 Sep 2020 18:10:00 GMT):
I am concerned about the ramifications for DIDComm of this DID core issue: https://github.com/w3c/did-core/issues/386 Joe A suggests that our (DIDComm designers) interpretation of the DID spec with regard to key rotation has no basis in the spec language. He may or may not be right, but I’m pretty confident that neither does the other interpretation he points out. I consider that other interpretation, where you can’t look up any DID state except the current one, to be totally unacceptable, to the point where I think we would have to abandon support for DIDs if key rotation is not supported. This is a BIG deal. Please be proactive about reading this ticket, commenting, and mustering support for our interpretation among like-minded community members.

trustcerts (Mon, 14 Sep 2020 10:52:19 GMT):
Has joined the channel.

trustcerts (Mon, 14 Sep 2020 10:52:20 GMT):
Hello everyone, i try to understand the communication between the client and the node and found the different types of requests (https://hyperledger-indy.readthedocs.io/projects/node/en/latest/requests.html), but what underlying protocol (http, grpc, etc.) and endpoint is used?

swcurran (Mon, 14 Sep 2020 13:44:34 GMT):
It uses ZMQ, AFAIK. Devs can check me on that, but pretty sure :-)

swcurran (Mon, 14 Sep 2020 13:44:39 GMT):
Hey folks --- on the Indy Contributors call this coming Tuesday (September 15, 8AM Pacific, 5PM CET), we're going to be follow up on the first two actions needed out of the Indy Interop-athon -- CI/CD migration to GitHub Actions and the did:indy DID Method spec. We'll also review the outputs from the Indy Interop-athon as we turn talk into action. Please join us on the Contributors call this Tuesday -- https://wiki.hyperledger.org/display/indy/2020-09-15+Indy+Contributors+Call -- Zoom: https://zoom.us/j/244779296

Bizyroth (Mon, 14 Sep 2020 13:46:17 GMT):
Has joined the channel.

Bizyroth (Mon, 14 Sep 2020 13:46:17 GMT):
Hello I'm trying to find which specification documents is used for the implementation of the Proof verification mechanism. I found documentation in different places (https://github.com/hyperledger/indy-hipe/tree/master/text/0109-anoncreds-protocol https://github.com/hyperledger/ursa-docs/tree/master/specs ). What is the reference paper used for the implementation?

hanumankumar (Mon, 14 Sep 2020 14:46:04 GMT):
Hi All, Do we have any fab token example using Java SDK?if we have please can you share with me. Cheers, Hanu

jorgeRodriguez (Tue, 15 Sep 2020 02:10:09 GMT):
Has joined the channel.

smiya 1 (Tue, 15 Sep 2020 07:32:03 GMT):
Has joined the channel.

hanumankumar (Tue, 15 Sep 2020 12:27:51 GMT):
Hi All, Can we design a utility token using hyperledger fabric without fab token? If we can do please can you share me the existing ways to do. Cheers, Hanu

Lucampelli (Tue, 15 Sep 2020 17:54:28 GMT):
Hello! I have a doubt about schema and credential definitions in the walkthroughs. I'm creating my own example to learn about the different flows. Is schema definition and cred_def issuance only allowed by the stewart trust anchor? I've sent requests for "TRUST ANCHOR" verinyms by the stewart, but these did's return unauthorized when sending schema definitions... am I doing something wrong?

swcurran (Tue, 15 Sep 2020 18:11:13 GMT):
No. Indy has a flexible authorization system for writes to the ledger. It's configurable, and in the Sovrin system (for example), there are 6 explicit roles: - readers -- anyone - author -- anyone with a DID on the ledger - network monitor -- has a DID and can query nodes to get detailed status about them - endorser -- has a DID and is permitted to sign transactions from authors and endorsers (including themselves) such that they can be written to the ledger. The endorser pays for all transactions that they sign and are written. - steward -- can operate a node on the ledger and contribute to validation - trustee -- can do what an endorser can do and can sign transactions to write endorser DIDs. So the TRUST ANCHOR is an Endorser on Sovrin, and so can sign and write it's on transactions, but a "normal" DID, cannot. In the sandbox environment, there is a TRUSTEE that you can create the DID for -- the seed for the DID is available for all.

Lucampelli (Tue, 15 Sep 2020 18:20:44 GMT):
Also, the indy_node pool from /docker/pool_start.sh is giving an error called CommonInvalidState

swcurran (Tue, 15 Sep 2020 18:21:02 GMT):
FYI - this Thursday (6AM Pacific / 3PM CET) we'll be holding an Indy Node / Indy Plenum release planning meeting for the next release of the ledger part of Indy. This is a great opportunity for anyone wanting to get involved in Indy development to pick up some tasks to contribute with support from existing contributors. We'll also be using this release to migrate the indy-node release process to GitHub Actions. Please join us on the normal Indy Contributors call Zoom -- https://zoom.us/j/244779296 See you there!

swcurran (Tue, 15 Sep 2020 18:21:58 GMT):
Regards your first question - No - nothing wrong. Indy has a flexible authorization system for writes to the ledger. It's configurable, and in the Sovrin system (for example), there are 6 explicit roles: - readers -- anyone - author -- anyone with a DID on the ledger - network monitor -- has a DID and can query nodes to get detailed status about them - endorser -- has a DID and is permitted to sign transactions from authors and endorsers (including themselves) such that they can be written to the ledger. The endorser pays for all transactions that they sign and are written. - steward -- can operate a node on the ledger and contribute to validation - trustee -- can do what an endorser can do and can sign transactions to write endorser DIDs. So the TRUST ANCHOR is an Endorser on Sovrin, and so can sign and write it's on transactions, but a "normal" DID, cannot. In the sandbox environment, there is a TRUSTEE that you can create the DID for -- the seed for the DID is available for all.

Lucampelli (Tue, 15 Sep 2020 18:24:12 GMT):
So even if stewart sends verinyms for an user for example, if the role is TRUST_ANCHOR the user can't create schemas, nor sign it's transactions, but if it's a TRUSTEE, it can?

Lucampelli (Tue, 15 Sep 2020 18:36:42 GMT):
Ok... so for some context, my code creates the stewart's wallet, then creates a regulatory agent, RA. Creates it's wallet, and stewart (maybe the gov) sends a verynim to the ledger appointing RA as Trust_anchor. RA then creates a schema (that's working) and now wants to create a cred_def based on it's schema, but i'm getting indy.error.CommonInvalidStructure.

Lucampelli (Tue, 15 Sep 2020 18:37:39 GMT):
also thanks for the response

the_identity_guy (Wed, 16 Sep 2020 03:34:12 GMT):
Hello Indy team! Does Indy have a full cylce (issue, manage, verify and revoke) support of zero knowledge proof claims using verifiable credentials? Is there any article/link that explains what is currently supported, or is in the pipeline?

the_identity_guy (Wed, 16 Sep 2020 03:34:12 GMT):
Hello Indy team! Does Indy have a full cycle (issue, manage, verify and revoke) support of zero knowledge proof claims using verifiable credentials? Is there any article/link that explains what is currently supported, or is in the pipeline?

avi23 (Wed, 16 Sep 2020 12:04:23 GMT):
Hello Folks, @swcurran can anyone please put some sight on how the master secret works in Indy to obtain the user anonymity? how can we hide user details like email , name etc using master secret key while presenting it for verification to verifier?

Lucampelli (Wed, 16 Sep 2020 14:05:36 GMT):
Ok.. solved one issue. The cred_def problem happened because i used the id instead of the json... didn't notice it. Now this is the problem: {'identifier': '5mhbTjB662MT599k6dpB4P', 'op': 'REQNACK', 'reason': "client request invalid: InvalidClientRequest('validation error " '[ClientClaimDefSubmitOperation]: cannot be smaller than 1 ' "(ref=0)',)", 'reqId': 1600265018440896430} i'm using only the "user" credentials, like did and wallet, not the stewart, since i made the "user" a trust_anchor... any leads?

swcurran (Wed, 16 Sep 2020 14:08:46 GMT):
It's crypto magic, but here is a quick summary. At issue time, the holder uses their link secret to create a piece of "blinded" data that they give to the issuer, who in turn puts it in the credential. At proving time, the holder uses their link secret to "unblind" that data. That proves to the verifier that the holder has the link secret that was used to blind the data that went into the credential. Further, that unblinded data is given to the verifier such that it is different in every proof, and does not create a unique identifier for the credential (Privacy preserving). That's at a high level -- you would have to go to the crypto level to understand more. Wikipedia of Commitment Schemes (https://en.wikipedia.org/wiki/Commitment_scheme) will give you more on the data that the holder gives to the issuer to be put in the credential and later proven.

swcurran (Wed, 16 Sep 2020 14:23:45 GMT):
On the TRUST_ANCHOR question -- I believe that is the generic Indy term for Endorser that is used in my definition above.

swcurran (Wed, 16 Sep 2020 14:24:47 GMT):
On the other issues -- I don't work these days at that level, so am not sure what issues you are hitting.

Lucampelli (Wed, 16 Sep 2020 14:25:33 GMT):
well thanks! you already helped a lot

smiya 1 (Thu, 17 Sep 2020 01:25:42 GMT):
``` Hello! ```

smiya 1 (Thu, 17 Sep 2020 01:28:28 GMT):
``` hello! I studied at the following site. - https://indy.readthedocs.io/en/latest/ - https://github.com/hyperledger/indy-sdk/blob/master/docs/getting-started/indy-walkthrough.md - We tried running the nodejs sample. But I couldn't understand. If there is a document that explains in detail, please introduce it. Especially the following explanation: - How to set Validator Node and Observer Node. - The relationship between Credential Schema and Credential Definition. - Explanation of indy-walkthrough . - Implementation when wallet of Alice, Faber, Acme, Thriftis in different Node in indy-walkthrough scenario. ```

smiya 1 (Thu, 17 Sep 2020 01:28:28 GMT):
hello! I studied at the following site. - https://indy.readthedocs.io/en/latest/ - https://github.com/hyperledger/indy-sdk/blob/master/docs/getting-started/indy-walkthrough.md - We tried running the nodejs sample. But I couldn't understand. If there is a document that explains in detail, please introduce it. Especially the following explanation: - How to set Validator Node and Observer Node. - The relationship between Credential Schema and Credential Definition. - Explanation of indy-walkthrough . - Implementation when wallet of Alice, Faber, Acme, Thriftis in different Node in indy-walkthrough scenario.

smiya 1 (Thu, 17 Sep 2020 01:28:28 GMT):
hello! I studied at the following site. - https://indy.readthedocs.io/en/latest/ - https://github.com/hyperledger/indy-sdk/blob/master/docs/getting-started/indy-walkthrough.md - I tried running the nodejs sample. But I couldn't understand. If there is a document that explains in detail, please introduce it. Especially the following explanation: - How to set Validator Node and Observer Node. - The relationship between Credential Schema and Credential Definition. - Explanation of indy-walkthrough . - Implementation when wallet of Alice, Faber, Acme, Thriftis in different Node in indy-walkthrough scenario.

CavitOzbay (Thu, 17 Sep 2020 08:56:53 GMT):
Hi everyone, I have a question about Indy's api calls about proof creation and verification. In order to prove that multiple credentials have the same attribute value without disclosing the attribute, "add_common_attribute" funtion of Ursa may be used. However, as far as I know, there is no way to add such functionality or trigger to call function "add_common_attribute" from Indy's api calls. Am I wrong?

etaleo (Thu, 17 Sep 2020 11:19:08 GMT):
Has joined the channel.

etaleo (Thu, 17 Sep 2020 11:19:09 GMT):
Hey there! I am trying to get the Python3 demo to work: https://github.com/hyperledger/indy-sdk/tree/master/vcx/wrappers/python3/demo After install the necessary requirements and starting a cloud agent via "cargo run config/sample-config.json", I get the following error when running faber.py: * https://privatebin.net/?5149379525afb412#7NZkjBow4frdUZvkGeZNcYtsLwdcmcpgJXSfiaeHPN5C (Pool Timeout) Do I maybe need to alter the dummy agent's sample config somehow?

etaleo (Thu, 17 Sep 2020 11:19:59 GMT):
Hey there! I am trying to get the Python3 demo to work: https://github.com/hyperledger/indy-sdk/tree/master/vcx/wrappers/python3/demo After install the necessary requirements and starting a cloud agent via "cargo run config/sample-config.json", I get the following error when running faber.py: * https://privatebin.net/?5149379525afb412#7NZkjBow4frdUZvkGeZNcYtsLwdcmcpgJXSfiaeHPN5C (Pool Timeout) Do I maybe need to alter the dummy agent's sample config somehow?

etaleo (Thu, 17 Sep 2020 11:19:59 GMT):
Hey there! I am trying to get the Python3 demo to work: https://github.com/hyperledger/indy-sdk/tree/master/vcx/wrappers/python3/demo After install the necessary requirements and starting a cloud agent via "cargo run config/sample-config.json", I get the following error when running faber.py: * https://zerobin.net/?20be0102115020cb#16Uw/MJFpjLMKiTEaZaZ8aLQMCaFR//96Ns21vpXT+k= (Pool Timeout) Do I maybe need to alter the dummy agent's sample config somehow?

etaleo (Thu, 17 Sep 2020 11:19:59 GMT):
Hey there! I am trying to get the Python3 demo to work: https://github.com/hyperledger/indy-sdk/tree/master/vcx/wrappers/python3/demo After install the necessary requirements and starting a cloud agent via "cargo run config/sample-config.json", I get the following error when running faber.py: * https://zerobin.net/?20be0102115020cb#16Uw/MJFpjLMKiTEaZaZ8aLQMCaFR//96Ns21vpXT+k= (Pool Timeout) Do I maybe need to alter the dummy agent's sample config somehow? editing the main message for visibility: Thanks to @sheru faber.py is running now and waiting for alice.py. Now, alice.py is throwing this error: https://zerobin.net/?a935e7add43fcb41#qkGowcBtYCah+NedXMOVPBwwmGw9xeqfno4m5xAt1Hs= (ErrorCode.InvalidConfiguration: Cannot parse config: unknown variant `4.0`, expected one of `1.0`, `2.0`, `3.0`)

etaleo (Thu, 17 Sep 2020 11:19:59 GMT):
~Hey there! I am trying to get the Python3 demo to work: https://github.com/hyperledger/indy-sdk/tree/master/vcx/wrappers/python3/demo After install the necessary requirements and starting a cloud agent via "cargo run config/sample-config.json", I get the following error when running faber.py: * https://zerobin.net/?20be0102115020cb#16Uw/MJFpjLMKiTEaZaZ8aLQMCaFR//96Ns21vpXT+k= (Pool Timeout) Do I maybe need to alter the dummy agent's sample config somehow?~ editing the main message for visibility: Thanks to @sheru, faber.py is running now and waiting for alice.py. Now, alice.py is throwing this error: https://zerobin.net/?a935e7add43fcb41#qkGowcBtYCah+NedXMOVPBwwmGw9xeqfno4m5xAt1Hs= (ErrorCode.InvalidConfiguration: Cannot parse config: unknown variant `4.0`, expected one of `1.0`, `2.0`, `3.0`)

etaleo (Thu, 17 Sep 2020 11:19:59 GMT):
Hey there! I am trying to get the Python3 demo to work: https://github.com/hyperledger/indy-sdk/tree/master/vcx/wrappers/python3/demo After install the necessary requirements and starting a cloud agent via "cargo run config/sample-config.json", I get the following error when running faber.py: * https://zerobin.net/?20be0102115020cb#16Uw/MJFpjLMKiTEaZaZ8aLQMCaFR//96Ns21vpXT+k= (Pool Timeout) Do I maybe need to alter the dummy agent's sample config somehow? editing the main message for visibility: Thanks to @sheru, faber.py is running now and waiting for alice.py. Now, alice.py is throwing this error: https://zerobin.net/?a935e7add43fcb41#qkGowcBtYCah+NedXMOVPBwwmGw9xeqfno4m5xAt1Hs= (ErrorCode.InvalidConfiguration: Cannot parse config: unknown variant `4.0`, expected one of `1.0`, `2.0`, `3.0`)

etaleo (Thu, 17 Sep 2020 11:19:59 GMT):
Hey there! I am trying to get the Python3 demo to work: https://github.com/hyperledger/indy-sdk/tree/master/vcx/wrappers/python3/demo After install the necessary requirements and starting a cloud agent via "cargo run config/sample-config.json", I get the following error when running faber.py: * https://zerobin.net/?20be0102115020cb#16Uw/MJFpjLMKiTEaZaZ8aLQMCaFR//96Ns21vpXT+k= (Pool Timeout) Do I maybe need to alter the dummy agent's sample config somehow? edit: Thanks to @sheru, faber.py is running now and waiting for alice.py. Now, alice.py is throwing this error: https://zerobin.net/?a935e7add43fcb41#qkGowcBtYCah+NedXMOVPBwwmGw9xeqfno4m5xAt1Hs= (ErrorCode.InvalidConfiguration: Cannot parse config: unknown variant `4.0`, expected one of `1.0`, `2.0`, `3.0`)

etaleo (Thu, 17 Sep 2020 11:19:59 GMT):
Hey there! I am trying to get the Python3 demo to work: https://github.com/hyperledger/indy-sdk/tree/master/vcx/wrappers/python3/demo After install the necessary requirements and starting a cloud agent via "cargo run config/sample-config.json", I get the following error when running faber.py: * https://zerobin.net/?20be0102115020cb#16Uw/MJFpjLMKiTEaZaZ8aLQMCaFR//96Ns21vpXT+k= (Pool Timeout) Do I maybe need to alter the dummy agent's sample config somehow?

Lucampelli (Thu, 17 Sep 2020 22:12:59 GMT):
Enabling Debug Mode, show me that it's a ref attribute set during the build_cred_def_request: request_json: "{\"reqId\":1600379007864815695,\"identifier\":\"PCdtL6qyjs43cVhBDPnwd4\",\"operation\":{\"ref\":0,\"data\":{\"primary\":{\"n

Lucampelli (Thu, 17 Sep 2020 22:27:05 GMT):
is there any restrictions about a single user creating the schema and the cred_def?

swcurran (Thu, 17 Sep 2020 22:50:51 GMT):
No restrictions -- that's been a common case. We hope people will use shared schema over time -- easier for verifiers, but not required.

sheru (Fri, 18 Sep 2020 07:36:19 GMT):
Please check the genesis file/url for which ledger network you are using is up and running.

SuperSeiyan (Fri, 18 Sep 2020 08:36:38 GMT):
Can you try to remove `~/.indy_client` then try again?

avi23 (Fri, 18 Sep 2020 09:20:00 GMT):
thanks @swcurran for detail explanation, though i have 2 more direct questions around alice faber demo

avi23 (Fri, 18 Sep 2020 09:21:41 GMT):
1. Does Master KEY used to get generated at the time of starting the agent itself in the backend process by indy?

avi23 (Fri, 18 Sep 2020 09:24:12 GMT):
2. Does the master key & master secret key are two different things here ? and does every actor knows the holder's master secret key at the time of verification?

etaleo (Fri, 18 Sep 2020 14:15:55 GMT):
@sheru Thank you! Now faber.py is running and waiting for alice.py. alice.py is throwing this error: https://zerobin.net/?a935e7add43fcb41#qkGowcBtYCah+NedXMOVPBwwmGw9xeqfno4m5xAt1Hs= (ErrorCode.InvalidConfiguration: Cannot parse config: unknown variant `4.0`, expected one of `1.0`, `2.0`, `3.0`)

etaleo (Fri, 18 Sep 2020 14:15:55 GMT):
@sheru Thank you! Now faber.py is running and waiting for alice.py. Now, alice.py is throwing this error: https://zerobin.net/?a935e7add43fcb41#qkGowcBtYCah+NedXMOVPBwwmGw9xeqfno4m5xAt1Hs= (ErrorCode.InvalidConfiguration: Cannot parse config: unknown variant `4.0`, expected one of `1.0`, `2.0`, `3.0`)

etaleo (Fri, 18 Sep 2020 14:15:55 GMT):
@sheru Thank you!

swcurran (Fri, 18 Sep 2020 15:00:46 GMT):
The terminology is not precise. Master Secret is what is used in credential exchange. Master Key I think in this context is the wallet key -- the key for encrypting/decrypting items in the wallet/storage. Both are known only to the Holder/Prover. The Master Secret (aka Link Secret) is used to prove the Prover created the value that was put into the issued credential.

AryamaI (Fri, 18 Sep 2020 15:08:06 GMT):
@swcurran please confirm so at the start of agent only master key is als generated? no way for a user to see rehat his wallet maste key is?

AryamaI (Fri, 18 Sep 2020 15:08:06 GMT):
@swcurran please confirm so at the start of agent only master key is also generated? no way for a user to see what his wallet master key is?

swcurran (Fri, 18 Sep 2020 17:58:41 GMT):
I think that is the case, but I'll ping @andrew.whitehead to confirm.

andrew.whitehead (Fri, 18 Sep 2020 18:06:55 GMT):
The master key can refer to a couple things. You can use a wallet password, in which case it derives a wallet encryption key using argon2 each time you open it. You can also configure the wallet with a raw 32-byte key and bypass the key derivation.

andrew.whitehead (Fri, 18 Sep 2020 18:09:20 GMT):
Actually I don't think aca-py exposes a way for you to use a raw key, that's only at the lower level. But the encryption key is derived directly from the wallet password

AryamaI (Sat, 19 Sep 2020 03:15:17 GMT):
Thank you for the explanations @swcurran and @andrew.whitehead

avi23 (Sat, 19 Sep 2020 07:30:26 GMT):
Thanks @andrew.whitehead so does is mean that master secret key is also getting generated internally in the process and as developer we don't have to do anything extra here?

daruru (Sat, 19 Sep 2020 10:10:23 GMT):
Has joined the channel.

daruru (Sat, 19 Sep 2020 10:10:23 GMT):
Hi, i´m following this repo "https://github.com/hyperledger/aries-mobileagent-xamarin" for do a Xamarin proyect, but i have problem when i run the app and i try to create a wallet, it said to me "failed to create wallet : the subsystem network fall" , before it i try to run the mediator dotnet run , but the shell said "Unable to load DLL 'indy' or one of its dependencies: Can´t find the specifil module (0x8007007E)" . Can anybody help me ? thanks

andrew.whitehead (Sat, 19 Sep 2020 16:00:24 GMT):
Yes, it’s automatic

rileyphughes (Sat, 19 Sep 2020 20:25:19 GMT):
you can ask this question in #aries-framework-dotnet , most of the ama-x maintainers are there

daruru (Mon, 21 Sep 2020 09:33:05 GMT):
thanks

PatrikStas (Tue, 22 Sep 2020 09:30:20 GMT):
Heads up @sheru and @SuperSeiyan We in Absa have forked LibVCX and took the development off hyperledger https://github.com/AbsaOSS/libvcx So you might want to consider using this, as far as I know Evernym is not intending further development of libvcx development in hyperledger. Absa has been main contributor to LibVCX for a while before we decided to fork.

PatrikStas (Tue, 22 Sep 2020 09:30:20 GMT):
Heads up @sheru and @SuperSeiyan We in Absa have forked LibVCX and took the development off hyperledger https://github.com/AbsaOSS/libvcx So you might want to consider using this, as far as I know Evernym is not intending further development of libvcx development in hyperledger. Absa has been main contributor to LibVCX (along Evernym) for a while before we decided to fork.

PatrikStas (Tue, 22 Sep 2020 09:36:35 GMT):
If you'd like to follow the fork, please let me know and I can help you to onboard. If you are using python vcx wrapper you might have to do some small updates

AryamaI (Tue, 22 Sep 2020 11:31:01 GMT):
Hello everyone, can somebody tell me on how we can setup aries cloudagent python docker agents with the indy nodes pool instead of von-network ?

AryamaI (Tue, 22 Sep 2020 11:31:01 GMT):
Hello everyone, can somebody tell me on how we can setup aries cloudagent python docker agents with the indy nodes pool instead of von-network ? Followed this doc for indy -ppol setup https://github.com/hyperledger/indy-sdk#how-to-start-local-nodes-pool-with-docker. But facing issue in understanding how to connect to aries docker agent containers

AryamaI (Tue, 22 Sep 2020 11:31:01 GMT):
Hello everyone, can somebody tell me how we can setup aries cloud agent python docker agents with the indy nodes pool instead of von-network? Followed this doc for indy -pool set up https://github.com/hyperledger/indy-sdk#how-to-start-local-nodes-pool-with-docker. But facing issue in understanding how to connect to aries docker agent containers

sheru (Tue, 22 Sep 2020 12:35:12 GMT):
once the nodes running in the local machine most probably you will get see the services on port 9000. you can found the genesis info in localhost:9000/genesis , use this path and run the aca-py agent and pass the genesis url in the envs of the docker.

AryamaI (Tue, 22 Sep 2020 12:45:04 GMT):
Hi @sheru , the ports exposed are from 9701-9708. Even so. I cant see the genesis info in any of these ports when trying to open from browser

swcurran (Tue, 22 Sep 2020 18:52:36 GMT):
The port 9000 web service is a feature of von-network, so you don't get that with plain indy. You need to get a genesis file for the configuration you have -- with the correct IP addresses and ports. I think (but am not certain) that is available in folder where the nodes are running. I would think the Indy docs for starting up a raw network would cover from where to get the genesis file. Once you have that, you need to copy it somewhere that the ACA-Py agent can get to it and use the mechanism to use GENESIS_FILE with the path to the file.

SuperSeiyan (Wed, 23 Sep 2020 05:26:49 GMT):
@PatrikStas sure, it will check that fork

SuperSeiyan (Wed, 23 Sep 2020 05:26:49 GMT):
@PatrikStas Sure, it will check that fork. If there is any issues or discussion where should be the main channel? ' It would be great if we will have the main channel for discussion for the Libvcx only.

SuperSeiyan (Wed, 23 Sep 2020 05:26:49 GMT):
@PatrikStas Sure, it will check that fork. If there is any issues or discussion where should be the main channel? ' It would be great if we will have a main channel for discussion for the Libvcx only.

AryamaI (Wed, 23 Sep 2020 06:33:25 GMT):
Hello @swcurran ,I started the indy pool nodes with this sommands`docker build -f ci/indy-pool.dockerfile -t indy_pool . docker run -itd -p 9701-9708:9701-9708 indy_pool`

AryamaI (Wed, 23 Sep 2020 06:33:58 GMT):
Hello @swcurran ,I started the indy pool nodes with this sommands`docker build -f ci/indy-pool.dockerfile -t indy_pool . docker run -itd -p 9701-9708:9701-9708 indy_pool` `` dwsd``

AryamaI (Wed, 23 Sep 2020 06:35:46 GMT):
Hello @swcurran ,I started the indy pool nodes with this commands *1. docker build -f ci/indy-pool.dockerfile -t indy_pool . 2.docker run -itd -p 9701-9708:9701-9708 indy_pool* But I dont see this dockerfile mentioned here using the gensis file anywhere for startign of nodes. And when I try to start agents with the genesis file provided in the sdk repo, its not working

AryamaI (Wed, 23 Sep 2020 06:35:46 GMT):
Hello @swcurran ,I started the indy pool nodes with this commands 1. docker build -f ci/indy-pool.dockerfile -t indy_pool . 2.docker run -itd -p 9701-9708:9701-9708 indy_pool But I dont see this dockerfile mentioned here using the genesis file anywhere for starting of nodes. And when I try to start agents with the genesis file provided in the indy-sdk repo, its not working and saying something like error retrieving genesis information from ledger

AryamaI (Wed, 23 Sep 2020 06:35:46 GMT):
Hello @swcurran ,I started the indy pool nodes with this commands 1. docker build -f ci/indy-pool.dockerfile -t indy_pool . 2.docker run -itd -p 9701-9708:9701-9708 indy_pool But I dont see this dockerfile mentioned here using the genesis file anywhere for starting of nodes. And when I try to start agents with the genesis file provided in the indy-sdk repo, its not working and saying something like error retrieving genesis information from ledger

AryamaI (Wed, 23 Sep 2020 06:35:46 GMT):
Hello @swcurran ,I started the indy pool nodes with this commands 1. docker build -f ci/indy-pool.dockerfile -t indy_pool . 2.docker run -itd -p 9701-9708:9701-9708 indy_pool But I dont see this dockerfile mentioned here using the genesis file anywhere for starting of nodes. And when I try to start agents with the genesis file provided in the indy-sdk repo, its not working and saying something like error retrieving genesis information from ledger

AryamaI (Wed, 23 Sep 2020 06:35:46 GMT):
Hello @swcurran ,I started the indy pool nodes with these commands: 1. docker build -f ci/indy-pool.dockerfile -t indy_pool . 2.docker run -itd -p 9701-9708:9701-9708 indy_pool But I dont see this dockerfile mentioned here using the genesis file anywhere for starting of nodes. And when I try to start agents with the genesis file provided in the indy-sdk repo, its not working and saying something like error retrieving genesis information from ledger

AryamaI (Wed, 23 Sep 2020 10:05:26 GMT):
Hello, Can someone confirm me if we can write anything to the indy ledger apart from Did, DidDoc, schema, credential definitions and revocation registries? I would like to know if we can write any other info and query its value from indy ledger

miguelmesquita (Wed, 23 Sep 2020 11:56:52 GMT):
hello everyone, Is there a way to send a complex type on a credential other then a string - has an attribute - following a schema?

Lucampelli (Wed, 23 Sep 2020 12:30:14 GMT):
Found the problem... the schema must be uploaded to the ledger and then retrieved from the ledger, in ored to be formatted correctly... so the walkthrough example will not work...

Lucampelli (Wed, 23 Sep 2020 12:30:14 GMT):
Found the problem... the schema must be uploaded to the ledger and then retrieved from the ledger, in order to be formatted correctly... so the walkthrough example will not work...

PatrikStas (Wed, 23 Sep 2020 12:39:25 GMT):
If you go with Absa's fork, you can create github issue on our repo. Alternatively you can text me directly

PatrikStas (Wed, 23 Sep 2020 12:40:35 GMT):
I've created this proposal https://lists.hyperledger.org/g/indy/topic/proposal_to_move_absa_libvcx/77033698?p=,,,20,0,0,0::recentpostdate%2Fsticky,,,20,2,0,77033698 to discuss possibility of moving our fork back under Hyperledger in favor of the "original" LibVCX currently under IndySDK repo.

PatrikStas (Wed, 23 Sep 2020 12:41:24 GMT):
I think if that happens, we could ask hyperledger to create here channel for libvcx as well

vasantkk (Wed, 23 Sep 2020 14:23:49 GMT):
Has joined the channel.

vasantkk (Wed, 23 Sep 2020 14:23:50 GMT):
hey I wanted to know what is the difference between metadium and sovrin foundation? Is metadium open source?

vasantkk (Wed, 23 Sep 2020 14:24:05 GMT):
is it possible to integrate hyperledger indy with ethereum ids?

vasantkk (Wed, 23 Sep 2020 14:24:19 GMT):
can sovrin integrate with ethereum ids?

blidd (Wed, 23 Sep 2020 14:35:15 GMT):
Has joined the channel.

swcurran (Wed, 23 Sep 2020 16:05:13 GMT):
You don't need the genesis files to start the nodes, but you need it to start the instances of ACA-Py that you are running.

swcurran (Wed, 23 Sep 2020 16:06:08 GMT):
The error you are getting is that the IP address is probably wrong. You need to determine the IP address of the nodes (their public IP) and edit the genesis file to put those in.

swcurran (Wed, 23 Sep 2020 16:07:02 GMT):
This is why von-network exists -- it does all that for you. Plus the web server network browser.

mccown (Wed, 23 Sep 2020 19:12:41 GMT):
Just a reminder about the Identity Implementer's WG Call tomorrow morning (Sep 24th @ 9am MT). This is a chance to get updates on a variety of WG calls that you may have missed this week. Our guest presenters will be Kevin Griffin and Philip Feairheller from Scoir and they will be giving a presentation of Canis "The Extensible, RESTful Credentialing Hub". Here are links to the Wiki and the Zoom meeting. Wiki: https://wiki.hyperledger.org/display/IWG/2020-09-24+Identity+Implementers+WG+Call Zoom: https://zoom.us/j/244779296

vasantkk (Thu, 24 Sep 2020 06:47:10 GMT):
I wanted to ask questions regarding indy project

vasantkk (Thu, 24 Sep 2020 07:10:43 GMT):
https://lists.hyperledger.org/g/indy/message/300 what is the status of this project?

GianlucaPinto (Thu, 24 Sep 2020 13:40:07 GMT):
Has joined the channel.

GianlucaPinto (Thu, 24 Sep 2020 13:40:08 GMT):
hi all, how can i do for get the timestamp of a credentials ? thanks

Lucampelli (Thu, 24 Sep 2020 14:02:55 GMT):
Hello, how can I create my own pool in a server for my University's lab, and have an instance o Indy-SDK connect to it?

Lucampelli (Thu, 24 Sep 2020 14:02:55 GMT):
Hello, how can I create my own pool in a server for my University's lab, and have an instance o Indy-SDK connect to it, so that we can develop and test on it?

AryamaI (Fri, 25 Sep 2020 11:15:31 GMT):
Thank you @swcurran . Can you confirm going forward to build indy application at the production level. Is it recommended to use mainnet or proceed with our own indy node vanilla setup

ldej (Fri, 25 Sep 2020 12:10:31 GMT):
Has joined the channel.

esplinr (Fri, 25 Sep 2020 13:32:32 GMT):
Have you looked at the Docker image? https://github.com/hyperledger/indy-node/tree/master/build-scripts/ubuntu-1604

esplinr (Fri, 25 Sep 2020 13:33:13 GMT):
I haven't heard an update on that in a long time.

danielhardman (Fri, 25 Sep 2020 14:51:05 GMT):
On the mailing list for Indy, there was a proposal to move Absa LibVCX under the Hyperledger umbrella, and possibly to remove the libvcx folder that currently sits under Indy SDK. How can we make a formal decision on the proposal? Should we put it to a vote on a community cal, or vote on it here, or something else?

swcurran (Fri, 25 Sep 2020 18:32:17 GMT):
I personally would not use an internal indy node network for a production app. A key component of decentralized identity is that the keys are publicly available and not controlled by one entity. Sovrin governs the mainnet, but the nodes are operated by many different organizations.

AryamaI (Fri, 25 Sep 2020 18:39:37 GMT):
Ok thank you @swcurran for the input

swcurran (Fri, 25 Sep 2020 23:18:49 GMT):
I should have said in the above: "... and *the publication of the keys is* not controlled by one entity" The original is not phrased well.

etaleo (Sat, 26 Sep 2020 12:44:31 GMT):
:wave: I am trying to get indy-node to run locally following the "Dev Setup" guide: https://github.com/hyperledger/indy-node/blob/master/docs/source/setup-dev.md When finally running "pytest ." in the indy-node directory, it fails with "ImportError: No module named '_lzma'". This seems (after a quick google search) to indicate a problem with the python environment. 1) The guide is aimed at Ubuntu 16.04, while I am running on Linux Mint 19.3. May this be the cause of the error, and if so, how can I get it to work? In the bigger picture, I'm trying to get a better understanding of the communication between the Indy pool and clients. I've found out so far that it uses ZeroMQ, and as a next step I wanted to get indy-node to run locally and add debug output to print out all ZeroMQ messages that are being sent and received. 2) Are there some useful resources in the documentation that cover the topic of which messages are being sent/received via ZeroMQ?

etaleo (Sat, 26 Sep 2020 12:44:31 GMT):
:wave: I am trying to get indy-node to run locally following the "Dev Setup" guide: https://github.com/hyperledger/indy-node/blob/master/docs/source/setup-dev.md When finally running "pytest ." in the indy-node directory, it fails with "ImportError: No module named '_lzma'". This seems (after a quick google search) to indicate a problem with the python environment. 1) The guide is aimed at Ubuntu 16.04, while I am running on Linux Mint 19.3. May this be the cause of the error, and if so, (how) can I get it to work on my machine? In the bigger picture, I'm trying to get a better understanding of the communication between the Indy pool and clients. I've found out so far that it uses ZeroMQ, and as a next step I wanted to get indy-node to run locally and add debug output to print out all ZeroMQ messages that are being sent and received. 2) Are there some useful resources in the documentation that cover the topic of which messages are being sent/received via ZeroMQ?

etaleo (Sat, 26 Sep 2020 12:44:31 GMT):
:wave: I am trying to get indy-node to run locally following the "Dev Setup" guide: https://github.com/hyperledger/indy-node/blob/master/docs/source/setup-dev.md When finally running "pytest ." in the indy-node directory, it fails with "ImportError: No module named '_lzma'". This seems (after a quick google search) to indicate a problem with the python environment. 1) The guide is aimed at Ubuntu 16.04, while I am running Linux Mint 19.3. May this be the cause of the error, and if so, (how) can I get it to work on my machine? In the bigger picture, I'm trying to get a better understanding of the communication between the Indy pool and clients. I've found out so far that it uses ZeroMQ, and as a next step I wanted to get indy-node to run locally and add debug output to print out all ZeroMQ messages that are being sent and received. 2) Are there some useful resources in the documentation that cover the topic of which messages are being sent/received via ZeroMQ?

etaleo (Sat, 26 Sep 2020 12:44:31 GMT):
:wave: I am trying to get indy-node to run locally following the "Dev Setup" guide: https://github.com/hyperledger/indy-node/blob/master/docs/source/setup-dev.md When finally running "pytest ." in the indy-node directory, it fails with "ImportError: No module named '_lzma'". This seems (after a quick google search) to indicate a problem with the python environment. 1) The guide is aimed at Ubuntu 16.04, while I am running Linux Mint 19.3 (based on Ubuntu 18.04). May this be the cause of the error, and if so, (how) can I get it to work on my machine? In the bigger picture, I'm trying to get a better understanding of the communication between the Indy pool and clients. I've found out so far that it uses ZeroMQ, and as a next step I wanted to get indy-node to run locally and add debug output to print out all ZeroMQ messages that are being sent and received. 2) Are there some useful resources in the documentation that cover the topic of which messages are being sent/received via ZeroMQ?

WadeBarnes (Sat, 26 Sep 2020 14:46:21 GMT):
It may be better to start by understanding the intercommunication from a higher level first. That will also provide you with a better idea of the components and infrastructure involved. A good place to start is https://github.com/hyperledger/aries-cloudagent-python/blob/master/DevReadMe.md

arsulegai (Sun, 27 Sep 2020 12:04:03 GMT):
Hello Indy community from the Hyperledger India Chapter, we are calling for speakers to engage with the community in Asia Pacific, Europe and Africa. Please see our event details here https://www.linkedin.com/feed/update/urn:li:activity:6715897303481372672 Calling the tech enthusiasts, maintainers to be part of it.

adrijanrogan (Mon, 28 Sep 2020 07:32:21 GMT):
Has joined the channel.

adrijanrogan (Mon, 28 Sep 2020 07:32:21 GMT):
Hi, is there a reason that getting a schema in ledger.rs allows for a null DID, but cache.rs does not? - ledger: https://github.com/hyperledger/indy-sdk/blob/9a123f57550477b0720f36385236e3706eb3d315/libindy/src/api/ledger.rs#L671 - cache: https://github.com/hyperledger/indy-sdk/blob/9a123f57550477b0720f36385236e3706eb3d315/libindy/src/api/cache.rs#L90

adrijanrogan (Mon, 28 Sep 2020 07:32:21 GMT):
Hi, is there a reason that getting a schema in `ledger.rs` allows for a null DID, but `cache.rs` does not? - ledger: https://github.com/hyperledger/indy-sdk/blob/9a123f57550477b0720f36385236e3706eb3d315/libindy/src/api/ledger.rs#L671 - cache: https://github.com/hyperledger/indy-sdk/blob/9a123f57550477b0720f36385236e3706eb3d315/libindy/src/api/cache.rs#L90

AryamaI (Tue, 29 Sep 2020 07:56:32 GMT):
Hello @swcurran , can u confirm this pt please?

krgko (Tue, 29 Sep 2020 09:53:07 GMT):
Has joined the channel.

esplinr (Tue, 29 Sep 2020 13:10:09 GMT):
https://github.com/hyperledger/indy-node/blob/master/docs/source/transactions.md#domain-ledger

AryamaI (Tue, 29 Sep 2020 13:18:22 GMT):
Hello @esplinr , thanks for sharing the link above, I have checked it and see the transactions are for schema, credenital_defs. Can you tell me if I want to store not these standard indy data but rather ike store lets say some transaction done outside of indy, but store that transaction info on indy ledger like a financial transaction or something like that as an immutable record onto the ledger. Trying to understand if we can use indy ledger for more features than idenity

AryamaI (Tue, 29 Sep 2020 13:20:51 GMT):
Hello @esplinr , thanks for sharing the link above, I have checked it and see the transactions are for schema, credenital_defs. Can you tell me if I want to store for example the financial transaction data done outside indy onto the ledger as an immutable record. Trying to understand if we can use indy ledger for more features than idenity

esplinr (Tue, 29 Sep 2020 13:22:49 GMT):
Indy is an identity specific ledger.

esplinr (Tue, 29 Sep 2020 13:22:58 GMT):
So you would have to model the data as an existing Indy transaction.

esplinr (Tue, 29 Sep 2020 13:24:26 GMT):
Another approach is to creating a ledger plugin for your needs. This is a good example: https://github.com/sovrin-foundation/token-plugin

esplinr (Tue, 29 Sep 2020 13:25:05 GMT):
Or you can use indy-plenum as a general purpose block-chain without the indy-node transactions. https://github.com/hyperledger/indy-plenum/

esplinr (Tue, 29 Sep 2020 13:27:09 GMT):
I would like to hear more about your "financial transaction data" use case. There are a few groups working on similar problems: https://github.com/hyperledger/indy-hipe/pull/160

AryamaI (Tue, 29 Sep 2020 13:30:25 GMT):
@esplinr when you say model data as indy transaction that means either make a create schema, create credential definition transactions something like this rt . But I can't do a custom transaction?

esplinr (Tue, 29 Sep 2020 13:31:03 GMT):
Correct. Indy excludes custom transactions to discourage people putting random junk on the ledger.

AryamaI (Tue, 29 Sep 2020 13:35:41 GMT):
@esplinr The problem statement which we are looking into is a transaction done between two peers in a network, where before making the transaction the peers are both validated using indy identity. Once validated transaction happens and then store the transaction info onto the blockchain indy ledger if possible

AryamaI (Tue, 29 Sep 2020 13:35:41 GMT):
@esplinr The problem statement which we are looking into is a transaction done between two peers in a network which is not indy, where before making the transaction the peers are both validated using indy identity. Once validated transaction happens and then store the transaction info onto the blockchain indy ledger if possible

AryamaI (Tue, 29 Sep 2020 13:35:41 GMT):
@esplinr The problem statement which we are looking into is a transaction done between two peers in a network which is not indy network, where before making the transaction the peers are both validated using indy identity. Once validated transaction happens and then store the transaction info onto the blockchain indy ledger if possible

esplinr (Tue, 29 Sep 2020 13:38:26 GMT):
The identity information isn't stored on Indy. The credential is defined on Indy, and the identity information is transmitted peer-to-peer. This preserves individual privacy. It also makes it easy to use those credentials as part of any workflow, such as validating participants to transactions on a separate DLT network.

esplinr (Tue, 29 Sep 2020 13:38:55 GMT):
But if you really wanted to store additional information on Indy, then the best approach would be to write a ledger plugin.

esplinr (Tue, 29 Sep 2020 13:39:56 GMT):
https://github.com/hyperledger/indy-plenum/blob/master/docs/source/plugins.md

AryamaI (Tue, 29 Sep 2020 13:48:41 GMT):
@esplinr can you confirm if on the indy test/mainnet this new transaction type(plugin) functionality we can implement.

esplinr (Tue, 29 Sep 2020 13:50:43 GMT):
There is no Indy Test / MainNet.

esplinr (Tue, 29 Sep 2020 13:50:54 GMT):
Do you mean the Sovrin StagingNet / MainNet?

AryamaI (Tue, 29 Sep 2020 13:51:02 GMT):
Yes

esplinr (Tue, 29 Sep 2020 13:51:02 GMT):
You would have to propose the usage of your plugin, and get enough stewards to accept it.

esplinr (Tue, 29 Sep 2020 13:51:30 GMT):
The Sovrin network is designed to not allow non-identity transactions. Stewards don't want to be responsible for that data.

AryamaI (Tue, 29 Sep 2020 13:52:43 GMT):
Ok. If this functionality we want to incorporate then we can do it in our own network ledger rt as per the documentation?

esplinr (Tue, 29 Sep 2020 13:53:49 GMT):
It is possible, but might not be the preferred approach. In general, I would favor managing the data separate from the identity stuff.

esplinr (Tue, 29 Sep 2020 13:54:10 GMT):
You need to be careful to avoid correlation that undermines privacy.

AryamaI (Tue, 29 Sep 2020 13:55:24 GMT):
Ok. Thank you for the insights and suggestions @esplinr

pfeairheller (Tue, 29 Sep 2020 21:28:37 GMT):
I am trying to submit write transactions to Sovrin buildnet and for that I need to provide the TAA. For that I need the acceptance mechanism. When I attempt to read the mechanism list (TXN #7) I am getting no results. I'm using indy_vdr for these transactions. Any help would be greatly appreciated!

pfeairheller (Tue, 29 Sep 2020 21:52:35 GMT):
Being the impatient guy that I am I kept poking at this. At first I found the following article https://sovrin.org/preparing-for-the-sovrin-transaction-author-agreement/ and assumed that maybe the mechanism could be _anything_. So I passed something and got an error. And happily I found in the error message the following: `"Txn Author Agreement acceptance mechanism is inappropriate: provided session_instantiation, expected one of ['at_submission', 'click_agreement', 'for_session', 'on_file', 'product_eula', 'service_agreement', 'wallet_agreement']`. I now know the acceptable values.

pfeairheller (Tue, 29 Sep 2020 21:53:01 GMT):
I am still curious why the request to list the AML isn't working.

swcurran (Tue, 29 Sep 2020 22:18:03 GMT):
@andrew.whitehead ^^^^^

andrew.whitehead (Wed, 30 Sep 2020 00:52:18 GMT):
Are you looking on the right ledger? The TAA and AML transactions are on the config ledger, aka. subledger 2

pfeairheller (Wed, 30 Sep 2020 00:56:55 GMT):
Running the same request for TAA and AML. And the TAA works. I'll check in the morning to be sure.

andrew.whitehead (Wed, 30 Sep 2020 00:57:29 GMT):
What wrapper/interface are you using?

Dan (Wed, 30 Sep 2020 17:51:25 GMT):
Has left the channel.

lynn.bendixsen (Wed, 30 Sep 2020 21:31:01 GMT):
Using the indy-cli works to get the AML using `ledger get-acceptance-mechanisms` so you might be able to find example code in the indy-sdk/cli code to help. Start here maybe? https://github.com/hyperledger/indy-sdk/blob/master/cli/src/commands/ledger.rs

alexsiqueira (Wed, 30 Sep 2020 22:49:36 GMT):
Has joined the channel.

zickau (Thu, 01 Oct 2020 08:51:26 GMT):
Does anyone have a link on how to get write access on the Sovrin StagingNet for interoperability testings?

pfeairheller (Thu, 01 Oct 2020 10:40:32 GMT):
https://selfserve.sovrin.org/

pfeairheller (Thu, 01 Oct 2020 10:40:46 GMT):
Instructions here: https://docs.google.com/document/d/1sXZoN18lpFoAF075QoptofwDV_1otUylPGFKRQnA56E/edit

pfeairheller (Thu, 01 Oct 2020 10:44:00 GMT):
Ok, thanks for looking into that. I'm using Go wrapper on indy_vdr, calling `indy_vdr_build_get_acceptance_mechanisms_request`

pfeairheller (Thu, 01 Oct 2020 10:44:23 GMT):
I should've thought to try the cli.

dhuseby (Thu, 01 Oct 2020 16:13:37 GMT):
@esplinr @nage I just got a report back from LF legal highlighting some licensing issues in the Indy code base: https://lfscanning.org/reports/hyperledger/indy-2020-09-01-bebd6334-7d9e-4113-baf3-802b2113075e.html

dhuseby (Thu, 01 Oct 2020 16:14:42 GMT):
the JIRA has three issues filed against the repo: https://jira.hyperledger.org/projects/LSI/issues/LSI-17 https://jira.hyperledger.org/projects/LSI/issues/LSI-18 https://jira.hyperledger.org/projects/LSI/issues/LSI-19

andrew.whitehead (Thu, 01 Oct 2020 17:36:22 GMT):
A small thing but the indy crate also advertises itself as MIT/Apache-2 when it's actually just Apache-2: https://crates.io/crates/indy

esplinr (Thu, 01 Oct 2020 17:42:54 GMT):
I don't have access to the LSI issues in Jira.

esplinr (Thu, 01 Oct 2020 17:42:54 GMT):
@dhuseby I don't have access to the LSI issues in Jira.

andrew.whitehead (Thu, 01 Oct 2020 17:45:31 GMT):
It looks like maybe the Go wrapper should be using -1 for the timestamp (meaning not specified) but it's passing 0

pfeairheller (Thu, 01 Oct 2020 18:57:43 GMT):
Ah, ok.

pfeairheller (Thu, 01 Oct 2020 18:57:48 GMT):
I'll fix that

Lucampelli (Fri, 02 Oct 2020 15:41:55 GMT):
Thanks for the heads up! Now one more question appeared, how to decentralize it? I mean run it in more than one server? I know that in order to connect to it from outside all I need is the genesis.txn file from it, and the client will connect on it's own. But what would happen if I, let's say have two servers, A and B. Ran the docker on A, creating the genesisA.txn file, transfered the file to B, and used it as the genesis file for creating the pool in B. What would happen?

esplinr (Fri, 02 Oct 2020 15:44:20 GMT):
The genesis file should contain the information for all the initial pool nodes.

esplinr (Fri, 02 Oct 2020 15:44:56 GMT):
https://stackoverflow.com/questions/59896480/hyperledger-indy-create-genesis-transaction-file

dcspinho (Mon, 05 Oct 2020 00:17:23 GMT):
Has joined the channel.

dcspinho (Mon, 05 Oct 2020 00:17:24 GMT):
Hi everyone. I'm new here. I'm a recent graduated and I attend the last year of the master’s degree in Telecommunications and Informatics Engineering. I’m currently developing my dissertation with the theme "evaluation of blockchain technology in Identity Management context in Mobility as a Service". I already have a network to test (Hyperledger Indy), but I have some doubt. Some Hyperledger links don't redirect to the source code, like https://github.com/hyperledger/indy-sdk/blob/master/docs/how-tos/write-did-and-query-verkey/README.md Can you help me?

esplinr (Mon, 05 Oct 2020 15:42:28 GMT):
It's usually easiest to answer specific questions. https://github.com/hyperledger/indy-sdk/tree/master/docs/how-tos/write-did-and-query-verkey

swcurran (Mon, 05 Oct 2020 18:51:57 GMT):
On this Wednesday's Aries Morning call we are going to talk about Indy/Aries Revocation 2.0 by looking at the different options that have been put forward for scaled ZKP revocation. I've put together a document (link below) that I'd like especially those that already understand the options to review and make sure that I have my facts correct. Others interested in the topic are encouraged to join and I'll provide a brief summary of the topic and approaches as the basis for the conversation. The call is this Wednesday, October 7 at 7AM Pacific / 4PM CET using the *NEW* Aries Morning Call Zoom Link: https://zoom.us/my/hyperledger.community?pwd=STZQd0xMZU9xRVVOVnpQM3JNQ2dqZz09 Document for prereading: https://docs.google.com/document/d/1QHTxCrv0HUwLaqbOMabVoIj1Rc_lpF5WCd2l3QZcq2w/edit?usp=sharing

swcurran (Mon, 05 Oct 2020 18:51:57 GMT):
On this Wednesday's Aries Morning call we are going to talk about Indy/Aries Revocation 2.0 by looking at the different options that have been put forward for scaled ZKP revocation. I've put together a document (link below) that I'd like especially those that already understand the options to review and make sure that I have my facts correct. Others interested in the topic are encouraged to join and I'll provide a brief summary of the topic and approaches as the basis for the conversation. The call is this Wednesday, October 7 at 7AM Pacific / 4PM CET using the *NEW* Aries Morning Call Zoom Link: TO BE ADDED Document for prereading: https://docs.google.com/document/d/1QHTxCrv0HUwLaqbOMabVoIj1Rc_lpF5WCd2l3QZcq2w/edit?usp=sharing

swcurran (Mon, 05 Oct 2020 18:51:57 GMT):
On this Wednesday's Aries Morning call we are going to talk about Indy/Aries Revocation 2.0 by looking at the different options that have been put forward for scaled ZKP revocation. I've put together a document (link below) that I'd like especially those that already understand the options to review and make sure that I have my facts correct. Others interested in the topic are encouraged to join and I'll provide a brief summary of the topic and approaches as the basis for the conversation. The call is this Wednesday, October 7 at 7AM Pacific / 4PM CET using the *NEW* Aries A Morning Call Zoom Link: https://zoom.us/my/hyperledger.community.3?pwd=UE90WHhEaHRqOGEyMkV3cldKa2d2dz09 Document for prereading: https://docs.google.com/document/d/1QHTxCrv0HUwLaqbOMabVoIj1Rc_lpF5WCd2l3QZcq2w/edit?usp=sharing

vasantkk (Tue, 06 Oct 2020 13:12:18 GMT):
Hi Folks! I wanted to have a comparative differentiation between hyperledger indy vs Uport of Consensys vs Metadium, I have asked this question on the sovrin forum https://forum.sovrin.org/t/metadium-vs-sovrin/1496 here is the link and I still don't have answers. Indy and Metadium support multiple dids whereas uport supports only single did right?

vasantkk (Tue, 06 Oct 2020 13:23:34 GMT):
There are dozens of identity-related blockchain projects out there, and everytime we hear of a new blockchain-ID project we roll our eyes thinking, “another one?!” Civic, SelfKey, Metadium, Blockpass, VeriME, uPort, Sovrin, UniquID, BlockAuth, and the list goes on and on. How to differentiate from all these projects?

tuckcodes (Wed, 07 Oct 2020 02:32:19 GMT):
Has joined the channel.

tuckcodes (Wed, 07 Oct 2020 02:32:19 GMT):
:point_up_2: I second @vasantkk question

daidoji (Wed, 07 Oct 2020 02:56:47 GMT):
I believe the idea is that by allowing multiple DIDs and DID methods wallets and agents aren't necessarily tied to infrastructure or particular implementations

daidoji (Wed, 07 Oct 2020 02:57:53 GMT):
that's just my beginner's understanding though. Other people should be more able to compare actual features and spec implementations of all these systems. You might try the W3C identity WG. I think they were discussing this on one of their mailing lists or chats

andrew.whitehead (Wed, 07 Oct 2020 03:34:51 GMT):
I’m not really sure what it means for a ledger to only allow one DID. In the case of Indy/Aries it’s encouraged to use single-purpose DIDs for communication between agents, which ought to be possible no matter what the backing identity ledger is. As for a comparison between the different ledgers, I’m not aware of one, but I’m sure it would be welcome.

Sandeepk40 (Wed, 07 Oct 2020 04:24:06 GMT):
Has joined the channel.

vasantkk (Wed, 07 Oct 2020 07:19:17 GMT):
what is the difference between indy vs ethereum

daidoji (Wed, 07 Oct 2020 13:07:34 GMT):
They are two completely different types of ledgers.

vasantkk (Wed, 07 Oct 2020 13:09:59 GMT):
what makes erc725 different from indy?

vasantkk (Wed, 07 Oct 2020 13:10:05 GMT):
or SSI as a concept?

daidoji (Wed, 07 Oct 2020 13:16:53 GMT):
erc725 is a set of identity standards for working on etherum. Indy is a ledger technology meant to work with W3C VC as implemented with DIDs. SSI is a broad concept that covers lots of different forms of decentralized identity systems and how to go about making those systems useable on the Internet.

daidoji (Wed, 07 Oct 2020 13:17:15 GMT):
but like I said, the W3C Identity WG I think has a document or something you can ask about with an FAQ on all these things (or should)

daidoji (Wed, 07 Oct 2020 13:17:50 GMT):
I'm not even guaranteeing my answers are correct as I'm pretty new to this space as well and there are some subtleties.

vasantkk (Wed, 07 Oct 2020 13:19:01 GMT):
@daidoji yes I am aware of this I wanted to know about the subtleties and minor differences

daidoji (Wed, 07 Oct 2020 13:21:13 GMT):
word

vasantkk (Wed, 07 Oct 2020 13:24:59 GMT):
?

daidoji (Wed, 07 Oct 2020 13:28:25 GMT):
oh sorry, that's American slang for something like "roger" or "affirmative" or "ahh"

daidoji (Wed, 07 Oct 2020 13:28:34 GMT):
I forget the grammatical term

daidoji (Wed, 07 Oct 2020 13:28:47 GMT):
just an affirmation that I heard you and understand

vasantkk (Wed, 07 Oct 2020 13:30:44 GMT):
any idea of what is the status of other closed source SSI projects like civi, fractal, trustedId this is a closed source fork on sovrin wanted to know what is the status of it and also other than connect.me any apps on sovrin? @daidoji

daidoji (Wed, 07 Oct 2020 13:31:54 GMT):
I'm sorry, I don't know much about all those and there are a lot of sovrin apps, but probably not that many with much adoption yet

daidoji (Wed, 07 Oct 2020 13:31:59 GMT):
as is true for the whole space I think

mccown (Wed, 07 Oct 2020 19:51:50 GMT):
Just a reminder about the Identity Implementer's WG Call tomorrow morning (Oct 8th @ 9am MT). In this call, we review updates from a variety of industry WGs that you may have missed this week. For our discussion this week, I am planning to describe some cryptographic primitives and then we can all discuss if / how they might be of use to identity. Here are links to the Wiki and the Zoom meeting. Wiki: https://wiki.hyperledger.org/display/IWG/2020-10-08+Identity+Implementers+WG+Call Zoom: https://zoom.us/j/244779296

vasantkk (Thu, 08 Oct 2020 10:24:11 GMT):
How does sovrin ensure that the Issuer is not fake and not giving out fake certificates?

daidoji (Thu, 08 Oct 2020 13:42:43 GMT):
@vasantkk If the issuer is not trusted in the ecosystem then people will not accept its certificates. If the issuer is attempting to impersonate another issuer (like a fake US gov't issuer issuing passports) then they will likely not hold the right keys that the US gov't signs in its issuing documents (or delegates)

csunitha (Thu, 08 Oct 2020 15:20:28 GMT):
Has joined the channel.

pankajn (Thu, 08 Oct 2020 17:36:55 GMT):
Has joined the channel.

swcurran (Thu, 08 Oct 2020 18:16:45 GMT):
@daidoji has it right. Anyone can issue credentials to holders, and holders can present them to verifiers. That allows the verifier to trust what they get from the holder -- they can't be tampered with, are bound to the holder and the issuer is known. What the verifier has to decide if they trust the issuer. They may know the issuer because they only accepted credentials from certain issuers, or they have to somehow (there are techniques) for deciding if the issuer can be trusted. This the human trust side of VCs. As noted, an issuer can pretend to be another issuer without taking control of the identifier (and associated keys) used to issue the credential.

vasantkk (Fri, 09 Oct 2020 07:08:33 GMT):
is indy a blockchain/ledger or a did standard a bit confused from this article https://medium.com/linum-labs/the-whos-who-of-decentralized-identity-systems-433b2dd9a195

swcurran (Fri, 09 Oct 2020 12:56:09 GMT):
Indy is an open source ledger focused on identity, and in particular, verifiable credentials. You can run an instance of Indy to operate a ledger, such as the ones operated by sovrin.org. It is a "generic" blockchain underneath (plenum RBFT). with a transaction interface (indy-node) that limits the types of transactions to those specifically related to identity and verifiable credential use cases. Entities (people, organizations, things) run agents that communicate with the ledger to read and write data. It is public (anyone can read) permissioned (nodes of the network are permissioned to join consensus).

vasantkk (Fri, 09 Oct 2020 13:18:56 GMT):
is civic wallet open source cause I see this https://www.identity.com/

vasantkk (Fri, 09 Oct 2020 13:51:33 GMT):
for private key recovery does indy use shamir secret key sharing for recovery?

swcurran (Fri, 09 Oct 2020 17:41:38 GMT):
The current indy-sdk has a storage component ("indy-wallet") that supports exporting an encrypted version of wallet data with a key. What you do with that is up to you, so it would be a higher level component that would implement the storage of the export and how to handle the key for recovery. So recovery strategies are at a layer up from Indy.

lijiachuan (Sat, 10 Oct 2020 11:09:06 GMT):
hi @swcurran , I would like to have your suggestion about another scenario of sharing digital identity, for example, to be able to use one service provided from one service provider, user needs to provide some identities such as name, email or even bank information, we know that this needs to be shared securely with user's consent, and when user doesn't want to share this any more, he/she can revoke this consent. Is there any recommended approach to implement this? Traditionally this information is shared and stored in service provider's database, am not sure whether there is any other better way. Kindly advise. Thanks.

ZappaBoy (Sun, 11 Oct 2020 12:54:26 GMT):
Has joined the channel.

vasantkk (Mon, 12 Oct 2020 07:25:38 GMT):
anyone has ideas on how key generation happens on ShoCard? Can keys be generated by user or does he have to depend upon ShoCard server for key generation?

vasantkk (Mon, 12 Oct 2020 07:37:11 GMT):
anyone using shocard?

swcurran (Mon, 12 Oct 2020 08:24:16 GMT):
There is nothing inherent in SSI for the use case you are defining. The user has been issued some credentials. A verifier asks the user for proof of those credentials via a proof request. The user (holder/prover) decides to construct provide and send the proof. At that point, the verifier has received the data and can do with it what they want -- there is nothing in the technology to prevent that, and there is nothing the holder/prover can do to "revoke" having given over the information. Think of the same scenario with paper credentials -- showing your driver's licence. You can't go back later and say "Forget that I showed you my licence". Where this goes is the governance framework under which the participants are operating (if any). Under what conditions (e.g. GDPR, etc.) was the proof requested, and the proof sent. Finally -- to be clear, an issuer of a credential can often revoke the credential to prevent the holder/prover from being able to continue to prove the credential they were issued.

Nabil-oussa (Mon, 12 Oct 2020 08:35:38 GMT):
Has joined the channel.

Nabil-oussa (Mon, 12 Oct 2020 08:35:38 GMT):
Hi all: Currently I am on a project which brings together your project on the hyperledger Indy platform, my question which is the mobile agent that you used to develop the holder part. My question is precisely what type of mobile agent I can use and which is functional Aries MAX (Aries Mobile Agent Xamarin) - Xamarin / .Net, Aries React-Native Mobile Agent (ARNIMA) - React Native or React Native Indy SDK (rn-indy-sdk) - React Native. Thanks in advance for your help

ankita.p17 (Mon, 12 Oct 2020 08:40:36 GMT):
Has joined the channel.

gnarula (Mon, 12 Oct 2020 08:59:45 GMT):
Hi everyone! I've a few questions around DKMS (https://github.com/hyperledger/aries-rfcs/blob/master/concepts/0051-dkms/dkms-v4.md). The document mentions an "Agent Policy Registry" for multi-device management. Is that implemented on Indy? Also, does libindy provide any recovery mechanisms? In particular, I'm more interested in social key recovery using Shamir Secret Sharing.

gnarula (Mon, 12 Oct 2020 09:06:00 GMT):
I think I also read some discussion about key rotation the other day but I'd like to know if my understanding's correct - Does Indy only support resolving the latest key for a DID every time? Is there a way to resolve the key at timestamp=x? I don't think indy-vdr exposes any such API (yet?). I'm curious to know if that's a limitation of the current design or it is by design? @danielhardman @andrewwhi

gnarula (Mon, 12 Oct 2020 09:06:00 GMT):
I think I also read some discussion about key rotation the other day but I'd like to know if my understanding's correct - Does Indy only support resolving the latest key for a DID every time? Is there a way to resolve the key at timestamp=x? I don't think indy-vdr exposes any such API (yet?). I'm curious to know if that's a limitation of the current design or it is by design? @danielhardman @andrew.whitehead

askemeline (Mon, 12 Oct 2020 13:10:06 GMT):
Has joined the channel.

askemeline (Mon, 12 Oct 2020 13:10:06 GMT):
HI there, I got a trouble when I installed indy-cli on ubuntu `Unable to locate package indy-cli ` Some know how to solve it?

vladyslavmunin (Mon, 12 Oct 2020 13:11:51 GMT):
rich schema

lijiachuan (Mon, 12 Oct 2020 13:19:54 GMT):
Ok, got it, so actually indy should only be used for this type of credential issuance, request and verify, but how the data will be used after verify is out of the scope but needs to be defined by another approach/framework.

tejgundavelli (Mon, 12 Oct 2020 22:33:49 GMT):
Has joined the channel.

darkchylde (Tue, 13 Oct 2020 07:32:14 GMT):
Has joined the channel.

darkchylde (Tue, 13 Oct 2020 07:32:14 GMT):
Hi would like to get some feedback ? I have raised an issue in indy-node repo regarding poor documentation for running production capable indy network and its governance. https://github.com/hyperledger/indy-node/issues/1625

vineeta (Tue, 13 Oct 2020 07:45:51 GMT):
Has joined the channel.

swcurran (Tue, 13 Oct 2020 16:52:22 GMT):
I recall that some independent testing was done on Indy, as was some performance testing. Does anyone know where any documents produced from those efforts exist?

lynn.bendixsen (Wed, 14 Oct 2020 17:10:40 GMT):
Evernym did some performance testing, and kept track of their results for a long period of time. https://docs.google.com/spreadsheets/d/1DTjDsLSysFBiKU-9z4-IzunJk4wEy44hE_PGZYxnN_8/edit#gid=343667650

swcurran (Wed, 14 Oct 2020 17:35:06 GMT):
Thanks. I'm guessing no interpretation document is available. This is very helpful, but requires a lot of interpretation.

MathewJoseph (Thu, 15 Oct 2020 08:46:26 GMT):
Has joined the channel.

MathewJoseph (Thu, 15 Oct 2020 08:46:26 GMT):
Hi i am currently facing difficulties using Indy while connecting to the indy pool from android device. Whenever i try to open the pool ledger the app suddenly quites and gives A/libc: Fatal signal 6 (SIGABRT) error. I have downloaded the compiled binaries and added it in jni. All other wallet related operations are successfull apart from open pool. Can someone please help guide me in solving the problem ? I am using v15 of indy-sdk

MathewJoseph (Thu, 15 Oct 2020 08:46:26 GMT):
Hi i am currently facing difficulties using Indy while connecting to the indy pool from android device. Whenever i try to open the pool ledger the app suddenly quits and gives A/libc: Fatal signal 6 (SIGABRT) error. I have downloaded the compiled binaries and added it in jni. All other wallet related operations are successfull apart from open pool. Can someone please help guide me in solving the problem ? I am using v15 of indy-sdk

MathewJoseph (Thu, 15 Oct 2020 08:49:14 GMT):
The project is build using react native and i have create a bridge java package which will expose the indy functionalities to the javascript side. I tried debugging the openPoolLedger function call and found that it fails when it calls `indy_open_pool_ledger` api function.

MathewJoseph (Thu, 15 Oct 2020 08:49:14 GMT):
The project is build using react native and i have created a bridge java package which will expose the indy functionalities to the javascript side. I tried debugging the openPoolLedger function call and found that it fails when it calls `indy_open_pool_ledger` api function.

MathewJoseph (Thu, 15 Oct 2020 08:50:39 GMT):
The pool creation was successfully from my understanding as i was able to see config.json and poolname.txn files inside pool folder of apk folder

TimoGlastra (Thu, 15 Oct 2020 10:54:44 GMT):
@MathewJoseph Which version of RN are you using? This is a problem with react-native > 0.61.5 currently. We're having the same problem with aries-framework-javascript. See https://github.com/facebook/react-native/issues/29930 and https://github.com/react-native-community/upgrade-support/issues/114 . So assuming you're on RN > 0.61.5 for now the "fix" will be to downgrade to version 0.61.5

MathewJoseph (Thu, 15 Oct 2020 11:03:09 GMT):
@TimoGlastra I am using RN 0.63.2 for my project. Oh now it all makes sense i had a sample project of 0.61.5 and open pool was working fine in it but when i integrated it to the main app it started crashing even though everything was same.

MathewJoseph (Thu, 15 Oct 2020 11:04:59 GMT):
Thank you so much for the tip. I have being tearing my hair off for the last 2-3 days thinking why its not working in the main app.

ArturPhilipp (Thu, 15 Oct 2020 13:44:22 GMT):
Has joined the channel.

ArturPhilipp (Thu, 15 Oct 2020 13:47:23 GMT):
Hey everybody. I have a question regarding the performance of indy based networks. I found an older JIRA issue (https://jira.hyperledger.org/browse/INDY-1343%208) which mentions, that a possible write throughput of 10 tx/sec was possible, which seems not so much to me. Are there any other performance benchmarks out there (with newer versions of indy and not from 2018)? I searched a lot, but was unable to find anything adequate...

ArturPhilipp (Thu, 15 Oct 2020 13:47:23 GMT):
Hey everybody. I have a question regarding the performance of indy based networks. I found an older JIRA issue (https://jira.hyperledger.org/browse/INDY-1343%208) which mentions, that a write-throughput of 10 tx/sec was possible, which seems not so much to me. Are there any other performance benchmarks out there (with newer versions of indy and not from 2018)? I searched a lot, but was unable to find anything adequate...

ArturPhilipp (Thu, 15 Oct 2020 13:47:23 GMT):
Hey everybody. I have a question regarding the performance of indy based networks. I found an older JIRA issue (https://jira.hyperledger.org/browse/INDY-1343%208) which mentions, that a write-throughput of 10 tx/sec was possible, which seems not so much to me. Are there any other performance benchmarks out there (with newer versions of indy and not from 2018)? I searched a lot, but was unable to find anything up2date...

ArturPhilipp (Thu, 15 Oct 2020 14:02:20 GMT):
Are there any newer tests since the ones mentioned in https://jira.hyperledger.org/browse/INDY-1343%208 are from 2018?

ArturPhilipp (Thu, 15 Oct 2020 14:02:20 GMT):
Are there any newer tests since the ones mentioned in https://jira.hyperledger.org/browse/INDY-1343%208 are from 2018? Not sure if during the evolution of indy brought some performance benefits...

ArturPhilipp (Thu, 15 Oct 2020 14:02:20 GMT):
Are there any newer tests since the ones mentioned in https://jira.hyperledger.org/browse/INDY-1343%208 are from 2018? Not sure if the evolution of indy brought some performance benefits...

lynn.bendixsen (Thu, 15 Oct 2020 16:02:23 GMT):
Looks like tests have been run more recently, but that the link I provided does not have those results listed... perhaps @esplinr can help?

braduf (Thu, 15 Oct 2020 22:41:46 GMT):
Has joined the channel.

ArturPhilipp (Fri, 16 Oct 2020 05:49:33 GMT):
I would appreciate that very much! :)

kim27 (Fri, 16 Oct 2020 21:51:03 GMT):
Has joined the channel.

Shubham95 (Sun, 18 Oct 2020 19:17:20 GMT):
Has joined the channel.

Shubham95 (Sun, 18 Oct 2020 19:18:15 GMT):
I am having hard time setting up indy sdk in ubuntu 20.04 .

Shubham95 (Sun, 18 Oct 2020 19:18:15 GMT):
I am having hard time setting up indy sdk in ubuntu 20.04 . Can someone please help

pixelschnitzel (Mon, 19 Oct 2020 07:21:15 GMT):
Hi. The GitHub documentation tells, that binaries for indy exist for Ubuntu 16.04 and 18.04. You can compile it by yourself for 20.04.

pixelschnitzel (Mon, 19 Oct 2020 07:21:15 GMT):
Hi. The GitHub documentation tells, that binaries for indy exist for Ubuntu 16.04 and 18.04. You can compile it by yourself for 20.04. You can check the Sovrin-Repo by yourself, it only has xenial and bionic: https://repo.sovrin.org/deb/dists/

Shubham95 (Mon, 19 Oct 2020 07:38:25 GMT):
Okay thanks

gnarula (Mon, 19 Oct 2020 21:26:17 GMT):
Hi! Does anyone know how to set the log level for libindy in iOS?

lynn.bendixsen (Mon, 19 Oct 2020 23:11:18 GMT):
I am not really sure but I have the following in my iOS env and I think it was for libindy... RUST_LOG=indy=trace

gnarula (Mon, 19 Oct 2020 23:12:22 GMT):
thanks! I tried setting that in my scheme but it didn't work :( I managed to debug my issue anyway

kh_touati (Tue, 20 Oct 2020 11:09:51 GMT):
Hi, I am looking for the newest roadmap of the indy project (mainly Indy network and ledger). I did not find an updated one. Could you please share it here. Thanks in advance

Priya.P (Tue, 20 Oct 2020 11:43:49 GMT):
Has joined the channel.

Priya.P (Tue, 20 Oct 2020 11:46:03 GMT):
I am trying to run getting started dockerized application.I am getting this error. Can anyone help me here. Collecting ipython-notebook Could not find a version that satisfies the requirement ipython-notebook (from versions: ) No matching distribution found for ipython-notebook You are using pip version 8.1.1, however version 20.2.4 is available. You should consider upgrading via the 'pip install --upgrade pip' command. ERROR: Service 'jupyter' failed to build: The command '/bin/sh -c pip3 install -U pip ipython-notebook ipython==7.9 setuptools jupyter python3-indy==1.11.0' returned a non-zero code: 1

ianco (Tue, 20 Oct 2020 13:32:18 GMT):
@Priya.P which demo are you trying to run?

Priya.P (Tue, 20 Oct 2020 14:04:13 GMT):
@ianco I am trying this sample https://hyperledger-indy.readthedocs.io/projects/sdk/en/latest/docs/getting-started/run-getting-started.html

Priya.P (Tue, 20 Oct 2020 14:05:58 GMT):
I am using docker v19.03.13 & docker-compose v1.25.3 Is there any other prerequisite to run this sample?

ianco (Tue, 20 Oct 2020 15:51:56 GMT):
I'm not sure I've run into the same error running the jupyter samples (running on Mac OSX)

ianco (Tue, 20 Oct 2020 15:52:44 GMT):
Are you just starting out or do you have a lot of experience with Indy/Aries?

ianco (Tue, 20 Oct 2020 15:53:20 GMT):
Also curious about your goals - are you just learning, or do you have a particular applicaiton in mind?

esplinr (Tue, 20 Oct 2020 17:00:28 GMT):
Most performance test results are captured in the tickets linked to https://jira.hyperledger.org/browse/INDY-1343 Such as https://jira.hyperledger.org/browse/INDY-1717 Then we largely rolled it into our regular testing process with video discussions: https://www.youtube.com/watch?v=984qU5q9e1Y&list=PLRp0viTDxBWGLdZk0aamtahB9cpJGV7ZF&index=46 https://www.youtube.com/watch?v=tyEDaeeUgfA&list=PLRp0viTDxBWGLdZk0aamtahB9cpJGV7ZF&index=38 https://youtu.be/OSpssHQXlOI https://www.youtube.com/watch?v=G-qlQkRJVWo  We haven't done as much performance testing in 2020, as the pace of development has slowed.

Priya.P (Wed, 21 Oct 2020 13:42:07 GMT):
I have just started with indy have worked with other blockchain frameworks. Basically we want to understand working of indy before we start using it for our identity management usecase.

khalifa (Wed, 21 Oct 2020 13:51:41 GMT):
Hi All, I want to share with you this medium tutorial (https://medium.com/@khalifa.toumi/how-to-use-hyperledger-aries-cloud-agent-for-a-classical-workflow-issuer-holder-verifier-9dd595f2f847) that describes the different steps to have a simple workflow between a verifier, issuer and a holder. Do not hesitate to comment it. I hope that it may help someone.

dgt1nsty (Wed, 21 Oct 2020 14:08:58 GMT):
Has joined the channel.

dgt1nsty (Wed, 21 Oct 2020 14:26:07 GMT):
Hi I tried integration tests on local nodes on

dgt1nsty (Wed, 21 Oct 2020 15:34:23 GMT):
hi to all I'm new to hyperleger-indy and I tried to test the indy-sdk codebase by starting a pool of local nodes using docker but when I run integration tests it failed from 1 test namely "test services::pool::merkle_tree_factory::tests::pool_worker_build_node_state_works_for_new_format ..." What may cause this error and how can I solve it.

ianco (Wed, 21 Oct 2020 15:34:37 GMT):
I recommending starting with Aries rather than Indy - Aries in an agent layer that sits on top of Indy and eventually will interoperate across other verifiable credential networks as well

dgt1nsty (Wed, 21 Oct 2020 15:34:52 GMT):
hi to all I'm new to hyperleger-indy and I tried to test the indy-sdk codebase by starting a pool of local nodes using docker but when I run integration tests it failed from 1 test namely "test services::pool::merkle_tree_factory::tests::pool_worker_build_node_state_works_for_new_format ..." What may cause this error and how can I solve it.

ianco (Wed, 21 Oct 2020 15:34:58 GMT):
There are some good Aries toturials that also cover the underlying Indy material

ianco (Wed, 21 Oct 2020 15:35:37 GMT):
A good place to start is the Aries python cloudagent demo's: https://github.com/hyperledger/aries-cloudagent-python/tree/master/demo

ianco (Wed, 21 Oct 2020 15:36:15 GMT):
Also there are a couple of EdX courses that cover Indy/Aries in more detail and explain the difference between a verifiable credentials network and your typical blockchain

ianco (Wed, 21 Oct 2020 15:36:36 GMT):
https://www.edx.org/course/identity-in-hyperledger-aries-indy-and-ursa

ianco (Wed, 21 Oct 2020 15:36:47 GMT):
https://www.edx.org/course/becoming-a-hyperledger-aries-developer

dgt1nsty (Wed, 21 Oct 2020 15:49:01 GMT):
is there any doc that describes how the nodes are communicating with each other in a technical way in other words. what are the operations happening at the back while communicating? Thanks in advance

ianco (Wed, 21 Oct 2020 16:33:09 GMT):
The Indy nodes are based on Indy Plenum, which is described here: https://github.com/hyperledger/indy-plenum

Jonathancj (Wed, 21 Oct 2020 23:55:03 GMT):
Has joined the channel.

Priya.P (Thu, 22 Oct 2020 11:27:27 GMT):
Thanks:thumbsup:

csunitha (Thu, 22 Oct 2020 13:41:32 GMT):
Has left the channel.

dgt1nsty (Thu, 22 Oct 2020 15:09:48 GMT):
I'll look it up, thanks

Koushik (Thu, 22 Oct 2020 15:26:47 GMT):
Hi guys, I am doing tech spec overview on Hyperledger Indy specifically indy nodes, can anyone point me to technical documents or answer on what is the average read per second for Hyperledger Indy.

daidoji (Thu, 22 Oct 2020 15:34:26 GMT):
@Koushik, See this message from October https://chat.hyperledger.org/channel/indy?msg=2dPGdExLAiq47Qydc

ColinClark (Sun, 25 Oct 2020 20:39:18 GMT):
Has joined the channel.

ColinClark (Sun, 25 Oct 2020 20:42:19 GMT):
there's an error in the getting-started.dockerfile - change "ipython-notebook" to "notebook" and re-run docker-compose up and you should be fine

ColinClark (Sun, 25 Oct 2020 20:43:10 GMT):
i've submitted a PR for this - jupyter has updated the package and doesn't include "ipython" in the name any longer

ColinClark (Sun, 25 Oct 2020 20:47:17 GMT):
You can try my repo if you want to run the example - it's here https://github.com/Xaiow/indy-sdk

pavanMude 1 (Tue, 27 Oct 2020 06:46:58 GMT):
Has joined the channel.

pavanMude 1 (Tue, 27 Oct 2020 06:46:58 GMT):
Hi I am currently facing issues on Native iOS after installing pods and adding requisite files to the project, after following indy integration steps still it throws this error `error: module compiled with Swift 4.2.1 cannot be imported by the Swift 5.3 compiler /Users/IndySdk/ios/Pods/libindy-objc/libindy-objc/Indy.framework/Modules/Indy.swiftmodule/arm64.swiftmodule. `Libindy-objc` is not maintained frequently it seems. What can be done here, Can we expect a new release to avoid swift compiler error !?

gnarula (Tue, 27 Oct 2020 07:28:52 GMT):
yeah, I ran into the above as well. Unfortunately the only way around it for now is to build libindy-objc yourself using the xcode/swift you have locally to avoid version conflicts

pavanMude 1 (Tue, 27 Oct 2020 08:05:12 GMT):
@gnarula Thanks for your response, Can you please elaborate, the procedure that you had followed while dealing with this issue and how did you overcome this error!? I appreciate your help here.

gnarula (Tue, 27 Oct 2020 08:20:35 GMT):
@pavanMude 1 Clone indy-sdk from https://github.com/hyperledger/indy-sdk and cd into `wrappers/ios/libindy-pod`. Do a `pod install` there and then open the `Indy.xcworkspace` file using XCode. Build the `Indy` Project and then Archive it. View the archive in finder, right click, show archive-contents and then copy 'Indy.framework' from '/Products/Library/Frameworks' and import it to your iOS project

pavanMude 1 (Tue, 27 Oct 2020 08:35:35 GMT):
@gnarula Thanks for your help, I will give a try then.

pavanMude 1 (Tue, 27 Oct 2020 09:03:16 GMT):
@gnarula While installing pods from cloned `libindy-pod` project, it throws warning `[!] The Podfile contains framework or static library targets (Indy), for which the Podfile does not contain host targets (targets which embed the framework).` Do you have idea why it is happening !?

esplinr (Tue, 27 Oct 2020 14:21:14 GMT):
Remember that the Indy Contributors call is in 40 minutes. North America doesn't start winter time until this weekend, so it's at a different time for the European attendees. Join here: https://zoom.us/my/hyperledger.community?pwd=STZQd0xMZU9xRVVOVnpQM3JNQ2dqZz09

swcurran (Tue, 27 Oct 2020 17:37:43 GMT):
FYI - We plan on having weekly `did:indy` Method Specification Drafting meetings on a weekly basis. I propose that the meetings be held at 7AM Pacific Time on Tuesdays, which is the hour before the bi-weekly Indy Contributors call. Request: Please let me know in replies if you plan on participating, and if so, if the time is OK with you, or what times would you prefer? Thanks!

daidoji (Tue, 27 Oct 2020 17:56:57 GMT):
this might be a silly question, but what will be the broad differences between `did:indy` and `did:sov`?

daidoji (Tue, 27 Oct 2020 17:58:23 GMT):
or what is the motivation for `did:indy`?

swcurran (Tue, 27 Oct 2020 18:13:44 GMT):
The main motivation is that `did:indy` will have a component to identify the network instance on which to resolve the DID. That is missing from `did:sov`. So an indy did will like look something like this: `did:indy:7GF4e:1231243wer...` where the `7GF4e` is derived from a hash of something that will identify the network. The exact mechanism to be used is to be determined.

swcurran (Tue, 27 Oct 2020 18:14:39 GMT):
That allows DIDs to be resolvable on any published Indy network -- Sovrin, Sovrin Staging, IDUnion, etc.

daidoji (Tue, 27 Oct 2020 18:45:45 GMT):
ahhh, that makes sense. Thanks!

mickra (Wed, 28 Oct 2020 10:27:09 GMT):
Has joined the channel.

Lucampelli (Wed, 28 Oct 2020 11:59:31 GMT):
Good day! I was wondering if it's possible to create a proof predicate comparing strings? I attempted setting the predicate to "=" and received the error telling me that it must be any of ">=","<=","<",">", so I was wondering if that is possible with proof predicates or only in attributes? Thanks

esplinr (Wed, 28 Oct 2020 12:47:43 GMT):
Currently it only supports integer comparisons, so the string has to be encoded as an integer. There is work underway to define those encodings on the ledger, but it's not done yet.

swcurran (Wed, 28 Oct 2020 12:57:16 GMT):
Trickier yet -- a sortable encoding.

Lucampelli (Wed, 28 Oct 2020 13:28:21 GMT):
for now I'm just decimal encoding the string, so the way would be to encode the string and use it in two predicates? One <= and the other >=. because if it's true on both it means its ==. Right?

swcurran (Wed, 28 Oct 2020 13:34:15 GMT):
That works. Note that there is also a `restriction` that you can use for the equality of the value if that is all you are looking for.

Lucampelli (Wed, 28 Oct 2020 14:01:33 GMT):
How do I use then? Can't seem to find it in the docs...

Lucampelli (Wed, 28 Oct 2020 14:01:40 GMT):
Also, Thanks!

AlphaX (Wed, 28 Oct 2020 17:06:02 GMT):
Has joined the channel.

AlphaX (Wed, 28 Oct 2020 17:06:03 GMT):
Hi, is there any tool to evaluate performance of Hyperledger Indy framework ? like Hyperledger Caliper For Hyperledger Fabric

troyronda (Wed, 28 Oct 2020 17:42:48 GMT):
Has left the channel.

esplinr (Wed, 28 Oct 2020 18:59:44 GMT):
https://github.com/hyperledger/indy-test-automation/tree/master/analysis

wenjing (Wed, 28 Oct 2020 21:28:09 GMT):
Has joined the channel.

wenjing (Wed, 28 Oct 2020 21:28:09 GMT):
I'm going through @esplinr 's indy-cli How-to to access sovrin buildernet. And I'm stuck in the step to write a schema. I set taa mechanism to be for_session in the cliconfig file. What else do I need? pool(buildernet):my_wallet:did(XD7...wvv):indy> ledger schema name=vswFirstSchema version=1.0 attr_names=FirstName,LastName,Address,Birthdate,SSN Transaction author agreement Acceptance Mechanism isn't set. pool(buildernet):my_wallet:did(XD7...wvv):indy> pool(buildernet):my_wallet:did(XD7...wvv):indy> ledger txn-acceptance-mechanisms aml={"taaAcceptanceMechanism":"for_session"} version=2.0 Transaction has been rejected: Not enough TRUSTEE signatures

wenjing (Wed, 28 Oct 2020 21:35:34 GMT):
Btw, here is the How-to I was following, except I ignored the payment-address, guessing it's no longer required. https://sovrin.org/wp-content/uploads/Token-Test-Instructions.pdf. What am I missing? Thanks!

Lucampelli (Wed, 28 Oct 2020 21:57:53 GMT):
which restriction? right now I could only find that it restricts for schema id and other things but not the value

swcurran (Wed, 28 Oct 2020 22:52:08 GMT):
Restrictions come in 6 forms: - schema ID - schema Issuer (DID) - Cred Def ID - Cred Def Issuer (DID) - attribute name (e.g. claim name) - attribute value So if you Cred Def ID and attribute value, you have the equivalent of an "equal" predicate. You are exposing the data, but that's OK if it's equality you are after -- if you a claim is equal to value, you know the value.

Lucampelli (Wed, 28 Oct 2020 22:52:24 GMT):
found about the "attr::name::value" restriction, but right now, with libindy 1.15 xenial, I'm getting this error, as well as the cred handle not finding the credential even though it does exists: Error occurred: Error: Invalid structure Caused by: Requested restriction validation failed for "{"Name": Some("Luca Campelli")}" attributes Caused by: $and operator validation failed. Caused by: $eq operator validation failed for tag: "attr::name::value", value: "Luca Campelli" Caused by: Unknown Filter Type

Lucampelli (Wed, 28 Oct 2020 22:52:24 GMT):
found about the "attr::name::value" restriction, but right now, with libindy 1.15 xenial, I'm getting this error, as well as the cred handle not finding the credential even though it does exists: Error occurred: Error: Invalid structure Caused by: Requested restriction validation failed for "{"Name": Some("Luca Campelli")}" attributes Caused by: $and operator validation failed. Caused by: $eq operator validation failed for tag: "attr::name::value", value: "Lucampelli" Caused by: Unknown Filter Type

Lucampelli (Wed, 28 Oct 2020 22:52:24 GMT):
found about the "attr::name::value" restriction, but right now, with libindy 1.15 xenial, I'm getting this error, as well as the cred handle not finding the credential even though it does exists: Error occurred: Error: Invalid structure Caused by: Requested restriction validation failed for "{"Name": Some("Lucampelli")}" attributes Caused by: $and operator validation failed. Caused by: $eq operator validation failed for tag: "attr::name::value", value: "Lucampelli" Caused by: Unknown Filter Type

Lucampelli (Wed, 28 Oct 2020 22:52:55 GMT):
also thank you for the help!

brentzundel (Thu, 29 Oct 2020 14:28:53 GMT):
You need to start indy-cli with the command `indy-cli --config config.json`

brentzundel (Thu, 29 Oct 2020 14:29:24 GMT):
where config.json is `{ "taaAcceptanceMechanism": string - transaction author agreement acceptance mechanism to be used when sending write transactions to the Ledger. }`

brentzundel (Thu, 29 Oct 2020 14:29:24 GMT):
where config.json is `{ "taaAcceptanceMechanism": string - transaction author agreement acceptance mechanism to be used when sending write transactions to the Ledger. }`

wenjing (Thu, 29 Oct 2020 16:40:09 GMT):
Thanks. embarrassingly silly. Can i safely skip the token plugin now for buildernet and stagingnet? Just want to get a confirm to be sure.

brentzundel (Thu, 29 Oct 2020 16:43:55 GMT):
on staging net and builder net you have the option of either use the token plugin, or to use transaction endorsers

brentzundel (Thu, 29 Oct 2020 16:45:17 GMT):
this website can help you get setup with tokens or endorser status on staging net: https://selfserve.sovrin.org/

brentzundel (Thu, 29 Oct 2020 16:45:17 GMT):
this website can help you get setup with tokens or endorser status on builder net or staging net: https://selfserve.sovrin.org/

wenjing (Thu, 29 Oct 2020 16:54:18 GMT):
Ok. I plugged in DID and verkey (again left the payment address field blank) in the selfserve page and it seems to work. Once this DID becomes an endorser, it can do write operations without tokens. Is that right understanding?

brentzundel (Thu, 29 Oct 2020 16:54:43 GMT):
that is correct

wenjing (Thu, 29 Oct 2020 16:57:32 GMT):
Thanks! very helpful to clear up some doubts. Is there a good place to figure out how the mainnet handle payments? like api definition?

brentzundel (Thu, 29 Oct 2020 16:58:07 GMT):
tokens are not currently used on mainnet, all transactions need to be signed by an endorser

wenjing (Thu, 29 Oct 2020 17:00:44 GMT):
I see. Presumably the endorser would create a "normal" payment account with sovrin to handle fees? Are there any other conditions for anyone to become an endorser?

brentzundel (Thu, 29 Oct 2020 17:02:11 GMT):
That is my understanding, but for specifics you'll need to reach out to Sovrin directly. I'd recommend asking questions on chat.sovrin.org

wenjing (Thu, 29 Oct 2020 17:03:27 GMT):
Great! Will do. Thanks so much for clearing it up for me.

brentzundel (Thu, 29 Oct 2020 17:03:42 GMT):
happy to be of asistance

wenjing (Thu, 29 Oct 2020 17:37:15 GMT):
Another problem popped up. The "ledger cred-def" command requires "primary" which is said to be the "primary key in json format". What is this? how to find it or create it?

lynn.bendixsen (Thu, 29 Oct 2020 17:58:29 GMT):
The `ledger cred-def` command cannot be run from the indy-cli without using an external program to build/create the primary key. I suggest switching completely to code to reliably create the cred-def.

lynn.bendixsen (Thu, 29 Oct 2020 18:00:21 GMT):
payment address is optional for writing txns to the network, but is required if you are testing token functionality.

wenjing (Thu, 29 Oct 2020 18:48:52 GMT):
Thanks. That's what I thought as I follow the How-to coding series in the doc. By the way, for anyone who wants the details - the anoncred paper is where the ultimate answer is. The indy-cli tutorial probably should find a way to remove its reference by a higher level of simplification.

swcurran (Thu, 29 Oct 2020 22:09:37 GMT):
FYI - the first of a (hopefully short-lived) series of meetings about defining the Indy DID Method Spec has been scheduled for this coming Tuesday November 3 at 7AM Pacific / 15:00 UTC. Calendar information for the event is here: https://lists.hyperledger.org/g/indy/viewevent?repeatid=32661&eventid=955634&calstart=2020-11-03 Information about this meeting series is here: https://wiki.hyperledger.org/display/indy/Drafting+the+Indy+DID+Method+Specification On the agenda this week is getting started and a discussion on the contents of the component of the DID. Please join in and help with this collaborative effort.

adrianblakey (Fri, 30 Oct 2020 14:59:58 GMT):
Has joined the channel.

adrianblakey (Fri, 30 Oct 2020 14:59:58 GMT):
How do I get python talking to libindy on a Mac? I have followed the instructions and installed it from source and have the ENVR's DYLD_LIBRARY_PATH and LD_LIBRARY_PATH pointed to /Users/ajb/indy-sdk/libindy/target/debug which has the dynlib in it. But running python - the traceback ... ERROR:indy.libindy:_load_cdll: Can't load libindy: dlopen(libindy.dylib, 6): image not found

adrianblakey (Fri, 30 Oct 2020 15:16:21 GMT):
Fixed it - I think LIBRARY_PATH looks like it's the setting

ianco (Fri, 30 Oct 2020 15:17:49 GMT):
I link from the `/usr/local/lib` directory rather than updating library paths

ianco (Fri, 30 Oct 2020 15:19:07 GMT):
`ln -s /usr/local/lib/` I think

ianco (Fri, 30 Oct 2020 15:19:07 GMT):
`ln -s /usr/local/lib/` I think

simnic (Mon, 02 Nov 2020 09:38:20 GMT):
Has joined the channel.

dcspinho (Mon, 02 Nov 2020 13:39:03 GMT):
Hello. How can you create nodes in Indy?

dgt1nsty (Mon, 02 Nov 2020 14:00:05 GMT):
Hi, I'm trying to run indy getting-started docker image and I get tis error "Could not find a version that satisfies the requirement ipython-notebook (from versions: ) No matching distribution found for ipython-notebook". Is there anyone know how to solve it?

ianco (Mon, 02 Nov 2020 14:03:30 GMT):
There's a PR that fixes this: https://github.com/hyperledger/indy-sdk/pull/2261

dgt1nsty (Mon, 02 Nov 2020 15:09:28 GMT):
Thanks a lot

dgt1nsty (Mon, 02 Nov 2020 15:09:40 GMT):
I didn't see it

swcurran (Tue, 03 Nov 2020 00:18:46 GMT):
REMINDER: The first of a (hopefully short-lived) series of meetings about defining the Indy DID Method Spec has been scheduled for Tuesday November 3 at 7AM Pacific / 15:00 UTC. Calendar information for the event is here: https://lists.hyperledger.org/g/indy/viewevent?repeatid=32661&eventid=955634&calstart=2020-11-03 Information about this meeting series is here: https://wiki.hyperledger.org/display/indy/Drafting+the+Indy+DID+Method+Specification On the agenda this week is getting started and a discussion on the contents of the component of the DID. Please join in and help with this collaborative effort.

MathewJoseph (Tue, 03 Nov 2020 06:51:02 GMT):
HI i am working on a RN project which uses indy. I am trying to implement the issue credential flow and the second phase of generating the credential request from prover side is causing issue for me. I debugged the java code in android studio and found that the issue is happening in the rust indy api lib. https://github.com/hyperledger/indy-sdk/blob/a21383cb960a008ce5ba976a2054c323696b5946/libindy/src/api/anoncreds.rs#L945 The param check in the indy_prover_create_credential_req api function is throwing invalid structure error

MathewJoseph (Tue, 03 Nov 2020 06:51:50 GMT):
I used the data i passed to the rust function and found this error *thread ‘main’ panicked at ‘called `Result::unwrap()` on an `Err` value: IndyError { error_code: CommonInvalidStructure, message: “Error: Invalid structure\n Caused by: UrsaCryptoError: Invalid Credential key correctness proof\n”, indy_backtrace: Some(“”) }’, src/issue-credential.rs:118:197 note: run*

MathewJoseph (Tue, 03 Nov 2020 06:53:05 GMT):
I deep dived into the rust code and found that *pub struct CredentialKeyCorrectnessProof { c: BigNumber, xz_cap: BigNumber, xr_cap: Vec<(String, BigNumber)>, }*

MathewJoseph (Tue, 03 Nov 2020 06:54:27 GMT):
The same cred offer is working fine in python

MathewJoseph (Tue, 03 Nov 2020 06:55:42 GMT):
"{"schema_id":"Th7MpTaRZVRYnPiabds81Y:2:gvt:1.0","cred_def_id":"EabLitju7VQTWq4wvMAXCN:3:CL:Th7MpTaRZVRYnPiabds81Y:2:gvt:1.0:TAG1","key_correctness_proof":{"c":"40685515695154179680883769139453278081987856777737920424998910544498716222582","xz_cap":"761950796450129830660034884686396544185774140916564349322608842194328090154127107434271258164460300363965615031182005136096780762668103512194930556286411024648272674466780996422156354311020664349804854517842957976621578530489275063044752172026031057692477973093566848427083378735950985892218851949164525941506407849622273458900430689583133148118345255576225799029051249158606880680383275101407092517783765318528539715522080374579427448627995755501931910699457130184541136929399565731053153097235045240063013589875860565066847685959138416393100208708449941452580681497902062709688787975979358367072742130796444245977387280499493384403736189741107266794014921082839526449138727896041936885268169","xr_cap":[["age","454911915797063572388147061966289926165510462404132938163182046712820267720561405974865845547314122708104375845420855123590909325761249167501733376846425176775577881739947662084409492654068455487597394058768840521934362223826461929791653623297181668290355472258985669970588097164249682669511327106627256793690288511485650704303756297200850597838453280474661070308920954535184816286285759769721930319172467966417775018017040534554730245400646896367401535653099474638491174753798957146246840967238729026312779407092801053150678054764146016578402078762250612395040322355100218618058942323505651603544571760943578304280406683804308466499336941270559291727348264559518971728661653899927956233508044"],["sex","140940371822738645180674207551267673834613723767666504025922287595181120210573506007078222646842690765534547450478810683625785435709102821226080415336434261180605169928939405747034146475248618527613458268513967193024920026737934376915934736193281590126412359298223542130510750543562229976561662355026094479627098202424411183464088805052837426196398963060874180552455033203006978998726366570958264421142666193456701612171708296124474919108829505108806880302310448280970601293189260854902825385602228557957124814755087833847735835996321992788878122305706158111749763991012776841624990568157970743371630170885668445123263860985392983153589181554832103444708333894325402571304367364541618347628897"],["height","289415187741673171682460711285965871166343099295573175019222813916532675434618331186338168643244728159363169848640719531935474442062630912777654056523263554752321455972340155631586477858647211511080108686613577078405537695307278666993224459942353979046429028697259556784109594421960530160064431579495866465360002917195760424022989530872209569962046419701435261157225245920434495860289317269035597925097234957351078570543629040803810975819600168614390599962926624616490666934916275690024286508827772635687241923502496433674260729411557054881057345430701038627629057082826327016884685210621340328815200697882692786942804919011076246006632849099737435827722648352300542033169251535898706516577148"],["master_secret","309503826043584880750886106014789731596212127298629611408133792294414702244393733586772289443061258553089796078662353583178706638458288477280114983101062723671399529208836613760049941742815242419406046171535723657096609896644143776964004125607973470395183305910515460221442102658347400533894452116400520536363150672413190252046896296022897581466824275527793945551991867700397178291498472384813750000913335385737641621137184079835017596696817832570707404217631128215999806882839642953314114691919194775822377258206510443100105534706423924957615378504571292011038759517719171212441368947561377943355266191763367816487619399911430041839499531450085089491366299056479947785697261696608087592972532"],["name","830153019745784290518852568736874877958064296421032969533343638618493454495536061582371488602834120939526389344818420550862217081503879095881786941587644515325334080236137842133879668343582050173775368707460724684651916455173353796122234046182397620833211415505324170504582973453023668491620277941573355657403959653818093509021818675867406086675411557167339985561545510650787519549251945835812520663938750077934977408439608073439460045489277636018101163430832895869899698359220542901008551025818237870755364400348707073428464808046115698917867134611286777084976674871758186590252876874468192165167642967430911233125745906378062394922645510277818457343703654545102116950580714763252890688004517"]]},"nonce":"757050520951517159804421"}"

MathewJoseph (Tue, 03 Nov 2020 06:57:11 GMT):
Can someone please give some pointers it would be very much helpful . Thank you

husnain (Tue, 03 Nov 2020 11:41:01 GMT):
Has joined the channel.

Lucampelli (Tue, 03 Nov 2020 12:13:22 GMT):
Hello! How can I reset my ledger to a newly created state? I'm using the indy sdk way of opening a docker network, and even if i delete the docker image, the credentials I created are still there!

Lucampelli (Tue, 03 Nov 2020 12:13:42 GMT):
Also I deleted the /tmp/indy/pool.txn

Lucampelli (Tue, 03 Nov 2020 12:13:46 GMT):
And nothing

Lucampelli (Tue, 03 Nov 2020 13:27:12 GMT):
Hello! If a Steward Creates a Trust anchor role for a did "a", and "a" creates a trust anchor did for "b". 1. Can steward revoke trust anchor for "a"? 2. Can steward revoke trust anchor for "b"? 3. If stewart revokes "a", does it also revokes "b"?

dcspinho (Tue, 03 Nov 2020 14:31:45 GMT):
Hello. When I do "ledger nym did=xxxx role=STEWARD verkey=yyyy" I get the error "Transaction author agreement Acceptance Mechanism isn't set. but I don't know how can I do it. Can you help me?

TimoGlastra (Tue, 03 Nov 2020 14:34:46 GMT):
Hi all, I have a question related to requesting proof for multiple credentials of the same type. E.g. lets say, as a holder, I have 10 vaccination credentials in my wallet. All from the same schema, but different credential_definitions. Each credential represents a single vaccination I received. When I go to my doctor I want to show proof of all received vaccinations. It seems like this is not possible with the current version of indy. Has anyone already put any thought into this scenario? Seems like a somewhat common use case, that I’m curious to see how it will work

daidoji (Tue, 03 Nov 2020 14:36:53 GMT):
you can show them one at a time? Do you want to combine proofs?

TimoGlastra (Tue, 03 Nov 2020 15:01:39 GMT):
One at a time would be possible, but this introduces some overhead and (I think) also eliminates the possibility to check all credentials are issued to the same holder (using link/master secret)

daidoji (Tue, 03 Nov 2020 15:06:26 GMT):
I agree on the overhead, on the second part I think you would just need to keep state. No reason that you can't check their provenance all originates with the same holder.

daidoji (Tue, 03 Nov 2020 15:06:47 GMT):
although maybe I'm misinformed about that. I am pretty new to this project

WadeBarnes (Tue, 03 Nov 2020 15:20:46 GMT):
You need to use the `--config` option when starting the `cli`: https://github.com/hyperledger/indy-sdk/blob/master/docs/configuration.md#options

WadeBarnes (Tue, 03 Nov 2020 15:21:18 GMT):
Content of you config file should be at least: ``` { "taaAcceptanceMechanism": "for_session" } ```

dcspinho (Tue, 03 Nov 2020 15:29:01 GMT):
Thanks. That worked. Now I have to create more TRUSTEE signatures. Error "Transaction has been rejected: not enough TRUSTEE signatures".

WadeBarnes (Tue, 03 Nov 2020 15:29:59 GMT):
Which network are you trying to run these commands on?

WadeBarnes (Tue, 03 Nov 2020 15:29:59 GMT):
Which network are you trying to run these transactions on?

dcspinho (Tue, 03 Nov 2020 15:31:44 GMT):
I dont' know if you are asking this: I using pool ledger I think

dcspinho (Tue, 03 Nov 2020 15:32:31 GMT):
I'm*

WadeBarnes (Tue, 03 Nov 2020 15:34:29 GMT):
When you run `pool connect` what pool are you specifying?

swcurran (Tue, 03 Nov 2020 17:46:17 GMT):
https://github.com/bcgov/von-network implements a full indy sandbox network and has a number of documents about working with the network. That's a place to start. If you have more specific questions, let us know.

dcspinho (Wed, 04 Nov 2020 00:43:09 GMT):
I'm following this https://github.com/hyperledger/indy-node/blob/master/acceptance/indy-cli-batches/expected/AS-03-01-trust-anchor.expected I specified AS-0301-endorser. Should I use predefined ones?

dcspinho (Wed, 04 Nov 2020 00:45:54 GMT):
Thanks. Can you tell me witch steps I have to take to create DID, DID document, schemas, etc and to authenticate an user?

dcspinho (Wed, 04 Nov 2020 00:46:32 GMT):
First I need to have the network to create nodes right? Then what should I do?

dcspinho (Wed, 04 Nov 2020 00:47:06 GMT):
which*

liwanghong (Wed, 04 Nov 2020 03:35:37 GMT):
Has joined the channel.

liwanghong (Wed, 04 Nov 2020 03:35:37 GMT):
I have one question about roles in indy-node

liwanghong (Wed, 04 Nov 2020 03:37:08 GMT):
One Trustee can change other trustee to steward or endorser

san1ker (Wed, 04 Nov 2020 06:48:05 GMT):
Hi team!

san1ker (Wed, 04 Nov 2020 06:53:09 GMT):
Hi team! While studying, I found writing credential definition takes more time and has larger time deviation than writing schema to the ledger. (Time dev of writing schema is very uniform) Can I ask why?? Thanks!!!

liwanghong (Wed, 04 Nov 2020 10:06:40 GMT):
I don't know the extractly flow of add new trustee, is it need approved by other trustees?

dgt1nsty (Wed, 04 Nov 2020 14:04:14 GMT):
is trustee and trust anchor are the same thing?

dgt1nsty (Wed, 04 Nov 2020 14:05:33 GMT):
if it is according to doc of sdk becoming a Trust Anchor requires contacting a person or organization who already has the Trust Anchor role on the ledger

swcurran (Wed, 04 Nov 2020 16:19:02 GMT):
A trustee is the highest level access on the system and are (usually) required for creating any elevated role DIDs, and for operations that affect the network -- e.g. initiating upgrades of the software run by the nodes. An Endorser (formerly Trust Anchor) can write or endorse the writing of "regular" DIDs (DIDs with no special privileges) and verifiable credential related objects (schema, cred def, rev reg and rev entries). An endorser may have to sign a legal agreement, and pay for any transaction that they endorse. Those are required by Sovrin. The roles are configurable, as are the number of signatures required.

crypto_beep (Wed, 04 Nov 2020 17:10:35 GMT):
Has joined the channel.

geonnave (Wed, 04 Nov 2020 19:05:22 GMT):
Hi, I was navigating the indy codebase, trying to understand how DID Documents are registered, but I was not able to identify where it creates the DDo Creation transaction specified in the Sovrin DID Method¹. Any ideas? Thanks. ¹ https://sovrin-foundation.github.io/sovrin/spec/did-method-spec-template.html#crud-operation-definitions

mccown (Wed, 04 Nov 2020 20:21:08 GMT):
Just a reminder about the Identity Implementer's WG Call tomorrow morning (Nov 5th @ 9am MT). In this call, we review updates from a variety of industry WGs that you may have missed this week. For our presentation, Paul Knowles will give a presentation & demo of the Overlays Data Capture Architecture (OCDA) Here are links to the Wiki and Zoom meeting: Wiki: https://wiki.hyperledger.org/display/IWG/2020-11-5+Identity+Implementers+WG+Call Zoom: https://zoom.us/my/hyperledger.community.3?pwd=UE90WHhEaHRqOGEyMkV3cldKa2d2dz09

dgt1nsty (Thu, 05 Nov 2020 08:38:34 GMT):
I see, thanks for clarification

sj1 4 (Thu, 05 Nov 2020 08:41:47 GMT):
hyperledger fabric

AlphaX (Thu, 05 Nov 2020 19:24:52 GMT):
hi everyone can anybody let me know that how can we analyze performance of Hyperledger indy based

AlphaX (Thu, 05 Nov 2020 19:25:28 GMT):
soluion

daidoji (Thu, 05 Nov 2020 19:26:52 GMT):
https://chat.hyperledger.org/channel/indy?msg=2dPGdExLAiq47Qydc

AlphaX (Thu, 05 Nov 2020 19:31:41 GMT):
can we integrate caliper with indy for bench marking

daidoji (Thu, 05 Nov 2020 19:38:04 GMT):
no clue, have to ask someone else I'm afraid.

AlphaX (Thu, 05 Nov 2020 19:39:46 GMT):
one man in #general told me that i have to wite a connector to integrate indy caliper

swcurran (Thu, 05 Nov 2020 20:57:41 GMT):
I don't know of anyone that has done anything with Caliper and Indy. There is some detail in the JIRA's listed above about how the test was done.

adityasanth (Fri, 06 Nov 2020 05:42:54 GMT):
Has joined the channel.

adityasanth (Fri, 06 Nov 2020 05:42:55 GMT):
Hi, I am building an identity solution for my school and we want it to be following a common standard that is universally verifiable. I have a few questions, will be great if I can get a response - 1. Does using Indy also mean I am compliant with w3c VC standards? 2. We are planning to go ahead with sov which provides me with a did of type sov, should I be looking at aries implementation? 3. When Indy too has a concept of channels authcrypt and decrypt etc. Is there a reason why I should look at aries? 4. Can I monitor the transactions of did type sov in the Indy scan net?

darkchylde (Mon, 09 Nov 2020 06:24:26 GMT):
Question, is it possible to query secrets stored in an indy wallet ?

darkchylde (Mon, 09 Nov 2020 06:24:55 GMT):
for example signing keys

askolesov (Mon, 09 Nov 2020 10:37:23 GMT):
Has joined the channel.

smithbk (Mon, 09 Nov 2020 13:16:35 GMT):
It is my understanding that this is intentionally NOT possible because it would prevent being able to implement the wallet using a backing HSM

shasnain (Mon, 09 Nov 2020 19:19:07 GMT):
Has joined the channel.

dcspinho (Tue, 10 Nov 2020 01:45:15 GMT):
Hello. I have the error "Transaction has been rejected: Not enough TRUSTEE signatures". Before I : 1. Connect to pool 2. Open wallet 3. Use TRUSTEE DID 4. Did "ledger nym did=... role=STEWARD verkey=..."

swcurran (Tue, 10 Nov 2020 02:35:12 GMT):
Don't have the instruction handy, but there is a command that dumps the config rules. Perhaps on the ledger you are using it takes more than one Trustee to add a Steward?

swcurran (Tue, 10 Nov 2020 02:35:48 GMT):
If not that, then I would check on whether the DID you are using is a Trustee DID.

paul.bastian (Tue, 10 Nov 2020 11:59:15 GMT):
ledger get-auth-rules

paul.bastian (Tue, 10 Nov 2020 11:59:37 GMT):
check the required signatures for the corresponding transaction

Arsh-Sandhu (Wed, 11 Nov 2020 14:32:02 GMT):
Has joined the channel.

gnarula (Thu, 12 Nov 2020 07:34:42 GMT):
I've a question around an Indy read transaction's state proof. The multi_signature timestamp refers to the timestamp the state was last updated at right? I wonder if the state is still updated and that timestamp changes irrespective of any transactions being posted on to the ledger every fixed time interval? A quick analysis suggests so but I wanted to be sure

gnarula (Thu, 12 Nov 2020 07:37:50 GMT):
and is the time interval 5 minutes at the moment?

kei32bit (Fri, 13 Nov 2020 01:12:35 GMT):
Has joined the channel.

gnarula (Fri, 13 Nov 2020 15:51:53 GMT):
nvm, in case anyone's looking for it, it's called STATE_FRESHNESS_UPDATE_INTERVAL

Shubham95 (Fri, 13 Nov 2020 17:04:34 GMT):
Can anyone provide indy-sdk build libraries for ubuntu 20.04

AlexanderHoughton (Fri, 13 Nov 2020 18:41:27 GMT):

Verifying_Problems__Verifying.txt

AlexanderHoughton (Fri, 13 Nov 2020 18:41:52 GMT):

Verifying_Problems__Verifying.txt

AlexanderHoughton (Fri, 13 Nov 2020 18:41:53 GMT):

Verifying_Problems__Node.js.txt

AlexanderHoughton (Fri, 13 Nov 2020 18:41:54 GMT):

Verifying_Problems__Issued.txt

AlexanderHoughton (Fri, 13 Nov 2020 18:42:20 GMT):
Hi all, I've been developing an Android Aries agent mplementation, along with a corresponding issuing/verifying agent server in Express.js and C++, on Windows. On the server side, try as I might, the C Indy function callbacks don't seem to be called, and I've been coding C/C++ for 24 years, so I'm at a bit of a loss as to why. As a temporary measure, I am talking to Indy by using ZeroMQ socket connections to the Indy Node.js wrapper and as insane as it sounds, this is working rather well. :) The Android app is working nicely, and works with pretty much all of the BC Gov. issuers and verifiers, such as the Unverified Person Issuer and the Health Gateway verifier. The server acts as a router when dealing with the BC sites and also works as an issuer for several test credentials. I am almost done adding credential verification functionality to the server and have made sure that the schema_id, cred_def_id etc. are correctly used as keys for the corresponding JSON array entries, where needed. Indy is no longer yelling at me about missing or malformed data, but it would appear that the signatures do not match as 'false' is returned from Indy.verifierVerifyProof. The encoded values in the issued credential match the values in the proof json (which I'm assuming is good), so I am rather confused as to what the problem is. I've attached the following: Verifying_Problems__Issued.txt - JSON console log of issued credential Verifying_Problems__Verifying.txt - JSON console log of data sent to JS call indy.verifierVerifyProof Verifying_Problems__Node.js - JS function call that calls Indy.verifierVerifyProof. I've been through the ACA Py source code, stared at anoncreds.rs (I don't know Rust), and generally tried to figure this out, but I am embarrassed to admit that I'm rather stuck. Please help, so that I can get this fixed and move on to implementing credential revocation. Many thanks in advance.

AlexanderHoughton (Fri, 13 Nov 2020 18:42:20 GMT):
Hi all, I've been developing an Android Aries agent mplementation, along with a corresponding issuing/verifying agent server in Express.js and C++, on Windows. On the server side, try as I might, the C Indy function callbacks don't seem to be called, and I've been coding C/C++ for 24 years, so I'm at a bit of a loss as to why. As a temporary measure, I am talking to Indy by using ZeroMQ socket connections to the Indy Node.js wrapper and as insane as it sounds, this is working rather well. :) The Android app is working nicely, and works with pretty much all of the BC Gov. issuers and verifiers, such as the Unverified Person Issuer and the Health Gateway verifier. The server acts as a router when dealing with the BC sites and also works as an issuer for several test credentials. I am almost done adding credential verification functionality to the server and have made sure that the schema_id, cred_def_id etc. are correctly used as keys for the corresponding JSON array entries, where needed. Indy is no longer yelling at me about missing or malformed data, but it would appear that the signatures do not match as 'false' is returned from Indy.verifierVerifyProof. The encoded values in the issued credential match the values in the proof json (which I'm assuming is good), so I am rather confused as to what the problem is. I've attached the following: Verifying_Problems__Issued.txt - JSON console log of issued credential Verifying_Problems__Verifying.txt - JSON console log of data sent to JS call indy.verifierVerifyProof Verifying_Problems__Node.js - JS function call that calls Indy.verifierVerifyProof. I've been through the ACA Py source code, stared at anoncreds.rs (I don't know Rust), and generally tried to figure this out, but I am embarrassed to admit that I'm rather stuck. Please help, so that I can get this fixed and move on to implementing credential revocation. Many thanks in advance.

AlexanderHoughton (Fri, 13 Nov 2020 18:42:20 GMT):
Hi all, I've been developing an Android Aries agent mplementation, along with a corresponding issuing/verifying agent server in Express.js and C++, on Windows. On the server side, try as I might, the C Indy function callbacks don't seem to be called, and I've been coding C/C++ for 24 years, so I'm at a bit of a loss as to why. As a temporary measure, I am talking to Indy by using ZeroMQ socket connections to the Indy Node.js wrapper and as insane as it sounds, this is working rather well. :) The Android app is working nicely, and works with pretty much all of the BC Gov. issuers and verifiers, such as the Unverified Person Issuer and the Health Gateway verifier. The server acts as a router when dealing with the BC sites and also works as an issuer for several test credentials. I am almost done adding credential verification functionality to the server and have made sure that the schema_id, cred_def_id etc. are correctly used as keys for the corresponding JSON array entries, where needed. Indy is no longer yelling at me about missing or malformed data, but it would appear that the signatures do not match as 'false' is always returned from Indy.verifierVerifyProof. The encoded values in the issued credential match the values in the proof json (which I'm assuming is good), so I am rather confused as to what the problem is. I've attached the following: Verifying_Problems__Issued.txt - JSON console log of issued credential Verifying_Problems__Verifying.txt - JSON console log of data sent to JS call indy.verifierVerifyProof Verifying_Problems__Node.js - JS function call that calls Indy.verifierVerifyProof. I've been through the ACA Py source code, stared at anoncreds.rs (I don't know Rust), and generally tried to figure this out, but I am embarrassed to admit that I'm rather stuck. Please help, so that I can get this fixed and move on to implementing credential revocation. Many thanks in advance.

AlexanderHoughton (Fri, 13 Nov 2020 18:42:20 GMT):
Hi all, I've been developing an Android Aries agent mplementation, along with a corresponding issuing/verifying agent server in Express.js and C++, on Windows. On the server side, try as I might, the C Indy function callbacks don't seem to be called, and I've been coding C/C++ for 24 years, so I'm at a bit of a loss as to why. As a temporary measure, I am talking to Indy by using ZeroMQ socket connections to the Indy Node.js wrapper and as insane as it sounds, this is working rather well. :) The Android app is working nicely, and works with pretty much all of the BC Gov. issuers and verifiers, such as the Unverified Person Issuer and the Health Gateway verifier. The server acts as a router when dealing with the BC sites and also works as an issuer for several test credentials. I am almost done adding credential verification functionality to the server and have made sure that the schema_id, cred_def_id etc. are correctly used as keys for the corresponding JSON array entries, where needed. Indy is no longer yelling at me about missing or malformed data, but it would appear that the signatures do not match as 'false' is always returned from Indy.verifierVerifyProof, no exception is thrown. The encoded values in the issued credential match the values in the proof json (which I'm assuming is good), so I am rather confused as to what the problem is. I've attached the following: Verifying_Problems__Issued.txt - JSON console log of issued credential Verifying_Problems__Verifying.txt - JSON console log of data sent to JS call indy.verifierVerifyProof Verifying_Problems__Node.js - JS function call that calls Indy.verifierVerifyProof. I've been through the ACA Py source code, stared at anoncreds.rs (I don't know Rust), and generally tried to figure this out, but I am embarrassed to admit that I'm rather stuck. Please help, so that I can get this fixed and move on to implementing credential revocation. Many thanks in advance.

AlexanderHoughton (Fri, 13 Nov 2020 18:42:20 GMT):
Hi all, I've been developing an Android Aries Java agent mplementation, along with a corresponding issuing/verifying agent server in Express.js and C++, on Windows. On the server side, try as I might, the C Indy function callbacks don't seem to be called, and I've been coding C/C++ for 24 years, so I'm at a bit of a loss as to why. As a temporary measure, I am talking to Indy by using ZeroMQ socket connections to the Indy Node.js wrapper and as insane as it sounds, this is working rather well. :) The Android app is working nicely, and works with pretty much all of the BC Gov. issuers and verifiers, such as the Unverified Person Issuer and the Health Gateway verifier. The server acts as a router when dealing with the BC sites and also works as an issuer for several test credentials. I am almost done adding credential verification functionality to the server and have made sure that the schema_id, cred_def_id etc. are correctly used as keys for the corresponding JSON array entries, where needed. Indy is no longer yelling at me about missing or malformed data, but it would appear that the signatures do not match as 'false' is always returned from Indy.verifierVerifyProof, no exception is thrown. The encoded values in the issued credential match the values in the proof json (which I'm assuming is good), so I am rather confused as to what the problem is. I've attached the following: Verifying_Problems__Issued.txt - JSON console log of issued credential Verifying_Problems__Verifying.txt - JSON console log of data sent to JS call indy.verifierVerifyProof Verifying_Problems__Node.js - JS function call that calls Indy.verifierVerifyProof. I've been through the ACA Py source code, stared at anoncreds.rs (I don't know Rust), and generally tried to figure this out, but I am embarrassed to admit that I'm rather stuck. Please help, so that I can get this fixed and move on to implementing credential revocation. Many thanks in advance.

AlexanderHoughton (Fri, 13 Nov 2020 18:42:20 GMT):
Hi all, I've been developing an Android Java Aries agent mplementation, along with a corresponding issuing/verifying agent server in Express.js and C++, on Windows. On the server side, try as I might, the C Indy function callbacks don't seem to be called, and I've been coding C/C++ for 24 years, so I'm at a bit of a loss as to why. As a temporary measure, I am talking to Indy by using ZeroMQ socket connections to the Indy Node.js wrapper and as insane as it sounds, this is working rather well. :) The Android app is working nicely, and works with pretty much all of the BC Gov. issuers and verifiers, such as the Unverified Person Issuer and the Health Gateway verifier. The server acts as a router when dealing with the BC sites and also works as an issuer for several test credentials. I am almost done adding credential verification functionality to the server and have made sure that the schema_id, cred_def_id etc. are correctly used as keys for the corresponding JSON array entries, where needed. Indy is no longer yelling at me about missing or malformed data, but it would appear that the signatures do not match as 'false' is always returned from Indy.verifierVerifyProof, no exception is thrown. The encoded values in the issued credential match the values in the proof json (which I'm assuming is good), so I am rather confused as to what the problem is. I've attached the following: Verifying_Problems__Issued.txt - JSON console log of issued credential Verifying_Problems__Verifying.txt - JSON console log of data sent to JS call indy.verifierVerifyProof Verifying_Problems__Node.js - JS function call that calls Indy.verifierVerifyProof. I've been through the ACA Py source code, stared at anoncreds.rs (I don't know Rust), and generally tried to figure this out, but I am embarrassed to admit that I'm rather stuck. Please help, so that I can get this fixed and move on to implementing credential revocation. Many thanks in advance.

AlexanderHoughton (Fri, 13 Nov 2020 18:42:28 GMT):

Verifying_Problems__Verifying.txt

AlexanderHoughton (Fri, 13 Nov 2020 18:42:30 GMT):

Verifying_Problems__Node.js.txt

AlexanderHoughton (Fri, 13 Nov 2020 18:42:31 GMT):

Verifying_Problems__Issued.txt

AlexanderHoughton (Fri, 13 Nov 2020 18:49:32 GMT):
Oh, I remember noticing that there was once a bug report for the Indy sdk where Indy.verifierVerifyProof was always returning true due to a casting issue (something about reading 4 bytes instead of 1 due to casting to an int or something). Could this be a similar issue, assuming it's not me simpley being thick? :P

AlexanderHoughton (Fri, 13 Nov 2020 18:49:32 GMT):
Oh, I remember noticing that there was once a bug report for the Indy sdk where Indy.verifierVerifyProof was always returning true due to a casting issue (something about reading 4 bytes instead of 1 due to casting to an int or something). Could this be a similar issue, assuming it's not me simply being thick? :P

Leila_M (Fri, 13 Nov 2020 19:53:16 GMT):
Has joined the channel.

Leila_M (Fri, 13 Nov 2020 19:53:16 GMT):
Hi, is there anyone who is expert in Sovrin and has time for a short interview for a master thesis regarding this subject and self-sovereign identities? I would appreciate it.

ianco (Mon, 16 Nov 2020 04:36:39 GMT):
I notice your proof request has no restictions: ```Proof Request: { name: "OnlineAccount", nonce: "298273534653100073159555", requested_attributes: { email: { name: "email" }, password: { name: "password" }, website: { name: "website" } }, version: "1.0" } ```

ianco (Mon, 16 Nov 2020 04:37:41 GMT):
Try adding `restrictions:[{"schema_name":"OnlineAccount"}]` (I don't recall theexact syntax offhand) or some other resrictions based on the schema or cred def you've defined

bstsc (Mon, 16 Nov 2020 08:02:26 GMT):
Has joined the channel.

bstsc (Mon, 16 Nov 2020 08:19:37 GMT):
Hi everyone. I am in trouble with how key rotation works on Indy. In key compromise case it is clear. However, when everything is valid, how does it work? If an issuer issues a credential and then rotates its keys before verification process, how can a verifier verify this credential?

daidoji (Mon, 16 Nov 2020 13:52:21 GMT):
I think the credentials issued with the compromised key can't be verified at that point.

AlexanderHoughton (Mon, 16 Nov 2020 14:49:39 GMT):
Thanks for the heads up! It turned out that the code for adding the restrictions in the request proof presentation message handler was commented out. I added a restriction by credential definition id: `proof_request_json: { "name":"proof_req_1", "nonce":"465299308595843519695681", "requested_attributes": { "email":{"name":"email", "restrictions":[{"cred_def_id":"V4SGRU86Z58d6TV7PBUe6f:3:CL:50645:tag1"}]}, "website":{"name":"website", "restrictions":[{"cred_def_id":"V4SGRU86Z58d6TV7PBUe6f:3:CL:50645:tag1"}]} }, "version":"1.0" }`

AlexanderHoughton (Mon, 16 Nov 2020 14:49:39 GMT):
Thanks for the heads up! It turned out that the code for adding the restrictions in the request proof presentation message handler was commented out. I added a restriction by credential definition id: ``` proof_request_json: { "name":"proof_req_1", "nonce":"465299308595843519695681", "requested_attributes": { "email":{"name":"email", "restrictions":[{"cred_def_id":"V4SGRU86Z58d6TV7PBUe6f:3:CL:50645:tag1"}]}, "website":{"name":"website", "restrictions":[{"cred_def_id":"V4SGRU86Z58d6TV7PBUe6f:3:CL:50645:tag1"}]} }, "version":"1.0" } ```

AlexanderHoughton (Mon, 16 Nov 2020 14:49:39 GMT):
Thanks for the heads up! It turned out that the code for adding the restrictions in the Android Java request proof presentation message handler was commented out. I added a restriction by credential definition id: ``` proof_request_json: { "name":"proof_req_1", "nonce":"465299308595843519695681", "requested_attributes": { "email":{"name":"email", "restrictions":[{"cred_def_id":"V4SGRU86Z58d6TV7PBUe6f:3:CL:50645:tag1"}]}, "website":{"name":"website", "restrictions":[{"cred_def_id":"V4SGRU86Z58d6TV7PBUe6f:3:CL:50645:tag1"}]} }, "version":"1.0" } ```

AlexanderHoughton (Mon, 16 Nov 2020 14:52:59 GMT):
Unfortunately, it didn't change the result, it still fails to verify. I solved the C++ callback problem and spent the weekend implementing the Indy calls so that my Agent server is now 100% C++. The result is the same, it stil l doesn't verify. Please see attached data, many thanks for looking at this.

AlexanderHoughton (Mon, 16 Nov 2020 14:52:59 GMT):
Unfortunately, it didn't change the result, it still fails to verify. I solved the C++ callback problem and spent the weekend implementing the Indy calls so that my Agent server is now 100% C++. The result is the same, it still doesn't verify. Please see attached data, many thanks for looking at this.

AlexanderHoughton (Mon, 16 Nov 2020 14:53:21 GMT):

cpp_server_indy_verifier_verify_proof.txt

AlexanderHoughton (Mon, 16 Nov 2020 15:16:27 GMT):
Ooops! Please ignore revoc_reg_defs and revoc_regs in the above txt file, I accidently pasted teh credential definitions again. revoc_reg_defs and revoc_regs are in fact empty.

AlexanderHoughton (Mon, 16 Nov 2020 15:17:59 GMT):

cpp_server_indy_verifier_verify_proof.txt

ianco (Mon, 16 Nov 2020 16:53:07 GMT):
Hmmm the only thing I notice is that the attribute names seem to be inconsistent (i.e. the names in the proof request, not the attribute names from the credential)

ianco (Mon, 16 Nov 2020 16:54:09 GMT):
i.e. in the initial request_presentation attachment you have `attr1_referent` an attr2_referent`, but then in the proof_request_json you name them `email` and `website`

ianco (Mon, 16 Nov 2020 16:54:09 GMT):
i.e. in the initial request_presentation attachment you have `attr1_referent` and `attr2_referent`, but then in the proof_request_json you name them `email` and `website`

ianco (Mon, 16 Nov 2020 16:54:51 GMT):
Maybe the verifier isn't recognizing the attributes that are being returned?

AlexanderHoughton (Mon, 16 Nov 2020 19:00:34 GMT):
I fixed this, but it made no difference, here's the updated proof request json:

AlexanderHoughton (Mon, 16 Nov 2020 19:00:38 GMT):
``` proof_request_data: { "name":"Covid19TestResults", "nonce":39592, "requested_attributes": { "collected_date":{"name":"collected_date", "restrictions":[{"schema_id":"V4SGRU86Z58d6TV7PBUe6f:2:Covid19TestResults:1.0"}]}, "facility":{"name":"facility", "restrictions":[{"schema_id":"V4SGRU86Z58d6TV7PBUe6f:2:Covid19TestResults:1.0"}]}, "organisms_isolated":{"name":"organisms_isolated", "restrictions":[{"schema_id":"V4SGRU86Z58d6TV7PBUe6f:2:Covid19TestResults:1.0"}]}, "result":{"name":"result", "restrictions":[{"schema_id":"V4SGRU86Z58d6TV7PBUe6f:2:Covid19TestResults:1.0"}]}, "source":{"name":"source", "restrictions":[{"schema_id":"V4SGRU86Z58d6TV7PBUe6f:2:Covid19TestResults:1.0"}]}, "specimen_source":{"name":"specimen_source", "restrictions":[{"schema_id":"V4SGRU86Z58d6TV7PBUe6f:2:Covid19TestResults:1.0"}]}, "status":{"name":"status", "restrictions":[{"schema_id":"V4SGRU86Z58d6TV7PBUe6f:2:Covid19TestResults:1.0"}]}, "test_name":{"name":"test_name", "restrictions":[{"schema_id":"V4SGRU86Z58d6TV7PBUe6f:2:Covid19TestResults:1.0"}]} }, "requested_predicates":{}, "version":"1.0" } ```

AlexanderHoughton (Mon, 16 Nov 2020 19:50:45 GMT):
I am now very embarrassed! I fixed this: ``` proof_request_data: { "name":"OnlineAccount2", "nonce":16660, "requested_attributes": { "email":{"name":"email", "restrictions":[{"schema_id":"V4SGRU86Z58d6TV7PBUe6f:2:OnlineAccount2:1.0"}]}, "website":{"name":"website", "restrictions":[{"schema_id":"V4SGRU86Z58d6TV7PBUe6f:2:OnlineAccount2:1.0"}]} }, "requested_predicates":{}, "version":"1.0" } ```

AlexanderHoughton (Mon, 16 Nov 2020 19:52:02 GMT):
...but it still fails to verify. I'll keep thrashing. away.

reviz0r (Mon, 16 Nov 2020 20:25:57 GMT):
Has joined the channel.

Lucampelli (Mon, 16 Nov 2020 20:28:59 GMT):
Hello! How can I reset my ledger to a newly created state? I'm using the indy sdk way of opening a docker network, and even if i delete the docker image, and the pool.txn file, the credentials I created are still there! Where in my computer is the ledger saved to?

Lucampelli (Mon, 16 Nov 2020 20:31:00 GMT):
Hello! If a Steward Creates a Trust anchor role for a did "a", and "a" creates a trust anchor did for "b". 1. Can steward revoke trust anchor for "a"? 2. Can steward revoke trust anchor for "b"? 3. If stewart revokes "a", does it also revokes "b"?

paul.bastian (Mon, 16 Nov 2020 21:53:10 GMT):
The credentials are signed by key material attributed to the credential definition. public did keys(NYM) are separate from those

paul.bastian (Tue, 17 Nov 2020 00:17:05 GMT):
Hello, I'm currently looking through Indy documentation and wondering what the status of this page is: https://github.com/hyperledger/indy-node/blob/master/docs/source/requests-new.md ? It's named "new", yet changes are 16 month ago(and therefore older than the requests page). Can anyone comment on the status of this?

swcurran (Tue, 17 Nov 2020 00:50:20 GMT):
Some central concepts about this. Authorizations on the ledger are configurable, so in theory, the answer to your questions may be affected by the configuration of the specific ledger. That said, in general the terms with the permissions are "Trustee", "Steward" and "Endorser" (formerly "Trust Anchor"). In general, Trustees will create all Stewards and Endorsers. Stewards will ONLY be able to do operations that affect their node of the network, and would NOT create Endorsers. Further, Endorsers can create DIDs for Authors (no ability to create other DIDs), but that's it. To your questions: 0. Change "Steward" to "Trustee" across the intro and the questions. 0. "a" is an Endorser and cannot create "b" another Endorser (caveat -- MAYBE the rules can be set up for that, but unlikely). 1. A Trustee may have authority to remove the "Endorser" privilege of "a" to just an Author. A Trustee cannot change the verkey of the Endorser, and cannot delete the DID altogether. 2. Since "b" can only be created by the Trustee (not "a" unless "a" is a Trustee), the Trustee can operate as per the paragraph above about "a". 3. There is no "delegation" flow involving the two DIDs -- they are independent of one another. Hope that helps.

swcurran (Tue, 17 Nov 2020 00:51:07 GMT):
I believe these are the accurate list of requests. They were "new" at the time :-). Doesn't make too much sense now.

AlexanderHoughton (Tue, 17 Nov 2020 02:03:15 GMT):
So, after 3 hour Zoom call with Emiliano (who I expect to be naming a child after), the issue has been solved.

AlexanderHoughton (Tue, 17 Nov 2020 02:03:15 GMT):
So, after a 3 hour Zoom call with Emiliano (who I expect to be naming a child after at some point), the issue has been solved.

AlexanderHoughton (Tue, 17 Nov 2020 02:04:48 GMT):
In case someone else gets stuck in the same in the same way I did, don't mix up yer key and names in the proof presentation like I did.

AlexanderHoughton (Tue, 17 Nov 2020 02:04:48 GMT):
In case someone else gets stuck in the same in the same way, don't mix up yer key and names in the proof presentation like I did.

AlexanderHoughton (Tue, 17 Nov 2020 02:04:52 GMT):

keys_and_names_mixup.png

AlexanderHoughton (Tue, 17 Nov 2020 02:05:33 GMT):
(The one on the left is me getting it wrong)

lynn.bendixsen (Tue, 17 Nov 2020 22:34:35 GMT):
Removing the /var/lib/indy//data directory on each node will reset the network to the genesis state. To do this, stop indy-node service on every node, remove the directory on every node, then start indy-node again on every node.

bstsc (Wed, 18 Nov 2020 06:12:57 GMT):
Thank you, but then, which keys are rotated periodically and which ones remain same? What about the master key?

mccown (Wed, 18 Nov 2020 20:27:14 GMT):
The Identity Implementer's WG is meeting tomorrow morning (Nov 19th @ 9am MT / 4pm UTC). In this call, we will review updates from several industry WGs, which you may have missed this week. For our presentation, Stephen Curran (BCGov) will present on how libindy is evolving into multiple shared libraries. Here are links to the Wiki and Zoom meeting: Wiki: https://wiki.hyperledger.org/display/IWG/2020-11-19+Identity+Implementers+WG+Call Zoom: https://zoom.us/my/hyperledger.community.3?pwd=UE90WHhEaHRqOGEyMkV3cldKa2d2dz09

rrishmawi (Thu, 19 Nov 2020 10:33:18 GMT):
Hello experts, I have a generic question, I am not sure if this is the right place to ask about it. In Verifiable Credentials, how a VC is correlated with an entity like a user? I read about Blinded Linked Secrets? but I could not find a lot of information about it anywhere? I am really surprised that such proving ownership is not talked about a lot in the documentation s and the W3C standards? Am I missing something? Does Hyperledger Indy have a different method for VC ownership? is there a documentation i am missing that explain those concepts?

ashishtripathi (Thu, 19 Nov 2020 12:50:43 GMT):
Has joined the channel.

paul.bastian (Thu, 19 Nov 2020 14:07:15 GMT):
NYM verkeys are easy to be rotated regularly and serve for connection establishment. I'm unsure what you mean with master key. The link secret?

paul.bastian (Thu, 19 Nov 2020 14:15:25 GMT):
Indy uses anoncreds. Credentials are not issued to a certain DID, but to the blinded link secret. the holder proves that 2 credentials belong to him by presenting the same signed link secret in both creds

paul.bastian (Thu, 19 Nov 2020 14:15:27 GMT):
https://forum.sovrin.org/t/proving-claim-ownership-without-sharing-did/860

bestape (Thu, 19 Nov 2020 23:34:31 GMT):
Has joined the channel.

rrishmawi (Fri, 20 Nov 2020 09:14:19 GMT):
Thank a lot Paul. I see now. But I have two questions: First, is this protocol fips compliant or has any sort of compliance or validation done to it? Second question which is not related: When an Issuer signs a Verifiable credentials (which include the blinded link secret), what DID public key is it signed with? is it the pairwise DID with the holder? or is it a special kind of DID that is published on the blockchain? In other words? does the Issuer use the same published DID for DID connections? or those are separate?

paul.bastian (Fri, 20 Nov 2020 15:10:30 GMT):
communication connection is done with DID public keys which translates to NYM verkeys, credential signing is done with credential definition CL keys

paul.bastian (Fri, 20 Nov 2020 15:11:39 GMT):
concerning FIPS I have no idea and you may ask as another question outside this thread to get more attention, there have been some reviews

alokrajiv (Sat, 21 Nov 2020 07:30:34 GMT):
Has joined the channel.

EtricKombat (Sat, 21 Nov 2020 12:55:47 GMT):
Hi Team, i have a question... any plans on Indy LTS version? Like fabric has.

esplinr (Sat, 21 Nov 2020 19:35:09 GMT):
@EtricKombat No. But we could make those plans if someone wants to volunteer to backport the fixes.

AlphaX (Sun, 22 Nov 2020 19:15:10 GMT):
What is the maximum number of transactions per second that can be written to the hyperledger indy?

pankajn (Mon, 23 Nov 2020 12:40:30 GMT):
credential offer

esplinr (Mon, 23 Nov 2020 15:11:58 GMT):
In 2019, we proved a sustained load of 10 writes per second on a 25 node cluster while serving thousands of reads per second.

bstsc (Mon, 23 Nov 2020 21:14:09 GMT):
Yes, which is used to produce NYMs.

pankajn (Tue, 24 Nov 2020 08:50:59 GMT):
In the getting started example (https://github.com/hyperledger/indy-sdk/blob/master/samples/python/src/getting_started.py) anoncreds.issuer_create_credential_offer (at line 274) returns a Credential Offer for Alice. This I believe is the connection invite. I am unable to link this code to the creation of pairwise DID (as I was expecting). Would really appreciate if someone could point out in the code where in the process of creating a connection invite do pairwise DID get created.

TimoGlastra (Tue, 24 Nov 2020 09:19:01 GMT):
Indy SDK doesn't know about the concept of connections. Hyperledger Indy concerns with the issuing, holding, proving and verifying of credentials. It can also create and manage keys for you, which can be used in connections. Hyperledger indy doesn't concern with exchange of credentials or DIDs. Connections are a concept from Hyperledger Aries, for which you can use one of the available aries frameworks (aries-cloudagent-python, aries-framework-dotnet, aries-framework-go, ...)

pankajn (Tue, 24 Nov 2020 09:20:22 GMT):
ok I see. Many thanks, much appreciated!

Lucampelli (Tue, 24 Nov 2020 12:57:29 GMT):
Thanks for the reply, but accessing the docker image via: "docker run --rm -it --entrypoint=/bin/bash indy_pool" the only folder inside /var/lib/indy is sandbox, and there is no data in it...

rrishmawi (Tue, 24 Nov 2020 13:59:58 GMT):
Hey Paul, i really thank you for your answers. I have another question if you can answer: How relying party (verifier) do trust some issuer in a DID worlds? I am familiar with the concept of governance trust frameworks and members directory, and that those might be hosted on a blockchain, but how technically a relying party can find the list of trusted issuers (anchors) and who signs them?

lynn.bendixsen (Tue, 24 Nov 2020 15:09:19 GMT):
Sorry @Lucampelli I don't know Docker well enough yet to answer that one.

WadeBarnes (Tue, 24 Nov 2020 15:13:14 GMT):
@Lucampelli, would https://github.com/bcgov/von-network work better for your purposes. It's easier to manage.

piyushmaheshwari65 (Wed, 25 Nov 2020 04:19:16 GMT):
Has joined the channel.

Shubham95 (Thu, 26 Nov 2020 10:17:18 GMT):
Hello everyone I am new to indy so was installing indy-sdk on ubuntu 20.04 but it doesn't have any official release, so can anyone please guide me how to install it.

TimoGlastra (Thu, 26 Nov 2020 11:57:20 GMT):
This docker file installs it for ubuntu 18 (https://github.com/hyperledger/aries-framework-javascript/blob/master/Dockerfile). I've changed 18 to 20 and libindy itself installs fine. I'm not sure which wrapper you intend to use but it get's stuck on installing the NodeJS wrapper because ubuntu doesn't ship python 2 anymore

Lucampelli (Thu, 26 Nov 2020 13:02:02 GMT):
Thaks for your help and suggestions, I will be looking into it.

Lucampelli (Thu, 26 Nov 2020 13:02:02 GMT):
Thanks for your help and suggestions, I will be looking into it.

IgnaceLoomans (Thu, 26 Nov 2020 13:20:56 GMT):
Has joined the channel.

IgnaceLoomans (Thu, 26 Nov 2020 13:20:56 GMT):
Hello does anybody here have any experience on creating a connection to the sovrin network, I am struggling on discovering how exactly to achieve this!

ozkul (Thu, 26 Nov 2020 17:14:58 GMT):
Has joined the channel.

ozkul (Thu, 26 Nov 2020 17:14:58 GMT):
Hello I want to upgrade indy pool and nodes .Do I have to add sha256 value for POOL_UPGRADE command

ozkul (Thu, 26 Nov 2020 17:15:15 GMT):
Can we run the upgrade command without sha256? Like this ; ledger pool-upgrade name=upgrade1121 version=1.12.1 action=start timeout=15 force=true

regy14 (Fri, 27 Nov 2020 13:02:56 GMT):
Has joined the channel.

tarun.sharma (Sat, 28 Nov 2020 02:51:56 GMT):
I am trying to run getting-started and no matter what I do, I am facing this error: The following packages have unmet dependencies: indy-plenum : Depends: python3-orderedset (= 2.0) but 2.0.3 is to be installed Depends: python3-psutil (= 5.4.3) but 5.6.6 is to be installed Depends: python3-psutil (= 5.4.3) but 5.6.6 is to be installed Depends: python3-pympler (= 0.5) but 0.8 is to be installed Can somebody help? I tried both on macOS and Ubuntu 16.04

saeveritt (Sat, 28 Nov 2020 09:48:01 GMT):
Has joined the channel.

sauveergoel (Sat, 28 Nov 2020 14:39:57 GMT):
I am facing this issue while adding a new node using indy-cli `Transaction has been rejected: validation error [ClientNodeOperationData]: invalid network ip address (0.0.0.0) (node_ip=0.0.0.0)`

ManfredM (Sun, 29 Nov 2020 13:18:05 GMT):
Has joined the channel.

ManfredM (Sun, 29 Nov 2020 13:23:08 GMT):
I have faced the same problem and changed the indy-pool.dockerfile in the ci directory as follows: 1. add aptitude and apt-utils to the install environment 2. change the versions of each module as follows: #ARG indy_plenum_ver=1.12.1~dev989 #ARG indy_node_ver=1.12.1~dev1172 ARG indy_plenum_ver=1.13.0~dev1220 ARG indy_node_ver=1.13.0~dev1220 ARG python3_indy_crypto_ver=0.4.5 ARG indy_crypto_ver=0.4.5 ARG python3_pyzmq_ver=18.1.0 3. change the installation command after the "ARG" section: RUN apt-get update -y && aptitude install -f -y \ python3-pyzmq=${python3_pyzmq_ver} \ indy-plenum=${indy_plenum_ver} \ indy-node=${indy_node_ver} \ python3-indy-crypto=${python3_indy_crypto_ver} \ libindy-crypto=${indy_crypto_ver} \ ursa \ python3-ursa \ vim With those changes, I was able to build the docker image. It also starts properly. However, I'm running into a different problem now, don't know if this is related to my changes. (Credential definition is panicking, no idea if this is caused by using different versions as in the original file, will have to figure this out).

ManfredM (Sun, 29 Nov 2020 13:43:03 GMT):
Ok, just tried with the following changes to the dockerfile (note that the indy_plenum and indy_node versions are as in the original version on github): ARG indy_plenum_ver=1.12.1~dev989 ARG indy_node_ver=1.12.1~dev1172 ARG python3_indy_crypto_ver=0.4.5 ARG indy_crypto_ver=0.4.5 ARG python3_pyzmq_ver=18.1.0 ARG python3_orderedset_ver=2.0 ARG python3_psutils_ver=5.4.3 ARG python3_pympler_ver=0.5 RUN apt-get update -y && aptitude install -f -y \ python3-pyzmq=${python3_pyzmq_ver} \ python3-orderedset=${python3_orderedset_ver} \ python3-psutil=${python3_psutils_ver} \ python3-pympler=${python3_pympler_ver} \ indy-plenum=${indy_plenum_ver} \ indy-node=${indy_node_ver} \ python3-indy-crypto=${python3_indy_crypto_ver} \ libindy-crypto=${indy_crypto_ver} \ ursa \ python3-ursa \ vim It builds, it runs, but I still have the panic error when creating cred defs (using Java wrapper).

ManfredM (Sun, 29 Nov 2020 13:47:32 GMT):
Hi Experts, has anyone been able to run the file found in https://github.com/hyperledger/indy-sdk/blob/v1.11.0/samples/java/src/main/java/Anoncreds.java without problems? I get a panic error message: "thread '' panicked at 'attempted to leave type `linked_hash_map::Node` uninitialized, which is invalid', /build/rust/src/rustc-1.48.0-src/library/core/src/mem/mod.rs:658:9" I have also got the full stack trace, if it helps. The panic occurs in line 53 of the file: AnoncredsResults.IssuerCreateAndStoreCredentialDefResult createCredDefResult = issuerCreateAndStoreCredentialDef(issuerWallet, issuerDid, schemaJson, credDefTag, null, credDefConfigJson).get(); Any hints what I could do about that? I'm running a local indy docker image, using a locally built libindy, debug version (from indy-sdk project). However, the same error also occurs when I use libindy-1.15.0-1 from Arch User Repos. (The stack trace says a little more with the locally built version, though, the AUR version is a release version without debug info).

msembinelli (Sun, 29 Nov 2020 21:29:45 GMT):
Has joined the channel.

msembinelli (Sun, 29 Nov 2020 21:29:45 GMT):
Hi All, I was wondering if anyone could point me to documentation around which port ranges/ports are used for indy nodes in the node pools in the various networks (builder, staging, mainnet). Based off the genesis files, we can see ports such as 9701, 9702 etc. But is there any explicit spec or documentation around this? This matters a lot for network firewalls and what traffic we should be allowing through. Thanks!

WadeBarnes (Sun, 29 Nov 2020 22:01:37 GMT):
9700-9799 is the complete range of possible ports.

WadeBarnes (Sun, 29 Nov 2020 22:06:42 GMT):
Can't put my finger in the exact documentation.

WadeBarnes (Sun, 29 Nov 2020 22:06:42 GMT):
Can't put my finger on the exact documentation.

msembinelli (Mon, 30 Nov 2020 00:05:43 GMT):
thats what I figured. Thanks!

kimura111 (Mon, 30 Nov 2020 02:19:19 GMT):
Has joined the channel.

MohitRakhade (Mon, 30 Nov 2020 02:44:16 GMT):
Has joined the channel.

tarun.sharma (Mon, 30 Nov 2020 14:16:59 GMT):
Thanks, I will try this now.

kaijuneer (Mon, 30 Nov 2020 16:49:14 GMT):
Has joined the channel.

kaijuneer (Mon, 30 Nov 2020 16:51:33 GMT):
Hello, I am a dapp developer on Ethereum but a completely newbie on Hyperledger. I am currently developing a dapp and planning to integrate Hyperledger Indy identity management system into my dapp. Just wondering if Indy is open to anyone to use, or is there some specific process that I have to go through before I could start using it?

daidoji (Mon, 30 Nov 2020 16:53:28 GMT):
indy is open to anyone to use but indy is the name of the technology, instances of indy (like the Sovrin chain `did:sov` method) each have their own protocol for integration if you want write access. Usually the public has read access.

brentzundel (Mon, 30 Nov 2020 17:27:24 GMT):
A schema is a small JSON object, whose size varies according to the number of attributes, each of which is a single string. Conversely, a cred def is a large JSON object, with a 256-bit integer for each attribute in the schema. a cred def for a schema with 2 attributes will be much smaller than a cred def for a schema with 20 attributes. Are you noticing the time difference between cred defs of similar size, or are the cred defs differently sized?

brentzundel (Mon, 30 Nov 2020 17:35:04 GMT):
The rules for what permissions are required to create DIDs of different roles are different for each ledger. On Sovrin main net, a Steward does not have permission to create an Endorser (Trust Anchor). An Endorser also does not have permission to create another Endorser.

GuilhermeFunchal (Mon, 30 Nov 2020 17:37:54 GMT):
Has joined the channel.

GuilhermeFunchal (Mon, 30 Nov 2020 17:37:55 GMT):
Hello, What blockchain network options could you use with Indy ?

MohitRakhade (Tue, 01 Dec 2020 04:06:20 GMT):
I am using this repo and started with indy : https://github.com/sovrin-foundation/indy-dev.git

MohitRakhade (Tue, 01 Dec 2020 04:06:54 GMT):
But after running `sudo make run`

MohitRakhade (Tue, 01 Dec 2020 04:07:25 GMT):
I am facing following error, please help me out. I just started with Indy

MohitRakhade (Tue, 01 Dec 2020 04:07:41 GMT):

$ sudo make start
./scripts/start.sh 
entering  /home/lenovo/Desktop/Indy/indy-dev directory
Unable to find image 'indy_dev_pool:latest' locally
docker: Error response from daemon: pull access denied for indy_dev_pool, repository does not exist or may require 'docker login': denied: requested access to the resource is denied.
See 'docker run --help'.
Unable to find image 'indy_dev:latest' locally
docker: Error response from daemon: pull access denied for indy_dev, repository does not exist or may require 'docker login': denied: requested access to the resource is denied.
See 'docker run --help'.
"docker stop" requires at least 1 argument.
See 'docker stop --help'.

MohitRakhade (Tue, 01 Dec 2020 04:08:22 GMT):
$ sudo make start ./scripts/start.sh entering /home/lenovo/Desktop/Indy/indy-dev directory Unable to find image 'indy_dev_pool:latest' locally docker: Error response from daemon: pull access denied for indy_dev_pool, repository does not exist or may require 'docker login': denied: requested access to the resource is denied. See 'docker run --help'. Unable to find image 'indy_dev:latest' locally docker: Error response from daemon: pull access denied for indy_dev, repository does not exist or may require 'docker login': denied: requested access to the resource is denied. See 'docker run --help'. "docker stop" requires at least 1 argument. See 'docker stop --help'.

MohitRakhade (Tue, 01 Dec 2020 04:08:59 GMT):
`$ sudo make start ./scripts/start.sh entering /home/lenovo/Desktop/Indy/indy-dev directory Unable to find image 'indy_dev_pool:latest' locally docker: Error response from daemon: pull access denied for indy_dev_pool, repository does not exist or may require 'docker login': denied: requested access to the resource is denied. See 'docker run --help'.`

KaushalIngle (Tue, 01 Dec 2020 06:59:17 GMT):
Has joined the channel.

KaushalIngle (Tue, 01 Dec 2020 07:02:53 GMT):
Hi Experts, Needed your opinion on Indyscan? It looks good, just wanted to confirm from you guys if its a stable and safe tool to use? This is the link :- https://github.com/Patrik-Stas/indyscan Any faced any issues with this? Thanks for you time.

KaushalIngle (Tue, 01 Dec 2020 07:02:53 GMT):
Hi Experts, Needed your opinion on Indyscan? It looks good, just wanted to confirm from you guys if its a stable and safe tool to use? This is the link :- https://github.com/Patrik-Stas/indyscan Has any of you faced any issues with this? Thanks for you time.

kaijuneer (Tue, 01 Dec 2020 11:30:57 GMT):
Hi guys, if I want to intergrate Hyperledger Indy to my Ethereum dapp, is that possible? Does anyone have any useful resources/links to learn how to use Hyperledger Indy? I've gone through the github but still find the learning entry barrier quite high

wip-abramson (Tue, 01 Dec 2020 11:57:09 GMT):
Hi @kaijuneer that is cool! I have thought about this a few times in the past. I would love to know your usecase, and happy to try help you get up to speed.

wip-abramson (Tue, 01 Dec 2020 11:57:18 GMT):
This edx course is a good place to start

wip-abramson (Tue, 01 Dec 2020 11:57:33 GMT):
Also this repo has some good code tutorials https://github.com/OpenMined/PyDentity

GuilhermeFunchal (Tue, 01 Dec 2020 11:58:53 GMT):
What options can I use for optional indy network ?

wip-abramson (Tue, 01 Dec 2020 12:02:48 GMT):
https://www.edx.org/course/becoming-a-hyperledger-aries-developer oops :)

kaijuneer (Tue, 01 Dec 2020 12:03:55 GMT):
Hi @wip-abramson thanks for the reply! My dapp is a ticketing system on Ethereum (so we could regulate the secondary market etc). I would like to integrate some sort of decentralised identity management use case to the dapp for events attendees, so I just come across Indy. But since I'm new to hyperledger everything seems quite foreign to me right now haha. So would like get started on learning indy and also understand whether it is possible to integrate a permissioned identity management into a dapp built on a permissionless network

kaijuneer (Tue, 01 Dec 2020 12:05:08 GMT):
That seems like good resources! I'll check them out, really appreciate your help

wip-abramson (Tue, 01 Dec 2020 14:18:24 GMT):
Well Indy is just the ledger - https://indyscan.io/home/SOVRIN_MAINNET you want to use Hyperledger Aries which is at the application layer. Allowing connect to other agents, issue, store and present credentials etc

PatrikStas (Tue, 01 Dec 2020 14:36:46 GMT):
Hi KaushalIngle, author here. It's running on indyscan.io prety stable against Sovrin networks, in past I've encountered issue where indysdk was reporting no new transactions on the ledger, even though there actually were. Perhaps it was ledger issue, or maybe something in indysdk, I don't know. It didn't happen for several weeks now. In general it's running with minimal devops maintenance on my side for couple months.

PatrikStas (Tue, 01 Dec 2020 14:37:51 GMT):
As for the stability of the code and APIs, it can change, there's features I would like to add and it might require changes how the data is stored.

PatrikStas (Tue, 01 Dec 2020 14:38:58 GMT):
That probably won't happen for at least couple months as I am right now overloaded with other business :-)

Lucampelli (Tue, 01 Dec 2020 14:41:14 GMT):
Well, thanks for your reply, but that raises two questions: First: In this case scenario you described, how are new Trust Anchors (Endorsers) created? Second: In my case scenario, in the test network where everything is allowed, how those 3 questions I asked work?

lynn.bendixsen (Tue, 01 Dec 2020 14:59:49 GMT):
Hi KaushalIngle, I mad a fork of Indyscan for the Indicio Networks and it is running great for me. indyscan.indiciotech.io To make a prototype and see it working on my own machine was a matter of a few hours as Patrik did a great job of making that quite simple to do. It took a bit more effort to get the different parts running as services on a public, AWS, Ubuntu 18.04 server and I am happy to share the notes I took while building that with anyone who is interested. (I am not an expert on Linux services, but it is all up and running for a few weeks now)

lynn.bendixsen (Tue, 01 Dec 2020 14:59:49 GMT):
Hi KaushalIngle, I made a fork of Indyscan for the Indicio Networks and it is running great for me. indyscan.indiciotech.io To make a prototype and see it working on my own machine was a matter of a few hours as Patrik did a great job of making that quite simple to do. It took a bit more effort to get the different parts running as services on a public, AWS, Ubuntu 18.04 server and I am happy to share the notes I took while building that with anyone who is interested. (I am not an expert on Linux services, but it is all up and running for a few weeks now)

seriousm 3 (Tue, 01 Dec 2020 15:50:01 GMT):
Has joined the channel.

kaijuneer (Tue, 01 Dec 2020 16:26:43 GMT):
I see, I am currently following the edx course you suggested on Aries. Perhaps will have more questions further on :)

hwrdtm (Tue, 01 Dec 2020 16:45:03 GMT):
Has joined the channel.

hwrdtm (Tue, 01 Dec 2020 16:45:03 GMT):
Hello everyone! I was wondering if there is a test Indy Tails Server spun up publicly for people to develop against? Maybe the BC Government has something similar? Any advice & pointers would be appreciated, thanks!

brentzundel (Tue, 01 Dec 2020 16:45:44 GMT):
On Sovrin main net, a Trustee must sign the transaction to create an Endorser. I'm not sure what the auth rules are for the other Sovrin networks. I do know that indy-cli has a command to retrive the auth rules ( I think it is: `get_auth_rules`)

WadeBarnes (Tue, 01 Dec 2020 16:54:03 GMT):
`dev`: https://tails-server-dev.pathfinder.gov.bc.ca `test`: https://tails-server-test.pathfinder.gov.bc.ca

hwrdtm (Tue, 01 Dec 2020 16:54:51 GMT):
Great, thanks! Would you know the difference between the 2 environments? Or which one should be used when?

WadeBarnes (Tue, 01 Dec 2020 16:55:24 GMT):
https://chat.hyperledger.org/channel/aries-cloudagent-python?msg=l2ql5LHcfeOLpNYyU

WadeBarnes (Tue, 01 Dec 2020 16:56:31 GMT):
There are `dev`, `test`, and `prod` instances. `dev` is subject to frequent change.

WadeBarnes (Tue, 01 Dec 2020 16:57:31 GMT):
`dev` to develop against the latest, `test` if you need something a bit more stable.

hwrdtm (Tue, 01 Dec 2020 16:58:00 GMT):
This is very helpful, thank you.

cmendoza (Tue, 01 Dec 2020 21:25:06 GMT):
Has joined the channel.

GuilhermeFunchal (Wed, 02 Dec 2020 00:03:40 GMT):
Does anyone know alternatives for using von-network with Indy?

dishan (Wed, 02 Dec 2020 01:13:32 GMT):
Has joined the channel.

KaushalIngle (Wed, 02 Dec 2020 05:52:08 GMT):
Hi PatrikStas, Thanks for the rundown, quick response and for creating the tool. I was on the lookout for something similar. I will go ahead with it and will be on the lookout for the ledger issue. Hi lynn.bendixsen, Thinking of making a fork and using it as per needed. Good to know that it can be done easily. I would be really grateful if you could share your notes. It will make my work that much easier. Thanks a lot.

KaushalIngle (Wed, 02 Dec 2020 05:52:08 GMT):
Hi PatrikStas, Thanks for the rundown, quick response and for creating the tool. I was on the lookout for something similar. I will go ahead with it and will be on the lookout for the ledger issue.

KaushalIngle (Wed, 02 Dec 2020 05:52:46 GMT):
Hi lynn.bendixsen, Thinking of making a fork and using it as per needed. Good to know that it can be done easily. I would be really grateful if you could share your notes. It will make my work that much easier. Thanks a lot.

lynn.bendixsen (Wed, 02 Dec 2020 14:29:21 GMT):
The Readme's that Patrik has are pretty good and they might work best for what you want to do, but here are the detailed notes I took: https://docs.google.com/document/d/1noHYmudrIyeqDh_Gitjo8LGXAC9duo7OpgYomvIGkoQ/edit?usp=sharing Look at Appendix A for how I set up our public IndyScan server for the Indicio networks.

SubhadeepBanerjee (Wed, 02 Dec 2020 14:57:35 GMT):
Has joined the channel.

SubhadeepBanerjee (Wed, 02 Dec 2020 14:57:50 GMT):
how to install indy-cli in ubuntu 20.04

SubhadeepBanerjee (Wed, 02 Dec 2020 14:57:59 GMT):
sudo add-apt-repository "deb https://repo.sovrin.org/sdk/deb xenial stable" sudo add-apt-repository "deb https://repo.sovrin.org/deb xenial stable"

SubhadeepBanerjee (Wed, 02 Dec 2020 14:58:41 GMT):
I can't do this because I use focal fossa

mccown (Thu, 03 Dec 2020 04:36:46 GMT):
The Identity Implementer's WG is meeting tomorrow morning (Dec 3rd @ 9am MT / 4pm UTC). In this call, we will review updates from several industry WGs, which you may have missed this week. For our presentation, Nicole Khor (Anonyome Labs; University of Queensland) will present a part of her ongoing research into integrating SSI technologies with existing browser-based authentication protocols (e.g., WebAuthn). She will talk about browser integration, protocol requirements, etc. Here are links to the Wiki and Zoom meeting: Wiki: https://wiki.hyperledger.org/display/IWG/2020-12-3+Identity+Implementers+WG+Call Zoom: https://zoom.us/my/hyperledger.community.3?pwd=UE90WHhEaHRqOGEyMkV3cldKa2d2dz09

KaushalIngle (Thu, 03 Dec 2020 05:07:52 GMT):
Thanks a lot for the notes lynn.bendixsen. Along with your notes and Patrik's readme, it will be a breeze for me. Thanks a lot again.

sauveergoel (Thu, 03 Dec 2020 08:15:22 GMT):
Can anyone help me with the running genesis file for staging and builder NET?

SubhadeepBanerjee (Thu, 03 Dec 2020 13:20:25 GMT):

Clipboard - December 3, 2020 6:49 PM

janpieterz (Thu, 03 Dec 2020 16:06:31 GMT):
Has joined the channel.

godofmyownreligion (Thu, 03 Dec 2020 16:10:58 GMT):
Has joined the channel.

lynn.bendixsen (Thu, 03 Dec 2020 16:19:35 GMT):
It looks like your nym is in your wallet, but has not yet been written to the BuilderNet ledger. https://selfserve.sovrin.org has a form you can fill in (ignore the token part) to get your nym on the ledger as an Endorser.

Alexandru-MihailAdam (Mon, 07 Dec 2020 07:31:02 GMT):
Has joined the channel.

Alexandru-MihailAdam (Mon, 07 Dec 2020 08:52:56 GMT):
Hello everyone. I couldn't find any demo / tutorial about creating a PoC app using Indy-SDK (Java wrapper) to connect to Sovrin network. For now, I don't want to deploy my own network. Can anyone help me with some resources / directions? Thx

swcurran (Mon, 07 Dec 2020 14:31:26 GMT):
You can get an Endorser DID on the Sovrin BuilderNet or StagingNet from this site -- https://selfserve.sovrin.org/ You can ignore the payment address pieces. For production use case with Sovrin MainNet, there is a formal process to get an Endorser DID and transaction fees associated with using it.

PascalHeitz (Wed, 09 Dec 2020 10:15:19 GMT):
Has joined the channel.

ArturPhilipp (Thu, 10 Dec 2020 12:24:01 GMT):
Hi everybody. Hope this question fits to this channel since it is about mobile mobile wallet apps: I found this Sovrin elaboration by @danielhardman regarding disaster recovery for mobile wallets (https://sovrin.org/wp-content/uploads/2019/03/What-if-someone-steals-my-phone-110319.pdf). A "device revocation feature" is mentioned, which I do not find in any other "not Sovrin" docs so far. Is this a (still existing) "Sovrin only" feature, or is there some equivalent in the general Hyperledger Indy space? And if there is not: When a wallet apps offer some kind of export / backup functionality, how can one make sure, that the backup is not imported as copy to a wallet app of a total different user (in order to "copy" identities)? I am pretty sure, that this is not the first time this kind of question was posted, but I would appreciate any hint very much...

Alexandru-MihailAdam (Thu, 10 Dec 2020 13:55:46 GMT):
I'm trying to write a schema on the Builder Network on Sovrin. I received a endorser DID, used it, but when i'm trying to submitRequest i get the following error: `{"op":"REQNACK","reason":"client request invalid: insufficient number of valid signatures, 1 is required but 0 valid and 1 invalid have been provided. The following signatures are invalid: did=KgHAvbiM4y6RHqJZHx2EGX, signature=2TEPnSQRUNmTjAj8zfQukTXqX7fZpQBKzyoAdAXqR5VJC1bG61vy6Zmy6RqzxX11VRLMtwQAh9JJyXwop9aD1zKH","reqId":1607608160786862225,"identifier":"KgHAvbiM4y6RHqJZHx2EGX"}`

Alexandru-MihailAdam (Thu, 10 Dec 2020 13:56:32 GMT):
I'm the Author of the transaction, also the endorser. Should I make another operations ?

Alexandru-MihailAdam (Thu, 10 Dec 2020 14:15:23 GMT):
PS: I checked my DID on indyscan.io as a NYM tx and it's published on the ledger

ArturPhilipp (Thu, 10 Dec 2020 14:30:06 GMT):
Are you using Indy CLI for the schema transaction? Are you connected to the correct pool (buildernet)?

Alexandru-MihailAdam (Fri, 11 Dec 2020 06:35:51 GMT):
No, I'm using Java SDK. I tried with Indy CLI and it works, but from SDK, it doesn't.

Alexandru-MihailAdam (Fri, 11 Dec 2020 07:19:38 GMT):
Forget it, I submitted wrong signed request.. beginner mistakes.

awais.ahmad (Fri, 11 Dec 2020 08:49:08 GMT):
Hi Guys, Just a quick question, is there an example of revocation implemented in aries ? How do i go about implementing revocation in aries ?

ArturPhilipp (Fri, 11 Dec 2020 09:34:31 GMT):
no worries :)

daidoji (Fri, 11 Dec 2020 14:07:14 GMT):
There's an #aries channel and one for each client too that you might have better luck if you're asking for specifics. That being said the aries-cloud-agent does have revocation example in their demo.

awais.ahmad (Fri, 11 Dec 2020 14:08:28 GMT):
@daidoji thanks! I will google what that one is.

awais.ahmad (Fri, 11 Dec 2020 14:09:22 GMT):
ah ok you mean python agent

daidoji (Fri, 11 Dec 2020 14:11:33 GMT):
well more like the REST agent haha, but yeah. acapy is the shorthand name.

awais.ahmad (Fri, 11 Dec 2020 14:14:03 GMT):
thanks. I appreciate it. I am trying to translate that agent to the dotnet framework and it doesnt seem to have an implementation of revocation.

swcurran (Fri, 11 Dec 2020 14:58:30 GMT):
Aries Cloud Agent Python has full support for revocation as an issuer/verifier, so you can use that. I'm not sure of the status in the other frameworks. The /demo directory has some example code showing revocation in action.

ianco (Fri, 11 Dec 2020 15:00:21 GMT):
(It's documented in the "AliceGetsAPhone" demo)

rileyphughes (Fri, 11 Dec 2020 15:01:28 GMT):
@awais.ahmad aries framework dotnet does support revocation.

Lucampelli (Mon, 14 Dec 2020 11:34:49 GMT):
Thank you, still, my original question remains. Is the trust anchor role revokable? And is it Transitive if so?

etaleo (Mon, 14 Dec 2020 13:46:13 GMT):
Is https://hyperledger-indy.readthedocs.io/projects/node/en/latest/ outdated? It refers to "© Copyright 2018, Hyperledger Revision ebf330f8." If yes, is it planned to update it?

etaleo (Mon, 14 Dec 2020 13:46:13 GMT):
Is https://hyperledger-indy.readthedocs.io/projects/node/en/latest/ outdated? It refers to "© Copyright 2018, Hyperledger Revision ebf330f8." If so, is it planned to update it?

swcurran (Mon, 14 Dec 2020 15:26:33 GMT):
indy continues to be maintained. We'll check on that string, as the last release was September, 2020.

brentzundel (Mon, 14 Dec 2020 15:35:27 GMT):
all roles are revokable, revocation is not transitive.

dcspinho (Tue, 15 Dec 2020 03:36:16 GMT):
Hello. I'm testing von network and I get "Error initializing pool ledger" on browser ledger. Can anyone help me?

swcurran (Tue, 15 Dec 2020 04:18:25 GMT):
Probably need to see the steps you are using (notably -- have you done the `build` before the `start`, and to see the logs after that. What is your environment? Docker on what platform?

dcspinho (Tue, 15 Dec 2020 12:46:54 GMT):

logsBuildStart.txt

dcspinho (Tue, 15 Dec 2020 12:48:31 GMT):
i'm using ubuntu 16.04.7 LT

deas (Tue, 15 Dec 2020 19:55:52 GMT):
Has joined the channel.

deas (Tue, 15 Dec 2020 19:55:52 GMT):
Hi all. I decided to try some of the indy tutorials directly after learning a bunch through #aries-cloudagent-python. I am trying the getting-started one (https://hyperledger-indy.readthedocs.io/projects/sdk/en/latest/docs/getting-started/run-getting-started.html) from a freshly cloned copy of `indy-sdk`, launching from the `indy-sdk/docs/getting-started/` directory. Bringing up the network via `docker-compose up` fails with the following error. Looks like version problems for `indy-plenum` dependencies. I haven't tried to fix this on my own yet - but thought this group might like to know since it's the most basic walkthrough piece a newcomer might encounter. ``` The following packages have unmet dependencies: indy-plenum : Depends: python3-orderedset (= 2.0) but 2.0.3 is to be installed Depends: python3-psutil (= 5.4.3) but 5.6.6 is to be installed Depends: python3-psutil (= 5.4.3) but 5.6.6 is to be installed Depends: python3-pympler (= 0.5) but 0.8 is to be installed E: Unable to correct problems, you have held broken packages. ERROR: Service 'indy_pool' failed to build : The command '/bin/sh -c apt-get update -y && apt-get install -y python3-pyzmq=${python3_pyzmq_ver} indy-plenum=${indy_plenum_ver} indy-node=${indy_node_ver} python3-indy-crypto=${python3_indy_crypto_ver} libindy-crypto=${indy_crypto_ver} vim' returned a non-zero code: 100 ```

deas (Tue, 15 Dec 2020 21:36:51 GMT):
Just realized this may have been a better fit for #indy-sdk and checked there. Sure enough - problem was reported and answer was found. See the thread here https://chat.hyperledger.org/channel/indy-sdk/thread/cWq34By6rBvyGkAiR?jump=q4KbJYrDTrPqeCmwf Basic gist: need to edit the `indy-sdk/ci/indy-pool.dockerfile` (to pin appropriate versions of the dependencies) and `indy-sdk/docs/getting-started/docker-compose.yml` (to get f-strings support on Python 3.6). Issue's open on github, so I trust this'll get included soon.

GuilhermeFunchal (Wed, 16 Dec 2020 18:13:16 GMT):
Hello! How can I add a new node to indy_pool_transactions ?

GuilhermeFunchal (Thu, 17 Dec 2020 19:43:36 GMT):
Hello, I need help to create and add new nodes in Indy network. I'm create for nodes with command :

GuilhermeFunchal (Thu, 17 Dec 2020 19:43:56 GMT):
generate_indy_pool_transactions --nodes 4 --clients 0 --nodeNum 1 --ips '192.168.2.2,192.168.2.3,192.168.2.4.192.168.2.5' --network=sandbox

GuilhermeFunchal (Thu, 17 Dec 2020 19:45:00 GMT):
How I can add another node ? How I can create the node with role Steward ?

marc0olo (Fri, 18 Dec 2020 11:31:35 GMT):
Has joined the channel.

marc0olo (Fri, 18 Dec 2020 11:35:30 GMT):
Hi, I haven't worked with Indy so far and I want to know how I can updated the DID document that that it follows this convention for the did-communication service: https://github.com/hyperledger/aries-rfcs/tree/master/features/0067-didcomm-diddoc-conventions#service-conventions What's the quickest and easiest way to achieve that? I read about "NYM" transactions so far. But not sure how to achieve what I want. Maybe someone can give me a quick hint. Thanks! :)

paul.bastian (Fri, 18 Dec 2020 12:55:26 GMT):
try using the indy-cli

paul.bastian (Fri, 18 Dec 2020 12:55:40 GMT):
thats easy way for testing out stuff in the beginning

paul.bastian (Fri, 18 Dec 2020 12:58:07 GMT):
Translation from Indy network data to DID doc is handled in the sovrin DID Method so far, and you can use a n ATTRIB for adding a service endpoint. However this is pretty outdated and there is currently a group working on a brand new Indy DID Method that will work for all indy networks

Shweta1 (Mon, 21 Dec 2020 14:51:51 GMT):
Has joined the channel.

Shweta1 (Mon, 21 Dec 2020 14:51:52 GMT):
Hi Team,

Shweta1 (Mon, 21 Dec 2020 14:58:57 GMT):
I am trying to create 3 indy node individual docker container and then create indy ledger browser server.For this I used ci/indy-pool.dockerfile and created three docker container. and then throgh this container I initialize and start node and then create genesis transaction.I followed follwing steps but getting poolledgertimeout error. docker build -f indy-node.dockerfile -t indy-node1 . docker run -itd -p 9701-9702:9701-9702 indy-node1 docker build -f indy-node.dockerfile -t indy-node2 . docker run -itd -p 9703-9704:9703-9704 indy-node2 Go to Node 1 docker container init_indy_keys --name Node1 start_indy_node Node1 127.0.0.1 9701 127.0.0.1 9702 & generate_indy_pool_transactions --nodes 2 --clients 3 --nodeNum 1 --ips '127.0.0.1,127.0.0.1' Go to Node 1 docker container init_indy_keys --name Node2 start_indy_node Node2 127.0.0.1 9703 127.0.0.1 9704 & generate_indy_pool_transactions --nodes 2 --clients 3 --nodeNum 2 --ips '127.0.0.1,127.0.0.1' up indy local ledger docker build -f indy-local-server.dockerfile -t indy-webserver . docker run -dti -v /opt/von-network:/von-network -p 9000:9000 indy-webserver go to webserver container GENESIS_FILE=../my_genesis.txn PORT=9000 python -m server.server but http://Ipaddress:9000 gives poolledgertimeout error.

Shweta1 (Tue, 22 Dec 2020 06:48:40 GMT):
getting this error File "/von-network/server/anchor.py", line 221, in _open_pool self._pool = await pool.open_pool_ledger(pool_name, json.dumps(pool_cfg)) File "/usr/local/lib/python3.6/dist-packages/indy/pool.py", line 88, in open_pool_ledger open_pool_ledger.cb) indy.error.PoolLedgerTimeout The above exception was the direct cause of the following exception: Traceback (most recent call last): File "/von-network/server/anchor.py", line 317, in open await self._open_pool() File "/von-network/server/anchor.py", line 223, in _open_pool raise AnchorException("Error opening pool ledger connection") from e server.anchor.AnchorException: Error opening pool ledger connection 2020-12-22 06:47:34,033|INFO|server.py|--- Trust anchor initialized ---

andrew.whitehead (Tue, 22 Dec 2020 06:54:22 GMT):
I think you need to substitute whatever docker's internal host IP address is, von-network uses this to find it: `docker run --rm --net=host eclipse/che-ip`

andrew.whitehead (Tue, 22 Dec 2020 06:55:04 GMT):
instead of 127.0.0.1

Shweta1 (Tue, 22 Dec 2020 07:23:28 GMT):
Thanks andrew,is this internal host Ip address should be mentioned in genesis file?

andrew.whitehead (Tue, 22 Dec 2020 07:25:43 GMT):
Yes it needs to be

andrew.whitehead (Tue, 22 Dec 2020 07:25:51 GMT):
Also I should mention that 2 nodes is not enough for consensus

Shweta1 (Tue, 22 Dec 2020 07:26:52 GMT):
Thanks a lot andrew,I am creating 3 nodes and let you know.

Shweta1 (Tue, 22 Dec 2020 07:51:47 GMT):

genesis.txt

Shweta1 (Tue, 22 Dec 2020 07:52:30 GMT):
I have attached my genesis file.But issue is not resolved.getting same error File "/von-network/server/anchor.py", line 221, in _open_pool self._pool = await pool.open_pool_ledger(pool_name, json.dumps(pool_cfg)) File "/usr/local/lib/python3.6/dist-packages/indy/pool.py", line 88, in open_pool_ledger open_pool_ledger.cb) indy.error.PoolLedgerTimeout The above exception was the direct cause of the following exception: Traceback (most recent call last): File "/von-network/server/anchor.py", line 317, in open await self._open_pool() File "/von-network/server/anchor.py", line 223, in _open_pool raise AnchorException("Error opening pool ledger connection") from e server.anchor.AnchorException: Error opening pool ledger connection

Shweta1 (Tue, 22 Dec 2020 10:18:04 GMT):
Please have a look into this issue?

rocky_2020 (Tue, 22 Dec 2020 11:06:52 GMT):
Has joined the channel.

andrew.whitehead (Tue, 22 Dec 2020 18:03:13 GMT):
At least 4 nodes are required. Docker networking can be complicated, I can't really diagnose that. Also you don't need to rebuild the docker image for each node

Shweta1 (Wed, 23 Dec 2020 12:49:58 GMT):
Thanks andrew,it's working with 3 nodes.I diagonose the problem.Actually I am starting node by providing IP 127.0.0.1 it should be 0.0.0.0 like this "start_indy_node Node1 0.0.0.0 9701 0.0.0.0 9702 &".YesI dont need to create separate image.Thanks a lot again

vijaypatel (Wed, 23 Dec 2020 23:17:06 GMT):
Has joined the channel.

vijaypatel (Wed, 23 Dec 2020 23:20:25 GMT):
I have question about the Indicio Network. 1. are they based on Hyperledger Indy? 2. How do I connect, and can I test on it? 3. can credentials issued from Sovrin net be verified with Indicio net and vice versa

vosrey (Sat, 26 Dec 2020 02:15:48 GMT):
Has joined the channel.

BrianRichter (Sat, 26 Dec 2020 08:26:24 GMT):
Has joined the channel.

EtricKombat (Sat, 26 Dec 2020 14:04:50 GMT):

1608991487267.aac

iLico (Sun, 27 Dec 2020 09:51:49 GMT):
Has joined the channel.

rileyphughes (Mon, 28 Dec 2020 16:42:31 GMT):
1. you should ask indicio directly for best answers :) 2. yes, it's HL indy based 3. you can connect via instructions on their website https://indicio.tech/indicio-testnet/ or, the easiest way is to sign up for a trinsic account and use a trinsic cloud agent (based on aries) http://trinsic.studio 4. credentials issued from one network need to be verified using that same network. however an agent that issues credentials on Sovrin could also verify credentials coming from Indicio or visa-versa. Hope that makes sense

ghabxph (Thu, 31 Dec 2020 16:46:22 GMT):
Has joined the channel.

ghabxph (Sat, 02 Jan 2021 08:11:24 GMT):
Hello Indy Team1

ghabxph (Sat, 02 Jan 2021 08:13:40 GMT):

Clipboard - January 2, 2021 4:13 PM

ghabxph (Sat, 02 Jan 2021 08:13:54 GMT):
Hello Indy Team! I am playing on indy-sdk and self-run indy node. I have a question ``` {"identifier":"Th7MpTaRZVRYnPiabds81Y","reason":"client request invalid: UnauthorizedClientRequest('Not enough TRUSTEE signatures',)","reqId":1609574476133587173,"op":"REJECT"} ``` How to add a trustee? No particular purpose. I am fully exploring the sdk and learn more facts as I go through

ghabxph (Sat, 02 Jan 2021 08:13:54 GMT):
Hello Indy Team! I am playing on indy-sdk and self-run indy node. I have a question ``` {"identifier":"Th7MpTaRZVRYnPiabds81Y","reason":"client request invalid: UnauthorizedClientRequest('Not enough TRUSTEE signatures',)","reqId":1609574476133587173,"op":"REJECT"} ``` How to add a trustee? I have no particular purpose or specific direction. I am just exploring the sdk and learn facts as I go through.

ghabxph (Sat, 02 Jan 2021 08:14:10 GMT):

Clipboard - January 2, 2021 4:13 PM

vosrey (Sat, 02 Jan 2021 23:40:25 GMT):
I'm following this instruction here to start TheOrgBook (https://github.com/bcgov/von-network/blob/master/docs/Writing%20Transactions%20to%20a%20Ledger%20for%20an%20Un-privileged%20Author.md#start-the-theorgbook) , but I'm running into this error: ` pull access denied for postgresql, repository does not exist or may require 'docker login': denied: requested access to the resource is denied ` anyone knows what's going on? and how can I fix this?

vosrey (Sat, 02 Jan 2021 23:40:25 GMT):
I'm following Becoming Aries Developer course, and I'm doing a lab and following this instruction here to start TheOrgBook (https://github.com/bcgov/von-network/blob/master/docs/Writing%20Transactions%20to%20a%20Ledger%20for%20an%20Un-privileged%20Author.md#start-the-theorgbook) , but I'm running into this error: ` pull access denied for postgresql, repository does not exist or may require 'docker login': denied: requested access to the resource is denied ` anyone knows what's going on? and how can I fix this?

ghabxph (Sun, 03 Jan 2021 04:39:03 GMT):
Hi. I just managed to run my own indy node using von-network: https://ledger-dev.ecred.tech/ Is this your goal?

ghabxph (Sun, 03 Jan 2021 04:39:03 GMT):
Hi. I just managed to run my own indy node using von-network: https://ledger-dev.ecred.tech/ Is your goal is to run an indy node using von-network? Then please try: ``` cd ./von-network ./manage build ./manage start ```

ghabxph (Sun, 03 Jan 2021 04:40:14 GMT):
``` cd ./von-network ./manage build ./manage start ``` Have you tried this?

ghabxph (Sun, 03 Jan 2021 04:40:14 GMT):
``` cd ./von-network ./manage build ./manage start ```

ghabxph (Sun, 03 Jan 2021 04:44:47 GMT):
> pull access denied for postgresql, repository does not exist or may require 'docker login': denied: requested access to the resource is denied The error implies that the image name is wrong. Find the Dockerfile and change postgresql to postgres. That error implies a typo... There's no postgresql in dockerhub, but there's postgres https://hub.docker.com/_/postgres

ghabxph (Sun, 03 Jan 2021 04:44:47 GMT):
> pull access denied for postgresql, repository does not exist or may require 'docker login': denied: requested access to the resource is denied The error implies that the docker image name is wrong. Find the Dockerfile and change postgresql to postgres. That error implies a typo... There's no postgresql in dockerhub, but there's postgres https://hub.docker.com/_/postgres

ghabxph (Sun, 03 Jan 2021 04:44:47 GMT):
> pull access denied for postgresql, repository does not exist or may require 'docker login': denied: requested access to the resource is denied The error implies that the docker image name is wrong. Find the Dockerfile and change postgresql to postgres. That error implies a typo... There's no postgresql in dockerhub, but there's postgres https://hub.docker.com/_/postgres But I do think that your goal is to run von-network, so if that's the case, then do what I suggest.

swcurran (Sun, 03 Jan 2021 16:45:09 GMT):
@WadeBarnes ^^^ Something to add to the list -- we need to eliminate TheOrgBook and the Issuer agent from the tutorial. Let's talk about what we can do instead.

WadeBarnes (Mon, 04 Jan 2021 13:51:19 GMT):
Looking

WadeBarnes (Mon, 04 Jan 2021 14:00:00 GMT):
@vosrey, It looks like you may have missed a step: ``` pushd TheOrgBook/docker ./manage build ``` The would have built the `postgresql` image the docker-compose file is looking for.

WadeBarnes (Mon, 04 Jan 2021 14:02:19 GMT):
https://github.com/bcgov/von-network/blob/master/docs/Writing%20Transactions%20to%20a%20Ledger%20for%20an%20Un-privileged%20Author.md#clone-the-repos-and-build https://github.com/bcgov/TheOrgBook/blob/master/docker/manage#L196 https://github.com/bcgov/TheOrgBook/blob/master/docker/docker-compose.yml#L185 https://github.com/bcgov/TheOrgBook/blob/master/docker/docker-compose.yml#L201

WadeBarnes (Mon, 04 Jan 2021 14:04:45 GMT):
@swcurran, We should update the instructions to use https://github.com/bcgov/aries-vcr

WadeBarnes (Mon, 04 Jan 2021 14:12:17 GMT):
https://github.com/bcgov/von-network/issues/135

pankajn (Mon, 04 Jan 2021 14:31:54 GMT):
Hello and a wish everyone a very happy new year 🙂 I am trying to query a saved schema from the ledger via the Java wrapper. I am using 1.15.0 version. The buildGetSchemaRequest method succeeds and when I call the parseGetSchemaResponse I get the following error, looking for some pointers on where to continue my investigation as when I look at the rust code it seems to be looking for the "op" as you can see below it is not in the JSON, is this a version mismatch between the sdk and the indy-node... or something else? Err(IndyError { inner: Error("missing field op", line: 0, column: 0) Structure doesn't correspond to type. Most probably not found Item not found on ledger })-------[Thread-0] DEBUG org.hyperledger.indy.sdk.LibIndy.native.ledger_command_executor - src/commands/ledger.rs:384 | ParseGetSchemaResponse command received [Thread-0] DEBUG org.hyperledger.indy.sdk.LibIndy.native.indy.commands.ledger - src/commands/ledger.rs:808 | parse_get_schema_response >>> get_schema_response: "{\"reqId\":1609608508230043543,\"identifier\":\"HFDD8BsSkNAPV16tNvcAnS\",\"operation\":{\"type\":\"107\",\"dest\":\"HFDD8BsSkNAPV16tNvcAnS\",\"data\":{\"name\":\"Test\",\"version\":\"1.0\"}},\"protocolVersion\":2}" [Thread-0] INFO org.hyperledger.indy.sdk.LibIndy.native.indy.services.ledger - src/services/ledger/mod.rs:458 | parse_response() => Err(IndyError { inner: Error("missing field op", line: 0, column: 0) Structure doesn't correspond to type. Most probably not found Item not found on ledger })

swcurran (Mon, 04 Jan 2021 15:23:58 GMT):
Thanks!

lynn.bendixsen (Mon, 04 Jan 2021 15:25:24 GMT):
Hello! To answer your specific question: To add a TRUSTEE you must use an existing TRUSTEE DID. Depending on how the Auth_rules are set up, you might need more than one TRUSTEE signature to add a new TRUSTEE. The screenshot you sent adds a little bit of confusion, though. A Trust Anchor is not the same as a TRUSTEE. The older term "Trust Anchor" has been replaced by "Transaction Endorser" (or in a lot of the doc's it's just referred to as "ENDORSER") Adding an ENDORSER DID to your test ledger also requires one or more TRUSTEE signatures.

lynn.bendixsen (Mon, 04 Jan 2021 15:40:41 GMT):
Hi @vijaypatel, I work for Indicio and manage their networks and Riley has answered correctly. I haven't tried connecting to the Indicio TestNet using a Trinsic agent yet, but I am happy to help if you have any other questions regarding the Indicio Networks.

vosrey (Mon, 04 Jan 2021 19:30:43 GMT):
Thanks everyone. I really did miss the step :)

ghabxph (Tue, 05 Jan 2021 00:05:44 GMT):
Hi Indy Team, I have a question: - 1. Do DID generated by libindy have a corresponding Did Document? Is there a way to see it via libindy? - 2. I assume that there's a corresponding private key with every verkey. Is there a way to see it via libindy? I am assuming that there's did doc from every DID that libindy generates because of this document: https://sovrin-foundation.github.io/sovrin/spec/did-method-spec-template.html And the peer did: https://identity.foundation/peer-did-method-spec/

ghabxph (Tue, 05 Jan 2021 00:13:51 GMT):
Hi @lynn.bendixsen thanks! I managed to figure out the DID of a trustee, and it's seed, thus I was able to add my very own trustee. For everyone, here's your clue: http://greenlight.bcovrin.vonx.io/browse/domain 1. Run a new node using von-network. 2. Browse localhost:9000/browse/domain and look for DID that is marked as trustee 3. Hunt the seed of that DID in google 4. build_nym_request --> role='TRUSTEE' and send via sign_and_submit_request

ghabxph (Tue, 05 Jan 2021 00:13:51 GMT):
Hi @lynn.bendixsen thanks! I managed to figure out the DID of a trustee, and it's seed, thus I was able to add my very own trustee. For everyone, here's your clue: http://greenlight.bcovrin.vonx.io/browse/domain 1. Run a new node using von-network. 2. Browse localhost:9000/browse/domain and look for DID that is marked as trustee 3. Hunt the seed of that DID in google 4. build_nym_request --> role='TRUSTEE' and send via sign_and_submit_request I recommend von-network as your indy node so that you'll have way to browse to indy blockchain. von-network is basically indy-node with UI. https://github.com/bcgov/von-network

ghabxph (Tue, 05 Jan 2021 00:14:14 GMT):

Clipboard - January 5, 2021 8:13 AM

ghabxph (Tue, 05 Jan 2021 00:16:07 GMT):
Simply put: - 1. I found this DID: V4SGRU86Z58d6TV7PBUe6f with a role of trustee. - 2. Thanks to google, I found it's seed: 000000000000000000000000Trustee1 - 3. Generate a DID using create_and_store_my_did and use the seed above - 4. Control your indy node!

swcurran (Tue, 05 Jan 2021 00:47:45 GMT):
For 1, the answer is yes and no :-). Within LibIndy, there are three elements associated with the DID -- the verkey (public key), the private key and the DID identifier. The latter is derived from the first verkey for a DID, so if you change the verkey and private key (e.g. rotate the keys), the DID identifier remains the same. A DIDDoc can be constructed using three elements, but there is literal document stored. Once you publish the NYM to an Indy Ledger, a resolver (such as the universal resolver) will retrieve those elements and return a formatted DIDDoc. If you also add an ATTRIB to the ledger with an "endpoint" as the data value, the Universal Resolver will include that as a fourth element of the DIDDoc. Likewise, an endpoint can be added to a Peer DID. 2. There is a private key. Libindy does not give you a way to get that, as it would be insecure. The less handling of the private key the better.

alvaradojl (Tue, 05 Jan 2021 08:01:08 GMT):
hello! I was wondering if someone of you folks could help me with this question? https://stackoverflow.com/questions/65574974/whats-the-method-in-libindy-to-create-a-credential-proposal

alvaradojl (Tue, 05 Jan 2021 08:01:47 GMT):
basically I cannot find the method in libindy to create a credential-proposal in the same way I found the create credential-offer for example

alvaradojl (Tue, 05 Jan 2021 08:02:31 GMT):
so I'm wondering what does it take to create this object, I'm using libindy and a python wrapper on top of it

alvaradojl (Tue, 05 Jan 2021 08:02:31 GMT):
so I'm wondering what does it take to create this object, I'm using libindy sdk and a python wrapper on top of it

alvaradojl (Tue, 05 Jan 2021 08:02:31 GMT):
so I'm wondering what does it take to create this object, I'm using libindy sdk and a custom python wrapper on top of it

ghabxph (Tue, 05 Jan 2021 08:11:49 GMT):
The idea is: 1. Create a schema 2. Create a credential definition **Create Schema** anoncreds.issuer_create_schema ledger.build_schema_request ledger.sign_and_submit_request **Create cred def** (cred_def_id, cred_def_json) = await anoncreds.issuer_create_and_store_credential_def Send the: cred_def_id and cred_def_json to the prover Here's your guide: https://github.com/hyperledger/indy-sdk/blob/master/docs/how-tos/issue-credential/python/issue_credential.py#L102

ghabxph (Tue, 05 Jan 2021 08:15:53 GMT):
Now, I haven't fully explored Aries yet, but I think the idea is to send that cred def id and cred def json to the prover to send the link secret and so on.

ghabxph (Tue, 05 Jan 2021 08:15:53 GMT):
Now, I haven't fully explored Aries yet, but I think the idea is to send that cred def id and cred def json to the prover by whatever means for prover to do further processing like sending the link secret.

alvaradojl (Tue, 05 Jan 2021 08:17:22 GMT):

Clipboard - January 5, 2021 9:17 AM

alvaradojl (Tue, 05 Jan 2021 08:17:24 GMT):
thanks @ghabxph yes all that works, but as in the following diagram

alvaradojl (Tue, 05 Jan 2021 08:18:06 GMT):
there is a step before you reply to credential-offer (which is coming from the issuer

alvaradojl (Tue, 05 Jan 2021 08:19:19 GMT):
so now i want to implement the flow, starting from the "holder" which is sending a credential proposal

alvaradojl (Tue, 05 Jan 2021 08:19:59 GMT):
they also mentioned it could be similar to have a credential preview, maybe you know how to create that credential preview?

daidoji (Tue, 05 Jan 2021 20:12:06 GMT):
Hello, in the Aries spec https://github.com/hyperledger/aries-rfcs/blob/master/concepts/0104-chained-credentials/README.md for chained credentials it says some work has already been done in Indy in regards to this work. Was hoping to implement this in acapy but would need someone to direct me along the way. I asked in the #aries-cloudagent-python but if anyone in here has information on what it would take to do this I'd love to get it.

proteche (Wed, 06 Jan 2021 07:20:16 GMT):
Has joined the channel.

proteche (Wed, 06 Jan 2021 07:20:17 GMT):
how to do backup of indy node data?

gagepoulson (Wed, 06 Jan 2021 15:34:34 GMT):
Has joined the channel.

Lucampelli (Fri, 08 Jan 2021 19:07:57 GMT):
Cool! Thanks!

Lucampelli (Mon, 11 Jan 2021 12:41:10 GMT):
Good Day! I would like to ask if Credentials ever make to the public Ledger, or are they only stored in the wallets of the one who created it and the one who received it?

Lucampelli (Mon, 11 Jan 2021 13:00:04 GMT):
Good Day! A question that popped up was: Considering one's wallet is stored in a physical medium in one's possession, be it a computer (with the Indy Client), a SmartCard, a SmartPhone whatever. What can be done to recover a wallet in case of device destruction? Would indy be working side by side with another Ledger in order to keep track of all credentials and DID's for this case?

daidoji (Mon, 11 Jan 2021 14:07:40 GMT):
@Lucampelli , I think Indy was designed so that credentials never go on the ledger. Just public dids, revocation registries, schema defs, credential defs

daidoji (Mon, 11 Jan 2021 14:08:13 GMT):
As to where they're stored, I think that's entirely up to the good faith of the issuer, verifier, and holder

swcurran (Mon, 11 Jan 2021 15:10:27 GMT):
As @daidoji noted, the credentials are issued to the holder and held in their secure storage. That storage can be backed up and restored as needed. That said, both the secure storage of the credentials, and the handling of the backups has to be done with care to prevent others from getting access to the private keys and the credentials. As noted, the credentials never go on a ledger -- the ledger holds identifiers and public keys of issuers and metadata about the credentials -- credential schema and revocation data.

lynn.bendixsen (Mon, 11 Jan 2021 16:50:30 GMT):
Which "data" are you referring to? Just the ledgers? What I usually recommend is to get a complete snapshot so that the node can be restored quickly to contribution level. The data for the ledgers themselves are restored automatically from other nodes within a short period of time of restarting, even if the ledgers need restored from scratch.

ianco (Mon, 11 Jan 2021 18:49:31 GMT):
Hi Indy folks, just want to remind everyone we have a release candidate for indy sdk (version is `1.16-rc-170`) available for testing. Release artifacts are all on https://repo.sovrin.org/ Please try to test by Tuesday Jan 19 so we can make a decision on promoting to an official `1.16` release

HarshMultani (Tue, 12 Jan 2021 19:12:14 GMT):
Has joined the channel.

coffeethecup (Wed, 13 Jan 2021 10:29:28 GMT):
Has joined the channel.

mccown (Wed, 13 Jan 2021 21:37:14 GMT):
The Identity Implementer's WG is meeting tomorrow morning (Jan 14th @ 9am MT / 4pm UTC). In this call, we will review updates from several industry WGs, which you may have missed this week. For our presentation, Patrik Stas (@PatrikStas) will provide an introduction to the Indyscan project and how it helps monitor the health and security of Indy-based ledgers. Here are links to the Wiki and Zoom meeting: Wiki: https://wiki.hyperledger.org/display/IWG/2021-01-14+Identity+Implementers+WG+Call Zoom: https://zoom.us/my/hyperledger.community.3?pwd=UE90WHhEaHRqOGEyMkV3cldKa2d2dz09

agrawalaman (Thu, 14 Jan 2021 06:48:50 GMT):
Has joined the channel.

IamRohitMittal (Sat, 16 Jan 2021 04:31:37 GMT):
Has joined the channel.

IamRohitMittal (Sat, 16 Jan 2021 04:31:38 GMT):
Hi Team

IamRohitMittal (Sat, 16 Jan 2021 04:32:47 GMT):
I am working on a identity management POC. Any demo repo will be appreciated, or documentation on same will be helping

mapa 5 (Sat, 16 Jan 2021 15:59:54 GMT):
https://www.epfl.ch/labs/dedis/wp-content/uploads/2020/09/report-2020-1-Alexandre-Deleze-hacking_sovereign_identity.pdf

IamRohitMittal (Mon, 18 Jan 2021 07:11:14 GMT):
Thanks mate :)

vardan10 (Tue, 19 Jan 2021 04:49:54 GMT):
I generated a image and ran a container using this docker file on a separate virtual machine: https://github.com/hyperledger/indy-sdk/blob/master/ci/indy-pool.dockerfile But I am unable to connect to the Indy pool from My PC. Should any ports be opened on the VM apart from the usual 9701-9708?

etaleo (Tue, 19 Jan 2021 11:56:12 GMT):
I've wanted to see how DID documents are stored on Indy (if at all), but it seems that there are stored some kind of credentials (?) For example, in the GET_ATTRIB example in https://github.com/hyperledger/indy-node/blob/master/docs/source/requests.md you can request the "dateOfBirth" attribute. Where is this dateOfBirth coming from? (Is it some part of a VC? I don't think so, as there is no VC specified, and in my understanding VC aren't stored on the ledger.) What is it exactly that is stored as "NYM" (and referenced by a DID) in Indy (in the context of SSI)?

enroll7 (Tue, 19 Jan 2021 12:07:07 GMT):
Has joined the channel.

etaleo (Tue, 19 Jan 2021 12:09:51 GMT):
So, as far as I understand, Indy does not follow the concept of DID documents and Verifiable Credentials (yet) as specified by W3, but instead has a different way of handling identites and related information to them, right? I'm still confused about where the similarites and differences are exactly.

TimoGlastra (Tue, 19 Jan 2021 12:19:36 GMT):
`dateOfBirth` is an example attribute. To my understanding an ATTRIB is just a key value storage tied to a did. If you for example look at the universal did resolver for `did:sov` it uses a get nym + get attrib to retrieve the did + endpoint (https://github.com/decentralized-identity/uni-resolver-driver-did-sov/blob/main/src/main/java/uniresolver/driver/did/sov/DidSovDriver.java#L148-L202). There is currently being worked on a `did:indy` method that does align with the DID spec, however I believe it will still use nym and attrib requests under the hood

etaleo (Tue, 19 Jan 2021 12:37:42 GMT):
Thank you. How is Indy able to resolve the dateOfBirth request, or in other words how does Indy know where the information is stored / how does Indy relate the information to the DID?

sauveergoel (Wed, 20 Jan 2021 19:08:39 GMT):
can anyone help on the below error: > thread '' panicked at 'called `Result::unwrap()` on an `Err` value: Error(Db(DbError { severity: "ERROR", parsed_severity: Some(Error), code: SqlState("42P01"), message: "relation \"metadata\" does not exist", detail: None, hint: None, position: Some(Normal(19)), where_: None, schema: None, table: None, column: None, datatype: None, constraint: None, file: Some("parse_relation.c"), line: Some(1180), routine: Some("parserOpenTable") }))', src/postgres_storage.rs:912:33 stack backtrace: 0: rust_begin_unwind at /rustc/e1884a8e3c3e813aada8254edfa120e85bf5ffca/library/std/src/panicking.rs:495:5 1: core::panicking::panic_fmt at /rustc/e1884a8e3c3e813aada8254edfa120e85bf5ffca/library/core/src/panicking.rs:92:14 2: core::option::expect_none_failed at /rustc/e1884a8e3c3e813aada8254edfa120e85bf5ffca/library/core/src/option.rs:1268:5 3: core::result::Result::unwrap at /rustc/e1884a8e3c3e813aada8254edfa120e85bf5ffca/library/core/src/result.rs:973:23 4: ::open_wallet at ./src/postgres_storage.rs:912:19 5: ::open_storage at ./src/postgres_storage.rs:2033:9 6: indystrgpostgres::PostgresWallet::open at ./src/lib.rs:220:29 7: 8: 9: 10: 11: 12: 13: 14: 15: 16: 17: start_thread 18: clone note: Some details are omitted, run with `RUST_BACKTRACE=full` for a verbose backtrace. Segmentation fault (core dumped)

mailgourav (Thu, 21 Jan 2021 05:00:31 GMT):
Has joined the channel.

mailgourav (Thu, 21 Jan 2021 05:00:32 GMT):
All, trying to do some performance testing credentials issuance with revocation registry size as 50K, It is taking approx 3 mins just to create the Credentials Definition and then 5 mins just to issue a single credential. So it is too slow. But with no revocation registry credential issuance is almost immediate. I am using aries-cloudagent:py36-1.14-1_0.5.1 and using 4 node Indy VON Network from https://github.com/bcgov/von-network. Wallet database being used here is SQLLite only. The CPU and memory utilization are all with in the limit, don't think it has any thing to do with infra. Any pointers on how to optimize this ?

x0axz (Thu, 21 Jan 2021 12:29:05 GMT):
Hi, I wanted to run multiple issuers on different servers. The first server is working fine, where the **IssuerDid** is `Th7MpTaRZVRYnPiabds81Y`. For second server, I change the **IssuerDid** to `Th7MpTaRZVRYnPiyxwv92Z`, and when I try to create a schema, it return this msg. ``` { success: false, error: "Ledger operation rejected" } ``` When I change `Th7MpTaRZVRYnPiyxwv92Z` in Genesis file, it returns this msg on Schema creation time. ``` { success: false, error: "The SDK library experienced an unexpected internal error." } ```

x0axz (Thu, 21 Jan 2021 12:29:05 GMT):
Hi, I wanted to run multiple issuers on different servers. The first server is working fine, where the **IssuerDid** is `Th7MpTaRZVRYnPiabds81Y`. For second server, I change the **IssuerDid** to `Th7MpTaRZVRYnPiyxwv92Z`, and when I try to create a schema, it return this msg. ``` { success: false, error: "Ledger operation rejected" } ``` When I change `Th7MpTaRZVRYnPiyxwv92Z` in Genesis file, it returns this msg on Schema creation time. ``` { success: false, error: "The SDK library experienced an unexpected internal error." } ```

x0axz (Thu, 21 Jan 2021 12:29:05 GMT):
Hi, I wanted to run multiple issuers on different servers. The first server is working fine, where the **IssuerDid** is `Th7MpTaRZVRYnPiabds81Y`. For second server, I changed the **IssuerDid** to `Th7MpTaRZVRYnPiyxwv92Z`, and when I try to create a schema, it return this msg. ``` { success: false, error: "Ledger operation rejected" } ``` When I change `Th7MpTaRZVRYnPiyxwv92Z` in Genesis file, it returns this msg on Schema creation time. ``` { success: false, error: "The SDK library experienced an unexpected internal error." } ```

timbu (Thu, 21 Jan 2021 14:04:56 GMT):
Has joined the channel.

daidoji (Thu, 21 Jan 2021 14:29:54 GMT):
Question about https://jira.hyperledger.org/browse/IS-1545 and https://jira.hyperledger.org/browse/INDY-2327 and a wider question that doesn't seem to have a clear answer. What exactly is the delta between W3C VC spec and Indy credentials? I know enough to know that they're different but don't know enough to know how to list out all those differences. Is it just this Rich Schema context that's missing or what? Is there a list somewhere or is this something I just have to crawl the specs for to determine myself?

daidoji (Thu, 21 Jan 2021 16:10:59 GMT):
The second looks like a bug but normally one doesn't need to mess with the genesis file I don't think so I'm not exactly sure what safeguards there are around that. Can both servers write to the same ledger that you're using??

x0axz (Thu, 21 Jan 2021 16:19:17 GMT):
no.. different servers and different ledger

daidoji (Thu, 21 Jan 2021 16:21:00 GMT):
ahh then the issue is that those public dids are bound to the ledger. If you create the did on one ledger and then try to use it on an issuer that's reading from the second ledger you're likely to get a failure.

swcurran (Thu, 21 Jan 2021 20:03:41 GMT):
Indy credentials are not W3C standard. The signature schema (CL Signatures) could be made W3C standard, but that has not been done, and there was some work on that with Rich Schema, but I believe that has been abandoned. Our plan (ACA-Py/BC Gov and others) is to instead move to BBS+ Signatures with JSON-LD to both achieve W3C Standard format with ZKP and Selective Disclosure support. To now, none of the W3C Standard formats or implementations (other than BBS+) achieves ZKP or selective disclosure, which we think is crucial. We continue to use Aries using Indy VCs for now because (a) the tech stack is rich and complete and interoperable based on Aries Interop 1.0, (b) there is a large community of collaborators and deployments, and (c) the Aries protocols give us a the protocols needed to exchange credentials and proof while evolving the VC payloads (e.g. from Indy Anoncreds to BBS+). Hope that helps.

daidoji (Thu, 21 Jan 2021 20:07:55 GMT):
@swcurran that helps a lot. Did the Aries-Go implementation take another route? Like where do Indy Anoncreds end and the Aries interop profile say begin?

swcurran (Thu, 21 Jan 2021 20:11:36 GMT):
AFGo skipped the use of Indy, anoncreds and the AIP 1.0 protocols. Until recently they did not support ZKP or selective disclosure, but recently added support for BBS+, which will provide those capabilities. On the down side, because they skipped Indy and AIP 1.0, we're not able to use AFGo in any use cases with ACA-Py or the Aries.Net framework. We expect that to change over the next few months, but it's going to be a bit yet.

daidoji (Thu, 21 Jan 2021 20:51:10 GMT):
@swcurran thanks again. Here's another question that might help clear up the last bit of misunderstanding I have. Lets say there's another credential format tomorrow. "ieee vc" or something like that. Would the AIP be able to support that too? Like are the credential formats (and crypto primitives) tightly bound to the protocols?

swcurran (Thu, 21 Jan 2021 20:56:03 GMT):
The goal of Aries 2.0 is to make it easy to extend. The credential exchange protocols define the interactions, and leave the payloads open, so that handlers for different payloads (formats) can be added, and the protocols still work. Of course, a specific deployment would have to have support for the format and it's underlying crypto (no magic :-) ), but it can be plugged in, and the protocols allow negotiation about what format to use -- e.g. payloads that both support. So an offer might say -- "I have a credential to give you and I support formats anoncreds, JWTs and ieee vc -- what format do you want?"

daidoji (Thu, 21 Jan 2021 21:05:37 GMT):
okay, I think that makes more sense.

daidoji (Thu, 21 Jan 2021 21:06:13 GMT):
right now though acapy could only say "I support anoncreds" and aries-go can only say "I support w3c vc jwts" right?

wip-abramson (Fri, 22 Jan 2021 10:14:14 GMT):
This is an interesting thread. I am wondering what is the point of supporting the VC Data Model for the credential side with signature schemes like CL and BBS+. The whole point is creating a fresh presentation each time so no one needs to interoperate with my VC. I see the value in moving towards a standard presentation format. Maybe I am missing something though, is the idea that it makes it easier to move wallets?

x0axz (Fri, 22 Jan 2021 10:31:05 GMT):
I have multiple issuers on different server, and I have Mobile App as holder, so I need to run mediator on every issuer's server? as I did for pool (add genesis file for every issuer).

x0axz (Fri, 22 Jan 2021 10:31:05 GMT):
I have multiple issuers on different servers, and I have Mobile App as holder. If Mobile has to talk to of these servers, do I need to run mediator on every issuer's server? and then the wallet will be created on every issuer server? as it does for single issuer.

x0axz (Fri, 22 Jan 2021 10:31:08 GMT):
I have multiple issuers on different server, and I have Mobile App as holder, so I need to run mediator on every issuer's server? as I did for pool (add genesis file for every issuer).

Toktar (Fri, 22 Jan 2021 13:34:17 GMT):
Hello all, Thanks to the `test_new_view_combinations` test, we found an issue in the View Change consensus node. This is a medium risk and does not entail irreparable consequences for the pool. However, we should remember that this problem exists. See more details in the issue https://github.com/hyperledger/indy-plenum/issues/1506

mailgourav (Fri, 22 Jan 2021 14:05:30 GMT):
Got it working with the latest Aries release, I see the registry size is restricted to 1000 in the latest version. Thanks

daidoji (Fri, 22 Jan 2021 14:25:02 GMT):
I think the idea of the VC data model is to not be locked down to one particular implementation (or standard spec).

daidoji (Fri, 22 Jan 2021 14:25:02 GMT):
I think the ideal of the VC data model is to not be locked down to one particular implementation (or standard spec).

daidoji (Fri, 22 Jan 2021 14:26:00 GMT):
Like in an ideal future I can take any credential in the wild, it'll be spec'd to VC data model, a few standard crypto primitives, and I'll be able to use those primitives and the data model without caring how its generated

daidoji (Fri, 22 Jan 2021 14:33:51 GMT):
I mean it depends on your use case, but in general all the agents mediator or not, would need to be able to read from the same ledger (and write if they're establishing new credentials or public dids)

swcurran (Fri, 22 Jan 2021 15:01:31 GMT):
Agree with the ideal, but the phrase "a few standard crypto primitives" is where it gets interesting. That has to be balanced with a desire to have a VC with attributes like ZKP and selective disclosure. TLS-style crypto doesn't get you those characteristics, but it is achievable with crypto that has been around since the early 2000s and is used elsewhere. Where do you draw the line?

swcurran (Fri, 22 Jan 2021 15:01:31 GMT):
Agree with the ideal, but the phrase "a few standard crypto primitives" is where it gets interesting. That has to be balanced with a desire to have a VC with attributes like ZKP and selective disclosure. TLS-style crypto doesn't get you those characteristics. They are achievable with crypto that has been around since the early 2000s and is used elsewhere. Where do you draw the line?

daidoji (Fri, 22 Jan 2021 15:14:45 GMT):
a question for wiser and more knowledgeable minds than I, I'm afraid.

daidoji (Fri, 22 Jan 2021 15:14:50 GMT):
:-)

wip-abramson (Fri, 22 Jan 2021 15:20:06 GMT):
Yeah you can only take the credentials with advanced crypto is you "understand" how to. Very different from just storing essentially a document with a signature against it. So my question is why go to the effort of conforming to the data model for the credential object. Whats the value add?

swcurran (Fri, 22 Jan 2021 15:27:17 GMT):
Wow, Will, I'm impressed. A question like that on a Friday afternoon. :-) It is a good question. I think the drivers of the standard started with the ideal above, decided they didn't need the "fancy" stuff and went ahead. That said, BBS+ Signatures does allow sticking to the format while achieving the desirable characteristics. But at the cost that the parties must support that crypto. And on it goes...

wip-abramson (Fri, 22 Jan 2021 15:49:00 GMT):
Hahahaha yeah it certainly feels like the VC Data model was never designed with this crypto in mind. I guess standards are always a compromise. I would love to follow along with the work going into BBS+ in aca-py. Did you select someone to work on this contract yet?

zossimov (Wed, 27 Jan 2021 10:03:49 GMT):
Hi, I'm trying to **Write DID on Ledger** by following this: https://github.com/hyperledger/indy-sdk/blob/master/docs/how-tos/write-did-and-query-verkey/cs/WriteDIDAndQueryVerkey.cs On Step 2, the Pool is created successfully, but at the time of opening pool ledger, it returns this error. `IndyException: An unmapped error with the code '307' was returned by the SDK.`

x0axz (Wed, 27 Jan 2021 18:10:32 GMT):
Hi all, a Pool with Steward Role is running on AWS server controlled by me. Now, I wanted to add multiple Trust Anchors, which are running on different servers. I did one experiment, I used the genesis file form AWS Server in Trust Anchor server (which has 4 nodes with Steward Server IP), and didn't start docker and was able to Create Schema & Credential Definition. Now, I've some questions. 1) Was that a right approach? if not, then what's the right way to create network like this? 2) Is there any need to run a docker on Trust anchor (I didn't run earlier)? 3) To add the Trust Anchor, have to add node_ip of the Trust Anchor's server in genesis file on Steward server? 4) How to add and remove the Trust Anchor in network? 5) How to generate DID for Trust Anchor? 6) If network is running, and I have add a new Trust Anchor, how to add it? Is there anything else I should know about?

x0axz (Wed, 27 Jan 2021 18:10:32 GMT):
Hi all, a Pool with Steward Role is running on AWS server controlled by me. Now, I wanted to add multiple Trust Anchors, which are running on different servers. I did one experiment, I used the genesis file form AWS Server in Trust Anchor server (which has 4 nodes with Steward Server IP), and didn't start docker and was able to Create Schema & Credential Definition. Now, I've some questions. 1) Was that a right approach? if not, then what's the right way to create network like this? 2) Is there any need to run a docker on Trust anchor (I didn't run earlier)? 3) To add the Trust Anchor, have to add node_ip of the Trust Anchor's server in genesis file on Steward server? 4) How to add and remove the Trust Anchor in network? 5) How to generate DID for Trust Anchor? 6) If network is running, and I have to add a new Trust Anchor, how to add it? Is there anything else I should know about?

x0axz (Wed, 27 Jan 2021 18:10:32 GMT):
Hi all, a Pool with Steward Role is running on AWS server controlled by me. Now, I wanted to add multiple Trust Anchors, which are running on different servers. I did one experiment, I used the genesis file form AWS Server in Trust Anchor server (which has 4 nodes with Steward Server IP), and didn't start docker and was able to Create Schema & Credential Definition. Now, I've some questions. 1) Is there any need to run a docker on Trust anchor (I didn't run earlier)? 3) To add the Trust Anchor, have to add node_ip of the Trust Anchor's server in genesis file on Steward server? 4) How to add and remove the Trust Anchor in network? 5) 6) If network is running, and I have to add a new Trust Anchor, how to add it? Is there anything else I should know about?

x0axz (Wed, 27 Jan 2021 18:10:32 GMT):
Hi all, a Pool with Steward Role is running on AWS server controlled by me. Now, I wanted to add multiple Trust Anchors, which are running on different servers. I did one experiment, I used the genesis file form AWS Server in Trust Anchor server (which has 4 nodes with Steward Server IP), and didn't start docker and was able to Create Schema & Credential Definition. Now, I've some questions. 1) Is there any need to run a docker on Trust anchor (I didn't run earlier)? 2) How to generate DID for Trust Anchor? 3) How to add and remove the Trust Anchor in network? 4) If network is running, and I have to add a new Trust Anchor, how to add it?

mwherman2000 (Thu, 28 Jan 2021 15:54:17 GMT):
Are there still Thursday Indy working group calls?

esplinr (Fri, 29 Jan 2021 21:56:42 GMT):
https://wiki.hyperledger.org/display/indy/Indy+Contributors+Meeting

esplinr (Fri, 29 Jan 2021 21:56:50 GMT):
Next Tuesday is the next one.

esplinr (Fri, 29 Jan 2021 21:57:43 GMT):
You might also be thinking of the Identity Implementors calls: https://wiki.hyperledger.org/display/IWG/Identity+Implementers+WG+Call I think the next call is in two weeks.

mwherman2000 (Sat, 30 Jan 2021 16:02:12 GMT):
Thk u ...looks like a lot of the calls have changed relative to about a year ago: https://wiki.hyperledger.org/display/HYP/Calendar+of+Public+Meetings Thk u @danielhardman for the link,

mwherman2000 (Sun, 31 Jan 2021 00:03:27 GMT):
Hi all, I’ve finished what I believe is an interesting document describing an effective solution for the SSI Personal Data Usage Licensing (SSI-PDUL) problem and would appreciate your comments and feedback. You can edit a Google Docs copy of my Word master version of the whitepaper here: https://docs.google.com/document/d/1yP0OgDhq-3TTIien5gsETi4jZdyiKbwE/edit If you prefer to read a PDF version, you can download it from here: https://hyperonomy.com/2021/01/27/self-sovereign-identity-personal-data-usage-licensing-ssi-pdul-model-solution-concept-review-draft-0-27/ One, what I believe to be an authoritative, comment that I’ve received so far is: Otherwise, I think it's a pretty unstudied problem. Abstract The purpose of this solution concept whitepaper is to provide the first complete description of the motivations, key concepts, problem statement, and solution approach for implementing the Self-Sovereign Identity Personal Data Usage Licensing (SSI-PDUL) Model, a functional component of the Self-Sovereign Identity Model (SSI Model). This solution concept addresses the following user scenario: How Alice User, an App User and Identity Owner, and Bob Developer, an App Developer and App Controller, might negotiate the use of Alice’s personal digital identifiers and any associated personal identity data by Bob’s app, based on Self-Sovereign Identity Model Usage Principles. The scope of the Self-Sovereign Identity Personal Data Usage Licensing (SSI-PDUL) Model is personal digital identifiers and any associated identity data presented by Alice to the App. It does not include the permissioning of data internal to the App (although the natural extension of the solution to internal data is an obvious one). The intended audience for this whitepaper is a broad range of professionals interested in furthering the application and use of the SSI Model in software apps, agents, and services using an SSI Personal Data Usage Licensing (SSI-PDUL) Model approach. This includes software architects, application developers, and user experience (UX) specialists; as well as people involved in a broad range of standards efforts related to decentralized identity, verified credentials, and secure storage. The work documented here was performed under the auspices of the Trusted Digital Web project in the Hyperonomy Digital Identity Lab of Parallelspace Corporation. Thank you, Michael Best regards, Michael Herman Self-Sovereign Blockchain Architect Trusted Digital Web

mwherman2000 (Sun, 31 Jan 2021 00:03:27 GMT):
Hi all, I’ve finished what I believe is an interesting document describing an effective solution for the SSI Personal Data Usage Licensing (SSI-PDUL) problem and would appreciate your comments and feedback. - You can edit a Google Docs copy of my Word master version of the whitepaper here: https://docs.google.com/document/d/1yP0OgDhq-3TTIien5gsETi4jZdyiKbwE/edit - If you prefer to read a PDF version, you can download it from here: https://hyperonomy.com/2021/01/27/self-sovereign-identity-personal-data-usage-licensing-ssi-pdul-model-solution-concept-review-draft-0-27/ One, what I believe to be an authoritative, comment that I’ve received so far is: Otherwise, I think it's a pretty unstudied problem. Abstract The purpose of this solution concept whitepaper is to provide the first complete description of the motivations, key concepts, problem statement, and solution approach for implementing the Self-Sovereign Identity Personal Data Usage Licensing (SSI-PDUL) Model, a functional component of the Self-Sovereign Identity Model (SSI Model). This solution concept addresses the following user scenario: How Alice User, an App User and Identity Owner, and Bob Developer, an App Developer and App Controller, might negotiate the use of Alice’s personal digital identifiers and any associated personal identity data by Bob’s app, based on Self-Sovereign Identity Model Usage Principles. The scope of the Self-Sovereign Identity Personal Data Usage Licensing (SSI-PDUL) Model is personal digital identifiers and any associated identity data presented by Alice to the App. It does not include the permissioning of data internal to the App (although the natural extension of the solution to internal data is an obvious one). The intended audience for this whitepaper is a broad range of professionals interested in furthering the application and use of the SSI Model in software apps, agents, and services using an SSI Personal Data Usage Licensing (SSI-PDUL) Model approach. This includes software architects, application developers, and user experience (UX) specialists; as well as people involved in a broad range of standards efforts related to decentralized identity, verified credentials, and secure storage. The work documented here was performed under the auspices of the Trusted Digital Web project in the Hyperonomy Digital Identity Lab of Parallelspace Corporation. Thank you, Michael Best regards, Michael Herman Self-Sovereign Blockchain Architect Trusted Digital Web

mwherman2000 (Sun, 31 Jan 2021 00:03:27 GMT):
Hi all, I’ve finished what I believe is an interesting document describing an effective solution for the SSI Personal Data Usage Licensing (SSI-PDUL) problem and would appreciate your comments and feedback. - You can edit a Google Docs copy of my Word master version of the whitepaper here: https://docs.google.com/document/d/1yP0OgDhq-3TTIien5gsETi4jZdyiKbwE/edit - If you prefer to read a PDF version, you can download it from here: https://hyperonomy.com/2021/01/27/self-sovereign-identity-personal-data-usage-licensing-ssi-pdul-model-solution-concept-review-draft-0-27/ One, what I believe to be an authoritative, comment that I’ve received so far is: _Otherwise, I think it's a pretty unstudied problem._ Abstract The purpose of this solution concept whitepaper is to provide the first complete description of the motivations, key concepts, problem statement, and solution approach for implementing the Self-Sovereign Identity Personal Data Usage Licensing (SSI-PDUL) Model, a functional component of the Self-Sovereign Identity Model (SSI Model). This solution concept addresses the following user scenario: How Alice User, an App User and Identity Owner, and Bob Developer, an App Developer and App Controller, might negotiate the use of Alice’s personal digital identifiers and any associated personal identity data by Bob’s app, based on Self-Sovereign Identity Model Usage Principles. The scope of the Self-Sovereign Identity Personal Data Usage Licensing (SSI-PDUL) Model is personal digital identifiers and any associated identity data presented by Alice to the App. It does not include the permissioning of data internal to the App (although the natural extension of the solution to internal data is an obvious one). The intended audience for this whitepaper is a broad range of professionals interested in furthering the application and use of the SSI Model in software apps, agents, and services using an SSI Personal Data Usage Licensing (SSI-PDUL) Model approach. This includes software architects, application developers, and user experience (UX) specialists; as well as people involved in a broad range of standards efforts related to decentralized identity, verified credentials, and secure storage. The work documented here was performed under the auspices of the Trusted Digital Web project in the Hyperonomy Digital Identity Lab of Parallelspace Corporation. Thank you, Michael Best regards, Michael Herman Self-Sovereign Blockchain Architect Trusted Digital Web__

mwherman2000 (Sun, 31 Jan 2021 00:03:27 GMT):
Hi all, I’ve finished what I believe is an interesting document describing an effective solution for the SSI Personal Data Usage Licensing (SSI-PDUL) problem and would appreciate your comments and feedback. - You can edit a Google Docs copy of my Word master version of the whitepaper here: https://docs.google.com/document/d/1yP0OgDhq-3TTIien5gsETi4jZdyiKbwE/edit - If you prefer to read a PDF version, you can download it from here: https://hyperonomy.com/2021/01/27/self-sovereign-identity-personal-data-usage-licensing-ssi-pdul-model-solution-concept-review-draft-0-27/ One, what I believe to be an authoritative, comment that I’ve received so far is: _Otherwise, I think it's a pretty unstudied problem._ Abstract The purpose of this solution concept whitepaper is to provide the first complete description of the motivations, key concepts, problem statement, and solution approach for implementing the Self-Sovereign Identity Personal Data Usage Licensing (SSI-PDUL) Model, a functional component of the Self-Sovereign Identity Model (SSI Model). This solution concept addresses the following user scenario: _How Alice User, an App User and Identity Owner, and Bob Developer, an App Developer and App Controller, might negotiate the use of Alice’s personal digital identifiers and any associated personal identity data by Bob’s app, based on Self-Sovereign Identity Model Usage Principles._ The scope of the Self-Sovereign Identity Personal Data Usage Licensing (SSI-PDUL) Model is personal digital identifiers and any associated identity data presented by Alice to the App. It does not include the permissioning of data internal to the App (although the natural extension of the solution to internal data is an obvious one). The intended audience for this whitepaper is a broad range of professionals interested in furthering the application and use of the SSI Model in software apps, agents, and services using an SSI Personal Data Usage Licensing (SSI-PDUL) Model approach. This includes software architects, application developers, and user experience (UX) specialists; as well as people involved in a broad range of standards efforts related to decentralized identity, verified credentials, and secure storage. The work documented here was performed under the auspices of the Trusted Digital Web project in the Hyperonomy Digital Identity Lab of Parallelspace Corporation. Thank you, Michael Best regards, Michael Herman Self-Sovereign Blockchain Architect Trusted Digital Web__

mwherman2000 (Sun, 31 Jan 2021 00:03:27 GMT):
Hi all, I’ve finished what I believe is an interesting document describing an effective solution for the SSI Personal Data Usage Licensing (SSI-PDUL) problem and would appreciate your comments and feedback. - You can edit a Google Docs copy of my Word master version of the whitepaper here: https://docs.google.com/document/d/1yP0OgDhq-3TTIien5gsETi4jZdyiKbwE/edit - If you prefer to read a PDF version, you can download it from here: https://hyperonomy.com/2021/01/27/self-sovereign-identity-personal-data-usage-licensing-ssi-pdul-model-solution-concept-review-draft-0-27/ One, what I believe to be an authoritative, comment that I’ve received so far is: _Otherwise, I think it's a pretty unstudied problem._ Abstract The purpose of this solution concept whitepaper is to provide the first complete description of the motivations, key concepts, problem statement, and solution approach for implementing the Self-Sovereign Identity Personal Data Usage Licensing (SSI-PDUL) Model, a functional component of the Self-Sovereign Identity Model (SSI Model). This solution concept addresses the following user scenario: _How Alice User, an App User and Identity Owner, and Bob Developer, an App Developer and App Controller, might negotiate the use of Alice’s personal digital identifiers and any associated personal identity data by Bob’s app, based on Self-Sovereign Identity Model Usage Principles._ The scope of the Self-Sovereign Identity Personal Data Usage Licensing (SSI-PDUL) Model is personal digital identifiers and any associated identity data presented by Alice to the App. It does not include the permissioning of data internal to the App (although the natural extension of the solution to internal data is an obvious one). The intended audience for this whitepaper is a broad range of professionals interested in furthering the application and use of the SSI Model in software apps, agents, and services using an SSI Personal Data Usage Licensing (SSI-PDUL) Model approach. This includes software architects, application developers, and user experience (UX) specialists; as well as people involved in a broad range of standards efforts related to decentralized identity, verified credentials, and secure storage. The work documented here was performed under the auspices of the Trusted Digital Web project in the Hyperonomy Digital Identity Lab of Parallelspace Corporation. Thank you, Michael Best regards, Michael Herman Self-Sovereign Blockchain Architect Trusted Digital Web

mwherman2000 (Sun, 31 Jan 2021 00:03:27 GMT):
Hi all, I’ve finished what I believe is an interesting document describing an effective solution for the *SSI Personal Data Usage Licensing (SSI-PDUL) problem* and would appreciate your comments and feedback. - You can edit a Google Docs copy of my Word master version of the whitepaper here: https://docs.google.com/document/d/1yP0OgDhq-3TTIien5gsETi4jZdyiKbwE/edit - If you prefer to read a PDF version, you can download it from here: https://hyperonomy.com/2021/01/27/self-sovereign-identity-personal-data-usage-licensing-ssi-pdul-model-solution-concept-review-draft-0-27/ One, what I believe to be an authoritative, comment that I’ve received so far is: _Otherwise, I think it's a pretty unstudied problem._ Abstract The purpose of this solution concept whitepaper is to provide the first complete description of the motivations, key concepts, problem statement, and solution approach for implementing the Self-Sovereign Identity Personal Data Usage Licensing (SSI-PDUL) Model, a functional component of the Self-Sovereign Identity Model (SSI Model). This solution concept addresses the following user scenario: _How Alice User, an App User and Identity Owner, and Bob Developer, an App Developer and App Controller, might negotiate the use of Alice’s personal digital identifiers and any associated personal identity data by Bob’s app, based on Self-Sovereign Identity Model Usage Principles._ The scope of the Self-Sovereign Identity Personal Data Usage Licensing (SSI-PDUL) Model is personal digital identifiers and any associated identity data presented by Alice to the App. It does not include the permissioning of data internal to the App (although the natural extension of the solution to internal data is an obvious one). The intended audience for this whitepaper is a broad range of professionals interested in furthering the application and use of the SSI Model in software apps, agents, and services using an SSI Personal Data Usage Licensing (SSI-PDUL) Model approach. This includes software architects, application developers, and user experience (UX) specialists; as well as people involved in a broad range of standards efforts related to decentralized identity, verified credentials, and secure storage. The work documented here was performed under the auspices of the Trusted Digital Web project in the Hyperonomy Digital Identity Lab of Parallelspace Corporation. Thank you, Michael Best regards, Michael Herman Self-Sovereign Blockchain Architect Trusted Digital Web

mwherman2000 (Sun, 31 Jan 2021 19:26:41 GMT):
Hopefully, it’s not sacrilege to suggest something like the following might be useful: _any file formats for the secure export/import of Verified Credentials_? …something like the file formats used for exporting/importing X.509 certificates (https://en.wikipedia.org/wiki/X.509#Certificate_filename_extensions).

mwherman2000 (Sun, 31 Jan 2021 19:26:41 GMT):
Hopefully, it’s not sacrilege to suggest something like the following might be useful: _any file formats for the secure export/import of Verified Credentials_? …something like the file formats used for exporting/importing X.509 certificates (https://en.wikipedia.org/wiki/X.509#Certificate_filename_extensions ...e.g. https://en.wikipedia.org/wiki/PKCS_12).

mwherman2000 (Sun, 31 Jan 2021 19:33:03 GMT):
The scenario is bootstrapping an app with a copy of a VC for you as "Member" or "Citizen" of the app. The attributes would only be shared within the app on a traditional SSI selective disclosure basis ...so it the very initial use case where you're claiming an instance of a mobile or desktop as "your own" ...an identity bootstrapping process ...where you're not beholding to any issuer or service provider once the credential has been issued. You can take it and present it to any app you want to.

mwherman2000 (Sun, 31 Jan 2021 19:33:03 GMT):
The scenario is bootstrapping an app with a copy of a VC for you as a "Member" or "Citizen" of the app. The attributes would only be shared within the app on a traditional SSI selective disclosure basis ...so it is the very initial use case where you're claiming an instance of a mobile or desktop as "your own" ...an identity bootstrapping process ...where you're not beholding to any issuer or service provider once the credential has been issued (modulo verification). You can take it and present it to any app you want to.

mwherman2000 (Sun, 31 Jan 2021 19:33:03 GMT):
The scenario is bootstrapping an app with a copy of a VC for you as a "Member" or "Citizen" of the app. The attributes would only be shared within the app on a traditional SSI selective disclosure basis ...so it is the very initial use case where you're claiming an instance of a mobile or desktop as "your own" ...an identity bootstrapping/onboarding process ...where you're not beholding to any issuer or service provider once the credential has been issued (modulo verification). You can take it and present it to any app you want to.

RazaDen (Sun, 31 Jan 2021 20:56:53 GMT):
Has joined the channel.

andrew.whitehead (Sun, 31 Jan 2021 22:52:46 GMT):
I don't really see how that would work, as the credential wouldn't have any binding to the new identity you're creating. Maybe if it was self-issued at the onset.

mwherman2000 (Mon, 01 Feb 2021 00:53:08 GMT):
Alice User would initially use an Issuer to get her Member or Citizen credential then she would persist it to some sort of secure, password protected file format like PKS12.

huytn.it (Mon, 01 Feb 2021 04:02:23 GMT):
Has joined the channel.

huytn.it (Mon, 01 Feb 2021 04:38:17 GMT):
I'm studying "Indy story walkthrough" 1. At "Step 4, Onboarding Faber ... by Steward" section "Connecting the establishment", "5.Steward sends connection request to Faber" -> my question is how steward send connection request to Faber???

huytn.it (Mon, 01 Feb 2021 04:38:17 GMT):
I'm studying "Indy story walkthrough" 1. At "Step 4, Onboarding Faber ... by Steward" section "Connecting the establishment", "5.Steward sends connection request to Faber" -> my question is how steward send connection request to Faber? section "Getting verinym" "4.Faber sends the encrypted message to the Steward" -> how to faber sends encrypted message to the steward? 2. How to Alice uses the her card to connect to other networks? example: she wants to apply a job to ABC which is on sorvin network.

RonaldReagan (Mon, 01 Feb 2021 05:54:28 GMT):
Has joined the channel.

abuaesh (Mon, 01 Feb 2021 07:03:09 GMT):
Has joined the channel.

mwherman2000 (Mon, 01 Feb 2021 10:17:01 GMT):
@huytn.it Take a look at this explanation of the Alice Buys A Car scenario: https://github.com/mwherman2000/indy-gsg-ea/blob/master/python/doc/getting_started-enterpise.md It's a bit dated but I think it will answer your question.

karthiksamaganam (Mon, 01 Feb 2021 14:43:59 GMT):
Has joined the channel.

morticianmili (Mon, 01 Feb 2021 16:39:28 GMT):
Has joined the channel.

madhugoundla (Tue, 02 Feb 2021 01:26:06 GMT):
Has joined the channel.

madhugoundla (Tue, 02 Feb 2021 01:26:06 GMT):
Hi, how can i start my own indy network? for the most part i only saw how to setup dev but never really saw any particular production installation guide.

dcspinho (Tue, 02 Feb 2021 01:41:28 GMT):
Hi. to authenticate do we need aries for agents?

ascatox (Tue, 02 Feb 2021 08:41:42 GMT):
Has joined the channel.

daidoji (Tue, 02 Feb 2021 14:12:46 GMT):
To run your own ledger you'd have to follow the indy setup rules (or you can start from https://github.com/bcgov/von-network and work your way up to running your own ledger)

madhugoundla (Tue, 02 Feb 2021 18:28:44 GMT):
Thank you, let me try the von network. Where is the indy setup rules documentation?

daidoji (Tue, 02 Feb 2021 18:37:51 GMT):
not sure, in the docs somewhere but it is a bit disorganized. Maybe someone else can show specifically. von-network is probably the path of least resistance though.

lynn.bendixsen (Tue, 02 Feb 2021 23:36:09 GMT):
The von-network is a great place to start to learn about and test an Indy network. When you are ready, though, I have some documents that can help you to set up your "production" network. Start here: https://docs.google.com/document/d/1Tg4dAEtC78TxG9AsIby_CfpbeOicK_YMKznSQOvtIVU/edit

madhugoundla (Wed, 03 Feb 2021 14:47:42 GMT):
@lynn.bendixsen Thank you so much, i will start with von network. That is a great documentation. Once i finish with von, i will go through the documentation. Thank you again for taking your time and sharing the information.

sownak (Wed, 03 Feb 2021 17:45:27 GMT):
You can also check https://blockchain-automation-framework.readthedocs.io/en/latest/example/indy-refapp.html which has samples to create an Indy network in minikube

luis.marado (Thu, 04 Feb 2021 10:50:11 GMT):
Has joined the channel.

Helen_Garneau (Thu, 04 Feb 2021 15:09:28 GMT):
Has joined the channel.

zossimov (Thu, 04 Feb 2021 20:34:11 GMT):
Hi, in von-network web app, status is loading in *Validator Node Status* section. cc: @WadeBarnes @swcurran

WadeBarnes (Thu, 04 Feb 2021 20:36:38 GMT):
Are you saying it stays at `loading` and never switches to:

WadeBarnes (Thu, 04 Feb 2021 20:36:39 GMT):

Clipboard - February 4, 2021 12:35 PM

zossimov (Thu, 04 Feb 2021 20:42:04 GMT):
@WadeBarnes yes

zossimov (Thu, 04 Feb 2021 20:47:11 GMT):
and in *Authenticate New DID* section, getting an error *Error: identity not registered.*

WadeBarnes (Thu, 04 Feb 2021 20:48:29 GMT):
Where are you running it, and what was the command you used to start it?

zossimov (Thu, 04 Feb 2021 20:49:23 GMT):
`AWS Machine ... sudo ./manage start AWS_MACHINE_IP_ADDRESS WEB_SERVER_HOST_PORT=80 "LEDGER_INSTANCE_NAME=My Ledger" &`

zossimov (Thu, 04 Feb 2021 20:49:23 GMT):
AWS Machine ... `sudo ./manage start AWS_MACHINE_IP_ADDRESS WEB_SERVER_HOST_PORT=80 "LEDGER_INSTANCE_NAME=My Ledger" &`

WadeBarnes (Thu, 04 Feb 2021 20:50:51 GMT):
Have you tried restarting? `./manage stop`, `./manage start AWS_MACHINE_IP_ADDRESS WEB_SERVER_HOST_PORT=80 "LEDGER_INSTANCE_NAME=My Ledger" &`

zossimov (Thu, 04 Feb 2021 20:51:48 GMT):
I used `./manage down`.. let me try it

WadeBarnes (Thu, 04 Feb 2021 20:52:26 GMT):
That is typically an indication one or more nodes have not started correctly or the browser container has not connected to the nodes. The latter can be a timing issue where the nodes have not completely initialized before the browser starts and tries to connect.

zossimov (Thu, 04 Feb 2021 20:57:11 GMT):
after using `./manage stop`, the loading stopped, but now again loading started.. why?

zossimov (Thu, 04 Feb 2021 20:57:11 GMT):
after using `./manage stop` and start the von-network, the loading stopped, but now again loading started.. why?

zossimov (Thu, 04 Feb 2021 20:58:30 GMT):
and also I used the genesis file in aries-dotnet-app, and try to create schema, getting an error `An unmapped error with the code '307' was returned by the SDK.`

zossimov (Thu, 04 Feb 2021 20:58:30 GMT):
and also I used the genesis file in aries-dotnet app, and try to create schema, it returns an error `An unmapped error with the code '307' was returned by the SDK.`

zossimov (Thu, 04 Feb 2021 20:58:30 GMT):
and also I used the genesis file in aries-dotnet app, and try to create schema, it returned an error `An unmapped error with the code '307' was returned by the SDK.`

zossimov (Thu, 04 Feb 2021 20:59:46 GMT):
aries-dotnet app `Startup.cs` config, `IssuerDid = "Th7MpTaRZVRYnPiabds81Y"; IssuerKeySeed = "000000000000000000000000Steward1";`

WadeBarnes (Thu, 04 Feb 2021 21:21:58 GMT):
`./manage down` deleted that volumes. `./manage stop` leaves them intact. Start von-network. Leave it for a bit. It can take a minute or so for everything to initialize. If, after a few minutes, it still says `loading`. Use `./manage stop` , and then start it again. Typically the nodes will come up faster and things will initialize in the right order.

WadeBarnes (Thu, 04 Feb 2021 21:33:28 GMT):
Also the IP address you are using for `AWS_MACHINE_IP_ADDRESS`. Is it the internal, or external IP address for that machine?

WadeBarnes (Thu, 04 Feb 2021 21:33:28 GMT):
Also the IP address you are using for `AWS_MACHINE_IP_ADDRESS`. Is it the internal, or external IP address for that VM?

WadeBarnes (Thu, 04 Feb 2021 21:45:38 GMT):
You saw this documentation; correct? https://github.com/bcgov/von-network#aws-ec2-security-considerations

zossimov (Thu, 04 Feb 2021 22:39:12 GMT):
external

bshambaugh (Thu, 04 Feb 2021 23:30:56 GMT):
Has joined the channel.

bshambaugh (Thu, 04 Feb 2021 23:30:57 GMT):
pasting this from #general : What is the the most expedient means to get going with publishing a DID document to the ledger? I found this: https://github.com/hyperledger/indy-sdk/blob/master/README.md which seems to be a prerequsite for this https://sovrin.org/wp-content/uploads/Token-Test-Instructions.pdf . I hope I am not confusing hyperledger indy and sovrin though. Evernym built indy, sovrin uses it along with its governance. afaik. I'd imagine new stuff is never easy. I think. Oh wow, this seems heavy. I'm dowloading 200MB of files, installing Rust, etc. Is there a docker image for everything or just for a local pool, or is a local pool all I need? I imagine the answer is somewhere along the lines of: it depends. If it helps, the machine I'm using is a beast (circa a few years ago). The pipe out of the net is small though.

bshambaugh (Thu, 04 Feb 2021 23:31:53 GMT):
swcurran suggested I got over here. I know I am basically saying, explain gravity. Where is the 1st place to go?

bshambaugh (Thu, 04 Feb 2021 23:31:53 GMT):
swcurran suggested I got over here. I know I am basically saying, fafhsdjgsdghsgssgdggsgy. Where is the 1st place to go?

bshambaugh (Thu, 04 Feb 2021 23:31:53 GMT):
swcurran suggested I got over here. I know I am basically saying, fafhsdjgsdghsgssgdggsgy. Where is the 1st place to go? The simplest thing.

bshambaugh (Thu, 04 Feb 2021 23:31:53 GMT):
swcurran suggested I got over here. I know I am basically saying, fafhsdjgsdghsgssgdggsgy. Where is the 1st place to go? The simplest most light weight thing.

bshambaugh (Thu, 04 Feb 2021 23:31:53 GMT):
swcurran suggested I got over here. I know I am basically saying, fafhsdjgsdghsgssgdggsgy. Where is the 1st place to go? The simplest most light weight thing. I'm still looking. Hopefully soon I have a better question.

bshambaugh (Thu, 04 Feb 2021 23:31:53 GMT):
swcurran suggested I got over here. I imagine I am basically saying, fafhsdjgsdghsgssgdggsgy. Where is the 1st place to go? The simplest most light weight thing. I'm still looking. Hopefully soon I have a better question.

swcurran (Fri, 05 Feb 2021 00:51:33 GMT):
So -- some thoughts. Indy does not support the DID spec, focuses mainly on verifiable credential exchange using AnonCreds, where only the Issuer of VCs needs to have a DID on the ledger. To create a DID on a public ledger, you can use the Indy CLI to generate a DID and then go to https://selfserve.sovrin.org/ to publish it. The same process occurs on Sovrin MainNet, but requires more legal overhead -- the governance. To add more to the DID (such as an Endpoint), you can use the CLI. However, most people are using Indy because they want an agent to issue, hold, prove or verify VCs. For that, you would probably want to get into Aries. In other words -- just creating a DID on a public ledger is not really "a thing" in Indy -- it's usually associated with VCs, and for VCs, there is a lot more moving parts. Hope that helps get you to the next step...

bshambaugh (Fri, 05 Feb 2021 01:24:31 GMT):
This does help a bit. I'm sure it will become more useful as I go. Thanks.

bshambaugh (Fri, 05 Feb 2021 01:31:09 GMT):
I got indy-cli installed

bshambaugh (Fri, 05 Feb 2021 04:04:15 GMT):
The tutorials to create a verinym talk about creating a DID, but what if I want customize the DID document. I am not entirely sure what it should look like: { "@context": "https://www.w3.org/ns/did/v1", "id": "did:example:F9C36F8964623378BDC068D4BCE07ED17C8FA486F9AC0C2613CA3C8C306D7BB6", "authentication": [{ "id": "did:example:F9C36F8964623378BDC068D4BCE07ED17C8FA486F9AC0C2613CA3C8C306D7BB#keys-1", "type": "JsonWebKey2020", "controller": "did:example:F9C36F8964623378BDC068D4BCE07ED17C8FA486F9AC0C2613CA3C8C306D7BB6", "publicKeyJwk": { "kty": "EC", "crv": "P-256", "x": "F9C36F8964623378BDC068D4BCE07ED17C8FA486F9AC0C2613CA3C8C306D7BB6", "y": "1CD36717B8AC5E4FEA8AD23DC8D0783C2318EE4AD7A80DB6E0026AD0B072A24F" } }] }

bshambaugh (Fri, 05 Feb 2021 04:04:15 GMT):
{ "@context": "https://www.w3.org/ns/did/v1", "id": "did:sov:ffsgerhdagwtgrsfvsgdhs", "authentication": [{ "id": "did:example:F9C36F8964623378BDC068D4BCE07ED17C8FA486F9AC0C2613CA3C8C306D7BB#keys-1", "type": "JsonWebKey2020", "controller": "did:example:F9C36F8964623378BDC068D4BCE07ED17C8FA486F9AC0C2613CA3C8C306D7BB6", "publicKeyJwk": { "kty": "EC", "crv": "P-256", "x": "F9C36F8964623378BDC068D4BCE07ED17C8FA486F9AC0C2613CA3C8C306D7BB6", "y": "1CD36717B8AC5E4FEA8AD23DC8D0783C2318EE4AD7A80DB6E0026AD0B072A24F" } }] }

bshambaugh (Fri, 05 Feb 2021 04:04:15 GMT):
{ "@context": "https://www.w3.org/ns/did/v1", "id": "did:sov:WRfXPg8dantKVubE3HX8pw", "authentication": [{ "id": "did:example:F9C36F8964623378BDC068D4BCE07ED17C8FA486F9AC0C2613CA3C8C306D7BB#keys-1", "type": "JsonWebKey2020", "controller": "did:example:F9C36F8964623378BDC068D4BCE07ED17C8FA486F9AC0C2613CA3C8C306D7BB6", "publicKeyJwk": { "kty": "EC", "crv": "P-256", "x": "F9C36F8964623378BDC068D4BCE07ED17C8FA486F9AC0C2613CA3C8C306D7BB6", "y": "1CD36717B8AC5E4FEA8AD23DC8D0783C2318EE4AD7A80DB6E0026AD0B072A24F" } }] }

bshambaugh (Fri, 05 Feb 2021 04:23:37 GMT):
want: verinym with custom did document

wiktNat (Fri, 05 Feb 2021 08:53:47 GMT):
Has joined the channel.

WadeBarnes (Fri, 05 Feb 2021 15:08:38 GMT):
Does this help? https://chat.hyperledger.org/channel/indy?msg=aA5ig3QNRzTyBpGGm

zossimov (Fri, 05 Feb 2021 15:34:25 GMT):
the Inbound and Outbound traffic is enabled on public EC2 address. I stop, down, build and start again and it worked.. but after few minutes later it again goes to loading status.. tried many times.. now, it's running fine from about 10 minutes..

zossimov (Fri, 05 Feb 2021 15:34:58 GMT):

zossimov - Fri Feb 05 2021 20:34:52 GMT+0500 (Pakistan Standard Time).txt

zossimov (Fri, 05 Feb 2021 15:42:11 GMT):

logs.txt

zossimov (Fri, 05 Feb 2021 15:43:22 GMT):
I'm trying to connect my **aries-dotnet app** with the pool, but it returns an error `An unmapped error with the code '307' was returned by the SDK.` below are the app configurations. **Startup.cs** ``` builder.RegisterAgent(c => { c.AgentName = "Issuer"; c.EndpointUri = "http://12.333.444.567:4000"; c.WalletConfiguration = new WalletConfiguration { Id = "Issuer" }; c.WalletCredentials = new WalletCredentials { Key = "IssuerKey" }; c.IssuerDid = "Th7MpTaRZVRYnPiabds81Y"; c.IssuerKeySeed = "000000000000000000000000Steward1"; c.GenesisFilename = "AriesTest.txn"; c.PoolName = "Aries"; c.ProtocolVersion = 2; }); ``` **launchSettings.json** ``` { "profiles": { "AriesWebApp": { "commandName": "Project", "launchBrowser": true, "environmentVariables": { "ASPNETCORE_ENVIRONMENT": "Development" }, "applicationUrl": "http://0.0.0.0:4000" } } } ```

zossimov (Fri, 05 Feb 2021 15:57:47 GMT):
any help regarding this?

WadeBarnes (Fri, 05 Feb 2021 15:58:50 GMT):
The folks over on the #aries-framework-dotnet channel would be more suited to help you with that issue.

WadeBarnes (Fri, 05 Feb 2021 15:59:34 GMT):
What's the IP of your `von-network` instance so I can have a quick review?

WadeBarnes (Fri, 05 Feb 2021 16:00:35 GMT):
Are you able to register new DIDs?

zossimov (Fri, 05 Feb 2021 16:09:22 GMT):
http://54.251.146.111/

zossimov (Fri, 05 Feb 2021 16:10:07 GMT):
no yet.. one questions, how to register new DIDs? and what are they for? how to use them?

zossimov (Fri, 05 Feb 2021 16:10:07 GMT):
not yet.. one questions, how to register new DIDs? and what are they for? how to use them?

WadeBarnes (Fri, 05 Feb 2021 16:15:52 GMT):
That's a more fundamental question.

WadeBarnes (Fri, 05 Feb 2021 16:16:35 GMT):
Good place to start; https://github.com/hyperledger/aries-cloudagent-python/blob/main/docs/GettingStartedAriesDev/README.md

WadeBarnes (Fri, 05 Feb 2021 16:16:58 GMT):

Clipboard - February 5, 2021 8:15 AM

WadeBarnes (Fri, 05 Feb 2021 16:17:26 GMT):
Looks like the ledger browser is not connected to the ledger nodes fully.

WadeBarnes (Fri, 05 Feb 2021 16:18:18 GMT):
You'll need to figure that out. That may have something to do with your agent not being able to connect.

WadeBarnes (Fri, 05 Feb 2021 16:18:51 GMT):
Have you spun up a `von-network` instance locally and familarized yourself with it?

WadeBarnes (Fri, 05 Feb 2021 16:18:51 GMT):
Have you spun up a `von-network` instance locally and familiarized yourself with it?

zossimov (Fri, 05 Feb 2021 16:19:28 GMT):
yes... few months earlier..

zossimov (Fri, 05 Feb 2021 16:20:05 GMT):
if I followed the documentation, why it's not working?

zossimov (Fri, 05 Feb 2021 16:20:55 GMT):
the seed you entered was random? and repeating my question where to use this did?

WadeBarnes (Fri, 05 Feb 2021 16:26:40 GMT):
Yes the seed I entered was random. I was just testing registration.

WadeBarnes (Fri, 05 Feb 2021 16:27:36 GMT):
Read through this, it should help answer you questions about DIDs; https://github.com/hyperledger/aries-cloudagent-python/blob/main/docs/GettingStartedAriesDev/README.md

zossimov (Fri, 05 Feb 2021 16:27:37 GMT):
I actually read some documentation on indy-node https://hyperledger-indy.readthedocs.io/projects/node/en/latest/transactions.html

WadeBarnes (Fri, 05 Feb 2021 16:28:46 GMT):
Those docs are too low level, you should start with the higher level docs I provided.

zossimov (Fri, 05 Feb 2021 16:29:14 GMT):
yes, I read them..

WadeBarnes (Fri, 05 Feb 2021 16:29:44 GMT):
In most cases you don't need to understand things at the low level since the Aries frameworks abstract the complexities.

zossimov (Fri, 05 Feb 2021 16:32:09 GMT):
I'm actually working on aries-dontet from 1 year and developed the mobile app, and api for issuer & verifier.. for testing we used the docker file from indy-sdk project / libindy e.t.c to deploy project on AWS.. and it was working smoothly (schema, cred definition id, credentials. presentation proof)..

zossimov (Fri, 05 Feb 2021 16:32:09 GMT):
I'm actually working on aries-dotnet from 1 year and developed the mobile app, and api for issuer & verifier.. for testing we used the docker file from indy-sdk project / libindy e.t.c to deploy project on AWS.. and it was working smoothly (schema, cred definition id, credentials. presentation proof)..

WadeBarnes (Fri, 05 Feb 2021 16:33:15 GMT):
@jcourt, Are you able to help troubleshoot the `von-network` / AWS issue here?

zossimov (Fri, 05 Feb 2021 16:33:20 GMT):
now, we want a large scale network for multiple issuers, that's when we found von-network.. it's the right choice, right?

zossimov (Fri, 05 Feb 2021 16:33:20 GMT):
now, we want a large scale network for multiple issuers on different servers, that's when we found von-network.. it's the right choice, right?

WadeBarnes (Fri, 05 Feb 2021 16:34:27 GMT):
Oh, NO. Not for a production scale and quality network.

zossimov (Fri, 05 Feb 2021 16:35:16 GMT):
WHAT??? what you will suggest??

WadeBarnes (Fri, 05 Feb 2021 16:35:18 GMT):
You're going to want to look at using something like the Sovrin Network; https://sovrin.org/

zossimov (Fri, 05 Feb 2021 16:36:02 GMT):
yes, but for now, we want our own network similar to sovrin (not that large)..

zossimov (Fri, 05 Feb 2021 16:36:30 GMT):
is there any open source project, we could use (we can develop for ourselves)..

zossimov (Fri, 05 Feb 2021 16:36:30 GMT):
is there any open source project, we could use (we can develop for ourselves)?

WadeBarnes (Fri, 05 Feb 2021 16:37:18 GMT):
Sovrin BuilderNet and StagingNet are free to use.

WadeBarnes (Fri, 05 Feb 2021 16:38:13 GMT):
If you're just looking for a provisional network then `von-network` is fine.

zossimov (Fri, 05 Feb 2021 16:38:34 GMT):
Our Requirement!!! We want multiple Issuers (who can offer/issuer/reject/revoke credentials) on different servers, connected to a pool on our main server. we would be the only one who can add/remove issuer. Is it possible?

zossimov (Fri, 05 Feb 2021 16:41:31 GMT):
Couldn't find Sovrin BuilderNet and StagingNet..

WadeBarnes (Fri, 05 Feb 2021 16:42:06 GMT):
`von-network` would meet those requirements other than the permission bit. There are no permissions/rules setup on it to stop someone who knows the default seed from doing whatever they want.

WadeBarnes (Fri, 05 Feb 2021 16:42:32 GMT):
https://sovrin.org/overview/

WadeBarnes (Fri, 05 Feb 2021 16:42:50 GMT):
https://selfserve.sovrin.org/

WadeBarnes (Fri, 05 Feb 2021 16:43:01 GMT):
Ignore the payment address field.

zossimov (Fri, 05 Feb 2021 16:43:14 GMT):
can't we setup on our servers?

zossimov (Fri, 05 Feb 2021 16:43:53 GMT):
sorry for taking your time, but as for our requirement (we want to run it on servers and control the network), what will u suggest?

WadeBarnes (Fri, 05 Feb 2021 16:46:45 GMT):
@lynn.bendixsen, I can't seem to put my finger on the document you have for setting up a new network.

WadeBarnes (Fri, 05 Feb 2021 17:07:12 GMT):
Found it; https://docs.google.com/document/d/1XE2QOiGWuRzWdlxiI9LrG9Am9dCfPXBXnv52wGHorNE

zossimov (Fri, 05 Feb 2021 17:12:19 GMT):
Thank you so much... I will try it and will update.. again thanks :)

zossimov (Fri, 05 Feb 2021 17:12:19 GMT):
Thank you so much... I will try it and will update.. again thanks for your time :)

abuaesh (Sun, 07 Feb 2021 13:01:10 GMT):
Pasting from #general: Trying to run python tests on indy, but I get this error: FileNotFoundError: Could not find module 'indy.dll' (or one of its dependencies). Try using the full path with constructor syntax.

abuaesh (Mon, 08 Feb 2021 08:40:33 GMT):
Fixed it. The tests should be run from indy-sdk\libindy\tests path. Also the command to test is: cargo test -- --test-threads=1

paul.bastian (Mon, 08 Feb 2021 17:30:09 GMT):
hi everyone, as the SSI ecosystem grows, we feel there is a growing need for a standardized wallet security ecosystem and therefore a working group called "Wallet Security" is to be launched under DIF umbrella. In particular we think that: - we should get a common understanding, terminology about wallet security in the community - talk about wallet security architectures to analyze similarities - create a specification and interface to communicate about wallet capabilities, security, regulation-conformance and other points of interoperability - define mechanism to enable wallet security assertions, certification and ways to prove them - define specifications about wallet user authentication, ways how to ensure them and how to communicate them to issuers/verifiers If you are interested about this topic and like to comment, participate, criticize: - please review the Wallet Security WG Charter: https://docs.google.com/document/d/18H2hVjHZEBjbnzod8tLogJIEzySdecbk9d-QBJaqHP0/edit - subscribe to the mailing list: https://lists.identity.foundation/g/wallet-security After gathering initial feedback, we will have an open discussion meeting soon. I hope this is in the best interests of the community and will get the necessary feedback to make an impact. Best regards, Paul Bastian

esplinr (Mon, 08 Feb 2021 19:48:35 GMT):
Thank you for the invitation Paul!

abuaesh (Wed, 10 Feb 2021 20:24:54 GMT):
Fixed by placing all the dll files in the system32 folder

abuaesh (Wed, 10 Feb 2021 20:27:52 GMT):
Hi all, I get this error trying to run the tests for indy-sdk python wrappers: "tests\ledger\test_build_txn_author_agreement_request.py Windows fatal exception: access violation" Any idea how to fix that? I work on windows 10 hyper-V image.

mccown (Wed, 10 Feb 2021 22:07:46 GMT):
The Identity Implementer's WG is meeting tomorrow (Feb 11th @ 9am MT / 4pm UTC). In this call, we will review updates from several industry WGs, which you may have missed this week. For our presentation, Riley Hughes from Trinsic is presenting some of their new developments (https://trinsic.id). Here are links to the Wiki and Zoom meeting: Wiki: https://wiki.hyperledger.org/display/IWG/2021-02-11+%3A+Identity+Implementers+WG+Call Zoom: https://zoom.us/my/hyperledger.community.3?pwd=UE90WHhEaHRqOGEyMkV3cldKa2d2dz09

abuaesh (Thu, 11 Feb 2021 15:01:53 GMT):
After creating a pool, I am trying to connect to it but I get: pool p1 can't be connected. Any help please?

Lucampelli (Thu, 11 Feb 2021 17:15:09 GMT):
Hello! Is it possible, in a scenario where, let's say that SSI (with Indy) was deployed on a country, for 2 people to have the same DID? In the Self Soverign Identity documents I've read from Indy and Sovrin, it's stated that a user can create as many DID's as it wants, and by studying the code a little, the creation of DID's is made offline, and if that DID needs to become public, it is made through the verynim steps described in the getting started. So, in my mind, it is possible for 2 distinct people to have credentials made for the same DID. Unlikely, but possible.

swcurran (Thu, 11 Feb 2021 17:20:23 GMT):
Define possible? :-). Bottom line is that it's not a concern and there is no need for centrally produced identifiers. I would suggest you look into UUIDs -- the same principles apply.

mwherman2000 (Thu, 11 Feb 2021 22:43:33 GMT):
When are Hyperledger Indy/Sovrin VCs better than Ethereum smart contracts for NFEs/NFTs (non-fungible entities/tokens)? It seems obvious but I don't have a detailed/worked out answer. One project I'm associated with wants to use the ERC-721 Non-Fungible Token Standard on Ethereum but I believe VCs are a better route to take. Part of the desire to stay on Ethereum is there is quite a vibrant NFT community on Ethereum and lots of different EC-721 tokens. What are the considerations/decision points/knock-offs? ...off to do some Googling.

mwherman2000 (Thu, 11 Feb 2021 22:46:04 GMT):
ERC-721 Non-Fungible Token Standard on Ethereum: https://eips.ethereum.org/EIPS/eip-721

abuaesh (Fri, 12 Feb 2021 06:39:06 GMT):
I believe you'd want to compare Indy with ERC725; the standard used for SSI on Ethereum. Here is what I missed after moving from working Ethereum to work with Indy: A vibrant community that hardly lets you go to sleep with unanswered questions. A solid ground of documentation and plethora of resources. They have at least one article for every problem you may run into. Now, I can get stuck in a technical problem for days and feel invisible when I ask questions here. I'm not sure if that's because I am using Windows(so not many people ran into the same problems), or because I am looking in the wrong places.

mwherman2000 (Fri, 12 Feb 2021 07:17:12 GMT):
@abuaesh It's the other way around: I want to do NFT better on Indy (using VCs) than either ECR-721 or ECR-1155 on Ethereum. Is this possible? ...can I do the same or better on Indy? What are the pros and cons?

mwherman2000 (Fri, 12 Feb 2021 07:17:12 GMT):
@abuaesh It's the other way around: I want to do NFTs better on Indy (using VCs) than either ECR-721 or ECR-1155 on Ethereum. Is this possible? ...can I do the same or better on Indy? What are the pros and cons?

abuaesh (Fri, 12 Feb 2021 10:22:12 GMT):
Yes, it is possible, since you can write anything really in a VC. You may want to check Indy plenum for that: a general purpose DLT. Indy node is just an identity-related implementation of plenum. Will it be better? I don't know tbh. I am still trying to get something to work on Indy!

Lucampelli (Fri, 12 Feb 2021 12:24:17 GMT):
Well, can two distinct people have in their wallets two similar DID's? Let's say you and I, you graduated university, created a DID for your diploma, eg. DID 1234, I completed a cooking course, and also created a DID for receiving the certificate, eg. DID 1234.

mwherman2000 (Fri, 12 Feb 2021 12:43:42 GMT):
@abuaesh The following is a little bit dated but if it's conceptual issues you're struggling with, the following may be helpful: https://github.com/mwherman2000/indy-gsg-ea/blob/master/python/doc/getting_started-enterpise.md

mwherman2000 (Fri, 12 Feb 2021 12:44:18 GMT):
@abuaesh p.s. It *is* nice to wake to find an answer waiting in the morning :-)

Priya.P (Mon, 15 Feb 2021 10:23:57 GMT):
Hi everyone,

Priya.P (Mon, 15 Feb 2021 10:31:49 GMT):
Hi everyone, I am using this doc to understand how the revocation works internally. I dint understand quite clearly about revocation registery if these are 3 nos 3,5,7 prime nos that form accumulator then what i read from the docs tails file will have following structure array &accumulator value [3,5,7]-->105 Is this understanding correct? now 3,5,7 corresponds to name id dob respectively now if this credential is revoked how this tails file will change &what prover sends to verifier when creating proof. can anyone please help me here

Priya.P (Mon, 15 Feb 2021 10:31:49 GMT):
Hi everyone, I am using this doc to understand how the revocation works internally. https://github.com/hyperledger/indy-sdk/blob/master/docs/concepts/revocation/cred-revocation.md I dint understand quite clearly about revocation registery if these are 3 nos 3,5,7 prime nos that form accumulator then what i read from the docs tails file will have following structure array &accumulator value [3,5,7]-->105 Is this understanding correct? now 3,5,7 corresponds to name id dob respectively now if this credential is revoked how this tails file will change &what prover sends to verifier when creating proof. can anyone please help me here

Priya.P (Mon, 15 Feb 2021 10:31:49 GMT):
Hi everyone, I am using this doc to understand how the revocation works internally. https://github.com/hyperledger/indy-sdk/blob/master/docs/concepts/revocation/cred-revocation.md I dint understand quite clearly about revocation registery if these are 3 nos 3,5,7 prime nos that form accumulator then what i read from the docs tails file will have following structure array &accumulator value [3,5,7]-->105 Is this understanding correct? now 3,5,7 corresponds to credential1 credential2 credential3 respectively now if this credential1 is revoked how this tails file will change &what prover sends to verifier when creating proof. can anyone please help me here

Priya.P (Mon, 15 Feb 2021 10:31:49 GMT):
Hi everyone, I am using this doc to understand how the revocation works internally. https://github.com/hyperledger/indy-sdk/blob/master/docs/concepts/revocation/cred-revocation.md I dint understand quite clearly about revocation registery if these are 3 nos 3,5,7 prime nos that form accumulator then what i read from the docs tails file will have following structure array &accumulator value [3,5,7]-->105 Is this understanding correct? now 3,5,7 corresponds to credential1 credential2 credential3 respectively ``` now if this credential1 is revoked how this tails file will change ``` ``` what prover sends to verifier when creating proof ``` ``` Prover knows the math problem. What does this refer to in above case with prime number 3,5,7 ``` can anyone please help me here

Priya.P (Mon, 15 Feb 2021 10:31:49 GMT):
Hi everyone, I am using this doc to understand how the revocation works internally. https://github.com/hyperledger/indy-sdk/blob/master/docs/concepts/revocation/cred-revocation.md I dint understand quite clearly about revocation registery if these are 3 nos 3,5,7 prime nos that form accumulator then what i read from the docs tails file will have following structure array &accumulator value [3,5,7]-->105 Is this understanding correct? now 3,5,7 corresponds to credential1 credential2 credential3 respectively now* if this credential1 is revoked how this tails file will change * *what prover sends to verifier when creating proof* *Prover knows the math problem. What does this refer to in above case with prime number 3,5,7* can anyone please help me here

Priya.P (Mon, 15 Feb 2021 10:31:49 GMT):
Hi everyone, I am using this doc to understand how the revocation works internally. https://github.com/hyperledger/indy-sdk/blob/master/docs/concepts/revocation/cred-revocation.md I dint understand quite clearly about revocation registery if these are 3 nos 3,5,7 prime nos that form accumulator then what i read from the docs tails file will have following structure array &accumulator value [3,5,7]-->105 Is this understanding correct? now 3,5,7 corresponds to credential1 credential2 credential3 respectively now* if this credential1 is revoked how this tails file will change ?* *what prover sends to verifier when creating proof?* *Prover knows the math problem. What does this refer to in above case with prime number 3,5,7?* can anyone please help me here

Priya.P (Mon, 15 Feb 2021 10:31:49 GMT):
Hi everyone, I am using this doc to understand how the revocation works internally. https://github.com/hyperledger/indy-sdk/blob/master/docs/concepts/revocation/cred-revocation.md I dint understand quite clearly about revocation registery if these are 3 nos 3,5,7 prime nos that form accumulator then what i read from the docs tails file will have following structure array &accumulator value [3,5,7]-->105 Is this understanding correct? now 3,5,7 corresponds to credential1 credential2 credential3 respectively now* if this credential1 is revoked how this tails file will change ?* *what prover sends to verifier when creating proof?* *Prover shows she can derive accumulator value the math problem. What does this refer to in above case with prime number 3,5,7?how does it show it can derive accumulator does it send private accumulator and witness as math problem?* can anyone please help me here

lynn.bendixsen (Mon, 15 Feb 2021 14:51:13 GMT):
Hi @abuaesh, It looks like we need some more details to be able to help you with this one. Csn you tell us what kind of pool you are trying to connect to? (test pool, docker pool, von image, public existing ledger, etc.) Usually that error comes from networking problems, or an issue in the genesis file used to create the pool handle. :)

lynn.bendixsen (Mon, 15 Feb 2021 14:51:13 GMT):
Hi @abuaesh, It looks like we need some more details to be able to help you with this one. Can you tell us what kind of pool you are trying to connect to? (test pool, docker pool, von image, public existing ledger, etc.) Usually that error comes from networking problems, or an issue in the genesis file used to create the pool handle. :)

swcurran (Mon, 15 Feb 2021 19:49:03 GMT):
You might take a look at this: https://github.com/hyperledger/indy-hipe/tree/master/text/0011-cred-revocation It has a fairly accessible description of how revocation works with tails files and the like.

pranav_kirtani (Tue, 16 Feb 2021 05:57:08 GMT):
are there any other public utilities like sovirin available?

pfeairheller (Tue, 16 Feb 2021 14:35:35 GMT):
@pranav_kirtani take a look at https://bedrockconsortium.org/

Shubham95 (Wed, 17 Feb 2021 07:20:46 GMT):
I am implementing aca-py with revocation but when I am starting faber with revocation it shows set tails base url. Can anyone please help me with how can I set tails base url.

RazaDen (Wed, 17 Feb 2021 07:48:46 GMT):
There is a command line parameter that you can supply to ACA-Py that tells it the URL of the tails server (separate process).

RazaDen (Wed, 17 Feb 2021 07:50:11 GMT):
Have a look at the following command line paremeters (as they are described by the help message of ACA-Py): ` --tails-server-base-url Sets the base url of the tails server in use. [env var: ACAPY_TAILS_SERVER_BASE_URL] --tails-server-upload-url Sets the base url of the tails server for upload, defaulting to the tails server base url. [env var: ACAPY_TAILS_SERVER_UPLOAD_URL]`

RazaDen (Wed, 17 Feb 2021 07:50:11 GMT):
Have a look at the following command line paremeters (as they are described by the help message of ACA-Py): ` --tails-server-base-url Sets the base url of the tails server in use. [env var: ACAPY_TAILS_SERVER_BASE_URL] --tails-server-upload-url Sets the base url of the tails server for upload, defaulting to the tails server base url. [env var: ACAPY_TAILS_SERVER_UPLOAD_URL] `

RazaDen (Wed, 17 Feb 2021 07:50:11 GMT):
Have a look at the following command line paremeters (as they are described by the help message of ACA-Py): ``` --tails-server-base-url Sets the base url of the tails server in use. [env var: ACAPY_TAILS_SERVER_BASE_URL] --tails-server-upload-url Sets the base url of the tails server for upload, defaulting to the tails server base url. [env var: ACAPY_TAILS_SERVER_UPLOAD_URL] ```

T_aanna (Wed, 17 Feb 2021 10:10:02 GMT):
Has joined the channel.

T_aanna (Wed, 17 Feb 2021 10:16:31 GMT):
Hi guys! I'm new to the hyperledger indy stuff and trying to implement a blockchain which can be used in the documentmanagement area (e.g. to varify signatures on documents in the document flow of a company). I started at 0 with Linux Foundation courses and tried to follow the Tutorials mentioned there. I got confused going through the docs. In some for example Python 3.5 is necessary in some a version higher than 3.5, in some you can run commands in your direct local environment in some a VM is necessary and even there a venv is sometimes recommended. I wondered if this is a must? are there any special requirements for using indy? or does it only vary because of the different tutorials and years they were published? Would be really great if someone could help me in the indy jungle :) Thanks

abuaesh (Wed, 17 Feb 2021 12:43:09 GMT):
There is also Indicio

abuaesh (Wed, 17 Feb 2021 12:43:09 GMT):
These are the ones I found so far: Indicio, IDunion, Bedrock.

abuaesh (Wed, 17 Feb 2021 12:46:48 GMT):
Indicio has a free testnet

abuaesh (Wed, 17 Feb 2021 13:21:29 GMT):
Trying to connect to a pool on Sovrin, so public existing ledger. Yes, I was using an incomplete genesis file(didn't copy the whole file from github). Fixed it now. Thanks!

abuaesh (Wed, 17 Feb 2021 13:27:05 GMT):
A DID is 32 characters long. Each character can be a digit or an alphabet letter->36 possibilities per character. A chance to generate a duplicate is one in a 32 power 36 chance.

abuaesh (Wed, 17 Feb 2021 13:28:03 GMT):
It is helpful. Thanks!

abuaesh (Wed, 17 Feb 2021 13:45:18 GMT):
Trying to write a credential schema to the ledger, I get the following error when the function ledger.sign_and_submit_request(pool_handle, wallet_handle, _did, schema_request) is called: {"identifier":"N5woRhHcnE7BBh3FwsPqFJ","reqId":1613568672368485900,"reason":"client request invalid: InvalidClientRequest('validation error [SchemaField]: invalid number of parts 1, should contain 2 or 3 (version=1)',)","op":"REQNACK"} Here is the schema request returned by ledger.build_schema_request(_did, schema): {"reqId":1613568672368485900,"identifier":"N5woRhHcnE7BBh3FwsPqFJ","operation":{"type":"101","data":{"name":"g","version":"1","attr_names":["c","b","a"]}},"protocolVersion":2} I am using python, on a Windows VM, connected to Sovrin.

abuaesh (Wed, 17 Feb 2021 13:47:09 GMT):
Hi there T_anna, yes, it took me a while before I could pass the installation test for Indy. Are you using Windows?

swcurran (Wed, 17 Feb 2021 15:43:47 GMT):
do you have a tails server running? There is an implementation here: https://github.com/bcgov/indy-tails-server It is trivial to run locally if you have docker -- clone the repo and then run `./manage start`

lynn.bendixsen (Wed, 17 Feb 2021 21:12:37 GMT):
"Versions must have at least 1 dot in them", but could contain as many as 2, according to Adam Burdett. The version string must look like a version, apparently. I looked for documentation to support it, but was unable to find it. The error message, though confusing, strongly implies that is the case, though.

swcurran (Wed, 17 Feb 2021 21:39:35 GMT):
Easiest way to run in Indy is using von-network https://github.com/bcgov/von-network Assuming you have docker/docker-compose/bash, you can be running a ledger in about 5 minutes.

swcurran (Wed, 17 Feb 2021 21:39:35 GMT):
Easiest way to run in Indy is using von-network https://github.com/bcgov/von-network Assuming you have docker/docker-compose/bash, you can be running a ledger in about 5 minutes with 4 commands, one of which is "cd"

Shubham95 (Thu, 18 Feb 2021 07:00:43 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=fuXcgfNNWYB6LzyPZ) I have done that already but still I am getting same error

RazaDen (Thu, 18 Feb 2021 09:10:47 GMT):
The tails server needs to have a "public IP", i.e., ACA-Py needs to be able to call it via its ip/domain name or w/e.

RazaDen (Thu, 18 Feb 2021 09:12:55 GMT):
Maybe that's the problem? Otherwise it's pretty straight forward. Let me give you an example where the tails server is configured to have a public endpoint, via ngrok: `--tails-server-base-url https://5cd2c2e7cde6.ngrok.io` This is for instance the parameter that we pass to one of our ACA-Py instances

T_aanna (Thu, 18 Feb 2021 13:17:22 GMT):
@abuaesh I'm using ubuntu 16.04

T_aanna (Thu, 18 Feb 2021 13:18:41 GMT):
@swcurran Thanks for the link! I followed the commands, but when trying to connect to the pool I get "Pool 'sandbox' does not exist", but got errors before.. Did I miss something here? Thanks a lot :)

swcurran (Thu, 18 Feb 2021 15:36:30 GMT):
Adding @WadeBarnes . There are a couple of things it might be.

WadeBarnes (Thu, 18 Feb 2021 15:55:04 GMT):
I need the answers to the questions I asked over here; https://chat.hyperledger.org/channel/indy-node?msg=RMrfK7GQkPt7CEy5s

T_aanna (Thu, 18 Feb 2021 16:38:29 GMT):
@WadeBarnes I'm trying to run the Tutorials on Ubuntu 16.04, docker version 18.09.7, build 2d0083d, docker-compose version 1.28.2

T_aanna (Thu, 18 Feb 2021 16:41:12 GMT):
@WadeBarnes at the moment I'm getting the pool sandbox error, mentioned above. Before that I tried to follow the IBM Tutorial, there I got errors that a specific version of docker, docker-compose etc. is necessary, which made no sense since I have the necessary versions installed.. Thanks a lot :)

lynn.bendixsen (Thu, 18 Feb 2021 17:03:14 GMT):
in other words, change "version=1" to "version=1.0" and your code should work.

WadeBarnes (Thu, 18 Feb 2021 17:42:53 GMT):
Are you able to get `von-network` up and running locally first? To the point were you're able to open the ledger browser and register a DID through the interface?

WadeBarnes (Thu, 18 Feb 2021 17:46:21 GMT):
Working example can be found here; http://dev.bcovrin.vonx.io/

WadeBarnes (Thu, 18 Feb 2021 17:46:43 GMT):

Clipboard - February 18, 2021 9:46 AM

T_aanna (Thu, 18 Feb 2021 18:45:26 GMT):
Hi! Following this guide https://github.com/IBM-Blockchain-Identity/indy-ssivc-tutorial I get errors on my ubuntu 16.04 when trying to build the network saying "Error while fetching server API version". After googling this issue one suggestion was to follow another guide, since the IBM guide is archived. When following this https://github.com/hyperledger-archives/education/blob/master/LFS171x/indy-material/nodejs/README.md I get the following erros: "ERROR: for node4 Cannot start service node4: driver failed programming external connectivity on endpoint nodejs_node4_1 (2b46b52b453ef2b4edef405e993534d83a5a5a5d9b53aba2cfcf4435c8855f55): Bind for 0.0.0.0:9708 failed: port is already allocated ERROR: Encountered errors while bringing up the project. /usr/local/lib/python3.5/site-packages/paramiko/transport.py:33: CryptographyDeprecationWarning: Python 3.5 support will be dropped in the next release of cryptography. Please upgrade your Python. from cryptography.hazmat.backends import default_backend Attaching to nodejs_node4_1, nodejs_node3_1, nodejs_node2_1, nodejs_node1_1 nodejs_node2_1 exited with code 128 nodejs_node1_1 exited with code 128 nodejs_node4_1 exited with code 128 nodejs_node3_1 exited with code 128 "

T_aanna (Thu, 18 Feb 2021 18:49:52 GMT):
I'm also getting the info that python3.5 is outdated. This info is actually not true, since "python --version" returns Python 3.6.xx

T_aanna (Thu, 18 Feb 2021 19:18:31 GMT):
Update: after resetting my docker containers i ran the command a second time, network came up, got the page where I can authenticate a new DID. Am I now done? Is this now a running von-network? Sorry for the dumb question, but since I got so many errors and different tutorials I am confused if this is it now..

abuaesh (Thu, 18 Feb 2021 21:25:47 GMT):
Yes! It works now--I mean giving a different reason for rejection: "op":"REJECT","reason":"client request invalid: InvalidClientTaaAcceptanceError('Txn Author Agreement acceptance is required for ledger with id 1',)"}

abuaesh (Thu, 18 Feb 2021 21:26:33 GMT):
@lynn.bendixsen I am using an endorser did to send the request.

abuaesh (Fri, 19 Feb 2021 08:53:51 GMT):
Trying to write a credential schema to the Sovrin Ledger, I get this response: {"identifier":"N5woRhHcnE7BBh3FwsPqFJ","reason":"client request invalid: InvalidClientTaaAcceptanceError('Txn Author Agreement acceptance is required for ledger with id 1',)","op":"REJECT","reqId":1613724753152741900} Here is the schema request I send: {"reqId":1613724753152741900,"identifier":"N5woRhHcnE7BBh3FwsPqFJ","operation":{"type":"101","data":{"name":"Government_ID","version":"0.1","attr_names":["issuing_state","first_name","date_of_birth","date_of_issue","nationality","id","last_name","date_of_expiry","gender"]}},"protocolVersion":2}

WadeBarnes (Fri, 19 Feb 2021 14:50:26 GMT):
That means you have `von-network` running properly. Now you can try connecting the agent and see how far you get with what you were doing before.

lynn.bendixsen (Fri, 19 Feb 2021 16:27:05 GMT):
You need to send the txn to the functions that add the TAA to the txn before sending it to a ledger that requires TAA. I will see if I can find the docs or a code snippet to help.

paul.bastian (Fri, 19 Feb 2021 18:43:19 GMT):
Hello folks, on March 1st, 6pm CET we will have an open call to kickoff the upcoming Wallet Security WG at DIF. I will present motivation, goals and a first roadmap. Very short summary: - standardized wallet security is necessary for sensitive credentials like id-cards, payment credentials or more - create a specification and interface to communicate about wallet capabilities, security, regulation-conformance and other points of security-relevant interoperability - define mechanism to enable wallet security assertions, certification and ways to prove them - define specifications about wallet user authentication, ways how to ensure them and how to communicate them to issuers/verifiers WG Charter:https://docs.google.com/document/d/18H2hVjHZEBjbnzod8tLogJIEzySdecbk9d-QBJaqHP0/edit Mailing list: https://lists.identity.foundation/g/wallet-security Topic: Wallet Security WG Kickoff Time: Mar 1, 2021 06:00 PM Amsterdam, Berlin, Rome, Stockholm, Vienna Join Zoom Meeting https://bundesdruckerei.zoom.us/j/88417582474?pwd=bXV1cE9udXNWK2VaYnFZdkNxaEx2UT09 Meeting ID: 884 1758 2474 Passcode: 391216 Join by Skype for Business https://bundesdruckerei.zoom.us/skype/88417582474

abuaesh (Sat, 20 Feb 2021 04:29:01 GMT):
Thanks Lynn. I found these 2 links talking about the TAA: https://github.com/hyperledger/indy-node/blob/master/docs/source/requests.md#transaction_author_agreement and https://github.com/hyperledger/indy-node/blob/master/docs/source/requests.md#common-write-request-structure But they don't really mention how to call the function doing this.

abuaesh (Sat, 20 Feb 2021 10:24:51 GMT):
Found something: indy_append_txn_author_agreement_acceptance_to_request Let me see if I can get it to work.

abuaesh (Sat, 20 Feb 2021 12:02:12 GMT):
*OK- Here is the error I get now: * _"reason":"client request invalid: InvalidClientTaaAcceptanceError('Txn Author Agreement acceptance time 1612822227 is too precise and is a privacy risk.',)","op":"REJECT"_ *Here is what I am doing: * _print("---------------------------------TAA PROCEDURE------------------------------------") # 1. Get Latest TAA from ledger taa_request = await ledger.build_get_txn_author_agreement_request(issuer['did'], None) latest_taa = await ledger.sign_and_submit_request(issuer['pool'], issuer['wallet'], issuer['did'], taa_request) print("\n\nStep A - TAA Retrieved: ") print(latest_taa) # 2. Get Acceptance Mechanisms List(AML) acceptance_mechanism_request = await ledger.build_get_acceptance_mechanisms_request(issuer['did'], None, None) aml = await ledger.sign_and_submit_request(issuer['pool'], issuer['wallet'], issuer['did'], acceptance_mechanism_request) print("\n\nStep B - AML Retrieved: ") print(aml) # 3. Append acceptance to your request latest_taa = json.loads(latest_taa) taa_digest = latest_taa['result']['data']['digest'] taa_version = latest_taa['result']['data']['version'] timestamp = calendar.timegm(time.gmtime()) #Time of accepting the TAA timestamp -= 1000000 request = await ledger.append_txn_author_agreement_acceptance_to_request(request, None, None, taa_digest, taa_version, timestamp) request = await ledger.sign_and_submit_request(issuer['pool'], issuer['wallet'], issuer['did'], request) print("\n\nStep C - TAA Appended to your Request: ") print(request) # 4. Return appended request_

abuaesh (Sat, 20 Feb 2021 12:02:12 GMT):
*OK- Here is the error I get now: * _"reason":"client request invalid: InvalidClientTaaAcceptanceError('Txn Author Agreement acceptance time 1612822227 is too precise and is a privacy risk.',)","op":"REJECT"_ *Here is what I am doing: * _print("---------------------------------TAA PROCEDURE------------------------------------") # 1. Get Latest TAA from ledger taa_request = await ledger.build_get_txn_author_agreement_request(issuer['did'], None) latest_taa = await ledger.sign_and_submit_request(issuer['pool'], issuer['wallet'], issuer['did'], taa_request) print("\n\nStep A - TAA Retrieved: ") print(latest_taa) # 2. Get Acceptance Mechanisms List(AML) acceptance_mechanism_request = await ledger.build_get_acceptance_mechanisms_request(issuer['did'], None, None) aml = await ledger.sign_and_submit_request(issuer['pool'], issuer['wallet'], issuer['did'], acceptance_mechanism_request) print("\n\nStep B - AML Retrieved: ") print(aml) # 3. Append acceptance to your request latest_taa = json.loads(latest_taa) taa_digest = latest_taa['result']['data']['digest'] taa_version = latest_taa['result']['data']['version'] timestamp = calendar.timegm(time.gmtime()) #Time of accepting the TAA request = await ledger.append_txn_author_agreement_acceptance_to_request(request, None, None, taa_digest, taa_version, timestamp) request = await ledger.sign_and_submit_request(issuer['pool'], issuer['wallet'], issuer['did'], request) print("\n\nStep C - TAA Appended to your Request: ") print(request) # 4. Return appended request_

Tanvi (Sat, 20 Feb 2021 14:34:18 GMT):
Has joined the channel.

Tanvi (Sat, 20 Feb 2021 14:48:33 GMT):
Hi everyone, I am trying out revocation with von tails server. For demonstration purposes i am using current nodejs revocation sample which uses tails file stored locally in .indy-client folder I am referring this readme file https://github.com/bcgov/indy-tails-server. i have deployed this server with docker I ran nodejs revocation sample (not deleting wallet) sample failed to revoke credential. So to test server i tried sending PUT request to von server but i am getting Revocation registry ID is not valid:GcjVZQ... Can anyone please help me here?

Tanvi (Sat, 20 Feb 2021 14:48:33 GMT):
Hi everyone, I am trying out revocation with von tails server. For demonstration purposes i am using current nodejs revocation sample which uses tails file stored locally in .indy-client folder I am referring this readme file https://github.com/bcgov/indy-tails-server. i have deployed this server with docker I ran nodejs revocation sample (not deleting wallet) sample failed to revoke credential. So to test server i tried sending PUT request with POSTMAN to von server but i am getting Revocation registry ID is not valid:GcjVZQ... Can anyone please help me here?

abuaesh (Sat, 20 Feb 2021 16:05:08 GMT):
Just figured that the timestamp shouldn't be the exact current time of accepting the TAA, to reduce corratability risk. So, the timestamp should be the time of the start of the day of accepting the TAA. More about it here: https://sovrin.org/preparing-for-the-sovrin-transaction-author-agreement/ "Similarly, the application should not provide the time the user accepts, but only the day. Correlation risk can be further reduced by tracking a different acceptance date per DID on the ledger, or regularly asking the user to reaccept the agreement in order to change the acceptance date—although we acknowledge these approaches may increase the user’s friction when writing to the ledger."

abuaesh (Sat, 20 Feb 2021 16:36:50 GMT):
Now this: "op":"REJECT","reason":"client request invalid: UnauthorizedClientRequest('The action is forbidden',)" I get it when this function is called: ledger.append_txn_author_agreement_acceptance_to_request Here is what my code looks like now: print("---------------------------------TAA PROCEDURE------------------------------------") # 1. Get Latest TAA from ledger taa_request = await ledger.build_get_txn_author_agreement_request(issuer['did'], None) latest_taa = await ledger.sign_and_submit_request(issuer['pool'], issuer['wallet'], issuer['did'], taa_request) print("\n\nStep A - TAA Retrieved: ") print(latest_taa) # 2. Get Acceptance Mechanisms List(AML) acceptance_mechanism_request = await ledger.build_get_acceptance_mechanisms_request(issuer['did'], None, None) aml = await ledger.sign_and_submit_request(issuer['pool'], issuer['wallet'], issuer['did'], acceptance_mechanism_request) print("\n\nStep B - AML Retrieved: ") print(json.loads(aml)['result']['data']['aml']) # 3. Append acceptance to your request latest_taa = json.loads(latest_taa) taa_digest = latest_taa['result']['data']['digest'] #d = datetime.datetime.now() #timestamp = time.mktime(d.timetuple()) aml = json.loads(aml) timestamp = aml['result']['txnTime'] timestamp = int (timestamp/(60*60*24)) #Need to set timestamp to beginning of the day to reduce corratability risk timestamp = timestamp*60*60*24 #return back to POSIX timestamp print (timestamp) # Values for Txn Author Agreement acceptance mechanism is one of: # ['at_submission', 'click_agreement', 'for_session', 'on_file', 'product_eula', 'service_agreement', 'wallet_agreement'] request = await ledger.append_txn_author_agreement_acceptance_to_request(request, None, None, taa_digest, 'at_submission', timestamp) request = await ledger.sign_and_submit_request(issuer['pool'], issuer['wallet'], issuer['did'], request) print("\n\nStep C - TAA Appended to your Request: ") print(request) # 4. Return appended request return request

abuaesh (Sat, 20 Feb 2021 16:36:50 GMT):
Now this: "op":"REJECT","reason":"client request invalid: UnauthorizedClientRequest('The action is forbidden',)" I get it when this function is called: ledger.append_txn_author_agreement_acceptance_to_request Here is what my code looks like now: print("---------------------------------TAA PROCEDURE------------------------------------") # 1. Get Latest TAA from ledger taa_request = await ledger.build_get_txn_author_agreement_request(issuer['did'], None) latest_taa = await ledger.sign_and_submit_request(issuer['pool'], issuer['wallet'], issuer['did'], taa_request) print("\n\nStep A - TAA Retrieved: ") print(latest_taa) # 2. Get Acceptance Mechanisms List(AML) acceptance_mechanism_request = await ledger.build_get_acceptance_mechanisms_request(issuer['did'], None, None) aml = await ledger.sign_and_submit_request(issuer['pool'], issuer['wallet'], issuer['did'], acceptance_mechanism_request) print("\n\nStep B - AML Retrieved: ") print(json.loads(aml)['result']['data']['aml']) # 3. Append acceptance to your request latest_taa = json.loads(latest_taa) taa_digest = latest_taa['result']['data']['digest'] #d = datetime.datetime.now() #timestamp = time.mktime(d.timetuple()) aml = json.loads(aml) timestamp = aml['result']['txnTime'] timestamp = int (timestamp/(60 * 60 * 24)) #Need to set timestamp to beginning of the day to reduce corratability risk timestamp = timestamp * 60 * 60 * 24 #return back to POSIX timestamp print (timestamp) # Values for Txn Author Agreement acceptance mechanism is one of: # ['at_submission', 'click_agreement', 'for_session', 'on_file', 'product_eula', 'service_agreement', 'wallet_agreement'] request = await ledger.append_txn_author_agreement_acceptance_to_request(request, None, None, taa_digest, 'at_submission', timestamp) request = await ledger.sign_and_submit_request(issuer['pool'], issuer['wallet'], issuer['did'], request) print("\n\nStep C - TAA Appended to your Request: ") print(request) # 4. Return appended request return request

abuaesh (Sat, 20 Feb 2021 16:36:50 GMT):
Now this: "op":"REJECT","reason":"client request invalid: UnauthorizedClientRequest('The action is forbidden',)" I get it when this function is called: ledger.append_txn_author_agreement_acceptance_to_request Here is what my code looks like now: print("---------------------------------TAA PROCEDURE------------------------------------") # 1. Get Latest TAA from ledger taa_request = await ledger.build_get_txn_author_agreement_request(issuer['did'], None) latest_taa = await ledger.sign_and_submit_request(issuer['pool'], issuer['wallet'], issuer['did'], taa_request) print("\n\nStep A - TAA Retrieved: ") print(latest_taa) # 2. Get Acceptance Mechanisms List(AML) acceptance_mechanism_request = await ledger.build_get_acceptance_mechanisms_request(issuer['did'], None, None) aml = await ledger.sign_and_submit_request(issuer['pool'], issuer['wallet'], issuer['did'], acceptance_mechanism_request) print("\n\nStep B - AML Retrieved: ") print(json.loads(aml)['result']['data']['aml']) # 3. Append acceptance to your request latest_taa = json.loads(latest_taa) taa_digest = latest_taa['result']['data']['digest'] aml = json.loads(aml) timestamp = aml['result']['txnTime'] timestamp = int (timestamp/(60 * 60 * 24)) #Need to set timestamp to beginning of the day to reduce corratability risk timestamp = timestamp * 60 * 60 * 24 #return back to POSIX timestamp print (timestamp) request = await ledger.append_txn_author_agreement_acceptance_to_request(request, None, None, taa_digest, 'at_submission', timestamp) request = await ledger.sign_and_submit_request(issuer['pool'], issuer['wallet'], issuer['did'], request) print("\n\nStep C - TAA Appended to your Request: ") print(request) # 4. Return appended request return request

abuaesh (Sat, 20 Feb 2021 16:57:02 GMT):
One year later, I am struggling with this. But I think you can take a look at my thread with Lynn down there.

abuaesh (Sat, 20 Feb 2021 17:03:18 GMT):
IT WORKED!!! I can see my schema on indyscan. That's why it is giving me forbidden action-because the schema is already written to the ledger!!! I can't believe it's finally working... alhamdulillah! I thought I will never get there! Thanks A LOT Lynn for your help! @lynn.bendixsen

Kshitij 7 (Mon, 22 Feb 2021 14:31:03 GMT):
Has joined the channel.

Kshitij 7 (Mon, 22 Feb 2021 14:31:04 GMT):
Hi Everyone, just cloned indy-sdk, completed all pre-requisites and while running negotiateProof.js in indy-sdk/docs/how-tos/negotiate-proof/nodejs - getting CommonInvalidState error. Any idea on how to resolve it ? The error is not descriptive enough to know more about the issue

Kshitij 7 (Mon, 22 Feb 2021 14:31:04 GMT):
Hi Everyone, just cloned indy-sdk, completed all pre-requisites and while running negotiateProof.js in indy-sdk/docs/how-tos/negotiate-proof/nodejs - getting CommonInvalidState error. This is coming at step# 8 at the stage of prover storing credentials. Any idea on how to resolve it ? The error is not descriptive enough to know more about the issue

Kshitij 7 (Mon, 22 Feb 2021 14:31:04 GMT):
Hi Everyone, just cloned indy-sdk, completed all pre-requisites and while running negotiateProof.js in indy-sdk/docs/how-tos/negotiate-proof/nodejs - getting CommonInvalidStructure error. This is coming at step# 8 at the stage of prover storing credentials. Any idea on how to resolve it ? The error is not descriptive enough to know more about the issue

Kshitij 7 (Mon, 22 Feb 2021 14:31:04 GMT):
Hi Everyone, just cloned indy-sdk, completed all pre-requisites and while running negotiateProof.js in indy-sdk/docs/how-tos/negotiate-proof/nodejs - getting CommonInvalidState error. This is coming at step# 8 at the stage of prover storing credentials. Any idea on how to resolve it ? The error is not descriptive enough to know more about the issue

abuaesh (Tue, 23 Feb 2021 17:23:01 GMT):
How do we know that the credential offer goes ONLY to the designated person. What if the offer(or a copy of it) falls in the wrong hands? They will still be able to generate a credential request, and get issued the credential(unless if the issuer keeps a list somewhere of DIDs that are legible to this offer). Is there something I am missing in the steps because I can't find something that links the offer to its targeted recipient.

swcurran (Tue, 23 Feb 2021 18:50:04 GMT):
The "best practices" in that is that you establish a DIDComm connection with an entity, and then use that channel to get some sort of proof of who they are. If possible, you would use the channel to do a proof request. With that done, binding the channel to the proof, you would use the same channel to issue the credential via the offer-request-issue sequence in the issue-credential protocol. Note as well that with AnonCreds (assuming you are using that), in the Offer and Request there are crypto graphic elements that bind the two parties to the credential. So when you issue the credential to the entity, you are binding it to that entity, and they will be the only one that can present the credential.

mmaashif (Wed, 24 Feb 2021 07:02:09 GMT):
Has joined the channel.

mmaashif (Wed, 24 Feb 2021 07:02:10 GMT):
Hi! I want to develop a Identity Management system, something that eliminates using National ID/Birth Reg card/driving license for seperate purpose. One single identity will work for all. Now can this be build on Indy? if can then what would be the architecture like?

mccown (Thu, 25 Feb 2021 05:36:48 GMT):
The Identity Implementer's WG is meeting on Thursday (Feb 25th @ 9am MT / 4pm UTC). In this call, we will review updates from several industry WGs, which you may have missed this week. Lohan Spies (CEO, did:x) will provide a presentation of tools & methods for monitoring the health and security of Indy-based ledgers. Here are links to the Wiki and Zoom meeting: Wiki: https://wiki.hyperledger.org/display/IWG/2021-02-25+:+Identity+Implementers+WG+Call Zoom: https://zoom.us/my/hyperledger.community.3?pwd=UE90WHhEaHRqOGEyMkV3cldKa2d2dz09

Shubham95 (Thu, 25 Feb 2021 06:48:49 GMT):
I am giving the path of genesis file for von-network in this I am getting module plenum not installed and when I am installing it using pip it shows error (error: command 'x86_64-linux-gnu-gcc' failed with exit status 1).

Shubham95 (Thu, 25 Feb 2021 06:48:49 GMT):
I am giving the path of genesis file for von-network in this I am getting module plenum not installed and when I am installing it using pip it shows error (error: command 'x86_64-linux-gnu-gcc' failed with exit status 1). Can anyone please help.

swcurran (Thu, 25 Feb 2021 15:41:30 GMT):
That's not the role of Indy. You don't want to put data about individuals, even identifiers and public keys on any blockchain. What you can do with Indy is issue a verifiable credential to every citizen that allows them to prove their National ID without any need to have verifiers integrating with the National ID service. The identifier does not need to be given to the verifiers for many (most?) use cases -- they just need to know the the person is a citizen or resident.

mvaibhav (Fri, 26 Feb 2021 06:03:58 GMT):
Has joined the channel.

dgt1nsty (Sat, 27 Feb 2021 12:37:17 GMT):
Hi everyone, I was reading indy-plenum doc and I didn't get what these statements saying "Client-to-node and node-to-node communication happens on CurveZMQ. The codebase has an abstraction called Stack that manages communication. It has several variants which offer different features." I understand that messages are send back and forth by peers' corresponding transient public keys obtained using curvezmq and are generated by curve-keygen. But what does following statement is saying "codebase has an abstraction called Stack that manages communication". What Stack stands for in this context and how it manages the communication? Can you please someone eleborate this statement so that I can make more sense.

nileshv (Tue, 02 Mar 2021 13:21:22 GMT):
Has joined the channel.

Lucampelli (Tue, 02 Mar 2021 18:46:29 GMT):
Hello! Are issued credentials stored also in the issuer's wallet? I mean, let's say Faber University Issued a Graduated Credential for me, will it also be stored in it's wallet? Or it depends on the implementation?

swcurran (Tue, 02 Mar 2021 19:25:20 GMT):
Depends on the implementation. In ACA-Py, there are three options for issued credential exchange record (which would contain the info in the credential). (1) Delete the record entirely, (2) keep the full exchange record in the wallet, (3) keep a minimal record that just contains just enough about the issuance to be able to revoke it using the ID of the record. Regardless, it is up to the controller to, for example, link the credential exchange ID to the entity your system that represents the holder. For example, Faber would keep in it's system the ID of the student and (at least) the ID the Credential Exchange ID so that if it needs to revoke the credential for the student it can find the information necessary to do so.

vineeta (Wed, 03 Mar 2021 04:24:35 GMT):
When the verifier creates revRegs json(proof of non revocation) to be sent to verify function timestamp added as a key to revRegs json What timetamp is expected here? https://www.npmjs.com/package/indy-sdk/v/1.16.0rev verifierVerifyProof ( proofRequest, proof, schemas, credentialDefsJsons, revRegDefs, revRegs ) -> valid revRegs: Json - all revocation registries json participating in the proof { "rev_reg_def1_id": { "timestamp1": , "timestamp2": , }, "rev_reg_def2_id": { "timestamp3": }, "rev_reg_def3_id": { "timestamp4": }, }

vineeta (Wed, 03 Mar 2021 04:24:35 GMT):
When the verifier creates revRegs json(proof of non revocation) to be sent to verify function timestamp added as a key to revRegs json What timetamp is expected here? https://www.npmjs.com/package/indy-sdk/v/1.16.0 verifierVerifyProof ( proofRequest, proof, schemas, credentialDefsJsons, revRegDefs, revRegs ) -> valid revRegs: Json - all revocation registries json participating in the proof { "rev_reg_def1_id": { "timestamp1": , "timestamp2": , }, "rev_reg_def2_id": { "timestamp3": }, "rev_reg_def3_id": { "timestamp4": }, }

vineeta (Wed, 03 Mar 2021 04:24:35 GMT):
When the verifier creates revRegs json(proof of non revocation) to be sent to verify function timestamp added as a key to revRegs json What timetamp is expected here? https://www.npmjs.com/package/indy-sdk/v/1.16.0 Should this timestamp match with deltatimestamp sent in proof? verifierVerifyProof ( proofRequest, proof, schemas, credentialDefsJsons, revRegDefs, revRegs ) -> valid revRegs: Json - all revocation registries json participating in the proof { "rev_reg_def1_id": { "timestamp1": , "timestamp2": , }, "rev_reg_def2_id": { "timestamp3": }, "rev_reg_def3_id": { "timestamp4": }, }

vineeta (Wed, 03 Mar 2021 04:24:35 GMT):
When the verifier creates revRegs json(proof of non revocation) to be sent to verify function timestamp added as a key to revRegs json What timetamp is expected here? https://www.npmjs.com/package/indy-sdk/v/1.16.0 Should this timestamp match with deltatimestamp sent in proof? *verifierVerifyProof ( proofRequest, proof, schemas, credentialDefsJsons, revRegDefs, revRegs ) -> valid* revRegs: Json - all revocation registries json participating in the proof { "rev_reg_def1_id": { "timestamp1": , "timestamp2": , }, "rev_reg_def2_id": { "timestamp3": }, "rev_reg_def3_id": { "timestamp4": }, }

vineeta (Wed, 03 Mar 2021 04:24:35 GMT):
When the verifier creates revRegs json(proof of non revocation by fetching delta with current timestamp for eg) to be sent to verify function timestamp added as a key to revRegs json What timetamp(timestamp1,timestamp2) is expected here? If timestamp returned by get current delta function is used then verification fails with common invalid structure. https://www.npmjs.com/package/indy-sdk/v/1.16.0 Should this timestamp match with deltatimestamp sent in proof? *verifierVerifyProof ( proofRequest, proof, schemas, credentialDefsJsons, revRegDefs, revRegs ) -> valid* revRegs: Json - all revocation registries json participating in the proof { "rev_reg_def1_id": { "timestamp1": , "timestamp2": , }, "rev_reg_def2_id": { "timestamp3": }, "rev_reg_def3_id": { "timestamp4": }, }

vineeta (Wed, 03 Mar 2021 04:24:35 GMT):
When the verifier creates revRegs json(proof of non revocation) to be sent to verify function timestamp added as a key to revRegs json What timetamp(timestamp1,timestamp2) is expected here? https://www.npmjs.com/package/indy-sdk/v/1.16.0 Should this timestamp match with deltatimestamp sent in proof? *verifierVerifyProof ( proofRequest, proof, schemas, credentialDefsJsons, revRegDefs, revRegs ) -> valid* revRegs: Json - all revocation registries json participating in the proof { "rev_reg_def1_id": { "timestamp1": , "timestamp2": , }, "rev_reg_def2_id": { "timestamp3": }, "rev_reg_def3_id": { "timestamp4": }, }

vineeta (Wed, 03 Mar 2021 04:24:35 GMT):
When the verifier creates revRegs json(proof of non revocation by fetching delta with current timestamp for eg) to be sent to verify function timestamp added as a key to revRegs json What timetamp(timestamp1,timestamp2) is expected here? https://www.npmjs.com/package/indy-sdk/v/1.16.0 Should this timestamp match with deltatimestamp sent in proof? *verifierVerifyProof ( proofRequest, proof, schemas, credentialDefsJsons, revRegDefs, revRegs ) -> valid* revRegs: Json - all revocation registries json participating in the proof { "rev_reg_def1_id": { "timestamp1": , "timestamp2": , }, "rev_reg_def2_id": { "timestamp3": }, "rev_reg_def3_id": { "timestamp4": }, }

vineeta (Wed, 03 Mar 2021 04:24:35 GMT):
When the verifier creates revRegs json(proof of non revocation by fetching delta with current timestamp for eg) to be sent to verify function timestamp added as a key to revRegs json What timetamp(timestamp1,timestamp2) is expected here? If timestamp of current delta is used then verification fails with common invalid structure. https://www.npmjs.com/package/indy-sdk/v/1.16.0 Should this timestamp match with deltatimestamp sent in proof? *verifierVerifyProof ( proofRequest, proof, schemas, credentialDefsJsons, revRegDefs, revRegs ) -> valid* revRegs: Json - all revocation registries json participating in the proof { "rev_reg_def1_id": { "timestamp1": , "timestamp2": , }, "rev_reg_def2_id": { "timestamp3": }, "rev_reg_def3_id": { "timestamp4": }, }

swcurran (Wed, 03 Mar 2021 15:28:39 GMT):
Sorry for the delay in respsonding. Have you read this document -- particular the part about the "interval" meaning in the proof, and the interval meaning in the query to the ledger? The differences between those create a lot of confusion and may be what you are seeing here: https://github.com/hyperledger/indy-hipe/tree/master/text/0011-cred-revocation

swcurran (Wed, 03 Mar 2021 15:30:18 GMT):
Basically, to create the proof of non-revocation, you have to have all of the deltas from ledger -- you have to know all of the inactive and revoked credentials so a witness can be constructed. If you don't get all of those, you can't make the proof.

vineeta (Wed, 03 Mar 2021 15:53:06 GMT):
yes I had checked this document wherein verifier sents the interval in non-revoked field but it doesnt explain about verification process. I had tried a scenario where proof is created by prover using now() as current timestamp for fetching delta (similar to what is done in indy-sdk samples/nodejs/revocationScenario) & at verifier this revReg object is created using timestamp from identifier as key So not clear on this part what does that timestamp mean. In this example delta is fetched based on current timestamp but timestamp key is from identifiers(timestampOfProof) https://github.com/hyperledger/indy-sdk/blob/8c669cff434f5d87e7a2e2e192423c67a9df0b33/samples/nodejs/src/anoncredsRevocationScenario.js#L418 https://github.com/hyperledger/indy-sdk/blob/8c669cff434f5d87e7a2e2e192423c67a9df0b33/samples/nodejs/src/anoncredsRevocationScenario.js#L464

vineeta (Wed, 03 Mar 2021 15:53:06 GMT):
yes I had checked this document and wherein verifier sents the interval in non-revoked field but it doesnt explain about verification process. I had tried a scenario where proof is created by prover using now() as current timestamp for fetching delta (similar to what is done in indy-sdk samples/nodejs/revocationScenario) & at verifier this revReg object is created using timestamp from identifier as key So not clear on this part what does that timestamp mean. In this example delta is fetched based on current timestamp but timestamp key is from identifiers(timestampOfProof) https://github.com/hyperledger/indy-sdk/blob/8c669cff434f5d87e7a2e2e192423c67a9df0b33/samples/nodejs/src/anoncredsRevocationScenario.js#L418 https://github.com/hyperledger/indy-sdk/blob/8c669cff434f5d87e7a2e2e192423c67a9df0b33/samples/nodejs/src/anoncredsRevocationScenario.js#L464

vineeta (Wed, 03 Mar 2021 15:53:06 GMT):
yes I had checked this document wherein verifier sents the interval in non-revoked field but it doesnt explain about verification process. I had tried a scenario where proof is created by prover using now() as current timestamp for fetching delta (similar to what is done in indy-sdk samples/nodejs/revocationScenario) & at verifier this revReg object is created using timestamp from identifier as key So not clear on this part what does that timestamp mean. In this example delta is fetched based on current timestamp but timestamp key is from identifiers(timestampOfProof) https://github.com/hyperledger/indy-sdk/blob/8c669cff434f5d87e7a2e2e192423c67a9df0b33/samples/nodejs/src/anoncredsRevocationScenario.js#L418 https://github.com/hyperledger/indy-sdk/blob/8c669cff434f5d87e7a2e2e192423c67a9df0b33/samples/nodejs/src/anoncredsRevocationScenario.js#L464

vineeta (Wed, 03 Mar 2021 15:57:06 GMT):
Reading through that document we thought that verifier creates revRegs{} for same timestamp as used by prover. Is this correct?

Tanvi (Wed, 03 Mar 2021 17:59:09 GMT):
Hi everyone, Is there a way to configure tails server url in revocation registry definition using indy-sdk. tails-location takes file system location by default. We were able to add location using Aries but can't figure out how to do it with indy SDK. Can anyone please help here

andrew.whitehead (Wed, 03 Mar 2021 18:23:44 GMT):
I don't think the SDK provides a method to do that. Currently you have to manually edit the JSON of the revocation registry definition

vineeta (Wed, 03 Mar 2021 18:35:23 GMT):
Yes. You can update revRegDef['value']['tailsLocation'] before building request here. indy.buildRevocRegDefRequest(did, revRegDef);

vineeta (Wed, 03 Mar 2021 18:35:23 GMT):
You can update revRegDef['value']['tailsLocation'] before building request here. indy.buildRevocRegDefRequest(did, revRegDef);

Vp10 (Fri, 05 Mar 2021 16:52:10 GMT):
Has joined the channel.

Shubham95 (Sun, 07 Mar 2021 06:27:09 GMT):
Please help me resolve this error : ImportError: cannot import name 'get_pool_genesis_txn_path' from 'utils' I am stuck at it for some time

victoryeo (Mon, 08 Mar 2021 09:15:27 GMT):
Has joined the channel.

victoryeo (Mon, 08 Mar 2021 09:15:28 GMT):
Hello, i am new to Indy. Indy blockchain is a public permissioned blockchain, is it true that everyone can read the data on the blockchain without authentication or authorization ?

victoryeo (Mon, 08 Mar 2021 09:16:44 GMT):
So any user identity data stored on Indy blockchain can be read by any one ??

WadeBarnes (Mon, 08 Mar 2021 13:44:53 GMT):
@victoryeo, Unlike other block chains, the credentials (the identity data itself) is not stored on the block chain.

WadeBarnes (Mon, 08 Mar 2021 13:44:53 GMT):
@victoryeo, Unlike other block chains, the credentials (the identity data itself) is not stored on the block chain. The data is issuer to the holder, who then stores the data in a secure wallet which they manage.

WadeBarnes (Mon, 08 Mar 2021 13:44:53 GMT):
@victoryeo, Unlike other block chains, the credentials (the identity data itself) is not stored on the block chain. The data is issued to the holder, who then stores the data in a secure wallet which they manage.

Divyansh10 (Mon, 08 Mar 2021 14:55:43 GMT):
Has joined the channel.

Divyansh10 (Mon, 08 Mar 2021 14:55:44 GMT):
Hello everone, i am very new to blockchain development..

Divyansh10 (Mon, 08 Mar 2021 14:57:36 GMT):
I have to develop Self sovreign identity for my internship project and while searching about it came across Hyperledger Indy.. so can anyone guide me where to start from..what to learn or what not to..anything will be useful

WadeBarnes (Mon, 08 Mar 2021 15:09:28 GMT):
https://github.com/hyperledger/aries-cloudagent-python https://www.edx.org/course/identity-in-hyperledger-aries-indy-and-ursa https://www.edx.org/course/becoming-a-hyperledger-aries-developer

Divyansh10 (Mon, 08 Mar 2021 15:11:23 GMT):
thank you so much

victoryeo (Tue, 09 Mar 2021 01:54:44 GMT):
thank you

mccown (Wed, 10 Mar 2021 20:58:53 GMT):
The Identity Implementer's WG is meeting on Thursday (Mar 11th @ 9am MT / 4pm UTC). In this call, we will review updates from several industry WGs, which you may have missed this week. Renata Toktar (Evernym) will give a presentation / demo on a new process called "ledger freeze" and how that works with Indy ledgers. Here are links to the Wiki and Zoom meeting: Wiki: https://wiki.hyperledger.org/display/IWG/2021-03-11+%3A+Identity+Implementers+WG+Call Zoom: https://zoom.us/my/hyperledger.community.3?pwd=UE90WHhEaHRqOGEyMkV3cldKa2d2dz09

BrianK7978 (Wed, 10 Mar 2021 21:46:39 GMT):
Has joined the channel.

mwherman2000 (Thu, 11 Mar 2021 14:52:27 GMT):

Clipboard - March 11, 2021 7:52 AM

mwherman2000 (Thu, 11 Mar 2021 14:52:27 GMT):
Two years ago today, John Jordan and the BCGov clan hosted their Hyperledger Indy Bootcamp...
Clipboard - March 11, 2021 7:52 AM

mwherman2000 (Thu, 11 Mar 2021 14:52:27 GMT):
Two years ago today, John Jordan and the BCGov clan hosted their Hyperledger Indy Bootcamp in Vancouver...
Clipboard - March 11, 2021 7:52 AM

mwherman2000 (Thu, 11 Mar 2021 14:52:27 GMT):
Two years ago today, John Jordan, the VON, and the BCGov clan hosted their Hyperledger Indy Bootcamp in Vancouver...
Clipboard - March 11, 2021 7:52 AM

mwherman2000 (Thu, 11 Mar 2021 14:52:31 GMT):
Two years ago today, John Jordan and the BCGov clan hosted their Hyperledger Indy Bootcamp...

mwherman2000 (Thu, 11 Mar 2021 14:54:22 GMT):
..and @TelegramSam described the #OpenToInnovationPrinciple: https://hyperonomy.com/2019/03/12/internet-protocols-and-standards-not-only-need-to-be-open-but-more-importantly-open-to-innovation/

haardikkk (Thu, 11 Mar 2021 22:08:43 GMT):
Has left the channel.

Shubham95 (Fri, 12 Mar 2021 00:44:13 GMT):
Can anyone please help with error Exception: Error: 500: Tails file for rev reg Tiw1AYNCdUfKDzSZFz9Kk4:4:Tiw1AYNCdUfKDzSZFz9Kk4:3:CL:102:faber.agent.degree_schema:CL_ACCUM:61a1810d-6d49-453f-afa5-4c36deace0db failed to upload: Exceeded maximum put attempts

Shubham95 (Fri, 12 Mar 2021 00:44:13 GMT):
Can anyone please help with error Exception: Error: 500: Tails file for rev reg Tiw1AYNCdUfKDzSZFz9Kk4:4:Tiw1AYNCdUfKDzSZFz9Kk4:3:CL:102:faber.agent.degree_schema:CL_ACCUM:61a1810d-6d49-453f-afa5-4c36deace0db failed to upload: Exceeded maximum put attempts I am stuck at it for quite sometime, I am using tails base server with acapy and getting this error on virtual machine

kukgini (Fri, 12 Mar 2021 09:13:23 GMT):
Can I run a read-only indy node and join the Sovrin network? If not, do you have any plans to do so?

kukgini (Fri, 12 Mar 2021 09:33:27 GMT):
Still, supporting `Untrusted Observers` is yet?

kukgini (Fri, 12 Mar 2021 09:33:27 GMT):
Still, supporting `Untrusted Observers` is yet? It seems that the design for it has been over 3 years. Is there any history of not implementing this in the meantime?

HighBrow (Sun, 14 Mar 2021 08:16:18 GMT):
Has joined the channel.

rome-sandra (Sun, 14 Mar 2021 19:29:51 GMT):
Has joined the channel.

coderatwork7 (Mon, 15 Mar 2021 08:52:35 GMT):
Has joined the channel.

ParminderParmar (Tue, 16 Mar 2021 00:02:10 GMT):
Has joined the channel.

ParminderParmar (Tue, 16 Mar 2021 00:02:11 GMT):
Hi Guys, I have a few questions about indy. if someone can help would be great. How can you connect your indy node to sovrin network? if yes, will sovrin charge for connectivity? If I will issue a credential using evernym, will I be able to load that credential in any wallet like Trinsic ? What would you recommend in terms of building applications - indy connected to sovrin, trinsic or evernym ? If i will spin my own network of indy, are there any wallets that will work automatically with it?

swcurran (Tue, 16 Mar 2021 03:39:34 GMT):
Answered on the indy-sdk channel.

asadhayat (Tue, 16 Mar 2021 11:49:20 GMT):
Has joined the channel.

chrisconway (Tue, 16 Mar 2021 16:43:37 GMT):
Has joined the channel.

chrisconway (Tue, 16 Mar 2021 16:43:38 GMT):
INFO|node.py|Node4 joined nodes set() but status is 2

asadhayat (Wed, 17 Mar 2021 15:00:41 GMT):
Hi, I am following https://hyperledger-indy.readthedocs.io/projects/sdk/en/latest/docs/getting-started/run-getting-started.html

asadhayat (Wed, 17 Mar 2021 15:01:39 GMT):
When i run `docker-compose up` in `indy-sdk/docs/getting-started`, i get this error after a while

asadhayat (Wed, 17 Mar 2021 15:02:06 GMT):
``` Successfully tagged indy_pool:latest WARNING: Image for service indy_pool was built because it did not already exist. To rebuild this image you must use `docker-compose build` or `docker-compose up --build`. Building jupyter Step 1/8 : FROM ubuntu:16.04 ---> 8185511cd5ad Step 2/8 : RUN useradd -ms /bin/bash indy ---> Using cache ---> 40ad383a9cf6 Step 3/8 : RUN apt-get update -y && apt-get install -y wget python3.5 python3-pip python-setuptools apt-transport-https ca-certificates software-properties-common ---> Using cache ---> 57ecf5e57640 Step 4/8 : WORKDIR /home/indy ---> Using cache ---> b5d85444635c Step 5/8 : RUN pip3 install -U pip ipython-notebook ipython==7.9 setuptools jupyter python3-indy==1.11.0 ---> Running in 4a5732f9a43b Collecting pip Downloading https://files.pythonhosted.org/packages/fe/ef/60d7ba03b5c442309ef42e7d69959f73aacccd0d86008362a681c4698e83/pip-21.0.1-py3-none-any.whl (1.5MB) Collecting ipython-notebook Could not find a version that satisfies the requirement ipython-notebook (from versions: ) No matching distribution found for ipython-notebook You are using pip version 8.1.1, however version 21.0.1 is available. You should consider upgrading via the 'pip install --upgrade pip' command. ERROR: Service 'jupyter' failed to build: The command '/bin/sh -c pip3 install -U pip ipython-notebook ipython==7.9 setuptools jupyter python3-indy==1.11.0' returned a non-zero code: 1 asad@localhost:~/blockchain/indy/indy-sdk/docs/getting-started> ```

asadhayat (Wed, 17 Mar 2021 15:02:58 GMT):
I am using OpenSUSE running in virtualbox 6.1

HighBrow (Fri, 19 Mar 2021 12:41:54 GMT):
Hi everyone, although it is widely agreed that we don’t put personal data of individuals on Indy, in case someone ends up putting personal data either accidentally or maliciously, is there a workaround to somehow take care of it? Also, can “Right to be forgotten” be somehow managed while using Indy?

SeriousM 4 (Sat, 20 Mar 2021 11:15:54 GMT):
Has joined the channel.

deas (Mon, 22 Mar 2021 16:59:38 GMT):
I'd _really_ try to avoid a scenario where that's possible, personally. What are you using Indy for that enables someone to write personal data to the chain?

swcurran (Mon, 22 Mar 2021 17:25:20 GMT):
On Sovrin, there is an agreement with an Endorser and with an Author that explicitly excludes writing personal data to the ledger as a way to avoid the need for that. The only potential PII that could go Indy is an identifier (with nothing about that identifier controller/subject), and on Sovrin, only business entities where such an identifier is NOT personal are (legally) allowed to write to the ledger.

swcurran (Mon, 22 Mar 2021 17:26:04 GMT):
If those agreements are followed, there is no PII on the ledger, and not a need to be forgotten.

HighBrow (Tue, 23 Mar 2021 06:53:34 GMT):
@swcurran I agree that because of the legal agreement a business entity will not be writing any personal data on the ledger since that could have legal consequences. But in case of accidental writes, is there anything that could be done on the ledger?

HighBrow (Tue, 23 Mar 2021 07:00:27 GMT):
@deas I’m not enabling someone to write personal data on the chain. I am talking about the case when some personal data could actually land up on the ledger since as of now there are only legal implications that would discourage that to happen in the first place.

swcurran (Tue, 23 Mar 2021 13:51:43 GMT):
Correct. There has been a plan defined to "tombstone" data -- make it inaccessible from the network, but that has not been implemented. This link was posted to the feature overview: https://jira.hyperledger.org/browse/INDY-2082

Vp10 (Wed, 24 Mar 2021 11:17:01 GMT):
Is it possible to add tags directly with credential Id stored in wallet using addWalletRecordTags function from indy-sdk?

Vp10 (Wed, 24 Mar 2021 11:17:01 GMT):
Is it possible to add custom tags directly with credential Id stored in wallet using addWalletRecordTags function from indy-sdk?

Vp10 (Wed, 24 Mar 2021 11:17:01 GMT):
Is it possible to add custom tags to credential directly with credential Id stored in wallet using addWalletRecordTags function from indy-sdk?

Vp10 (Wed, 24 Mar 2021 11:17:01 GMT):
Is it possible to add custom tags to credential directly with its credential Id stored in wallet using addWalletRecordTags function from indy-sdk?

Vp10 (Wed, 24 Mar 2021 11:17:01 GMT):
Is it possible to add custom tags to credential stored in wallet with its credential Id using addWalletRecordTags function from indy-sdk?

mccown (Wed, 24 Mar 2021 19:31:37 GMT):
The Identity Implementer's WG is meeting tomorrow (Mar 25th @ 9am MT / 3pm UTC). In this call, we will review updates from several industry WGs, which you may have missed this week. Steve McCown (Anonyome Labs) will give a presentation on how to simplify creating language wrappers for Rust libraries to more easily support multiple platforms. Here are links to the Wiki and Zoom meeting: Wiki: https://wiki.hyperledger.org/display/IWG/2021-03-25+%3A+Identity+Implementers+WG+Call Zoom: https://zoom.us/my/hyperledger.community.3?pwd=UE90WHhEaHRqOGEyMkV3cldKa2d2dz09

nao (Fri, 26 Mar 2021 05:34:43 GMT):
Has joined the channel.

HighBrow (Sun, 28 Mar 2021 04:59:17 GMT):
Hi everyone, apart from supporting the cause of "Identity for All", are there any incentives for an organization to be a Steward in the Sovrin MainNet? Is any kind of transaction fees currently being shared with all the Stewards of the network?

HighBrow (Sun, 28 Mar 2021 04:59:17 GMT):
Hi everyone, apart from supporting the cause of "Identity for All", are there any incentives for an organization to be a Steward in the Sovrin MainNet? Is any kind of transaction fees currently being shared as a reward with all the Stewards of the network?

weiyaous (Sun, 28 Mar 2021 14:21:39 GMT):
Has joined the channel.

weiyaous (Sun, 28 Mar 2021 14:21:40 GMT):
Dear all, is there any tutorial or documentation for the indy-test-automation project? https://github.com/hyperledger/indy-test-automation

weiyaous (Sun, 28 Mar 2021 17:37:58 GMT):
Dear all, did anyone got this issue, while running test script from indy-test-automation/system/docker/prepare.sh. I think the problem is caused by calling indy-test-automation/system/docker/client/dockfile. I'll attach the screenshot. I'm wondering anyone can help me fix it. Thank you!

weiyaous (Sun, 28 Mar 2021 17:37:58 GMT):
Dear all, did anyone get this issue, while running test script from indy-test-automation/system/docker/prepare.sh. I think the problem is caused by calling indy-test-automation/system/docker/client/dockfile. I'll attach the screenshot. I'm wondering anyone can help me fix it. Thank you!

weiyaous (Sun, 28 Mar 2021 17:38:09 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=5JPHbkai8oWdyK424)
Clipboard - March 28, 2021 1:38 PM

WadeBarnes (Mon, 29 Mar 2021 12:10:38 GMT):
@askolesov, Any pointers ^?

usingetechnology (Wed, 31 Mar 2021 22:36:47 GMT):
Has joined the channel.

usingetechnology (Wed, 31 Mar 2021 22:36:47 GMT):
Not sure which channel would be best for this question. I was attempting to stand up a DID resolver on Openshift 4. (Note that I am abandoning this and waiting on resolver being included in act-py; however, if there is an easy answer, that'd be great). This appears to be one of the many issues related to openshift and the image user accounts, but hoping I can get pointed to something more concrete that I can address. I know the files for my genesis transactions are on the image and in the location specified by my config. I have built and run the same image outside of Openshift and it works. But this is my error stack: ```m[qtp1262822392-12] WARN servlet.ResolveServlet (doGet) - Driver reported for did:sov:WRfXPg8dantKVubE3HX8pw: Cannot create pool config "_": org.hyperledger.indy.sdk.IOException: An IO error occurred. uniresolver.ResolutionException: Cannot create pool config "_": org.hyperledger.indy.sdk.IOException: An IO error occurred. at uniresolver.driver.did.sov.DidSovDriver.openIndy(DidSovDriver.java:334) at uniresolver.driver.did.sov.DidSovDriver.resolve(DidSovDriver.java:123) at uniresolver.driver.servlet.ResolveServlet.doGet(ResolveServlet.java:71) at javax.servlet.http.HttpServlet.service(HttpServlet.java:687) at javax.servlet.http.HttpServlet.service(HttpServlet.java:790) at org.eclipse.jetty.servlet.ServletHolder.handle(ServletHolder.java:865) at org.eclipse.jetty.servlet.ServletHandler$CachedChain.doFilter(ServletHandler.java:1623) at org.eclipse.jetty.websocket.server.WebSocketUpgradeFilter.doFilter(WebSocketUpgradeFilter.java:214) at org.eclipse.jetty.servlet.ServletHandler$CachedChain.doFilter(ServletHandler.java:1610) at org.eclipse.jetty.servlet.ServletHandler.doHandle(ServletHandler.java:540) at org.eclipse.jetty.server.handler.ScopedHandler.handle(ScopedHandler.java:146) at org.eclipse.jetty.security.SecurityHandler.handle(SecurityHandler.java:548) at org.eclipse.jetty.server.handler.HandlerWrapper.handle(HandlerWrapper.java:132) at org.eclipse.jetty.server.handler.ScopedHandler.nextHandle(ScopedHandler.java:257) at org.eclipse.jetty.server.session.SessionHandler.doHandle(SessionHandler.java:1701) at org.eclipse.jetty.server.handler.ScopedHandler.nextHandle(ScopedHandler.java:255) at org.eclipse.jetty.server.handler.ContextHandler.doHandle(ContextHandler.java:1345) at org.eclipse.jetty.server.handler.ScopedHandler.nextScope(ScopedHandler.java:203) at org.eclipse.jetty.servlet.ServletHandler.doScope(ServletHandler.java:480) at org.eclipse.jetty.server.session.SessionHandler.doScope(SessionHandler.java:1668) at org.eclipse.jetty.server.handler.ScopedHandler.nextScope(ScopedHandler.java:201) at org.eclipse.jetty.server.handler.ContextHandler.doScope(ContextHandler.java:1247) at org.eclipse.jetty.server.handler.ScopedHandler.handle(ScopedHandler.java:144) at org.eclipse.jetty.server.handler.ContextHandlerCollection.handle(ContextHandlerCollection.java:220) at org.eclipse.jetty.server.handler.HandlerCollection.handle(HandlerCollection.java:126) at org.eclipse.jetty.server.handler.HandlerWrapper.handle(HandlerWrapper.java:132) at org.eclipse.jetty.server.Server.handle(Server.java:502) at org.eclipse.jetty.server.HttpChannel.handle(HttpChannel.java:370) at org.eclipse.jetty.server.HttpConnection.onFillable(HttpConnection.java:267) at org.eclipse.jetty.io.AbstractConnection$ReadCallback.succeeded(AbstractConnection.java:305) at org.eclipse.jetty.io.FillInterest.fillable(FillInterest.java:103) at org.eclipse.jetty.io.ChannelEndPoint$2.run(ChannelEndPoint.java:117) at org.eclipse.jetty.util.thread.QueuedThreadPool.runJob(QueuedThreadPool.java:765) at org.eclipse.jetty.util.thread.QueuedThreadPool$2.run(QueuedThreadPool.java:683) at java.base/java.lang.Thread.run(Thread.java:834)```

asadhayat (Wed, 31 Mar 2021 23:58:07 GMT):
Can anyone help me with this error i posted in #aries channel. Resharing here cause i think aries might not be the right channel https://chat.hyperledger.org/channel/aries?msg=fkMWWT2XtH9mbkR7c

WadeBarnes (Thu, 01 Apr 2021 14:31:49 GMT):
@asadhayat, The part of your question regarding the ledger browser is best posted as an issue on https://github.com/bcgov/von-network

WadeBarnes (Thu, 01 Apr 2021 14:33:09 GMT):
The second part of your question is best posted on the #aries-cloudagent-python channel.

WadeBarnes (Thu, 01 Apr 2021 14:34:44 GMT):
As for tutorials, the most up-to-date ones for `` will be found here; https://github.com/hyperledger/aries-cloudagent-python/blob/main/docs/GettingStartedAriesDev/README.md

WadeBarnes (Thu, 01 Apr 2021 14:34:44 GMT):
As for tutorials, the most up-to-date ones for `/aries-cloudagent-python` will be found here; https://github.com/hyperledger/aries-cloudagent-python/blob/main/docs/GettingStartedAriesDev/README.md

WadeBarnes (Thu, 01 Apr 2021 14:34:44 GMT):
As for tutorials, the most up-to-date ones for `aries-cloudagent-python` will be found here; https://github.com/hyperledger/aries-cloudagent-python/blob/main/docs/GettingStartedAriesDev/README.md

chrisconway (Thu, 01 Apr 2021 18:21:02 GMT):
I know this is a bit late, but things like this are relatively common while doing the getting started. Please see this issue for a dockerfile that might run better: https://github.com/hyperledger/indy-sdk/issues/2347

echsecutor (Mon, 05 Apr 2021 20:25:50 GMT):
Has joined the channel.

rpobulic (Wed, 07 Apr 2021 10:56:19 GMT):
Has joined the channel.

mccown (Wed, 07 Apr 2021 20:30:24 GMT):
The Identity Implementer's WG is meeting tomorrow (Apr 8th @ 9am MT / 3pm UTC). In this call, we will review updates from several industry WGs, which you may have missed this week. Daniel Bachenheimer (Principal Director at Accenture) will be presenting on the Known Traveler Digital Identity (KTDI; World Economic Forum initiative) and how it uses SSI / Hyperledger Indy stack. Wiki: https://wiki.hyperledger.org/display/IWG/2021-04-8+%3A+Identity+Implementers+WG+Call Zoom: https://zoom.us/my/hyperledger.community.3?pwd=UE90WHhEaHRqOGEyMkV3cldKa2d2dz09

CryptoMuck (Thu, 08 Apr 2021 12:24:13 GMT):
Has joined the channel.

CryptoMuck (Thu, 08 Apr 2021 12:46:50 GMT):
Hello all, while studying the Indy project I wondered if there is some kind of verifiable encryption implemented or if there are plans to implement it? So that for instance it would be possible for a prover to encrypt an identifying attribute from a credential in form of ZKP under the usage of a public key from a third party, so that this third party could decrypt the attribute in case of (proven) misbehaviour by the prover and identify him. Thank you!

swcurran (Thu, 08 Apr 2021 14:21:32 GMT):
At this point -- no plans for that. The prover can put self-asserted claims into a presentation, so in theory, I think that could work.

CryptoMuck (Thu, 08 Apr 2021 16:10:06 GMT):
Thank you very much. Are accepted proofs currently stored by the verifier in some form?

swcurran (Thu, 08 Apr 2021 18:22:53 GMT):
That's up to the verifier -- and the regulations under which they operate. Minimizing what the verifier holds can reduce their liability.

anil_helvaci (Fri, 09 Apr 2021 07:36:30 GMT):
Has joined the channel.

anil_helvaci (Fri, 09 Apr 2021 07:36:45 GMT):
Hi everyone, I am trying build the indy-sdk postgres-plugin(https://github.com/hyperledger/indy-sdk/tree/master/experimental/plugins/postgres_storage) by following the instructions in README.md but I am having some problems. Hope someone can help me. I am trying to build each project in the following order: 1) indy-sdk/libindy 2) indy-sdk/cli 3) indy-sdk/experimental/plugins/postgres_storage But the error message is always the same; error: failed to run custom build command for libsodium-sys v0.0.16` Caused by: process didn't exit successfully: C:\Users\anilh\work\one37\postgres-plugin\indy-sdk\libindy\target\debug\build\libsodium-sys-2c407bb769129617\build-script-build (exit code: 101) --- stdout cargo:rerun-if-env-changed=SODIUM_LIB_DIR cargo:rerun-if-env-changed=SODIUM_STATIC cargo:rerun-if-env-changed=LIBSODIUM_NO_PKG_CONFIG cargo:rerun-if-env-changed=PKG_CONFIG cargo:rerun-if-env-changed=LIBSODIUM_STATIC cargo:rerun-if-env-changed=LIBSODIUM_DYNAMIC cargo:rerun-if-env-changed=PKG_CONFIG_ALL_STATIC cargo:rerun-if-env-changed=PKG_CONFIG_ALL_DYNAMIC cargo:rerun-if-env-changed=PKG_CONFIG_PATH_x86_64-pc-windows-msvc cargo:rerun-if-env-changed=PKG_CONFIG_PATH_x86_64_pc_windows_msvc cargo:rerun-if-env-changed=HOST_PKG_CONFIG_PATH cargo:rerun-if-env-changed=PKG_CONFIG_PATH cargo:rerun-if-env-changed=PKG_CONFIG_LIBDIR_x86_64-pc-windows-msvc cargo:rerun-if-env-changed=PKG_CONFIG_LIBDIR_x86_64_pc_windows_msvc cargo:rerun-if-env-changed=HOST_PKG_CONFIG_LIBDIR cargo:rerun-if-env-changed=PKG_CONFIG_LIBDIR cargo:rerun-if-env-changed=PKG_CONFIG_SYSROOT_DIR_x86_64-pc-windows-msvc cargo:rerun-if-env-changed=PKG_CONFIG_SYSROOT_DIR_x86_64_pc_windows_msvc cargo:rerun-if-env-changed=HOST_PKG_CONFIG_SYSROOT_DIR cargo:rerun-if-env-changed=PKG_CONFIG_SYSROOT_DIR --- stderr thread 'main' panicked at 'called Result::unwrap() on an Err value: "Failed to run \"pkg-config\" \"--libs\" \"--cflags\" \"libsodium\": The system cannot find the file specified. (os error 2)"', C:\Users\anilh\.cargo\registry\src\github.com-1ecc6299db9ec823\libsodium-sys-0.0.16\build.rs:20:47 note: run with RUST_BACKTRACE=1 environment variable to display a backtrace warning: build failed, waiting for other jobs to finish... error: build failed` I am not familiar with rust but "sodiumoxide = {version = "0.0.16"}" is present in Cargo.toml file in all directories. I hope someone knows how to build postgres-plugin correctly 🙏 . Or is there an alternative way to run a dotnet agent with postgres?

abax1 (Mon, 12 Apr 2021 21:43:42 GMT):
Has joined the channel.

abax1 (Mon, 12 Apr 2021 21:43:43 GMT):
I am also getting this issue. That new suggested dockerfile has helped but I now get this error: ``` getting_started | Traceback (most recent call last): getting_started | File "/usr/local/bin/jupyter-notebook", line 6, in getting_started | from notebook.notebookapp import main getting_started | File "/usr/local/lib/python3.5/dist-packages/notebook/notebookapp.py", line 81, in getting_started | from .services.kernels.kernelmanager import MappingKernelManager, AsyncMappingKernelManager getting_started | File "/usr/local/lib/python3.5/dist-packages/notebook/services/kernels/kernelmanager.py", line 18, in getting_started | from jupyter_client.session import Session getting_started | File "/usr/local/lib/python3.5/dist-packages/jupyter_client/__init__.py", line 6, in getting_started | from .asynchronous import AsyncKernelClient # noqa getting_started | File "/usr/local/lib/python3.5/dist-packages/jupyter_client/asynchronous/__init__.py", line 1, in getting_started | from .client import AsyncKernelClient # noqa getting_started | File "/usr/local/lib/python3.5/dist-packages/jupyter_client/asynchronous/client.py", line 6, in getting_started | from jupyter_client.channels import HBChannel getting_started | File "/usr/local/lib/python3.5/dist-packages/jupyter_client/channels.py", line 47 getting_started | time_to_dead: float = 1.0 getting_started | ^ getting_started | SyntaxError: invalid syntax ```

ascatox (Wed, 14 Apr 2021 11:57:02 GMT):
Hi All! Is it possible to connect to an Indy installed on a server with a public address? Where can i find a config file to setup the server address?

RudraniAngira (Sat, 17 Apr 2021 00:32:03 GMT):
Has joined the channel.

waqaswahid1 (Mon, 19 Apr 2021 23:39:36 GMT):
Has joined the channel.

waqaswahid1 (Mon, 19 Apr 2021 23:39:38 GMT):
hello everyone how are you?

waqaswahid1 (Mon, 19 Apr 2021 23:48:42 GMT):
can someone guide me about hyperledger. which course is best to learn hyperledger indy.

bshambaugh (Tue, 20 Apr 2021 02:55:33 GMT):
I wish I knew more. I've seen stuff like this which may be helpful? https://hyperledger-indy.readthedocs.io/projects/sdk/en/latest/docs/getting-started/index.html , https://docs.trinsic.id/docs (uses hyperledger indy)

waqaswahid1 (Tue, 20 Apr 2021 06:39:44 GMT):
how can i store a thousand of users data on Hyperledger Indy using smart contract? can we make non fungible tokens and smart contracts on indy?

waqaswahid1 (Tue, 20 Apr 2021 06:40:35 GMT):
how can we make users' data as an NFT on hyperledger indy? How can we store and retrieve users NFTs privately?

deas (Tue, 20 Apr 2021 12:44:46 GMT):
@waqaswahid1 I'm not an Indy expert, but I'm pretty sure that's not really what Indy is for. The beauty of Indy is that none of the private information is stored on chain; all of the private data moves between (Aries) agents in an encrypted peer-to-peer fashion following the DIDcomm protocols. All that's stored on chain is the sets of the definitions and public identities needed to enable the verifiability of proof presentations made in those subsequent peer to peer transactions. Out of curiosity, what would be the goal of making user data into NFTs?

waqaswahid1 (Fri, 23 Apr 2021 10:01:43 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=S9vha3tSfoXbxZCiL) goal of making user data into NFTs is that user have the power to sell their personal data as an NFT to the companies.

waqaswahid1 (Fri, 23 Apr 2021 10:07:57 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=S9vha3tSfoXbxZCiL) to give power to the users to sell their personal on their own. but it is only possible when user data will be fully private and decentralized.

Vp10 (Fri, 23 Apr 2021 16:16:44 GMT):
Is this plug-in used only for web wallets or can also be used for mobile wallets like default SQL db? https://github.com/hyperledger/indy-sdk/blob/master/experimental/plugins/postgres_storage/README.md

Vp10 (Fri, 23 Apr 2021 16:16:44 GMT):
Is this plug-in used only for web wallets or can also be used for mobile wallets like default SQLite? https://github.com/hyperledger/indy-sdk/blob/master/experimental/plugins/postgres_storage/README.md

andrew.whitehead (Fri, 23 Apr 2021 17:37:56 GMT):
Postgres runs as separate database application, so it only suitable for mobile if you are connecting to a remote DB

andrew.whitehead (Fri, 23 Apr 2021 17:37:56 GMT):
Postgres runs as separate database application, so it's only suitable for mobile if you are connecting to a remote DB

vardan10 (Mon, 26 Apr 2021 08:12:18 GMT):
Is there a way to get did document from indy legder? I tried this API to create a request to fetch did doc as a trustee: https://github.com/hyperledger/indy-sdk/tree/master/wrappers/nodejs#buildgetddorequest--submitterdid-targetdid----requestresult But when I try to submit the request to ledger as a trustee i get following response: {"reason":"client request invalid: InvalidClientRequest('validation error [ClientAuthRuleOperation]: missed fields - field, constraint, auth_type, new_value, auth_action. ',)","reqId":1619421463489510700,"identifier":"V4SGRU86Z58d6TV7PBUe6f","op":"REQNACK"}

vardan10 (Mon, 26 Apr 2021 08:12:18 GMT):
Is there a way to get did document from indy legder? I tried this API to create a request to fetch did doc: https://github.com/hyperledger/indy-sdk/tree/master/wrappers/nodejs#buildgetddorequest--submitterdid-targetdid----requestresult But when I try to submit this request to ledger as a trustee i get following response: {"reason":"client request invalid: InvalidClientRequest('validation error [ClientAuthRuleOperation]: missed fields - field, constraint, auth_type, new_value, auth_action. ',)","reqId":1619421463489510700,"identifier":"V4SGRU86Z58d6TV7PBUe6f","op":"REQNACK"}

vardan10 (Mon, 26 Apr 2021 08:12:18 GMT):
Is there a way to get did document from indy legder? I tried this API to create a request to fetch did doc: https://github.com/hyperledger/indy-sdk/tree/master/wrappers/nodejs#buildgetddorequest--submitterdid-targetdid----requestresult But when I try to submit this request to ledger as a trustee, i get following response: {"reason":"client request invalid: InvalidClientRequest('validation error [ClientAuthRuleOperation]: missed fields - field, constraint, auth_type, new_value, auth_action. ',)","reqId":1619421463489510700,"identifier":"V4SGRU86Z58d6TV7PBUe6f","op":"REQNACK"}

LukaszSawicki (Mon, 26 Apr 2021 23:38:16 GMT):
Has joined the channel.

Shweta1 (Tue, 27 Apr 2021 13:53:11 GMT):
how to take backup of indy node data.I have created 4 indy nodes running as individual docker container.where exactly nodes transaction data is saving into docker container.I want to mount it to external volume in case of any crash.could you pls let me know where the indy data file present

Shweta1 (Thu, 29 Apr 2021 08:08:56 GMT):
how I can integrate indy node with postgres

dcspinho (Thu, 29 Apr 2021 16:19:45 GMT):
Hello. Someone know how to fix the error indy.error.WalletAlreadyExistsError?I've tryed this https://hyperledger-indy.readthedocs.io/projects/sdk/en/latest/docs/how-tos/trouble-shooting.html I can't find the directory .indy_client to remove https://github.com/hyperledger/indy-sdk/blob/master/samples/nodejs/README.md Thanks

lynn.bendixsen (Thu, 29 Apr 2021 16:50:52 GMT):
Hi Shweta1, The data, and key files are all under /var/lib/indy (txn or ledger data is further under in /data) and the config files are in /etc/indy.

dcspinho (Fri, 30 Apr 2021 11:59:33 GMT):
Hello. I'm trying indy sdk samples https://github.com/hyperledger/indy-sdk/tree/master/samples/python How can I know which nodes are in the pool? The indy APIS are this (link) ? https://github.com/hyperledger/indy-sdk/tree/master/wrappers/python/indy

sbirky (Fri, 30 Apr 2021 21:02:25 GMT):
Has joined the channel.

mccown (Wed, 05 May 2021 20:14:39 GMT):
The Identity Implementer's WG is meeting tomorrow (May 6th @ 9am MT / 3pm UTC). In this call, we will review updates from several industry WGs, which you may have missed this week. This week, James Ebert (Indicio) will be giving a presention on the new Aires Bifold project. Wiki: https://wiki.hyperledger.org/display/IWG/2021-05-06+%3A+Identity+Implementers+WG+Call Zoom: https://zoom.us/my/hyperledger.community.3?pwd=UE90WHhEaHRqOGEyMkV3cldKa2d2dz09

waqaswahid1 (Thu, 06 May 2021 08:58:45 GMT):
can anyone help me?

waqaswahid1 (Thu, 06 May 2021 09:00:06 GMT):
how to make DID of the user data which include the name, email, age, id no, and many other attributes? The DID should be accessible outside the world.

swcurran (Mon, 10 May 2021 17:52:36 GMT):
The following was posted by the W3C DID Working Group. Does anyone (e.g. @brentzundel @peacekeeper @drummondreed ) exactly what is needed here? I tried to understand it, but it is not obvious to me what is supposed to be in the DID Method document, and what testing is done. ``` The W3C DID Working Group achieved another milestone last week; all tests for the specification are now complete. This means that we are ready for DID Method implementers to submit their DID Method implementations for conformance testing against the specification. The deadline for submitting DID Method implementations is in two weeks (by May 25th, 2021). The instructions on how to submit a DID Method implementation can be found here: https://github.com/w3c/did-test-suite/#adding-your-did-implementation If you are a DID Method implementer, please get your DID Method implementations submitted in the next two weeks. ```

drummondreed (Mon, 10 May 2021 18:20:03 GMT):
@swcurran I am not an expert on the DID test suite (as I'm not a developer and thus have not been able to submit any tests). @peacekeeper is probably the best advisor on this followed by @brentzundel.

dcspinho (Mon, 10 May 2021 22:34:00 GMT):
hello. In Indy Indy Walkthrough is the steward that creates the entity identity (DID) https://hyperledger-indy.readthedocs.io/projects/sdk/en/latest/docs/getting-started/indy-walkthrough.html Is suposed to be steward or trustee?

trustcerts (Tue, 11 May 2021 08:39:22 GMT):
Hello everyone, is there a rocketchat for indy-node? I am interested in the status of the observer design approach for more scalable network: https://github.com/hyperledger/indy-plenum/blob/master/design/observers.md was there any progress during the last 3 years?

WadeBarnes (Tue, 11 May 2021 11:30:45 GMT):
#indy-node

Arsh-Sandhu (Fri, 14 May 2021 09:28:03 GMT):
Hello, is there any way with which we can get the private key associated with a DID in Indy. indy_create_and_store_my_did Function creates and stored the keys in the wallet and with KeyForDidAsync function we can get the verification key for the specified DID but this does not return the private key. so is it possible to Fetch the private key associated with the DID?

swcurran (Fri, 14 May 2021 15:10:05 GMT):
In Indy, use of the private key is embedded within Indy library calls, and the private key does not get exposed outside of those calls. I think there is even code that clears the memory where the private key was used to prevent it getting out. So, the basic answer is -- you can't. It's stored in encrypted storage, accessed by the routines that need to use it, and unavailable anywhere else.

charith1994 (Sat, 15 May 2021 17:04:46 GMT):
Has joined the channel.

Priya.P (Tue, 18 May 2021 14:31:34 GMT):
Hi, I am trying to install indy sdk on ubuntu18 .04 but getfing this error. Has anyone faced this issue before? Command used: sudo add-apt-repository "deb https://repo.sovrin.org/sdk/deb bionic stable" ERROR: Err:6 https://repo.sovrin.org/sdk/deb bionic InRelease The following signatures couldn't be verified because the public key is not available: NO_PUBKEY E8BDBE36C8C97811 Reading package lists... Done W: GPG error: https://repo.sovrin.org/sdk/deb bionic InRelease: The following signatures couldn't be verified because the public key is not available: NO_PUBKEY E8BDBE36C8C97811

Priya.P (Tue, 18 May 2021 14:31:34 GMT):
Hi, I am trying to install indy sdk on ubuntu18 .04 using indy SDK https://github.com/hyperledger/indy-sdk , but getfing this error. Has anyone faced this issue before? Command used:sudo apt-key adv --keyserver keyserver.ubuntu.com --recv-keys CE7709D068DB5E88 sudo add-apt-repository "deb https://repo.sovrin.org/sdk/deb bionic stable" ERROR: Err:6 https://repo.sovrin.org/sdk/deb bionic InRelease The following signatures couldn't be verified because the public key is not available: NO_PUBKEY E8BDBE36C8C97811 Reading package lists... Done W: GPG error: https://repo.sovrin.org/sdk/deb bionic InRelease: The following signatures couldn't be verified because the public key is not available: NO_PUBKEY E8BDBE36C8C97811

Priya.P (Tue, 18 May 2021 14:31:34 GMT):
Hi, I am trying to install indy sdk on ubuntu18 .04 using indy SDK docs here https://github.com/hyperledger/indy-sdk , but getfing this error. Has anyone faced this issue before? Command used:sudo apt-key adv --keyserver keyserver.ubuntu.com --recv-keys CE7709D068DB5E88 sudo add-apt-repository "deb https://repo.sovrin.org/sdk/deb bionic stable" ERROR: Err:6 https://repo.sovrin.org/sdk/deb bionic InRelease The following signatures couldn't be verified because the public key is not available: NO_PUBKEY E8BDBE36C8C97811 Reading package lists... Done W: GPG error: https://repo.sovrin.org/sdk/deb bionic InRelease: The following signatures couldn't be verified because the public key is not available: NO_PUBKEY E8BDBE36C8C97811

nserrao (Tue, 18 May 2021 17:54:30 GMT):
Has joined the channel.

lauravuo-techlab (Wed, 19 May 2021 08:22:13 GMT):
I remember having some issues in installation when switching from xenial to bionic. Nowadays I use this script for installation to ubuntu: https://github.com/findy-network/findy-wrapper-go/blob/dev/scripts/debian-libindy/install-indy.sh

mccown (Wed, 19 May 2021 22:03:56 GMT):
The Identity Implementer's WG is meeting tomorrow (May 20th @ 9am MT / 3pm UTC). In this call, we will review updates from several industry WGs, which you may have missed this week. This week, Ivan Temchenko will be giving a presentation on DIDComm-RS and how Jolocom is leveraging it for SSI solutions. Wiki: https://wiki.hyperledger.org/display/IWG/2021-05-20+%3A+Identity+Implementers+WG+Call Zoom: https://zoom.us/my/hyperledger.community.3?pwd=UE90WHhEaHRqOGEyMkV3cldKa2d2dz09

dcspinho (Thu, 20 May 2021 02:27:15 GMT):
Hi. I have this error on anoncreds.verifier_verify_proof. How can I fix this? Thanks INFO:indy.libindy.native.anoncreds_service: /home/indy/.cargo/registry/src/github.com-1ecc6299db9ec823/ursa-0.3.2/src/cl/verifier.rs:313 | Verifier verify proof -> done Traceback (most recent call last): File "/usr/lib/python3.6/runpy.py", line 193, in _run_module_as_main "__main__", mod_spec) File "/usr/lib/python3.6/runpy.py", line 85, in _run_code exec(code, run_globals) File "/home/maas4/web server py/src/main.py", line 43, in run_coroutine(main) File "/home/maas4/web server py/src/utils.py", line 52, in run_coroutine loop.run_until_complete(coroutine()) File "/usr/lib/python3.6/asyncio/base_events.py", line 488, in run_until_complete return future.result() File "/home/maas4/web server py/src/main.py", line 30, in main await getting_started.run() File "/home/maas4/web server py/src/getting_started.py", line 377, in run maas_service['revoc_regs_for_trip_preferences']) AssertionError

Priya.P (Thu, 20 May 2021 07:07:20 GMT):
@lauravuo-techlab ,Thanks it worked:thumbsup:

waqaswahid1 (Fri, 21 May 2021 04:35:39 GMT):

poolledgertimouterror.png

waqaswahid1 (Fri, 21 May 2021 07:28:14 GMT):
:indy:

jkrstic (Sat, 22 May 2021 18:10:32 GMT):
Has joined the channel.

esplinr (Tue, 25 May 2021 02:01:17 GMT):
The short answer is that nothing has been done to my knowledge. All the pieces exist in the ledger, but someone would need to write an observer client to pull the transactions without engaging in consensus.

leina.nazar (Tue, 25 May 2021 13:30:56 GMT):
Has joined the channel.

leina.nazar (Tue, 25 May 2021 13:30:56 GMT):
Is it possible to communicate between Hyperledger Fabric and Hyperledger Indy?

leina.nazar (Tue, 25 May 2021 13:31:14 GMT):
Can I read the data from Fabric and store it in Indy and vice-versa? My idea is to use Indy to authenticate identities and use Fabric to store data through chaincodes using the communicated credentials. Is Quilt the tool for it?

sauveergoel (Tue, 25 May 2021 18:13:18 GMT):
why is there three revocations IDs for one revocation registry?

sauveergoel (Tue, 25 May 2021 18:13:18 GMT):
why is there three revocations IDs for one credential definition?

swcurran (Tue, 25 May 2021 18:17:53 GMT):
Not sure what you mean by 3. There are two objects -- RevReg_Def and RevReg_Entry. The RevReg_Def is the metadata about the RevReg -- size, use type, ID, etc. RevReg_Entry -- are the updates to the RevReg as credentials are revoked -- the list of index values in the tails file that are revoked and the new accumulator for the entry.

sauveergoel (Tue, 25 May 2021 18:59:57 GMT):

Clipboard - May 26, 2021 12:29 AM

sauveergoel (Tue, 25 May 2021 19:00:57 GMT):
when i am creating a revocation reg for a given cred def, it is producing 3 different rev reg ids

swcurran (Tue, 25 May 2021 19:01:03 GMT):
Is that using ACA-Py?

sauveergoel (Tue, 25 May 2021 19:01:06 GMT):
yes

sauveergoel (Tue, 25 May 2021 19:01:06 GMT):
yes, it uses ACA-Py

sauveergoel (Tue, 25 May 2021 19:03:42 GMT):
the interesting fact is that when I created the rev reg id it gave me a different id and when I issued a credential, it attached the an other id, both were out of the three displayed, but the point is how to identify which id was used in the credential as the issue credential outputs don't carry rev reg info

sauveergoel (Tue, 25 May 2021 19:04:40 GMT):
I am using the bcgov indy-tails-server as well

swcurran (Tue, 25 May 2021 19:04:56 GMT):
I'm not sure about the three, but ACA-Py creates an extra RevReg so that there is never a need to stop issuing to create a RevReg (which can take awhile). When ACA-Py includes revocation on a Cred Def, it automatically creates two registries. Then it activates one, and uses it until it is all used up. When the first is full, it switches to the second one, and makes a third one in the background. That continues as needed. An issue is never delayed to make a RevReg.

swcurran (Tue, 25 May 2021 19:05:26 GMT):
So, it's not an Indy thing, it's done at the ACA-Py level.

sauveergoel (Tue, 25 May 2021 19:06:23 GMT):
ohh ok, but then how do we know which one has been used for which cred, when we are an endorser

swcurran (Tue, 25 May 2021 19:07:34 GMT):
When issuing, the "issue credential" process should be returning the id of the credential and it's revreg -- you have to know both to revoke. As well, if you keep the credential exchange record around in the issuer, both pieces of data are in that, and I believe (could be wrong...) that you can supply that when revoking, and the data will be automagically pulled from that record to do the revocation.

swcurran (Tue, 25 May 2021 19:08:26 GMT):
I'm confused on the three IDs that you first had (unless you have done a bunch of issuing).

sauveergoel (Tue, 25 May 2021 19:08:37 GMT):
thats my whole issue, `issue-credential` endpooints are not giving me the rev_reg_id back

swcurran (Tue, 25 May 2021 19:09:39 GMT):
Hmmm.... need people with more knowledge here -- hands on in the code. @esune , could you please weigh in here?

sauveergoel (Tue, 25 May 2021 19:13:44 GMT):

sauveergoel - Wed May 26 2021 00:43:41 GMT+0530 (India Standard Time).txt

sauveergoel (Tue, 25 May 2021 19:14:17 GMT):
this is the output of the /issue-credential/send API

sauveergoel (Tue, 25 May 2021 19:14:33 GMT):
it doesnot provide with rev_reg_id

swcurran (Tue, 25 May 2021 20:56:39 GMT):
Looks like it is still in the "Offer_sent" state -- which makes sense for the response to the API call -- it's an asynch operation to issue a credential. There should be a webhook come back to the controller when the credential itself has been issued. Also, you can query the credential exchange records and then look at the one for the issuance (assuming it has not been deleted).

esune (Wed, 26 May 2021 15:50:52 GMT):
@sauveergoel what version oaf aca-py are you using?

esune (Wed, 26 May 2021 15:50:52 GMT):
@sauveergoel what version of aca-py are you using?

esune (Wed, 26 May 2021 15:52:42 GMT):
As @swcurran stated, normally you would get `revocation_id` and `revoc_reg_id` as part of the response. Not sure whether they will show up if the state is prior to `issued`, as they might not have been written yet (I am not sure about this, but it is possible).

esune (Wed, 26 May 2021 15:52:42 GMT):
As @swcurran stated, normally you would get `revocation_id` and `revoc_reg_id` as part of the response. Not sure whether they will show up if the state is prior to `credential_issued`, as they might not have been written yet (I am not sure about this, but it is possible).

esune (Wed, 26 May 2021 15:54:05 GMT):
I would first make sure you can complete the flow and reach the `credential_issued` state on the credential exchange, and dig deeper afterwards.

PaulWen (Wed, 26 May 2021 18:53:12 GMT):
Has joined the channel.

waqaswahid1 (Thu, 27 May 2021 04:36:59 GMT):
question: lets say we have created DID and which is made through some json data. can we get back the data through DID.

daidoji (Thu, 27 May 2021 13:27:39 GMT):
Through indy no unless you manufacture that did yourself with some funny business. Through the default constructions though there isn't a way to get the data from the did by design (Indy community focuses a lot on privacy, data leakage, and anonymity).

SergiuM 1 (Thu, 27 May 2021 18:58:14 GMT):
Has joined the channel.

waqaswahid1 (Fri, 28 May 2021 04:56:52 GMT):
@daidoji is there any other way to get users credentials through DID?

daidoji (Fri, 28 May 2021 13:12:57 GMT):
I'm afraid I don't understand the question.

tgalal (Fri, 28 May 2021 16:01:22 GMT):
Has joined the channel.

mwherman2000 (Sat, 29 May 2021 17:41:44 GMT):
Does anyone know of a DID Core/VC or indysdk/libindy implementation that runs on top of a Plain Old Smart Contract Enabled Blockchain Ledger? Is it POSCEBL? https://github.com/IBM/hpass/blob/main/doc/did-spec.md is one example running on Fabric.

swcurran (Sun, 30 May 2021 20:59:32 GMT):
FYI - Arnaud (Chair of the Hyperledger TSC) requested a "top story" about each Hyperledger project. I've sent this in for Indy: ``` The top story in the Hyperledger Indy community this year has been the deployment of new public instances of Indy. While the most active and best known instance of Indy, Sovrin's Mainnet, continues to operate smoothly, a new global ledger was deployed by Indicio.tech, and the IDUnion group deployed their network in the European Union. The expansion of networks has triggered efforts to create a new DID Method for Indy (called "did:indy") to enable resolving Indy identifiers across all public Indy networks. ``` Feel free to provide feedback -- we can adjust as needed before the Hyperledger Global Forum.

weiyaous (Mon, 31 May 2021 20:28:23 GMT):
Dear all, I have a question regarding the observer node in Indy. When we read data from ledger through an observer node. How do we know the replicated ledger is not fake?

swcurran (Mon, 31 May 2021 20:45:32 GMT):
The observer node functionality has not yet been implemented. All reads are currently done against the ledger.

weiyaous (Mon, 31 May 2021 20:53:34 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=KrDf8n3RPepLZpD3u) Thanks, Stephen. May I ask from a design perspective? I found the idea about the observer node from Alex's slides. I think the observer nodes are good for reducing the validators to process reading request. I only curious about anyone have an idea about how to prevent the ledger on the observer node is not fake.

mwherman2000 (Mon, 31 May 2021 23:35:16 GMT):
(Repeating) my previous query) Are there projects out there who have implemented an Indy conformant API on top of an non-Indy blockchain/ledger platform?

mwherman2000 (Mon, 31 May 2021 23:35:16 GMT):
(Repeating my previous query) Are there projects out there who have implemented an Indy conformant API on top of an non-Indy blockchain/ledger platform?

mwherman2000 (Mon, 31 May 2021 23:35:16 GMT):
(Repeating my previous query) Are there projects out there who have implemented an Indy conformant API (or somewhat-conformant API) on top of an non-Indy blockchain/ledger platform?

swcurran (Tue, 01 Jun 2021 01:33:25 GMT):
Not that I'm aware of. There many ledgers that store DIDs (90-odd last check?), and many Indy ledger instances. But no ledgers that store the same objects as Indy, but using a different underlying blockchain.

conanoc (Wed, 02 Jun 2021 04:57:52 GMT):
You could compare the merkle root of the observer node to the merkle root of the real node.

CryptoMuck (Wed, 02 Jun 2021 12:28:03 GMT):
Hi all, I have a question regarding credential unlinkability. Based on the contents of a proof, a verifier cannot track which credentials are used to create the proof, I understand that. But if a prover wants to present a proof unlinkably multiple times to the same verifier, then it has to establish a new connection with the verifier for each presentation, correct? And I assume that on network level it must also be ensured that the individual presentations cannot be linked to each other (e.g. by changing IPs of the prover). Have I got all this right?

CryptoMuck (Wed, 02 Jun 2021 12:28:03 GMT):
Hi all, I have a question regarding credential unlinkability. Based on the contents of a proof, a verifier cannot track which credentials are used to create the proof, I understand that. But if a prover (AMA) wants to present a proof unlinkably multiple times to the same verifier (ACA Py), then it has to establish a new connection with the verifier for each presentation, correct? And I assume that on network level it must also be ensured that the individual presentations cannot be linked to each other (e.g. by changing IPs of the prover). Have I got all this right?

brentzundel (Wed, 02 Jun 2021 17:00:16 GMT):
Indy relies on state proofs for verification of read requests. This allows (currently) a read transaction to be answered by a single validator node, rather than requiring the read request to go through multiple validators. The same proof would allow for an observer node to be trusted.

mccown (Thu, 03 Jun 2021 12:57:04 GMT):
The Identity Implementer's WG is cancelled for today (6/3). Please join us when we reconvene in 2 weeks on 17 June 2021. :calendar:

mccown (Thu, 03 Jun 2021 12:57:04 GMT):
The Identity Implementer's WG is cancelled for today (6/3). Please join us when we reconvene in 2 weeks on 17 June 2021.

swcurran (Thu, 03 Jun 2021 16:39:34 GMT):
A little more detail -- to make sure that the transaction is valid, a client could query each validator node, compare the answers and decide if it is correct. The state proof is signed by multiple validators (I believe (T - 1)/3 nodes where T is the total number of nodes) so that a client only has to ask one validator node, get back the state proof and verifiy it.

swcurran (Thu, 03 Jun 2021 16:49:21 GMT):
There are several things that are not quite right here. From the proof, the verifier can tell what credentials the claims come from -- they have to know that to know who issued the credential and to verify the signatures on the data. `But if a prover (AMA) wants to present a proof unlinkably multiple times to the same verifier (ACA Py), then it has to establish a new connection with the verifier for each presentation, correct?` That's correct -- a person/entity could do that. Would be like signing up for a website using different email addresses. Linking via IP address might be possible - depends on how the message travelled from prover to verifier. With DIDComm, that could be multiple hops.

sergio.anguita (Fri, 04 Jun 2021 07:37:07 GMT):
Has joined the channel.

sergio.anguita (Fri, 04 Jun 2021 07:42:25 GMT):
Hi, I've been working with indy for a while and I notice kind of inconsistency in pool genesis format. I mean that, some versions have the pool_genesis content in where services port numbering is defined as string and others as integer. The problem arises when you try to load one of those files to connect to the pool, because you will get a common invalid state error message. For example: the sandox pool genesis (https://raw.githubusercontent.com/sovrin-foundation/sovrin/stable/sovrin/pool_transactions_sandbox_genesis) has ports encoded as string, while built-in development pool included in indy-sdk repo, has them defined as integers (https://github.com/hyperledger/indy-sdk/blob/master/samples/nodejs/src/util.js). So, what is the correct data type for this field?

swcurran (Fri, 04 Jun 2021 15:06:56 GMT):
Interesting. Thoughts @andrew.whitehead @dbluhm

dbluhm (Fri, 04 Jun 2021 16:15:45 GMT):
I'll check with Lynn Bendixsen; he manages our Indy networks at Indicio and was also a major player in the Sovrin networks

lynn.bendixsen (Fri, 04 Jun 2021 17:04:51 GMT):
All of the public networks that I have created genesis files for use strings for the ports, so I would assume that strings is the correct format. To create my genesis files I use: https://github.com/sovrin-foundation/steward-tools/tree/master/create_genesis and I have some step-by-step instructions to help with running that script: https://docs.google.com/document/d/1HXG6WOl_3K7k63CNdYEiTk1Y3eMiwLuhr5uTHg3RaoY

swcurran (Fri, 04 Jun 2021 18:11:39 GMT):
Presumably indy-sdk accepts both. @andrew.whitehead -- can you confirm that indy-vdr does as well?

andrew.whitehead (Fri, 04 Jun 2021 18:57:27 GMT):
It looks like client_port and node_port should both accept a string or number

sergio.anguita (Mon, 07 Jun 2021 08:35:19 GMT):
Im working on type safe application and needed to know the correct data type, just to avoid extra code for handling both datatypes. For what you said, using String is the most desirable way, am I right?

Bechir (Tue, 08 Jun 2021 19:00:44 GMT):
Has joined the channel.

Bechir (Tue, 08 Jun 2021 19:00:45 GMT):
hey everyone im very new to hyperledger indy and SSI and DID and i need to set it up is there a good tutorial/ course/article/ressource that can help me get started

ianco (Tue, 08 Jun 2021 19:04:32 GMT):
https://www.edx.org/course/identity-in-hyperledger-aries-indy-and-ursa

ianco (Tue, 08 Jun 2021 19:04:43 GMT):
https://www.edx.org/course/becoming-a-hyperledger-aries-developer

Bechir (Tue, 08 Jun 2021 19:05:01 GMT):
thank you very much can i ask you a question ?

Bechir (Tue, 08 Jun 2021 19:05:27 GMT):
is an SSI solution using the latest oauth a good idea that is doable in indy

Bechir (Tue, 08 Jun 2021 19:05:41 GMT):
and what is sovrin to indy

ianco (Tue, 08 Jun 2021 19:06:44 GMT):
Yes, in fact we've integrated with keycloak (on openshift) so you can authenticate with a verifiable credential

ianco (Tue, 08 Jun 2021 19:07:06 GMT):
SOVRIN is an implementation of a public Indy network

Bechir (Tue, 08 Jun 2021 19:07:33 GMT):
thank you so much for your time and efforts

Bechir (Tue, 08 Jun 2021 19:07:51 GMT):
im gonna be doing a whole research / scientific article on it

ianco (Tue, 08 Jun 2021 19:08:29 GMT):
https://github.com/bcgov/vc-authn-oidc

ianco (Tue, 08 Jun 2021 19:25:55 GMT):
Great!

vijaypatel (Tue, 08 Jun 2021 19:58:37 GMT):
Indicio offers trainings for indy that were very helpful for me and my devlmt team https://indicio.tech/training-packages/

chetanhere (Wed, 09 Jun 2021 17:04:42 GMT):
Has joined the channel.

Arsh-Sandhu (Fri, 11 Jun 2021 18:55:40 GMT):
Hi, is anybody able to build the Postgres wallet storage plug-in on windows? whenever i try to build, it always gives this error: error[E0463]: can't find crate for vcpkg. https://github.com/hyperledger/indy-sdk/issues/2248

sauveergoel (Fri, 11 Jun 2021 20:23:20 GMT):
can we export a postgreSQL wallet?

sauveergoel (Fri, 11 Jun 2021 20:25:20 GMT):
this was solved following your advice, thansk

ianco (Fri, 11 Jun 2021 21:18:55 GMT):
Yes you can use the indy cli to export the wallet. (All the wallets are compatible, so you can export from postgres and import into sqlite or the other way around)

tusharbansal (Sun, 13 Jun 2021 18:29:31 GMT):
Has joined the channel.

esune (Mon, 14 Jun 2021 17:53:40 GMT):
If you are moving from PostgreSQL to PostgreSQL you should be also able to use `pg_dump` and `pg_restore`

aito (Tue, 15 Jun 2021 09:21:18 GMT):
Has joined the channel.

aito (Tue, 15 Jun 2021 09:21:18 GMT):
restore

aito (Tue, 15 Jun 2021 10:11:18 GMT):
As I asked in #aries, can Indy manage something like "data DID" (or document DID)? I heard that ACA-py can't handle such concepts, is that also the case in Indy? In this case, "manage" means issuing a DID Document with data as the subject, controller as the data manager (person), etc. Some people have mentioned using Indy as a document management system. Have you done anything like this?

Shweta1 (Tue, 15 Jun 2021 10:56:28 GMT):
shweta

swcurran (Tue, 15 Jun 2021 13:41:06 GMT):
The ACA-Py answer was based on capabilities of Indy -- not for Aries/ACA-Py per se. Indy is only used for storing specific data related to the issuing of verifiable credentials. It is not for storing arbitrary documents, and does not support doing so. Currently, the DIDDoc returned from resolving an Indy DID is constrained to a very limited format.

aito (Tue, 15 Jun 2021 22:53:41 GMT):
Thanks!

mccown (Wed, 16 Jun 2021 20:48:06 GMT):
The Identity Implementer's WG is meeting tomorrow (June 17th @ 9am MT / 3pm UTC). In this call, we will review updates from several industry WGs, which you may have missed this week. This week, Bruce Conrad will be presenting on implementing Self-Sovereign Identity in IoT environments. Wiki: https://wiki.hyperledger.org/display/IWG/2021-06-17+:+Identity+Implementers+WG+Call Zoom: https://zoom.us/my/hyperledger.community.3?pwd=UE90WHhEaHRqOGEyMkV3cldKa2d2dz09

sergio.anguita (Thu, 17 Jun 2021 06:44:13 GMT):
is there any plans on the roadmap to release/build/design a Go SDK for libindy? Or is aries-framework-go capable enough to support all indy operations?

lauravuo-techlab (Thu, 17 Jun 2021 07:23:43 GMT):
we released our own go indy wrapper a while ago that we are using in our aries agency implementation: https://github.com/findy-network/findy-wrapper-go

lauravuo-techlab (Thu, 17 Jun 2021 07:23:43 GMT):
we published our own go indy wrapper a while ago that we are using in our aries agency implementation: https://github.com/findy-network/findy-wrapper-go

sergio.anguita (Thu, 17 Jun 2021 07:44:04 GMT):
oh, thats great idea. I'll take a look to that, cause I started with the same idea while ago.

sergio.anguita (Thu, 17 Jun 2021 07:44:04 GMT):
@lauravuo-techlab oh, thats great idea. I'll take a look to that, cause I started with the same idea while ago.

sergio.anguita (Thu, 17 Jun 2021 07:44:04 GMT):
@lauravuo-techlab oh, thats great. I'll take a look to that, cause I started with the same idea while ago.

pratapgowda (Thu, 17 Jun 2021 09:31:42 GMT):
Has joined the channel.

saverix (Fri, 18 Jun 2021 07:02:19 GMT):
Has joined the channel.

bruno.hivert (Fri, 18 Jun 2021 17:50:49 GMT):
Has joined the channel.

bruno.hivert (Fri, 18 Jun 2021 17:54:05 GMT):
Hello all, I'm using http://test.bcovrin.vonx.io/ as a ledger for a demo app, but if you connect to this url, you will quiclky realize that `Validator Node Status` is not showing anything 😶 And if you try to register a new Did, you get a 500 error. Is this some scheduled maintenance, or do I need to notify someone ? Thanks @swcurran for suggesting the right channel, by the way, obvious newby here

bruno.hivert (Fri, 18 Jun 2021 18:47:46 GMT):
Well, I noticed that the front-end is back up. Thanks/kudos to whoever fixed the problem.

esune (Fri, 18 Jun 2021 18:49:06 GMT):
I was going to message back now :slight_smile: I kicked the service and it looks like it should be working now, not (yet) sure what happened there - thanks for reporting.

swcurran (Fri, 18 Jun 2021 19:25:35 GMT):
FYI - we do have a monitor on the service. We just didn't respond fast enough this time. :-) Good to hear that people notice -- encourages us to do better.

WadeBarnes (Sat, 19 Jun 2021 13:01:33 GMT):
Apologies all, I though I had this sorted before 8am (PST) yesterday. I was traveling for appointments at the time. I've added additional alerting to better surface these sorts of issues in those situations. Thanks team BC for covering.

WadeBarnes (Sat, 19 Jun 2021 14:01:49 GMT):
Here is a ticket for the issue; https://github.com/bcgov/von-network/issues/148

gut (Wed, 23 Jun 2021 07:31:37 GMT):
Hi @swcurran Is there any architecture-vs-repository map to know everey development performed under Indy+Aries ecosystem?

swcurran (Wed, 23 Jun 2021 14:38:05 GMT):
Not that I know, but that doesn't sound too hard to create. Maintaining may be harder, but it's pretty known right now. Do you have an example of what you are thinking of from another ecosystem?

jacobsaur (Thu, 24 Jun 2021 00:08:12 GMT):
Hi all, I had a question about the quintessential zero-knowledge proof example of Alice proving that she is over 18 without revealing her DOB - given that current Indy predicates only work on int32s, what is the best way to represent dates in a credential such that they work for predicate proofs? I had one idea: at first I was trying to go with an approach of number of days since a certain epoch and being able to use common datetime functions to do the calculations, but it seemed rather complicated and there were a lot of subtle issues around timezones, leap seconds, which epoch to choose, etc. The approach I would like to propose is to apply simple modular arithmetic to map a date to an int of the form YYYYMMDD. This would allow easy conversion to int and back to date without any confusion that would result from using time libraries. The only gotchas I can see are that you couldn't represent dates before AD (which seems ok, or they could be negative which also seems ok), and that you'd need to have some error handling if somehow a credential was ever submitted with an invalid mapping eg 20201332 (13th month, 32nd day, etc) - although some error handling on credential issuing should address that. To construct the proof the verifier would just convert today's date minus 18 years to int form, and then check that the int-date DOB on the credential is < that. What do people think? Ideally I'd like to use a standard date convention so if there's not already one, I could add a proposal to https://github.com/hyperledger/aries-rfcs/blob/master/concepts/0074-didcomm-best-practices/README.md#date-time-conventions (or create a new RFC)

jacobsaur (Thu, 24 Jun 2021 00:08:12 GMT):
Hi all, I had a question about the quintessential zero-knowledge proof example of Alice proving that she is over 18 without revealing her DOB - given that current Indy predicates only work on int32s, what is the best way to represent dates in a credential such that they work for predicate proofs? I had one idea: at first I was trying to go with an approach of number of days since a certain epoch and being able to use common datetime functions to do the calculations, but it seemed rather complicated and there were a lot of subtle issues around timezones, leap seconds, which epoch to choose, etc. The approach I would like to propose is to apply simple modular arithmetic to map a date to an int of the form YYYYMMDD. This would allow easy conversion to int and back to date without any confusion that would result from using time libraries. The only gotchas I can see are that you couldn't represent dates before AD (which seems ok, or they could be negative which also seems ok), and that you'd need to have some error handling if somehow a credential was ever submitted with an invalid mapping eg 20201332 (13th month, 32nd day, etc) - although some error handling on credential issuing should address that. To construct the proof the verifier would just convert today's date minus 18 years to int form, and then check that the int-date DOB on the credential is < that. What do people think? Ideally I'd like to use a standard date convention so if there's not already one, I could add a proposal to https://github.com/hyperledger/aries-rfcs/blob/master/concepts/0074-didcomm-best-practices/README.md#date-time-conventions (or create a new RFC). (And I should specify that 32 bit unixtime wouldn't work, as we want to include dates outside of 1901-2038)

sergio.anguita (Thu, 24 Jun 2021 09:47:58 GMT):
dates can also considered as numeric values using epoch format, which is more standarized way to figure out if a value is bigger or less than certain threshold. In that way, the epoch is truncated to int32 size to make the calculations. Does this fit for your needs?

sergio.anguita (Thu, 24 Jun 2021 09:47:58 GMT):
@jabocdates can also considered as numeric values using epoch format, which is more standarized way to figure out if a value is bigger or less than certain threshold. In that way, the epoch is truncated to int32 size to make the calculations. Does this fit for your needs?

sergio.anguita (Thu, 24 Jun 2021 09:47:58 GMT):
@jacobsaur dates can also considered as numeric values using epoch format, which is more standarized way to figure out if a value is bigger or less than certain threshold. In that way, the epoch is truncated to int32 size to make the calculations. Does this fit for your needs?

sergio.anguita (Thu, 24 Jun 2021 09:49:45 GMT):
i think you can still apply the unix with a offset substraction

sergio.anguita (Thu, 24 Jun 2021 09:49:45 GMT):
i think you can still apply the unix format with a 69 year (1970-1901) offset implementation

jacobsaur (Thu, 24 Jun 2021 16:45:32 GMT):
I believe the epoch format for unix timestamps can only cover dates between 1901-2038 (unless you're referring to another format), and we need to be able to handle dates outside that range (eg 30-year mortgages in the future, land titles in the past, etc).

swcurran (Thu, 24 Jun 2021 17:27:13 GMT):
I would love to see this. A couple of interesting challenges to throw into this: - It would be nice to be able to do this via a convention so that it can be used with minimal/no changes to Indy AnonCreds and the indy-sdk. - If you switch to epoch for the raw data, the format becomes ugly for display, such as in the wallet. It could lead to needing both an epoch date and a "human" date attribute. Would it work to have a convention of the field name and data format? For example, the name of the attribute must end in "DT" and the data element must be an ISO date format. Alternatively, perhaps it is sufficient that the string format being an ISO Date is sufficient. If you do that, then the encoded value can be an integer date, and the proof request predicate could assume it is a date. The encoding would then have to be formalized in an RFC, so that all participants were aware of it.

swcurran (Thu, 24 Jun 2021 17:27:13 GMT):
I would love to see this. A couple of interesting challenges to throw into this: - It would be nice to be able to do this via a convention so that it can be used with minimal/no changes to Indy AnonCreds and the indy-sdk. - If you switch to epoch/integer for the raw data, the format becomes ugly for display, such as in the wallet. It could lead to needing both an epoch/integer date and a "human" date attribute. Would it work to have a convention of the field name and data format? For example, the name of the attribute must end in "DT" and the data element must be an ISO date format. Alternatively, perhaps it is sufficient that the string format being an ISO Date is sufficient. If you do that, then the encoded value can be an integer date, and the proof request predicate could assume it is a date. The encoding would then have to be formalized in an RFC, so that all participants were aware of it.

swcurran (Thu, 24 Jun 2021 17:29:28 GMT):
FYI -- here is the current definition of how to encode data in Indy credentials: https://github.com/hyperledger/aries-rfcs/blob/5fdf3bbdc31881fd18162f7f2051b139537942a0/features/0037-present-proof/README.md#verifying-claims-of-indy-based-verifiable-credentials

swcurran (Thu, 24 Jun 2021 17:30:16 GMT):
This would extend that to include some "format-sniffing" to decide if something is a date and then encoding it as appropriate.

jacobsaur (Thu, 24 Jun 2021 21:57:32 GMT):
Would the approach I suggested of mapping to an int in the form YYYYMMDD address most of these issues - ie it doesn't require changes to Indy AnonCreds and it's more or less human readable? Granted some UIs will still probably want to format it into a date format appropriate for their locale but at least you can tell what date it is by looking at it.

swcurran (Thu, 24 Jun 2021 22:20:56 GMT):
Doh...I misread what you wrote and was making it more complicated. That's a great idea and it does solve the human readable problem -- nice! Nevermind what I said. This is excellent and solves the issue very nicely! `#respect` OK -- lets push it.

gut (Fri, 25 Jun 2021 13:08:42 GMT):
No, I don't have any examples. Sorry for that. In some way, it is already done in each repo references. I'll try to gather it myself and share it as it could be helpful for newcomers.

swcurran (Fri, 25 Jun 2021 18:09:26 GMT):
If you get something started that seems useful, please share and I'll try to supplement it.

deulu (Sun, 27 Jun 2021 10:05:08 GMT):
Has joined the channel.

deulu (Mon, 28 Jun 2021 09:40:44 GMT):
Hi everyone, I have a question about indy node performance. When we perform a load test with indy java wrapper, the tps values we get are very low (for example, the tps value for nym request is around 50-60) and when we monitor,we saw that java threads are waiting on submit request step. Does anyone have a similar load test and get different values? Or if you can get much higher tps values, what would you suggest to check? Not only write request but also read request is giving similar tps results. We made our test in our private network ( which 4 nodes on same machine different docker containers and machine resources 40 GB ram,30 CPU, 105 GB storage.)

gut (Mon, 28 Jun 2021 12:18:20 GMT):

Repo-arch visual map.png

gut (Mon, 28 Jun 2021 12:18:25 GMT):
This is the first draft of what I had in mind. The final purpose is to know which projects are used by who while creating an SSI ecosystem. I've sent to you a PM in parallel.

swcurran (Mon, 28 Jun 2021 12:53:32 GMT):
Very nice! That's a clever presentation of the data.

jacobsaur (Mon, 28 Jun 2021 16:54:14 GMT):
Thanks - would you advice raising a new RFC for this, or just add a new example to https://github.com/hyperledger/aries-rfcs/blob/master/concepts/0074-didcomm-best-practices/README.md#date-time-conventions . My thoughts are that we could have an _dateint (or _intdate?) suffix, since it's similar to the _date example but just in int format.

jacobsaur (Mon, 28 Jun 2021 16:54:14 GMT):
Thanks - would you advise raising a new RFC for this, or just add a new example to https://github.com/hyperledger/aries-rfcs/blob/master/concepts/0074-didcomm-best-practices/README.md#date-time-conventions . My thoughts are that we could have an _dateint (or _intdate?) suffix, since it's similar to the _date example but just in int format.

jacobsaur (Mon, 28 Jun 2021 16:54:14 GMT):
Thanks - would you advise raising a new RFC for this, or just add a new example to https://github.com/hyperledger/aries-rfcs/blob/master/concepts/0074-didcomm-best-practices/README.md#date-time-conventions . My thoughts are that we could have an "_dateint" (or "_intdate"?) suffix, since it's similar to the _date example but just in int format.

swcurran (Mon, 28 Jun 2021 17:07:01 GMT):
I was thinking it should go in either or both of https://github.com/hyperledger/aries-rfcs/tree/master/concepts/0441-present-proof-best-practices and this https://github.com/hyperledger/aries-rfcs/tree/master/features/0592-indy-attachments In a quick scan of 0592, I don't see the "verifying claims" section that was in 0037 (link from my earlier comment), so I think we should add that in, and include this idea there and in the examples. As well, we need the caveat of this not handling time zones and all that stuff, so the test is accurate to +/- 24 hours. I think we should verify the process and document it so that people have a clear example of "To prove a person is over 21, do the following" covering the attribute in the schema, the credential data and the proof request. The RFC you mentioned is a general one about DIDComm messages, not about VCs, so I recommend NOT adding it there. We need to focus on clarifying the handling of this for Indy AnonCreds.

jacobsaur (Mon, 28 Jun 2021 19:41:09 GMT):
Ok great, thanks. I think I'll start with a PR off of https://github.com/hyperledger/aries-rfcs/tree/master/concepts/0441-present-proof-best-practices with a section below the timestamps

praneel1819 (Tue, 29 Jun 2021 04:05:30 GMT):
Has joined the channel.

jacobsaur (Tue, 29 Jun 2021 17:12:48 GMT):
I've submitted a PR here: https://github.com/hyperledger/aries-rfcs/pull/687 The only thing that seems a little bit off is that the RFC is focused on proof presentations, while this also has a component about how the issuer should represent dates in their issued credentials (but it's in the service of eventually creating a proof). Comments welcome!

pranavkirtani88 (Wed, 30 Jun 2021 12:18:49 GMT):
I want to create a REST service using indy-sdk, what is a good way to scale indy.openwallet?

swcurran (Wed, 30 Jun 2021 13:30:42 GMT):
What are you trying to scale -- what particular bottlenecks are you hitting?

WadeBarnes (Wed, 30 Jun 2021 13:59:55 GMT):
@swcurran, I started responding to the same question here; https://chat.hyperledger.org/channel/indy-sdk?msg=33NG35WnQzZSYYA6a

tusharbansal (Wed, 30 Jun 2021 18:36:42 GMT):
Hello Everyone, I am trying to run Von network on the AWS instance. On executing "./manage start" I am able to run it. Even with AWS instance's public address I am able to see the VON Dashboard and when I do /genesis , I am able to see genesis file where client port IP and node port IP as AWS Instance's private address instead of public address. How can I get public IP instead of private IP in my Genesis file? I tried to give external genesis file which has client IP and Node IP as my AWS Instance's public address but, I am getting error "Error initializing pool ledger " on VON Dashboard. Could anyone please guide me here?

WadeBarnes (Wed, 30 Jun 2021 18:51:37 GMT):
https://github.com/bcgov/von-network#running-the-network-on-a-vps Step 5: ``` # This command requires the publicly accessible ip address of the machine `public_ip_address` # WEB_SERVER_HOST_PORT maps the docker service port to a public port on the machine # LEDGER_INSTANCE_NAME sets the display name of the ledger on the page headers. ./manage start public_ip_address WEB_SERVER_HOST_PORT=80 "LEDGER_INSTANCE_NAME=My Ledger" ```

WadeBarnes (Wed, 30 Jun 2021 18:53:08 GMT):
An example of hte start command for http://test.bcovrin.vonx.io/ ``` ./manage start 138.197.138.255 WEB_SERVER_HOST_PORT=80 "LEDGER_INSTANCE_NAME=BCovrin Test" "INFO_SITE_TEXT=vonx.io" "INFO_SITE_URL=https://vonx.io/" ```

WadeBarnes (Wed, 30 Jun 2021 18:53:08 GMT):
An example of the start command for http://test.bcovrin.vonx.io/ ``` ./manage start 138.197.138.255 WEB_SERVER_HOST_PORT=80 "LEDGER_INSTANCE_NAME=BCovrin Test" "INFO_SITE_TEXT=vonx.io" "INFO_SITE_URL=https://vonx.io/" ```

mccown (Wed, 30 Jun 2021 19:34:32 GMT):
The Identity Implementer's WG is meeting tomorrow (July 1st @ 9am MT / 3pm UTC). In this call, we will review updates from several industry WGs, which you may have missed this week. This week, Steve McCown (Anonyome Labs) will present on automatically-generating language wrappers for Rust libraries, such as didcomm_rs. Language wrappers include: Swift, Kotlin, Python, and C++. Wiki: https://wiki.hyperledger.org/display/IWG/2021-07-01+%3A+Identity+Implementers+WG+Call Zoom: https://zoom.us/my/hyperledger.community.3?pwd=UE90WHhEaHRqOGEyMkV3cldKa2d2dz09

MikeSmith (Thu, 01 Jul 2021 05:18:51 GMT):
Has joined the channel.

TapuDas (Thu, 01 Jul 2021 06:55:13 GMT):
Has joined the channel.

TapuDas (Thu, 01 Jul 2021 06:55:14 GMT):
hello everyone.I am doing research on hypeledger indy for my udergrad thesis.Is there any working implementation of indy? I've tried to implement indy using this link: https://github.com/cloudcompass/ToIPLabs/tree/master/src/indy-material/nodejs .

TapuDas (Thu, 01 Jul 2021 06:55:14 GMT):
hello everyone.I am doing research on hypeledger indy for my udergrad thesis.Is there any working implementation of indy? I've tried to implement indy using this link: https://github.com/cloudcompass/ToIPLabs/tree/master/src/indy-material/nodejs .But I can't tweak this as my need. For different node version and npm packages it gives error I can't understand

TapuDas (Thu, 01 Jul 2021 06:55:14 GMT):
hello everyone.I am doing research on hypeledger indy for my udergrad thesis. Is there any working implementation of indy? I've tried to implement indy using this link: https://github.com/cloudcompass/ToIPLabs/tree/master/src/indy-material/nodejs .But I can't tweak this as my need. For different node version and npm packages it gives error I can't understand

tusharbansal (Thu, 01 Jul 2021 08:01:32 GMT):
Thanks for the response. I tried this method as well, I am not getting any pool initialization error but *I was still getting private IP in client IP and node IP my genesis file.* This private IP in genesis File causing issue when I am passing ledger URL(AWS instance's public URL) in Aries Agent to start up. Could you please guide me here?

tusharbansal (Thu, 01 Jul 2021 08:01:32 GMT):
Thanks for the response. I tried this method as well, I am not getting any pool initialization error but *I was still getting private IP in client IP and node IP in my genesis file.* This private IP in genesis File causing issue when I am passing ledger URL(AWS instance's public URL) in Aries Agent to start up. Could you please guide me here?

swcurran (Thu, 01 Jul 2021 14:13:56 GMT):
Lots and lots. To run an indy ledger, look at https://github.com/bcgov/von-network, and then to run agents, https://github.com/hyperledger/aries-cloudagent-python

jacobsaur (Thu, 01 Jul 2021 18:42:58 GMT):
Thanks for all the feedback, the PR is approved, now just needs to be merged: https://github.com/hyperledger/aries-rfcs/pull/687

gaberasturi (Fri, 02 Jul 2021 11:24:53 GMT):
Has joined the channel.

rjones (Fri, 02 Jul 2021 23:11:51 GMT):

rjones (Fri, 02 Jul 2021 23:11:51 GMT):

pmn2090 (Sat, 03 Jul 2021 01:08:19 GMT):
Has joined the channel.

vakster (Sat, 03 Jul 2021 07:01:03 GMT):
Has joined the channel.

BravoNatalie (Sat, 03 Jul 2021 10:55:01 GMT):
Has joined the channel.

mahengct (Sat, 03 Jul 2021 13:07:46 GMT):
Has joined the channel.

AsifMesbah (Sun, 04 Jul 2021 01:00:45 GMT):
Has joined the channel.

AsifMesbah (Sun, 04 Jul 2021 01:09:40 GMT):
@TapuDas I think it will work

mahengct (Sun, 04 Jul 2021 13:54:12 GMT):
Could someone please show me a demo video of their INDY setup for my usecase: "eKYC sharing"

mahengct (Sun, 04 Jul 2021 13:54:16 GMT):
?

IoanSlavu (Sun, 04 Jul 2021 17:21:29 GMT):
Has joined the channel.

AsifMesbah (Sun, 04 Jul 2021 18:19:27 GMT):
Can anyone help me please regarding this issue? https://stackoverflow.com/questions/68244464/node-gyp-rebuild-error-on-docker-hyperledger-indy

sg1408 (Sun, 04 Jul 2021 19:01:33 GMT):
Has joined the channel.

bphan (Mon, 05 Jul 2021 05:00:24 GMT):
Has joined the channel.

mturetchi (Mon, 05 Jul 2021 10:53:36 GMT):
Has joined the channel.

sharkrit (Mon, 05 Jul 2021 18:20:50 GMT):
Has joined the channel.

QuasiStellar (Mon, 05 Jul 2021 19:00:35 GMT):
Has joined the channel.

LikithS (Mon, 05 Jul 2021 19:51:40 GMT):
Has joined the channel.

sakthi.raju (Tue, 06 Jul 2021 04:39:14 GMT):
Has joined the channel.

mahendraswain (Tue, 06 Jul 2021 05:15:07 GMT):
Has joined the channel.

shanest (Tue, 06 Jul 2021 05:25:59 GMT):
Has joined the channel.

yujinpark (Tue, 06 Jul 2021 05:50:18 GMT):
Has joined the channel.

vshawn (Tue, 06 Jul 2021 07:06:54 GMT):
Has joined the channel.

chhokra (Tue, 06 Jul 2021 09:38:57 GMT):
Has joined the channel.

cpsatav (Tue, 06 Jul 2021 10:14:17 GMT):
Has joined the channel.

vasile135 (Tue, 06 Jul 2021 15:36:45 GMT):
Has joined the channel.

CalebLiu (Wed, 07 Jul 2021 02:48:52 GMT):
Has joined the channel.

drewmattic (Wed, 07 Jul 2021 04:58:03 GMT):
Has joined the channel.

chhokra (Wed, 07 Jul 2021 08:44:03 GMT):
While following https://hyperledger-indy.readthedocs.io/projects/sdk/en/latest/docs/getting-started/run-getting-started.html, Is anyone running into an issue regarding Jupyter failing to build . "Traceback (most recent call last): File "/usr/local/bin/pip3", line 7, in from pip._internal.cli.main import main File "/usr/local/lib/python3.5/dist-packages/pip/_internal/cli/main.py", line 58 sys.stderr.write(f"ERROR: {exc}") ^ SyntaxError: invalid syntax ERROR: Service 'jupyter' failed to build: The command '/bin/sh -c pip3 install -U pip ipython-notebook ipython==7.9 setuptools jupyter python3-i"

gt88 (Wed, 07 Jul 2021 11:40:04 GMT):
Has joined the channel.

gaberasturi (Wed, 07 Jul 2021 11:55:07 GMT):
Hi, It's possible to query from indy-cli for all schemas in a DOMAIN ledger? like in VON Ledger Browser filtering by txn_type=101. Many thanks.

tychobrailleur (Wed, 07 Jul 2021 12:26:05 GMT):
Has joined the channel.

alanveloso (Wed, 07 Jul 2021 12:52:15 GMT):
Has joined the channel.

swcurran (Wed, 07 Jul 2021 13:31:43 GMT):
No. There are no query mechanisms on Indy. VON Ledger browser and IndyScan query one by one through all the ledger objects, creates a database and use that to do queries like the one you want to do.

gaberasturi (Wed, 07 Jul 2021 13:36:03 GMT):
Thank you very much Stephen.

shanest (Wed, 07 Jul 2021 13:53:38 GMT):
Sorry for newb question. I have stucked here for days. I even tried to search on past messages but I still unable to get indy-cli working on MacOS Big Sur. I am using .zshrc to set up the environment path but zsh still returns command not found. Here's how my .zshrc looks like `export LIBRARY_PATH=~/Users/capuchin/SDK/indy/libindy_1.16.0/lib/ export LD_LIBRARY_PATH=~/Users/capuchin/SDK/indy/libindy_1.16.0/lib/ export LIBINDY_DIR=~/Users/capuchin/SDK/indy/libindy_1.16.0/lib/` Can someone point to me what did I do wrong? and also runnin indy-cli executable file returns `capuchin@macintosh ~ % /Users/capuchin/SDK/indy/indy-cli ; exit; dyld: Library not loaded: /Users/jenkins/workspace/indy-sdk_indy-sdk-cd_rc/libindy/target/release/deps/libindy.dylib Referenced from: /Users/capuchin/SDK/indy/indy-cli Reason: image not found zsh: abort /Users/capuchin/SDK/indy/indy-cli`

ianco (Wed, 07 Jul 2021 14:06:24 GMT):
Looks like you're using absolute paths, so you shouldn't include the `~` prefix

ianco (Wed, 07 Jul 2021 14:07:34 GMT):
... also rather than updating your `PATH` env variables try creating links to the `/usr/local/lib` directory (which is searched by default)

ianco (Wed, 07 Jul 2021 14:08:39 GMT):
e.g. `ln -s /Users/capuchin/SDK/indy/libindy_1.16.0/lib/somefile.dylib /usr/local/lib/somefile.dylib`

ReufRujevic (Wed, 07 Jul 2021 14:08:44 GMT):
Has joined the channel.

shanest (Wed, 07 Jul 2021 14:22:25 GMT):
I have copied "libindy.lib" and "libindy.a" into /usr/local/lib and updated .zshrc to

shanest (Wed, 07 Jul 2021 14:25:08 GMT):
I have copied "libindy.lib" and "libindy.a" into /usr/local/lib and updated .zshrc to `export LIBRARY_PATH="/usr/local/lib/" export LD_LIBRARY_PATH="/usr/local/lib/" export LIBINDY_DIR="/usr/local/lib/" export PATH="/usr/local/opt/openssl@1.1/bin:$PATH" export LDFLAGS="-L/usr/local/opt/openssl@1.1/lib" export CPPFLAGS="-I/usr/local/opt/openssl@1.1/include"` but it still won't recognized the lib. Running the executable indy-cli file returns `dyld: Library not loaded: /usr/local/opt/libsodium/lib/libsodium.18.dylib Referenced from: /usr/local/lib/libindy.dylib Reason: image not found `

ianco (Wed, 07 Jul 2021 15:19:10 GMT):
Make sure you install all the dependencies: https://github.com/hyperledger/indy-sdk#macos

AndreEsteves (Thu, 08 Jul 2021 10:56:17 GMT):
Has joined the channel.

mkm 14 (Thu, 08 Jul 2021 14:00:03 GMT):
Has joined the channel.

lynn.bendixsen (Thu, 08 Jul 2021 14:04:35 GMT):
Here's a document I use to help people install indy-cli on mac: https://docs.google.com/document/d/1mJ3r3uH9jIWkA59X6bV7eWUlWmQ0NerSuPSFv4Lh7vQ/edit#heading=h.hdfj5odxmp7o I don't think its required, but I also have the following in my ~/.profile file: export SODIUM_LIB_DIR=/usr/local/lib

Cryptohash (Thu, 08 Jul 2021 16:11:07 GMT):
Has joined the channel.

Najetbh83 (Thu, 08 Jul 2021 19:33:56 GMT):
Has joined the channel.

hoang-innomizetech (Fri, 09 Jul 2021 03:56:49 GMT):
Has joined the channel.

AndrewStarling (Sat, 10 Jul 2021 03:58:53 GMT):
Has joined the channel.

ritikbhandari (Sat, 10 Jul 2021 09:48:21 GMT):
Has joined the channel.

atirikt (Sat, 10 Jul 2021 09:54:10 GMT):
Has joined the channel.

CopyBilla (Sat, 10 Jul 2021 14:18:07 GMT):
Has joined the channel.

danielszego (Sat, 10 Jul 2021 18:47:00 GMT):
Has joined the channel.

AkshatCy2012sn (Sat, 10 Jul 2021 21:14:31 GMT):
Has joined the channel.

tim-trutech (Sat, 10 Jul 2021 21:27:39 GMT):
Has joined the channel.

TomRobertson1 (Sun, 11 Jul 2021 17:11:48 GMT):
Has joined the channel.

eharris128 (Sun, 11 Jul 2021 23:03:06 GMT):
Has joined the channel.

ksanjayk (Mon, 12 Jul 2021 02:21:47 GMT):
Has joined the channel.

davidviejopomata (Mon, 12 Jul 2021 14:20:50 GMT):
Has joined the channel.

IV12 (Mon, 12 Jul 2021 15:09:54 GMT):
Has joined the channel.

jmaric (Mon, 12 Jul 2021 15:47:59 GMT):
Has joined the channel.

piyushap (Mon, 12 Jul 2021 16:21:49 GMT):
Has joined the channel.

piyushap (Mon, 12 Jul 2021 16:22:10 GMT):
Hello

piyushap (Mon, 12 Jul 2021 16:59:38 GMT):
i have to set up hyperledger-indy project ..anyone guide me how to setup node and write chaincode

WadeBarnes (Mon, 12 Jul 2021 18:05:45 GMT):
Easiest way to spin up an indy-node based ledger for development; https://github.com/bcgov/von-network/

WadeBarnes (Mon, 12 Jul 2021 18:06:58 GMT):
Hosted ledger options; https://sovrin.org/overview/ You can start on BuilderNet or StagingNet here for free; https://selfserve.sovrin.org/

Aurelio 8 (Mon, 12 Jul 2021 20:21:13 GMT):
Has joined the channel.

NancyL 4 (Mon, 12 Jul 2021 20:24:03 GMT):
Has joined the channel.

mfrodriguezc (Mon, 12 Jul 2021 23:36:18 GMT):
Has joined the channel.

LabibFarag (Tue, 13 Jul 2021 05:47:02 GMT):
Has joined the channel.

surajsingla333 (Tue, 13 Jul 2021 07:03:01 GMT):
Has joined the channel.

surajsingla333 (Tue, 13 Jul 2021 07:16:55 GMT):
Hey all, I am trying to locally run von network and aries cloud agents. So to begin with, I am using sovrin network - stagging network genesis block to start the von network locally. Now when I start the aries-cloud-agent with the local LEDGER_URL and it gives me TAA error while creating schema. The did is registered on the sovrin staging network using `https://selfserve.sovrin.org/nym`. Not sure why I am still facing the TAA error since von network code was recently updated to incorporate those changes. Anyone has any idea about this?

bt333 (Tue, 13 Jul 2021 11:42:30 GMT):
Has joined the channel.

WadeBarnes (Tue, 13 Jul 2021 12:05:15 GMT):
It sounds like you're connecting to Sovrin StagingNet and not a `von-network` ledger instance. From what I can gather you're setting up `von-network` as a ledger browser for Sovrin StagingNet.

WadeBarnes (Tue, 13 Jul 2021 12:06:23 GMT):
What's the error you are getting regarding the TAA? When a TAA is enforced on a network you need to accept the agreement using one of the accepted options.

lfrattini (Tue, 13 Jul 2021 19:26:13 GMT):
Has joined the channel.

alebaffa (Wed, 14 Jul 2021 05:15:47 GMT):
Has joined the channel.

TuureLainen (Wed, 14 Jul 2021 06:25:32 GMT):
Has joined the channel.

MartinAulich (Wed, 14 Jul 2021 07:14:56 GMT):
Has joined the channel.

pranavkirtani88 (Wed, 14 Jul 2021 08:05:42 GMT):
Hello, where can I get in contact with developers that have worked on the vc-authn-oidc repo on github? https://github.com/bcgov/vc-authn-oidc

swcurran (Wed, 14 Jul 2021 08:12:07 GMT):
@esune is the one most familiar with that repo and deploying it.

PatrickHerbke-TUBerlin (Wed, 14 Jul 2021 08:22:49 GMT):
Has joined the channel.

PatrickHerbke-TUBerlin (Wed, 14 Jul 2021 08:51:11 GMT):
Hi all, we are working as Consortialpartner in the project IDunoin and in future also in the project EBSI. In context of education, we want to „pimp“ the schema definition (and credential defintion) to have a full High School Diploma (Abiturzeugnis) as verifiable credential. Is there a way to modify the existing schema such as: from: ` "attrNames": ["age", "sex", "height", "name"] ` to: ` ( "attrNames") or „credentialSubject": { // TODO: Add learner... "degree": { "type": "HighSchoolDiploma", "grade": "2.1", // Grade conversion based on country. Adopted from https://github.com/european-commission-europass/Europass-Learning-Model "issuedDate":"2020-03-30T00:00:00+02:00" // Ausstellungsdatum des Zeugnisses }, "achievements": [ { "learningAchievements": [ // Schulfächer und pro Semester {{ "educationSubject": "Mathematics", "codeISCED": "0541", "semesterGrading": { "semester1": "6", // 6 points Semester 1 "semester2": "10", }}]},{ "foreignLanguages": [ // Fremdsprachen { "educationSubject": "English", "codeISCED": "0232", "grade":"C1" },} ….. ` Bests, Patrick p.s.: Example from https://hyperledger-indy.readthedocs.io/projects/sdk/en/latest/docs/how-tos/save-schema-and-cred-def/README.html STEP3

PatrickHerbke-TUBerlin (Wed, 14 Jul 2021 08:52:14 GMT):
Hi all, we are working as Consortialpartner in the project IDunoin and in future also in the project EBSI. In context of education, we want to „pimp“ the schema definition (and credential defintion) to have a full High School Diploma (Abiturzeugnis) as verifiable credential. Is there a way to modify the existing schema such as: from: ` "attrNames": ["age", "sex", "height", "name"] ` to: ` ( "attrNames") or „credentialSubject": { // TODO: Add learner... "degree": { "type": "HighSchoolDiploma", "grade": "2.1", // Grade conversion based on country. Adopted from https://github.com/european-commission-europass/Europass-Learning-Model "issuedDate":"2020-03-30T00:00:00+02:00" // Ausstellungsdatum des Zeugnisses }, "achievements": [ { "learningAchievements": [ // Schulfächer und pro Semester {{ "educationSubject": "Mathematics", "codeISCED": "0541", "semesterGrading": { "semester1": "6", // 6 points Semester 1 "semester2": "10", }}]},{ "foreignLanguages": [ // Fremdsprachen { "educationSubject": "English", "codeISCED": "0232", "grade":"C1" },} ….. ` Bests, Patrick p.s.: Example from https://hyperledger-indy.readthedocs.io/projects/sdk/en/latest/docs/how-tos/save-schema-and-cred-def/README.html STEP3

PatrickHerbke-TUBerlin (Wed, 14 Jul 2021 08:54:12 GMT):
Hi all, we are working as Consortialpartner in the project IDunoin and in future also in the project EBSI. In context of education, we want to „pimp“ the schema definition (and credential defintion) to have a full High School Diploma (Abiturzeugnis) as verifiable credential. Is there a way to modify the existing schema such as: from: `"attrNames": ["age", "sex", "height", "name"]` to: `( "attrNames") or „credentialSubject": { // TODO: Add learner... "degree": { "type": "HighSchoolDiploma", "grade": "2.1", // Grade conversion based on country. Adopted from https://github.com/european-commission-europass/Europass-Learning-Model "issuedDate":"2020-03-30T00:00:00+02:00" // Ausstellungsdatum des Zeugnisses }, "achievements": [ { "learningAchievements": [ // Schulfächer und pro Semester {{ "educationSubject": "Mathematics", "codeISCED": "0541", "semesterGrading": { "semester1": "6", // 6 points Semester 1 "semester2": "10", }}]},{ "foreignLanguages": [ // Fremdsprachen { "educationSubject": "English", "codeISCED": "0232", "grade":"C1" },} …..` Bests, Patrick p.s.: Example from https://hyperledger-indy.readthedocs.io/projects/sdk/en/latest/docs/how-tos/save-schema-and-cred-def/README.html STEP3

PatrickHerbke-TUBerlin (Wed, 14 Jul 2021 08:55:15 GMT):

Bildschirmfoto 2021-07-14 um 10.54.39.png

PatrickHerbke-TUBerlin (Wed, 14 Jul 2021 08:55:40 GMT):
Hi all, we are working as Consortialpartner in the project IDunoin and in future also in the project EBSI. In context of education, we want to „pimp“ the schema definition (and credential defintion) to have a full High School Diploma (Abiturzeugnis) as verifiable credential. Is there a way to modify the existing schema such as: from: ` "attrNames": ["age", "sex", "height", "name"] ` to: ` ( "attrNames") or „credentialSubject": { // TODO: Add learner... "degree": { "type": "HighSchoolDiploma", "grade": "2.1", // Grade conversion based on country. Adopted from https://github.com/european-commission-europass/Europass-Learning-Model "issuedDate":"2020-03-30T00:00:00+02:00" // Ausstellungsdatum des Zeugnisses }, "achievements": [ { "learningAchievements": [ // Schulfächer und pro Semester {{ "educationSubject": "Mathematics", "codeISCED": "0541", "semesterGrading": { "semester1": "6", // 6 points Semester 1 "semester2": "10", }}]},{ "foreignLanguages": [ // Fremdsprachen { "educationSubject": "English", "codeISCED": "0232", "grade":"C1" },} ….. ` Bests, Patrick p.s.: Example from https://hyperledger-indy.readthedocs.io/projects/sdk/en/latest/docs/how-tos/save-schema-and-cred-def/README.html STEP3

PatrickHerbke-TUBerlin (Wed, 14 Jul 2021 08:55:40 GMT):
Hi all, we are working as Consortialpartner in the project IDunoin and in future also in the project EBSI. In context of education, we want to „pimp“ the schema definition (and credential defintion) to have a full High School Diploma (Abiturzeugnis) as verifiable credential. Is there a way to modify the existing schema such as: from: `"attrNames": ["age", "sex", "height", "name"]` to: `( "attrNames") or „credentialSubject": { // TODO: Add learner... "degree": { "type": "HighSchoolDiploma", "grade": "2.1", // Grade conversion based on country. Adopted from https://github.com/european-commission-europass/Europass-Learning-Model "issuedDate":"2020-03-30T00:00:00+02:00" // Ausstellungsdatum des Zeugnisses }, "achievements": [ { "learningAchievements": [ // Schulfächer und pro Semester {{ "educationSubject": "Mathematics", "codeISCED": "0541", "semesterGrading": { "semester1": "6", // 6 points Semester 1 "semester2": "10", }}]},{ "foreignLanguages": [ // Fremdsprachen { "educationSubject": "English", "codeISCED": "0232", "grade":"C1" },} …..` Bests, Patrick p.s.: Example from https://hyperledger-indy.readthedocs.io/projects/sdk/en/latest/docs/how-tos/save-schema-and-cred-def/README.html STEP3

PatrickHerbke-TUBerlin (Wed, 14 Jul 2021 08:55:40 GMT):
Hi all, we are working as Consortialpartner in the project IDunoin and in future also in the project EBSI. In context of education, we want to „pimp“ the schema definition (and credential defintion) to have a full High School Diploma (Abiturzeugnis) as verifiable credential. Is there a way to modify the existing schema such as: from: `"attrNames": ["age", "sex", "height", "name"]` to: `( "attrNames") or „credentialSubject": { // TODO: Add learner... "degree": { "type": "HighSchoolDiploma", "grade": "2.1", // Grade conversion based on country. Adopted from https://github.com/european-commission-europass/Europass-Learning-Model "issuedDate":"2020-03-30T00:00:00+02:00" // Ausstellungsdatum des Zeugnisses }, "achievements": [ { "learningAchievements": [ // Schulfächer und pro Semester {{ "educationSubject": "Mathematics", "codeISCED": "0541", "semesterGrading": { "semester1": "6", // 6 points Semester 1 "semester2": "10", }}]},{ "foreignLanguages": [ // Fremdsprachen { "educationSubject": "English", "codeISCED": "0232", "grade":"C1" },}…..` Bests, Patrick p.s.: Example from https://hyperledger-indy.readthedocs.io/projects/sdk/en/latest/docs/how-tos/save-schema-and-cred-def/README.html STEP3

PatrickHerbke-TUBerlin (Wed, 14 Jul 2021 08:55:40 GMT):
Hi all, we are working as Consortialpartner in the project IDunoin and in future also in the project EBSI. In context of education, we want to „pimp“ the schema definition (and credential defintion) to have a full High School Diploma (Abiturzeugnis) as verifiable credential. Is there a way to modify the existing schema such as: from: `"attrNames": ["age", "sex", "height", "name"]` to: see screenshot Bests, Patrick p.s.: Example from https://hyperledger-indy.readthedocs.io/projects/sdk/en/latest/docs/how-tos/save-schema-and-cred-def/README.html STEP3

PatrickHerbke-TUBerlin (Wed, 14 Jul 2021 09:04:58 GMT):

Bildschirmfoto 2021-07-14 um 11.04.34.png

PatrickHerbke-TUBerlin (Wed, 14 Jul 2021 09:50:21 GMT):

Bildschirmfoto 2021-07-14 um 11.50.06.png

domwoe (Wed, 14 Jul 2021 09:56:31 GMT):
For indy credentials not really. Indy creds are just lists of attributes. So as soon as you want to define an array with an unspecified size (as your example suggests) you have a problem.

PatrickHerbke-TUBerlin (Wed, 14 Jul 2021 10:00:51 GMT):
is there a more detailed documentation somewhere than https://hyperledger-indy.readthedocs.io/projects/sdk/en/latest/docs/how-tos/save-schema-and-cred-def/README.html? Would our approach (array lists) generally not work or can code passages be adapted to implement this?

PatrickHerbke-TUBerlin (Wed, 14 Jul 2021 10:04:03 GMT):
The current schema restrictions wouldn't allow any more complex certification, such as high school diploma (Abiturzeugnis). We don't want to store just signed pdf files in our wallet solution...do you have a tip, which solutions would offer themselves for it?

domwoe (Wed, 14 Jul 2021 10:09:38 GMT):
Currently you could either (1) Prepare a schema/creddef that supports a fixed number of arrays items and make them explicit attributes0.degree.type ..., but this makes proof requests pretty ugly. (2) have the whole attributes block as one stringified attribute, but then you have to disclose the whole block

domwoe (Wed, 14 Jul 2021 10:10:36 GMT):
The main issue is that a credential definition includes cryptographic matterial per attribute. Hence, the number of attributes is fixed

domwoe (Wed, 14 Jul 2021 10:11:32 GMT):
There was work in Rich Schemas, but it's more likely that we switch to W3C Credentials with BBS+ Signatures

domwoe (Wed, 14 Jul 2021 10:11:32 GMT):
There was work on Rich Schemas, but it's more likely that the community switches to W3C Credentials with BBS+ Signatures to support such use cases

swcurran (Wed, 14 Jul 2021 10:17:09 GMT):
Note that such use cases with BBS+ are going to "fun". While you can have an arbitrary length array, for selective disclosure, all of the attributes become a list, and you pick the elements of the list to disclose. Thus, to expose an entry (set of values) in the array, you will have to determine the indices in the list for those items. That said, a user interface can be constructed for the Holder to make the selection of the values in they array they want to disclose and then have the code calculate the list indices needed.

swcurran (Wed, 14 Jul 2021 10:17:09 GMT):
Note that such use cases with BBS+ are going to "fun". While you can have an arbitrary length array, for selective disclosure, all of the attributes become a list, and you pick the elements of the list to disclose. Thus, to expose an entry (set of values) in the array, you will have to determine the indices in the list for those items. That said, a user interface can be constructed for the Holder to make the selection of the values in the array they want to disclose and then have the code calculate the list indices needed.

domwoe (Wed, 14 Jul 2021 10:18:02 GMT):
Another approach would be to issue individual credentials for each array item :)

swcurran (Wed, 14 Jul 2021 10:19:13 GMT):
And interestingly, I just raised a PR to the "issue-credential" RFC to support such a use case -- issuing multiple credentials with an instance of the protocol :-). https://github.com/hyperledger/aries-rfcs/pull/692

swcurran (Wed, 14 Jul 2021 10:19:13 GMT):
And interestingly, I just raised a PR to the "issue-credential" RFC to support such a use case -- issuing multiple credentials within an instance of the protocol :-). https://github.com/hyperledger/aries-rfcs/pull/692

domwoe (Wed, 14 Jul 2021 10:21:42 GMT):
I'm looking forward to support BBS+ Creds in the BPA :)

umeraqeel (Wed, 14 Jul 2021 10:36:01 GMT):
Has joined the channel.

PatrickHerbke-TUBerlin (Wed, 14 Jul 2021 11:53:19 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=G2QHC5veXontdSx76) Great ! :-) Maybe we then will have the opp. to realize this.

jkbpvsc (Wed, 14 Jul 2021 12:41:08 GMT):
Has joined the channel.

tonybka (Wed, 14 Jul 2021 14:03:54 GMT):
Has joined the channel.

XINGYUZHU (Wed, 14 Jul 2021 14:37:01 GMT):
Has joined the channel.

janrockdev (Wed, 14 Jul 2021 15:53:07 GMT):
Has joined the channel.

janrockdev (Wed, 14 Jul 2021 15:55:39 GMT):
Has left the channel.

HakanTUB (Wed, 14 Jul 2021 17:05:26 GMT):
Has joined the channel.

Lucampelli (Wed, 14 Jul 2021 18:49:22 GMT):
Hello! two quick questions: Are the Hyperledger Indy's credentials actually Smart Contracts, or does Indy actually use Smart Contracts?

swcurran (Wed, 14 Jul 2021 18:58:49 GMT):
No and No. Indy credentials do not go on a ledger. Only public keys for signing / verifying credentials, credential schema metadata, and revocation data go on the ledger.

Lucampelli (Wed, 14 Jul 2021 19:01:20 GMT):
Nice, thank you! One last question: In a setting where a university needs a credential in order to be valid for issuing a diploma. If the university's credencial were to be revoked, and the university ceased function, how to check if the university was "valid" by the time the diploma was sent, if it is needed to obtain the proof from the university in order to check that?

Lucampelli (Wed, 14 Jul 2021 19:02:22 GMT):
I'd say maybe the university generates and sends the proof, together with the diploma credencial... what do you think?

Lucampelli (Wed, 14 Jul 2021 19:03:01 GMT):
This way the student would be able to send that proof with the diploma proof, and someone would be able to check both right?

mccown (Wed, 14 Jul 2021 23:10:06 GMT):
The Identity Implementer's WG is meeting tomorrow (July 15th @ 9am MT / 3pm UTC). In this call, we will review updates from several industry WGs, which you may have missed. This week, Timo Glastra (Co-Founder, Animo Solutions) will present their OSS & SSI work and how it drives their national collaboration to facilitate citizen volunteers during emergency situations. Wiki: https://wiki.hyperledger.org/display/IWG/2021-07-01+%3A+Identity+Implementers+WG+Call Zoom: https://zoom.us/my/hyperledger.community.3?pwd=UE90WHhEaHRqOGEyMkV3cldKa2d2dz09

rcraswell (Wed, 14 Jul 2021 23:13:00 GMT):
Has joined the channel.

TIMO_ReDWit (Thu, 15 Jul 2021 05:42:03 GMT):
Has joined the channel.

bharathkumar_m (Thu, 15 Jul 2021 14:13:58 GMT):
Has joined the channel.

alextrafimau (Thu, 15 Jul 2021 14:24:47 GMT):
Has joined the channel.

mccown (Thu, 15 Jul 2021 14:42:14 GMT):
The Identity Implementer's WG is starting at the top of the hour (9am MT / 3pm UTC). Today, Timo Glastra (Co-Founder, Animo Solutions) is presenting on their OSS & SSI work and how they are using that in a national collaboration to help citizen volunteers during emergencies. Zoom: https://zoom.us/my/hyperledger.community.3?pwd=UE90WHhEaHRqOGEyMkV3cldKa2d2dz09 Wiki: https://wiki.hyperledger.org/display/IWG/2021-07-15+%3A+Identity+Implementers+WG+Call

sgpokala (Thu, 15 Jul 2021 16:44:24 GMT):
Has joined the channel.

brett7297 (Thu, 15 Jul 2021 17:35:49 GMT):
Has joined the channel.

pranjal23 (Thu, 15 Jul 2021 19:04:08 GMT):
Has joined the channel.

pranjal23 (Thu, 15 Jul 2021 19:09:24 GMT):
hello, I need some help using the command "./network.sh down" , I'm facing some issues while running this command in git bash

pranjal23 (Thu, 15 Jul 2021 19:09:32 GMT):

Screenshot (45).png

deas (Thu, 15 Jul 2021 19:20:58 GMT):
This looks like you're having trouble with #fab.

deas (Thu, 15 Jul 2021 19:20:58 GMT):
This looks like you're having trouble with #fabric not #indy . Right?

pranjal23 (Thu, 15 Jul 2021 19:21:35 GMT):
fabric

deas (Thu, 15 Jul 2021 19:22:08 GMT):
Yeah - I'd recommend trying in #fabric or another Fabric-related channel, then.

pranjal23 (Thu, 15 Jul 2021 19:23:12 GMT):
I tried no response from the channel

Lucampelli (Thu, 15 Jul 2021 20:52:57 GMT):
What i said is impossible in Indy right?...

swcurran (Fri, 16 Jul 2021 06:31:43 GMT):
There has been some talk of that -- being able to include a proof within a credential. The RFC 104 chained credential talks about that use case. https://github.com/hyperledger/aries-rfcs/tree/master/concepts/0104-chained-credentials That said, I'm not aware of an implementation of this. Would love to see it as it could add a lot of interesting capabilities, including the one you mention.

richarde 7 (Fri, 16 Jul 2021 06:51:34 GMT):
Has joined the channel.

dominichawton (Fri, 16 Jul 2021 10:20:33 GMT):
Has joined the channel.

Sowjanya_Segu (Fri, 16 Jul 2021 10:45:50 GMT):
Has joined the channel.

DmitryHalahayeu (Fri, 16 Jul 2021 13:23:57 GMT):
Has joined the channel.

mdhasan (Fri, 16 Jul 2021 13:48:06 GMT):
Has joined the channel.

Lupusalix (Fri, 16 Jul 2021 13:57:20 GMT):
Has joined the channel.

Laichonious (Fri, 16 Jul 2021 16:20:03 GMT):
Has joined the channel.

pSchlarb (Fri, 16 Jul 2021 16:43:32 GMT):
Has joined the channel.

ChandramouliMaduri (Fri, 16 Jul 2021 20:58:23 GMT):
Has joined the channel.

mostafafares (Fri, 16 Jul 2021 21:46:55 GMT):
Has joined the channel.

hackerunet (Sat, 17 Jul 2021 01:51:08 GMT):
Has joined the channel.

phamhung (Sat, 17 Jul 2021 02:00:11 GMT):
Has joined the channel.

ederro (Sat, 17 Jul 2021 02:08:49 GMT):
Has joined the channel.

noe305 (Sat, 17 Jul 2021 02:10:53 GMT):
Has joined the channel.

itsbalamurali (Sat, 17 Jul 2021 08:45:48 GMT):
Has joined the channel.

piratesmile (Sat, 17 Jul 2021 12:28:05 GMT):
Has joined the channel.

RajanHarshit (Sat, 17 Jul 2021 15:37:36 GMT):
Has joined the channel.

otfreetime (Sun, 18 Jul 2021 06:37:28 GMT):
Has joined the channel.

kaushambi (Sun, 18 Jul 2021 06:40:33 GMT):
Has joined the channel.

igorkrupczynski (Sun, 18 Jul 2021 08:31:50 GMT):
Has joined the channel.

fghiasy (Mon, 19 Jul 2021 00:59:59 GMT):
Has joined the channel.

bhavesh.c (Mon, 19 Jul 2021 04:58:02 GMT):
Has joined the channel.

chhokra (Mon, 19 Jul 2021 08:22:56 GMT):
https://hyperledger-indy.readthedocs.io/projects/node/en/latest/add-node.html While following this tutorial. We are sucessfully able to set up a static indy pool, but we want to add nodes dynamically as given in this document. But, if our node is on a different EC2 instance, how can we give the access of genesis file as it doesn't exist on this new instance?

grim 37 (Mon, 19 Jul 2021 08:27:10 GMT):
Has joined the channel.

swcurran (Mon, 19 Jul 2021 08:30:16 GMT):
The genesis file is a static document once established, and so can be put in an accessible place. For example, the Sovrin network genesis files are on GitHub.

grim 37 (Mon, 19 Jul 2021 08:44:45 GMT):
Has left the channel.

Steit (Mon, 19 Jul 2021 09:59:16 GMT):
Has joined the channel.

Shweta1 (Mon, 19 Jul 2021 11:58:07 GMT):
Hi Team, I have setup indy nodes on multi host.2 indy nodes on host1 and 2 indy nodes on host2. How I can add New nodes into this environement.How i can use genesis_url.Also can i remove nodes from this network? Thanks in advance !!

chhokra (Mon, 19 Jul 2021 12:07:53 GMT):
To add to this question, is there a specific procedure to be followed for adding new nodes from a host different from host 1 and host 2? When trying to do so while following https://hyperledger-indy.readthedocs.io/projects/node/en/latest/add-node.html my node crashes

Shweta1 (Mon, 19 Jul 2021 12:09:02 GMT):
what error you are getting?

chhokra (Mon, 19 Jul 2021 12:11:45 GMT):
The error log on my third host shows the following messages "Node3 found the following missing connections : Node2,Node1. The node is not ready yet so view change will not be proposed now, but rescheduled"

SwapnaliDive (Mon, 19 Jul 2021 12:20:02 GMT):
Has joined the channel.

SwapnaliDive (Mon, 19 Jul 2021 12:20:02 GMT):
I tried starting the node3 on host3 just by copying the pool_transactions_genesis and domain_transactions_genesis from host1 of node1 , is this the proper way of doing this ??

Shweta1 (Mon, 19 Jul 2021 12:20:14 GMT):
How you are providing exiting genesis_pool?

Shweta1 (Mon, 19 Jul 2021 12:20:14 GMT):
How you are providing exiting genesis_pool?New nodes to be added into existing genesis pool

SwapnaliDive (Mon, 19 Jul 2021 12:22:01 GMT):
Yes I am following https://hyperledger-indy.readthedocs.io/projects/node/en/latest/add-node.html#node-preparation , for step one I copied files of Node1 and provided on host3 to start Node3

Shweta1 (Mon, 19 Jul 2021 12:24:05 GMT):
even i tried in the same way ,but not works

GuntherWalden (Mon, 19 Jul 2021 13:13:16 GMT):
Has joined the channel.

WadeBarnes (Mon, 19 Jul 2021 14:03:02 GMT):
Resources; https://github.com/hyperledger/indy-node/issues/1674

WadeBarnes (Mon, 19 Jul 2021 14:03:58 GMT):
If you just want to spin up an `indy-node` based network for testing use; https://github.com/bcgov/von-network

MALodder (Mon, 19 Jul 2021 14:53:01 GMT):
QQ - what are the revocation requirements from the indy community, please be specific like 1- Holder privacy 2- Max revocation proof size 4KB 3- Max computation time 1s 4- Single network request to become up-to-date In finishing the revocation proposal for anoncreds 2.0 I want to make sure I didn't miss anything

MatthiasGeihs.Perun (Mon, 19 Jul 2021 15:07:59 GMT):
Has joined the channel.

AtulPandeySaw (Tue, 20 Jul 2021 05:31:44 GMT):
Has joined the channel.

Prerana72 (Tue, 20 Jul 2021 05:43:25 GMT):
Has joined the channel.

JohnWick2704 (Tue, 20 Jul 2021 06:16:59 GMT):
Has joined the channel.

tusharbansal (Tue, 20 Jul 2021 06:38:06 GMT):
Hello Team, We are trying to create an indy-network where we are facing issue adding a node into an existing pool. So far our network setup is working for "Remote Test Network" but when trying to add another node from different aws instance then getting error as mentioned below. We are following Hyperledger indy documentation "https://hyperledger-indy.readthedocs.io/projects/node/en/latest/add-node.html#node-preparation" Could someone please guide us here?

tusharbansal (Tue, 20 Jul 2021 06:45:28 GMT):

Screenshot from 2021-07-19 16-51-57.png

tusharbansal (Tue, 20 Jul 2021 06:45:28 GMT):
Hello Team, We are trying to create an indy-network where we are facing issue adding a node into an existing pool. So far our network setup is working for "Remote Test Network" but when trying to add another node from different aws instance then getting error as mentioned below. We are following Hyperledger indy documentation "https://hyperledger-indy.readthedocs.io/projects/node/en/latest/add-node.html#node-preparation" Could someone please guide us here?
Screenshot from 2021-07-19 16-51-57.png

RositsaKrasteva (Tue, 20 Jul 2021 08:34:03 GMT):
Has joined the channel.

MaternaDev (Tue, 20 Jul 2021 08:48:12 GMT):
Has joined the channel.

RyanEckersfield (Tue, 20 Jul 2021 08:52:58 GMT):
Has joined the channel.

indisateesh (Tue, 20 Jul 2021 11:05:21 GMT):
Has joined the channel.

Prerana72 (Tue, 20 Jul 2021 11:52:50 GMT):
Hello Team, We are developing edge agent for android. We are using libindy library for android taken from " https://repo.sovrin.org/android/libindy/stable/1.16.0/". When we try to store the credentials using indy_prover_store_credential, we see the following error- "indyorg.hyperledger.indy.sdk.InvalidStructureException: A value being processed is not valid." Please note that we have ensured the following: 1. we are passing valid JSON string values for all the input params. For "rev_reg_def_json" we passed null. 2. ensured the JSON strings carry the value null instead of "null", wherever applicable. Any pointers to how we can find out which param may be causing the issue? The library does not specify that in the error. We tried to build the libindy for android by adding few logs. But we could not successfully compile it. The 'cargo build' is successful but the 'build-libindy-android.sh' is failing with error - "can't open file 'build/tools/make_standalone_toolchain.py' :[Errno 2] No such file or directory?" We are building on ubuntu 20.0.4. Did anyone face this issue or knows what could be the issue?

WadeBarnes (Tue, 20 Jul 2021 12:29:23 GMT):
A few things to look at: - Indy-node uses ports 9700-9799. I see 8701 and 8702 in your screenshot. - There is a distinct difference between internal and external client and node IP:Port addresses, especially on AWS. When initializing the node you need to use the internal addresses and ports. When registering the node on the network with `ledger node` you need to use external (publicly accessible) addresses. - The genesis file must be constructed with publicly accessible addresses. - After that you need to make sure you have your internal and external addresses mapped correctly.

WadeBarnes (Tue, 20 Jul 2021 12:30:02 GMT):
Some additional resources: https://github.com/hyperledger/indy-node/issues/1674

DeepaliGupta (Tue, 20 Jul 2021 12:54:28 GMT):
Has joined the channel.

gautamarora24 (Tue, 20 Jul 2021 16:14:54 GMT):
Has joined the channel.

dhoffman32 (Tue, 20 Jul 2021 18:11:38 GMT):
Has joined the channel.

romenon2703 (Wed, 21 Jul 2021 05:43:58 GMT):
Has joined the channel.

ashishtripathi (Wed, 21 Jul 2021 05:46:47 GMT):
Hello Everyone, We have been working on Identity management with ERC1056, with the current requirement we are planning to use Ursa's revocation registry for the VC revocation. I wanted to know if the Credential definition or any other definitions in Ursa are JSON-LD compatible? Thanks in Advance!

ashishtripathi (Wed, 21 Jul 2021 05:46:47 GMT):
Hello Everyone, We have been working on Identity management with ERC1056, with the current requirement we are planning to use Ursa's revocation registry for the VC revocation. I wanted to know if the Credential definition or any other definitions in Ursa are JSON-LD compatible? Also, would it be possible to use Ursa's implementation of VCs and ZKP for any different chain other that IndyPool maybe Ethereum? Thanks in Advance!

nehajain20683 (Wed, 21 Jul 2021 07:19:55 GMT):
Has joined the channel.

SwapnaliDive (Wed, 21 Jul 2021 07:20:07 GMT):
Thank you so much @WadeBarnes We are able to add new node from different aws instances now. Great help.

nehajain20683 (Wed, 21 Jul 2021 07:23:25 GMT):
Thank you @WadeBarnes

yaozhaoyuan (Wed, 21 Jul 2021 07:57:05 GMT):
Has joined the channel.

bhaaratkhatri (Wed, 21 Jul 2021 08:22:16 GMT):
Has joined the channel.

Maginaro (Wed, 21 Jul 2021 10:02:23 GMT):
Has joined the channel.

WadeBarnes (Wed, 21 Jul 2021 12:18:58 GMT):
:thumbsup:

NancyL 4 (Wed, 21 Jul 2021 14:08:23 GMT):
Hello Everyone, I would like to use Indy SDK for Java, can someone help me please?

indirajith (Wed, 21 Jul 2021 14:33:47 GMT):
Has joined the channel.

domwoe (Wed, 21 Jul 2021 15:31:00 GMT):
Does someone know off hand the size of a tails file given the size of the revocation registry?

yo-va-ni (Wed, 21 Jul 2021 18:06:55 GMT):
Has joined the channel.

domwoe (Thu, 22 Jul 2021 07:47:48 GMT):
If anyone is curious, it's about 1MB/4000 entries

ideree (Thu, 22 Jul 2021 11:41:45 GMT):
Has joined the channel.

swcurran (Thu, 22 Jul 2021 14:11:05 GMT):
Thanks. We've checked that at points, but I couldn't find the data.

swcurran (Thu, 22 Jul 2021 14:11:18 GMT):
:man_facepalming_light_skin_tone:

septcinquante (Thu, 22 Jul 2021 15:21:41 GMT):
Has joined the channel.

mianko (Thu, 22 Jul 2021 15:52:29 GMT):
Has joined the channel.

faizulbari (Thu, 22 Jul 2021 16:00:29 GMT):
Has joined the channel.

AmarSrivastava1 (Thu, 22 Jul 2021 16:16:52 GMT):
Has joined the channel.

MikeSmith (Thu, 22 Jul 2021 17:21:25 GMT):
Hello Everyone, I am a beginner in Hyperledger Indy and trying to understand the basics. There is one question that I am not able to relate or understand. What is the purpose of adding a node to an indy pool? Does the node gets added as validator node and become a part of consensus? If so what is the reward of becoming a validator node? Could anyone please guide me here?

MikeSmith (Thu, 22 Jul 2021 17:25:45 GMT):
Hello Everyone, I was trying to setup Indy node and as per the documentation I found only Ubuntu 16 is preferred. Is there any way to make the setup on Ubuntu 20?

lynn.bendixsen (Thu, 22 Jul 2021 17:39:04 GMT):
Adding nodes to an Indy pool helps with consensus as every node currently added becomes a participator in consensus. Because of the BFT algorithm that Indy uses for consensus, the more nodes that are part of the pool, the more robust, or stable the network is (or more tolerant of nodes going down or unavailable). The best benefit of running a node on a public Indy network is the good feeling you get about donating to a great cause! Think of the prestige and the marketing opportunities! Participating also makes the network diverse, decentralized and viable, so if the product you are working on needs a network, it pays to support at least your favorite one, and maybe more than one, as we head into an era of multi-network compatibility and usage. Node Operators (Stewards) do not have any special write privileges on the major public networks at this time, they simply manage their node. Some networks might return some of the eventual profits of the network to Node Operators, but even then it is unlikely that it will do much more than cover costs IMO.

MikeSmith (Thu, 22 Jul 2021 17:59:10 GMT):
@lynn.bendixsen Thank you so much for your answer.

MikeSmith (Thu, 22 Jul 2021 18:00:19 GMT):
Hello Everyone, I was trying to setup Indy node and as per the documentation I found only Ubuntu 16 is preferred. Is there any way to make the setup on Ubuntu 20? Is it even possible?

swcurran (Thu, 22 Jul 2021 18:13:01 GMT):
Not today, but Real Soon Now. The main focus of the Indy Maintainers right now is a new/modern CI/CD process, and the production of a Ubuntu 20.04 node instance. We're onto the CD part of the work.

MikeSmith (Thu, 22 Jul 2021 18:16:28 GMT):
@swcurran Thanks for your reply. Can we expect the release within this year?

MikeSmith (Thu, 22 Jul 2021 18:16:28 GMT):
@swcurran Thanks for your reply. When is community planning to release it? This year?

cpux07 (Thu, 22 Jul 2021 18:49:09 GMT):
Has joined the channel.

MikeSmith (Thu, 22 Jul 2021 19:04:37 GMT):
Hi Team, I am making a local setup of an Indy node and now I want to integrate an explorer like Indyscan or Etherscan. Could anyone guide me if there is any tool that can help me achieve this task?

playswithblocks (Thu, 22 Jul 2021 19:06:31 GMT):
Has joined the channel.

bt333 (Thu, 22 Jul 2021 19:08:28 GMT):
Hello Team, I am making a local setup of an Indy node and now I want to integrate an explorer like indyscan or etherscan. Could anyone guide me if there is any tool that can help me achieve this task?

allysonugarte (Thu, 22 Jul 2021 19:21:05 GMT):
Has joined the channel.

swcurran (Thu, 22 Jul 2021 19:32:39 GMT):
Yes -- we hope within a month.

lynn.bendixsen (Thu, 22 Jul 2021 19:32:41 GMT):
Indyscan, written by Patrik Stas is open-source, and that is the baseline I used to do https://indyscan.indiciootech.io Patrik made it pretty easy to change just a few things to make it work for your own ledger. Here's a link to my notes that I made while doing this first on my MAC locally, then on a public server: https://docs.google.com/document/d/1noHYmudrIyeqDh_Gitjo8LGXAC9duo7OpgYomvIGkoQ

lynn.bendixsen (Thu, 22 Jul 2021 19:33:53 GMT):
@bt333 ^^

LucyLow (Fri, 23 Jul 2021 01:10:07 GMT):
Has joined the channel.

MikeSmith (Fri, 23 Jul 2021 04:59:16 GMT):
@lynn.bendixsen Thanks for the help.

ManojTaleka (Fri, 23 Jul 2021 06:29:27 GMT):
Has joined the channel.

GrahamLeach (Fri, 23 Jul 2021 08:17:28 GMT):
Has joined the channel.

gut (Fri, 23 Jul 2021 08:34:01 GMT):

gut (Fri, 23 Jul 2021 08:34:01 GMT):
Dear all. As I continue studying Indy+Aries projects I'm getting lost. Are the frameworks a substitute to the SDK+Agent stack or should they be combined? I'm seeing duplicate functionality between all these projects.

gut (Fri, 23 Jul 2021 08:40:09 GMT):
And if I use aries-vcx, does it add one more agent to my system?

Shania-xian (Fri, 23 Jul 2021 10:28:15 GMT):
Has joined the channel.

Jess_Hurwitz (Fri, 23 Jul 2021 22:42:49 GMT):
Has joined the channel.

Alameenb (Sat, 24 Jul 2021 00:15:36 GMT):
Has joined the channel.

ashutoshsharma01 (Sat, 24 Jul 2021 07:02:26 GMT):
Has joined the channel.

ashutosh1 (Sat, 24 Jul 2021 12:35:39 GMT):
Has joined the channel.

db2127 (Sat, 24 Jul 2021 15:04:14 GMT):
Has joined the channel.

slsanc (Sat, 24 Jul 2021 16:32:12 GMT):
Has joined the channel.

mirshahzad (Sat, 24 Jul 2021 16:33:57 GMT):
Has joined the channel.

mengxi123 (Sun, 25 Jul 2021 05:17:02 GMT):
Has joined the channel.

qiqo (Sun, 25 Jul 2021 05:17:10 GMT):
Has joined the channel.

NajetBENHAMOUDA (Sun, 25 Jul 2021 11:15:46 GMT):
Has joined the channel.

lavaicer (Sun, 25 Jul 2021 12:04:59 GMT):
Has joined the channel.

shumpeishimizu (Mon, 26 Jul 2021 04:21:04 GMT):
Has joined the channel.

Shweta1 (Mon, 26 Jul 2021 06:38:55 GMT):

Clipboard - July 26, 2021 12:00 PM

Shweta1 (Mon, 26 Jul 2021 06:38:57 GMT):
Hi Team.When I trying to add new indy node into existing indy network,getting attached error.Even i am not able to add new Trustee/Steward into ledger.Getting same error.

PatrikStas (Mon, 26 Jul 2021 10:16:01 GMT):
Kudos to lynn for writing the guide! However I'd just add it's you can start Indyscan+indy locally thttps://github.com/Patrik-Stas/indyscan/#start-locally-in-localhost

PatrikStas (Mon, 26 Jul 2021 10:16:01 GMT):
Kudos to lynn for writing the guide! However I'd just add it's you can start Indyscan+indy locally esentially running 1 command and then adjusting your /etc/hosts thttps://github.com/Patrik-Stas/indyscan/#start-locally-in-localhost

PatrikStas (Mon, 26 Jul 2021 10:16:01 GMT):
Kudos to lynn for writing the guide! However I'd just add it's you can start Indyscan+indy locally esentially running 1 command and then adjusting your /etc/hosts https://github.com/Patrik-Stas/indyscan/#start-locally-in-localhost

PatrikStas (Mon, 26 Jul 2021 10:16:01 GMT):
Kudos to lynn for writing the guide! However I'd just add it's you can start Indyscan+indy locally esentially running 1 command and then adjusting your /etc/hosts https://github.com/Patrik-Stas/indyscan/tree/master/start

PatrikStas (Mon, 26 Jul 2021 10:16:01 GMT):
Kudos to Lynn for writing the guide! However I'd just add it's you can start Indyscan+indy locally esentially running 1 command and then adjusting your /etc/hosts https://github.com/Patrik-Stas/indyscan/tree/master/start

amarnadh (Mon, 26 Jul 2021 10:53:22 GMT):
Has joined the channel.

aspannag (Mon, 26 Jul 2021 11:41:04 GMT):
Has joined the channel.

HelliHell (Mon, 26 Jul 2021 14:56:24 GMT):
Has joined the channel.

Shweta1 (Mon, 26 Jul 2021 15:12:36 GMT):

Clipboard - July 26, 2021 8:33 PM

Shweta1 (Mon, 26 Jul 2021 15:12:40 GMT):

Clipboard - July 26, 2021 8:33 PM

Shweta1 (Mon, 26 Jul 2021 15:12:45 GMT):
Hi Team,when I adding new node into existing indy network,new node config goes succesfully to ledger.but when i try to see pool ,getting error.Even i can't see new validator node into ledger

sukrutsathaye (Mon, 26 Jul 2021 16:24:15 GMT):
Has joined the channel.

cel (Mon, 26 Jul 2021 19:37:55 GMT):
Has joined the channel.

cpsatav (Tue, 27 Jul 2021 06:52:48 GMT):
Hello Everyone ! I am looking for case for deploying Sovrin on Mainnet Can anybody help me with resources please ?

aymankhan (Tue, 27 Jul 2021 10:36:19 GMT):
Has joined the channel.

WadeBarnes (Tue, 27 Jul 2021 13:03:44 GMT):
Information on Sovrin MainNet can be found here; https://sovrin.org/overview/

WadeBarnes (Tue, 27 Jul 2021 13:03:59 GMT):
Are you looking for something more specific?

ueesha (Tue, 27 Jul 2021 14:15:30 GMT):
Has joined the channel.

ShashankKulkarni1 (Tue, 27 Jul 2021 14:41:48 GMT):
Has joined the channel.

Shweta1 (Tue, 27 Jul 2021 17:33:03 GMT):

Clipboard - July 27, 2021 10:51 PM

Shweta1 (Tue, 27 Jul 2021 17:33:03 GMT):

Clipboard - July 27, 2021 10:51 PM

Shweta1 (Tue, 27 Jul 2021 17:33:03 GMT):

Clipboard - July 27, 2021 10:51 PM

DecentralLabs (Tue, 27 Jul 2021 19:41:05 GMT):
Has joined the channel.

robnik911 (Tue, 27 Jul 2021 21:10:20 GMT):
Has joined the channel.

robnik911 (Tue, 27 Jul 2021 21:25:39 GMT):
Hi all! I would need some help with running and integrating the indy-node or sovrin in a real project. I am working on it almost for a week now, but several packages and links are broken or deprecated, even the getting started docker file from (https://github.com/hyperledger/indy-sdk/blob/master/docs/getting-started/run-getting-started.md) through the Indi-CLI eventually I managed to generate an own DID, but not even with that I was able to connect to the created sandbox in the end, only to a sandboxes, which were in the list already. I guess there was some issue with the IP addresses. Please, I am a junior software developer, and not a big master of computer science. However it would be really important to me, to integrate some way the DID based connection in our project, but need some help or further explanation, from someone who already did that.

robnik911 (Tue, 27 Jul 2021 21:25:39 GMT):
Hi all! I would need some help with running and integrating the indy-node or sovrin in a real project. I am working on it almost for a week now, but several packages and links are broken or deprecated, even the getting started docker file from (https://github.com/hyperledger/indy-sdk/blob/master/docs/getting-started/run-getting-started.md) through the Indi-CLI eventually I managed to generate an own DID, but not even with that I was able to connect to the created sandbox in the end, only to a sandboxes, which were in the list already. I guess there was some issue with the IP addresses. Please, I am a junior software developer, and not a big master of computer science. However it would be really important to me, to integrate some way the DID based connection in our project, but need some help or further explanation, from someone who already did that. Ubuntu 20.04.2 LTS

swcurran (Tue, 27 Jul 2021 22:50:43 GMT):
Chances are you are looking to create an agent -- a component to use verifiable credentials -- issue, hold or verify them. Assuming so, you should be looking at Aries, not Indy. Aries enable you to build agents to exchange verifiable credentials. Take a look at the getting started part of the https://github.com/hyperledger/aries-cloudagent-python README. We've put together a course on becoming an Aries developer on edX -- https://www.edx.org/course/becoming-a-hyperledger-aries-developer

robnik911 (Tue, 27 Jul 2021 22:54:07 GMT):
Thank you so much, I really appreciate your answer and help. Will continue on this thread. :thumbsup:

knagware (Wed, 28 Jul 2021 04:38:44 GMT):
Has joined the channel.

cpsatav (Wed, 28 Jul 2021 06:52:19 GMT):
Thanks for the reply, My concern is to look for a used case of Sovrin apart from BC Gov case.

gaberasturi (Wed, 28 Jul 2021 07:45:48 GMT):
Hello, I have one doubt about log levels in VON-NETWORK. At docker-compose level we have two log related environment variables. - LOG_LEVEL=${LOG_LEVEL} - RUST_LOG=${RUST_LOG} I suppose the levels are: INFO, WARNING, ERROR, DEBUG, TRACE But how can I enable diferents levels? or how do we need to config then? i mean how is the "string" needed to enable f.e debug level. Its posible to have granularity of logs only enabling at some of them Indy, zqm, cripto,...? Following the next link https://jira.hyperledger.org/browse/INDY-1104 We are able to enable debug on each node editing `/etc/indy/indy_config.py` and adding `logLevel = 0` but want to know if there is an easier way Thanks in advanced.

sergio.anguita (Wed, 28 Jul 2021 11:23:56 GMT):
@bt333 have you tried something like https://indyscan.io/home/SOVRIN_MAINNET?

sergio.anguita (Wed, 28 Jul 2021 11:24:25 GMT):
its opensourced at github (https://github.com/Patrik-Stas/indyscan) - ISC LICENSE

WadeBarnes (Wed, 28 Jul 2021 11:40:58 GMT):
Could you please open a new issue on `von-network` for this please; https://github.com/bcgov/von-network/issues. We'll be happy to answer your questions there.

WadeBarnes (Wed, 28 Jul 2021 11:46:12 GMT):
Have you looked at [von-network](https://github.com/bcgov/von-network)? It's a indy-node based ledger designed for use during local development. It has a ledger browser built in so your can review the transactions on the ledger.

WadeBarnes (Wed, 28 Jul 2021 11:47:04 GMT):
Example hosted instance; http://test.bcovrin.vonx.io/

WadeBarnes (Wed, 28 Jul 2021 11:50:39 GMT):
When running in AWS it's important the nodes are initialized with their internal IP addresses, while when registered with the network you need to use the public IP address so nodes are able to reach each other.

WadeBarnes (Wed, 28 Jul 2021 11:53:08 GMT):
You can use `nc` to test the connection from each node to the other nodes. example `nc -vz `

WadeBarnes (Wed, 28 Jul 2021 11:55:58 GMT):
Use cases are somewhat endless. Any use case that can benefit from verifiable credentials: - Identity - Licensing - Proof of origin - Authentication and Authorization - etc.

WadeBarnes (Wed, 28 Jul 2021 11:57:16 GMT):
Are you specifically looking for case studies?

Shweta1 (Wed, 28 Jul 2021 13:12:09 GMT):
Thanks Wade,it is connected.but how pool_transactions_gensis will be update.this file have only first four nodes

WadeBarnes (Wed, 28 Jul 2021 13:15:12 GMT):
The `pool_transactions_gensis` file will only contain what is originally (manually) written into it, it does not get updated automatically as new nodes are added to the pool. It is only used to bootstrap connections to the ledger. Once a connection is made a node or client can read to pool ledger to get the full list and state of the nodes in the pool.

WadeBarnes (Wed, 28 Jul 2021 13:22:15 GMT):
For example, the gensis file for Sovrin BuilderNet only contains four entries; https://raw.githubusercontent.com/sovrin-foundation/sovrin/stable/sovrin/pool_transactions_builder_genesis

WadeBarnes (Wed, 28 Jul 2021 13:22:15 GMT):
For example, the gensis file for Sovrin BuilderNet only contains four entries; https://raw.githubusercontent.com/sovrin-foundation/sovrin/stable/sovrin/pool_transactions_builder_genesis. While the network has 19 nodes currently.

Shweta1 (Wed, 28 Jul 2021 13:22:46 GMT):
Thanks WadeBarnes!! so this pool_transactions_genesis will remain same.When ever i am initializing new node,this pool transaction file should be provided first.

WadeBarnes (Wed, 28 Jul 2021 13:24:00 GMT):
Yes

ChristosPaschalidis (Wed, 28 Jul 2021 14:48:59 GMT):
Has joined the channel.

bt333 (Wed, 28 Jul 2021 17:12:51 GMT):
Thanks, it really helped

bt333 (Wed, 28 Jul 2021 17:15:09 GMT):
Hello Team, Can anyone please help me with N = 3f+1 , How does it actually works? Is it a possibility formula if there is 1 malicious node then there should be a total 4 nodes to handle it? I am trying to understand the process if a pool gets created of 4 loyal nodes and then 4 malicious nodes join the pool. How it can be prevented?

swcurran (Wed, 28 Jul 2021 18:12:57 GMT):
It means if you have N nodes, you can handle up to f failures, where f = round((N - 1)/3). So if you have a 25 node network, you can sustain up to f = 8 failures. Likewise, if you have a 4 node network, you can sustain f =1 failures. The balance is that the more nodes, the slower the consensus, so you have to balance speed with network resiliency. A failure is either a node outage or a detected malicious node.

zenilvaghasiya (Wed, 28 Jul 2021 22:40:55 GMT):
Has joined the channel.

mccown (Thu, 29 Jul 2021 03:53:38 GMT):
The Identity Implementer's WG is meeting today / tomorrow (July 28th @ 9am MT / 3pm UTC). In this call, we will review updates from several industry WGs, which you may have missed. This week, Horacio Nunez (Kiva) will present an overview of Kiva’s SSI work and how it drives their innovative products. Wiki: https://wiki.hyperledger.org/display/IWG/2021-07-29+%3A+Identity+Implementers+WG+Call Zoom: https://zoom.us/my/hyperledger.community.3?pwd=UE90WHhEaHRqOGEyMkV3cldKa2d2dz09

gaberasturi (Thu, 29 Jul 2021 06:22:41 GMT):
Of course. Thank you.

nedaa2020 (Thu, 29 Jul 2021 07:19:48 GMT):
Has joined the channel.

LeslytH (Thu, 29 Jul 2021 08:01:25 GMT):
Has joined the channel.

joarder97 (Thu, 29 Jul 2021 09:14:14 GMT):
Has joined the channel.

Drilmo (Thu, 29 Jul 2021 11:58:15 GMT):
Has joined the channel.

Shweta1 (Thu, 29 Jul 2021 12:36:16 GMT):
How can i remove indy node from existng network.How this "ledger node target .." command work?how can i remove node from indy-cli command

thongtran (Thu, 29 Jul 2021 15:57:18 GMT):
Has joined the channel.

bizsecure (Thu, 29 Jul 2021 16:15:40 GMT):
Has joined the channel.

bizsecure (Thu, 29 Jul 2021 16:15:40 GMT):
Good Morning, we noticed the sovrin repository is not available anymore. Any advice on what might have changed? repo link https://repo.sovrin.org/repository/maven-public/com/facebook/react/react-native/maven-metadata.xml

agentcooper75 (Thu, 29 Jul 2021 18:31:00 GMT):
Has joined the channel.

AnuragChauhan (Fri, 30 Jul 2021 03:29:17 GMT):
Has joined the channel.

JuanEstrada (Fri, 30 Jul 2021 03:36:38 GMT):
Has joined the channel.

rsharmajp (Fri, 30 Jul 2021 07:59:53 GMT):
Has joined the channel.

pp371 (Fri, 30 Jul 2021 09:42:26 GMT):
Has joined the channel.

WadeBarnes (Fri, 30 Jul 2021 12:10:42 GMT):
Use the `ledger node target` with `role=`.

Shweta1 (Fri, 30 Jul 2021 12:41:47 GMT):
Thanks WadeBarnes!!can i edit node ip.like i want to start same node on different host machine.so for that first i have to stop node and then again add to to the network?

Shweta1 (Fri, 30 Jul 2021 12:41:47 GMT):
Thanks WadeBarnes!!can i edit existing validator indy node IP. I want to start same node on different host machine.so for that first i have to stop node1 and then again add this node to the network from another host?

WadeBarnes (Fri, 30 Jul 2021 13:05:35 GMT):
Same process, run `ledger node target` with the new IP address.

janathraveendra (Fri, 30 Jul 2021 15:07:50 GMT):
Has joined the channel.

SelvamaniGovindaraj (Fri, 30 Jul 2021 16:01:55 GMT):
Has joined the channel.

proberge (Fri, 30 Jul 2021 19:34:05 GMT):
Has joined the channel.

Minnu (Fri, 30 Jul 2021 20:15:39 GMT):
Has joined the channel.

akash242891 (Sun, 01 Aug 2021 12:35:50 GMT):
Has joined the channel.

smfk (Sun, 01 Aug 2021 18:52:58 GMT):
Has joined the channel.

fboris (Sun, 01 Aug 2021 19:07:49 GMT):
Has joined the channel.

udaya.rkrish (Sun, 01 Aug 2021 21:44:55 GMT):
Has joined the channel.

KentBull (Mon, 02 Aug 2021 01:54:03 GMT):
Has joined the channel.

NgugsJohn (Mon, 02 Aug 2021 02:32:03 GMT):
Has joined the channel.

maurolionp (Mon, 02 Aug 2021 06:39:07 GMT):
Has joined the channel.

wcc19840827 (Mon, 02 Aug 2021 07:55:33 GMT):
Has joined the channel.

Shweta1 (Mon, 02 Aug 2021 12:59:47 GMT):
Thansks WadeBarnes !!!when I am going to edit existing node ip,I need to update only the node ip .Rest parameters will remain same.so who will update node ip.Existing steward or do we need to create new Steward.I am getting error "Transaction has already reject ,already has a node"

WadeBarnes (Mon, 02 Aug 2021 13:05:48 GMT):
When changing an existing node's configuration you don't specify all of the parameters, only some. For example removing a node from the network would be `ledger node target= alias= services=`. `arget= alias=` always required.

WadeBarnes (Mon, 02 Aug 2021 13:05:48 GMT):
When changing an existing node's configuration you don't specify all of the parameters, only some. For example removing a node from the network would be `ledger node target= alias= services=`. `arget= alias=` are always required.

WadeBarnes (Mon, 02 Aug 2021 13:05:48 GMT):
When changing an existing node's configuration you don't specify all of the parameters, only some. For example removing a node from the network would be `ledger node target= alias= services=`. `target= alias=` are always required.

Shweta1 (Mon, 02 Aug 2021 13:06:51 GMT):
existing steward can edit node ?

WadeBarnes (Mon, 02 Aug 2021 13:09:57 GMT):
The steward of the node can edit it's settings, yes.

Shweta1 (Mon, 02 Aug 2021 13:12:40 GMT):

Clipboard - August 2, 2021 6:33 PM

WadeBarnes (Mon, 02 Aug 2021 13:15:10 GMT):
That's because there is a rule that a steward can only have a single node on any give network.

WadeBarnes (Mon, 02 Aug 2021 13:15:10 GMT):
That's because there is a rule that a steward can only have a single node on any given network.

Shweta1 (Mon, 02 Aug 2021 13:21:54 GMT):

Clipboard - August 2, 2021 6:42 PM

Shweta1 (Mon, 02 Aug 2021 13:21:54 GMT):

Clipboard - August 2, 2021 6:42 PM

Shweta1 (Mon, 02 Aug 2021 13:21:54 GMT):

Clipboard - August 2, 2021 6:42 PM

WadeBarnes (Mon, 02 Aug 2021 14:12:02 GMT):
Are you trying to add a new node, or edit an existing node. You've stated both.

DanielHaren (Mon, 02 Aug 2021 14:20:47 GMT):
Has joined the channel.

WadeBarnes (Mon, 02 Aug 2021 14:29:42 GMT):
Switched to DM.

AlessandraC (Mon, 02 Aug 2021 14:33:10 GMT):
Has joined the channel.

GabrieleVitae (Mon, 02 Aug 2021 14:53:19 GMT):
Has joined the channel.

lynn.bendixsen (Mon, 02 Aug 2021 16:24:45 GMT):
Thanks for the tip @PatrikStas !

cod3rfabric (Mon, 02 Aug 2021 16:35:01 GMT):
Has joined the channel.

unnati (Mon, 02 Aug 2021 17:32:16 GMT):
Has joined the channel.

KentBull (Mon, 02 Aug 2021 17:53:28 GMT):
This makes sense. Good overview. Thanks @lynn.bendixsen

praveenkpandu (Mon, 02 Aug 2021 22:40:58 GMT):
Has joined the channel.

rubix982 (Tue, 03 Aug 2021 02:30:46 GMT):
Has joined the channel.

areyal (Tue, 03 Aug 2021 05:06:06 GMT):
Has joined the channel.

pavithra8895 (Tue, 03 Aug 2021 05:49:09 GMT):
Has joined the channel.

pavithra8895 (Tue, 03 Aug 2021 06:07:01 GMT):
someone please help me I am getting this error on windows 10 and the command I am using is "./network.sh deployCC -ccn basic -ccp ../asset-transfer-basic/chaincode-javascript/ -ccl javascript" and the error I am getting is : Error: chaincode install failed with status: 500 - failed to invoke backing implementation of 'InstallChaincode': could not build chaincode: docker build failed: docker image inspection failed: cannot connect to Docker endpoint Chaincode installation on peer0.org1 has failed Deploying chaincode failed Docker version 20.10.7, build f0df350 docker-compose version 1.29.2, build 5becea4c I am new to hyledger-fabric

bh4rtp (Tue, 03 Aug 2021 07:04:50 GMT):
Has joined the channel.

AboliShinde (Tue, 03 Aug 2021 08:34:03 GMT):
Has joined the channel.

Bandana 1 (Tue, 03 Aug 2021 13:08:30 GMT):
Has joined the channel.

BrunoMartins3 (Tue, 03 Aug 2021 13:48:04 GMT):
Has joined the channel.

PedroBranco (Tue, 03 Aug 2021 16:44:32 GMT):
Has joined the channel.

LAYLeangsros (Tue, 03 Aug 2021 17:01:19 GMT):
Has joined the channel.

rud9321 (Tue, 03 Aug 2021 19:32:31 GMT):
Has joined the channel.

LuisaB 2 (Wed, 04 Aug 2021 05:19:24 GMT):
Has joined the channel.

mcourtin (Wed, 04 Aug 2021 07:41:46 GMT):
Has joined the channel.

jlochman (Wed, 04 Aug 2021 09:03:13 GMT):
Has joined the channel.

SwapnaliDive (Wed, 04 Aug 2021 10:02:22 GMT):
Hi Team, I am trying to setup von network locally and following the steps which are given here _https://github.com/bcgov/von-network#running-the-web-server-on-your-local-machine. _ When I am executing *"GENESIS_URL=https://raw.githubusercontent.com/sovrin-foundation/sovrin/master/sovrin/pool_transactions_sandbox_genesis LEDGER_SEED=000000000000000000000IanCostanzo PORT=9000 python -m server.server" * It is giving error for _No module named 'indy' _ And we resolved this using pip install python3_indy Later we got error for No module named 'plenum' And we tried to resolve using * pip install indy-plenum* But It is failing with messages below _Failed building wheel for indy-plenum Failed building wheel for orderedset Failed building wheel for psutil_ and so on.. Could someone help us here? our python --version is 3.5.2 and python3 --version is 3.6.13 and pip and pip3 --version is 20.3.4 (python 3.6)

amitpadmani-awts (Wed, 04 Aug 2021 12:39:19 GMT):
Has joined the channel.

WadeBarnes (Wed, 04 Aug 2021 13:06:18 GMT):
Is there a particular reason you are running it directly on your machine rather than running it in the docker environment; `./manage build`, `./manage start`? Using the docker environment contains all of the dependencies.

WadeBarnes (Wed, 04 Aug 2021 13:09:36 GMT):
If you are just trying to connect the ledger browser to another network , you are better off doing it this way; https://github.com/bcgov/von-network#running-the-web-server-in-docker-against-another-ledger

SwapnaliDive (Wed, 04 Aug 2021 13:14:12 GMT):
Thanks for your reply. We are trying to use von network web app where we can review the transctions occuring on the ledger. We have set up our indy network and now we want a monitoring service. We are trying to run von network locally so that we can pass our genesis file. But during the setup itself we are getting above errors.

WadeBarnes (Wed, 04 Aug 2021 13:57:11 GMT):
It might just be easier to make your genesis file web accessible so `von-network` can download it. Or, update the code and docker environment to be able to accept (mount) a genesis file. Then you could contribute that back to the project for others to use.

WadeBarnes (Wed, 04 Aug 2021 13:58:11 GMT):
For Network/Node monitoring there is this project; https://github.com/hyperledger/indy-node-monitor

amunoz (Wed, 04 Aug 2021 19:53:23 GMT):
Has joined the channel.

cmstonecse (Thu, 05 Aug 2021 03:29:31 GMT):
Has joined the channel.

woodymk (Thu, 05 Aug 2021 09:33:03 GMT):
Has joined the channel.

sagaracharya (Thu, 05 Aug 2021 10:28:18 GMT):
Has joined the channel.

MikeSmith (Thu, 05 Aug 2021 10:33:02 GMT):
Hello Team, I wanted to know if I create a pool on aws instance using the steps mentioned in hyperledger indy documentation `https://hyperledger-indy.readthedocs.io/projects/node/en/latest/start-nodes.html` So my poolTransaction genesis file which gets created at path /var/lib/indy/sandbox, If i want to make it web accessible All I need to do is replace the nodeIp and clientIP from 127.0.0.1 to AWS instance's public address. Could someone confirm? Is there anything else required to make genesis web accessible?

abraxasccs (Thu, 05 Aug 2021 11:02:57 GMT):
Has joined the channel.

rsebastiany (Thu, 05 Aug 2021 11:27:48 GMT):
Has joined the channel.

WadeBarnes (Thu, 05 Aug 2021 11:49:54 GMT):
Resources; https://github.com/hyperledger/indy-node/issues/1674

MithunH 3 (Thu, 05 Aug 2021 12:17:28 GMT):
Has joined the channel.

merakh (Thu, 05 Aug 2021 17:39:32 GMT):
Has joined the channel.

oussema.gharsallaoui (Thu, 05 Aug 2021 17:48:40 GMT):
Has joined the channel.

patricioibarrag (Thu, 05 Aug 2021 18:56:29 GMT):
Has joined the channel.

swcurran (Fri, 06 Aug 2021 22:47:31 GMT):
My cut of the Indy 2021 Q3 Quarterly Report for the Hyperledger TSC can be found here: https://wiki.hyperledger.org/display/TSC/2021+Q3+Hyperledger+Indy Comments, questions, corrections and additions are welcome!

DanteCullari (Sat, 07 Aug 2021 07:09:31 GMT):
Has joined the channel.

wifimarketingsolutions (Sat, 07 Aug 2021 09:34:54 GMT):
Has joined the channel.

JesusMG (Sat, 07 Aug 2021 15:42:02 GMT):
Has joined the channel.

vova777 (Sat, 07 Aug 2021 18:57:30 GMT):
Has joined the channel.

IshanChowdhury (Sun, 08 Aug 2021 10:04:34 GMT):
Has joined the channel.

jljordan_bcgov (Sun, 08 Aug 2021 18:05:25 GMT):
Thanks for the report. I'm really glad to see the pipeline getting cleaned up which will allow progress for new features. Is it possible to share transaction activity on the staging and production networks for Sovrin as one of the Long running DLTs?

swcurran (Sun, 08 Aug 2021 19:22:21 GMT):
The activity information for Sovrin is available for [MainNet](https://indyscan.io) and here for [StagingNet](https://indyscan.io/home/SOVRIN_STAGINGNET). Sovrin also has a Metrics page here, with links to each network: https://sovrin.org/ssi-metrics-dashboards/

jljordan_bcgov (Sun, 08 Aug 2021 20:44:54 GMT):
👍🏼

luisabrunhosa (Sun, 08 Aug 2021 21:55:45 GMT):
Has joined the channel.

ahmedalsalih (Mon, 09 Aug 2021 03:32:48 GMT):
Has joined the channel.

nskishore1993 (Mon, 09 Aug 2021 07:13:15 GMT):
Has joined the channel.

anhduck (Mon, 09 Aug 2021 08:41:24 GMT):
Has joined the channel.

Mr.S (Mon, 09 Aug 2021 09:43:29 GMT):
Has joined the channel.

Amit18 (Tue, 10 Aug 2021 07:01:00 GMT):
Has joined the channel.

MikeSmith (Tue, 10 Aug 2021 10:33:17 GMT):
Hello Team, Can someone help me to understand this? When we create a pool so we start with few number of nodes those come inside the pool genesis file. Are those nodes steward or trustee ? If these are not trustee then where does trustee exist? Also, what if my all pool *genesis * nodes are down then what will be the impact?

MikeSmith (Tue, 10 Aug 2021 10:33:17 GMT):
Hello Team, Can someone please help me to understand this? When we create a pool so we start with few number of nodes those come inside the pool genesis file. Are those nodes steward or trustee ? If these are not trustee then where does trustee exist? Also, what if my all pool *genesis * nodes are down then what will be the impact?

PuneetSivananda (Tue, 10 Aug 2021 11:36:50 GMT):
Has joined the channel.

Amit18 (Tue, 10 Aug 2021 12:45:21 GMT):
I Hello everyone, I am working on indy node backup.i am using /var/lib/indy/sandbox/Node1/data/... to restore data. 4 nodes are running on host1.Again I am trying to run these 4 nodes on host2.is it possible to restore all indy nodes data from host1 machine to host2.

WadeBarnes (Tue, 10 Aug 2021 13:45:39 GMT):
Nodes only have a single role (at the moment), they are validators. The role of Steward and Trustee is for DIDs. A Steward DID has the permissions to manage their node's configuration on the ledger, while a Trustee is able to perform more privileged operations. The basic roles and permissions can be found here; https://github.com/hyperledger/indy-node/blob/master/docs/source/auth_rules.md

WadeBarnes (Tue, 10 Aug 2021 13:45:59 GMT):
Auth rules can be customized for a given network.

nmoskaleva (Tue, 10 Aug 2021 18:56:53 GMT):
Has joined the channel.

roshan.joseph (Tue, 10 Aug 2021 22:00:31 GMT):
Has joined the channel.

fernandobender (Tue, 10 Aug 2021 22:48:10 GMT):
Has joined the channel.

kkaur01 (Tue, 10 Aug 2021 23:48:31 GMT):
Has joined the channel.

conanoc (Wed, 11 Aug 2021 01:50:18 GMT):
ANNOUNCEMENT: An example of issuing vaccination certificates using Indy This example consists of an android app and a web app showing how to build applications with Indy SDK. Hope this repository helps newcomers in Indy. Any comments would be appreciated. https://github.com/conanoc/indy-sample-agents

VikiAnand (Wed, 11 Aug 2021 04:14:57 GMT):
Has joined the channel.

ghoshbishakh (Wed, 11 Aug 2021 06:50:09 GMT):
Has joined the channel.

ghoshbishakh (Wed, 11 Aug 2021 06:50:10 GMT):
Hi all, I am trying to add a new kind of transaction or modify NYM transactions to include some additional information. Can anyone point me to the relevant places in the code to do that? I am trying to understand indy-node and indy-plenum codebases for a couple of weeks. Looks like I need to add the new type of transaction to plenum. Do I need to make changes to the indy-sdk also?

AndreyShevchenko (Wed, 11 Aug 2021 11:35:09 GMT):
Has joined the channel.

SandeepVankala (Wed, 11 Aug 2021 12:30:45 GMT):
Has joined the channel.

SandeepVankala (Wed, 11 Aug 2021 12:34:39 GMT):
Has left the channel.

swcurran (Wed, 11 Aug 2021 18:12:54 GMT):
Are you able to share what you are trying to add, or are you doing this on a fork for your own use?

swcurran (Wed, 11 Aug 2021 18:14:27 GMT):
Not sure where in the indy-node/plenum code. If you are modifying the NYM transaction, I think (not certain) that you only have to change indy-node. If you are using indy-sdk to send transactions, then yes, you also have to update indy-sdk with the new information being sent.

BrianRani (Wed, 11 Aug 2021 19:13:50 GMT):
Has joined the channel.

bentoumia (Wed, 11 Aug 2021 21:53:59 GMT):
Has joined the channel.

krupakar_linux (Thu, 12 Aug 2021 06:58:01 GMT):
Has joined the channel.

krupakar_linux (Thu, 12 Aug 2021 06:58:03 GMT):
We normally using docker volumes outside of container, and trying to remount when running on a new instances. We face challenge that the existing data files have the old instance ip addresses and not able to resolve on the new instances.

manfredmeyer (Thu, 12 Aug 2021 09:21:25 GMT):
Has joined the channel.

marcocomo (Thu, 12 Aug 2021 11:16:31 GMT):
Has joined the channel.

Amit18 (Thu, 12 Aug 2021 12:52:19 GMT):
How I can restore indy node data on another host machine?

Amit18 (Thu, 12 Aug 2021 12:52:19 GMT):
How I can restore indy node data on another host machine?Nodes are running on two different host machines and remote host ip is provided into pool transaction genesis file

Amit18 (Thu, 12 Aug 2021 12:52:19 GMT):
How I can restore indy node data .Nodes are running on two different host machines and remote host ip is provided into pool transaction genesis file.In case if host machines crashed,how I can restore indy node data on new host.

mccown (Thu, 12 Aug 2021 13:45:45 GMT):
Just a reminder that the Identity Implementer’s WG call is cancelled for today (Aug 12th). Please join us for our next meeting on Aug 26th. Thanks!

lynn.bendixsen (Thu, 12 Aug 2021 14:32:46 GMT):
If all pool genesis nodes are down, then any agent or entity that only has the genesis file will not be able to connect to the network.

lynn.bendixsen (Thu, 12 Aug 2021 14:42:43 GMT):
Hi @Amit18, At a somewhat high level, if only some of your Nodes crash, then the data directory on those Nodes will be automatically populated when you bring the Nodes back up. This is called "catchup", I think, and it happens on Node startup or when you first add a node to the network. If all nodes go down, then complications might arise that I am not 100% sure about. There are some checks that are performed based on information in the audit ledger. I am pretty sure that an identical copy of the Data directory could be restored to every node on the network and then it would work, but my information might be outdated. Try it out and let us know :)

lynn.bendixsen (Thu, 12 Aug 2021 14:50:57 GMT):
That's a tough one because not even the genesis files will work for getting connected to that network. Changing the IP addresses of all of your genesis file nodes, means that you need to create new genesis files to be able to connect. The Indy Blockchain architecture does not seem to mesh well with the scenario that you propose...

Amit18 (Thu, 12 Aug 2021 15:02:16 GMT):
yes i tried to mount existing /var/lib/indy/sandbox/Node1/data folder to node1 started on new host.I updated new node ip in pool_transacton_genesis and node_info where ip is present.but that doesn't work

Amit18 (Thu, 12 Aug 2021 15:02:16 GMT):
yes i tried to mount existing /var/lib/indy/sandbox/Node1/data folder to node1 started on new host.I updated new node ip in pool_transacton_genesis and node_info where ip is present.but that doesn't work.i can't see older data.but have new 10 records only.

lynn.bendixsen (Thu, 12 Aug 2021 15:11:19 GMT):
Updating the genesis file (changing it) is not allowed. It violates the blockchain "immutable ledgers" notion. The only way to update a genesis file is to extend it to be larger so that it includes the additional changes made , for example, to update the IP addresses while the ledger is live. The only other way to do what you are talking about is to reinstantiate the ledger, then replay all of the transactions that you want to have on your new ledger.

lynn.bendixsen (Thu, 12 Aug 2021 15:11:19 GMT):
Updating the genesis file (changing it) is not allowed. It violates the blockchain "immutable ledgers" notion. The only way to update a genesis file is to extend it to be larger so that it includes the additional changes made , for example, to update the IP addresses while the ledger is live. The only other way to do what you are talking about is to reinstantiate the ledger with your new genesis file, then replay all of the transactions that you want to have on your new ledger.

lynn.bendixsen (Thu, 12 Aug 2021 15:11:19 GMT):
Updating the genesis file (changing the transactions in it) is not allowed. It violates the blockchain "immutable ledgers" notion. The only way to update a genesis file is to extend it to be larger so that it includes the additional changes made , for example, to update the IP addresses while the ledger is live. The only other way to do what you are talking about is to reinstantiate the ledger with your new genesis file, then replay all of the transactions that you want to have on your new ledger.

krupakar_linux (Thu, 12 Aug 2021 16:00:17 GMT):
Ideally, the IP addresses where we are running the node shouldn't be stored in ledger and assuming them as part of configuration files. In hyperledger fabric we use alias names for peers and orderers. is there any notation that we can use in Indy node?

Shweta1 (Thu, 12 Aug 2021 16:28:28 GMT):
I have 4 nodes in the network.My host machine is going to rehydrate I have updated Node1 and Node2 IP.but when I try to update Node3 IP getting error "Transaction rejected" After that I am not able to do transaction

andreolf1 (Thu, 12 Aug 2021 16:49:11 GMT):
Has joined the channel.

phalves (Thu, 12 Aug 2021 17:08:12 GMT):
Has joined the channel.

lynn.bendixsen (Thu, 12 Aug 2021 21:15:21 GMT):
Transaction rejected occurs when there are not enough nodes in the pool that are in consensus. Without more details, it's difficult for me to tell what happened to the other nodes (for a 4 node network, you need 3 to be up and participating in consensus). Your log files might help you determine what has happened to the other nodes. /var/log/indy//.log

lynn.bendixsen (Thu, 12 Aug 2021 21:22:00 GMT):
None that I know of. Indy does not allow for anything like that, ,and I am not sure why, but it's probably for security reasons.

phearaeun (Fri, 13 Aug 2021 02:26:03 GMT):
Has joined the channel.

ahmetcakmak606 (Fri, 13 Aug 2021 08:19:02 GMT):
Has joined the channel.

ataberkozek (Fri, 13 Aug 2021 08:30:39 GMT):
Has joined the channel.

schlagtim (Fri, 13 Aug 2021 10:28:48 GMT):
Has joined the channel.

OzodbekXomidbekov (Fri, 13 Aug 2021 12:02:21 GMT):
Has joined the channel.

OzodbekXomidbekov (Fri, 13 Aug 2021 12:06:15 GMT):
Hello everyone,

OzodbekXomidbekov (Fri, 13 Aug 2021 12:08:00 GMT):
Hello everyone, I am a new in Hyperledger Indy. I am trying to install Hyperledger Indy on Windows following this link

OzodbekXomidbekov (Fri, 13 Aug 2021 12:10:17 GMT):
Hello everyone, I am a new in Hyperledger Indy. I am trying to install Hyperledger Indy on Windows following this link. https://hyperledger-indy.readthedocs.io/projects/sdk/en/latest/docs/build-guides/windows-build.html I cannot handle this task "Create an empty static library project in Visual Studio and add sqlite.c file and 2 headers from extracted archive. ". I cannot add sqlite.c file to the static library. Can anyone help me to install Hyperledger Indy please ? Thank you in advance !

WadeBarnes (Fri, 13 Aug 2021 13:40:43 GMT):
What are your plans? There are easier ways to start with Hyperledger Indy/Aries that use containerization. Installing on bare metal, especially on Windows, can be challenging.

WadeBarnes (Fri, 13 Aug 2021 13:41:49 GMT):
I'd recommend starting with the quick-start and tutorials here; https://github.com/hyperledger/aries-cloudagent-python

SwapnaliDive (Fri, 13 Aug 2021 13:49:07 GMT):
Hello Team, We are trying to migrate one node from one aws instance to another. We have total 5 nodes running when we are trying to migrate 5th node which is running at 5th instance to 6th instance then we are not getting success. Below are the steps that we are following : 1) At 5th instance we are shutting down our 5th node 2) At 6th instance we are copying pool_transactions_genesis and domain_transactions_genesis from 5th instance. 3) On 6th instance we are running -- -- sudo init_indy_node Node5 9711 9712 -- sudo start_indy_node Node5 9711 9712 -- Then from indy-cli, using node owner did we are adding Node5 ledger node target= client_port=9712 client_ip= alias=Node5 node_ip= node_port=9711 services=VALIDATOR blskey= blskey_pop= After following the above steps , other nodes able to find 5th node from new_ip but when we are starting 5th node from 6th instance it is not connecting with other nodes. Could anyone please verify the above steps and guide us ???

WadeBarnes (Fri, 13 Aug 2021 14:16:21 GMT):
When initializing the node on the new host, you need to include the random seed that was generated from the first time you ran the init_indy_node command on the previous host, otherwise a new random seed is generated and all of the keys will be different, so it will appear to the network as a new node with a duplicate name (alias).

WadeBarnes (Fri, 13 Aug 2021 14:16:57 GMT):
The format of the command you need to use is; `sudo -i -u indy init_indy_node `

WadeBarnes (Fri, 13 Aug 2021 14:18:10 GMT):
When you first ran the init command the output would have included a line like this: ``` Generating keys for random seed b'FA7b1cc42Da11B8F4BC83990cECF63aD' ```

WadeBarnes (Fri, 13 Aug 2021 14:19:06 GMT):
It's that seed (the value between the single quotes) you need when re-running the init on the new host.

sebaleoperez (Fri, 13 Aug 2021 14:59:08 GMT):
Has joined the channel.

surya7 (Sat, 14 Aug 2021 14:17:58 GMT):
Has joined the channel.

nishantchandla (Sat, 14 Aug 2021 15:02:31 GMT):
Has joined the channel.

nits2k (Sun, 15 Aug 2021 02:44:42 GMT):
Has joined the channel.

ApoorvSingal (Sun, 15 Aug 2021 17:08:03 GMT):
Has joined the channel.

jeaninejue1 (Sun, 15 Aug 2021 21:54:05 GMT):
Has joined the channel.

fftkim (Mon, 16 Aug 2021 04:30:18 GMT):
Has joined the channel.

ghoshbishakh (Mon, 16 Aug 2021 06:54:19 GMT):
@swcurran thank you for the help! really appreciate it! I am doing this on a fork for my own use. I am trying to add some verification methods in the NYM transaction for including it in the DID. Currently the NYM has only verkey. I want to add more information (which are not attributes).

esther_pan (Mon, 16 Aug 2021 07:39:18 GMT):
Has joined the channel.

mklarmann (Mon, 16 Aug 2021 08:31:19 GMT):
Has joined the channel.

aditya.21991 (Mon, 16 Aug 2021 09:43:26 GMT):
Has joined the channel.

SwapnaliDive (Mon, 16 Aug 2021 10:18:47 GMT):
@WadeBarnes Thank you so much, your solution perfectly working for me. I am able to migrate nodes now.

SwapnaliDive (Mon, 16 Aug 2021 10:41:07 GMT):
Hello Team, I also want to know is it possible to migrate those nodes as well which comes under genesisFile? If yes, does genesis file gets updated automatically?

Amit18 (Mon, 16 Aug 2021 11:36:13 GMT):
is it possible to migrate all nodes from host1 to host2.All nodes are valdiators.

Amit18 (Mon, 16 Aug 2021 11:36:13 GMT):
is it possible to migrate all nodes from host1 to host2.All nodes are valdiators.After Nod1,Node2,Node3 migration,i am able to do ledger transaction.But after succesful Nod4 migration,I am getting error.sometimes Node1 is not reachable or Node 2,Node3,Node4.After several restart of indy nodes,no nodes are reachable to each other

WadeBarnes (Mon, 16 Aug 2021 12:28:59 GMT):
You need to update the genesis file manually.

bt333 (Mon, 16 Aug 2021 13:38:45 GMT):
Could you please confirm this? So, If there is a pool of 100 nodes where 10 are the founder nodes and one of founder node migrated to different IP which will lead to update the genesis file manually. After the manual updation of Genesis file all the remaining nodes will need to update their genesis file manually? They will have to follow pool connection process again? Does it mean that it is recommended that founder nodes should remain at same ip and port for forever?

swcurran (Mon, 16 Aug 2021 13:43:52 GMT):
Interesting. There is a plan to create `did:indy`, and one of the 2 major purposes for that effort is to do what you are doing. Here is the definition of what is planned: https://hackmd.io/@icZC4epNSnqBbYE0hJYseA/S1eUS2BQw I think the change is in indy-node only (on the ledger side), but am not certain. And there would be indy-sdk/indy-vdr changes on the client side to retrieve the data.

luandt1809 (Tue, 17 Aug 2021 04:08:57 GMT):
Has joined the channel.

N_Mizuki (Tue, 17 Aug 2021 05:12:47 GMT):
Has joined the channel.

hoily.dev (Tue, 17 Aug 2021 07:09:27 GMT):
Has joined the channel.

atodorov_kiwi (Tue, 17 Aug 2021 07:53:28 GMT):
Has joined the channel.

GuoshenKuang (Tue, 17 Aug 2021 09:14:36 GMT):
Has joined the channel.

samie (Tue, 17 Aug 2021 11:18:37 GMT):
Has joined the channel.

KevinKerkhoven (Tue, 17 Aug 2021 11:24:11 GMT):
Has joined the channel.

bharadwajambati172 (Tue, 17 Aug 2021 12:19:54 GMT):
Has joined the channel.

GereonHinz (Tue, 17 Aug 2021 13:34:45 GMT):
Has joined the channel.

HamzaHijazi (Tue, 17 Aug 2021 17:14:28 GMT):
Has joined the channel.

kevinhill (Tue, 17 Aug 2021 19:55:51 GMT):
Has joined the channel.

luandt1809 (Wed, 18 Aug 2021 04:20:45 GMT):

Clipboard - August 18, 2021 11:19 AM

luandt1809 (Wed, 18 Aug 2021 04:21:56 GMT):
I installed it, but can't use it according to the instructions in the documentation https://hyperledger-indy.readthedocs.io/projects/sdk/en/latest/docs/design/006-cli-plugins/README.html

BouXez (Wed, 18 Aug 2021 09:35:01 GMT):
Has joined the channel.

bsustudent (Wed, 18 Aug 2021 19:39:15 GMT):
Has joined the channel.

sudng (Wed, 18 Aug 2021 21:49:59 GMT):
Has joined the channel.

hiroyukihonda (Thu, 19 Aug 2021 01:08:28 GMT):
Has joined the channel.

HUOH 1 (Thu, 19 Aug 2021 02:15:59 GMT):
Has joined the channel.

OzodbekXomidbekov (Thu, 19 Aug 2021 06:49:12 GMT):
Dear Wade Barnes, Thank you for your answer! My plans are install Hyperledger indy, run test version. Do some project. Maybe related to DIDs. My thesis subjects is in Hypereledger Indy and DIDs, so I have to do some project on it. I do not know why there is no much information about Indy on youtube.

aravindrajck (Thu, 19 Aug 2021 07:51:51 GMT):
Has joined the channel.

kaveri (Thu, 19 Aug 2021 08:28:35 GMT):
Has joined the channel.

tuyendev (Thu, 19 Aug 2021 09:48:02 GMT):
Has joined the channel.

DianaTique (Thu, 19 Aug 2021 17:36:55 GMT):
Has joined the channel.

mona15 (Thu, 19 Aug 2021 22:12:40 GMT):
Has joined the channel.

huangxiaoqi (Fri, 20 Aug 2021 01:48:24 GMT):
Has joined the channel.

AnuragKumar10 (Fri, 20 Aug 2021 04:08:24 GMT):
Has joined the channel.

MatanNahmani (Fri, 20 Aug 2021 05:06:59 GMT):
Has joined the channel.

norswap (Fri, 20 Aug 2021 08:53:23 GMT):
Has joined the channel.

yigemaser (Fri, 20 Aug 2021 10:42:58 GMT):
Has joined the channel.

hishmad (Fri, 20 Aug 2021 11:10:38 GMT):
Has joined the channel.

bsu (Fri, 20 Aug 2021 11:45:15 GMT):
Has joined the channel.

w1g0 (Fri, 20 Aug 2021 13:28:40 GMT):
Has joined the channel.

Crispin-Sharkaroo (Fri, 20 Aug 2021 18:04:34 GMT):
Has joined the channel.

elman27 (Fri, 20 Aug 2021 19:11:11 GMT):
Has joined the channel.

yogesh.yadav (Sun, 22 Aug 2021 17:12:21 GMT):
Has joined the channel.

mikexiour (Mon, 23 Aug 2021 06:39:14 GMT):
Has joined the channel.

Anasalamin (Mon, 23 Aug 2021 07:00:24 GMT):
Has joined the channel.

malarflower23 (Mon, 23 Aug 2021 07:37:24 GMT):
Has joined the channel.

cinhtw (Mon, 23 Aug 2021 09:45:20 GMT):
Has joined the channel.

OzodbekXomidbekov (Mon, 23 Aug 2021 10:58:52 GMT):
Hi everyone, I am new Hyperledger Indy so no idea to do some project Can anyone help me with my thesis to do a project in Hyperledger Indy ?

OzodbekXomidbekov (Mon, 23 Aug 2021 10:58:52 GMT):
Hi everyone, I am new Hyperledger Indy so no idea to do some project Can anyone help me with my thesis to do a project in Hyperledger Indy or DIDs ?

wonderwoman11 (Mon, 23 Aug 2021 12:40:04 GMT):
Has joined the channel.

ffendt (Mon, 23 Aug 2021 13:02:03 GMT):
Has joined the channel.

svitzebec (Mon, 23 Aug 2021 13:43:53 GMT):
Has joined the channel.

biswa1 (Mon, 23 Aug 2021 18:24:07 GMT):
Has joined the channel.

no_one13 (Mon, 23 Aug 2021 20:13:35 GMT):
Has joined the channel.

Oliver_Tahir (Tue, 24 Aug 2021 05:02:40 GMT):
Has joined the channel.

Matannah 1 (Tue, 24 Aug 2021 05:17:41 GMT):
Has joined the channel.

mirgaon (Tue, 24 Aug 2021 06:46:36 GMT):
Has joined the channel.

tinkerbell (Tue, 24 Aug 2021 07:26:58 GMT):
Has joined the channel.

Jianglh (Tue, 24 Aug 2021 08:50:41 GMT):
Has joined the channel.

hfjohann (Tue, 24 Aug 2021 08:55:47 GMT):
Has joined the channel.

RuiYang (Tue, 24 Aug 2021 10:03:03 GMT):
Has joined the channel.

deathgod86 (Tue, 24 Aug 2021 10:42:31 GMT):
Has joined the channel.

fdiarra (Tue, 24 Aug 2021 11:04:45 GMT):
Has joined the channel.

drewmarshburn (Tue, 24 Aug 2021 12:00:05 GMT):
Has joined the channel.

Shweta1 (Tue, 24 Aug 2021 14:46:41 GMT):
Hi lynn,how i can restore all indy nodes data in case indy network crash.is it possible to create same network with same node config and mount existing indy node data docker volume or migrate all nodes from one host to other

kirstygoodary (Tue, 24 Aug 2021 15:25:49 GMT):
Has joined the channel.

anuragsingh (Tue, 24 Aug 2021 18:31:42 GMT):
Has joined the channel.

jjordan-quantum (Tue, 24 Aug 2021 19:49:38 GMT):
Has joined the channel.

j-s (Tue, 24 Aug 2021 20:46:22 GMT):
Has joined the channel.

prathampatil8119 (Wed, 25 Aug 2021 04:24:32 GMT):
Has joined the channel.

VisheshRaj (Wed, 25 Aug 2021 07:37:36 GMT):
Has joined the channel.

AlkaidCC (Wed, 25 Aug 2021 10:11:46 GMT):
Has joined the channel.

huzhiyang (Wed, 25 Aug 2021 12:29:30 GMT):
Has joined the channel.

huzhiyang (Wed, 25 Aug 2021 12:31:50 GMT):
hello every,I am a new bee to hyperledger Indy, after I installed Indy , I tried to use the command dion@dion-Lenovo:~/下载/indy-sdk/docs/getting-started$ docker-compose up, however there exist a bug which I cannot fix it

huzhiyang (Wed, 25 Aug 2021 12:32:28 GMT):
dion@dion-Lenovo:~/下载/indy-sdk/docs/getting-started$ docker-compose up Building jupyter Step 1/9 : FROM ubuntu:16.04 ---> 38b3fa4640d4 Step 2/9 : RUN useradd -ms /bin/bash indy ---> Using cache ---> 0ce3a35325e3 Step 3/9 : RUN apt-get update -y && apt-get install -y wget python3.5 python3-pip python-setuptools apt-transport-https ca-certificates software-properties-common ---> Using cache ---> a20238d5c57c Step 4/9 : WORKDIR /home/indy ---> Using cache ---> 092f66c64283 Step 5/9 : RUN pip3 install --upgrade pip ---> Using cache ---> 332b60b509c6 Step 6/9 : RUN pip3 install -U pip ipython-notebook ipython==7.9 setuptools jupyter python3-indy==1.11.0 ---> Running in f93cb035534c Traceback (most recent call last): File "/usr/local/bin/pip3", line 7, in from pip._internal.cli.main import main File "/usr/local/lib/python3.5/dist-packages/pip/_internal/cli/main.py", line 57 sys.stderr.write(f"ERROR: {exc}") ^ SyntaxError: invalid syntax ERROR: Service 'jupyter' failed to build: The command '/bin/sh -c pip3 install -U pip ipython-notebook ipython==7.9 setuptools jupyter python3-indy==1.11.0' returned a non-zero code: 1

huzhiyang (Wed, 25 Aug 2021 12:33:06 GMT):
Would there be anyone so nicely to help me ?

WadeBarnes (Wed, 25 Aug 2021 13:11:11 GMT):
@huzhiyang, You may find it easier to get started with https://github.com/hyperledger/aries-cloudagent-python The tutorials and getting started guides are more up to date. It provides a layer of abstraction that hides many of the intricate complexities on Indy.

balaji7777 (Wed, 25 Aug 2021 15:22:34 GMT):
Has joined the channel.

ShubhamUmate (Wed, 25 Aug 2021 17:01:14 GMT):
Has joined the channel.

ArdianAbazi (Wed, 25 Aug 2021 18:11:00 GMT):
Has joined the channel.

ArdianAbazi (Wed, 25 Aug 2021 18:11:00 GMT):
(Indy error code: [0] {\"code\":4,\"extra\":null,\"message\":\"Unknown request handle\"})"}

mccown (Wed, 25 Aug 2021 19:47:28 GMT):
The Identity Implementer's WG is meeting today / tomorrow (Aug 26th @ 9am MT / 3pm UTC). In this call, we will review updates from several industry WGs, which you may have missed. This week, Horacio Nunez (Kiva) will present a new open source desktop tool that will connect secure biometrics with SSI-powered backend services. Wiki: https://wiki.hyperledger.org/display/IWG/2021-08-26+%3A+Identity+Implementers+WG+Call Zoom: https://zoom.us/my/hyperledger.community.3?pwd=UE90WHhEaHRqOGEyMkV3cldKa2d2dz09

mccown (Wed, 25 Aug 2021 19:47:28 GMT):
The Identity Implementer's WG is meeting tomorrow (Aug 26th @ 9am MT / 3pm UTC). In this call, we will review updates from several industry WGs, which you may have missed. This week, Horacio Nunez (Kiva) will present a new open source desktop tool that will connect secure biometrics with SSI-powered backend services. Wiki: https://wiki.hyperledger.org/display/IWG/2021-08-26+%3A+Identity+Implementers+WG+Call Zoom: https://zoom.us/my/hyperledger.community.3?pwd=UE90WHhEaHRqOGEyMkV3cldKa2d2dz09

NimiCorp (Wed, 25 Aug 2021 20:55:19 GMT):
Has joined the channel.

ashishspg (Thu, 26 Aug 2021 05:22:22 GMT):
Has joined the channel.

AskarAlmabek (Thu, 26 Aug 2021 06:02:22 GMT):
Has joined the channel.

Hasib_Linux (Thu, 26 Aug 2021 07:28:01 GMT):
Has joined the channel.

JongHyunPark (Thu, 26 Aug 2021 08:38:03 GMT):
Has joined the channel.

SatyamGupta1 (Thu, 26 Aug 2021 13:56:49 GMT):
Has joined the channel.

ElhamShammar (Thu, 26 Aug 2021 15:30:45 GMT):
Has joined the channel.

noopur2393 (Thu, 26 Aug 2021 15:58:00 GMT):
Has joined the channel.

lynn.bendixsen (Thu, 26 Aug 2021 16:21:56 GMT):
I think that is possible, but really everything would have to be the same, so it would probably be quite difficult to succeed. (Each node would have to restore their own original same keys, same IP's, same aliases, etc) I am not sure about the migrate, all of what I have worked with is multiple separately managed nodes for networks.

pranjay (Thu, 26 Aug 2021 21:51:17 GMT):
Has joined the channel.

Princekumar4812 (Fri, 27 Aug 2021 03:02:21 GMT):
Has joined the channel.

bhuwanupadhyay (Fri, 27 Aug 2021 06:37:11 GMT):
Has joined the channel.

mohsinanees (Fri, 27 Aug 2021 09:51:06 GMT):
Has joined the channel.

sashankh (Fri, 27 Aug 2021 18:58:05 GMT):
Has joined the channel.

j-s (Fri, 27 Aug 2021 20:10:01 GMT):
Hi, I'm trying to connect the indy sample to a production network, instead of the local docker indy pool. The network is set up and I can connect to it via the indy CLI, using the network's genesis file. But when I try to use that same genesis file to connect from my indy-sdk environment (which is local to my machine, same as the indy CLI) it times out (`UnhandledPromiseRejectionWarning: IndyError: PoolLedgerTimeout`). I think this could mean that the network is not configured to accept all the right traffic... but if that's the case how come my indy CLI from the same machine _can_ connect? Has anyone had this issue? Does anyone have an idea?

ShafiqIslam (Fri, 27 Aug 2021 22:34:38 GMT):
Has joined the channel.

thonexin (Sat, 28 Aug 2021 04:41:22 GMT):
Has joined the channel.

TrishPat (Sat, 28 Aug 2021 07:55:35 GMT):
Has joined the channel.

jk-tonycui (Sun, 29 Aug 2021 02:14:52 GMT):
Has joined the channel.

samson9993 (Sun, 29 Aug 2021 08:19:53 GMT):
Has joined the channel.

maurosala (Mon, 30 Aug 2021 07:09:49 GMT):
Has joined the channel.

dsam82 (Mon, 30 Aug 2021 08:53:39 GMT):
Has joined the channel.

rroman 1 (Mon, 30 Aug 2021 09:12:34 GMT):
Has joined the channel.

peppelinux (Mon, 30 Aug 2021 14:04:26 GMT):
Has joined the channel.

WillimJames (Tue, 31 Aug 2021 05:56:25 GMT):
Has joined the channel.

pipscobie (Tue, 31 Aug 2021 11:16:04 GMT):
Has joined the channel.

giorgiomecca (Tue, 31 Aug 2021 12:31:16 GMT):
Has joined the channel.

JJeffrey (Tue, 31 Aug 2021 14:35:26 GMT):
Has joined the channel.

birjoossh (Wed, 01 Sep 2021 00:53:18 GMT):
Has joined the channel.

clemhumb (Wed, 01 Sep 2021 06:20:02 GMT):
Has joined the channel.

roviso (Wed, 01 Sep 2021 07:05:06 GMT):
Has joined the channel.

sunchiye (Wed, 01 Sep 2021 07:46:35 GMT):
Has joined the channel.

ohthemboats (Wed, 01 Sep 2021 08:03:49 GMT):
Has joined the channel.

MahaQahtani (Wed, 01 Sep 2021 08:38:23 GMT):
Has joined the channel.

SheharYar (Wed, 01 Sep 2021 20:43:21 GMT):
Has joined the channel.

Harsh_Vekariya (Thu, 02 Sep 2021 05:31:49 GMT):
Has joined the channel.

ShekharGuptaEN (Thu, 02 Sep 2021 05:47:21 GMT):
Has joined the channel.

martinmaurer (Thu, 02 Sep 2021 06:11:09 GMT):
Has joined the channel.

ArchitaDasgupta (Thu, 02 Sep 2021 11:45:25 GMT):
Has joined the channel.

smyyihvn (Thu, 02 Sep 2021 15:08:15 GMT):
Has joined the channel.

ramu450 (Fri, 03 Sep 2021 05:12:52 GMT):
Has joined the channel.

Omen 8 (Fri, 03 Sep 2021 05:34:42 GMT):
Has joined the channel.

AdityaGupta150 (Fri, 03 Sep 2021 12:15:59 GMT):
Has joined the channel.

anil8753 (Fri, 03 Sep 2021 15:21:54 GMT):
Has joined the channel.

leebradley (Fri, 03 Sep 2021 15:42:06 GMT):
Has joined the channel.

FoamCactus (Fri, 03 Sep 2021 15:54:09 GMT):
Has joined the channel.

alagiesellu (Fri, 03 Sep 2021 20:27:03 GMT):
Has joined the channel.

kangme (Fri, 03 Sep 2021 22:11:27 GMT):
Hi! I am Meng Kang, a Master’s student in Electrical Engineering supervised by Dr.Victria Lemieux. We are looking for people with a background in blockchain, cryptography, or software engineering to volunteer for an anonymous survey. You will first watch a presentation video of our software design research project ‘A Decentralized Identity-Based Blockchain Solution for Privacy-Preserving Licensing of Individual-Owned Data to Prevent Unauthorized Secondary Data Usage’, and then provide your feedback by filling out an online questionnaire. Here is the link for the presentation video: https://drive.google.com/file/d/1X7epdOwUB0SvDOlKrDeGU8l8eKwWSiLd/view?usp=sharing Here is the link to the survey: https://ubc.ca1.qualtrics.com/jfe/form/SV_3C4Yp2FPiHufmM6 Here is the PDF version of the video: https://docs.google.com/presentation/d/1pbjmTVxWe6HQ4Jge-qR0-xS3nn2rIRAW/edit?usp=sharing&ouid=105944806570308548495&rtpof=true&sd=true The video will take you 20 minutes to watch, and the survey will take you another 20 minutes to complete. Thank you for your time and feedback on our research. Please contact me via meng.kang@ubc.ca if you have further interest in this study.

rlnrajesh (Sat, 04 Sep 2021 03:27:10 GMT):
Has joined the channel.

frank-student (Sat, 04 Sep 2021 03:41:46 GMT):
Has joined the channel.

kutya (Sat, 04 Sep 2021 09:47:06 GMT):
Has joined the channel.

MiryangJung (Sat, 04 Sep 2021 14:41:21 GMT):
Has joined the channel.

pseudoyu (Sat, 04 Sep 2021 15:10:22 GMT):
Has joined the channel.

bh4rtp (Sat, 04 Sep 2021 15:46:49 GMT):
Hi, is there a full Alice example for local network tested using Indy-cli commands? I need understand the concepts of Indy. Thank you in advance!

rohitcha (Sun, 05 Sep 2021 09:05:42 GMT):
Has joined the channel.

RahulRajBarnwal (Sun, 05 Sep 2021 16:21:29 GMT):
Has joined the channel.

bhathiya (Sun, 05 Sep 2021 17:58:03 GMT):
Has joined the channel.

scnale (Sun, 05 Sep 2021 23:39:04 GMT):
Has joined the channel.

taccatisid (Mon, 06 Sep 2021 03:30:43 GMT):
Has joined the channel.

BovTricera (Mon, 06 Sep 2021 04:11:43 GMT):
Has joined the channel.

XINGYUZHU (Mon, 06 Sep 2021 06:44:17 GMT):
Has left the channel.

ShaharinAhmed (Mon, 06 Sep 2021 09:52:58 GMT):
Has joined the channel.

iffathossain (Mon, 06 Sep 2021 09:53:01 GMT):
Has joined the channel.

omerabbas (Mon, 06 Sep 2021 11:09:05 GMT):
Has joined the channel.

Junior22 (Mon, 06 Sep 2021 11:48:05 GMT):
Has joined the channel.

kangme (Tue, 07 Sep 2021 01:22:32 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=byJiytXSe2Q2w6umq) We are seeking more volunteers to participate in this study

michael1203 (Tue, 07 Sep 2021 01:46:40 GMT):
Has joined the channel.

sasikanuri (Tue, 07 Sep 2021 03:15:11 GMT):
Has joined the channel.

MarcoOliveira (Tue, 07 Sep 2021 10:06:37 GMT):
Has joined the channel.

regiseloi (Tue, 07 Sep 2021 18:25:34 GMT):
Has joined the channel.

BGerb27 (Wed, 08 Sep 2021 11:29:49 GMT):
Has joined the channel.

bfrech (Wed, 08 Sep 2021 12:00:08 GMT):
Has joined the channel.

fangxin (Wed, 08 Sep 2021 13:12:44 GMT):
Has joined the channel.

danilp (Wed, 08 Sep 2021 15:04:09 GMT):
Has joined the channel.

WadeBarnes (Wed, 08 Sep 2021 16:52:41 GMT):
`von-network` now has a set of tagged releases so you can reference stable versions; https://github.com/bcgov/von-network/releases

mccown (Wed, 08 Sep 2021 17:52:29 GMT):
The Identity Implementer's WG is meeting tomorrow (Sep 9th @ 9am MT / 3pm UTC). In this call, we will review updates from several industry WGs, which you may have missed. This week, Jim StClair (Lumedic.io) will present an overview of how SSI technologies are progressing through ISO/TC 307 to become a part of ISO standards.  Wiki: https://wiki.hyperledger.org/display/IWG/2021-09-9+%3A+Identity+Implementers+WG+Call Zoom: https://zoom.us/my/hyperledger.community.3?pwd=UE90WHhEaHRqOGEyMkV3cldKa2d2dz09

jan-law (Wed, 08 Sep 2021 21:36:37 GMT):
Has joined the channel.

Jason.Leach (Wed, 08 Sep 2021 22:06:04 GMT):
Has joined the channel.

MarcoOliveira (Wed, 08 Sep 2021 22:56:33 GMT):
Has left the channel.

kangme (Wed, 08 Sep 2021 23:24:02 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=byJiytXSe2Q2w6umq) We are seeking more volunteers to participate in this study

Kharvok (Thu, 09 Sep 2021 00:43:08 GMT):
Has joined the channel.

subashnair (Thu, 09 Sep 2021 05:21:41 GMT):
Has joined the channel.

gseront (Thu, 09 Sep 2021 08:19:42 GMT):
Has joined the channel.

mianhammad (Thu, 09 Sep 2021 08:47:43 GMT):
Has joined the channel.

mianhammad (Thu, 09 Sep 2021 09:08:51 GMT):
Hi Everyone! I want to benchmark our ethereum private network and would like to use the caliper for this. I have the configuration file for it but it's giving me error "You must provide the json interface of the contract when instantiating a contract object" I have pasted my network configuration file code and command. Can you please help me out in solving this. `{ "caliper": { "blockchain": "ethereum", "command" : {} }, "ethereum": { "url": "ws://localhost:7545", "contractDeployerAddress": "0x5B14D7452573edB2ABe4941286618Fc09F1Ae085", "contractDeployerAddressPassword": "a0969bdb33adfb905d0d94cf4d6b1f63ebee447a17b80972f16e178bd0b219df", "fromAddress": "0x5B14D7452573edB2ABe4941286618Fc09F1Ae085", "fromAddressPassword": "a0969bdb33adfb905d0d94cf4d6b1f63ebee447a17b80972f16e178bd0b219df", "transactionConfirmationBlocks": 2, "contracts": { "simple": { "path": "../../../src/ethereum/simple/simple.json", "estimateGas": true, "gas": { "query": 100000, "transfer": 70000 } } } } } ` Command caliper launch manager --caliper-workspace ./ --caliper-networkconfig ./networks/ethereum/1node-clique/networkconfig.json --caliper-benchconfig benchmarks/scenario/simple/config.yaml --caliper-flow-only-test --caliper-fabric-gateway-enabled

akankshadixit (Thu, 09 Sep 2021 10:31:26 GMT):
Has joined the channel.

tkuhrt (Thu, 09 Sep 2021 14:48:36 GMT):
Have you asked in the #caliper channel?

DenverNaicker (Thu, 09 Sep 2021 14:52:25 GMT):
Has joined the channel.

justiny (Fri, 10 Sep 2021 01:27:50 GMT):
Has joined the channel.

Fang.Fang (Fri, 10 Sep 2021 02:13:31 GMT):
Has joined the channel.

LakshmanPalli (Fri, 10 Sep 2021 15:40:44 GMT):
Has joined the channel.

ricc.zappoli (Fri, 10 Sep 2021 16:16:02 GMT):
Has joined the channel.

Charly88554 (Fri, 10 Sep 2021 20:41:02 GMT):
Has joined the channel.

gameprofits (Fri, 10 Sep 2021 22:17:35 GMT):
Has joined the channel.

BlockchainTrainer (Sat, 11 Sep 2021 02:58:26 GMT):
Has joined the channel.

Pankil_Patel (Sat, 11 Sep 2021 06:27:52 GMT):
Has joined the channel.

fernandosilvap96 (Sun, 12 Sep 2021 00:14:48 GMT):
Has joined the channel.

HarithKza (Sun, 12 Sep 2021 03:47:23 GMT):
Has joined the channel.

IbrahimBroachwala (Sun, 12 Sep 2021 11:28:00 GMT):
Has joined the channel.

Tkonno (Mon, 13 Sep 2021 00:58:08 GMT):
Has joined the channel.

takehypotenuses (Mon, 13 Sep 2021 01:12:41 GMT):
Has joined the channel.

rsayon (Mon, 13 Sep 2021 04:05:56 GMT):
Has joined the channel.

mianhammad (Mon, 13 Sep 2021 06:35:41 GMT):
Hi Everyone! How can I setup hyperledger caliper for ethereum? Can anyone guide me?

vlaca (Mon, 13 Sep 2021 08:56:35 GMT):
Has joined the channel.

siladu (Mon, 13 Sep 2021 12:45:23 GMT):
Has joined the channel.

SargisDudaklyan (Mon, 13 Sep 2021 17:44:12 GMT):
Has joined the channel.

mukulverm4 (Mon, 13 Sep 2021 18:50:43 GMT):
Has joined the channel.

j-s (Mon, 13 Sep 2021 21:27:20 GMT):
Hi, it seems like there are many potential avenues to implement wallet recovery in the case of loss or breach, though no method is implemented at the Indy level (source: https://chat.hyperledger.org/channel/indy?msg=9mcnDczAp8WwMpRXo).

j-s (Mon, 13 Sep 2021 21:27:20 GMT):
Hi, it seems like there are many potential avenues to implement wallet recovery in the case of loss or breach, though no method is implemented at the Indy level (source: https://chat.hyperledger.org/channel/indy?msg=9mcnDczAp8WwMpRXo). Does anyone have a suggestion for (open source) wallet recovery protocols/solutions? Even if they only work for a particular wallet implementation?

j-s (Mon, 13 Sep 2021 21:27:20 GMT):
Hi, it seems like there are many potential avenues to implement wallet recovery in the case of loss or breach, though no method is implemented at the Indy level (source: https://chat.hyperledger.org/channel/indy?msg=9mcnDczAp8WwMpRXo). Does anyone have a suggestion for (open source) wallet *recovery protocols*/solutions? Even if they only work for a particular wallet implementation?

swcurran (Mon, 13 Sep 2021 22:12:47 GMT):
@tomislav -- Is the backup and recovery part of the Trinsic wallet open source? I recall that it works across esatus and LISSI implementations, correct? @sebastian

rajashekarkodi (Tue, 14 Sep 2021 01:35:56 GMT):
Has joined the channel.

rajashekarkodi (Tue, 14 Sep 2021 01:35:57 GMT):
Hi I'm new to Indy and installation and testing was done successful on EC2 instances with public ip's but now to productionize the same for client we need to use private subnet to host indy network. So can we use AWS NAT gateway to make this communication ?

ajitspendse (Tue, 14 Sep 2021 04:03:56 GMT):
Has joined the channel.

fabio.henriques (Tue, 14 Sep 2021 08:33:40 GMT):
Has joined the channel.

gfunchal (Tue, 14 Sep 2021 12:28:34 GMT):
Has joined the channel.

blessingkp (Tue, 14 Sep 2021 17:15:17 GMT):
Has joined the channel.

hjpn (Tue, 14 Sep 2021 19:10:07 GMT):
Has joined the channel.

asif123 (Wed, 15 Sep 2021 02:24:25 GMT):
Has joined the channel.

claracavalcante (Wed, 15 Sep 2021 07:46:45 GMT):
Has joined the channel.

fkossiaras (Wed, 15 Sep 2021 09:27:00 GMT):
Has joined the channel.

hjpn (Wed, 15 Sep 2021 09:41:47 GMT):
Has left the channel.

moisesjaramillo (Wed, 15 Sep 2021 16:28:31 GMT):
Has joined the channel.

bobbh (Wed, 15 Sep 2021 16:57:52 GMT):
Has joined the channel.

CherryHsu (Thu, 16 Sep 2021 02:55:52 GMT):
Has joined the channel.

mnill2 (Thu, 16 Sep 2021 06:21:38 GMT):
Has joined the channel.

kingstenzzz (Thu, 16 Sep 2021 07:04:06 GMT):
Has joined the channel.

MiryangJung (Thu, 16 Sep 2021 10:00:53 GMT):
Occasionally, writing data to the indy ledger does not cause errors, but data is not registered in the ledger. The ledger is using a docker. Do you know anything about this Situation?

WadeBarnes (Thu, 16 Sep 2021 12:25:19 GMT):
@MiryangJung, Are you able to provides details and the steps you are following that lead up to the issue?

WadeBarnes (Thu, 16 Sep 2021 12:25:19 GMT):
@MiryangJung, Are you able to provide details and the steps you are following that lead up to the issue?

smbkd6 (Thu, 16 Sep 2021 15:00:06 GMT):
Has joined the channel.

fabio.henriques (Thu, 16 Sep 2021 15:42:44 GMT):
Has left the channel.

MiryangJung (Thu, 16 Sep 2021 15:46:34 GMT):
There was a problem only with Node1. Perhaps it is an internal network problem.

aaronrbg (Thu, 16 Sep 2021 18:28:04 GMT):
Has joined the channel.

vberg 2 (Fri, 17 Sep 2021 08:48:16 GMT):
Has joined the channel.

sharukh (Fri, 17 Sep 2021 11:21:58 GMT):
Has joined the channel.

OscarTBallester (Fri, 17 Sep 2021 13:07:44 GMT):
Has joined the channel.

MullaAlmas (Fri, 17 Sep 2021 14:41:12 GMT):
Has joined the channel.

chikai (Fri, 17 Sep 2021 14:57:07 GMT):
Has joined the channel.

javieronishi (Fri, 17 Sep 2021 16:08:33 GMT):
Has joined the channel.

mollayii (Fri, 17 Sep 2021 16:32:04 GMT):
Has joined the channel.

sanskarkhare (Fri, 17 Sep 2021 18:08:04 GMT):
Has joined the channel.

Clippo 2 (Sat, 18 Sep 2021 04:26:19 GMT):
Has joined the channel.

Nufail (Sat, 18 Sep 2021 06:00:24 GMT):
Has joined the channel.

handao (Sat, 18 Sep 2021 06:30:13 GMT):
Has joined the channel.

iros75 (Sat, 18 Sep 2021 20:37:23 GMT):
Has joined the channel.

bardia (Sun, 19 Sep 2021 05:39:03 GMT):
Has joined the channel.

littlemouse (Sun, 19 Sep 2021 05:55:21 GMT):
Has joined the channel.

MrChainBlockz (Sun, 19 Sep 2021 06:52:42 GMT):
Has joined the channel.

arrey_damn (Sun, 19 Sep 2021 09:32:25 GMT):
Has joined the channel.

AmshumaanVarma (Sun, 19 Sep 2021 13:14:41 GMT):
Has joined the channel.

jcarlosusa (Mon, 20 Sep 2021 02:47:47 GMT):
Has joined the channel.

cattanep (Mon, 20 Sep 2021 12:44:03 GMT):
Has joined the channel.

Noc0r (Mon, 20 Sep 2021 12:51:47 GMT):
Has joined the channel.

chesta0703 (Mon, 20 Sep 2021 18:09:10 GMT):
Has joined the channel.

gravin10 (Mon, 20 Sep 2021 19:36:12 GMT):
Has joined the channel.

Zarquan314 (Tue, 21 Sep 2021 05:23:02 GMT):
Has joined the channel.

F4NES (Tue, 21 Sep 2021 08:16:37 GMT):
Has joined the channel.

TejaBanton (Tue, 21 Sep 2021 13:50:19 GMT):
Has joined the channel.

hiromaily (Tue, 21 Sep 2021 15:10:47 GMT):
Has joined the channel.

JonathanScialpi (Tue, 21 Sep 2021 17:42:49 GMT):
Has joined the channel.

JonathanScialpi (Tue, 21 Sep 2021 17:47:12 GMT):
Hi, I am getting a docker error when trying to run BAF via minikube: Starting build process... Adding env variables... Validatin network yaml error: Cannot find schema '/home/blockchain-automation-framework/platforms/network-schema.json' Anyone know how to fix or suggestions?

JonathanScialpi (Tue, 21 Sep 2021 17:47:27 GMT):
Followed this guide: https://blockchain-automation-framework.readthedocs.io/en/latest/developer/baf_minikube_setup.html

JonathanScialpi (Tue, 21 Sep 2021 17:47:48 GMT):
I used hyperledger-indy/configuration/samples/network-minikube.yaml

scaraclette (Tue, 21 Sep 2021 21:43:44 GMT):
Has joined the channel.

a-i-to (Wed, 22 Sep 2021 06:19:38 GMT):
Has joined the channel.

IgorSim (Wed, 22 Sep 2021 08:22:42 GMT):
Has joined the channel.

kongwu 2 (Wed, 22 Sep 2021 08:50:33 GMT):
Has joined the channel.

bardia (Wed, 22 Sep 2021 09:08:06 GMT):
How do I fix the following error? :tears: This happens when I fail to login 20 times failed to enroll user: enroll failed: enroll failed: Response from server: Error Code: 73 - Incorrect password entered 10 times, max incorrect password limit of 10 reached

kosarlatif (Wed, 22 Sep 2021 10:26:27 GMT):
Has joined the channel.

oding (Wed, 22 Sep 2021 14:21:00 GMT):
Has joined the channel.

CollinMeese (Wed, 22 Sep 2021 16:16:02 GMT):
Has joined the channel.

Andre_Augusto (Wed, 22 Sep 2021 16:27:22 GMT):
Has joined the channel.

PowellAndy (Wed, 22 Sep 2021 17:31:30 GMT):
Has joined the channel.

SamKrew (Wed, 22 Sep 2021 17:52:57 GMT):
Has joined the channel.

mccown (Wed, 22 Sep 2021 20:06:27 GMT):
The Identity Implementer's WG is meeting tomorrow (Sep 23rd @ 9am MT / 3pm UTC). In this call, we will review updates from several SSI industry WGs. This week, Xavier Vila (SICPA) will present an overview of SICPA and their efforts to create and launch new identity solutions. Wiki: https://wiki.hyperledger.org/display/IWG/2021-09-23+%3A+Identity+Implementers+WG+Call Zoom: https://zoom.us/my/hyperledger.community.3?pwd=UE90WHhEaHRqOGEyMkV3cldKa2d2dz09

ssmxmss (Thu, 23 Sep 2021 07:52:49 GMT):
Has joined the channel.

Rafailk (Thu, 23 Sep 2021 08:00:03 GMT):
Has joined the channel.

vonpmg (Thu, 23 Sep 2021 08:31:17 GMT):
Has joined the channel.

GiovanniZanol (Thu, 23 Sep 2021 12:41:37 GMT):
Has joined the channel.

Prerana72 (Thu, 23 Sep 2021 13:57:27 GMT):
Hi team, We are using stable libindy.so for android and see some issue with storing credentials on wallet. For debugging, We are trying to build the libindy for android following the steps mentioned here (https://github.com/hyperledger/indy-sdk/blob/master/docs/build-guides/android-build.md), with additional logs. We used v1.15 of source code from github for this. We were able to build it, but are seeing a different error while in the call to indy_add_wallet_record. This error was not seen when we used the stable libindy from here(https://repo.sovrin.org/android/libindy/stable/1.15.0/) So, the question is, how are the stable build libraries built? Are there different scripts to generate them? We followed the steps mentioned in the github link above.

danielpenagos (Thu, 23 Sep 2021 15:38:31 GMT):
Has joined the channel.

raushan.kumar (Thu, 23 Sep 2021 17:10:54 GMT):
Has joined the channel.

lcc 3 (Fri, 24 Sep 2021 06:51:23 GMT):
Has joined the channel.

Bertho (Fri, 24 Sep 2021 10:23:54 GMT):
Has joined the channel.

lainio (Fri, 24 Sep 2021 11:47:07 GMT):
Has joined the channel.

alvarodelaiglesia (Fri, 24 Sep 2021 13:40:38 GMT):
Has joined the channel.

t3x (Fri, 24 Sep 2021 13:48:28 GMT):
Has joined the channel.

ArnoldsBogdanovs (Fri, 24 Sep 2021 14:06:53 GMT):
Has joined the channel.

spandolfi (Fri, 24 Sep 2021 14:32:00 GMT):
Has joined the channel.

JNeedham (Fri, 24 Sep 2021 15:05:17 GMT):
Has joined the channel.

shadownetz (Sat, 25 Sep 2021 01:32:11 GMT):
Has joined the channel.

mohammed.alzuharey (Sat, 25 Sep 2021 11:11:57 GMT):
Has joined the channel.

chesta (Sat, 25 Sep 2021 13:34:38 GMT):
Has joined the channel.

PeterSkogh (Sat, 25 Sep 2021 20:33:01 GMT):
Has joined the channel.

GergelyLengyel (Sat, 25 Sep 2021 21:49:33 GMT):
Has joined the channel.

GergelyLengyel (Sat, 25 Sep 2021 21:54:40 GMT):
Has left the channel.

sasschicken69 (Sun, 26 Sep 2021 11:36:44 GMT):
Has joined the channel.

brandongomezsanchez (Mon, 27 Sep 2021 00:45:04 GMT):
Has joined the channel.

Adisa123 (Mon, 27 Sep 2021 04:38:33 GMT):
Has joined the channel.

karayel (Mon, 27 Sep 2021 07:35:19 GMT):
Has joined the channel.

adn-misa (Mon, 27 Sep 2021 09:58:22 GMT):
Has joined the channel.

sapthasurendran (Mon, 27 Sep 2021 12:51:28 GMT):
Has joined the channel.

SwapnaliDive (Mon, 27 Sep 2021 13:48:06 GMT):
Hello Team, I created two pools using indy cli namely pool1 and pool2. Why I can not use the created schemas on pool1 to create credential defination on pool2. In this case, I am using the same Indy Node Setup. I am not able to undestand conceptually Indy Network , Pool setup with Indy nodes, pool setup with indy-cli. What is the exact difference. Can someone please help to enlighten the concepts. Thank you.

weiiv (Mon, 27 Sep 2021 15:00:56 GMT):
Has joined the channel.

srihari.r (Mon, 27 Sep 2021 15:38:03 GMT):
Has joined the channel.

Lucampelli (Mon, 27 Sep 2021 23:36:43 GMT):
Hi I'm having trouble with the "How to start local nodes pool with docker" Ledger... somehow a build_get_nym_request is failing with "indy.error.CommonInvalidStructure". The thing is... it was working just fine until I sent a build_nym_request and sign_and_submit_request.... then it simply broke... Is there a way to reset the local pool docker ledger? what have I broke?

Lucampelli (Tue, 28 Sep 2021 00:29:49 GMT):
As misteriously as it appeared, the error is gone... beats me...

HokitTang (Tue, 28 Sep 2021 02:18:02 GMT):
Has joined the channel.

Oznfc (Tue, 28 Sep 2021 08:33:26 GMT):
Has joined the channel.

XM.Jin (Tue, 28 Sep 2021 08:47:37 GMT):
Has joined the channel.

SamRutten (Tue, 28 Sep 2021 10:09:17 GMT):
Has joined the channel.

ranjan_paul (Tue, 28 Sep 2021 13:16:15 GMT):
Has joined the channel.

Oznfc (Tue, 28 Sep 2021 13:38:10 GMT):
Hi , ı have a question. I'm working on "vc-authn-oidc/demo" . This example about email . How can I change it . For example "name surname address" .

kylegrobinson (Tue, 28 Sep 2021 17:28:47 GMT):
Has joined the channel.

vsadriano (Wed, 29 Sep 2021 10:30:27 GMT):
Has left the channel.

weiiv (Wed, 29 Sep 2021 14:20:05 GMT):
Has left the channel.

lynn.bendixsen (Wed, 29 Sep 2021 19:18:13 GMT):
Hello, In the Indy CLI the abbreviated "pool" is actually a "pool handle" that points to a network or pool of nodes. So IF both pool1 and pool2 from your example used the same genesis file during setup, then they both refer to the same network and schemas created while you were connected to pool1 CAN be used to create credential-definitions on pool2, because in this case they are referencing the same network. From the Indy Node perspective, the network contains a "Pool Ledger' which is a record of all of the nodes added to that network. So from this perspective a "pool" is the group of nodes in the same network. Does that help?

matheus27 (Wed, 29 Sep 2021 20:14:39 GMT):
Has joined the channel.

Parthavs (Thu, 30 Sep 2021 01:01:59 GMT):
Has joined the channel.

SwapnaliDive (Thu, 30 Sep 2021 11:58:10 GMT):
Thank you @lynn.bendixsen It gave me more clarity

barnold (Thu, 30 Sep 2021 13:43:27 GMT):
Has joined the channel.

NkNs (Thu, 30 Sep 2021 18:23:32 GMT):
Has joined the channel.

sauveergoel (Thu, 30 Sep 2021 18:46:32 GMT):
What kind of security audits/infrastructure should be followed with indy ledger?

AdrianAvila (Thu, 30 Sep 2021 21:32:58 GMT):
Has joined the channel.

JagannathAlur (Fri, 01 Oct 2021 01:11:21 GMT):
Has joined the channel.

bardia (Fri, 01 Oct 2021 10:11:12 GMT):
Has left the channel.

AnuragKumar_10 (Fri, 01 Oct 2021 11:16:53 GMT):
Has joined the channel.

anilhelvaci (Fri, 01 Oct 2021 11:23:28 GMT):
Has joined the channel.

dmckay (Fri, 01 Oct 2021 14:11:13 GMT):
Has joined the channel.

geekcharming (Fri, 01 Oct 2021 16:45:31 GMT):
Has joined the channel.

azhang919 (Fri, 01 Oct 2021 17:26:38 GMT):
Has joined the channel.

mahnoormmalik (Fri, 01 Oct 2021 17:32:27 GMT):
Has joined the channel.

CrystalChun (Fri, 01 Oct 2021 17:35:32 GMT):
Has joined the channel.

imchaves10 (Fri, 01 Oct 2021 19:10:34 GMT):
Has joined the channel.

Leeyoungone (Fri, 01 Oct 2021 23:09:17 GMT):
Has joined the channel.

elenipartakki (Fri, 01 Oct 2021 23:24:02 GMT):
Has joined the channel.

guri (Sat, 02 Oct 2021 04:10:58 GMT):
Has joined the channel.

Z3R0097 (Sat, 02 Oct 2021 06:22:58 GMT):
Has joined the channel.

RakshitaKaulgud (Sat, 02 Oct 2021 18:42:24 GMT):
Has joined the channel.

fallenvalkyrie17 (Sat, 02 Oct 2021 19:14:15 GMT):
Has joined the channel.

OmarKhodr (Sun, 03 Oct 2021 06:52:58 GMT):
Has joined the channel.

quyenld176859 (Sun, 03 Oct 2021 08:39:28 GMT):
Has joined the channel.

DanielHaren1 (Sun, 03 Oct 2021 11:42:34 GMT):
Has joined the channel.

Pankaj_1 (Mon, 04 Oct 2021 06:31:48 GMT):
Has joined the channel.

glehmann (Mon, 04 Oct 2021 07:05:34 GMT):
Has joined the channel.

Priyadaharan107 (Mon, 04 Oct 2021 17:20:07 GMT):
Has joined the channel.

analia_meira (Mon, 04 Oct 2021 18:17:47 GMT):
Has joined the channel.

ringo156 (Tue, 05 Oct 2021 04:18:42 GMT):
Has joined the channel.

sbyits (Tue, 05 Oct 2021 05:04:25 GMT):
Has joined the channel.

shanest (Tue, 05 Oct 2021 05:48:59 GMT):
i'm running VON by checkout into von_network repo and ./manage start; all 4 nodes are up and running on my localhost:9000 but load seemed to take few minutes on every page / mouse click. Is this normal?

HakanTUB (Tue, 05 Oct 2021 16:21:44 GMT):
Hello, is there a way for me to register a DID from another DID method as a DID controller to an Indy existing Indy DID?

shivamraj6288 (Tue, 05 Oct 2021 19:51:47 GMT):
Has joined the channel.

conanoc (Wed, 06 Oct 2021 02:26:43 GMT):
Can you elaborate?

guidam (Wed, 06 Oct 2021 08:22:46 GMT):
Has joined the channel.

hugebing (Wed, 06 Oct 2021 09:05:33 GMT):
Has joined the channel.

HakanTUB (Wed, 06 Oct 2021 11:31:48 GMT):
Of course. I have a DID in ebsi ledger and want to associate that DID with another DID in indy ledger

HakanTUB (Wed, 06 Oct 2021 11:32:27 GMT):
this is to make a proof that the DID issuing credentials in indy network has the right to do so. And that right is basically coming from did:ebsi

HakanTUB (Wed, 06 Oct 2021 11:32:49 GMT):
I am trying to solve the root of trust by creating an association with these two dids

x6ein (Wed, 06 Oct 2021 20:52:18 GMT):
Has joined the channel.

conanoc (Thu, 07 Oct 2021 02:48:20 GMT):
You could write an indy DID in the DID doc. of ebsi if ebsi allows writing such metadata in the DID doc.

conanoc (Thu, 07 Oct 2021 03:05:38 GMT):
You also can add attributes in Indy DID https://hyperledger-indy.readthedocs.io/projects/node/en/latest/requests.html#attrib In this case, you should add a proof as an attribute, for example, a signature of the owner of did:ebsi signing the indy DID. Then, you could prove that the owner of the indy DID is the owner of ebsi DID.

CristianLorenzetto (Thu, 07 Oct 2021 06:05:40 GMT):
Has joined the channel.

damianlluch (Thu, 07 Oct 2021 08:56:44 GMT):
Has joined the channel.

Aneesha (Thu, 07 Oct 2021 08:59:12 GMT):
Has joined the channel.

mccown (Thu, 07 Oct 2021 13:39:06 GMT):
The Identity Implementer's WG for today (Oct 7th) has been cancelled. Enjoy IIW next week and I look forward to seeing you in our next meeting on Oct 21st!

jyfregramer (Thu, 07 Oct 2021 17:45:05 GMT):
Has joined the channel.

nataliasofia (Thu, 07 Oct 2021 20:29:22 GMT):
Has joined the channel.

yianco (Thu, 07 Oct 2021 22:28:22 GMT):
Has joined the channel.

ThinCrusts (Fri, 08 Oct 2021 01:37:34 GMT):
Has joined the channel.

stephaneJ (Fri, 08 Oct 2021 07:50:44 GMT):
Has joined the channel.

Camilo12987 (Fri, 08 Oct 2021 15:21:30 GMT):
Has joined the channel.

jackmacloud (Sat, 09 Oct 2021 06:35:59 GMT):
Has joined the channel.

Lucampelli (Sun, 10 Oct 2021 20:11:27 GMT):
Hello! I'm having a little bit of trouble with verifying a proof.

Lucampelli (Sun, 10 Oct 2021 20:11:27 GMT):
Hello! I'm having a little bit of trouble with verifying a proof. { "name": "Proof Request for CED_1.0", "requested_attributes": { "attr1_referent": { "name": "University Name", "restrictions": [ { "issuer_did": "6oNMYEhuupttTUa3TwxvYP", "schema_id": "6oNMYEhuupttTUa3TwxvYP:2:CED:1.0", "attr::universityname::value": "UFSC" } ] }, "attr2_referent": { "name": "University DID", "restrictions": [ { "issuer_did": "6oNMYEhuupttTUa3TwxvYP", "schema_id": "6oNMYEhuupttTUa3TwxvYP:2:CED:1.0", "attr::universitydid::value": "3113323/0001-99" } ] }, "attr3_referent": { "name": "University CNPJ", "restrictions": [ { "issuer_did": "6oNMYEhuupttTUa3TwxvYP", "attr::universitycnpj::value": "Kxdw7YRR2EVGFG22bv1iAw", "schema_id": "6oNMYEhuupttTUa3TwxvYP:2:CED:1.0" } ] } }, "requested_predicates": { "predicate1_referent": { "name": "Expiry", "p_value": 1633662000, "restrictions": [ { "issuer_did": "6oNMYEhuupttTUa3TwxvYP", "schema_id": "6oNMYEhuupttTUa3TwxvYP:2:CED:1.0" } ], "p_type": ">" } }, "nonce": "499654565955", "version": "1.0" } I create a proof request, and execute "anoncreds.prover_get_credentials_for_proof_req" without trouble, receiving all valid credentials for it.

Lucampelli (Sun, 10 Oct 2021 20:11:27 GMT):
Hello! I'm having a little bit of trouble with verifying a proof. I create a proof request, and execute "anoncreds.prover_get_credentials_for_proof_req" without trouble, receiving all valid credentials for it.

Lucampelli (Sun, 10 Oct 2021 20:11:27 GMT):
Hello! I'm having a little bit of trouble with verifying a proof. I create a proof request, and execute "anoncreds.prover_get_credentials_for_proof_req" without trouble, receiving all valid credentials for it. I then create the the proof with "anoncreds.prover_create_proof" without trouble. The problem comes when I try to verify the proof. Since I use validation for attributes as well, I wonder why it is not working. I will place all payloads in as comments on this thread. I'd appreciate if any one could help me...

Lucampelli (Sun, 10 Oct 2021 20:11:27 GMT):
Hello! I'm having a little bit of trouble with verifying a proof. I create a proof request, and execute "anoncreds.prover_get_credentials_for_proof_req" without trouble, receiving all valid credentials for it. I then create the the proof with "anoncreds.prover_create_proof" without trouble. The problem comes when I try to verify the proof. Since I use validation for attributes as well, I wonder why it is not working. I will place all payloads in as comments on this thread. I'd appreciate if any one could help me... Also there is no problem with showing anything of worth in the attributes since these are test values.

Lucampelli (Sun, 10 Oct 2021 20:29:02 GMT):
Proof Request: { "name": "Proof Request for CED_1.0", "requested_attributes": { "attr1_referent": { "name": "University Name", "restrictions": [ { "issuer_did": "6oNMYEhuupttTUa3TwxvYP", "schema_id": "6oNMYEhuupttTUa3TwxvYP:2:CED:1.0", "attr::universityname::value": "UFSC" } ] }, "attr2_referent": { "name": "University DID", "restrictions": [ { "issuer_did": "6oNMYEhuupttTUa3TwxvYP", "schema_id": "6oNMYEhuupttTUa3TwxvYP:2:CED:1.0", "attr::universitydid::value": "3113323/0001-99" } ] }, "attr3_referent": { "name": "University CNPJ", "restrictions": [ { "issuer_did": "6oNMYEhuupttTUa3TwxvYP", "attr::universitycnpj::value": "Kxdw7YRR2EVGFG22bv1iAw", "schema_id": "6oNMYEhuupttTUa3TwxvYP:2:CED:1.0" } ] } }, "requested_predicates": { "predicate1_referent": { "name": "Expiry", "p_value": 1633662000, "restrictions": [ { "issuer_did": "6oNMYEhuupttTUa3TwxvYP", "schema_id": "6oNMYEhuupttTUa3TwxvYP:2:CED:1.0" } ], "p_type": ">" } }, "nonce": "499654565955", "version": "1.0" }

Lucampelli (Sun, 10 Oct 2021 20:30:08 GMT):

Lucampelli - Sun Oct 10 2021 17:29:59 GMT-0300 (Brasilia Standard Time).txt

Lucampelli (Sun, 10 Oct 2021 20:31:31 GMT):

Proof_Request.txt

Lucampelli (Sun, 10 Oct 2021 20:31:49 GMT):

Lucampelli - Sun Oct 10 2021 17:31:42 GMT-0300 (Brasilia Standard Time).txt

Lucampelli (Sun, 10 Oct 2021 20:32:22 GMT):

Proof.txt

Lucampelli (Sun, 10 Oct 2021 20:34:33 GMT):

Proof_Request.txt

Lucampelli (Sun, 10 Oct 2021 20:35:46 GMT):

Proof.txt

Lucampelli (Sun, 10 Oct 2021 20:36:44 GMT):
DEBUG:indy.libindy:_get_error_details: <<< error_details: {'backtrace': '', 'message': 'Error: Proof rejected\n Caused by: Requested restriction validation failed for "{"University Name": Some("UFSC")}" attributes\n Caused by: $or operator validation failed. All conditions were failed.\n'}

Lucampelli (Sun, 10 Oct 2021 20:37:06 GMT):
Credential: {"referent":"1fc76701-ab19-4aa6-86d5-319229c304cc","attrs":{"Expiry":"1641006000","University DID":"Kxdw7YRR2EVGFG22bv1iAw","University Name":"UFSC","University CNPJ":"3113323/0001-99"},"schema_id":"6oNMYEhuupttTUa3TwxvYP:2:CED:1.0","cred_def_id":"6oNMYEhuupttTUa3TwxvYP:3:CL:13:TAG634","rev_reg_id":null,"cred_rev_id":null}

baxihemant (Mon, 11 Oct 2021 01:45:02 GMT):
Has joined the channel.

Bhawikchawda (Mon, 11 Oct 2021 06:49:17 GMT):
Has joined the channel.

OmarSaad (Mon, 11 Oct 2021 07:07:52 GMT):
Has joined the channel.

HakanTUB (Mon, 11 Oct 2021 11:34:56 GMT):
I will take a look at this, thanks a lot!

Sa_Block98 (Mon, 11 Oct 2021 14:01:59 GMT):
Has joined the channel.

chris2222 (Mon, 11 Oct 2021 14:07:16 GMT):
Has joined the channel.

TimoGlastra (Mon, 11 Oct 2021 14:31:00 GMT):
Question with regards to endorsing transactions. When trying to submit a transaction as a DID that is not ENDORSER I get the following error (which makes sense). However one of the OR requirements doesn't make fully sense to me. It mentions *OR 1 signature of any role is required and needs to be owner with additional metadata fees attrib_add* Could someone explain to me what this means and how this requirement could be achieved? :) ``` Ledger rejected transaction request: client request invalid: UnauthorizedClientRequest('Rule for this action is: 1 TRUSTEE signature is required and needs to be owner OR 1 STEWARD signature is required and needs to be owner OR 1 ENDORSER signature is required and needs to be owner OR 1 NETWORK_MONITOR signature is required and needs to be owner OR 1 signature of any role is required and needs to be owner with additional metadata fees attrib_add OR 1 ENDORSER signature is required AND 1 signature of any role is required and needs to be owner\nFailed checks:\nConstraint: 1 TRUSTEE signature is required and needs to be owner, Error: Not enough TRUSTEE signatures\nConstraint: 1 STEWARD signature is required and needs to be owner, Error: Not enough STEWARD signatures\nConstraint: 1 ENDORSER signature is required and needs to be owner, Error: Not enough ENDORSER signatures\nConstraint: 1 NETWORK_MONITOR signature is required and needs to be owner, Error: Not enough NETWORK_MONITOR signatures\nConstraint: 1 signature of any role is required and needs to be owner with additional metadata fees attrib_add, Error: Fees are required for this txn type\nConstraint: 1 ENDORSER signature is required AND 1 signature of any role is required and needs to be owner, Error: Not enough ENDORSER signatures',) ```

Lucampelli (Mon, 11 Oct 2021 15:28:59 GMT):
Hello! An interesting question about the workings of revocation registries popped up. For example there are 3 entities: A, B and C. A is organizationally superior to B, and B > C. This means that A manages B, and B manages C. B then issues a credential T to C, and sets up the revocation registry R for it. In a scenario where A sees the credential as being irregularly issued, how can A access the revocation registry set up by B, in order to revoke that credential? Can B, when publishing the revocation registry to the ledger configure authorization for A to publish on it as well? How can that be done?

moorada (Mon, 11 Oct 2021 15:38:16 GMT):
Has joined the channel.

MANARKASSOU (Mon, 11 Oct 2021 20:17:03 GMT):
Has joined the channel.

JoseeeM (Mon, 11 Oct 2021 22:04:27 GMT):
Has joined the channel.

action-sj (Tue, 12 Oct 2021 01:28:10 GMT):
Has joined the channel.

zero0097 (Tue, 12 Oct 2021 04:50:24 GMT):
Has joined the channel.

prasadmnd (Tue, 12 Oct 2021 09:49:02 GMT):
Has joined the channel.

prasadmnd (Tue, 12 Oct 2021 09:49:49 GMT):
I have a question, with blockchain Indy can you store information besides the did, or my other assumption is that it needs to be used with other blockchain to store additional information, Thanks

WeCanBeFriends (Tue, 12 Oct 2021 14:38:02 GMT):
Has joined the channel.

daidoji (Tue, 12 Oct 2021 15:29:37 GMT):
on Indy nothing gets stored on chain besides the did, schemas, cred defs, and revocation pointers. Its designed that way so that privacy and anonymity are the default. If you want to store data on chain then you have to use some other technology explicitly.

swcurran (Tue, 12 Oct 2021 15:29:43 GMT):
The idea with Indy is that you don't put data on the blockchain -- you only put public information, notably a DID, and schema information. Nothing about the verifiable credential goes on the ledger.

frostyfrog (Tue, 12 Oct 2021 20:23:39 GMT):
Has joined the channel.

axays007 (Wed, 13 Oct 2021 05:22:53 GMT):
Has joined the channel.

FrancescoUrbano (Wed, 13 Oct 2021 08:01:18 GMT):
Has joined the channel.

Ilikecodes (Wed, 13 Oct 2021 17:04:56 GMT):
Has joined the channel.

priamv (Wed, 13 Oct 2021 18:11:35 GMT):
Has joined the channel.

Dazpel (Wed, 13 Oct 2021 18:19:00 GMT):
Has joined the channel.

rolandtyler (Wed, 13 Oct 2021 20:43:37 GMT):
Has joined the channel.

bs (Wed, 13 Oct 2021 21:07:42 GMT):
Has joined the channel.

alekseishcherbin (Thu, 14 Oct 2021 07:22:56 GMT):
Has joined the channel.

iorveth (Thu, 14 Oct 2021 08:51:06 GMT):
Has joined the channel.

FrancescoMemola (Thu, 14 Oct 2021 09:50:57 GMT):
Has joined the channel.

thejas.pavithran (Thu, 14 Oct 2021 14:35:31 GMT):
Has joined the channel.

EC983 (Thu, 14 Oct 2021 19:02:10 GMT):
Has joined the channel.

lynn.bendixsen (Thu, 14 Oct 2021 20:15:48 GMT):
Hi Timo, Short Answer - You must own the DID you are adding or editing the attrib for, and add the ledger fees associated with the fee named "attrib_add" for that specific requirement. More details: The error is essentially listing all of the possible ways to authorize the requested transaction and is based on the Auth_Rules for the ledger you are using. For example, the Sovrin BuilderNet currently has Auth_Rules that would produce that kind of output. You can see all of the auth_rules using: `indy-cli> ledger get-auth-rule` The part of the message you asked about refers to any owner of the DID that wants to change or add an attrib has to pay the ledger fee associated with the request (the fee named attrib_add). I looked on the Buildernet to see what the current fee is and it appears that the fee table for the "sov" payment method has been removed, so that effectively invalidates that method of satisfying the auth_rule. The only method currently valid on the builldernet for an "Author" (no role) to get their attrib added or changed is to have an endorser sign that transaction. The same is not true for the Sovrin StagingNet or the Sovrin MainNet, so as you progress towards "production" know that the rules will change (for Sovrin's networks). On the Sovrin Staging and Main networks (And on all Indicio networks), the owner of a DID can change their attribs as much as they would like to. @WadeBarnes might want to look into fixing that inconsistency. Even more data for those that are still with me :) (Sorry for the huge post)This has been an issue historically. The network operators want to charge for attrib writes to the ledger, but the original implementation had a bug in the auth_rule restricting an Author from being able to change their attrib regardless of what signature was on the txn. There was a "quick fix" applied to remove the bug but that fix made it "free" for authors to add/edit their attribs (can't track them down to ask for payment). Then the overall issue was never re-addressed or fixed for the desired ability to charge for it. BuilderNet seems to have addressed the bug and the ability to charge for the txn. It looks like the correct fix was applied there, but then never propagated to the other Sovrin ledgers.

TimoGlastra (Thu, 14 Oct 2021 21:32:29 GMT):
Thanks for this really detailed answer Lynn! This is really helpful :) We're currently testing a setup where we let an endorser submit a new DID to the ledger, but then want to use that DID as the public DID on an ACA-Py agent. When we start ACA-Py it will automatically try to update the endpoint on the ledger (which is a desired feature because we want the endpoint on the ledger), but currently rejects with the following error. Would you say this won't happen on either Sovrin staging or main net? Do you know if there are any plans to change this?

swcurran (Thu, 14 Oct 2021 23:07:34 GMT):
FYI -- @ianco is working on automatic/configured endorsing in ACA-Py, so that an endorser can sign any transaction needed by an author that they agree to endorse.

ianco (Thu, 14 Oct 2021 23:13:22 GMT):
FYI Right now we have endorser support built in for Schema, Cred Def and Revocation functions, but not (yet) for updating DID attrs

ianco (Thu, 14 Oct 2021 23:14:11 GMT):
As I understand from @lynn.bendixsen 's answer, if you are the DID owner you can write the transaction (with fees) without any endorser signature

ianco (Thu, 14 Oct 2021 23:14:13 GMT):
OR

ianco (Thu, 14 Oct 2021 23:14:28 GMT):
You can get the transaction endorsed, and then write it to the ledger

ianco (Thu, 14 Oct 2021 23:14:46 GMT):
I'll add this (endorser support for DID attrs) to the TODO list

lynn.bendixsen (Thu, 14 Oct 2021 23:29:08 GMT):
@TimoGlastra The way it currently stands, I am pretty sure you won't get an error on Sovrin's Staging or MainNet. You could test the behavior on Indicio's Testnet, if you like, as it is currently set up with the same over-permissive Auth_Rule for attribs. The way it currently works makes it really hard for Sovrin to collect the listed attrib fees for their networks, though, so if I had to guess, I would say not to count on it being that way for the long term. It's really up to Sovrin on how they manage that though. While I am answering your specific question, the other posters in this thread are pointing at what is probably the real answer to your issue. (Auto-endorsing would likely help to resolve the error you are seeing in your use case.)

hyper-curious (Fri, 15 Oct 2021 02:06:15 GMT):
Has joined the channel.

swcurran (Fri, 15 Oct 2021 02:46:49 GMT):
Not sure what you mean Lynn about the fees. There should be no rules that even mention fees as I would assume that is tied to the token. We'll look at that. As far as tracing the fees, it's pretty easy -- the ATTRIB is tied to a DID that was endorsed by an Endorser (or is an Endorser). That's how must cover the ATTRIB transaction on ledgers that have fees.

spaturu2305 (Fri, 15 Oct 2021 02:47:16 GMT):
Has joined the channel.

natedl98 (Sat, 16 Oct 2021 03:22:50 GMT):
Has joined the channel.

NaufalAlexander (Sat, 16 Oct 2021 04:26:33 GMT):
Has joined the channel.

kondo (Sat, 16 Oct 2021 05:05:52 GMT):
Has joined the channel.

ThomasTeo (Sat, 16 Oct 2021 08:09:32 GMT):
Has joined the channel.

AyhanSunar (Sat, 16 Oct 2021 14:12:31 GMT):
Has joined the channel.

hellboy03 (Sat, 16 Oct 2021 18:18:02 GMT):
Has joined the channel.

kdahya (Sun, 17 Oct 2021 02:50:45 GMT):
Has joined the channel.

sagarmylavarapu (Sun, 17 Oct 2021 10:26:43 GMT):
Has joined the channel.

manvendra (Sun, 17 Oct 2021 19:19:42 GMT):
Has joined the channel.

lfklupp (Mon, 18 Oct 2021 03:15:52 GMT):
Has joined the channel.

John.Greenan (Mon, 18 Oct 2021 04:38:47 GMT):
Has joined the channel.

indong (Mon, 18 Oct 2021 06:08:00 GMT):
Has joined the channel.

starjara (Mon, 18 Oct 2021 08:10:16 GMT):
Has joined the channel.

lev_jac (Mon, 18 Oct 2021 09:50:26 GMT):
Has joined the channel.

hossein.namazian (Mon, 18 Oct 2021 15:30:32 GMT):
Has joined the channel.

lawrenceabeyta (Mon, 18 Oct 2021 21:53:11 GMT):
Has joined the channel.

Vanderscycle (Mon, 18 Oct 2021 22:30:20 GMT):
Has joined the channel.

multijik (Tue, 19 Oct 2021 03:18:18 GMT):
Has joined the channel.

pattycode (Tue, 19 Oct 2021 05:03:11 GMT):
Has joined the channel.

SwapnaliDive (Tue, 19 Oct 2021 06:11:08 GMT):
Hello Team, What are the considerations who should keep in mind / what are the things who should take care while deploying indy network on production level Thanks in advance.

SwapnaliDive (Tue, 19 Oct 2021 06:11:08 GMT):
Hello Team, What are the considerations one should keep in mind / what are the things one should take care while deploying indy network on production level Thanks in advance.

indongkim (Tue, 19 Oct 2021 06:29:53 GMT):
Has joined the channel.

amitchandra01 (Tue, 19 Oct 2021 08:08:24 GMT):
Has joined the channel.

mateokurti (Tue, 19 Oct 2021 08:41:26 GMT):
Has joined the channel.

Luis_Manuel_Cortes_Tirado (Tue, 19 Oct 2021 10:43:46 GMT):
Has joined the channel.

WadeBarnes (Tue, 19 Oct 2021 11:48:35 GMT):
Have a look at the available managed production level ledgers before considering hosting your own. https://sovrin.org/overview/

EliasKim (Tue, 19 Oct 2021 13:57:37 GMT):
Has joined the channel.

EliasKim (Tue, 19 Oct 2021 13:59:59 GMT):
Hey all, I'm trying to install the indy sdk on wsl running ubuntu 18.04 following the instructions on https://github.com/hyperledger/indy-sdk/blob/master/README.md , but whenever I try to update it says that the repository doesn't have a release file. Does anybody know what could be causing this problem?

EliasKim (Tue, 19 Oct 2021 14:00:39 GMT):
It also throws errors about the certificate not being trusted

WadeBarnes (Tue, 19 Oct 2021 14:11:07 GMT):
Can you please provide the specific list of commands you are running that lead up to the error?

EliasKim (Tue, 19 Oct 2021 14:13:46 GMT):
Yes! sudo apt-key adv --keyserver keyserver.ubuntu.com --recv-keys CE7709D068DB5E88 then sudo add-apt-repository "deb https://repo.sovrin.org/sdk/deb bionic master" then when I get the error is: sudo apt-get update

EliasKim (Tue, 19 Oct 2021 14:14:00 GMT):
I tried to follow the github readme as close as possible

WadeBarnes (Tue, 19 Oct 2021 14:22:00 GMT):
run `cat /etc/apt/sources.list` and see if `deb https://repo.sovrin.org/sdk/deb bionic master` is in the list.

WadeBarnes (Tue, 19 Oct 2021 14:22:43 GMT):
If not run `echo "deb https://repo.sovrin.org/sdk/deb bionic master" >> /etc/apt/sources.list`, you may need to use `sudo`.

WadeBarnes (Tue, 19 Oct 2021 14:23:10 GMT):
Then `apt-get update`.

EliasKim (Tue, 19 Oct 2021 14:23:18 GMT):

Clipboard - October 19, 2021 9:23 AM

EliasKim (Tue, 19 Oct 2021 14:23:25 GMT):
yep!

WadeBarnes (Tue, 19 Oct 2021 14:24:22 GMT):
I ran the commands without error. What is the error you are seeing specifically?

EliasKim (Tue, 19 Oct 2021 14:24:50 GMT):
I'll provide a screenshot, just a second

EliasKim (Tue, 19 Oct 2021 14:25:29 GMT):

Clipboard - October 19, 2021 9:25 AM

WadeBarnes (Tue, 19 Oct 2021 14:27:27 GMT):
The repositories it's complaining about, are they listed in your `/etc/apt/sources.list`?

WadeBarnes (Tue, 19 Oct 2021 14:27:54 GMT):
The `Release` and `InRelease` ones specifically?

EliasKim (Tue, 19 Oct 2021 14:29:31 GMT):
No, they are not

EliasKim (Tue, 19 Oct 2021 14:30:42 GMT):
The only like that says sovrin is the bionic master that i sent a screenshot of

EliasKim (Tue, 19 Oct 2021 14:30:42 GMT):
The only line that says sovrin is the bionic master that i sent a screenshot of

WadeBarnes (Tue, 19 Oct 2021 14:36:36 GMT):
Do you get an error when you run `apt-key adv --keyserver keyserver.ubuntu.com --recv-keys CE7709D068DB5E88`?

WadeBarnes (Tue, 19 Oct 2021 14:36:58 GMT):
Like this `E: gnupg, gnupg2 and gnupg1 do not seem to be installed, but one of them is required for this operation`

EliasKim (Tue, 19 Oct 2021 14:38:27 GMT):
no

EliasKim (Tue, 19 Oct 2021 14:38:46 GMT):

Clipboard - October 19, 2021 9:38 AM

WadeBarnes (Tue, 19 Oct 2021 14:43:54 GMT):
Try installing `ca-certificates`

WadeBarnes (Tue, 19 Oct 2021 14:45:02 GMT):
On a a raw ubuntu 18.04 image I had to install that before the `apt-get update` would work properly.

EliasKim (Tue, 19 Oct 2021 14:45:32 GMT):
Wow, worked perfectly

EliasKim (Tue, 19 Oct 2021 14:45:38 GMT):
thanks so much for your help, wade!

WadeBarnes (Tue, 19 Oct 2021 14:45:49 GMT):
:thumbsup:

david.liu (Tue, 19 Oct 2021 16:47:32 GMT):
Has joined the channel.

OsamaAzmy (Tue, 19 Oct 2021 22:48:03 GMT):
Has joined the channel.

ralvarezalvarez (Wed, 20 Oct 2021 03:25:43 GMT):
Has joined the channel.

subin8899 (Wed, 20 Oct 2021 04:34:00 GMT):
Has joined the channel.

nbAmit (Wed, 20 Oct 2021 04:52:25 GMT):
Has joined the channel.

SwapnaliDive (Wed, 20 Oct 2021 05:59:38 GMT):
Yes I checked the sovrin production level ledgers. I just need to study the considerations required to bring up huge indy network together.

DarshilPatel (Wed, 20 Oct 2021 11:19:50 GMT):
Has joined the channel.

shanest (Wed, 20 Oct 2021 12:00:31 GMT):
Hi i'm trying to use this image right here https://hub.docker.com/r/bcgovimages/von-image/tags unlike the one hosted on Github, this one doesn't come with manage.py to run a localhost testnet. My question is how do I use this image, when checked there are only 3 items inside namely requirements.txt, logs and ledger which the later are empty folders.

deepakgerawipro (Wed, 20 Oct 2021 12:13:08 GMT):
Has joined the channel.

WadeBarnes (Wed, 20 Oct 2021 13:26:49 GMT):
Hi @shanest, What is it you are trying to do with the image? `von-image` has a number of different uses. If you are trying to run a local dev ledger, I'd recommend referring to [von-network](https://github.com/bcgov/von-network) and using the `./manage` script to `build`, `start`, and `stop` the ledger. If it's the [von-network-base](https://github.com/bcgov/von-network/pkgs/container/von-network-base) image you're referring to as the "one hosted in GitHub", `von-image` is the base image for that image; https://github.com/bcgov/von-network/blob/master/Dockerfile#L1

WadeBarnes (Wed, 20 Oct 2021 13:29:09 GMT):
Check out the **Create New Indy Network** link here https://github.com/hyperledger/indy-node/issues/1674

shanest (Wed, 20 Oct 2021 13:38:43 GMT):
Yep. im building a demo and trying to run local dev ledger. I'm assuming I could do the initialization of von-network within a single docker-compose file actually. In that case, I'll stick to using the former. Thanks Wade.

WadeBarnes (Wed, 20 Oct 2021 13:40:06 GMT):
`von-network` is based on docker-compose. It has everything you need to run an local dev ledger.

WadeBarnes (Wed, 20 Oct 2021 13:40:06 GMT):
`von-network` is based on docker-compose. It has everything you need to run a local dev ledger.

shanest (Wed, 20 Oct 2021 13:48:18 GMT):
Got it. i'm very new to software development in general actually. Your input is very much appreciated. Please bear with me

webspeller (Wed, 20 Oct 2021 19:44:54 GMT):
Has joined the channel.

webspeller (Wed, 20 Oct 2021 20:29:10 GMT):
Hi, I am trying to make an open source Authentication System which will be used as an authentication tool for a crypto-currency based private project. I am confused whether I should be using the Kadena chain (as it supports more transactions) or should I be using Aries or Indy? As the summary of explanation says that Aries is a tool, could it be used with Kadena as a blockchain?

alieghmazi (Wed, 20 Oct 2021 21:07:41 GMT):
Has joined the channel.

jason_jaan (Wed, 20 Oct 2021 22:54:52 GMT):
Has joined the channel.

leonfrank (Thu, 21 Oct 2021 07:24:13 GMT):
Has joined the channel.

tpluck (Thu, 21 Oct 2021 07:56:07 GMT):
Has joined the channel.

garthr (Thu, 21 Oct 2021 08:34:40 GMT):
Has joined the channel.

sjana0 (Thu, 21 Oct 2021 10:34:40 GMT):
Has joined the channel.

outsidethecode (Thu, 21 Oct 2021 13:20:11 GMT):
Has joined the channel.

mccown (Thu, 21 Oct 2021 14:33:35 GMT):
Quick Reminder (due to time changes): The Identity Implementer's WG is meeting at the top of the hour. Wiki: https://wiki.hyperledger.org/display/IWG/2021-10-21+%3A+Identity+Implementers+WG+Call Zoom: https://zoom.us/my/hyperledger.community.3?pwd=UE90WHhEaHRqOGEyMkV3cldKa2d2dz09

webspeller (Thu, 21 Oct 2021 17:53:28 GMT):
anybody?

webspeller (Thu, 21 Oct 2021 17:53:33 GMT):
No answers?

weihong.ou (Thu, 21 Oct 2021 18:00:10 GMT):
Has joined the channel.

swcurran (Thu, 21 Oct 2021 18:31:39 GMT):
Not much guidance to provide. Don't know anything about Kadena. If it supports DIDs, it could in theory be used with Aries, but getting others to use it will be a challenge.

webspeller (Thu, 21 Oct 2021 19:03:00 GMT):
Thanks!

webspeller (Thu, 21 Oct 2021 19:03:28 GMT):
I don't think it supports DID, what special is required in a Blockchain to support DID?

webspeller (Thu, 21 Oct 2021 19:04:05 GMT):
Thanks! I don't think it supports DID, what special is required in a Blockchain to support DID?

swcurran (Thu, 21 Oct 2021 19:07:17 GMT):
Client code that allows a transaction that stores or references a DID. https://www.w3.org/TR/did-core/

webspeller (Thu, 21 Oct 2021 19:11:20 GMT):
Let me confirm

webspeller (Thu, 21 Oct 2021 19:12:01 GMT):
What ab0ut the scalability and speed? Are Indy and Aries able to support the scale and are fast enough?

webspeller (Thu, 21 Oct 2021 19:12:01 GMT):
What about the scalability and speed? Are Indy and Aries able to support the scale and are fast enough?

webspeller (Thu, 21 Oct 2021 19:16:31 GMT):
Kadena runs 20 parallel bitcoin like blockchains which makes it massively scalable and quick to respond. Kadena supports stuff like traffic volume, response time/transaction finality, security, integrations with other systems etc

webspeller (Thu, 21 Oct 2021 19:16:31 GMT):
Kadena (kadena.io) runs 20 parallel bitcoin like blockchains which makes it massively scalable and quick to respond. Kadena supports stuff like traffic volume, response time/transaction finality, security, integrations with other systems etc

lynn.bendixsen (Thu, 21 Oct 2021 20:37:18 GMT):
"Huge indy network" is currently not advised. The current recommendation is 25 distributed nodes and you might be okay going as high as 32 but would likely run into performance issues if you go any higher than that.

lukasgem (Thu, 21 Oct 2021 20:46:16 GMT):
Has joined the channel.

cvarjao (Thu, 21 Oct 2021 22:24:45 GMT):
Has joined the channel.

sfiguereo (Fri, 22 Oct 2021 01:11:56 GMT):
Has joined the channel.

SwapnaliDive (Fri, 22 Oct 2021 05:29:07 GMT):
I agree. @lynn.bendixsen

andra1996 (Fri, 22 Oct 2021 08:25:47 GMT):
Has joined the channel.

TimoGlastra (Fri, 22 Oct 2021 09:14:29 GMT):
What do people use to securely generate seeds? Is there a recommended tool that I can use for this?

NikunjPurohit (Fri, 22 Oct 2021 13:47:05 GMT):
Has joined the channel.

khamdb (Fri, 22 Oct 2021 13:47:56 GMT):
Has joined the channel.

lynn.bendixsen (Fri, 22 Oct 2021 15:04:48 GMT):
On Linux (or Mac) I use `pwgen -s 32 -1` to generate my secure seeds. You could also use a password keeper program. 1Password has an option to generate a 32 character password (seed). And 1Password is a secure way to store the seeds too.

TimoGlastra (Fri, 22 Oct 2021 15:53:53 GMT):
Thanks! :)

andrew.whitehead (Fri, 22 Oct 2021 15:57:48 GMT):
You have a much wider range if you generate bytes and encode it as base64, os.urandom in python isn’t a bad option

TimoGlastra (Fri, 22 Oct 2021 15:58:53 GMT):
I was using openssl rand, does that also provide the wide range?

andrew.whitehead (Fri, 22 Oct 2021 15:59:15 GMT):
Yes I thought that might be a good option as well

GeluUngur (Fri, 22 Oct 2021 18:05:31 GMT):
Has joined the channel.

acantrell (Fri, 22 Oct 2021 19:12:44 GMT):
Has joined the channel.

palacios (Fri, 22 Oct 2021 19:48:36 GMT):
Has joined the channel.

swcurran (Fri, 22 Oct 2021 22:35:52 GMT):
Indy is the ledger. Aries allows you to build agents that are peer to peer and that talk to whatever ledger you are using. With VCs you don't us

swcurran (Fri, 22 Oct 2021 22:35:52 GMT):
Indy is the ledger. Aries allows you to build agents that are peer to peer and that talk to whatever ledger you are using. With VCs you don't write to the ledger when issuing or verifying credentials. You can issue a billion VCs using just a single ledger transaction, so the scale of something like financial transactions is different.

swcurran (Fri, 22 Oct 2021 22:37:17 GMT):
Nothing about a VC itself goes on the ledger, and nothing is written at ledger issuance time.

BrianBauska (Sat, 23 Oct 2021 05:16:31 GMT):
Has joined the channel.

marchelbudi.k (Sat, 23 Oct 2021 05:47:39 GMT):
Has joined the channel.

KiranLakhotia (Sat, 23 Oct 2021 08:49:55 GMT):
Has joined the channel.

webzest (Sat, 23 Oct 2021 12:29:03 GMT):
Has joined the channel.

jrojek (Sun, 24 Oct 2021 05:27:44 GMT):
Has joined the channel.

jrd10 (Sun, 24 Oct 2021 06:04:32 GMT):
Has joined the channel.

lincyaw (Sun, 24 Oct 2021 08:03:37 GMT):
Has joined the channel.

amankr1619 (Sun, 24 Oct 2021 20:21:37 GMT):
Has joined the channel.

JorgeLuisAcosta (Mon, 25 Oct 2021 07:00:46 GMT):
Has joined the channel.

dovanan95 (Mon, 25 Oct 2021 08:26:33 GMT):
Has joined the channel.

webspeller (Mon, 25 Oct 2021 08:29:22 GMT):
I heard of something like writing over the ledger when there are 100 or say million transactions done?? Delaying the frequency of input and accommodating the data in size of the block? Is it really happening or were those just the talks? I hope my questions are not bothering you.

AtlasSMT (Mon, 25 Oct 2021 10:11:08 GMT):
Has joined the channel.

shadiayoub (Mon, 25 Oct 2021 11:24:53 GMT):
Has joined the channel.

jean-claudedalot (Mon, 25 Oct 2021 13:50:08 GMT):
Has joined the channel.

flaviocal (Mon, 25 Oct 2021 14:24:36 GMT):
Has joined the channel.

spunkymandar (Mon, 25 Oct 2021 16:50:38 GMT):
Has joined the channel.

dmimaz (Mon, 25 Oct 2021 17:36:09 GMT):
Has joined the channel.

swcurran (Mon, 25 Oct 2021 17:36:21 GMT):
No problem. Since writing occurs when you want to define a type of credential, and not when you issue a credential, the volumes are way lower. The highest writing is in revoking credentials (if that is needed) and for most business cases, that happens once a day per type of credential.

swcurran (Mon, 25 Oct 2021 17:37:07 GMT):
Revocation is still something we'd like to scale, but it's not a ledger issue to scale it -- it's getting the right cryptography for revocation.

Tenno (Tue, 26 Oct 2021 03:42:20 GMT):
Has joined the channel.

kamalraj2000 (Tue, 26 Oct 2021 07:17:43 GMT):
Has joined the channel.

dheerajbudhiraja (Tue, 26 Oct 2021 08:09:02 GMT):
Has joined the channel.

webspeller (Tue, 26 Oct 2021 17:19:34 GMT):
so you mean to say that we can use Indy as a ledger and Aries as p2p tool between clients (company/institution/organisation and user/individual/group/employee) and use Kadena as a Blockchain for making a VC and use Indy and Kadena (managed through microservices)?Does it sound like a valid technical infra?

webspeller (Tue, 26 Oct 2021 17:20:39 GMT):
And are there any Slack/Discord channels I can connect to please? Or is there any app I can use to connect to this chat?

artemgorev (Tue, 26 Oct 2021 18:18:08 GMT):
Has joined the channel.

swcurran (Tue, 26 Oct 2021 19:02:08 GMT):
There is no need for a blockchain for VCs. The VCs are shared p2p -- the issuer creates a VC for a holder and gives it to that holder. The holder keeps it locally. A verifier requests a presentation that requires the use of the VC, and the holder constructs the proof based on the VC and gives it to the verifier. Nothing about the issuance or verification requires writing to a blockchain. Does that make sense?

trilawney (Tue, 26 Oct 2021 19:08:54 GMT):
Has joined the channel.

Lucampelli (Tue, 26 Oct 2021 21:07:34 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=93Qs9xBXba28jcYpy) Help?

Lucampelli (Tue, 26 Oct 2021 21:07:42 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=N7Go2RRvcGWgCkZnv) Help?

tomaspozo (Tue, 26 Oct 2021 21:55:29 GMT):
Has joined the channel.

webspeller (Tue, 26 Oct 2021 23:26:48 GMT):
I think I understand that, VC is a virtual coin which is developed and distributed by an organisation or identity or group to its members. They hold it and the value of coin rises according to the demand and usage. I don't think it will stay locally if it is part of a blockchain, afterall we are creating the VC out of an established coin/currency.

webspeller (Tue, 26 Oct 2021 23:26:48 GMT):
I think I understand that, VC is a virtual currency/coin which is developed and distributed by an organisation or identity or group to its members within an ecosystem. They hold it and the value of coin rises according to the demand and usage. I don't think it will stay locally if it is part of a blockchain, after all we are creating the VC out of an established coin/currency.

Eric.Lin (Tue, 26 Oct 2021 23:27:04 GMT):
Has joined the channel.

webspeller (Tue, 26 Oct 2021 23:31:12 GMT):
As per my understanding, We need blockchain to keep the data secure, we can call it somewhat secure database. But, If I am making an ecommerce website and keeping the identity and currency and its transactions and making smart contracts for distribution of currency according to some rules, then will that not be the use of Blockchain? VC is a subcoin (sub currency) which I can customize and make using the parent financial blockchain's rules and algo. Please correct me if I either missed something or didn't understand or misunderstood something.

Luis_Manuel_Cortes_Tirado (Wed, 27 Oct 2021 06:27:03 GMT):
Has left the channel.

lyc9702 (Wed, 27 Oct 2021 06:36:20 GMT):
Has joined the channel.

Ferdinand.wittmann (Wed, 27 Oct 2021 09:39:25 GMT):
Has joined the channel.

Ferdinand.wittmann (Wed, 27 Oct 2021 09:50:31 GMT):
Topic: Transaction Endorser for Sovrin Main Net. Hi I have developed a aca-py controller, that issues credentials currently on Sovrins BuilderNet. Our goal is to be able to issue credentials anchored on Sovrins Main Net. We therefore need an endorser for our transaction. We are Swiss-based non profit organisation and are now looking for an endorser that we can use together with the new endorsing functionality provided by aca-py. Is anybody working on this right now or nows of an endorser that cloud provide this functionality for us?

SwapnaliDive (Wed, 27 Oct 2021 10:42:27 GMT):
Hello Team, How can I generate genesis nodes transaction file using separate ips for node and client Node example with separate client_ip and node_ip - {"alias":"ev1","client_ip":"54.207.36.81","client_port":"9702","node_ip":"18.231.96.215","node_port":"9701","services":["VALIDATOR"]},"dest":"GWgp6huggos5HrzHVDy5xeBkYHxPvrRZzjPNAyJAqpjA"},"metadata":{"from":"J4N1K1SEB8uY2muwmecY5q"},"type":"0"},"txnMetadata":{"seqNo":1,"txnId":"b0c82a3ade3497964cb8034be915da179459287823d92b5717e6d642784c50e6"},"ver":"1"}

MontassarNAGHMOUCHI (Wed, 27 Oct 2021 11:35:53 GMT):
Has joined the channel.

WadeBarnes (Wed, 27 Oct 2021 13:17:55 GMT):
Would you be able to provide a bit more background on what you're doing?

SwapnaliDive (Wed, 27 Oct 2021 13:32:18 GMT):
For security reasons I read, we should keep node_ip address and client_ip as different, becuase if we connect cli with validator node and if the node_ip and client_ip address is same in that case user may access the steward node information

WadeBarnes (Wed, 27 Oct 2021 13:34:06 GMT):
Are you trying to build your own indy network? If so for what purpose?

WadeBarnes (Wed, 27 Oct 2021 13:36:31 GMT):
If for development, your better bet is [bcgov/von-network](https://github.com/bcgov/von-network). If for `beta`, `test`, or `prod`, there are the hosted Sovrin Networks; https://sovrin.org/overview/

SwapnaliDive (Wed, 27 Oct 2021 13:37:43 GMT):
If I see the soverin genesis file Link : https://sovrin-mainnet-browser.vonx.io/genesis how to set different addresses that too for genesis file.

SwapnaliDive (Wed, 27 Oct 2021 13:38:10 GMT):
I am building own indy network just to figure out how exactly the things work between the nodes.

WadeBarnes (Wed, 27 Oct 2021 13:39:49 GMT):
Resources for that are here; https://github.com/hyperledger/indy-node/issues/1674 The `Create New Indy Network` is the better of the two. It also provides a link to the script used to generate the genesis files.

WadeBarnes (Wed, 27 Oct 2021 13:41:43 GMT):
As for building a node that supports two IP addresses, you'll need to configure your nodes with two separate network interfaces and assign separate IPs.

viktor.pulyak (Wed, 27 Oct 2021 15:31:36 GMT):
Has joined the channel.

rolandw (Wed, 27 Oct 2021 17:09:36 GMT):
Has joined the channel.

swcurran (Thu, 28 Oct 2021 00:02:09 GMT):
We are definitely not talking about the same thing. I'm talking about Verifiable Credentials as VCs, and they are not a virtual currency or coin -- nothing to do with that. A VC is like a paper credential (e.g. a driver's licence, passport, university degree) that an issuer gives to a holder and the holder presents to a verifier. No cryptocurrency involved in the general use case.

Jotenks (Thu, 28 Oct 2021 10:16:27 GMT):
Has joined the channel.

gut (Thu, 28 Oct 2021 10:53:41 GMT):
Which is the best way to create an edge agent storing the wallet locally? Which project helps with that? Any "officially" accepted clue is welcomed.

important (Thu, 28 Oct 2021 11:05:10 GMT):
Has joined the channel.

SwapnaliDive (Thu, 28 Oct 2021 11:53:39 GMT):
Two separate network interfaces means two seperate aws server IPs ?

yukesong (Thu, 28 Oct 2021 12:20:46 GMT):
Has joined the channel.

WadeBarnes (Thu, 28 Oct 2021 13:51:04 GMT):
Correct

Acarcan (Thu, 28 Oct 2021 14:17:47 GMT):
Has joined the channel.

AlessandroPagliaro (Thu, 28 Oct 2021 15:06:02 GMT):
Has joined the channel.

Mellywins (Thu, 28 Oct 2021 15:28:18 GMT):
Has joined the channel.

webspeller (Thu, 28 Oct 2021 16:20:14 GMT):
oh.. I thought we are discussing the use case I mentioned.

webspeller (Thu, 28 Oct 2021 16:27:13 GMT):
I am trying to figure out the tech stack with the amount of knowledge I have. There are 2 projects 1) An Identity Platform, something like Evernym or Kiva

webspeller (Thu, 28 Oct 2021 16:28:05 GMT):
2) A web platform which will allow clients identified through project 1. This web platform will use currency ( a Coin on Kadena Blockchain).

webspeller (Thu, 28 Oct 2021 16:29:20 GMT):
My confusion is, Whether I should be using Kadena as a blockchain for: 1) both the projects 2) Either of the projects 3) for coin only.

webspeller (Thu, 28 Oct 2021 16:31:22 GMT):
If not Kadena, then should I be using Indy as blockchain and Aries as web/mobile p2p transaction tool (I believe it is something like Evernym video, in which they have made a mobile app and once they scan the QR code through mobile they get the certificate downloaded to their mobile and can use that further for transaction or communication with the organisation).

webspeller (Thu, 28 Oct 2021 16:52:26 GMT):
I am worried abt both the issuence of tokens on chain as well as verification of ID/person each time they connect to access the VC (verifiable credential).

webspeller (Thu, 28 Oct 2021 16:52:26 GMT):
I am worried abt both - the issuence of tokens on chain as well as verification of ID/person each time they connect to access the VC (verifiable credential).

TaylorBelk1 (Thu, 28 Oct 2021 16:57:52 GMT):
Has joined the channel.

chadmott (Thu, 28 Oct 2021 20:49:49 GMT):
Has joined the channel.

KaranKulwal (Fri, 29 Oct 2021 05:07:15 GMT):
Has joined the channel.

hblockchain (Fri, 29 Oct 2021 05:50:01 GMT):
Has joined the channel.

BabitaBisht1 (Fri, 29 Oct 2021 07:54:34 GMT):
Has joined the channel.

MikaLammi (Fri, 29 Oct 2021 08:01:00 GMT):
Has joined the channel.

marcus.schoeller (Fri, 29 Oct 2021 10:17:04 GMT):
Has joined the channel.

hailx (Fri, 29 Oct 2021 12:59:32 GMT):
Has joined the channel.

Natinasss (Fri, 29 Oct 2021 13:07:38 GMT):
Has joined the channel.

SwapnaliDive (Fri, 29 Oct 2021 13:13:39 GMT):
Thank you @WadeBarnes

RajeshNavagare (Fri, 29 Oct 2021 16:39:54 GMT):
Has joined the channel.

deoalade (Fri, 29 Oct 2021 17:16:53 GMT):
Has joined the channel.

omago123 (Sat, 30 Oct 2021 06:56:59 GMT):
Has joined the channel.

Kazuya.N (Sat, 30 Oct 2021 07:31:27 GMT):
Has joined the channel.

efrain (Sat, 30 Oct 2021 09:28:41 GMT):
Has joined the channel.

edouardfrance (Sat, 30 Oct 2021 14:40:05 GMT):
Has joined the channel.

Santiagotwm99 (Sat, 30 Oct 2021 21:18:54 GMT):
Has joined the channel.

JeremyChou0314 (Sun, 31 Oct 2021 11:12:35 GMT):
Has joined the channel.

tleonard1941 (Sun, 31 Oct 2021 17:06:14 GMT):
Has joined the channel.

rohan333 (Mon, 01 Nov 2021 03:55:24 GMT):
Has joined the channel.

conanoc (Mon, 01 Nov 2021 05:05:46 GMT):
AFAIK, every agents store the wallet locally, and there is no other option.

szbalazs (Mon, 01 Nov 2021 11:12:11 GMT):
Has joined the channel.

pperez (Mon, 01 Nov 2021 11:23:58 GMT):
Has joined the channel.

SwapnaliDive (Mon, 01 Nov 2021 12:07:51 GMT):
@swcurran Is it released ? Now, can we use it on ubuntu 20.04

keithyweethy (Mon, 01 Nov 2021 16:52:11 GMT):
Has joined the channel.

swcurran (Mon, 01 Nov 2021 17:21:37 GMT):
Unfortunately not yet. Progress continues to be made, but the work is not yet completed.

teerakiat (Tue, 02 Nov 2021 05:00:40 GMT):
Has joined the channel.

Rigwarl (Tue, 02 Nov 2021 06:26:36 GMT):
Has joined the channel.

vinodhini (Tue, 02 Nov 2021 06:32:40 GMT):
Has joined the channel.

Rigwarl (Tue, 02 Nov 2021 06:37:28 GMT):
Has left the channel.

y8l (Tue, 02 Nov 2021 11:57:16 GMT):
Has joined the channel.

rafaadan6 (Tue, 02 Nov 2021 18:08:34 GMT):
Has joined the channel.

rafaadan6 (Tue, 02 Nov 2021 18:10:16 GMT):
Hi! I want to deploy an enviroment with docker but I am a bit lost after checking the docs. How should I deploy the Public Ledger, with indy-node code, indy-sdk code or with VON Network?

rafaadan6 (Tue, 02 Nov 2021 18:10:26 GMT):
I want to use Aries Agent after that

imillar99 (Tue, 02 Nov 2021 20:43:53 GMT):
Has joined the channel.

aarjav (Tue, 02 Nov 2021 20:54:47 GMT):
Has joined the channel.

ElisJacobson (Tue, 02 Nov 2021 23:06:13 GMT):
Has joined the channel.

terexitarius (Wed, 03 Nov 2021 02:45:43 GMT):
Has joined the channel.

Howie001 (Wed, 03 Nov 2021 02:48:24 GMT):
Has joined the channel.

frankliawesome (Wed, 03 Nov 2021 03:37:57 GMT):
Has joined the channel.

jaysonr (Wed, 03 Nov 2021 03:39:30 GMT):
Has joined the channel.

SriniGovindaswamy (Wed, 03 Nov 2021 06:01:00 GMT):
Has joined the channel.

hhgfgg (Wed, 03 Nov 2021 12:05:55 GMT):
Has joined the channel.

JiriPeinlich (Wed, 03 Nov 2021 14:35:25 GMT):
Has joined the channel.

Stevedimaras (Wed, 03 Nov 2021 17:05:20 GMT):
Has joined the channel.

alfahami (Wed, 03 Nov 2021 17:55:33 GMT):
Has joined the channel.

hellipps (Wed, 03 Nov 2021 18:03:46 GMT):
Has joined the channel.

swcurran (Wed, 03 Nov 2021 18:29:05 GMT):
To get started, use von-network as your ledger and an Aries agent (perhap ACA-Py) as your "ledger client". The Aries agent embeds the indy-sdk code. That bootstrapping will help your understanding for building to a production product.

mccown (Wed, 03 Nov 2021 19:51:21 GMT):
The Identity Implementer's WG is meeting tomorrow (Nov 4th @ 9am MT / 3pm UTC). We will review a number of updates and milestones from several SSI industry WGs. I will also be giving a presentation on how to auto-generate language wrappers from Rust libraries (e.g., DIDComm_rs, indy-sdk, etc.). Wiki: https://wiki.hyperledger.org/display/IWG/2021-11-04+%3A+Identity+Implementers+WG+Call Zoom: https://zoom.us/my/hyperledger.community.3?pwd=UE90WHhEaHRqOGEyMkV3cldKa2d2dz09

rafaadan6 (Wed, 03 Nov 2021 20:45:21 GMT):
Hi Stephen, thanks for the reply!

rafaadan6 (Wed, 03 Nov 2021 20:46:30 GMT):
Hi Stephen, thanks for the reply! I will do that. For now, I'm doing the edX course Becoming a Hyperledger Aries Developer and I think it would help me a lot.

cemmurphy (Wed, 03 Nov 2021 21:59:32 GMT):
Has joined the channel.

minsecprisoner (Thu, 04 Nov 2021 09:53:49 GMT):
Has joined the channel.

CarlosGarcia.Briones (Thu, 04 Nov 2021 10:26:27 GMT):
Has joined the channel.

mafer_027 (Thu, 04 Nov 2021 15:38:06 GMT):
Has joined the channel.

deathdealer007 (Fri, 05 Nov 2021 03:11:16 GMT):
Has joined the channel.

ksgr5566 (Fri, 05 Nov 2021 04:30:16 GMT):
Has joined the channel.

alvone01 (Fri, 05 Nov 2021 06:58:44 GMT):
Has joined the channel.

surabhi17 (Fri, 05 Nov 2021 08:45:59 GMT):
Has joined the channel.

SwapnaliDive (Fri, 05 Nov 2021 10:39:10 GMT):
Hello Team, I am trying to install indy-node using below steps : sudo apt-key adv --keyserver keyserver.ubuntu.com --recv-keys CE7709D068DB5E88 sudo bash -c 'echo "deb https://repo.sovrin.org/deb xenial stable" >> /etc/apt/sources.list' sudo apt-get update sudo apt-get install indy-node I am facing the error while apt-get update : Failed to fetch https://repo.sovrin.org/sdk/deb/dists/xenial/stable/binary-amd64/Packages server certificate verification failed. CAfile: /etc/ssl/certs/ca-certificates.crt CRLfile: Can anyone please help to understand why this error is occuring ?? Do I have to install anything before doing this steps ??

tperes (Fri, 05 Nov 2021 10:56:14 GMT):
Has joined the channel.

WadeBarnes (Fri, 05 Nov 2021 13:57:06 GMT):
Update or install the `ca-certificates` package> ``` apt-get update && apt-get install -y ca-certificates ``` and then try again.

Makabongwe (Fri, 05 Nov 2021 14:39:29 GMT):
Has joined the channel.

elDanielain (Fri, 05 Nov 2021 15:59:55 GMT):
Has joined the channel.

SySagar (Sun, 07 Nov 2021 04:25:26 GMT):
Has joined the channel.

suisingLai (Sun, 07 Nov 2021 14:07:21 GMT):
Has joined the channel.

ceejaymi (Sun, 07 Nov 2021 18:42:13 GMT):
Has joined the channel.

Ezzah 1 (Mon, 08 Nov 2021 01:10:36 GMT):
Has joined the channel.

Jangto (Mon, 08 Nov 2021 03:27:46 GMT):
Has joined the channel.

ThinhPhan (Mon, 08 Nov 2021 07:28:00 GMT):
Has joined the channel.

sristhiassudani (Mon, 08 Nov 2021 09:22:21 GMT):
Has joined the channel.

NSagar (Mon, 08 Nov 2021 12:25:57 GMT):
Has joined the channel.

RyanJJones10 (Mon, 08 Nov 2021 17:13:45 GMT):
Has joined the channel.

swcurran (Tue, 09 Nov 2021 04:15:13 GMT):
I've done a pass at the TSC Quarterly Report on Indy for this past quarter. Please review and adjust as necessary on the Hyperledger Wiki: https://wiki.hyperledger.org/display/TSC/2021+Q4+Hyperledger+Indy

shin1yama (Tue, 09 Nov 2021 07:02:16 GMT):
Has joined the channel.

LakshmiMurali (Tue, 09 Nov 2021 07:50:21 GMT):
Has joined the channel.

Aljone (Tue, 09 Nov 2021 15:58:36 GMT):
Has joined the channel.

rockson (Tue, 09 Nov 2021 17:02:09 GMT):
Has joined the channel.

knguyenBC (Tue, 09 Nov 2021 17:17:59 GMT):
Has joined the channel.

ghastly (Tue, 09 Nov 2021 17:32:50 GMT):
Has joined the channel.

sdunstan (Tue, 09 Nov 2021 20:47:43 GMT):
Has joined the channel.

LEOISHere (Wed, 10 Nov 2021 00:03:22 GMT):
Has joined the channel.

ffabregas (Wed, 10 Nov 2021 03:56:07 GMT):
Has joined the channel.

DeepromSomkiadcharoen (Wed, 10 Nov 2021 03:56:07 GMT):
Has joined the channel.

Strohhut (Wed, 10 Nov 2021 17:27:06 GMT):
Has joined the channel.

loziniak (Thu, 11 Nov 2021 00:32:27 GMT):
Has joined the channel.

ChangmingLiu (Thu, 11 Nov 2021 01:38:01 GMT):
Has joined the channel.

newbieTech (Thu, 11 Nov 2021 02:56:28 GMT):
Has joined the channel.

uguroney (Thu, 11 Nov 2021 08:38:29 GMT):
Has joined the channel.

Monavari (Thu, 11 Nov 2021 11:18:16 GMT):
Has joined the channel.

leostereo (Thu, 11 Nov 2021 13:42:39 GMT):
Has joined the channel.

leostereo (Thu, 11 Nov 2021 13:42:50 GMT):
hello !!!

leostereo (Thu, 11 Nov 2021 13:43:20 GMT):
my name is leandro , im looking for some help beggining with DID.

ianco (Thu, 11 Nov 2021 13:50:35 GMT):
There are a couple of good edx courses:

ianco (Thu, 11 Nov 2021 13:50:36 GMT):
https://www.edx.org/course/identity-in-hyperledger-aries-indy-and-ursa

ianco (Thu, 11 Nov 2021 13:50:48 GMT):
https://www.edx.org/course/becoming-a-hyperledger-aries-developer

ianco (Thu, 11 Nov 2021 13:51:07 GMT):
... and the aries cloudagent python has a lot of demo's and tutorials:

ianco (Thu, 11 Nov 2021 13:51:07 GMT):
... and the aries cloudagent python repository has a lot of demo's and tutorials:

ianco (Thu, 11 Nov 2021 13:51:19 GMT):
https://github.com/hyperledger/aries-cloudagent-python

leostereo (Thu, 11 Nov 2021 13:53:43 GMT):
excelent!!

leostereo (Thu, 11 Nov 2021 13:53:50 GMT):
thanks

anuj-kh (Thu, 11 Nov 2021 14:01:40 GMT):
Has joined the channel.

sebposselt (Thu, 11 Nov 2021 14:44:53 GMT):
Has joined the channel.

fab-10 (Thu, 11 Nov 2021 15:10:10 GMT):
Has joined the channel.

ricc.zappoli (Thu, 11 Nov 2021 16:03:11 GMT):
Has left the channel.

ArhangelSky999 (Thu, 11 Nov 2021 16:47:21 GMT):
Has joined the channel.

knichols (Thu, 11 Nov 2021 21:59:13 GMT):
Has joined the channel.

maurorambo (Fri, 12 Nov 2021 00:34:02 GMT):
Has joined the channel.

ChrisPaterson (Fri, 12 Nov 2021 02:35:32 GMT):
Has joined the channel.

kukgini (Fri, 12 Nov 2021 02:54:02 GMT):
Has left the channel.

sekinosekai (Fri, 12 Nov 2021 09:17:01 GMT):
Has joined the channel.

RomanBorisov (Fri, 12 Nov 2021 14:04:59 GMT):
Has joined the channel.

somaningp (Fri, 12 Nov 2021 14:29:25 GMT):
Has joined the channel.

loziniak (Fri, 12 Nov 2021 15:38:50 GMT):
Has left the channel.

Andrewval (Fri, 12 Nov 2021 16:46:22 GMT):
Has joined the channel.

ChheunPhearaeun (Fri, 12 Nov 2021 17:48:40 GMT):
Has joined the channel.

Clippo 3 (Fri, 12 Nov 2021 17:56:22 GMT):
Has joined the channel.

matt.nelson.csi (Fri, 12 Nov 2021 20:56:07 GMT):
Has joined the channel.

luoyu 1 (Fri, 12 Nov 2021 22:32:26 GMT):
Has joined the channel.

chandra100100 (Sat, 13 Nov 2021 06:58:25 GMT):
Has joined the channel.

Beshir (Sat, 13 Nov 2021 08:44:54 GMT):
Has joined the channel.

VladimirRadeski (Sat, 13 Nov 2021 09:32:26 GMT):
Has joined the channel.

manolaz (Sun, 14 Nov 2021 04:51:26 GMT):
Has joined the channel.

xeon123 (Mon, 15 Nov 2021 08:54:54 GMT):
Has joined the channel.

Salt4f (Mon, 15 Nov 2021 09:46:53 GMT):
Has joined the channel.

SwapnaliDive (Mon, 15 Nov 2021 11:52:51 GMT):
Hello Team , How did the sovrin network integrated the payment mechanism on the network? Is is managed on-chain or off-chain? We saw the payment functions in indy-cli but couldn't decipher what was the currency and/or token used and how it's done. Can someone please guide us on the same

ybuser (Mon, 15 Nov 2021 12:06:48 GMT):
Has joined the channel.

matijasipek (Mon, 15 Nov 2021 12:21:35 GMT):
Has joined the channel.

amit1nayak (Mon, 15 Nov 2021 12:33:31 GMT):
Has joined the channel.

RajaDhanushkodi (Mon, 15 Nov 2021 12:48:59 GMT):
Has joined the channel.

sidharthan.j (Mon, 15 Nov 2021 14:08:36 GMT):
Has joined the channel.

sidharthan.j (Mon, 15 Nov 2021 14:14:40 GMT):
hi team . i am trying to deploy the indy test network by using the docker method with the cmd './pool_start.sh' but getting an error ``` The following packages have unmet dependencies: indy-plenum : Depends: python3-orderedset (= 2.0) but 2.0.3 is to be installed Depends: python3-psutil (= 5.4.3) but 5.6.6 is to be installed Depends: python3-psutil (= 5.4.3) but 5.6.6 is to be installed Depends: python3-pympler (= 0.5) but 0.8 is to be installed E: Unable to correct problems, you have held broken packages. ```

sidharthan.j (Mon, 15 Nov 2021 14:15:00 GMT):
is there a way to get over this..thanks in advance.

WadeBarnes (Mon, 15 Nov 2021 14:16:29 GMT):
The payment mechanism is on ledger. It's still experimental and is only available on BuilderNet and StagingNet. More information can be found here; https://sovrin.org/wp-content/uploads/Token-Test-Instructions.pdf

WadeBarnes (Mon, 15 Nov 2021 14:18:57 GMT):
The references to the token have been temporarily removed from the https://selfserve.sovrin.org/ page. You can still register for test tokens using the following `curl`: ``` curl --location --request POST 'https://selfserve.sovrin.org/nym' \ --header 'Content-Type: application/json' \ --data-raw '{ "network": "buildernet", "did": "your_did", "verkey": "" }' ```

WadeBarnes (Mon, 15 Nov 2021 14:18:57 GMT):
The references to the token have been temporarily removed from the https://selfserve.sovrin.org/ page. You can still register for test tokens using the following `curl`: ``` curl --location --request POST 'https://selfserve.sovrin.org/nym' \ --header 'Content-Type: application/json' \ --data-raw '{ "network": "buildernet", "did": "", "verkey": "" }' ```

MarioMichel (Mon, 15 Nov 2021 14:19:46 GMT):
Has joined the channel.

WadeBarnes (Mon, 15 Nov 2021 14:20:56 GMT):
[bcgov/von-network](https://github.com/bcgov/von-network) provides an easier way to run a test network for development.

matejstern (Mon, 15 Nov 2021 14:57:24 GMT):
Has joined the channel.

elenaizaguirre (Mon, 15 Nov 2021 15:54:06 GMT):
Has joined the channel.

matisalimba (Mon, 15 Nov 2021 17:24:38 GMT):
Has joined the channel.

RyanKoch (Mon, 15 Nov 2021 19:51:01 GMT):
Has joined the channel.

AlexTran (Mon, 15 Nov 2021 20:58:47 GMT):
Has joined the channel.

panicstyle (Tue, 16 Nov 2021 01:02:10 GMT):
Has joined the channel.

mamba2017 (Tue, 16 Nov 2021 02:24:05 GMT):
Has joined the channel.

SwapnaliDive (Tue, 16 Nov 2021 05:30:27 GMT):
Do sovrin manually charge the 10$ amount on registering the DID

PHAGUNJAIN (Tue, 16 Nov 2021 08:43:43 GMT):
Has joined the channel.

MiguelCabeza (Tue, 16 Nov 2021 10:17:07 GMT):
Has joined the channel.

WadeBarnes (Tue, 16 Nov 2021 13:39:10 GMT):
Yes

Helsonkumar (Tue, 16 Nov 2021 13:49:35 GMT):
Has joined the channel.

xlswirlds (Tue, 16 Nov 2021 15:31:16 GMT):
Has joined the channel.

smannathan (Tue, 16 Nov 2021 17:11:28 GMT):
Has joined the channel.

hegde.rahul (Wed, 17 Nov 2021 01:20:10 GMT):
Has joined the channel.

Bert (Wed, 17 Nov 2021 01:53:41 GMT):
Has joined the channel.

SwapnaliDive (Wed, 17 Nov 2021 04:22:52 GMT):
\

Arjen (Wed, 17 Nov 2021 09:23:29 GMT):
Has joined the channel.

beeraaron81 (Wed, 17 Nov 2021 13:01:11 GMT):
Has joined the channel.

beeraaron81 (Wed, 17 Nov 2021 13:03:50 GMT):
Has left the channel.

samsonlarsson (Wed, 17 Nov 2021 14:02:08 GMT):
Has joined the channel.

gaberasturi (Wed, 17 Nov 2021 15:57:18 GMT):
Hello, We have a doubt about credentials revocation and tails file and what is write into the ledger. We are using .NET implementation. When we register a revocable credential we see that in the ledger at REVOC_REG_DEF the tails fail location URI is composed by Issuer endpoint URI + /tails/.... `Tails file location: http://ISSUER_URI/tails/......` This location URI is working so we understand that Issuer is doing as "proxy" to tails server? because our tails server is in other location. Yes we are registering in the Issuer `RevocationServerUrl` pointing to the tails sever. As is pointed in https://github.com/hyperledger/indy-hipe/tree/master/text/0011-cred-revocation#putting-it-all-together > - Verifiers do not need to contact issuers or consult a revocation list to test revocation. We are assuming that nobody (Verifier or Prover) need to interact with Issuer in verfication protocol. Our doubt is if all is working as expected? or we have something bad configured. Thanks in advanced.

swcurran (Wed, 17 Nov 2021 20:32:22 GMT):
The issuer needs to publish the tails file for a revocation registry, and needs to write updates to the RevReg on the ledger (RevRegEntry transactions). The holder needs to get the tails file associated with the credential issued to them. This is a static file, so in theory, they only need to grab it once and hold on to it. To construct a proof, they need to get the RevReg state (usually, the latest) from the ledger, and use that with the tails file to create a proof of non-revocation. The verifier needs to get the RevReg state as well (just the accumulator from that) and use that to verify the proof of non-revocation. Hope that makes sense.

EmilioFL (Wed, 17 Nov 2021 22:13:16 GMT):
Has joined the channel.

shimizut (Thu, 18 Nov 2021 04:12:13 GMT):
Has joined the channel.

shimizut (Thu, 18 Nov 2021 04:33:28 GMT):
Hello, I am setting up environment for developing indy-node and meeting some problems: `pytest` outputs errors. When testing indy-plenum, I got the following output. ``` = 7 failed, 11453 passed, 307 skipped, 71 warnings, 1089 error in 2380.80 seconds = ``` I executed commands according to https://github.com/hyperledger/indy-node/blob/master/docs/source/setup-dev.md on ubuntu 16 https://releases.ubuntu.com/16.04/ , but the test `pytest . ` did not run. According to error messages, I executed `pip install --upgrade 'setuptools<45.0.0`, `pip install pip==9.0.3` and so on. Then, `pytest .` ran but showed errors. For example I got the following. Actually, plenum/config.py does not include `GENESIS_DIR` ``` E AttributeError: module 'plenum.config' has no attribute 'GENESIS_DIR' ``` If this question is not suitable here, I would appreciate it if you could tell me where I should write. Thanks in adnvace.

shimizut (Thu, 18 Nov 2021 04:33:28 GMT):
Hello, I am setting up environment for developing indy-node and meeting some problems: `pytest` outputs errors. When testing indy-plenum, I got the following output. ``` = 7 failed, 11453 passed, 307 skipped, 71 warnings, 1089 error in 2380.80 seconds = ``` I executed commands according to https://github.com/hyperledger/indy-node/blob/master/docs/source/setup-dev.md on ubuntu 16 https://releases.ubuntu.com/16.04/ , but the test `pytest . ` did not run. According to error messages, I executed `pip install --upgrade 'setuptools<45.0.0`, `pip install pip==9.0.3` and so on. Then, `pytest .` ran but showed errors. For example I got the following. Actually, plenum/config.py does not include `GENESIS_DIR` ``` E AttributeError: module 'plenum.config' has no attribute 'GENESIS_DIR' ``` Does this output show an installation problem? If this question is not suitable here, I would appreciate it if you could tell me where I should write. Thanks in adnvace.

shimizut (Thu, 18 Nov 2021 04:33:28 GMT):
Hello, I am setting up environment for developing indy-node and meeting some problems: `pytest` outputs errors. When testing indy-plenum, I got the following output. ``` = 7 failed, 11453 passed, 307 skipped, 71 warnings, 1089 error in 2380.80 seconds = ``` I executed commands according to https://github.com/hyperledger/indy-node/blob/master/docs/source/setup-dev.md on ubuntu 16 https://releases.ubuntu.com/16.04/ , but the test `pytest . ` did not run. According to error messages, I additionally executed `pip install --upgrade 'setuptools<45.0.0`, `pip install pip==9.0.3` and so on. Then, `pytest .` ran but showed errors. For example I got the following. Actually, plenum/config.py does not include `GENESIS_DIR` ``` E AttributeError: module 'plenum.config' has no attribute 'GENESIS_DIR' ``` Does this output show an installation problem? If this question is not suitable here, I would appreciate it if you could tell me where I should write. Thanks in adnvace.

shimizut (Thu, 18 Nov 2021 04:33:28 GMT):
Hello, I am setting up environment for developing indy-node and meeting some problems: `pytest` outputs errors. When testing indy-plenum, I got the following output. ``` = 7 failed, 11453 passed, 307 skipped, 71 warnings, 1089 error in 2380.80 seconds = ``` indy-node's test showed a similar result, which included fails and errors. I executed commands according to https://github.com/hyperledger/indy-node/blob/master/docs/source/setup-dev.md on ubuntu 16 https://releases.ubuntu.com/16.04/ , but the test `pytest . ` did not run. According to error messages, I additionally executed `pip install --upgrade 'setuptools<45.0.0`, `pip install pip==9.0.3` and so on. Then, `pytest .` ran but showed errors. For example I got the following. Actually, plenum/config.py does not include `GENESIS_DIR` ``` E AttributeError: module 'plenum.config' has no attribute 'GENESIS_DIR' ``` Does this output show an installation problem? If this question is not suitable here, I would appreciate it if you could tell me where I should write. Thanks in adnvace.

shimizut (Thu, 18 Nov 2021 04:33:28 GMT):
Hello, I am setting up an environment for developing indy-node and meeting some problems: `pytest .` outputs errors. When testing indy-plenum, I got the following output. ``` = 7 failed, 11453 passed, 307 skipped, 71 warnings, 1089 error in 2380.80 seconds = ``` indy-node's test showed a similar result, which included fails and errors. I executed commands according to https://github.com/hyperledger/indy-node/blob/master/docs/source/setup-dev.md on ubuntu 16 https://releases.ubuntu.com/16.04/ , but the test `pytest . ` did not run. According to error messages, I additionally executed `pip install --upgrade 'setuptools<45.0.0`, `pip install pip==9.0.3` and so on. Then, `pytest .` ran but showed errors. For example I got the following. Actually, plenum/config.py does not include `GENESIS_DIR` ``` E AttributeError: module 'plenum.config' has no attribute 'GENESIS_DIR' ``` Does this output show an installation problem? If this question is not suitable here, I would appreciate it if you could tell me where I should write. Thanks in adnvace.

shimizut (Thu, 18 Nov 2021 04:33:28 GMT):
Hello, I am setting up an environment for developing indy-node and meeting some problems: `pytest .` outputs errors. When testing indy-plenum, I got the following output. ``` = 7 failed, 11453 passed, 307 skipped, 71 warnings, 1089 error in 2380.80 seconds = ``` indy-node's test showed a similar result, which included fails and errors. I am using master branch in indy-node and indy-plenum. I executed commands according to https://github.com/hyperledger/indy-node/blob/master/docs/source/setup-dev.md on ubuntu 16 https://releases.ubuntu.com/16.04/ , but the test `pytest . ` did not run. According to error messages, I additionally executed `pip install --upgrade 'setuptools<45.0.0`, `pip install pip==9.0.3` and so on. Then, `pytest .` ran but showed errors. For example I got the following. Actually, plenum/config.py does not include `GENESIS_DIR` ``` E AttributeError: module 'plenum.config' has no attribute 'GENESIS_DIR' ``` Does this output show an installation problem? If this question is not suitable here, I would appreciate it if you could tell me where I should write. Thanks in adnvace.

gaberasturi (Thu, 18 Nov 2021 08:49:12 GMT):
Thank you very much @swcurran for the response, you give us some points that we don't have``` > This is a static file, so in theory, they only need to grab it once and hold on to it``` ``` ```

gaberasturi (Thu, 18 Nov 2021 08:58:24 GMT):
Thank you very much @swcurran for the response, you give us some points that we hadn't > This is a static file, so in theory, they only need to grab it once and `hold on to it` but our doubt is more related about if the issuer need to act like a proxy to reach the tail server. We expected tails location file is provided by tails server URL not through Issuer enpoint URI.

deoalade (Thu, 18 Nov 2021 09:41:20 GMT):
Hi all, I would like to understand why we include a timestamp in the requested predicates object when sending presentation, given that we would have checked the non-revoked status already if requested in proof request? ``` { "requested_attributes": { "additionalProp1": { "cred_id": "{{Referent}}", "revealed": true } }, "requested_predicates": { "additionalProp2": { "cred_id": "{{Referent}}", "timestamp": 1636649578 } }, "self_attested_attributes": {}, "trace": false } ```

deoalade (Thu, 18 Nov 2021 09:41:20 GMT):
Hi all, I had previously asked this question in the aries channel but was advised to ask this here instead: I would like to understand why we include a timestamp in the requested predicates object when sending presentation, given that we would have checked the non-revoked status already if requested in proof request? ``` { "requested_attributes": { "additionalProp1": { "cred_id": "{{Referent}}", "revealed": true } }, "requested_predicates": { "additionalProp2": { "cred_id": "{{Referent}}", "timestamp": 1636649578 } }, "self_attested_attributes": {}, "trace": false } ```

deoalade (Thu, 18 Nov 2021 09:43:20 GMT):
I would like to have a better understanding to help us understand how to properly implement it in our solution.

joymondal (Thu, 18 Nov 2021 10:34:15 GMT):
Has joined the channel.

SwapnaliDive (Thu, 18 Nov 2021 12:37:55 GMT):
Hello Team, My Indy Network is failing when I add new nodes into the network with TAA enabled.

SwapnaliDive (Thu, 18 Nov 2021 12:37:55 GMT):
Hello Team, Indy Network is failing when I add new nodes into the network with TAA enabled.

SwapnaliDive (Thu, 18 Nov 2021 12:37:55 GMT):
Hello Team, Indy Network is failing when I add new nodes into the network with TAA enabled. Shall we disable TAA services while adding new nodes to build Indy Network Are there any other precautions/considerations while building network with TAA enabled options.

mccown (Thu, 18 Nov 2021 12:48:52 GMT):
The Identity Implementer’s WG meeting for today (11/18) has been cancelled. Please join us in our next meeting on Dec 2nd (9am MT / 4pm UTC).

vasile135 (Thu, 18 Nov 2021 12:56:47 GMT):
Hello, We are developing a .NET web agent which has to support the revoke functionality. The ledger operation is rejected for some reason, when we try to send a Revocation Registry Entry to the network. We have performed the following steps: - Credential definition was successfully created, - The tails file was created; - The REVOC_REG_DEF transaction passes as expected. - The initial REVOC_REG_ENTRY is successfully sent. But when try to revoke a document, the REVOC_REG_ENTRY transaction is rejected. Does anybody have a clue what can be the problem? Thx in advance.

WadeBarnes (Thu, 18 Nov 2021 13:02:36 GMT):
What process are you using to build the network and add the nodes?

ArneGebert (Thu, 18 Nov 2021 14:31:05 GMT):
Has joined the channel.

swcurran (Thu, 18 Nov 2021 14:57:09 GMT):
Ah...the tails server URL is given to the issuer in ACA-Py so they know where to put the tails file. The holder is only told of the full URL of the tails file itself. Does that make sense? Note that the idea of a "tails server" is an add on to make publishing easier. In Indy, the only requirement is that the tails file be somewhere, and the credential given to the holder has a URL to the file.

gaberasturi (Thu, 18 Nov 2021 15:16:13 GMT):
Yes we registered/configured in the Issuer `RevocationServerUrl` pointing to the tails sever. f.e http://TAILS_SERVER/ so what we are expecting (maybe wrong) is that the URL to the tails file be http://TAILS_SERVER/tails/... instead of http://ISSUER_SERVER/tails/... and what we are getting registered in the ledger is Tails file location: http://ISSUER_SERVER/tails/...

swcurran (Thu, 18 Nov 2021 15:18:43 GMT):
Hmmm...sounds like something is wrong in the config. I'd say post an issue in either the ACA-Py or Tails Server repo with your configs.

swcurran (Thu, 18 Nov 2021 15:19:05 GMT):
Have you looked at the Alice-Faber demo in ACA-Py, with the revocation option set?

gaberasturi (Thu, 18 Nov 2021 15:26:21 GMT):
Hmmm maybe the problem is that we are NOT using ACA-Py but .NET? - VON-NETWORK Ledger - Indy Tails Server (with all tests passed) I going to review Alice-Faber demo in ACA-Py with the revocation option set.

gaberasturi (Thu, 18 Nov 2021 15:37:45 GMT):
Many thanks Stephen!

ayushraj8271 (Thu, 18 Nov 2021 15:52:47 GMT):
Has joined the channel.

hitesh.sharma.06 (Thu, 18 Nov 2021 17:03:45 GMT):
Has joined the channel.

AshokkumarMdx (Thu, 18 Nov 2021 17:46:21 GMT):
Has joined the channel.

Adyntec (Thu, 18 Nov 2021 20:09:37 GMT):
Has joined the channel.

conanoc (Fri, 19 Nov 2021 01:56:10 GMT):
What's the error message?

SwapnaliDive (Fri, 19 Nov 2021 04:37:17 GMT):
Step : sudo init_indy_node Node5 127.0.0.1 9709 127.0.0.1 9710 Step 2 : sudo start_indy_node Node5 127.0.0.1 9709 127.0.0.1 9710 Then I add node using indy-cli --config taaFile.json and finally add node using ledger node command

SwapnaliDive (Fri, 19 Nov 2021 04:37:17 GMT):
Error 1: Node4 discarding message {'op': 'REQNACK', 'reqId': 1, 'reason': 'client request invalid: InvalidClientMsgType()', 'identifier': None} because 'RequestNack' object has no attribute 'get' Error 2 : Node5 discarding message ({'instId': 0, 'op': 'COMMIT', 'ppSeqNo': 23, 'blsSigs': {'2': 'RAGPWRJtb9etdhpe5noTk9YfeVc7h5VpVyHK2jV5LWAoBU51vMBVdULRYpJHkJj5sGPr42yRtvYUcnZckxxyaPegetgDQCxu19bKetwnx12GbowAwgqUWhyyaqrb9WseMAWyq5timK1msSDf7KJxGnYC77ttmKYEgYbjjcn8C8Z8Bq'}, 'blsSig': ' ', 'viewNo': 1}, b'-2aEQy3=<9:V.^B.sB#>dXd^DGfVU69[2yMSLsc9') because InvalidClientMsgType()

vasile135 (Fri, 19 Nov 2021 06:32:26 GMT):
"Ledger operation rejected", no error code or something...

conanoc (Fri, 19 Nov 2021 07:41:12 GMT):
You may find more detailed error message from the log file of your indy node.

SanjayRS (Fri, 19 Nov 2021 07:44:05 GMT):
Has joined the channel.

maciej_ol (Fri, 19 Nov 2021 10:41:31 GMT):
Has joined the channel.

bondasoup (Fri, 19 Nov 2021 12:43:12 GMT):
Has joined the channel.

vasile135 (Fri, 19 Nov 2021 13:15:28 GMT):
Following logs are written: 2021-11-19 12:10:54,155|INFO|message_processor.py|Node1 discarding message ({'protocolVersion': 2, 'reqId': 1637323853819319500, 'operation': {'type': '114', 'value': {'accum': '21 114A325B8DA8B3C7425D2E1B730A8BDA83A95273ED0C866148D341795CACB0217 21 128571509B55166431CCBBBC604FA1F5CE55CDCF7011C7B98C76F2A8FC1B0B0F6 6 6A7B79AC4625BCFFEE663ECDBBF2BBA0AB4ADCAD35437ADEBF742849EA8276F2 4 115BA504CD7B6CCEA5962A70580F655E8DB2452722F80AAA12A39CF918CB87E6 6 83A9AEEC65615C13E1A15B44035B1AE16032E45EAF232A0A7D413BA949889423 4 23FE2907DB8859E2254021B88E1D48C11ED93E10DC9FACC29CAAE384B8EFF938', 'revoked': [2], 'prevAccum': '21 10B4A91201088C56E025F16401EA908E8489E770597A3721C75B987411CE889C2 21 1322389DE6A6F9EFEF156FBA279D9DBFC1C8C1D24179C133DDBA89349E936FFCF 6 6FC3A6835DDDA3EB11185EC0F8032B3B6B07D2BD3238CD1789FB139011061B69 4 441752358690DCAC585BBB3E5B73AC0517303F9E5640CC3D02621D095E4D197F 6 5BB5CFDFA04E14C29F80A64AF74039DD8DE1F069A3AD2C4C351DFB0B43FE3E23 4 1689BF36E70836E8FFF892A1C4C8577094BEB30C4393DA6DC1F5D3AF9E53B848'}, 'revocRegDefId': 'RPVLWJ1WUdTiFiA7p7FQsD:4:RPVLWJ1WUdTiFiA7p7FQsD:3:CL:22:default:CL_ACCUM:1-100', 'revocDefType': 'CL_ACCUM'}, 'identifier': '6uTFna1HETg7pnV1HzYmsK', 'signature': 'QM9fRVHuJaMipPjpPm5ZWJR5BtV9ZtWxcFsPjsRmts87gRRVivaKko2Teo2pAQPgavEVduK6TdBYHQc9GwGR3jE'}, b'k+Qf6ILLjtfamG886yCT7+qFkx9y8H5XsIgqSOv0Fmk=') because could not authenticate, verkey for 6uTFna1HETg7pnV1HzYmsK cannot be found

vasile135 (Fri, 19 Nov 2021 14:24:23 GMT):
Was found which was the problem. We were creating the credential definition with the endpoint.Did, but the library Revoke method uses IssuerDid. Thus DIDs were different. Thx for logs suggestion, did the thing.

WadeBarnes (Fri, 19 Nov 2021 17:26:40 GMT):
What errors and behavior are you seeing after adding the node?

WadeBarnes (Fri, 19 Nov 2021 17:27:05 GMT):
You shouldn't have to disable TAA services when adding new nodes.

WadeBarnes (Fri, 19 Nov 2021 17:27:48 GMT):
We don't when adding/removing nodes from the Sovrin networks.

Tim98390383 (Fri, 19 Nov 2021 20:53:22 GMT):
Has joined the channel.

uvdsl (Fri, 19 Nov 2021 22:24:46 GMT):
Has joined the channel.

Kanishk-Kumar24 (Sat, 20 Nov 2021 05:45:20 GMT):
Has joined the channel.

lonis (Sat, 20 Nov 2021 18:09:02 GMT):
Has joined the channel.

ShumanCapote (Mon, 22 Nov 2021 05:12:12 GMT):
Has joined the channel.

ThomasRitter (Mon, 22 Nov 2021 05:28:23 GMT):
Has joined the channel.

SwapnaliDive (Mon, 22 Nov 2021 07:45:58 GMT):
Node4 discarding message {'op': 'REQNACK', 'reqId': 1, 'reason': 'client request invalid: InvalidClientMsgType()', 'identifier': None} because 'RequestNack' object has no attribute 'get'

SwapnaliDive (Mon, 22 Nov 2021 07:47:27 GMT):
Error 1 : Node4 discarding message {'op': 'REQNACK', 'reqId': 1, 'reason': 'client request invalid: InvalidClientMsgType()', 'identifier': None} because 'RequestNack' object has no attribute 'get' Error 2: Node5 discarding message ({'params': {'ledgerId': 0}, 'msg_type': 'LEDGER_STATUS', 'op': 'MESSAGE_RESPONSE', 'msg': {'ledgerId': 0, 'txnSeqNo': 5, 'viewNo': None, 'ppSeqNo': None, 'merkleRoot': 'AbMKrRWQixGoETfw6AY6tY3ALLyh1Vemho4LLphbA2AZ', 'protocolVersion': 2}}, b'?3qEFmOgUWAq&v2R#kSX38B=av>B8fg=Q2h{mN(+') because InvalidClientMsgType()

saurabhSAG (Mon, 22 Nov 2021 08:31:14 GMT):
Has joined the channel.

Sharonmunyao (Mon, 22 Nov 2021 09:46:40 GMT):
Has joined the channel.

SwapnaliDive (Mon, 22 Nov 2021 10:43:48 GMT):
Then when I try to add node7th I am not able to write on ledger

SwapnaliDive (Mon, 22 Nov 2021 10:44:22 GMT):

Screenshot from 2021-11-22 16-13-04.png

WadeBarnes (Mon, 22 Nov 2021 13:10:32 GMT):
Are you also registering the new nodes with the existing nodes using the `ledger node` command?

SwapnaliDive (Mon, 22 Nov 2021 13:12:22 GMT):
Yes @WadeBarnes

WadeBarnes (Mon, 22 Nov 2021 13:14:18 GMT):
You can add the nodes successfully when the TAA is disabled?

SwapnaliDive (Mon, 22 Nov 2021 13:15:13 GMT):
Yes @WadeBarnes

WadeBarnes (Mon, 22 Nov 2021 13:19:15 GMT):
Are you just using this as a test network?

SwapnaliDive (Mon, 22 Nov 2021 13:19:36 GMT):
Yes

SwapnaliDive (Mon, 22 Nov 2021 13:21:14 GMT):
I also tried and setup genesis nodes again , but still facing the same isue

SwapnaliDive (Mon, 22 Nov 2021 13:21:14 GMT):
I also tried and setup genesis nodes again , but still facing the same issue

SwapnaliDive (Mon, 22 Nov 2021 13:21:14 GMT):
@WadeBarnes I am facing the below error Consensus for digest 2fce9447b50d2698396b13138226d23a9eeb24f4f424d9f06f022966f5f1db65 was not achieved within 91.0379502128344 seconds. Primary node is Node3. Received Pre-Prepare from UNKNOWN. Received 3 valid Prepares from Node2:0, Node4:0, Node1:0. Received 0 valid Commits from noone.

SwapnaliDive (Mon, 22 Nov 2021 13:29:44 GMT):
@WadeBarnes Is this happening because of AWS instances issue ?

WadeBarnes (Mon, 22 Nov 2021 13:39:16 GMT):
Unlikely.

WadeBarnes (Mon, 22 Nov 2021 13:39:28 GMT):
Are all the nodes running on the same machine?

SwapnaliDive (Mon, 22 Nov 2021 13:40:23 GMT):
No , each node on separate machine

SwapnaliDive (Mon, 22 Nov 2021 13:40:23 GMT):
No , each node on separate instance

WadeBarnes (Mon, 22 Nov 2021 13:41:33 GMT):
Are the public IPs getting changed on the machines due to restarts?

SwapnaliDive (Mon, 22 Nov 2021 13:41:56 GMT):
No

WadeBarnes (Mon, 22 Nov 2021 13:44:48 GMT):
Would you be able to provide a full log of node 5 from the point where you've added it to the network?

SwapnaliDive (Mon, 22 Nov 2021 14:31:28 GMT):

Node5.txt

SwapnaliDive (Mon, 22 Nov 2021 14:31:47 GMT):
@WadeBarnes You can refer the above file for Node5 logs

Joesv (Mon, 22 Nov 2021 16:24:44 GMT):
Has joined the channel.

krishnabayanna (Mon, 22 Nov 2021 17:14:52 GMT):
Has joined the channel.

TheoKneen (Mon, 22 Nov 2021 17:25:27 GMT):
Has joined the channel.

RobinBeltran (Mon, 22 Nov 2021 18:49:35 GMT):
Has joined the channel.

jfbourque (Mon, 22 Nov 2021 19:18:55 GMT):
Has joined the channel.

WadeBarnes (Mon, 22 Nov 2021 19:39:42 GMT):
For Node5, double check the mapping of the external to internal IP/Ports. Make sure they are not reversed. Also what are the external IPs you are using to register Node5 with the network? I've checked Nodes1 to 4 and their ports seem to be open.

LeonQiao (Tue, 23 Nov 2021 02:08:26 GMT):
Has joined the channel.

bjwswangmail (Tue, 23 Nov 2021 03:24:15 GMT):
Has joined the channel.

g.mishra-blockgemini (Tue, 23 Nov 2021 05:21:26 GMT):
Has joined the channel.

Prasad8 2 (Tue, 23 Nov 2021 08:58:22 GMT):
Has joined the channel.

Prasad8 2 (Tue, 23 Nov 2021 09:04:00 GMT):
Can someone point me to the SDK library that helps how to create Pairwise DID's . I am trying to build VC application where issuer and holder share messages securely.

GoreTushar (Tue, 23 Nov 2021 11:24:47 GMT):
Has joined the channel.

skenaja (Tue, 23 Nov 2021 11:56:45 GMT):
Has joined the channel.

WadeBarnes (Tue, 23 Nov 2021 14:12:11 GMT):
You should have a look at https://github.com/hyperledger/aries-cloudagent-python. It's provides a higher level starting point for building applications and abstracts many of the complexities of the `indy-sdk`.

jyoti.patel (Tue, 23 Nov 2021 15:31:25 GMT):
Has joined the channel.

anipaul2 (Tue, 23 Nov 2021 19:08:23 GMT):
Has joined the channel.

dantetti (Tue, 23 Nov 2021 23:55:00 GMT):
Has joined the channel.

Prasad8 2 (Wed, 24 Nov 2021 04:22:57 GMT):
I am not going to use Aries agent , however develop a custom solution that agent to agent communicates securely via pairwise DID. In that case, would the link still be suitable ?

Prasad8 2 (Wed, 24 Nov 2021 04:22:57 GMT):
Thanks . I am not going to use Aries agent , however develop a custom solution that agent to agent communicates securely via pairwise DID. In that case, would the link still be suitable ?

hacke (Wed, 24 Nov 2021 07:34:00 GMT):
Has joined the channel.

KurosakiIchigo (Wed, 24 Nov 2021 09:14:58 GMT):
Has joined the channel.

Kmm 4 (Wed, 24 Nov 2021 13:49:10 GMT):
Has joined the channel.

ShengChillie (Wed, 24 Nov 2021 14:05:45 GMT):
Has joined the channel.

john.h.o.mantaring (Wed, 24 Nov 2021 15:02:17 GMT):
Has joined the channel.

WadeBarnes (Wed, 24 Nov 2021 15:15:58 GMT):
Yes

shinnosuke.sonoda (Thu, 25 Nov 2021 01:12:32 GMT):
Has joined the channel.

OzodbekXomidbekov (Thu, 25 Nov 2021 01:33:37 GMT):
Hello, Can anyone help me to install the Hyperledger Indy ? I tried many times, but I could not. PLEASE !

RitikaKakkar (Thu, 25 Nov 2021 04:53:49 GMT):
Has joined the channel.

abhisheksavaliya (Thu, 25 Nov 2021 06:03:02 GMT):
Has joined the channel.

SwapnaliDive (Thu, 25 Nov 2021 08:28:22 GMT):
@WadeBarnes Consensus for digest 2fce9447b50d2698396b13138226d23a9eeb24f4f424d9f06f022966f5f1db65 was not achieved within 91.0379502128344 seconds. Primary node is Node3. Received Pre-Prepare from UNKNOWN. Received 3 valid Prepares from Node2:0, Node4:0, Node1:0. Received 0 valid Commits from noone.

Haniff (Thu, 25 Nov 2021 09:33:10 GMT):
Has joined the channel.

gurkhaman (Thu, 25 Nov 2021 11:02:07 GMT):
Has joined the channel.

neves 2 (Thu, 25 Nov 2021 11:19:29 GMT):
Has joined the channel.

deetho (Thu, 25 Nov 2021 11:49:03 GMT):
Has joined the channel.

SwapnaliDive (Thu, 25 Nov 2021 11:55:06 GMT):
Hello Team, Using the TRUSTEE's DID I am setting the TAA as follows using incy-cli : 1) ledger txn-acceptance-mechanisms aml={"Session Agreement":"some description"} version=1 send=true 2) ledger txn-author-agreement text="some description" version=1 send=true ratification-timestamp=1637837478 After enabling the TAA I am starting the indy-cli like - indy-cli --config taa-config.json taa-config.json file have the below contents: { "taaAcceptanceMechanism": "Session Agreement" } After setting up TAA I am trying to add the new nodes into the network using indy-cli "ledger node" command which is failing network. We are not able to write on ledger. Can someone please confirm the method of enabling the TAA on the ledger ?

atsveloso (Thu, 25 Nov 2021 12:08:49 GMT):
Has joined the channel.

zhengxilong (Thu, 25 Nov 2021 12:21:09 GMT):
Has joined the channel.

WadeBarnes (Thu, 25 Nov 2021 13:59:04 GMT):
What are the external IPs you are using to register Node5 with the network?

SwapnaliDive (Thu, 25 Nov 2021 14:00:08 GMT):
instance 5 : Public - 18.118.55.243 | Private IPv4 - 172.31.16.37

WadeBarnes (Thu, 25 Nov 2021 14:03:56 GMT):
Porst 9701 and 9702?

WadeBarnes (Thu, 25 Nov 2021 14:03:56 GMT):
Ports 9701 and 9702?

SwapnaliDive (Thu, 25 Nov 2021 14:04:51 GMT):
ports 9709 and 9710

WadeBarnes (Thu, 25 Nov 2021 14:05:29 GMT):
The ports are showing as closed. Is Node5 running?

SwapnaliDive (Thu, 25 Nov 2021 14:05:38 GMT):
no

SwapnaliDive (Thu, 25 Nov 2021 14:05:47 GMT):
node5 is closed now

SwapnaliDive (Thu, 25 Nov 2021 14:05:47 GMT):
node5 is stopped now

SwapnaliDive (Thu, 25 Nov 2021 14:06:10 GMT):
how to check ports are running or not ?

SwapnaliDive (Thu, 25 Nov 2021 14:06:10 GMT):
Ports are working - nc -vz 3.34.192.230 9711 nc -vz 3.34.192.230 9712 I am still facing the error - 2021-11-25 14:23:36,187|WARNING|replicas.py|Consensus for digest 1a22f53f6922a3eb2b92a9f215bd6149763091dff3dc9cf3f361da183e6d1658 was not achieved within 118.244 87143754959 seconds. Primary node is Node3. Received Pre-Prepare from UNKNOWN. Received 3 valid Prepares from Node4:0, Node2:0, Node1:0. Received 0 valid Commits from noone. Transaction contents: {'signature': 'n4purnZ1UBWnM9a8T5VvHyRKJiprQJ4YfS3GgevBpgmJodueBdrCUiSZpNJPoqF8Xy5TPGV7Sz3k89kduJb2LBw', 'taaAcceptance': {'me chanism': 'Session Agreement', 'taaDigest': 'b3c0f48041cbf7e97fa372cfb924f43b42d6d48f02352bc8e06ed2caa333767c', 'time': 1637798400}, 'identifier': 'Ve8d4CxRBiyTc noWrCJMPw', 'protocolVersion': 2, 'reqId': 1637850097918945151, 'operation': {'type': '101', 'data': {'name': 'adhar', 'attr_names': ['name', 'age'], 'version': '1.0'}}}. 2021-11-25 14:23:39,833|INFO|message_processor.py|Node3 discarding message {'identifier': None, 'op': 'REQNACK', 'reqId': 1, 'reason': 'client request invalid: I nvalidClientMsgType()'} because 'RequestNack' object has no attribute 'get'

WadeBarnes (Thu, 25 Nov 2021 14:07:18 GMT):
`nc -vz `

SwapnaliDive (Thu, 25 Nov 2021 14:27:45 GMT):
Ports are working for me now. 2021-11-25 14:23:20,478|WARNING|replicas.py|Consensus for digest 1a22f53f6922a3eb2b92a9f215bd6149763091dff3dc9cf3f361da183e6d1658 was not achieved within 102.54031935427338 seconds. Primary node is Node3. Received Pre-Prepare from UNKNOWN. Received 3 valid Prepares from Node4:0, Node1:0, Node2:0. Received 0 valid Commits from noone. Transaction contents: {'operation': {'data': {'name': 'adhar', 'attr_names': ['name', 'age'], 'version': '1.0'}, 'type': '101'}, 'identifier': 'Ve8d4CxRBiyTcnoWrCJMPw', 'reqId': 1637850097918945151, 'signature': 'n4purnZ1UBWnM9a8T5VvHyRKJiprQJ4YfS3GgevBpgmJodueBdrCUiSZpNJPoqF8Xy5TPGV7Sz3k89kduJb2LBw', 'taaAcceptance': {'mechanism': 'Session Agreement', 'taaDigest': 'b3c0f48041cbf7e97fa372cfb924f43b42d6d48f02352bc8e06ed2caa333767c', 'time': 1637798400}, 'protocolVersion': 2}.

SwapnaliDive (Thu, 25 Nov 2021 14:29:37 GMT):
Ports are working - nc -vz 3.34.192.230 9711 nc -vz 3.34.192.230 9712

SwapnaliDive (Thu, 25 Nov 2021 14:29:46 GMT):
I am still facing the error - 2021-11-25 14:23:36,187|WARNING|replicas.py|Consensus for digest 1a22f53f6922a3eb2b92a9f215bd6149763091dff3dc9cf3f361da183e6d1658 was not achieved within 118.244 87143754959 seconds. Primary node is Node3. Received Pre-Prepare from UNKNOWN. Received 3 valid Prepares from Node4:0, Node2:0, Node1:0. Received 0 valid Commits from noone. Transaction contents: {'signature': 'n4purnZ1UBWnM9a8T5VvHyRKJiprQJ4YfS3GgevBpgmJodueBdrCUiSZpNJPoqF8Xy5TPGV7Sz3k89kduJb2LBw', 'taaAcceptance': {'me chanism': 'Session Agreement', 'taaDigest': 'b3c0f48041cbf7e97fa372cfb924f43b42d6d48f02352bc8e06ed2caa333767c', 'time': 1637798400}, 'identifier': 'Ve8d4CxRBiyTc noWrCJMPw', 'protocolVersion': 2, 'reqId': 1637850097918945151, 'operation': {'type': '101', 'data': {'name': 'adhar', 'attr_names': ['name', 'age'], 'version': '1.0'}}}. 2021-11-25 14:23:39,833|INFO|message_processor.py|Node3 discarding message {'identifier': None, 'op': 'REQNACK', 'reqId': 1, 'reason': 'client request invalid: I nvalidClientMsgType()'} because 'RequestNack' object has no attribute 'get'

WadeBarnes (Thu, 25 Nov 2021 14:32:19 GMT):
Is that log from Node%?

WadeBarnes (Thu, 25 Nov 2021 14:32:19 GMT):
Is that log from Node5?

SwapnaliDive (Thu, 25 Nov 2021 14:37:07 GMT):
These are from Node6

ap13 (Thu, 25 Nov 2021 15:28:50 GMT):
Has joined the channel.

krishna2752 (Thu, 25 Nov 2021 15:40:15 GMT):
Has joined the channel.

WadeBarnes (Thu, 25 Nov 2021 17:12:14 GMT):
You should focus on getting a single node connected successfully at a time. Otherwise you'll run into consensus issues, which you've already seend, and then you won't be able to do anything.

WadeBarnes (Thu, 25 Nov 2021 17:12:14 GMT):
You should focus on getting a single node connected successfully at a time. Otherwise you'll run into consensus issues, which you've already seen, and then you won't be able to do anything.

SwapnaliDive (Thu, 25 Nov 2021 17:13:06 GMT):
Yes @WadeBarnes

SwapnaliDive (Thu, 25 Nov 2021 17:13:06 GMT):
Yes @WadeBarnes Thank you.

SwapnaliDive (Thu, 25 Nov 2021 17:41:08 GMT):
@WadeBarnes The network issue got solved. Thank you for all your support

WadeBarnes (Thu, 25 Nov 2021 17:57:31 GMT):
:thumbsup:

SiddharthGupta (Fri, 26 Nov 2021 01:13:46 GMT):
Has joined the channel.

Prasad8 2 (Fri, 26 Nov 2021 07:57:16 GMT):
Hi All, Can you please help me with this query . I understand, the Indy wallet implementation has an embedded database for both mobile and non-mobile sdk. How to find the underlying default database configuration and schema tables .

csum112 (Fri, 26 Nov 2021 08:20:05 GMT):
Has joined the channel.

Kayaba.Keita (Fri, 26 Nov 2021 10:38:47 GMT):
Has joined the channel.

vasile135 (Fri, 26 Nov 2021 10:38:52 GMT):
Hello, everyone! When try to create a proof (AnonCreds.ProverCreateProofAsync) providing a mix of revocable and non-revocable credentials, it returns an error saying "A value being processed is not valid". Is this case supported by Indy? Thank you!

swcurran (Fri, 26 Nov 2021 15:09:02 GMT):
It certainly should be -- I've not heard anyone saying this had been a problem, and we did a lot of testing in implementing revocation in ACA-Py. Likely sharing the details is needed -- try to simplify your PR as much as possible.

vasile135 (Fri, 26 Nov 2021 15:24:22 GMT):
Thank you for response. I followed your advice from here: https://github.com/hyperledger/aries-framework-dotnet/issues/184#issuecomment-816187518. If I drop the revocation interval, I'm able to create a proof. But, in this case, when I call AnonCreds.VerifierVerifyProofAsync as verifier, an error is thrown, probably because verifier's proof request still has RevocationInterval. hmmm...

vasile135 (Fri, 26 Nov 2021 15:33:56 GMT):
My PR: ```{ "name":"Test", "version":"333789", "nonce":"710254563841187486574541", "requested_attributes":{ "Email":{ "name":"Email", "restrictions":[ ] }, "FirstName":{ "name":"FirstName", "restrictions":[ ] } }, "requested_predicates":{ }, "non_revoked":{ "from":1637929619, "to":1637929619 } } ``` I provide Email from a non-revocable credential and FirstName - from a revokable one.

vasile135 (Fri, 26 Nov 2021 15:33:56 GMT):
My PR: ```{ "name":"Test", "version":"333789", "nonce":"710254563841187486574541", "requested_attributes":{ "Email":{ "name":"Email", "restrictions":[ ] }, "FirstName":{ "name":"FirstName", "restrictions":[ ] } }, "requested_predicates":{ }, "non_revoked":{ "from":1637929619, "to":1637929619 } } ``` I provide Email from a non-revocable credential and FirstName - from a revocable one.

swcurran (Fri, 26 Nov 2021 16:41:28 GMT):
The key question is where the issue is coming from -- is it happening in Aries Framework .NET, or is it in Indy? It would be good to try this in ACA-Py to see what it does with this scenario. @esune @ianco -- have either of you ever tried this use case -- a PR asking for revocation, with two credentials being used in the Proof, one revocable, one not.

vasile135 (Fri, 26 Nov 2021 16:42:58 GMT):
Thank you, @swcurran

ianco (Fri, 26 Nov 2021 17:10:44 GMT):
I see that you're not providing any restrictions in your proof request. In this case the attributes are "self attested" and I'm not sure how revocation is handled in this scenario (I would expect an error, because self attested attributes can't be revoked)

ianco (Fri, 26 Nov 2021 17:11:20 GMT):
Try using the schema id, schema name or cred def is as a restriction and see how the proof is handled

vasile135 (Fri, 26 Nov 2021 17:48:20 GMT):
Thx, @ianco , it worked

ankitsaurabh (Fri, 26 Nov 2021 21:47:21 GMT):
Has joined the channel.

agrawalaman (Sat, 27 Nov 2021 08:01:14 GMT):
Hi, can someone tell me how do we ensure that the credential we issue cannot be tampered by the holder?

swcurran (Sat, 27 Nov 2021 21:22:29 GMT):
All of the attributes in the messages are signed by the issuer and given to the holder. If the holder tampers with any of the messages, the signatures will not verify. Note that in AnonCreds, there is the raw data (e.g. strings, dates, numbers) and there is the encoded value (always integers). It is the encoded value that is signed. So the verifier must also verify that the raw data can be encoded in the way the issuer had done so before signing the encoded values. The Indy verifier library does that.

Oznfc (Sun, 28 Nov 2021 12:16:41 GMT):
can anyone help me about running identy server , indy node on react example. ı runned indy node and verified my credential. added my mobile wallet and ı can login with oidc demo examples(https://github.com/bcgov/vc-authn-oidc/tree/master/demo). I want to connect to the itentiy server inside the oidc example from my react app.

iamchris1688 (Mon, 29 Nov 2021 03:43:53 GMT):
Has joined the channel.

SubashAnbazhagan (Mon, 29 Nov 2021 04:49:37 GMT):
Has joined the channel.

laolao (Mon, 29 Nov 2021 05:27:07 GMT):
Has joined the channel.

amithkk (Mon, 29 Nov 2021 06:05:41 GMT):
Has joined the channel.

s.vahidi (Mon, 29 Nov 2021 12:14:35 GMT):
Has joined the channel.

FotisMichalopoulos (Mon, 29 Nov 2021 13:51:49 GMT):
Has joined the channel.

MohammadHT (Mon, 29 Nov 2021 14:27:03 GMT):
Has joined the channel.

WadeBarnes (Mon, 29 Nov 2021 15:29:26 GMT):
@esune ^

esune (Mon, 29 Nov 2021 17:08:11 GMT):
`vc-authn` is a standard OIDC identity provider. How you configure your app will depend on the plugin you are using as OIDC client, this is an example of what `oidc-js` would take as input configuration: https://github.com/bcgov/essential-services-delivery/blob/main/openshift/templates/issuer-web/config/esr2/dev/config.json#L15-L34

Oznfc (Tue, 30 Nov 2021 08:01:38 GMT):
ı tried but not worked. I proof my cridential successfully. I'm gettinh this error :

Oznfc (Tue, 30 Nov 2021 08:01:46 GMT):
```UserManager.getUser: user not found in storage```

Oznfc (Tue, 30 Nov 2021 09:15:50 GMT):
I'm triying in my local network.

dalan 2 (Tue, 30 Nov 2021 10:07:28 GMT):
Has joined the channel.

RouteSagar (Tue, 30 Nov 2021 12:08:10 GMT):
Has joined the channel.

Oznfc (Tue, 30 Nov 2021 13:30:10 GMT):
ı solved somethings. My last error : ```Client not authorized for code flow`` ```

Oznfc (Tue, 30 Nov 2021 13:30:10 GMT):
ı solved somethings. My last error : ```Client not authorized for code flow ```

caglayan45 (Tue, 30 Nov 2021 20:00:07 GMT):
Has joined the channel.

JunkF (Wed, 01 Dec 2021 02:58:09 GMT):
Has joined the channel.

Oznfc (Wed, 01 Dec 2021 08:45:19 GMT):

Clipboard - 1 Aralık 2021 11:45

Oznfc (Wed, 01 Dec 2021 08:45:23 GMT):
hi how can ı add icon to credentials for style ? is it possible.

Oznfc (Wed, 01 Dec 2021 08:45:23 GMT):
hi how can ı add icon(style) to credentials like the id below? ? is it possible.

lventi (Wed, 01 Dec 2021 12:41:32 GMT):
Has joined the channel.

joey_gogo (Wed, 01 Dec 2021 13:50:13 GMT):
Has joined the channel.

BigmoneyJoe (Wed, 01 Dec 2021 15:52:38 GMT):
Has joined the channel.

JosephPotts (Wed, 01 Dec 2021 19:36:12 GMT):
Has joined the channel.

thusithathilina (Wed, 01 Dec 2021 23:27:49 GMT):
Has joined the channel.

swcurran (Thu, 02 Dec 2021 00:26:33 GMT):
Folks, on the Indy Contributors call this coming Tuesday, Dec. 7 (8AM Pacific, 5PM CET) we're going to be talking about the "did:indy" DID Method Specification and how to get it implemented. We have several organizations at least that are planning to start on that Real Soon Now and we're looking to see who else is interested in tracking or participating in this effort. Please join us! Meeting Details and Zoom Link: https://wiki.hyperledger.org/display/indy/Indy+Contributors+Meeting The DID Indy Specification: https://hyperledger.github.io/indy-did-method/

Rp27 (Thu, 02 Dec 2021 03:52:55 GMT):
Has joined the channel.

alnazif (Thu, 02 Dec 2021 04:27:16 GMT):
Has joined the channel.

mccown (Thu, 02 Dec 2021 05:36:03 GMT):
The Identity Implementer’s WG meeting for today (12/2) has been cancelled. Please join us in our next meeting on Dec 16th (9am MT / 4pm UTC).

Bhemeswararaoxcubelabs.com (Thu, 02 Dec 2021 06:27:15 GMT):
Has joined the channel.

BoJacobs (Thu, 02 Dec 2021 13:29:52 GMT):
Has joined the channel.

kudzaishezana (Thu, 02 Dec 2021 15:45:47 GMT):
Has joined the channel.

The_Science (Thu, 02 Dec 2021 18:28:15 GMT):
Has joined the channel.

F.Florian (Thu, 02 Dec 2021 23:11:39 GMT):
Has joined the channel.

PalanisamyChellappan (Fri, 03 Dec 2021 05:17:09 GMT):
Has joined the channel.

ipepoi (Fri, 03 Dec 2021 07:07:26 GMT):
Has joined the channel.

rohitsaikrishnan (Fri, 03 Dec 2021 08:22:32 GMT):
Has joined the channel.

rohitsaikrishnan (Fri, 03 Dec 2021 08:22:39 GMT):
Hi

rohitsaikrishnan (Fri, 03 Dec 2021 08:22:55 GMT):
I am trying to work with Sovrin's buildernet

rohitsaikrishnan (Fri, 03 Dec 2021 08:23:04 GMT):
I dont quite understand how it works

rohitsaikrishnan (Fri, 03 Dec 2021 08:23:32 GMT):
I have successfully created a wallet and generated DID and Verkey

rohitsaikrishnan (Fri, 03 Dec 2021 08:23:47 GMT):
and clicket submit and got endorser token

rohitsaikrishnan (Fri, 03 Dec 2021 08:24:05 GMT):
not really sure what this means and how does this provide SSI

rohitsaikrishnan (Fri, 03 Dec 2021 08:24:27 GMT):
further how do I sign any documents for verification across 2 modes..

rohitsaikrishnan (Fri, 03 Dec 2021 08:24:34 GMT):
please help. Thanks.

NB-PrasadKatkar (Fri, 03 Dec 2021 08:57:12 GMT):
Has joined the channel.

cmalbas (Fri, 03 Dec 2021 14:06:03 GMT):
Has joined the channel.

KodeKamper (Fri, 03 Dec 2021 14:15:39 GMT):
Has joined the channel.

x4v13r (Fri, 03 Dec 2021 15:44:30 GMT):
Has joined the channel.

SudeshBanskota (Fri, 03 Dec 2021 15:45:04 GMT):
Has joined the channel.

swcurran (Fri, 03 Dec 2021 17:09:08 GMT):
I suspect you need to understand more about Verifiable Credentials and then how VCs interact with ledgers for SSI. Perhaps take a look at this edX course from the Linux Foundation: https://www.edx.org/course/introduction-to-hyperledger-sovereign-identity-blockchain-solutions-indy-aries-and-ursa

lazaridis.com (Sat, 04 Dec 2021 07:08:56 GMT):
Has joined the channel.

Oznfc (Sun, 05 Dec 2021 08:50:19 GMT):

Clipboard - 5 Aralık 2021 11:47

saibot940 (Sun, 05 Dec 2021 14:38:29 GMT):
Has joined the channel.

bhupatbheda (Sun, 05 Dec 2021 17:52:18 GMT):
Has joined the channel.

babamatmat (Mon, 06 Dec 2021 04:21:50 GMT):
Has joined the channel.

tgalal (Mon, 06 Dec 2021 09:14:23 GMT):
Hey all. I was trying to find out what cryptographic construction is used for forming the `aggregated_proof.c_hash` and `aggregated_proof.c_list`? And do I get it right that the purpose of aggregated proof is to ensure that the proofs being verified were in fact generated together?

Shweta1 (Mon, 06 Dec 2021 12:52:13 GMT):

Clipboard - December 6, 2021 6:23 PM

Shweta1 (Mon, 06 Dec 2021 12:52:57 GMT):
Hi Team, I have added new validator node into network.i can see node into Pool but can't see as a validator node.Query from cli getting attached error.

Shweta1 (Mon, 06 Dec 2021 12:57:16 GMT):

Clipboard - December 6, 2021 6:28 PM

Shweta1 (Mon, 06 Dec 2021 12:57:16 GMT):
I am able to add new validator node into network successfully and i can see it in pool but when query from cli "ledger get-validator nodes=Node5" getting attached error.
Clipboard - December 6, 2021 6:28 PM

JOHNNYDOE (Mon, 06 Dec 2021 13:49:27 GMT):
Has joined the channel.

WadeBarnes (Mon, 06 Dec 2021 14:43:46 GMT):
This would indicate Node5 does not recognize the DID you are using for the query. It could be that it has not completed synchronizing with the other nodes, therefore it does not have record of that DID yet.

Shweta1 (Mon, 06 Dec 2021 14:55:58 GMT):
what is the issue when node is not synchronizing?I can see one info new node log "Node5:0 primary has been disconnected for too long"

Shweta1 (Mon, 06 Dec 2021 14:55:58 GMT):
what is the issue when node is not synchronizing?I can see one info in new node log "Node5:0 primary has been disconnected for too long"

WadeBarnes (Mon, 06 Dec 2021 14:58:24 GMT):
Typically it's because the new node can't communicate with the other nodes. Either because of connectivity issues between the nodes, or the node has not been registered with the network and therefore is not recognized as a valid node.

Shweta1 (Mon, 06 Dec 2021 15:52:36 GMT):

Clipboard - December 6, 2021 9:24 PM

Shweta1 (Mon, 06 Dec 2021 15:52:40 GMT):
all node are reachable and new node is registered into network too.but catchupService is not working properly for new node.Why this is not working .Attached info

WadeBarnes (Mon, 06 Dec 2021 15:53:48 GMT):
How many transactions are on the ledger and how long has Node5 been running?

WadeBarnes (Mon, 06 Dec 2021 15:56:48 GMT):
What is the output of `sudo validator-info` when run on Node5? What is the output when run on another node?

binodpandey (Mon, 06 Dec 2021 16:25:52 GMT):
Has joined the channel.

gustavosales (Mon, 06 Dec 2021 17:45:42 GMT):
Has joined the channel.

swcurran (Tue, 07 Dec 2021 01:05:00 GMT):
Reminder that 15 hours from now the Indy Contributors call, Tuesday, Dec. 7 (8AM Pacific, 5PM CET) where we're going to be talking about the "did:indy" DID Method Specification and how to get it implemented. We have several organizations at least that are planning to start on that Real Soon Now and we're looking to see who else is interested in tracking or participating in this effort. Please join us! Meeting Details and Zoom Link: https://wiki.hyperledger.org/display/indy/Indy+Contributors+Meeting The DID Indy Specification: https://hyperledger.github.io/indy-did-method/

tuan072090 (Tue, 07 Dec 2021 02:56:51 GMT):
Has joined the channel.

ksinkar (Tue, 07 Dec 2021 03:59:57 GMT):
Has joined the channel.

YeMyint (Tue, 07 Dec 2021 05:22:48 GMT):
Has joined the channel.

jainamsha (Tue, 07 Dec 2021 07:31:49 GMT):
Has joined the channel.

jandaman (Tue, 07 Dec 2021 08:57:31 GMT):
Has joined the channel.

TejaSatyanarayana (Tue, 07 Dec 2021 09:31:34 GMT):
Has joined the channel.

ashritha2410 (Tue, 07 Dec 2021 10:06:29 GMT):
Has joined the channel.

Ivanaa9 (Tue, 07 Dec 2021 10:13:47 GMT):
Has joined the channel.

urahara (Tue, 07 Dec 2021 10:17:32 GMT):
Has joined the channel.

carlosalca94 (Tue, 07 Dec 2021 14:14:31 GMT):
Has joined the channel.

PalanisamyChellappan (Tue, 07 Dec 2021 16:09:04 GMT):
Hello Everyone, I am new to this channel and Can anyone help me to install the Hyperledger Indy ? I tried many times, but I could not. PLEASE !

vasiksim (Tue, 07 Dec 2021 20:06:35 GMT):
Has joined the channel.

WadeBarnes (Tue, 07 Dec 2021 20:28:43 GMT):
Which of the Hyperledger Indy components are you trying to install/use, specifically. Indy SDK, Indy Node?

WadeBarnes (Tue, 07 Dec 2021 20:28:57 GMT):
What's your goal?

kennylau89 (Wed, 08 Dec 2021 01:27:05 GMT):
Has joined the channel.

PalanisamyChellappan (Wed, 08 Dec 2021 03:12:48 GMT):
Hi Wade, i am trying to setup the Hyperledger Indy for one of our POC. I am following the below documentation which eventually ended up with python related issues. The documentation which i am following are,

PalanisamyChellappan (Wed, 08 Dec 2021 03:12:48 GMT):
Hi Wade, i am trying to setup the Hyperledger Indy for one of our POC. I am following the below documentation which eventually ended up with python related issues. The documentation which i am following are, https://hyperledger-indy.readthedocs.io/projects/node/en/latest/setup-dev.html https://github.com/hyperledger/indy-node

PalanisamyChellappan (Wed, 08 Dec 2021 03:14:00 GMT):
Would be great if you could throw some light on this.

PalanisamyChellappan (Wed, 08 Dec 2021 03:16:12 GMT):
Hi Wade, i am trying to setup the Hyperledger Indy for one of our POC. I am following the below documentation which eventually ended up with python related issues. The documentation which i am following are, https://hyperledger-indy.readthedocs.io/projects/node/en/latest/setup-dev.html https://github.com/hyperledger/indy-node Would be great if you could throw some light on this.

AJAR (Wed, 08 Dec 2021 05:31:44 GMT):
Has joined the channel.

bajerm (Wed, 08 Dec 2021 08:40:13 GMT):
Has joined the channel.

StrangeDays (Wed, 08 Dec 2021 09:19:24 GMT):
Has joined the channel.

StrangeDays (Wed, 08 Dec 2021 09:19:45 GMT):
Hi2all

SujithSu 2 (Wed, 08 Dec 2021 12:18:46 GMT):
Has joined the channel.

MateC89 (Wed, 08 Dec 2021 13:30:45 GMT):
Has joined the channel.

WadeBarnes (Wed, 08 Dec 2021 14:13:06 GMT):
If you are just needing a ledger for development purposes or for a POC you could use; [von-network](https://github.com/bcgov/von-network).

WadeBarnes (Wed, 08 Dec 2021 14:13:34 GMT):
Another option is; [indy-node-container](https://github.com/IDunion/indy-node-container).

WadeBarnes (Wed, 08 Dec 2021 14:15:12 GMT):
You could also use the Sovrin Networks, starting with BuilderNet, which provide a path to a production hosted environment; https://sovrin.org/overview/

WadeBarnes (Wed, 08 Dec 2021 14:18:18 GMT):
If you still want to setup your own network there are some guides here; https://github.com/hyperledger/indy-node/issues/1674 Which are the basis for the updated documents in this PR; https://github.com/hyperledger/indy-node/pull/1689

PalanisamyChellappan (Wed, 08 Dec 2021 16:58:36 GMT):
i have tried von network. that also throws python library mismatch errors. i will try the another options and wll update to you.

WadeBarnes (Wed, 08 Dec 2021 17:00:07 GMT):
You shouldn't get python library mismatch errors with von-network. You run it by issuing two commands; ./manage build`, `./manage start`.

WadeBarnes (Wed, 08 Dec 2021 17:00:07 GMT):
You shouldn't get python library mismatch errors with von-network. You run it by issuing two commands; `./manage build`, `./manage start`.

WadeBarnes (Wed, 08 Dec 2021 17:01:01 GMT):
That is spins up the nodes and the ledger browser in containers.

PalanisamyChellappan (Wed, 08 Dec 2021 17:01:47 GMT):
./manage build throws the error.. will back with you the specific error details

WadeBarnes (Wed, 08 Dec 2021 17:02:37 GMT):
Are you using the latest version from the repo?

PalanisamyChellappan (Wed, 08 Dec 2021 17:02:51 GMT):
i am using master

PalanisamyChellappan (Wed, 08 Dec 2021 17:03:22 GMT):
i have tried with Ubuntu 18.04 and 16.04

WadeBarnes (Wed, 08 Dec 2021 17:07:51 GMT):
I just ran a `./manage rebuild` and it built without error.

PalanisamyChellappan (Wed, 08 Dec 2021 17:08:07 GMT):
which branch you had tried

WadeBarnes (Wed, 08 Dec 2021 17:08:33 GMT):
`main` , I renamed `master` this morning.

PalanisamyChellappan (Wed, 08 Dec 2021 17:08:58 GMT):
alright.. i will clone that now

PalanisamyChellappan (Wed, 08 Dec 2021 17:09:16 GMT):
i am using Ubuntu 18.04.6 LTS

WadeBarnes (Wed, 08 Dec 2021 17:09:53 GMT):
I built and ran von-network on Ubuntu 18.04 LTS yesterday.

PalanisamyChellappan (Wed, 08 Dec 2021 17:10:42 GMT):
ok.. python version is Python 3.6.13

PalanisamyChellappan (Wed, 08 Dec 2021 17:11:11 GMT):
following the below documentation

PalanisamyChellappan (Wed, 08 Dec 2021 17:11:13 GMT):
https://github.com/bcgov/von-network/blob/main/docs/UsingVONNetwork.md

WadeBarnes (Wed, 08 Dec 2021 17:12:27 GMT):
BTW if you want to host `von-network` on a machine and have other machines connect to the ledger you would run `./manage start WEB_SERVER_HOST_PORT=80`

WadeBarnes (Wed, 08 Dec 2021 17:13:21 GMT):
The version of Python on the host system should not really come into play since everything is running in containers.

PalanisamyChellappan (Wed, 08 Dec 2021 17:13:31 GMT):
ok

PalanisamyChellappan (Wed, 08 Dec 2021 17:14:09 GMT):
i cloned master

PalanisamyChellappan (Wed, 08 Dec 2021 17:14:19 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=AWrTQ59C7khEw8rTa) sure

mdshahbazalam (Wed, 08 Dec 2021 17:33:10 GMT):
Has joined the channel.

PalanisamyChellappan (Wed, 08 Dec 2021 17:38:30 GMT):
I am deploying the von-network in an single AWS EC2 instance

PalanisamyChellappan (Wed, 08 Dec 2021 17:38:58 GMT):
i have followed the below documentation and its built 4 nodes

PalanisamyChellappan (Wed, 08 Dec 2021 17:39:00 GMT):
https://github.com/bcgov/von-network/blob/main/docs/UsingVONNetwork.md

PalanisamyChellappan (Wed, 08 Dec 2021 17:41:18 GMT):
i got it.. thank you for guiding me

PalanisamyChellappan (Wed, 08 Dec 2021 17:41:23 GMT):

Clipboard - December 8, 2021 11:11 PM

WadeBarnes (Wed, 08 Dec 2021 17:42:03 GMT):
:thumbsup:

george.patterson (Thu, 09 Dec 2021 01:37:44 GMT):
Has joined the channel.

Jerry.Bach (Thu, 09 Dec 2021 03:22:18 GMT):
Has joined the channel.

JamesYL (Thu, 09 Dec 2021 05:38:10 GMT):
Has joined the channel.

nathanbn (Thu, 09 Dec 2021 11:48:02 GMT):
Has joined the channel.

Oznfc (Thu, 09 Dec 2021 12:07:13 GMT):
hi everyone , does anyone using keycloak ? ı want to login by identyserver without username and password.is it possible ? ı'm using von-oidc exxample.

WadeBarnes (Thu, 09 Dec 2021 12:21:56 GMT):
This is the example you are using? https://github.com/bcgov/vc-authn-oidc

WadeBarnes (Thu, 09 Dec 2021 12:22:19 GMT):
@esune^

Oznfc (Thu, 09 Dec 2021 12:25:06 GMT):
yess. ı asked under the issue(https://github.com/bcgov/vc-authn-oidc/issues/143) .

WadeBarnes (Thu, 09 Dec 2021 12:28:39 GMT):
Looks like @esune is already helping you.

Oznfc (Thu, 09 Dec 2021 13:12:56 GMT):
yes ı asked here maybe somebody has same problem last time so they can help me. But ı solved now.

Oznfc (Thu, 09 Dec 2021 13:13:37 GMT):
ı added new flow with my settings.

Oznfc (Thu, 09 Dec 2021 13:14:00 GMT):

Clipboard - 9 Aralık 2021 16:13

Oznfc (Thu, 09 Dec 2021 13:37:20 GMT):
now ı can login with qr code. I updated keycloak version to 15.0.2 . Than get invalid scope error. I chanced scope. I wrote email instead of vc_authn

murilo_silva (Thu, 09 Dec 2021 15:07:10 GMT):
Has joined the channel.

KMM 6 (Thu, 09 Dec 2021 15:13:09 GMT):
Has joined the channel.

diane.lockhart2 (Thu, 09 Dec 2021 16:26:07 GMT):
Has joined the channel.

nikusingh (Thu, 09 Dec 2021 18:01:55 GMT):
Has joined the channel.

Princekumar4812 (Thu, 09 Dec 2021 18:16:15 GMT):
Hello everyone, I want to create an Indy Test Ledger from scratch. Can anyone please suggest a pathway that I can follow?

WadeBarnes (Thu, 09 Dec 2021 19:07:03 GMT):
I'd recommend using: - https://github.com/bcgov/von-network or - https://github.com/IDunion/indy-node-container but if you're intent on building it from scratch there are these resources: - https://github.com/hyperledger/indy-node/issues/1674 which is the basis of - https://github.com/hyperledger/indy-node/pull/1689

mahdiabdian (Thu, 09 Dec 2021 20:40:22 GMT):
Has joined the channel.

BoranCar (Thu, 09 Dec 2021 21:45:50 GMT):
Has joined the channel.

mvasavi (Fri, 10 Dec 2021 03:07:24 GMT):
Has joined the channel.

Aqusa (Fri, 10 Dec 2021 03:42:38 GMT):
Has joined the channel.

Princekumar4812 (Fri, 10 Dec 2021 07:56:52 GMT):
Thank-you very much

rmdnsd (Fri, 10 Dec 2021 11:29:45 GMT):
Has joined the channel.

nebucaz (Fri, 10 Dec 2021 12:24:43 GMT):
Has joined the channel.

wlhowell (Fri, 10 Dec 2021 14:40:39 GMT):
Has joined the channel.

tabouzid (Fri, 10 Dec 2021 15:19:06 GMT):
Has joined the channel.

AutoMAttic (Fri, 10 Dec 2021 19:10:31 GMT):
Has joined the channel.

VarunLashkari (Fri, 10 Dec 2021 20:13:44 GMT):
Has joined the channel.

VarunLashkari (Fri, 10 Dec 2021 20:14:10 GMT):
which are the best hyperledger projects?

VarunLashkari (Fri, 10 Dec 2021 20:14:31 GMT):
that are still not archived

ianco (Fri, 10 Dec 2021 20:58:27 GMT):
All the Aries projects are the best

HyperledgerNewbie (Sat, 11 Dec 2021 23:52:04 GMT):
Has joined the channel.

brnhrd89 (Sun, 12 Dec 2021 03:18:25 GMT):
Has joined the channel.

shanest (Sun, 12 Dec 2021 11:51:25 GMT):
Hi, schema id is generally written as {issuer_did}:{schema_name}:{schema_version} like 2uiaGAtZFzCL15n7CcxEYY:2:my-license:1.0. What is ":2" after did referred to?

sairanjitaw (Sun, 12 Dec 2021 13:47:07 GMT):
Has joined the channel.

Oznfc (Sun, 12 Dec 2021 14:29:32 GMT):
hi . again me. :-) Can we change user role from client app ?

ianco (Sun, 12 Dec 2021 17:38:10 GMT):
`2` is the transaction type (it means a schema transaction)

shanest (Sun, 12 Dec 2021 20:00:30 GMT):
Much appreciated!

PaulHaynes (Mon, 13 Dec 2021 12:55:48 GMT):
Has joined the channel.

arshdebian (Mon, 13 Dec 2021 14:44:02 GMT):
Has joined the channel.

lynn.bendixsen (Mon, 13 Dec 2021 22:30:23 GMT):
Yes, the Indy-cli can do it, so sdk calls exist that can be used to change roles. Make sure to pass in (use) a DID with the right permissions for the role change requested, and check required permissions by looking at auth_rules.

talgar (Mon, 13 Dec 2021 23:33:21 GMT):
Has joined the channel.

kbyokin (Tue, 14 Dec 2021 07:22:31 GMT):
Has joined the channel.

MehdiIda (Tue, 14 Dec 2021 08:12:14 GMT):
Has joined the channel.

Oznfc (Tue, 14 Dec 2021 12:25:57 GMT):
@esune can ı use diffetent critential for different page with keycloak. for example I added scopes. when ı call oidc with ```read scope``` and if user proof cridential , user can read page. when ı call ```write scope``` and if user proof cridential , user can write something to page . is it possible or does any example about this . already ı added scopes to keycloak. when ı call with scope ı can read userdata and scopes . when user click the write page , app must wants to second cridential with ```write scope```

Oznfc (Tue, 14 Dec 2021 12:25:57 GMT):
@esune can ı use diffetent critential for different page with keycloak. for example I added scopes. when ı call oidc with ```read scope``` and if user proof cridential , user can read page. when ı call ```write scope``` and if user proof cridential , user can write something to page . is it possible or does any example about this . already ı added scopes to keycloak. when ı call with scope ı can read userdata and scopes . when user click the write page , app must wants to second cridential with ```write scope```

Oznfc (Tue, 14 Dec 2021 12:25:57 GMT):
@esune can ı use diffetent critential for different page with keycloak. for example I added scopes. when ı call oidc with ```read scope``` and if user proof cridential , user can read page. when ı call ```write scope``` and if user proof cridential , user can write something to page . is it possible or does any example about this . already ı added scopes to keycloak. when ı call with scope ı can read userdata and scopes . when user click the write page , app must wants to second cridential with ```write scope```

Oznfc (Tue, 14 Dec 2021 12:25:57 GMT):
@esune can ı use diffetent critential for different page with keycloak. for example I added scopes. when ı call oidc with ```read scope``` and if user proof cridential , user can read page. when ı call ```write scope``` and if user proof cridential , user can write something to page . is it possible or does any example about this . already ı added scopes to keycloak. when ı call with scope ı can read userdata and scopes . when user click the write page , app must wants to second cridential with ```write scope```

Oznfc (Tue, 14 Dec 2021 12:25:57 GMT):
@esune can ı use diffetent critential for different page with keycloak. for example I added scopes. when ı call oidc with read scope and if user proof cridential , user can read page. when ı call ```write scope``` and if user proof cridential , user can write something to page . is it possible or does any example about this . already ı added scopes to keycloak. when ı call with scope ı can read userdata and scopes . when user click the write page , app must wants to second cridential with ```write scope```

fabric (Tue, 14 Dec 2021 12:40:25 GMT):
Has joined the channel.

whcjst (Tue, 14 Dec 2021 13:34:17 GMT):
Has joined the channel.

Oznfc (Tue, 14 Dec 2021 15:48:22 GMT):
Actually, what I want to do is: When I log in with different certificates, I want to control the user role.For example, let's have the user role when I log in with user only id. When I log in with a driver's license, I want to have a driver role.

TarsoQueiroz (Tue, 14 Dec 2021 16:52:33 GMT):
Has joined the channel.

esune (Tue, 14 Dec 2021 17:08:19 GMT):
How do you do that in the backend will vary, I think if you're using Keycloak you can define functions to assign roles based on claims received by the IdP, but I have not done that in the past.

esune (Tue, 14 Dec 2021 17:08:40 GMT):
You could also have the role listed in the credential and have Keycloak read it as part of a claim.

MattProto (Tue, 14 Dec 2021 22:34:58 GMT):
Has joined the channel.

AshwiniD 1 (Wed, 15 Dec 2021 05:14:59 GMT):
Has joined the channel.

Oznfc (Wed, 15 Dec 2021 07:38:01 GMT):
``` You could also have the role listed in the credential and have Keycloak read it as part of a claim. ```

Oznfc (Wed, 15 Dec 2021 07:39:26 GMT):
what exactly did you say? How can ı do it ? add prop to schema then read and add keycloak ?

baron0426 (Wed, 15 Dec 2021 08:47:45 GMT):
Has joined the channel.

gpaliwal (Wed, 15 Dec 2021 09:07:36 GMT):
Hi, I am trying to provision 4 node Indy network as detailed out at https://github.com/bcgov/von-network, all the 4 nodes and web server on the same instance. The instance is on ubuntu 18.04, there is no public IP to the VM and I am using private IPs only to complete the setup. In the final setup I am seeing errors when these nodes are trying to communicate with each other. In the logs I can see they are trying to communicate over private IP only. Also I can see genesis file getting correctly generated with the private IP values

gpaliwal (Wed, 15 Dec 2021 09:07:36 GMT):
Hi, I am trying to provision 4 node Indy network as detailed out at https://github.com/bcgov/von-network, all the 4 nodes and web server on the same instance. The instance is on ubuntu 18.04, there is no public IP to the VM and I am using private IPs only to complete the setup. In the final setup I am seeing errors when these nodes are trying to communicate with each other. In the logs I can see they are trying to communicate over private IP only. Also I can see genesis file getting correctly generated with the private IP values. Any idea, what I might be missing here

Shweta1 (Wed, 15 Dec 2021 12:24:05 GMT):
HI team,is observer node functionality implemented?

WadeBarnes (Wed, 15 Dec 2021 18:05:48 GMT):
Can you provide additional information? What steps are you following specifically, and what errors are you seeing?

gpaliwal (Wed, 15 Dec 2021 18:20:23 GMT):
steps excuted are as 1. sudo ./manage build 2. sudo ./manage start I can see 5 docker containers getting created, 4 indy nodes and 1 web server In the logs of node1 I see : 021-12-15 18:10:46,684|INFO|motor.py|Node1 changing status from stopped to starting 2021-12-15 18:10:46,686|INFO|zstack.py|Node1 will bind its listener at 0.0.0.0:9701 2021-12-15 18:10:46,688|INFO|stacks.py|CONNECTION: Node1 listening for other nodes at 0.0.0.0:9701 2021-12-15 18:10:46,690|INFO|zstack.py|Node1C will bind its listener at 0.0.0.0:9702 2021-12-15 18:10:46,778|INFO|node.py|Node1 first time running... 2021-12-15 18:10:46,779|INFO|node.py|Node1 processed 0 Ordered batches for instance 0 before starting catch up 2021-12-15 18:10:46,780|INFO|node.py|Node1 processed 0 Ordered batches for instance 1 before starting catch up 2021-12-15 18:10:46,780|INFO|ordering_service.py|Node1:0 reverted 0 batches before starting catch up 2021-12-15 18:10:46,781|INFO|node_leecher_service.py|Node1:NodeLeecherService starting catchup (is_initial=True) 2021-12-15 18:10:46,782|INFO|node_leecher_service.py|Node1:NodeLeecherService transitioning from Idle to PreSyncingPool 2021-12-15 18:10:46,782|INFO|cons_proof_service.py|Node1:ConsProofService:0 starts 2021-12-15 18:10:46,789|INFO|kit_zstack.py|CONNECTION: Node1 found the following missing connections: Node3, Node4, Node2 2021-12-15 18:10:46,790|INFO|zstack.py|CONNECTION: Node1 looking for Node3 at WW.XXX.YYY.ZZZ:9705 2021-12-15 18:10:46,808|INFO|zstack.py|CONNECTION: Node1 looking for Node4 at WW.XXX.YYY.ZZZ:9707 2021-12-15 18:10:46,809|INFO|zstack.py|CONNECTION: Node1 looking for Node2 at WW.XXX.YYY.ZZZ:9703 2021-12-15 18:11:01,793|INFO|cons_proof_service.py|Node1:ConsProofService:0 asking for ledger status of ledger 0 2021-12-15 18:11:16,798|INFO|cons_proof_service.py|Node1:ConsProofService:0 asking for ledger status of ledger 0 2021-12-15 18:11:31,810|INFO|cons_proof_service.py|Node1:ConsProofService:0 asking for ledger status of ledger 0

gpaliwal (Wed, 15 Dec 2021 18:20:23 GMT):
steps excuted are as 1. sudo ./manage build 2. sudo ./manage start I can see 5 docker containers getting created, 4 indy nodes and 1 web server In the logs of node1 I see : 021-12-15 18:10:46,684|INFO|motor.py|Node1 changing status from stopped to starting 2021-12-15 18:10:46,686|INFO|zstack.py|Node1 will bind its listener at 0.0.0.0:9701 2021-12-15 18:10:46,688|INFO|stacks.py|CONNECTION: Node1 listening for other nodes at 0.0.0.0:9701 2021-12-15 18:10:46,690|INFO|zstack.py|Node1C will bind its listener at 0.0.0.0:9702 2021-12-15 18:10:46,778|INFO|node.py|Node1 first time running... 2021-12-15 18:10:46,779|INFO|node.py|Node1 processed 0 Ordered batches for instance 0 before starting catch up 2021-12-15 18:10:46,780|INFO|node.py|Node1 processed 0 Ordered batches for instance 1 before starting catch up 2021-12-15 18:10:46,780|INFO|ordering_service.py|Node1:0 reverted 0 batches before starting catch up 2021-12-15 18:10:46,781|INFO|node_leecher_service.py|Node1:NodeLeecherService starting catchup (is_initial=True) 2021-12-15 18:10:46,782|INFO|node_leecher_service.py|Node1:NodeLeecherService transitioning from Idle to PreSyncingPool 2021-12-15 18:10:46,782|INFO|cons_proof_service.py|Node1:ConsProofService:0 starts 2021-12-15 18:10:46,789|INFO|kit_zstack.py|CONNECTION: Node1 found the following missing connections: Node3, Node4, Node2 2021-12-15 18:10:46,790|INFO|zstack.py|CONNECTION: Node1 looking for Node3 at WW.XXX.YYY.ZZZ:9705 2021-12-15 18:10:46,808|INFO|zstack.py|CONNECTION: Node1 looking for Node4 at WW.XXX.YYY.ZZZ:9707 2021-12-15 18:10:46,809|INFO|zstack.py|CONNECTION: Node1 looking for Node2 at WW.XXX.YYY.ZZZ:9703 2021-12-15 18:11:01,793|INFO|cons_proof_service.py|Node1:ConsProofService:0 asking for ledger status of ledger 0 2021-12-15 18:11:16,798|INFO|cons_proof_service.py|Node1:ConsProofService:0 asking for ledger status of ledger 0 2021-12-15 18:11:31,810|INFO|cons_proof_service.py|Node1:ConsProofService:0 asking for ledger status of ledger 0 WW.XXX.YYY.ZZZ represents the private IP of the VM here

gpaliwal (Wed, 15 Dec 2021 18:24:59 GMT):
Similar errors I see in the logs of node2 as well 2021-12-15 18:19:17,850|INFO|motor.py|Node2 changing status from stopped to starting 2021-12-15 18:19:17,852|INFO|zstack.py|Node2 will bind its listener at 0.0.0.0:9703 2021-12-15 18:19:17,852|INFO|stacks.py|CONNECTION: Node2 listening for other nodes at 0.0.0.0:9703 2021-12-15 18:19:17,857|INFO|zstack.py|Node2C will bind its listener at 0.0.0.0:9704 2021-12-15 18:19:17,908|INFO|node.py|Node2 first time running... 2021-12-15 18:19:17,909|INFO|node.py|Node2 processed 0 Ordered batches for instance 0 before starting catch up 2021-12-15 18:19:17,911|INFO|node.py|Node2 processed 0 Ordered batches for instance 1 before starting catch up 2021-12-15 18:19:17,911|INFO|ordering_service.py|Node2:0 reverted 0 batches before starting catch up 2021-12-15 18:19:17,911|INFO|node_leecher_service.py|Node2:NodeLeecherService starting catchup (is_initial=True) 2021-12-15 18:19:17,912|INFO|node_leecher_service.py|Node2:NodeLeecherService transitioning from Idle to PreSyncingPool 2021-12-15 18:19:17,912|INFO|cons_proof_service.py|Node2:ConsProofService:0 starts 2021-12-15 18:19:17,916|INFO|kit_zstack.py|CONNECTION: Node2 found the following missing connections: Node1, Node3, Node4 2021-12-15 18:19:17,922|INFO|zstack.py|CONNECTION: Node2 looking for Node1 at WW.XXX.YYY.ZZZ:9701 2021-12-15 18:19:17,924|INFO|zstack.py|CONNECTION: Node2 looking for Node3 at WW.XXX.YYY.ZZZ:9705 2021-12-15 18:19:17,927|INFO|zstack.py|CONNECTION: Node2 looking for Node4 at WW.XXX.YYY.ZZZ:9707 2021-12-15 18:19:32,923|INFO|cons_proof_service.py|Node2:ConsProofService:0 asking for ledger status of ledger 0 2021-12-15 18:19:47,930|INFO|cons_proof_service.py|Node2:ConsProofService:0 asking for ledger status of ledger 0 2021-12-15 18:20:02,939|INFO|cons_proof_service.py|Node2:ConsProofService:0 asking for ledger status of ledger 0 2021-12-15 18:20:17,815|NOTIFICATION|primary_connection_monitor_service.py|Node2:0 primary has been disconnected for too long 2021-12-15 18:20:17,816|INFO|primary_connection_monitor_service.py|Node2:0 The node is not ready yet so view change will not be propose d now, but re-scheduled

WadeBarnes (Wed, 15 Dec 2021 18:26:30 GMT):
Are you using the latest version of `von-network`? I had this running on an ubuntu 18.04 VM the other day when I was testing updates.

gpaliwal (Wed, 15 Dec 2021 18:29:46 GMT):
I think it is from main branch executed these steps : curl -L https://github.com/bcgov/von-network/archive/main.zip > bcovrin.zip && \ unzip bcovrin.zip && \ cd von-network-main && \ chmod a+w ./server/

gpaliwal (Wed, 15 Dec 2021 18:32:29 GMT):
you think I should be try tagged release 1.7.2 instead of pulling it from main ?

gpaliwal (Wed, 15 Dec 2021 18:32:29 GMT):
you think, I should try tagged release 1.7.2 instead of pulling it from main ?

esune (Wed, 15 Dec 2021 19:04:26 GMT):
Correct, using a mapper save the role as an attribute instead and then add it to the JWT generated by Keycloak when you authenticate

gpaliwal (Wed, 15 Dec 2021 19:22:34 GMT):
I think the issue is probably related to the VM Only, tried with a new VM, it runs perfectly fine there

WadeBarnes (Wed, 15 Dec 2021 22:29:20 GMT):
:thumbsup:

mccown (Thu, 16 Dec 2021 00:25:11 GMT):
The final Identity Implementer's WG of 2021 is (tomorrow) Thurs Dec 16th @ 9am MT / 3pm UTC.  In addition to reviewing several updates from some key SSI industry WGs, Daniel Hardman (SICPA) will be presenting on the new Gossyp Protocol. Wiki: https://wiki.hyperledger.org/display/IWG/2021-12-16+%3A+Identity+Implementers+WG+Call Zoom: https://zoom.us/my/hyperledger.community.3?pwd=UE90WHhEaHRqOGEyMkV3cldKa2d2dz09

Anthony022 (Thu, 16 Dec 2021 02:50:23 GMT):
Has joined the channel.

beifangfazhanlu (Thu, 16 Dec 2021 08:33:00 GMT):
Has joined the channel.

StepniowskiP (Thu, 16 Dec 2021 08:52:14 GMT):
Has joined the channel.

Oznfc (Thu, 16 Dec 2021 09:37:35 GMT):
I guess I will have to interfere with the keycloak codes for this. Because it only saves simple things (username name surname). Maybe if I save the schema id, I can assign a role accordingly.

leonardoPereira (Thu, 16 Dec 2021 14:08:15 GMT):
Has joined the channel.

swcurran (Thu, 16 Dec 2021 16:55:29 GMT):
IT has not been implemented.

swcurran (Thu, 16 Dec 2021 16:55:29 GMT):
It has not been implemented.

tdiesler (Thu, 16 Dec 2021 17:24:25 GMT):
Has joined the channel.

tdiesler (Thu, 16 Dec 2021 17:25:14 GMT):
Folks, would this be the right place to ask (a basic) python wrapper Q?

tdiesler (Thu, 16 Dec 2021 17:25:14 GMT):
Folks, would this be the right place to ask (a basic) python wrapper Q? I have this code ... if ex.error_code == ErrorCode.PoolLedgerConfigAlreadyExistsError and get ... NameError: name 'ErrorCode' is not defined I find it here https://github.com/hyperledger/indy-sdk/blob/master/wrappers/python/indy/error.py and when I say error.ErrorCode.PoolLedgerConfigAlreadyExistsError it works What would I want to import so that ErrorCode is known?

tdiesler (Thu, 16 Dec 2021 17:25:14 GMT):
Folks, would this be the right place to ask (a basic) python wrapper Q? I have this code ... if ex.error_code == ErrorCode.PoolLedgerConfigAlreadyExistsError and get ... NameError: name 'ErrorCode' is not defined which I found here https://github.com/hyperledger/indy-sdk/blob/master/wrappers/python/indy/error.py and when I say error.ErrorCode.PoolLedgerConfigAlreadyExistsError it works What would I want to import so that ErrorCode is known?

tdiesler (Thu, 16 Dec 2021 17:25:14 GMT):
Folks, would this be the right place to ask (a basic) python wrapper Q? I have this code ... if ex.error_code == ErrorCode.PoolLedgerConfigAlreadyExistsError and get ... NameError: name 'ErrorCode' is not defined which I found here https://github.com/hyperledger/indy-sdk/blob/master/wrappers/python/indy/error.py When I use `error.ErrorCode.PoolLedgerConfigAlreadyExistsError` it works What would I want to import so that ErrorCode is known?

andrew.whitehead (Thu, 16 Dec 2021 17:36:10 GMT):
from Indy.error import ErrorCode I believe

tdiesler (Thu, 16 Dec 2021 17:37:30 GMT):
Yes, of course - Thanks

grzegorz.drozda (Thu, 16 Dec 2021 18:54:11 GMT):
Has joined the channel.

esune (Thu, 16 Dec 2021 21:05:37 GMT):
Yes, it would be something you need to do in Keycloak using mappers - I think

daiki.nakashima (Fri, 17 Dec 2021 04:18:29 GMT):
Has joined the channel.

Abcold (Fri, 17 Dec 2021 06:11:55 GMT):
Has joined the channel.

Amits1 (Fri, 17 Dec 2021 11:10:59 GMT):
Has joined the channel.

Amits1 (Fri, 17 Dec 2021 11:21:53 GMT):
Hi, i am creating 4 nodes using https://hyperledger-indy.readthedocs.io/projects/node/en/latest/start-nodes.html.I want to change trustee of this network.I don't want to use default trustee.(000000000000000000000000Trustee1 ).How can i do that?

Amits1 (Fri, 17 Dec 2021 11:21:53 GMT):
Hi, i am creating 4 nodes using https://hyperledger-indy.readthedocs.io/projects/node/en/latest/start-nodes.html.I want to change trustee of this network.I don't want to use default trustee.(000000000000000000000000Trustee1 ).How can i do that?Also how i can change Steward1,Steward2,Steward3,Steward4 seed?is any command to create domain transaction genesis file?

Amits1 (Fri, 17 Dec 2021 11:21:53 GMT):
Hi, i am creating 4 nodes using https://hyperledger-indy.readthedocs.io/projects/node/en/latest/start-nodes.html.I want to change trustee of this network.I don't want to use default trustee.(000000000000000000000000Trustee1 ).How can i do that?Also how i can change Steward1,Steward2,Steward3,Steward4 seed?is any command to create domain transaction genesis file? Many thanks !!

UnparagonedWisdom (Fri, 17 Dec 2021 13:02:27 GMT):
Has joined the channel.

WadeBarnes (Fri, 17 Dec 2021 15:59:14 GMT):
Depending on what you're trying to accomplish here are some recommendations; https://github.com/bcgov/von-network/issues/190#issuecomment-995918202

ThreeJam (Fri, 17 Dec 2021 18:02:54 GMT):
Has joined the channel.

barcellosrod (Fri, 17 Dec 2021 21:24:19 GMT):
Has joined the channel.

Amits1 (Sat, 18 Dec 2021 11:38:34 GMT):
Thanks for reply!!is there any document for indy network setup on production.

TheRealJimShady (Sat, 18 Dec 2021 15:33:55 GMT):
Has joined the channel.

EdditLemons (Sat, 18 Dec 2021 17:43:11 GMT):
Has joined the channel.

JeenaDevasia (Sun, 19 Dec 2021 13:23:35 GMT):
Has joined the channel.

WadeBarnes (Sun, 19 Dec 2021 15:54:44 GMT):
The link to the `Create New Indy Network` document.

ayushridhar (Sun, 19 Dec 2021 18:02:22 GMT):
Has joined the channel.

samih 5 (Sun, 19 Dec 2021 18:04:25 GMT):
Has joined the channel.

jkvargas (Sun, 19 Dec 2021 18:52:01 GMT):
Has joined the channel.

sbohanlf (Mon, 20 Dec 2021 14:06:18 GMT):
Has joined the channel.

FabienFF (Mon, 20 Dec 2021 16:35:30 GMT):
Has joined the channel.

MBurnes (Mon, 20 Dec 2021 23:42:47 GMT):
Has joined the channel.

conanoc (Tue, 21 Dec 2021 02:57:13 GMT):
I made a presentation about Hyperledger Indy to introduce it to my team. You can take it if you need it. https://www.slideshare.net/ssuser3993f3/hyperledger-indy-tutorial

Yangshuangxi (Tue, 21 Dec 2021 06:57:19 GMT):
Has joined the channel.

elchastro (Tue, 21 Dec 2021 10:46:26 GMT):
Has joined the channel.

skbalahere (Tue, 21 Dec 2021 16:49:35 GMT):
Has joined the channel.

a-p-petrosyan (Wed, 22 Dec 2021 08:11:42 GMT):
Has joined the channel.

ammaarpatel99 (Wed, 22 Dec 2021 10:07:16 GMT):
Has joined the channel.

Abhi-Mynam (Wed, 22 Dec 2021 11:34:09 GMT):
Has joined the channel.

DinuraSeneviratne1 (Wed, 22 Dec 2021 12:16:12 GMT):
Has joined the channel.

Oznfc (Wed, 22 Dec 2021 13:10:16 GMT):
Hi , does cridentials has user_did or wallet_did ?

Oznfc (Wed, 22 Dec 2021 13:11:49 GMT):
when user login with cridentials , can see user's DID?

swcurran (Wed, 22 Dec 2021 19:13:16 GMT):
BC Gov has posted a funded "Code With Us" Challenge for individuals/teams interested in updating Hyperledger Indy to support the new `did:indy` DID Method, specified [here](https://hyperledger.github.io/indy-did-method/). The opportunity and response guidelines are [posted here](https://digital.gov.bc.ca/marketplace/opportunities/code-with-us/e3dd1605-cc1d-4c30-a9ee-245940bccd0d). The total funding for the challenge is $70,000CDN and is divided into 4 phases. The first 3 phases require the use of Python working on the [Indy Node](https://github.com/hyperledger/indy-node) and [Indy Plenum](https://github.com/hyperledger/indy-plenum) repos, while the 4th phase requires Rust development in the [Indy VDR](https://github.com/hyperledger/indy-vdr) repo. Those responding may choose to apply for only the Python work, only the Rust work or both. Questions can be asked here or on the website hosting the opportunity. Applications for the challenge will be accepted up to January 10, 2022. and the work is expected to begin shortly after that date, with completion expected by March 31, 2022.

p.amrith (Thu, 23 Dec 2021 05:18:54 GMT):
Has joined the channel.

krishna1982 (Thu, 23 Dec 2021 06:20:21 GMT):
Has joined the channel.

Dixit 13 (Thu, 23 Dec 2021 07:06:44 GMT):
Has joined the channel.

KavyaR (Thu, 23 Dec 2021 11:44:40 GMT):
Has joined the channel.

tahyonline (Thu, 23 Dec 2021 16:40:53 GMT):
Has joined the channel.

SkandaGurunathanR (Thu, 23 Dec 2021 19:10:44 GMT):
Has joined the channel.

mgrant (Fri, 24 Dec 2021 12:14:53 GMT):
Has joined the channel.

nandrews283 (Fri, 24 Dec 2021 13:59:04 GMT):
Has joined the channel.

lvillar (Sun, 26 Dec 2021 03:30:53 GMT):
Has joined the channel.

friendlyhacker (Sun, 26 Dec 2021 07:12:36 GMT):
Has joined the channel.

skittlez1522 (Mon, 27 Dec 2021 03:32:38 GMT):
Has joined the channel.

skittlez1522 (Mon, 27 Dec 2021 03:33:04 GMT):
Can anyone say whether or not Indy has been abandoned?

sudharsan.s (Mon, 27 Dec 2021 09:42:17 GMT):
Has joined the channel.

yogi-cw (Mon, 27 Dec 2021 12:07:41 GMT):
Has joined the channel.

WadeBarnes (Mon, 27 Dec 2021 12:48:54 GMT):
It has not been abandoned.

marya1 (Mon, 27 Dec 2021 14:19:56 GMT):
Has joined the channel.

skittlez1522 (Mon, 27 Dec 2021 17:38:37 GMT):
OK, thanks! I was just making sure.

GrzegorzDrozda (Mon, 27 Dec 2021 19:18:55 GMT):
Has joined the channel.

omkar-mohanty (Tue, 28 Dec 2021 04:07:33 GMT):
Has joined the channel.

hejte (Tue, 28 Dec 2021 11:00:16 GMT):
Has joined the channel.

EmSpireInc (Tue, 28 Dec 2021 11:52:59 GMT):
Has joined the channel.

Xcerla (Tue, 28 Dec 2021 21:24:50 GMT):
Has joined the channel.

Vishwas1 (Wed, 29 Dec 2021 07:02:27 GMT):
Has joined the channel.

Vishwas1 (Wed, 29 Dec 2021 07:02:27 GMT):
Hi all, I am new to SSI. I am confused between Indy and Aries projects. What are their differences ?

WadeBarnes (Wed, 29 Dec 2021 12:31:34 GMT):
Aries is compatible with Indy, and in most cases, at the moment, it is built on top of Indy. The Aries frameworks provide many layers of abstraction that make it easier for you to concentrate on creating your application without worrying about the intricacies of what's going on under the hood at the Indy levels.

WadeBarnes (Wed, 29 Dec 2021 12:31:34 GMT):
Since you're new to all this Aries is the place to start.

WadeBarnes (Wed, 29 Dec 2021 12:31:34 GMT):
Since you're new to all this, Aries is the place to start.

Ranbowhu (Wed, 29 Dec 2021 14:25:52 GMT):
Has joined the channel.

skittlez1522 (Thu, 30 Dec 2021 01:26:32 GMT):
Hi. I am trying to get my NodeJS server connected to an indy pool I have connected to the Sovrin buildernet, but I cannot find any documentation on how to do that. Does anyone know of any docs that show how to do that?

conanoc (Thu, 30 Dec 2021 06:55:22 GMT):
https://stackoverflow.com/questions/61158552/hyperledger-indy-connecting-to-an-existing-pool

IraqReshi (Thu, 30 Dec 2021 07:27:29 GMT):
Has joined the channel.

dev2020 (Thu, 30 Dec 2021 09:09:04 GMT):
Has joined the channel.

skittlez1522 (Thu, 30 Dec 2021 16:25:58 GMT):
Thank you!!!

gaurav471 (Thu, 30 Dec 2021 19:06:16 GMT):
Has joined the channel.

MojoCoderBo (Fri, 31 Dec 2021 07:11:55 GMT):
Has joined the channel.

Ojas365 (Fri, 31 Dec 2021 09:16:25 GMT):
Has joined the channel.

parnika24 (Fri, 31 Dec 2021 10:59:12 GMT):
Has joined the channel.

kareemikassab (Fri, 31 Dec 2021 12:33:00 GMT):
Has joined the channel.

MaximeBarbet (Fri, 31 Dec 2021 15:16:54 GMT):
Has joined the channel.

vp10905353 (Fri, 31 Dec 2021 19:14:17 GMT):
Has joined the channel.

jsolderitsch (Fri, 31 Dec 2021 21:49:56 GMT):
Has joined the channel.

manjinder98 (Sat, 01 Jan 2022 14:36:47 GMT):
Has joined the channel.

bshada (Sat, 01 Jan 2022 20:36:04 GMT):
Has joined the channel.

kma-blockchain (Sun, 02 Jan 2022 09:20:05 GMT):
Has joined the channel.

yangzifeng1995 (Sun, 02 Jan 2022 10:14:14 GMT):
Has joined the channel.

ManjinderSingh (Sun, 02 Jan 2022 10:43:32 GMT):
Has joined the channel.

Nishantsharma12 (Sun, 02 Jan 2022 13:25:13 GMT):
Has joined the channel.

HackyCoder0951 (Mon, 03 Jan 2022 08:24:19 GMT):
Has joined the channel.

tanujdev (Mon, 03 Jan 2022 16:22:39 GMT):
Has joined the channel.

kinagimanju (Mon, 03 Jan 2022 16:42:30 GMT):
Has joined the channel.

richardspeter (Mon, 03 Jan 2022 19:05:44 GMT):
Has joined the channel.

loganb 6 (Mon, 03 Jan 2022 21:45:18 GMT):
Has joined the channel.

haruharu (Tue, 04 Jan 2022 01:10:29 GMT):
Has joined the channel.

zondervan.v.calvez (Tue, 04 Jan 2022 01:43:38 GMT):
Has joined the channel.

micoferdinand (Tue, 04 Jan 2022 01:50:08 GMT):
Has joined the channel.

aldous.arvin.alvarez (Tue, 04 Jan 2022 01:53:11 GMT):
Has joined the channel.

charellesandig (Tue, 04 Jan 2022 02:09:52 GMT):
Has joined the channel.

suomea (Tue, 04 Jan 2022 08:36:08 GMT):
Has joined the channel.

mr3e (Tue, 04 Jan 2022 09:23:40 GMT):
Has joined the channel.

vborisik (Tue, 04 Jan 2022 11:18:11 GMT):
Has joined the channel.

AlaeEddineLahlali (Tue, 04 Jan 2022 14:04:34 GMT):
Has joined the channel.

billet75 (Tue, 04 Jan 2022 15:14:29 GMT):
Has joined the channel.

arvieira (Tue, 04 Jan 2022 15:41:51 GMT):
Has joined the channel.

damiandennis (Tue, 04 Jan 2022 18:46:57 GMT):
Has joined the channel.

swcurran (Tue, 04 Jan 2022 19:39:15 GMT):
Lots going on! Mostly working CI/CD modernization right now -- it's been a roadblock. Next up is supporting for the "did:indy" DID Method: - Specification: https://hyperledger.github.io/indy-did-method/ - Opportunity to be paid to develop DID:indy: https://digital.gov.bc.ca/marketplace/opportunities/code-with-us/e3dd1605-cc1d-4c30-a9ee-245940bccd0d

Nocturne (Wed, 05 Jan 2022 00:22:10 GMT):
Has joined the channel.

PunleuChomnan (Wed, 05 Jan 2022 09:22:18 GMT):
Has joined the channel.

VenessaK (Wed, 05 Jan 2022 14:38:11 GMT):
Has joined the channel.

Zeddrich (Wed, 05 Jan 2022 15:59:11 GMT):
Has joined the channel.

PureBlack (Thu, 06 Jan 2022 01:07:48 GMT):
Has joined the channel.

esainty (Thu, 06 Jan 2022 01:20:07 GMT):
Has joined the channel.

pvsmagbojos (Thu, 06 Jan 2022 04:37:42 GMT):
Has joined the channel.

ntlm (Thu, 06 Jan 2022 05:38:38 GMT):
Has joined the channel.

obioram (Thu, 06 Jan 2022 06:36:50 GMT):
Has joined the channel.

ayush5936 (Thu, 06 Jan 2022 06:38:39 GMT):
Has joined the channel.

Teharjula (Thu, 06 Jan 2022 09:59:21 GMT):
Has joined the channel.

DIBYENDUBAG (Thu, 06 Jan 2022 11:51:23 GMT):
Has joined the channel.

Paul_Tabled (Thu, 06 Jan 2022 12:09:12 GMT):
Has joined the channel.

jas191 (Thu, 06 Jan 2022 14:12:47 GMT):
Has joined the channel.

MyeongheeJoung (Thu, 06 Jan 2022 18:37:47 GMT):
Has joined the channel.

MyeongheeJoung (Thu, 06 Jan 2022 18:40:52 GMT):
Hi I saw some article that Korean application "initial", which made by SKT is based on Hyperledger Indy. Could you explain how it works?

Mj (Thu, 06 Jan 2022 18:43:50 GMT):
And could university independently make did? ? then how it works

Mj (Thu, 06 Jan 2022 18:43:51 GMT):
?

Sergioco 2 (Thu, 06 Jan 2022 19:50:04 GMT):
Has joined the channel.

Tenno (Fri, 07 Jan 2022 01:57:11 GMT):
Has left the channel.

skittlez1522 (Fri, 07 Jan 2022 03:36:09 GMT):
Read this: https://www.edx.org/course/identity-in-hyperledger-aries-indy-and-ursa ``` ``` It's a lot of reading, but it explains Hyperledger Indy, along with Aries and Ursa very well. And you don't have to pay for the course. That just lets you get a certificate.

trustcerts (Fri, 07 Jan 2022 11:49:10 GMT):
Hello everyone, I am analyzing the attack vectors of a blockchain like ddos. Since indy uses two ports the internal consensus api cannot be accessed from the outside world. But since Indy is using ZeroMQ to handle client-server connections in a "hungry" way, the docs recommend to limit amount the parallel connections. https://hyperledger-indy.readthedocs.io/projects/node/en/latest/setup-iptables.html#using-raw-iptables-command-or-iptables-front-end So an attacker could open 500 parallel connections to a node to prevent new clients to interact with a node. The huge read request (just some json stuff to blow the size up) will go through something like toxy proxy to reduce my bandwidth, so the connections will be up for a long time. Has anyone else made tests like this before? Or is this problem already solved and the docs just outdated?

trustcerts (Fri, 07 Jan 2022 11:49:10 GMT):
Hello everyone, I am analyzing the attack vectors of a blockchain like ddos. Since indy uses two ports the internal consensus api cannot be accessed from the outside world. But since Indy is using ZeroMQ to handle client-server connections in a "hungry" way, the docs recommend to limit amount the parallel connections. https://hyperledger-indy.readthedocs.io/projects/node/en/latest/setup-iptables.html#using-raw-iptables-command-or-iptables-front-end So an attacker could open 500 parallel connections to a node to prevent new clients to interact with a node. The huge read request (just some json stuff to blow the size up) will go through something like toxy proxy to reduce my bandwidth, so the connections will be up for a long time. Has anyone else made tests like this before? Or is this problem already solved and the docs just outdated? The "classic" way to protect a web service like this would be horizontal scaling of the http endpoint, but for this scenario indy has to be a microservice architecture.

trustcerts (Fri, 07 Jan 2022 11:49:10 GMT):
Hello everyone, I am analyzing the attack vectors of a blockchain like ddos. Since indy uses two ports the internal consensus api can not be accessed from the outside world. But Indy is using ZeroMQ to handle client-server connections in a "hungry" way, that's why the docs recommend to limit amount the parallel connections. https://hyperledger-indy.readthedocs.io/projects/node/en/latest/setup-iptables.html#using-raw-iptables-command-or-iptables-front-end So an attacker could open 500 parallel connections to a node to prevent new clients to interact with a node. The huge read request (just some json stuff to blow the size up) will go through something like toxy proxy to reduce my bandwidth, so the connections will be up for a long time. Has anyone else made tests like this before? Or is this problem already solved and the docs just outdated? The "classic" way to protect a web service like this would be horizontal scaling of the http endpoint, but for this scenario indy has to be a microservice architecture.

mzago (Fri, 07 Jan 2022 16:04:36 GMT):
Has joined the channel.

pawlak (Fri, 07 Jan 2022 17:12:47 GMT):
Has joined the channel.

esplinr (Sat, 08 Jan 2022 05:04:07 GMT):
This is an inherent vulnerability with Indy's current architecture. That's why each validator should run a firewall. A DDOS that follows the deterministic leader through consensus view changes means that a sustained attack could bring down the network. But that would be very hard to achieve in practice. We did tests in 2019 where we took the network through thousands of view changes under high network load and the network recovered as soon as the attack halted. We've talked about changing the consensus protocol to make it more robust against such attacks using an approach like Aardvark or the Tendermint protocol. At this time, no one is volunteering to do that work.

esplinr (Sat, 08 Jan 2022 05:04:07 GMT):
This is an inherent vulnerability with Indy's current architecture. That's why each validator should run a firewall. A DDOS that follows the deterministic leader through consensus view changes means that a sustained attack could bring down the network. But that would be very hard to achieve in practice. We did tests in 2019 where we took the network through thousands of view changes under high network load and the network recovered as soon as the attack halted. Even during the attack, the other nodes were available for read-only access. We've talked about changing the consensus protocol to make it more robust against such attacks using an approach like Aardvark or the Tendermint protocol. At this time, no one is volunteering to do that work.

esplinr (Sat, 08 Jan 2022 05:04:07 GMT):
This is an inherent vulnerability with Indy's current architecture. That's why each validator should run a firewall. A DDOS that follows the deterministic leader through consensus view changes means that a sustained attack could bring down the network. But that would be very hard to achieve in practice. We did tests in 2019 where we took the network through thousands of view changes under high network load and the network recovered as soon as the attack halted. Even during the attack, the other nodes were available for read-only access. We've talked about changing the consensus protocol to make it more robust against such attacks using an approach like Aardvark or the Tendermint protocol. At this time, no one is volunteering to do that work. The history of #indy-ledger-next archives many of those conversations.

gouthamdk (Sat, 08 Jan 2022 07:03:23 GMT):
Has joined the channel.

iraq17 (Sat, 08 Jan 2022 09:47:37 GMT):
Has joined the channel.

neetusharma (Sat, 08 Jan 2022 11:26:00 GMT):
Has joined the channel.

trustcerts (Sat, 08 Jan 2022 14:39:13 GMT):
I am not talking about a DDos attack where you "bomb" a huge amount of traffic to overload the CPU and RAM of a node. This can be prevented, as you mentioned, with a rate limit for the connections of a firewall. But when I send N requests, where N is the max amount of parallel connections defined, each request is about 10MB and I am slowing down the bandwith to 1KB/s, I can block all incoming traffic for around 10.000 seconds and then run the requests again. The consensus isn't touched at that point, but no agent will be able to send new transactions to one of the nodes, doesn't matter if its a read or write request.

mjs19780 (Sat, 08 Jan 2022 20:14:40 GMT):
Has joined the channel.

SAYED94 (Sat, 08 Jan 2022 20:33:45 GMT):
Has joined the channel.

SAYED94 (Sat, 08 Jan 2022 21:39:48 GMT):
Hello everyone I am engaged in projects applying blockchain technology for data security. For instance, sensors deployed in the field, and they collected the measurement and stream to the data aggregator. Now my project scope is to secure this data exchange and data storage. I am very new in blockchain and IT, so i am seeking guideline help if anyone can so that i can start. You may recommend me a series of courses for beginners so that i start from the beginning. Thank you for your time!

wifisunset94 (Sat, 08 Jan 2022 22:06:03 GMT):
Has joined the channel.

sayraj (Sun, 09 Jan 2022 08:49:21 GMT):
Has joined the channel.

Tushar199 1 (Sun, 09 Jan 2022 13:26:11 GMT):
Has joined the channel.

akhmad_219 (Mon, 10 Jan 2022 08:27:50 GMT):
Has joined the channel.

iamit-singh (Mon, 10 Jan 2022 09:18:23 GMT):
Has joined the channel.

arieger (Mon, 10 Jan 2022 10:44:51 GMT):
Has joined the channel.

amitgarg12 (Mon, 10 Jan 2022 11:44:30 GMT):
Has joined the channel.

antauloglou (Mon, 10 Jan 2022 15:38:19 GMT):
Has joined the channel.

inttrading (Mon, 10 Jan 2022 17:11:44 GMT):
Has joined the channel.

samyakjain (Tue, 11 Jan 2022 04:42:54 GMT):
Has joined the channel.

KodeCracker (Tue, 11 Jan 2022 04:52:15 GMT):
Has joined the channel.

tongYuSir (Tue, 11 Jan 2022 06:49:37 GMT):
Has joined the channel.

hrishiballal (Tue, 11 Jan 2022 08:54:28 GMT):
Has joined the channel.

prashantr (Tue, 11 Jan 2022 09:26:11 GMT):
Has joined the channel.

Rahul11132133 (Tue, 11 Jan 2022 10:29:42 GMT):
Has joined the channel.

p10010 (Tue, 11 Jan 2022 12:10:09 GMT):
Has joined the channel.

samyak314 (Tue, 11 Jan 2022 12:35:59 GMT):
Has joined the channel.

Ammar1974 (Tue, 11 Jan 2022 18:20:46 GMT):
Has joined the channel.

raghunathanp95 (Tue, 11 Jan 2022 20:17:40 GMT):
Has joined the channel.

JeffCantores (Wed, 12 Jan 2022 02:56:07 GMT):
Has joined the channel.

ataberkozek (Wed, 12 Jan 2022 14:00:27 GMT):
Hi Everyone, is indy SDK compatible with OpenShift? If it is, is there a guide on how to deploy indy SDK on OpenShift?

lbradley (Wed, 12 Jan 2022 14:29:31 GMT):
Has joined the channel.

akshitgoyal (Wed, 12 Jan 2022 17:26:12 GMT):
Has joined the channel.

StefanBratanov93 (Wed, 12 Jan 2022 19:26:57 GMT):
Has joined the channel.

HaBerber (Thu, 13 Jan 2022 07:59:20 GMT):
Has joined the channel.

shaikh47 (Thu, 13 Jan 2022 10:12:01 GMT):
Has joined the channel.

trustcerts (Thu, 13 Jan 2022 13:35:50 GMT):
The Indy SDK is just a library you can use to build your own software. Or did you mean indy node, the software to run a blockchain node?

mccown (Thu, 13 Jan 2022 14:15:00 GMT):
The Identity Implementer’s WG meeting for today (Jan 13th) has been cancelled. Please join us in our next meeting on Jan 27th (9am MT / 4pm UTC).

SeanBohan (Thu, 13 Jan 2022 15:06:34 GMT):
FinCEN, FDIC to Hold 'Tech Sprint' for Digital Identity Tools The proliferation of scams, information leaks and synthetic identity fraud is presenting a major challenge to the online financial services industry, federal regulators said. https://www.coindesk.com/policy/2022/01/12/fincen-fdic-to-hold-tech-sprint-for-digital-identity-tools/

tdiesler (Thu, 13 Jan 2022 15:35:57 GMT):
Folks, trying to run the Java wrapper tests on MacOS, I see ... ``` TRACE org.hyperledger.indy.sdk.LibIndy.nativeindy.api.pool src/api/pool.rs:56 | indy_create_pool_ledger_config: <<< res: Success DEBUG org.hyperledger.indy.sdk.LibIndy.nativepool_command_executor src/commands/pool.rs:63 | Create command received DEBUG org.hyperledger.indy.sdk.LibIndy.nativeindy.commands.pool src/commands/pool.rs:139 | create >>> name: "default_pool", config: Some(PoolConfig { genesis_txn: "/var/folders/tv/bjwqm2bx59j5nw3nv1pwvffm0000gn/T/indy_client/temp.txn" }) TRACE org.hyperledger.indy.sdk.LibIndy.nativeindy.services.pool src/services/pool/mod.rs:68 | PoolService::create default_pool with config Some(PoolConfig { genesis_txn: "/var/folders/tv/bjwqm2bx59j5nw3nv1pwvffm0000gn/T/indy_client/temp.txn" }) dyld[80457]: missing symbol called /bin/sh: line 1: 80457 Abort trap: 6 /Library/Java/JavaVirtualMachines/jdk1.8.0_281.jdk/Contents/Home/jre/bin/java -jar /Users/tdiesler/git/indy-sdk/wrappers/java/target/surefire/surefirebooter5545266179096662125.jar /Users/tdiesler/git/indy-sdk/wrappers/java/target/surefire/surefire7682983429577780883tmp /Users/tdiesler/git/indy-sdk/wrappers/java/target/surefire/surefire_05781558020738526106tmp ``` Do I perhaps need a specific Java distro/version?

tdiesler (Thu, 13 Jan 2022 15:58:36 GMT):
I wanted to attend the weekly "Hyperledger Indy Working Group" meeting. Is this link not valid any more? (https://zoom.us/j/232861185)

rfleschenberg (Thu, 13 Jan 2022 16:01:53 GMT):
Has joined the channel.

rfleschenberg (Thu, 13 Jan 2022 16:02:04 GMT):
Hello everyone

rfleschenberg (Thu, 13 Jan 2022 16:03:45 GMT):
I'm currently investigating if it's possible to run indy nodes on Kubernetes. Hyperledger bevel seems to support this. As far as I understand, the genesis files need to contain node IPs (not hostnames). Does anyone know how they handle this, since the Kubernetes pod IPs can change at any time?

swcurran (Thu, 13 Jan 2022 16:06:04 GMT):
The Indy Working Group meeting has morphed into an "Identity Implementers WG" meeting on Thursday's, every 2nd week (but cancelled today). It covers broader issues in Identity. The Indy Contributor's call is next Tuesday and every second week. The Aries Working Group is every week on Wednesday morning. Those are both focused on building, deploying and using Indy and Aries, respectively. Links:

swcurran (Thu, 13 Jan 2022 16:06:10 GMT):
https://wiki.hyperledger.org/display/indy/Indy+Contributors+Meeting

swcurran (Thu, 13 Jan 2022 16:06:50 GMT):
https://wiki.hyperledger.org/display/ARIES/Aries+Working+Group

swcurran (Thu, 13 Jan 2022 16:07:44 GMT):
https://wiki.hyperledger.org/display/IWG/Identity+Implementers+WG+Call

tdiesler (Thu, 13 Jan 2022 16:20:01 GMT):
merci

swcurran (Thu, 13 Jan 2022 16:24:41 GMT):
FYI -- here is a full Hyperledger Calendar: https://wiki.hyperledger.org/display/HYP/Calendar+of+Public+Meetings

WadeBarnes (Thu, 13 Jan 2022 19:59:17 GMT):
They might be able to help you with that over on the #bevel channel. I'd be interested in how that is addressed too. I have not had the chance to dig into Bevel yet.

WadeBarnes (Thu, 13 Jan 2022 20:01:40 GMT):
The way indy-node wants to address things is not very compatible with k8s out of the box.

iliketurtles69 (Fri, 14 Jan 2022 05:09:13 GMT):
Has joined the channel.

huonghope (Fri, 14 Jan 2022 08:48:14 GMT):
Has joined the channel.

veaceslav.dimitroglo (Fri, 14 Jan 2022 10:07:17 GMT):
Has joined the channel.

kimzter (Fri, 14 Jan 2022 12:58:17 GMT):
Has joined the channel.

Cassius87 (Fri, 14 Jan 2022 13:23:53 GMT):
Has joined the channel.

SeanConnolly (Fri, 14 Jan 2022 17:39:19 GMT):
Has joined the channel.

dhruvsharma (Fri, 14 Jan 2022 19:22:13 GMT):
Has joined the channel.

AdamEikman (Fri, 14 Jan 2022 19:45:20 GMT):
Has joined the channel.

dakata (Sat, 15 Jan 2022 13:56:52 GMT):
Has joined the channel.

luisnaranjo733 (Sat, 15 Jan 2022 15:34:28 GMT):
Has joined the channel.

g4gekkouga (Sun, 16 Jan 2022 07:36:04 GMT):
Has joined the channel.

knagware9 (Sun, 16 Jan 2022 08:40:55 GMT):
Do we have some data on TPS of Hyperedger Indy network?

boozden (Sun, 16 Jan 2022 10:11:09 GMT):
Has joined the channel.

roberto.porfiro (Sun, 16 Jan 2022 11:11:38 GMT):
Has joined the channel.

WadeBarnes (Sun, 16 Jan 2022 16:22:23 GMT):
@esplinr, Do you have any performance data from any of the load tests you've done in the past?

HuzMS (Sun, 16 Jan 2022 22:42:31 GMT):
Has joined the channel.

conanoc (Mon, 17 Jan 2022 07:25:16 GMT):
You can get a static external IP of the pod from a service object in k8s. https://kubernetes.io/docs/tutorials/stateless-application/expose-external-ip-address/

ataberkozek (Mon, 17 Jan 2022 08:32:21 GMT):
Yeah i meant indy-node. I found out the scripts for openshift but build configs for indy client and indy agent gives an error. This is from the build log `STEP 15/22: COPY ./scripts/common/*.sh /home/indy/ --> 27a2132ad86 STEP 16/22: COPY ./scripts/agent/start.sh /home/indy/ error: build error: error building at STEP "COPY ./scripts/agent/start.sh /home/indy/": checking on sources under "/tmp/build/inputs/environment/openshift": copier: stat: "/scripts/agent/start.sh": no such file or directory` How can i solve this?

ataberkozek (Mon, 17 Jan 2022 11:28:10 GMT):
also the pods have this error: `Traceback (most recent call last): File "/usr/local/bin/generate_indy_pool_transactions", line 18, in TestNetworkSetup.bootstrapTestNodes(getConfig(), portsStart, nodeParamsFileName, File "/usr/local/lib/python3.5/dist-packages/indy_common/config_util.py", line 31, in getConfig CONFIG = _getConfig(PlenumConfig, general_config_dir, user_config_dir) File "/usr/local/lib/python3.5/dist-packages/indy_common/config_util.py", line 24, in _getConfig raise ex File "/usr/local/lib/python3.5/dist-packages/indy_common/config_util.py", line 21, in _getConfig user_config_dir=user_config_dir) File "/usr/local/lib/python3.5/dist-packages/plenum/common/config_util.py", line 52, in extend_with_default_external_config os.path.join(extendee.GENERAL_CONFIG_DIR, extendee.GENERAL_CONFIG_FILE))) Exception: NETWORK_NAME must be set in /etc/indy/indy_config.py =============================================================================== Starting indy node ... Starting indy-node service ... /usr/bin/env python3 -O /usr/local/bin/start_indy_node Node2 9703 9704 Traceback (most recent call last): File "/usr/local/bin/start_indy_node", line 12, in raise Exception("Provide name and two pairs of IP/port for running the node " Exception: Provide name and two pairs of IP/port for running the node and client stacks in form 'node_name node_ip node_port client_ip client_port'`

ataberkozek (Mon, 17 Jan 2022 11:54:37 GMT):
it seems that node_ip and client_ip are missing as the exception suggests. Where can i add these arguments?

AAlzaabi (Mon, 17 Jan 2022 14:41:17 GMT):
Has joined the channel.

newbusybee (Tue, 18 Jan 2022 01:01:56 GMT):
Has joined the channel.

sudojhill (Tue, 18 Jan 2022 05:37:15 GMT):
Has joined the channel.

DevalShah (Tue, 18 Jan 2022 07:05:12 GMT):
Has joined the channel.

adaephonben (Tue, 18 Jan 2022 08:40:16 GMT):
Has joined the channel.

christianatmueller (Tue, 18 Jan 2022 14:10:19 GMT):
Has joined the channel.

WadeBarnes (Tue, 18 Jan 2022 14:26:04 GMT):
The caveat there is the special requirements for creating the `external load balancer` resource.

Affin-Ajanthan (Tue, 18 Jan 2022 14:35:03 GMT):
Has joined the channel.

nglf (Tue, 18 Jan 2022 15:23:36 GMT):
Has joined the channel.

Ravjot (Tue, 18 Jan 2022 17:28:57 GMT):
Has joined the channel.

conanoc (Wed, 19 Jan 2022 01:49:49 GMT):
Yeah. Creating an external load balancer is not a special requirement, as it is also necessary to simply run a single web server in a k8s cluster.

HariN 35 (Wed, 19 Jan 2022 03:44:49 GMT):
Has joined the channel.

harikrishnashk (Wed, 19 Jan 2022 06:29:39 GMT):
Has joined the channel.

sumiteshnaithani (Wed, 19 Jan 2022 07:18:44 GMT):
Has joined the channel.

seancm69 (Wed, 19 Jan 2022 07:24:45 GMT):
Has joined the channel.

Tipu_Singh (Wed, 19 Jan 2022 07:41:33 GMT):
Has joined the channel.

sergedh (Wed, 19 Jan 2022 07:57:50 GMT):
Has joined the channel.

souksou (Wed, 19 Jan 2022 10:06:12 GMT):
Has joined the channel.

ouyangwenli (Wed, 19 Jan 2022 13:46:59 GMT):
Has joined the channel.

robbesettlemint (Wed, 19 Jan 2022 16:15:05 GMT):
Has joined the channel.

esplinr (Wed, 19 Jan 2022 16:38:31 GMT):
Indy's performance will depend hugely on how big the pool of validators is, how much RAM and compute power they have, and how fast their network connections are. Thus, there isn't a general "Indy performance" study. The team that builds Indy evaluates it in a specific configuration to guarantee acceptable performance before each release. They are trying to prove sustained reads of thousands of reads per second and >10 writes per second, with a validator pool of 25-ish, and they have consistently achieved those metrics. Less validators would increase write performance significantly. (10 writes per second sounds slow to me, although it is twice as fast as the global bitcoin network. However, Indy has a number of performance optimizations that are still teed up, that should easily allow this number to be 1-3 orders of magnitude faster when we need it. It just hasn't been a focus for now, partly because of how skewed traffic is toward reading.) The bottom line is that Indy's performance is predictable, and fast enough, for the use cases we have today. And it can be increased substantially when we need to. Exchanging credentials does not involve the ledger very much, though. Issuance doesn't require any write with each credential--only a write of the schema, cred def, and revocation registry that they all use. So you could issue 1000 or 10M credentials, and you still only have 3 ledger writes and no ledger reads (except maybe to populate a cache). When the credential is used to present proof, the verifier and the holder/prover each do a ledger read to check an accumulator value. These reads are likely to be cached for the holder, if they are proving regularly; in real usage, only the verifier's reads will probably have a linear correlation to ledger traffic. Because of the protections on correlation, performance metrics don't need to be a primary focus (but any and all help here is very much appreciated). Time correlation means that throughput spikes ought to be spread out over time, so we don't have to worry about "visa network" type numbers. We have been using 10 writes per second and 100 reads per second as our performance minimum (if we fall below that we are worried). Benchmarks can go much much faster of course, the Sovrin Foundation uses those numbers as a guideline for our production configuration on a global pool of 24 nodes. Also, regarding performance - 10 write transactions per second can look small compared to 1000s, but there is huge difference in benchmarking conditions: - other ledgers usually use just 4 nodes, often inside one LAN, while 10 TPS in plenum was measured with 25 nodes scattered across the world, which increases number of messages per batch 40-fold with PBFT/Aardvark and 240-fold with RBFT, and increases latencies - other ledgers often generate batches every 5-10 seconds, but plenum defaults to 1 batch per second, so this is 5-10 more batches per unit of time than with other projects, but significantly decreases latency - other ledgers use very simple test transactions, like incrementing some counter, while in plenum we test with writing cred defs, revocations etc with fees on top of that - many of these things use pretty heavyweight crypto If we conduct a test in a single AWS VPC with 4 nodes, batch per 5 seconds and some dummy transactions I believe we can easily achieve something in a 500 write transactions per second range with current code. read performance in Indy Ledger is pretty good, this is more than 5000 per sec on 25 nodes pool, and this is easily scalable with Observer Nodes (which are almost done BTW).

esplinr (Wed, 19 Jan 2022 16:39:43 GMT):
That's the standard response I send every time I've got this question since 2018. We haven't done detailed performance testing since August of 2019.

esplinr (Wed, 19 Jan 2022 16:39:43 GMT):
^ That's the standard response I send every time I've got this question since 2018. We haven't done detailed performance testing since August of 2019.

esplinr (Wed, 19 Jan 2022 16:40:03 GMT):
There are tickets with results in the old Hyperledger Jira.

esplinr (Wed, 19 Jan 2022 16:40:03 GMT):
There are tickets with charts of results in the old Hyperledger Jira.

esplinr (Wed, 19 Jan 2022 16:42:20 GMT):
That's exactly what I'm describing. Consensus can't complete, so writes can not be processed. But any node can provide a trustworthy result signed with the state proof from before the attack.

trustcerts (Wed, 19 Jan 2022 17:02:16 GMT):
I am trying to build a poc and will validate if it only slows down the network for a short time or if the attack can hold on longer. Results will be shared here :)

Gluwa_DavidLebee (Wed, 19 Jan 2022 17:42:11 GMT):
Has joined the channel.

ManishReddyN (Wed, 19 Jan 2022 18:03:16 GMT):
Has joined the channel.

norbertrogers (Wed, 19 Jan 2022 18:19:01 GMT):
Has joined the channel.

norbertrogers (Wed, 19 Jan 2022 18:35:16 GMT):
Im a uni student doing a project on verifiable DID. i need abit of guidance on where to start. is there like a walkthrough for indy to setting up 3 trust anchors and an api to verify a users ID. im new to this

mgdotonl (Thu, 20 Jan 2022 01:20:36 GMT):
Has joined the channel.

fl0x (Thu, 20 Jan 2022 06:32:13 GMT):
Has joined the channel.

conanoc (Thu, 20 Jan 2022 07:09:42 GMT):
I would recommend these two for you: https://www.slideshare.net/ssuser3993f3/hyperledger-indy-tutorial https://github.com/hyperledger/indy-sdk/blob/master/docs/getting-started/indy-walkthrough.md

sung-il (Thu, 20 Jan 2022 07:35:18 GMT):
Has joined the channel.

daohan (Thu, 20 Jan 2022 07:59:47 GMT):
Has joined the channel.

fsatsuki (Thu, 20 Jan 2022 08:00:19 GMT):
Has joined the channel.

guodadada (Thu, 20 Jan 2022 08:41:12 GMT):
Has joined the channel.

ESiat (Thu, 20 Jan 2022 09:03:41 GMT):
Has joined the channel.

SebastianKe (Thu, 20 Jan 2022 09:53:18 GMT):
Has joined the channel.

prasanthdotv (Thu, 20 Jan 2022 11:17:16 GMT):
Has joined the channel.

ShubhamShukla31752 (Thu, 20 Jan 2022 13:07:58 GMT):
Has joined the channel.

JNeedham (Thu, 20 Jan 2022 13:26:24 GMT):
Quick question -- all the links in https://hyperledger-indy.readthedocs.io/projects/sdk/en/latest/docs/how-tos/write-did-and-query-verkey/README.html seem to be empty. Is there a plan to release the code for these examples?

WadeBarnes (Thu, 20 Jan 2022 15:43:26 GMT):
https://chat.hyperledger.org/channel/indy-sdk?msg=K89qBSCxXr72RMSey

ScottHarris (Thu, 20 Jan 2022 15:51:38 GMT):
Has joined the channel.

adipata (Thu, 20 Jan 2022 15:55:27 GMT):
Has joined the channel.

ffuentes (Thu, 20 Jan 2022 16:00:49 GMT):
Has joined the channel.

mks4343 (Thu, 20 Jan 2022 16:02:24 GMT):
Has joined the channel.

shehanp72 (Thu, 20 Jan 2022 16:02:59 GMT):
Has joined the channel.

singhhp1069 (Thu, 20 Jan 2022 16:04:33 GMT):
Has joined the channel.

paulyp (Thu, 20 Jan 2022 16:05:00 GMT):
Has joined the channel.

rz (Thu, 20 Jan 2022 16:06:26 GMT):
Has joined the channel.

awuttke (Thu, 20 Jan 2022 16:07:46 GMT):
Has joined the channel.

Torben.Zick (Thu, 20 Jan 2022 16:14:57 GMT):
Has joined the channel.

malfaris (Thu, 20 Jan 2022 16:17:17 GMT):
Has joined the channel.

danielhardman (Thu, 20 Jan 2022 16:20:43 GMT):
Call to Participate - DIDComm User Group Now that the DIDComm v2 spec is nearing completion, and there are robust libraries in multiple programming languages, we are starting a user group to share learnings as we put DIDComm into production. We will organize community resources, produce a handbook, foster application-level protocol creation, maintain the didcomm.org website and repo, and recommend best practices. We invite anyone who's interested to register in the user group's email list or discord channel: Discord Invite: https://discord.gg/F4Qw7h6Sr9 Mailing list: https://lists.identity.foundation/g/didcomm-usergroup The first regular meeting will be held in January, with meeting selection to happen prior. Please fill out our survey: http://whenisgood.net/d8dkhgi Although this UG is sponsored by DIF, this is not an IP-protected context. We won't be developing specs or working on standards. Thus, you don't have to be a member of any particular org to join, and there are no legal terms to review or paperwork to sign. Please join us!

bp99 (Thu, 20 Jan 2022 16:33:15 GMT):
Has joined the channel.

sanjikumar (Thu, 20 Jan 2022 16:39:22 GMT):
Has joined the channel.

aymanejabrane (Thu, 20 Jan 2022 17:04:32 GMT):
Has joined the channel.

CharlesDS (Thu, 20 Jan 2022 17:29:39 GMT):
Has joined the channel.

esplinr (Thu, 20 Jan 2022 18:46:52 GMT):
The trick is following the leader through the consensus view changes.

jackiepeters (Fri, 21 Jan 2022 02:02:21 GMT):
Has joined the channel.

shashwot2 (Fri, 21 Jan 2022 04:55:58 GMT):
Has joined the channel.

GiorgosSiachamis (Fri, 21 Jan 2022 08:05:19 GMT):
Has joined the channel.

AnanayNarang (Fri, 21 Jan 2022 08:10:42 GMT):
Has joined the channel.

VNCheck (Fri, 21 Jan 2022 08:19:23 GMT):
Has joined the channel.

MinMail (Fri, 21 Jan 2022 09:32:24 GMT):
Has joined the channel.

JonasKreusch (Fri, 21 Jan 2022 12:02:43 GMT):
Has joined the channel.

Nischitha (Fri, 21 Jan 2022 13:22:25 GMT):
Has joined the channel.

RodrigoGarcia1 (Fri, 21 Jan 2022 14:18:10 GMT):
Has joined the channel.

Tipu_Singh (Fri, 21 Jan 2022 14:23:00 GMT):
hello ,Everyone I'm trying to create a wallet using(Indy-SDK) on local pool running on docker ,parallel connected the pool using (indy-cli). i think i created the wallet using (indy-sdk) but it's not showing the same wallet on (indy-cli) terminal and when i'm trying to create the same wallet using (indy-cli) then it's showing "Wallet already exists".

ianco (Fri, 21 Jan 2022 14:34:10 GMT):
If the wallet already exists then you need to "attach" it to the cli. Within the cli prompt, you can type "wallet help" or "wallet attach help" to see the options

ianco (Fri, 21 Jan 2022 14:34:19 GMT):
Once you "attach" the wallet then you can open it

mpg (Fri, 21 Jan 2022 14:40:27 GMT):
Has joined the channel.

mpg (Fri, 21 Jan 2022 14:41:11 GMT):
Has left the channel.

Tipu_Singh (Fri, 21 Jan 2022 14:42:43 GMT):
Solved :grinning: thanks

Gag 131 (Fri, 21 Jan 2022 17:07:25 GMT):
Has joined the channel.

angmunoz (Fri, 21 Jan 2022 19:02:22 GMT):
Has joined the channel.

ramkapoor (Fri, 21 Jan 2022 19:47:14 GMT):
Has joined the channel.

ebaldivias (Fri, 21 Jan 2022 21:16:05 GMT):
Has joined the channel.

Vamsikn (Fri, 21 Jan 2022 21:26:43 GMT):
Has joined the channel.

himanisingla (Fri, 21 Jan 2022 21:29:50 GMT):
Has joined the channel.

mgamal (Sat, 22 Jan 2022 09:42:06 GMT):
Has joined the channel.

pratikc (Sun, 23 Jan 2022 13:46:53 GMT):
Has joined the channel.

rtatton (Sun, 23 Jan 2022 15:58:22 GMT):
Has joined the channel.

dave-beck (Sun, 23 Jan 2022 17:17:47 GMT):
Has joined the channel.

anilraj27 (Sun, 23 Jan 2022 20:09:17 GMT):
Has joined the channel.

bweymouth (Mon, 24 Jan 2022 01:53:56 GMT):
Has joined the channel.

jaymhc (Mon, 24 Jan 2022 03:46:27 GMT):
Has joined the channel.

ataberkozek (Mon, 24 Jan 2022 06:44:31 GMT):
Hi Everyone, I want to deploy indy-node to OpenShift but indy-client and indy-agent pods are missing their start.sh scripts. Is there a branch of indy-node that i can use to deploy indy-node to OpenShift?

jjlee (Mon, 24 Jan 2022 07:22:44 GMT):
Has joined the channel.

knagware9 (Mon, 24 Jan 2022 07:24:28 GMT):
GReat Thanks for the insights, this looks nice

knagware9 (Mon, 24 Jan 2022 07:24:29 GMT):
Issuance doesn't require any write with each credential--only a write of the schema, cred def, and revocation registry that they all use. So you could issue 1000 or 10M credentials, and you still only have 3 ledger writes and no ledger reads (except maybe to populate a cache)

jechae (Mon, 24 Jan 2022 08:04:13 GMT):
Has joined the channel.

tdiesler (Mon, 24 Jan 2022 09:50:38 GMT):
Folks, when I create TRUST_ANCOR in the cli like this ... ``` wallet open Government key=government_wallet_key did new seed=000000000000000000000Government1 metadata="Government" did use 5iWVNb56iqMoGwnY62HB1j ledger nym did=5iWVNb56iqMoGwnY62HB1j verkey=~5Zyg8vCV5qQfdfRwQk8w94 role=TRUST_ANCHOR endorser=Th7MpTaRZVRYnPiabds81Y wallet close ``` it is not stored in the ledger. In Python I see that there is a reference to the STEWARD required. How do I do that in the cli?

norbertrogers (Mon, 24 Jan 2022 10:20:13 GMT):
thanks all

tdiesler (Mon, 24 Jan 2022 10:44:19 GMT):
In batch scripts I see commands with leading dash

tdiesler (Mon, 24 Jan 2022 10:44:19 GMT):
In batch scripts I see commands with leading dash ... ``` - ledger get-nym did=23LxaWGVajU7vMn7zcrHGK ``` How is this different from not using a dash?

WadeBarnes (Mon, 24 Jan 2022 14:04:17 GMT):
The `TRUST_ANCHOR` role has been renamed to `ENDORSER`. Also it looks like you're attempting to write the same DID you are using. You'll need to write the DID using a privileged DID that is already on the ledger. Are you connecting to a ledger? Which one? The error messages from the commands should give some hints.

WadeBarnes (Mon, 24 Jan 2022 14:05:53 GMT):
In the batch scripts prefixing a command with `-` tells the cli to continue even if the command fails, otherwise the batch process will stop on a failed command.

tdiesler (Mon, 24 Jan 2022 14:06:18 GMT):
great thanks

WadeBarnes (Mon, 24 Jan 2022 14:07:34 GMT):
What code are you looking at now?

WadeBarnes (Mon, 24 Jan 2022 14:10:49 GMT):
Here are a couple projects you may want to check out: - https://github.com/hyperledger/bevel - https://github.com/IDunion/indy-node-container

tdiesler (Mon, 24 Jan 2022 14:17:36 GMT):
Thanks. I think I figured it out https://github.com/tdiesler/nessus-indy/blob/ghi4/etc/getting-started-cli.batch The error msg when the nym already exists in the ledger is a little confusing ``` Transaction has been rejected: STEWARD can not touch role field since only the owner can modify it ``` I won't complain, because I'm just starting with this ;-)

WadeBarnes (Mon, 24 Jan 2022 14:18:21 GMT):
The IP addressing (what indy-node wants) and the URI addressing (what OpenShift wants) can be a little tricky. Also the ephemeral nature of containers running on OpenShift cause issues with the automated network upgrade process of indy-node. When a network upgrade command is issued, the nodes will perform the scheduled upgrade as requested, however this will not persist a rollout or restart, and the node running on OCP will revert to the software versions built into the image. This can cause consensus issues, especially if there is a breaking change. It can also pose a security risk if the upgrade was meant to resolve a security issue.

WadeBarnes (Mon, 24 Jan 2022 14:18:21 GMT):
The IP addressing (what indy-node wants) and the URI addressing (what OpenShift wants) can be a little tricky. Also the ephemeral nature of containers running on OpenShift cause issues with the automated network upgrade process of indy-node. When a network upgrade command is issued, the nodes will perform the scheduled upgrade as requested, however this will not persist a rollout or restart, and the node running on OCP will revert to the software versions built into the image. This can cause consensus issues, especially if there is a breaking change between releases. It can also pose a security risk if the upgrade was meant to resolve a security issue.

pbusch42 (Mon, 24 Jan 2022 16:50:12 GMT):
Has joined the channel.

WadeBarnes (Mon, 24 Jan 2022 17:54:49 GMT):
Virtual development environments have been integrated into the `ubuntu-20.04-upgrade` branches of [hyperledger/indy-node](https://github.com/hyperledger/indy-node/tree/ubuntu-20.04-upgrade) and [hyperledger/indy-plenum](https://github.com/hyperledger/indy-plenum/tree/ubuntu-20.04-upgrade). You can now spin up a fully functional development environment in your choice of a Visual Studio Code Devcontainer or a Gitpod. Thanks to @pSchlarb for his work on getting this done. Happy Coding!

ftrsau (Tue, 25 Jan 2022 03:15:59 GMT):
Has joined the channel.

errorfourten (Tue, 25 Jan 2022 03:25:47 GMT):
Has joined the channel.

ravip.singh (Tue, 25 Jan 2022 06:29:02 GMT):
Has joined the channel.

ataberkozek (Tue, 25 Jan 2022 07:18:25 GMT):
I was using: https://github.com/hyperledger/indy-node repo. OpenShift scripts are under environment/openshift. Do you have any repo recommendations (aside from bevel and indy-node-container) for prod environment for indy?

aj3 (Tue, 25 Jan 2022 09:06:33 GMT):
Has joined the channel.

tdiesler (Tue, 25 Jan 2022 10:57:09 GMT):
Folks, in indi-cli> ledger cred-def I see ... ``` ledger cred-def schema_id=1 signature_type=CL tag=1 primary={"n":"1","s":"2","rms":"3","r":{"age":"4","name":"5"},"rctxt":"6","z":"7"} ``` Where can I learn more about the primary key? Alle these `{"n":"1","s":"2","r":{"age":"4","name":"5"},"rctxt":"6","z":"7"}` seem to be required and `"rms":"3"` seems to be optional. What do these keys mean and how do they relate to the schema?

tdiesler (Tue, 25 Jan 2022 10:57:09 GMT):
Folks, in indi-cli> ledger cred-def I see ... ``` ledger cred-def schema_id=1 signature_type=CL tag=1 primary={"n":"1","s":"2","rms":"3","r":{"age":"4","name":"5"},"rctxt":"6","z":"7"} ``` Where can I learn more about the primary key? All these `{"n":"1","s":"2","r":{"age":"4","name":"5"},"rctxt":"6","z":"7"}` seem to be required and `"rms":"3"` seems to be optional. What do these keys mean and how do they relate to the schema?

tdiesler (Tue, 25 Jan 2022 10:57:09 GMT):
Folks, in indi-cli> ledger cred-def I see ... ``` ledger cred-def schema_id=1 signature_type=CL tag=1 primary={"n":"1","s":"2","rms":"3","r":{"age":"4","name":"5"},"rctxt":"6","z":"7"} ``` Where can I learn more about the primary key? All these `{"n":"1","s":"2","r":{"age":"4","name":"5"},"rctxt":"6","z":"7"}` seem to be required and `"rms":"3"` seems to be optional. What do these keys mean and how do they relate to the schema? https://github.com/hyperledger/indy-sdk/issues/2474

WadeBarnes (Tue, 25 Jan 2022 13:32:46 GMT):
I've linked to some sample code that should help you follow the process better.

tdiesler (Tue, 25 Jan 2022 13:48:23 GMT):
:sunflower:

WadeBarnes (Tue, 25 Jan 2022 13:48:57 GMT):
Those scripts were designed more for demonstrating things in a local OpenShift cluster where all the moving parts are running within the same OpenShift namespace. The scripts dynamically configure the nodes with the IP addresses assigned by OpenShift so everything can communicate. This does not translate very well to a production environment where the nodes in OpenShift need to communicate with the outside world. lol, I'm just now noticing this is based on work I did back in 2017.

WadeBarnes (Tue, 25 Jan 2022 13:54:08 GMT):
We never did finish automating the process, and we were running into the issue with exposing the nodes to the outside world via static IP addresses from OpenShift, which was overly tricky and complicated (at the time at least).

WadeBarnes (Tue, 25 Jan 2022 14:00:26 GMT):
At the time we were looking for a way to host a pool of nodes on OpenShift for test purposes. We ended up building out and deploying instances of [bcgov/von-network](https://github.com/bcgov/von-network) on Digital Ocean instead. For example; http://test.bcovrin.vonx.io/

WadeBarnes (Tue, 25 Jan 2022 14:06:25 GMT):
If you want to go the route of a containerized node instance the two links I provided are the best resources at the moment. If you want to go the more traditional route there is documentation here on how to setup a new network; https://github.com/hyperledger/indy-node/blob/master/docs/source/NewNetwork/NewNetwork.md You can also find Terraform and Ansible scripts here for setting up a node in AWS; https://github.com/CQEN-QDCE/Candy. Speifically: - https://github.com/CQEN-QDCE/Candy/tree/main/terraform/aws - https://github.com/CQEN-QDCE/Candy/tree/main/ansible/indy_node

tdiesler (Tue, 25 Jan 2022 15:20:50 GMT):
It seems that this primary key is generated by some higher level function. How would I generate/know these values if limited to indy-cli? The meaning of n, s, r, rctxt, z, rms is not document AFAICS. ``` {'id': 'DFuDqCYpeDNXLuc3MKooX3:3:CL:1', 'schemaId': '1', 'tag': 'tag', 'type': 'CL', 'value': {'primary': {'n': '109330248353590897319389070001241417443720503932971588287006464922388229178334793336510417955875881059871543804578087088276791800083305615121644511248337970131893077241681758874562815652760517127419781690778951536164836285458183775012778319181918711231775022070055487033667781997555780000964033080223070801890480454130338986221023468272183501187404600365439161142989665555780971754563754309510943630872371340817447612828969885782344790492881494591201224675720837407955083191745600556967629527217606100951074519486650521288615897758074209367922751727352657076430733961205538551220485786837438600982766853425186252644017', 'r': {'age': '73308673216472493575090598174963399948529629134006496210571752239581655779012920152879390449719574107939536194125393138485840251400105908560965343195463986660046396898151291844591150186299279740115095392914394151816733033421421781471599302173575425800443185010936709144943232555123378965813972641717797624815066173821197309970706537526529635178157107747878761192927131531306789978613595162689363565628723311165111774867707907441698784714369233660014225193452769768361969161868978698182387153720193894374602878251447258528578841422721713848011884183426934563483417227006937242841853827004235408539062962437389057557466', 'master_secret': '68361484436365673061072041624647294914425855902623496297375686496312847852294891269292033386069491663411635310339094263142698298689132782329339987266934634535728272777489921087755693951033987277703390817896541076260102142585504616782084149039016192790475658619096435661926936093335858078539362316056330917248469510038667267988488591837331636897350889581870219495748565872958952115592906821121236209205411654897745071778422593295566925294862162049245508656755995167332502259192551948308838679613532706595396718914348820966604241737677454847945523997559921590852327709685893644680377213866453241491125951811592264674385', 'name': '47956700491301067281258647205353751258983285450131303607847093059046267200221967853146726991503228245603799976265038836926279273756505065304549666555819258018152979494728703569302125080586313807600151910235972765806376471411615253734993567102101608092977530540217931683348869413718435281288986248345335344018542983222736766375500635662856035064726147913925488526522192025619312494021637149847804815197911363884756275007799195202034451124167367439201703744060644702991961222569708991445106308180298048676620063203473029844891286050419366140751942222786439647012450277116627707784993011702899356860514634641379792471881'}, 'rctxt': '85822465136763484750224903111930923618937001868097822592103216386643211499337898367784057120826066430186818827320375097357041704110495665025997489429045658165162147907795204346206877770526428837895201878621908877672713195000392657196945094035963718417455273866302452201821774685040248478392362268576673532904049764968773578332077261408115307685949804917889275660854872818155274730122917471074855046658920937068414071937936393976713715659099504318383054607840309655173598188730099851335185640343544202013064216095459208905638997738029996011970156760774708328892008498934772831156833716387364110926733423726356601343689', 's': '106345188679199089125597322701414631094937145030314613341378947140738361187874138161772329427608423907567247428077450395652493629986745542580472189756730607466150103643195464371742896667685304340461985111222457560719775350966587953014029845146432673104558897228576416276655693758576076167730137327545172086773323375530896947049176540552702297043676954245462417720544877461007526007923822646362581648140569373776087100702971388864141862849461540636711884781495543149191950953666057015608051814287517402916176848788512328682250948222207557803534778301129658621242780466275103890365613475979127357941024766695451697885176', 'z': '47107036540803311469435904351406772134330713253387217487860154283622977007467454258519685096545379994908750889467439958454982798929251155494690304601387525503696682822022729189520351495627997108947442055176478587213869104400238227223946797194814057365368583932995967095555966182151943614690693513901167896619168429109522378359328460889950835611914690332686684663213810934909998666094622585987936453973281713802447375961016451674180164951212112576812042153386412793986139385065180612574606273328511899790711670508882466261670565204415908876341282665166564296557762007159015012163706719247955095314823816627774810936820'}}, 'ver': '1.0'} ```

tdiesler (Tue, 25 Jan 2022 15:20:50 GMT):
It seems that this primary key is generated by some higher level function. How would I generate/know these values if limited to indy-cli? The meaning of n, s, r, rctxt, z, rms is not document AFAICS. ``` {'id': 'DFuDqCYpeDNXLuc3MKooX3:3:CL:1', 'schemaId': '1', 'tag': 'tag', 'type': 'CL', 'value': {'primary': {'n': '109330..4017', 'r': {'age': '733086..7466', 'master_secret': '68361..4385', 'name': '479567..1881'}, 'rctxt': '85822..43689', 's': '10634..85176', 'z': '47107..6820'}}, 'ver': '1.0'} ```

tdiesler (Tue, 25 Jan 2022 15:23:36 GMT):
Perhaps `ledger cred-def` requires some involved pre-steps to obtain the primary key?

WadeBarnes (Tue, 25 Jan 2022 15:33:24 GMT):
If you were limited to the `indy-cli` you would not be able to generate the cred-def. That's why the code I pointed out in `von-network`'s containerized `indy-cli` exists.

tdiesler (Tue, 25 Jan 2022 15:37:23 GMT):
Ah ok, I thought it would be possible to use indy-cli alone to step through getting started. Ultimately, I need to use Java anyway. Perhaps its best to start with that right away and resort to cli when needed. WDYT?

WadeBarnes (Tue, 25 Jan 2022 15:37:50 GMT):
I agree.

WadeBarnes (Tue, 25 Jan 2022 15:37:50 GMT):
I agree. Go the code route.

tdiesler (Tue, 25 Jan 2022 15:38:40 GMT):
This does a similar thing, right? https://github.com/hyperledger/indy-sdk/blob/master/samples/java/src/main/java/Anoncreds.java

WadeBarnes (Tue, 25 Jan 2022 15:41:11 GMT):
Yes. The cred-def is generated here; https://github.com/hyperledger/indy-sdk/blob/master/samples/java/src/main/java/Anoncreds.java#L50-L56

tdiesler (Tue, 25 Jan 2022 15:42:30 GMT):
Ok thanks again. Hopefully it won't be too long before I become productive

ataberkozek (Wed, 26 Jan 2022 08:59:46 GMT):
Thanks for the reply! This is going to help me very much.

aj3 (Wed, 26 Jan 2022 10:13:26 GMT):
Question: Rate limits Our use case involves building an agency on top of indy, via the java wrapper. I'm busy testing the throughput, and the test procedure involves creating a new wallet, storing a did in it, and making a nym request to write that DID to the ledger. Running the code in a container, and sending 100 requests, 10 concurrently at a time, maxes out at about 1 request per second. This is true for the default test network and for the sovrin builder net. Ive been narrowing down the issue to either being with the harddrive IO, or a limit on the indy side. Are there any rate limits on the indy side that could account for this?

aj3 (Wed, 26 Jan 2022 10:13:26 GMT):
*Question: Rate limits* Our use case involves building an agency on top of indy, via the java wrapper. I'm busy testing the throughput, and the test procedure involves creating a new wallet, storing a did in it, and making a nym request to write that DID to the ledger. Running the code in a container, and sending 100 requests, 10 concurrently at a time, maxes out at about 1 request per second. This is true for the default test network and for the sovrin builder net. Ive been narrowing down the issue to either being with the harddrive IO, or a limit on the indy side. Are there any rate limits on the indy side that could account for this?

WadeBarnes (Wed, 26 Jan 2022 13:09:44 GMT):
You are running into a limit in the indy-sdk. I can't recall the exact technical details, but many operations all go through a central blocking queue before being processed. @andrew.whitehead or @swcurran may be able to elaborate.

AravindVoruganti (Wed, 26 Jan 2022 14:16:16 GMT):
Has joined the channel.

andrew.whitehead (Wed, 26 Jan 2022 17:15:20 GMT):
Creating new wallets could have an additional delay as it derives the wallet key, depending on the configuration

mccown (Wed, 26 Jan 2022 22:05:01 GMT):
The Identity Implementer’s WG meeting for tomorrow (Jan 27th) has been cancelled. Please join us in our next meeting on Feb 10th (9am MT / 4pm UTC).

sandeep20 (Thu, 27 Jan 2022 04:32:36 GMT):
Has joined the channel.

Alen_Sebastian (Thu, 27 Jan 2022 09:14:31 GMT):
Has joined the channel.

winterking (Thu, 27 Jan 2022 09:50:12 GMT):
Has joined the channel.

StefanBratanov (Thu, 27 Jan 2022 09:52:26 GMT):
Has joined the channel.

rejnol93 (Thu, 27 Jan 2022 10:01:21 GMT):
Has joined the channel.

isavelli (Thu, 27 Jan 2022 14:27:07 GMT):
Has joined the channel.

danielhardman (Thu, 27 Jan 2022 18:05:26 GMT):
Re. the call to participate in a new DIDComm Users Group that I posted last week: We had a nice set of respondents to this survey. Unfortunately, the general consensus was that it is impossible to get us together! Almost every proposed timeslot had at least ¾ of the respondents saying they couldn’t come. We are busy! I’m going to try again, but with the following changes: * I’ve extended the calendar time until Feb 11, in case the early parts of a month are better for us than the end parts. * I’ve decided to look for 30 minute slots instead of 60. I think this might be better – the commitment to attend is less heavy, and perhaps calendars will be a little more flexible. I believe we can still get a lot done in 30 minutes, if we’re well organized. So, can we try again? http://whenisgood.net/d8dkhgi

knibals (Thu, 27 Jan 2022 20:13:27 GMT):
Has joined the channel.

dci (Thu, 27 Jan 2022 20:36:06 GMT):
Has joined the channel.

alamin25 1 (Thu, 27 Jan 2022 20:53:14 GMT):
Has joined the channel.

LunarM 1 (Thu, 27 Jan 2022 21:44:15 GMT):
Has joined the channel.

brettd (Fri, 28 Jan 2022 01:21:56 GMT):
Has joined the channel.

heysatya (Fri, 28 Jan 2022 07:01:53 GMT):
Has joined the channel.

Avijeet_15 (Fri, 28 Jan 2022 13:35:22 GMT):
Has joined the channel.

romanr (Fri, 28 Jan 2022 14:51:38 GMT):
Has joined the channel.

mzago (Fri, 28 Jan 2022 16:18:56 GMT):
Hello, I'm looking for technical documentation on key derivation in the context of Indy Wallets and Indy Agents communication. Specifically, I would like to fully understand the key derivation processes from the provided seed to all the various elements such as the DIDs, the verkeys, etc. Could you please share some references in such regard?

mzago (Fri, 28 Jan 2022 16:19:24 GMT):
I don't mind technical published papers

mzago (Fri, 28 Jan 2022 16:19:24 GMT):
I don't mind technical published papers, on the contrary, that would be of great help

swcurran (Fri, 28 Jan 2022 16:26:23 GMT):
@andrew.whitehead ^^^^

rejnol93 (Sat, 29 Jan 2022 11:26:59 GMT):
Has left the channel.

Trisham_Bharat_Patil (Sat, 29 Jan 2022 16:57:36 GMT):
Has joined the channel.

aminst (Sat, 29 Jan 2022 19:03:00 GMT):
Has joined the channel.

rajan-31 (Sat, 29 Jan 2022 19:20:18 GMT):
Has joined the channel.

cxk934 (Sun, 30 Jan 2022 13:42:46 GMT):
Has joined the channel.

Junior-nitcheu (Mon, 31 Jan 2022 11:36:29 GMT):
Has joined the channel.

rameshmiriyala (Mon, 31 Jan 2022 12:45:56 GMT):
Has joined the channel.

hamoT (Mon, 31 Jan 2022 15:25:05 GMT):
Has joined the channel.

sudarsan.immadi (Tue, 01 Feb 2022 01:30:46 GMT):
Has joined the channel.

SobhikaR (Tue, 01 Feb 2022 08:59:55 GMT):
Has joined the channel.

tdiesler (Tue, 01 Feb 2022 10:43:02 GMT):
In Java, I can create/store a Steward DID like this ... ``` String stewardSeed = new JSONObject().put("seed", "000000000000000000000000Steward1").toString(); CreateAndStoreMyDidResult didResult = Did.createAndStoreMyDid(ctx.stewardWallet, stewardSeed).get(); ctx.stewardDid = didResult.getDid(); ctx.stewardVkey = didResult.getVerkey(); logInfo("DID Steward: did={}, vkey={}", ctx.stewardDid, ctx.stewardVkey); String nymRequest = Ledger.buildNymRequest(ctx.stewardDid, ctx.stewardDid, ctx.stewardVkey, null, ROLE_STEWARD).get(); Ledger.signAndSubmitRequest(ctx.pool, ctx.stewardWallet, ctx.stewardDid, nymRequest).get(); ``` and then create and later get a Schema like this ... ``` String issuerDid = ctx.stewardDid; Wallet issuerWallet = ctx.stewardWallet; IssuerCreateSchemaResult schemaResult = Anoncreds.issuerCreateSchema(issuerDid, "Transcript", "1.2", new JSONArray(Arrays.asList("first_name","last_name","degree","status","year","average","ssn")).toString()).get(); logInfo(schemaResult.toString()); Ledger.buildSchemaRequest(issuerDid, schemaResult.getSchemaJson()).get(); ctx.transcriptSchemaId = schemaResult.getSchemaId(); String schemaRequest = Ledger.buildSchemaRequest(issuerDid, schemaResult.getSchemaJson()).get(); Ledger.signAndSubmitRequest(ctx.pool, issuerWallet, issuerDid, schemaRequest).get(); String getSchemaRequest = Ledger.buildGetSchemaRequest(ctx.faberDid, ctx.transcriptSchemaId).get(); String getSchemaResponse = Ledger.submitRequest(ctx.pool, getSchemaRequest).get(); ParseResponseResult parseSchemaResult = Ledger.parseGetSchemaResponse(getSchemaResponse).get(); logInfo(parseSchemaResult.toString()); ``` With all else equal (fresh pool nodes, removed indy-client state), this does no longer work, with a seed other than `000000000000000000000000Steward1`. ``` java.util.concurrent.ExecutionException: org.hyperledger.indy.sdk.pool.LedgerNotFoundException: Item not found on ledger exception. at java.util.concurrent.CompletableFuture.reportGet(CompletableFuture.java:357) at java.util.concurrent.CompletableFuture.get(CompletableFuture.java:1908) at io.nessus.indy.test.SchemaTest.createTranscriptSchema(SchemaTest.java:221) Caused by: org.hyperledger.indy.sdk.pool.LedgerNotFoundException: Item not found on ledger exception. at org.hyperledger.indy.sdk.IndyException.fromSdkError(IndyException.java:168) at org.hyperledger.indy.sdk.IndyJava$API.checkResult(IndyJava.java:92) ``` For example, if I change the seed to `000000000000000000000Government1` I get the above exception. is this perhaps because of some stale state that I'm not aware of?

MathiasBaumgartinger (Tue, 01 Feb 2022 10:53:02 GMT):
Has joined the channel.

etschelp (Tue, 01 Feb 2022 11:28:52 GMT):
Has joined the channel.

WadeBarnes (Tue, 01 Feb 2022 12:58:51 GMT):
@esplinr, It would be nice to get this info into the indy-node repo.

Biradi (Tue, 01 Feb 2022 13:01:56 GMT):
Has joined the channel.

WadeBarnes (Tue, 01 Feb 2022 13:07:31 GMT):
@tdiesler, What ledger are you using for your testing? Does the DID fo `000000000000000000000000Steward1` already exist on the ledger before the code is run? i.e. Is that DID setup when the pool is first created? Has the DID for `000000000000000000000Government1` been added to the ledger with endorser privileges?

WadeBarnes (Tue, 01 Feb 2022 13:07:31 GMT):
@tdiesler, What ledger are you using for your testing? Does the DID for `000000000000000000000000Steward1` already exist on the ledger before the code is run? i.e. Is that DID setup when the pool is first created? Has the DID for `000000000000000000000Government1` been added to the ledger with endorser privileges before you try using it to write the schema?

tdiesler (Tue, 01 Feb 2022 13:45:13 GMT):
Hi Wade, its this code here (https://github.com/tdiesler/nessus-indy/blob/main/itests/samples/src/test/java/io/nessus/indy/test/SchemaTest.java#L91) The ledger is empty, so is the indy client state. I'm currently investigate whether this could be a timing issue. i.e. the get schema request may not be able to get the schema just yet Is that possible?

tdiesler (Tue, 01 Feb 2022 13:49:16 GMT):
There is PoolUtils.ensurePreviousRequestApplied(...) - maybe that's it.

WadeBarnes (Tue, 01 Feb 2022 14:01:02 GMT):
What ledger are you using? A functional ledger is always going to have some information written to it, and typically that includes a set of Steward DIDs and a Trustee DID. Any other DIDs written to the ledger need to be "endorsed" by one or more of the foundational DIDs.

WadeBarnes (Tue, 01 Feb 2022 14:01:29 GMT):
Do you know about https://github.com/bcgov/von-network?

yusakrabin (Tue, 01 Feb 2022 14:09:36 GMT):
Has joined the channel.

tdiesler (Tue, 01 Feb 2022 14:23:17 GMT):
I'm creating the ledger with genesis Txs from PoolUtils. So far, I assumed that these nodes are not connected to a wider network and that Steward does not exist when the network is started. Let me verify that ...

tdiesler (Tue, 01 Feb 2022 14:23:40 GMT):
Thanks for that pointer I'll look into it

WadeBarnes (Tue, 01 Feb 2022 14:44:26 GMT):
You're currently using this method to run a pool; correct? ``` docker build -f ci/indy-pool.dockerfile -t indy_pool ci docker run -it -p 9701-9708:9701-9708 indy_pool ```

WadeBarnes (Tue, 01 Feb 2022 14:45:50 GMT):
I'd recommend running a `von-network` instance. It makes the transactions on the ledger far more visible.

WadeBarnes (Tue, 01 Feb 2022 14:46:29 GMT):
You'd have better insight into what "default" transactions exist on the ledger.

dovidkopel (Tue, 01 Feb 2022 14:48:39 GMT):
Has joined the channel.

tdiesler (Tue, 01 Feb 2022 14:59:16 GMT):
Yes, that's what I'm using. This code here tries to register 9 Steward DIDs. It works for Steward1-4 and then fails for Steward5-9 ``` for (int i = 1; i < 10; i++) { String stewardSeed = new JSONObject().put("seed", "000000000000000000000000Steward" + i).toString(); CreateAndStoreMyDidResult didResult = Did.createAndStoreMyDid(ctx.stewardWallet, stewardSeed).get(); ctx.stewardDid = didResult.getDid(); ctx.stewardVkey = didResult.getVerkey(); logInfo("DID Steward{}: did={}, vkey={}", i, ctx.stewardDid, ctx.stewardVkey); String nymRequest = Ledger.buildNymRequest(ctx.stewardDid, ctx.stewardDid, ctx.stewardVkey, null, ROLE_STEWARD).get(); String submitResponse = Ledger.signAndSubmitRequest(ctx.pool, ctx.stewardWallet, ctx.stewardDid, nymRequest).get(); logInfo("SubmitResponse: " + submitResponse); } ``` so it seems that not all of the Stewards are equal and the ledger already knows about the first for DIDs ``` 2022-02-01 15:44:56 INFO [io.nessus.indy.test.SchemaTest] - DID Steward9: did=FiYoKcGabuCtTN8fi9J8JD, vkey=92DLtcgtCreSEUu2ed8pWPcvjxwXXrzAWLq2mWVeS5An 2022-02-01 15:44:56 INFO [org.hyperledger.indy.sdk.LibIndy.nativeindy.services.ledger] - src/services/ledger/mod.rs:58 | build_nym_request() => Ok("{\"reqId\":1643726696732628000,\"identifier\":\"FiYoKcGabuCtTN8fi9J8JD\",\"operation\":{\"type\":\"1\",\"dest\":\"FiYoKcGabuCtTN8fi9J8JD\",\"verkey\":\"92DLtcgtCreSEUu2ed8pWPcvjxwXXrzAWLq2mWVeS5An\",\"role\":\"2\"},\"protocolVersion\":2}") 2022-02-01 15:44:59 INFO [io.nessus.indy.test.SchemaTest] - SubmitResponse: {"reqId":1643726696732628000,"reason":"client request invalid: UnauthorizedClientRequest(\"sender's DID FiYoKcGabuCtTN8fi9J8JD is not found in the Ledger\",)","identifier":"FiYoKcGabuCtTN8fi9J8JD","op":"REJECT"} ```

WadeBarnes (Tue, 01 Feb 2022 15:00:24 GMT):
That's because Steward1-4 already exist as part of the pool.

tdiesler (Tue, 01 Feb 2022 15:00:44 GMT):
Is that hard coded?

WadeBarnes (Tue, 01 Feb 2022 15:01:08 GMT):
Yeah, as part of the setup of the test pool.

tdiesler (Tue, 01 Feb 2022 15:04:56 GMT):
Anything else that already exists (e.g. some TRUST_ANCHOR perhaps)? Is that setup in GitHub somewhere? Is it still correct to assume that I run a local ledger, not connected to wider (external) network?

WadeBarnes (Tue, 01 Feb 2022 15:05:37 GMT):
Yes a Trustee (previously a TRUST_ANCHOR) also exists.

WadeBarnes (Tue, 01 Feb 2022 15:07:35 GMT):
This is the code that sets up the test ledger when you use the method you're using; https://github.com/hyperledger/indy-plenum/blob/master/plenum/common/test_network_setup.py

WadeBarnes (Tue, 01 Feb 2022 15:10:52 GMT):
It looks like there are 5 `Client` DIDs that also get generated automatically for that test ledger.

tdiesler (Tue, 01 Feb 2022 15:12:31 GMT):
Mercy. Starting with one of the Stewards, I should be able to create government, Faber, Acme & Thrift as TRUSTEE, right?

WadeBarnes (Tue, 01 Feb 2022 15:12:33 GMT):
`von-network` generates a cleaner ledger, just the 4 Stewards (one for each node) and a single Trustee, and provides a ledger browser so you can see exactly what is on the ledger.

WadeBarnes (Tue, 01 Feb 2022 15:13:57 GMT):
A Steward is not allowed to create a Trustee; https://github.com/hyperledger/indy-node/blob/master/docs/source/auth_rules.md

WadeBarnes (Tue, 01 Feb 2022 15:14:38 GMT):
You need to sign such a transaction with a Trustee DID.

WadeBarnes (Tue, 01 Feb 2022 15:16:29 GMT):
The DID for the default Trustee on the test ledger you are using should be `000000000000000000000000Trustee1`

tdiesler (Tue, 01 Feb 2022 15:18:33 GMT):
Getting started talks about TRUST_ANCHOR. I read somewhere that ENDORSER replaced TRUST_ANCHOR. Is that right?

WadeBarnes (Tue, 01 Feb 2022 15:19:02 GMT):
TRUSTEE replaced TRUST_ANCHOR

tdiesler (Tue, 01 Feb 2022 15:19:32 GMT):
ok

WadeBarnes (Tue, 01 Feb 2022 15:25:05 GMT):
You may want to have a look at the Aries frameworks, such as https://github.com/hyperledger/aries-cloudagent-python. They can make working with indy ledgers much easier, as they abstract many of the lower level complexities. Though there is not one for Java ATM.

tdiesler (Tue, 01 Feb 2022 15:25:52 GMT):
Do you want me to change the getting started Python workflow, such that it talks about TRUSTEE rather that TRUST_ANCHOR?

WadeBarnes (Tue, 01 Feb 2022 15:26:24 GMT):
That would be great!

MichaelSchaefer (Tue, 01 Feb 2022 15:30:02 GMT):
Has joined the channel.

tdiesler (Tue, 01 Feb 2022 15:31:07 GMT):
Ok. Government would need to have role STEWARD in order to create the Transcript + JobCertificate Schemas. Neither Government, Faber, Acme nor Thrift need to be a TRUSTEE. Role ENDORSER would also not be needed in getting started. Does that sound right?

WadeBarnes (Tue, 01 Feb 2022 15:33:35 GMT):
Stewards operate nodes. Trustees govern the network. These are the highest privileged DIDs. Endorsers are able to write Schemas and Cred_Defs to the ledger, or sign such transactions so they can be written by non-privileged DIDs.

WadeBarnes (Tue, 01 Feb 2022 15:36:07 GMT):
You want to ensure a DID has the least amount of privilege it needs to operate, which in many cases is no privilege, provided the resources it needs are already written to the ledger, either by a privileged DID or by having the txn signed by a privileged DID (e.g. by an Endorser) .

WadeBarnes (Tue, 01 Feb 2022 15:37:23 GMT):
Make sense?

WadeBarnes (Tue, 01 Feb 2022 15:39:11 GMT):
For demo purposes a DID that needs to write to the ledger is given the Endorser roll, however in real life this is not always necessary.

tdiesler (Tue, 01 Feb 2022 15:44:20 GMT):
How about, the preconfigured Trustee onboards Government as Trustee. Then, Government can onboard Faber, Acme and Thrift as Endorser so that they can create their respective Schemas and Cred_Defs. Alice would also need to get onboarded through Government.

WadeBarnes (Tue, 01 Feb 2022 16:16:19 GMT):
That would work.

jacobgorman613 (Tue, 01 Feb 2022 20:57:20 GMT):
Has joined the channel.

da3v21 (Wed, 02 Feb 2022 04:23:52 GMT):
Hi everyone, can genesis file of a network be updated without losing the network data, such as adding a new genesis node

da3v21 (Wed, 02 Feb 2022 04:23:52 GMT):
Hi everyone, can genesis file of a network be updated without losing the network data, such as adding a new genesis node. ``` What will be status of the network if the genesis nodes are down, but there are sufficient trustees running on the network ```

da3v21 (Wed, 02 Feb 2022 04:23:52 GMT):
Hi everyone, can genesis file of a network be updated without losing the network data, such as adding a new genesis node. What will be status of the network if the genesis nodes are down, but there are sufficient trustees running on the network

BalarajuBalu (Wed, 02 Feb 2022 06:08:42 GMT):
Has joined the channel.

Tipu_Singh (Wed, 02 Feb 2022 07:41:22 GMT):
hello everyone, I have a few basic questions like, in an INDY wallet every DID is different from the other on the entire globe? If yes then which method or algorithm ensures that every wallet has unique DIDs.

chetanambi (Wed, 02 Feb 2022 07:52:31 GMT):
Has joined the channel.

EduOliveira (Wed, 02 Feb 2022 10:40:26 GMT):
Has joined the channel.

GiacomoDisalvo (Wed, 02 Feb 2022 11:01:23 GMT):
Has joined the channel.

devratapuri (Wed, 02 Feb 2022 13:41:17 GMT):
Has joined the channel.

jayzhan (Wed, 02 Feb 2022 14:40:12 GMT):
Has joined the channel.

chetanhanda (Wed, 02 Feb 2022 16:24:09 GMT):
Has joined the channel.

ksl2europe (Wed, 02 Feb 2022 18:16:58 GMT):
Has joined the channel.

mayurthosar11 (Wed, 02 Feb 2022 19:20:43 GMT):
Has joined the channel.

wontondon (Wed, 02 Feb 2022 19:59:39 GMT):
Has joined the channel.

kakaupke (Thu, 03 Feb 2022 00:01:39 GMT):
Has joined the channel.

dioxis (Thu, 03 Feb 2022 01:05:00 GMT):
Has joined the channel.

vsh1982 (Thu, 03 Feb 2022 02:59:35 GMT):
Has joined the channel.

drichard2 (Thu, 03 Feb 2022 10:06:52 GMT):
Has joined the channel.

makikato (Thu, 03 Feb 2022 13:08:31 GMT):
Has joined the channel.

kaissfr (Thu, 03 Feb 2022 13:08:36 GMT):
Has joined the channel.

sohansamant (Thu, 03 Feb 2022 14:39:21 GMT):
Has joined the channel.

devg (Thu, 03 Feb 2022 15:09:54 GMT):
Has joined the channel.

Smalviya (Thu, 03 Feb 2022 15:14:52 GMT):
Has joined the channel.

Smalviya (Thu, 03 Feb 2022 15:16:26 GMT):
Form the prerequisite to the indy workstop -> the last pytest . command under ../indy_node is taking forever (over 30 mins) wihtout any error/warning on ubuntu 18.4 with 30 gigs disk

Siltes (Thu, 03 Feb 2022 15:17:33 GMT):
Has joined the channel.

Siltes (Thu, 03 Feb 2022 15:22:21 GMT):
Had some issues with the last pytest command as well, especially, it seems like it kills itself (error syntax: .py Killed )...

Smalviya (Thu, 03 Feb 2022 15:23:55 GMT):
how long it took for autokill

Siltes (Thu, 03 Feb 2022 15:25:40 GMT):
Depends on the run, sometimes it killed itself when reaching 5% of the tests, other times it reached 7% test completion, another time 71%... Regarding the timing of the individual test, it was being done in the background so I don't have some time measure... :/

Smalviya (Thu, 03 Feb 2022 15:26:14 GMT):

Clipboard - February 3, 2022 8:56 PM

Smalviya (Thu, 03 Feb 2022 15:26:35 GMT):
5% for me from close to an hour

Siltes (Thu, 03 Feb 2022 15:27:48 GMT):

tests.png

Siltes (Thu, 03 Feb 2022 15:27:55 GMT):
Here's my longest run

Siltes (Thu, 03 Feb 2022 15:28:42 GMT):
As you can see it's strange since it seems to take forever at the test_nym_auth_rules.py but it does not kill it

Siltes (Thu, 03 Feb 2022 15:29:15 GMT):
While the second persistence test kills itself kind of immediatly

Siltes (Thu, 03 Feb 2022 15:35:33 GMT):
I will run it with the verbose option, will see what's happening under the hood hopefully

JamesSchulte (Thu, 03 Feb 2022 15:36:49 GMT):
Hi everyone, today's Hyperledger workshop "Hyperledger Indy Deep Dive" will be starting shortly at 8:00am PST. As a reminder, we will be using this channel to answer questions or provide tech support throughout the workshop.

agiove (Thu, 03 Feb 2022 15:57:35 GMT):
Has joined the channel.

lynn.bendixsen (Thu, 03 Feb 2022 15:58:05 GMT):
You will probably need to increase disk space to 50 and maybe need to increase RAM to 3-4G

sbohanlf (Thu, 03 Feb 2022 16:00:03 GMT):
We are livestreaming on Youtube and recording the workshop and will share the video in a follow-up email to participants. You can find the livestream here: https://www.youtube.com/watch?v=9TOkm1H40fs

sohansamant (Thu, 03 Feb 2022 16:03:42 GMT):
Hi, even my test failed after running till 89%

sohansamant (Thu, 03 Feb 2022 16:03:52 GMT):
Error looks something like this..

sohansamant (Thu, 03 Feb 2022 16:03:52 GMT):
-- Docs: https://docs.pytest.org/en/stable/warnings.html ============================================================================================= short test summary info ============================================================================================== FAILED test/pool_restart/test_node_control_tool_for_restart.py::test_communication_with_node_control_tool - OSError: [Errno 98] Address already in use FAILED test/upgrade/test_node_control_tool_performs_migrations.py::testNodeControlPerformsMigrations - OSError: [Errno 98] Address already in use FAILED test/upgrade/test_node_control_tool_processes_invalid_data.py::test_node_control_tool_processes_invalid_json - Failed: DID NOT RAISE FAILED test/upgrade/test_node_control_tool_processes_invalid_data.py::test_node_control_tool_processes_invalid_version[some-pkg-1.2.3.4.5] - Failed: DID NOT RAISE FAILED test/upgrade/test_node_control_tool_processes_invalid_data.py::test_node_control_tool_processes_invalid_version[indy-node-1.2.3.4.5] - Failed: DID NOT RAISE FAILED test/upgrade/test_node_control_tool_removes_backups.py::testNodeControlRemovesBackups - OSError: [Errno 98] Address already in use FAILED test/upgrade/test_node_control_tool_resolves_dep_top_level.py::test_node_as_depend - OSError: [Errno 98] Address already in use ======================================================================= 7 failed, 3030 passed, 57 skipped, 44 warnings in 1481.26s (0:24:41) =======================================================================

lynn.bendixsen (Thu, 03 Feb 2022 16:05:33 GMT):
Success! (7 failed is expected, you are good to go)

sohansamant (Thu, 03 Feb 2022 16:06:55 GMT):
oh ok! Thanks!

grittyronin (Thu, 03 Feb 2022 16:06:59 GMT):
Has joined the channel.

adambabs (Thu, 03 Feb 2022 16:10:23 GMT):
Has joined the channel.

sergm (Thu, 03 Feb 2022 16:11:39 GMT):
Has joined the channel.

kyilintun (Thu, 03 Feb 2022 16:11:50 GMT):
Has joined the channel.

gosankar (Thu, 03 Feb 2022 16:13:26 GMT):
Has joined the channel.

senthil.akshay27 (Thu, 03 Feb 2022 16:15:02 GMT):
Has joined the channel.

mepeltier (Thu, 03 Feb 2022 16:15:06 GMT):
Has joined the channel.

matisalimba (Thu, 03 Feb 2022 16:20:40 GMT):
I'm still struggling a bit to make sense of Indy and Aries. If I say Indy is the DLT itself, the ledger (as fabric would be) and Aries is a set of library/tools to build clients and interfaces to work with Indy, is this a correct understanding?

TelegramSam (Thu, 03 Feb 2022 16:21:30 GMT):
@matisalimba That is a good way to describe the line there.

TelegramSam (Thu, 03 Feb 2022 16:21:55 GMT):
There are tools that Indy has natively (like the Indy CLI), but production/consumer facing tools are built with Aries and other projects.

devin007 (Thu, 03 Feb 2022 16:23:20 GMT):
Has joined the channel.

ngwlf (Thu, 03 Feb 2022 16:23:38 GMT):
Has joined the channel.

sanjichained (Thu, 03 Feb 2022 16:26:36 GMT):
Has joined the channel.

sbohanlf (Thu, 03 Feb 2022 16:29:48 GMT):
from the Youtube stream: "How far would you say Indy and Aries are from being ready for commercial use? I've found it difficult sometimes in the documentation to see"

TelegramSam (Thu, 03 Feb 2022 16:33:30 GMT):
Current production deployments exist using both Indy and Aries.

TelegramSam (Thu, 03 Feb 2022 16:34:23 GMT):
The complexity depends on the nature of the project, and that affects how much development is required on top of the existing codebases.

Siltes (Thu, 03 Feb 2022 16:51:12 GMT):
What's the name of the design pattern again? Sorry

etschelp (Thu, 03 Feb 2022 16:51:22 GMT):
command

Siltes (Thu, 03 Feb 2022 16:51:53 GMT):
Just Command Design Pattern?

etschelp (Thu, 03 Feb 2022 16:52:34 GMT):
yes, command pattern

Siltes (Thu, 03 Feb 2022 16:52:44 GMT):
Thanks :)

JamesSchulte (Thu, 03 Feb 2022 16:53:07 GMT):
​Question from livestream: "Could you give an example of a use case, where Indy would be more suitable over Fabric? I find it difficult to see how Fabric differs when it has the MSP component in it. Perhaps both can be used?"

swcurran (Thu, 03 Feb 2022 17:00:10 GMT):
Indy is about Identity and the availability of public keys and metadata in order to exchange VCs. If you are doing identity/verifiable credentials, you are better to use Indy and Aries. Verifiable Credentials are held by the Holder and only shared when they want -- they cannot be retrieved by a verifier -- the verifier has to ask the holder for the data. If you are publishing data on a blockchain, and you want shared access to that data with the timeline and immutablity of the data, then Fabric is right. The only crossovers that I know: - Using a verifiable credential to access a Fabric ledger, so you know who is using the ledger. In other words, the classic authentication/authorization, but you use VCs instead of "traditional" means. - Using Fabric as a place to store DIDs instead of Indy. That gets you DID functionality, and in some VC use cases, that is enough.

swcurran (Thu, 03 Feb 2022 17:00:10 GMT):
Indy is about Identity and the availability of public keys and metadata in order to exchange VCs. If you are doing identity/verifiable credentials, you are better to use Indy and Aries. Verifiable Credentials are held by the Holder and only shared when they want -- they cannot be retrieved by a verifier in an ad hoc way -- the verifier has to ask the holder for the data. If you are publishing data on a blockchain, and you want shared access to that data with the timeline and immutablity of the data, then Fabric is right. The only crossovers that I know: - Using a verifiable credential to access a Fabric ledger, so you know who is using the ledger. In other words, the classic authentication/authorization, but you use VCs instead of "traditional" means. - Using Fabric as a place to store DIDs instead of Indy. That gets you DID functionality, and in some VC use cases, that is enough.

swcurran (Thu, 03 Feb 2022 17:00:10 GMT):
Indy is about Identity and the availability of public keys and metadata in order to exchange VCs. If you are doing identity/verifiable credentials, you are better to use Indy and Aries. Verifiable Credentials are held by the Holder and only shared when they want -- they cannot be retrieved by a verifier in an ad hoc way -- the verifier has to ask the holder for the verifiable data. If you are publishing data on a blockchain, and you want shared access to that data with the timeline and immutablity of the data, then Fabric is right. The only crossovers that I know: - Using a verifiable credential to access a Fabric ledger, so you know who is using the ledger. In other words, the classic authentication/authorization, but you use VCs instead of "traditional" means. - Using Fabric as a place to store DIDs instead of Indy. That gets you DID functionality, and in some VC use cases, that is enough.

swcurran (Thu, 03 Feb 2022 17:00:10 GMT):
Indy is about Identity and the availability of public keys and metadata in order to exchange VCs. If you are doing identity/verifiable credentials, you are better to use Indy and Aries. Verifiable Credentials are held by the Holder and only shared when they want -- they cannot be retrieved by a verifier in an ad hoc way -- the verifier has to ask the holder for the verifiable data. If you are publishing data on a blockchain, and you want shared access to that data with the timeline and immutablity of the data, then Fabric is right. The only crossovers that I know: - Using a verifiable credential to access a Fabric ledger, so you know who is using the ledger. In other words, classic authentication/authorization, but you use VCs instead of "traditional" means. - Using Fabric as a place to store DIDs instead of Indy. That gets you DID functionality, and in some VC use cases, that is enough.

Ben208N (Thu, 03 Feb 2022 17:03:03 GMT):
Has joined the channel.

sbohanlf (Thu, 03 Feb 2022 17:05:21 GMT):
great reply - added to YT stream chat

TelegramSam (Thu, 03 Feb 2022 17:10:29 GMT):
Note: when using a seed to create a DID, be careful choosing your seed! It should be sufficiently hard to guess.

Smalviya (Thu, 03 Feb 2022 17:15:55 GMT):
thanks @lynn.bendixsen bumping up the ram to 4 gigs and disk to 60 gigs helped, i got 6 errors which as per you is fine :)

fl0x (Thu, 03 Feb 2022 17:23:07 GMT):
i want to build the indy and i got the message that the devcontainter path is not find it. have somebody an idea what is the problem ?

fl0x (Thu, 03 Feb 2022 17:23:07 GMT):
i want to build the indy-node according to the documentation( docker build -t indy-node-dev -f .devcontainer/Dockerfile .devcontainer/) and i got the message that the devcontainter path is not find it. have somebody an idea what is the problem ? The docker is installed and works

Smalviya (Thu, 03 Feb 2022 17:25:31 GMT):
are you able to run docker as non-root.

DibbsZA (Thu, 03 Feb 2022 17:26:05 GMT):
You may have not checked out the right branch

ArturPhilipp (Thu, 03 Feb 2022 17:26:09 GMT):
I am pretty sure, it is because you did not checkout the ubuntu 20.04 upgrade branch

ArturPhilipp (Thu, 03 Feb 2022 17:26:10 GMT):
git clone -b ubuntu-20.04-upgrade

ArturPhilipp (Thu, 03 Feb 2022 17:26:24 GMT):
(because I had the issue earlier as well and it was because of the wromg branch :)

DibbsZA (Thu, 03 Feb 2022 17:27:30 GMT):
same - copy paste fail :joy:

fl0x (Thu, 03 Feb 2022 17:31:47 GMT):
[ ](https://chat.hyperledger.org/channel/indy?msg=vtKnTSzSsYq6SHrLe) Thanks .. that was the error

richardspeter (Thu, 03 Feb 2022 17:56:47 GMT):
I am facing an issue with indy-cli installation

richardspeter (Thu, 03 Feb 2022 17:57:04 GMT):
sudo add-apt-repository "deb https://repo.sovrin.org/sdk/deb xenial stable"

richardspeter (Thu, 03 Feb 2022 17:57:44 GMT):
I see errors related to expired certificates

dioxis (Thu, 03 Feb 2022 17:58:23 GMT):
Got the same error as @richardspeter

richardspeter (Thu, 03 Feb 2022 17:58:47 GMT):
rr:11 https://repo.sovrin.org/sdk/deb bionic Release Certificate verification failed: The certificate is NOT trusted. The certificate chain uses expired certificate. Could not handshake: Error in the certificate verification. [IP: 3.17.126.236 443] Err:12 https://repo.sovrin.org/sdk/deb xenial Release Certificate verification failed: The certificate is NOT trusted. The certificate chain uses expired certificate. Could not handshake: Error in the certificate verification. [IP: 3.17.126.236 443] Err:13 https://repo.sovrin.org/deb xenial Release Certificate verification failed: The certificate is NOT trusted. The certificate chain uses expired certificate. Could not handshake: Error in the certificate verification. [IP: 3.17.126.236 443] Reading package lists... Done

richardspeter (Thu, 03 Feb 2022 17:59:19 GMT):
Could you please help me resolve this issue?

sbohanlf (Thu, 03 Feb 2022 18:05:55 GMT):
question from YT: ​Hi! In a use case supporting self-sovereign identities, where users pass contracts between each other and sign these, could you in brief describe how Indy could fit, form an architectural pow?

swcurran (Thu, 03 Feb 2022 18:06:44 GMT):
@WadeBarnes

kwollo (Thu, 03 Feb 2022 18:07:33 GMT):
Has joined the channel.

da3v21 (Thu, 03 Feb 2022 18:08:51 GMT):
Can master secret be accessed in aries-askar?

lynn.bendixsen (Thu, 03 Feb 2022 18:09:22 GMT):
:)

kwollo (Thu, 03 Feb 2022 18:10:56 GMT):
can you add a genesis node to the pool via indy cli using the "pool upgrade" command?

burdettadam (Thu, 03 Feb 2022 18:12:56 GMT):
Great question! While I have not worked much with askar yet, I believe that is a defining difference with libindy/sdk and aries-askar. Askar gives more flexibility in storage and allows direct access to the wallets secrets including the Link or master secret.

WadeBarnes (Thu, 03 Feb 2022 18:18:56 GMT):
You are likely missing the certificate update before you try to install the packages.

WadeBarnes (Thu, 03 Feb 2022 18:19:02 GMT):
Example;

WadeBarnes (Thu, 03 Feb 2022 18:19:02 GMT):
https://github.com/hyperledger/indy-node/blob/ubuntu-20.04-upgrade/.devcontainer/Dockerfile#L18-L25

da3v21 (Thu, 03 Feb 2022 18:19:17 GMT):
great! thanks

WadeBarnes (Thu, 03 Feb 2022 18:19:56 GMT):
You can also run into certificate issues if you don't update your `ca-certificates` package. Example; https://github.com/hyperledger/indy-node/blob/ubuntu-20.04-upgrade/.devcontainer/Dockerfile#L13

lynn.bendixsen (Thu, 03 Feb 2022 18:24:18 GMT):
No. first, "Pool upgrade" is used to do a software upgrade to a new version of indy-node and other dependencies. Genesis files are created before a Network is instantiated and remain static. You can add nodes to a network, but you do not add them to the genesis file.

lynn.bendixsen (Thu, 03 Feb 2022 18:25:05 GMT):
pool-upgrade is a remote way to tell all the nodes to update to the newest version.

WadeBarnes (Thu, 03 Feb 2022 18:34:00 GMT):
Working example: ``` FROM ubuntu:20.04 RUN apt-get update -y && apt-get install -y \ gnupg \ ca-certificates RUN apt-key adv --keyserver keyserver.ubuntu.com --recv-keys CE7709D068DB5E88 RUN echo "deb https://repo.sovrin.org/sdk/deb bionic stable" >> /etc/apt/sources.list RUN apt-get update -y && apt-get install -y \ libindy \ indy-cli ```

WadeBarnes (Thu, 03 Feb 2022 18:34:00 GMT):
Working (Docker) example: ``` FROM ubuntu:20.04 RUN apt-get update -y && apt-get install -y \ gnupg \ ca-certificates RUN apt-key adv --keyserver keyserver.ubuntu.com --recv-keys CE7709D068DB5E88 RUN echo "deb https://repo.sovrin.org/sdk/deb bionic stable" >> /etc/apt/sources.list RUN apt-get update -y && apt-get install -y \ libindy \ indy-cli ```

WadeBarnes (Thu, 03 Feb 2022 18:34:14 GMT):
That works with 18.04 too.

dioxis (Thu, 03 Feb 2022 18:34:55 GMT):
apt-key adv --keyserver keyserver.ubuntu.com --recv-keys 9692C00E657DDE61 helpt for me its missing on this site: https://wiki.hyperledger.org/display/events/Hyperledger+Aries+Training+Workshop+Prerequisites in the Install Indy-CLI section

dioxis (Thu, 03 Feb 2022 18:36:07 GMT):
Thank you very much @WadeBarnes

WadeBarnes (Thu, 03 Feb 2022 18:39:15 GMT):
I think what's missing from that page is an install or upgrade to the `ca-certificates` package. `CE7709D068DB5E88` is the correct signing key, but it's not always recognized without the updated ca package. This started happening late last year.

kwollo (Thu, 03 Feb 2022 18:40:06 GMT):
To return to my original question in zoom: So if a genesis node needs to be replaced, is there no way to move it to a new one?

kwollo (Thu, 03 Feb 2022 18:41:48 GMT):
This means that you cannot subsequently delete a genesis node and add a new one?

dioxis (Thu, 03 Feb 2022 18:45:02 GMT):
Can confirm that. Done a reset to my VM. Upgrade to the ca-certificates package works too.

WadeBarnes (Thu, 03 Feb 2022 18:45:59 GMT):
`9692C00E657DDE61` is the signing key for the new Hyperledger repository where all the latest `indy-node` and `indy-plenum` packages are now being published buy the new GitHub Actions Workflows.

lynn.bendixsen (Thu, 03 Feb 2022 18:48:47 GMT):
You can remove 1 or 2 of the original nodes and be okay, but there is not an easy way to replace them. options: 1. "Reset" the ledger (loses all data on all ledgers, but allows for new genesis files if you choose.) 2. I have actually helped update the Pool ledger for a network once, by taking the complete existing pool ledger (180+txns) and telling all network users to update to that new file as the genesis.

lynn.bendixsen (Thu, 03 Feb 2022 18:48:47 GMT):
You can remove 1 or 2 of the original nodes and be okay, but there is not an easy way to replace them. options: 1. "Reset" the ledger (loses all data on all sub-ledgers, but allows for new genesis files if you choose.) 2. I have actually helped update the Pool ledger for a network once, by taking the complete existing pool ledger (180+txns) and telling all network users to update to that new file as the genesis.

lynn.bendixsen (Thu, 03 Feb 2022 18:55:19 GMT):
so, the original genesis nodes are never replaced (immutable ledger). You can add "the whole network" as "genesis nodes" using option 2

lynn.bendixsen (Thu, 03 Feb 2022 18:55:19 GMT):
so, the original genesis nodes can never be removed if you want to keep the other ledger transactions (immutable ledger). You can add "the whole network" as "genesis nodes" using option 2

kwollo (Thu, 03 Feb 2022 18:57:05 GMT):
All right, I have understood that now. When a new validator node is added, does it change anything in the two genesis files? Or in general when adding a new node.

lynn.bendixsen (Thu, 03 Feb 2022 19:06:05 GMT):
When a new Node is added, it updates the pool sub-ledger on the network, but NOT the genesis files. The genesis files remains static.

lynn.bendixsen (Thu, 03 Feb 2022 19:09:10 GMT):
When a client connects to the network, it uses the pool genesis file, then the first thing it does next is to download the rest of the pool ledger, and now uses that for connecting to all the network nodes to run any other commands.

kwollo (Thu, 03 Feb 2022 19:22:50 GMT):
Is it possible to read the pool_genesis with one of your tools, like indy-cli oder indy-vdr, only if the domain_genesis is available?

lynn.bendixsen (Thu, 03 Feb 2022 19:30:04 GMT):
Hmm, not sure I understand the question. I am pretty sure that the indy-cli does not have anything that displays the pool genesis file. The pool genesis is downloaded from wherever the network administrators have provided it and can be viewed there or on your workstation after downloading it. The domain genesis is actually only used by Nodes of the network and usually only when they add themselves to the network. The tools need you to provide the pool genesis file to be able to connect to a network.

richardspeter (Thu, 03 Feb 2022 19:38:48 GMT):
Hi WadeBarnes

WadeBarnes (Thu, 03 Feb 2022 19:39:07 GMT):
HI

richardspeter (Thu, 03 Feb 2022 19:40:06 GMT):
Are we expected to use Ubuntu 18.04 for trying out Indy at the moment?

WadeBarnes (Thu, 03 Feb 2022 19:40:42 GMT):
What version were you using. I was just providing an example.

richardspeter (Thu, 03 Feb 2022 19:41:29 GMT):
Ubuntu 19.10

WadeBarnes (Thu, 03 Feb 2022 19:43:10 GMT):
Looks like you might run into some trouble there.

richardspeter (Thu, 03 Feb 2022 19:43:39 GMT):
ok

WadeBarnes (Thu, 03 Feb 2022 19:43:52 GMT):
I've tried my example with 18.04 and 20.04. I just tried with 19.10 and ran into security errors.

richardspeter (Thu, 03 Feb 2022 19:44:28 GMT):
i can try upgrading to 20.04

WadeBarnes (Thu, 03 Feb 2022 19:44:49 GMT):
Specifically I ran into issues installing `gnupg`

WadeBarnes (Thu, 03 Feb 2022 19:44:49 GMT):
Specifically I ran into issues installing `gnupg`, with 19.10

WadeBarnes (Thu, 03 Feb 2022 19:45:32 GMT):
Which is a prerequisite for running `apt-key adv --keyserver keyserver.ubuntu.com --recv-keys CE7709D068DB5E88`

richardspeter (Thu, 03 Feb 2022 19:45:56 GMT):
I performed these 2:

richardspeter (Thu, 03 Feb 2022 19:45:58 GMT):
sudo apt-key adv --keyserver keyserver.ubuntu.com --recv-keys CE7709D068DB5E88 sudo apt-key adv --keyserver keyserver.ubuntu.com --recv-keys 9692C00E657DDE61

richardspeter (Thu, 03 Feb 2022 19:47:11 GMT):
and then when I add repositories, I see the certificate errors

WadeBarnes (Thu, 03 Feb 2022 19:47:16 GMT):
I also ran into issues installing `ca-certificates` on 19.10

richardspeter (Thu, 03 Feb 2022 19:47:21 GMT):
sudo add-apt-repository "deb https://repo.sovrin.org/sdk/deb xenial stable" sudo add-apt-repository "deb https://repo.sovrin.org/deb xenial stable"

WadeBarnes (Thu, 03 Feb 2022 19:48:12 GMT):
You would need to update the `ca-certificates` package. But as I said I ran into errors installing that on 19.10

richardspeter (Thu, 03 Feb 2022 19:48:56 GMT):
sudo apt-get install -y ca-certificates Reading package lists... Done Building dependency tree Reading state information... Done ca-certificates is already the newest version (20190110ubuntu0.19.10.1). The following packages were automatically installed and are no longer required: gir1.2-gst-plugins-base-1.0 gir1.2-rb-3.0 gir1.2-totem-1.0 gir1.2-totemplparser-1.0 libdmapsharing-3.0-2 libgpod-common libgpod4 liblirc-client0 libnetplan0 libsgutils2-2 Use 'sudo apt autoremove' to remove them. 0 upgraded, 0 newly installed, 0 to remove and 2 not upgraded.

WadeBarnes (Thu, 03 Feb 2022 19:49:08 GMT):
19.10 is no longer supported.

richardspeter (Thu, 03 Feb 2022 19:49:12 GMT):
ok

WadeBarnes (Thu, 03 Feb 2022 19:49:59 GMT):
`Ubuntu 19.10 (Eoan Ermine) End of Life reached on July 17 2020`

richardspeter (Thu, 03 Feb 2022 19:51:00 GMT):
ok, would you recommend upgrading to 20.04 or testing on 18.04

WadeBarnes (Thu, 03 Feb 2022 19:51:28 GMT):
20.04

richardspeter (Thu, 03 Feb 2022 19:51:33 GMT):
ok

richardspeter (Thu, 03 Feb 2022 19:51:47 GMT):
thanks a lot! Let me try once..

ngwlf (Thu, 03 Feb 2022 19:51:57 GMT):
Great event - keep it up!

WadeBarnes (Thu, 03 Feb 2022 19:52:47 GMT):
Both indy-node and indy-plenum have virtual 20.04 development environments in the `ubuntu-20.04-upgrade` branches.

WadeBarnes (Thu, 03 Feb 2022 19:52:47 GMT):
Both `indy-node` and `indy-plenum` have virtual 20.04 development environments in the `ubuntu-20.04-upgrade` branches.

Smalviya (Thu, 03 Feb 2022 19:53:13 GMT):
Great event :)

WadeBarnes (Thu, 03 Feb 2022 19:54:13 GMT):
FYI, I'm going offline in a few minutes, so I may not respond to any other questions until early tomorrow.

richardspeter (Thu, 03 Feb 2022 19:54:34 GMT):
sure, no problem

DurgaKVS (Thu, 03 Feb 2022 19:55:03 GMT):
Has joined the channel.

dbluhm (Thu, 03 Feb 2022 19:57:23 GMT):
Thanks for everyone who attended today's workshop! If you'd like to get involved in Indy Node, you can find some "good first issues" in the github repo here: https://github.com/hyperledger/indy-node/labels/good%20first%20issue :slightly_smiling_face:

dbluhm (Thu, 03 Feb 2022 19:57:23 GMT):
Thanks to everyone who attended today's workshop! If you'd like to get involved in Indy Node, you can find some "good first issues" in the github repo here: https://github.com/hyperledger/indy-node/labels/good%20first%20issue :slightly_smiling_face:

kwollo (Thu, 03 Feb 2022 20:32:50 GMT):
Thank you very much for your answers. I'm just wondering how to regenerate the genesis files when I'm in an existing network and I need to change a genesis node. E.g. if there is an irrreprable damage on a node and all the data is gone. So a new genesis node has to come online to reach the minimum number for operation. Then I would replace the failed node with init_indy_keys script and set a new name and seed on the new server. Then just generate the new genesis files with generate_indy_pool_transactions, right? Finally distribute the genesis files in the network: 1. will the new node run directly with the existing network? Can it download all the transactions from the network? 2. or do I have to create the new node 1:1 with the same name, seed, IP and import a backup of all the transactions to get it running again?

kwollo (Thu, 03 Feb 2022 20:35:28 GMT):
3. Or can i have a different name and seed and need a backup of the transactions to get it running again?

swcurran (Thu, 03 Feb 2022 22:51:27 GMT):
I've created and published the Hyperledger TSC Indy Quarterly Report here: https://wiki.hyperledger.org/display/TSC/2022+Q1+Hyperledger+Indy Please review and let me know if I have missed anything or have anything wrong. Thanks!

swcurran (Thu, 03 Feb 2022 22:57:08 GMT):
My initial reaction is that it is not a primary use case for Indy/Aries, which are focused more on the verifiable credential model (e.g. this model as documented on the Trust Over IP website - https://trustoverip.org/toip-model/). In general, Indy instances are NOT used to store DIDs (identifiers+public keys) for individuals, so although there is signing as a concept in Hyperledger, not signing as you might think of it in this use case. I'm sure there are ways to do this with Indy in that you have keys as a foundational component, but it is really not a core use case supported today.

swcurran (Thu, 03 Feb 2022 22:57:08 GMT):
My initial reaction is that it is not a primary use case for Indy/Aries, which are focused more on the verifiable credential model (e.g. this model as documented on the Trust Over IP website - https://trustoverip.org/toip-model/). In general, Indy instances are NOT used to store DIDs (identifiers+public keys) for individuals, so although there is signing as a concept in Hyperledger, not signing as you might think of it in this specific use case. I'm sure there are ways to do the back and forth contract signing (basically, multi-sig) with Indy in that you have keys as a foundational component, but it is not a core use case supported today.

swcurran (Thu, 03 Feb 2022 23:03:19 GMT):
Both Indy Wallet and Askar are designed to make it easy to use the private keys in the storage, but not easy to access those keys. Signing and encryption is done by passing in the relevant objects and getting them back out signed/encrypted, without the private keys being accessible to the user. We want to prevent developers from doing bad things. Security precautions are used, like clearing memory after the operation and before data is provided back to the caller. The advantage that Askar brings is that it has more flexibility in the types keys handled, and the operations supported (enabling different crypto schemes, like BBS+ Signatures), uses some newer database handling for more efficient operations, and is separated from other Indy operations. In libindy, all operations share a common lock, which can cause a bottleneck.

swcurran (Thu, 03 Feb 2022 23:03:40 GMT):
The Master Secret is treated like a private key in Askar.

swcurran (Thu, 03 Feb 2022 23:03:40 GMT):
The Master Secret is treated like a private key in Askar and Indy.

richardspeter (Fri, 04 Feb 2022 00:34:30 GMT):
I upgraded my OS from 19.10 to 20.04. It resolved the issues related to certificates.

richardspeter (Fri, 04 Feb 2022 00:36:27 GMT):
I am experiencing an issue with indy-cli

richardspeter (Fri, 04 Feb 2022 00:37:31 GMT):
The following packages have unmet dependencies: indy-cli : Depends: libindy (= 1.16.0)

richardspeter (Fri, 04 Feb 2022 00:39:36 GMT):
While trying to install libindy, there are issues related to libsodium18

richardspeter (Fri, 04 Feb 2022 00:39:47 GMT):
The following packages have unmet dependencies: libindy : Depends: libsodium18 but it is not installable

richardspeter (Fri, 04 Feb 2022 00:41:51 GMT):
I installed libsodium ( from https://libsodium.gitbook.io/doc/installation). I tried with libsodium-1.0.18 and libsodium-1.0.14.

richardspeter (Fri, 04 Feb 2022 00:43:24 GMT):
However I am still not able to install indy-cli and libindy

richardspeter (Fri, 04 Feb 2022 00:43:43 GMT):
ls /usr/local/lib libsodium.a libsodium.la libsodium.so libsodium.so.18 libsodium.so.18.4.0 libsodium.so.23 libsodium.so.23.3.0 nodejs pkgconfig python2.7 python3.8

richardspeter (Fri, 04 Feb 2022 00:43:57 GMT):
Am I doing something wrong here?

lynn.bendixsen (Fri, 04 Feb 2022 00:46:36 GMT):
If it's just one node, then I wouldn't worry about updating the genesis files unless you want to lose everything and start over. If you can get the same seed/IPs and name on a new node, then put the old genesis files in the right place and it will come up and catch up to the other nodes (best option). Instructions here (very near the end of 3.2.3) have a note about how to run init-indy-node and get the same seeds/keys: https://docs.google.com/document/d/1y0rW78_I-bRkH3qFN5kcP58jJH23Zahc363h18SmJ48 If you create and add a new node to network that is in consensus, then the same thing will happen, all of the sub-ledgers are downloaded to the new node until it is in sync with the network and a participating member. You shouldn't ever need a backup of the transactions copied around manually.

DibbsZA (Fri, 04 Feb 2022 06:46:34 GMT):
I get a Page not Found error?

DibbsZA (Fri, 04 Feb 2022 06:49:57 GMT):
You mislabeled it Q4 https://wiki.hyperledger.org/display/TSC/2022+Q4+Hyperledger+Indy

DibbsZA (Fri, 04 Feb 2022 06:50:53 GMT):
Very forward thinking of you :stuck_out_tongue_winking_eye:

JonasRasmussen (Fri, 04 Feb 2022 08:20:18 GMT):
Has joined the channel.

ThomasFR89 (Fri, 04 Feb 2022 08:20:26 GMT):
Has joined the channel.

LucasCanaro (Fri, 04 Feb 2022 08:29:03 GMT):
Has joined the channel.

iamchirag45 (Fri, 04 Feb 2022 12:18:56 GMT):
Has joined the channel.

WadeBarnes (Fri, 04 Feb 2022 12:27:17 GMT):
What have you done differently than this example: https://chat.hyperledger.org/channel/indy?msg=GjBqXAzHKTpwhZWqR

iamchirag45 (Fri, 04 Feb 2022 12:43:26 GMT):

Screenshot (104).png

WadeBarnes (Fri, 04 Feb 2022 13:27:38 GMT):
Would you be able to provide a record of the steps you preformed and the output of each step?

WadeBarnes (Fri, 04 Feb 2022 13:27:51 GMT):
cc @pSchlarb

Prasad8 2 (Fri, 04 Feb 2022 14:07:14 GMT):
Hello All, I am trying to install Indy SDK packages. For windows, i just downloaded and imported the libindy dll files inside of my java project and i can call the indy methods. They are working fine. Now i am trying to move the java application to ubuntu and needed linux installation of libindy files . so i followed *Ubuntu based distributions (Ubuntu 16.04 and 18.04)* section from https://github.com/hyperledger/indy-sdk . I am getting an error when installing *libvcx* The following packages have unmet dependencies: libvcx : Depends: libindy (= 1.15.0~1628) E: Unable to correct problems, you have held broken packages. My ubuntu version : bionic for 18.04 Ubuntu . Please help me.

Prasad8 2 (Fri, 04 Feb 2022 14:07:14 GMT):
Hello All, I am trying to install Indy SDK packages. For windows, i just downloaded and imported the libindy dll files inside of my java project and i can call the indy methods. They are working fine. Now i am trying to move the java application to ubuntu and needed linux installation of libindy files . so i followed *Ubuntu based distributions (Ubuntu 16.04 and 18.04)* section from https://github.com/hyperledger/indy-sdk . I am getting an error when installing *libvcx* The following packages have unmet dependencies: libvcx : Depends: libindy (= 1.15.0~1628) E: Unable to correct problems, you have held broken packages. My ubuntu version : bionic for 18.04 Ubuntu . Please help me.

WadeBarnes (Fri, 04 Feb 2022 14:20:58 GMT):
Example (using Docker): ``` FROM ubuntu:18.04 RUN apt-get update -y && apt-get install -y \ gnupg \ ca-certificates RUN apt-key adv --keyserver keyserver.ubuntu.com --recv-keys CE7709D068DB5E88 RUN echo "deb https://repo.sovrin.org/sdk/deb bionic stable" >> /etc/apt/sources.list RUN apt-get update -y && apt-get install -y \ libindy ```

WadeBarnes (Fri, 04 Feb 2022 14:26:26 GMT):
Created a Gist here; https://gist.github.com/WadeBarnes/c0ec30be1040ade584489ba4f7b91e23

WadeBarnes (Fri, 04 Feb 2022 14:35:28 GMT):
PR for doc update here; https://github.com/hyperledger/indy-sdk/pull/2478

swcurran (Fri, 04 Feb 2022 15:56:34 GMT):
:joy:

swcurran (Fri, 04 Feb 2022 15:56:41 GMT):
Fixed now...doh!!!

swcurran (Fri, 04 Feb 2022 15:56:59 GMT):
Thanks for letting me know.

freddylarag (Fri, 04 Feb 2022 16:39:07 GMT):
Has joined the channel.

LucasCosta (Fri, 04 Feb 2022 19:16:52 GMT):
Has joined the channel.

richardspeter (Sat, 05 Feb 2022 02:44:26 GMT):
I managed to go past the error by following the instructions here - https://github.com/hyperledger/indy-node/blob/master/docs/source/setup-dev.md

richardspeter (Sat, 05 Feb 2022 02:45:29 GMT):
I included "deb http://us.archive.ubuntu.com/ubuntu xenial main universe" to my /etc/apt/sources.list

richardspeter (Sat, 05 Feb 2022 02:48:10 GMT):
After that, installation of libsodium18, libindy and indy-cli took place.

rizy (Sat, 05 Feb 2022 03:21:39 GMT):
Has joined the channel.

Prasad8 2 (Sat, 05 Feb 2022 09:01:21 GMT):
Thanks. The Linux packages works for Ubuntu 20 also or is it strictly allowed to use only with Ubuntu 16 and 18 ?

Prasad8 2 (Sat, 05 Feb 2022 09:14:43 GMT):
Hi all, I am following anoncreds Java file to issue credentials. In anoncreds, the offer credentials json is sent to holder with only the attributes and without any actual credentials data, however the Aries agent cloud py has offer credentials that needed the actual data to be sent in the credentials preview . What is the actual flow if holder wants to start the conversation asking for a credential ? My flow is : holder requesting credentials > issuer offering credentials > Holder accepts or decline to end the flow. Pls help me.

DeepakMuthekar (Sat, 05 Feb 2022 11:48:50 GMT):
Has joined the channel.

WadeBarnes (Sat, 05 Feb 2022 14:33:55 GMT):
I believe it works for 20.04 as well.

iamchirag45 (Sat, 05 Feb 2022 14:45:47 GMT):
@WadeBarnes @pSchlarb Thanku for reply, i am attaching a drive link to a text file which contains all the steps in detail with output that i have performed while creating this network. https://drive.google.com/file/d/1xVQOkYgFOUoG6NpsqDS_W2zNZ9RguH4Y/view and any suggestions, any kind of advice is welcome .

iamchirag45 (Sat, 05 Feb 2022 14:47:22 GMT):
also i am using Ubuntu 16

alih27140 (Sat, 05 Feb 2022 15:53:04 GMT):
Has joined the channel.

VasileiosK (Sat, 05 Feb 2022 17:19:54 GMT):
Has joined the channel.

jmb (Sat, 05 Feb 2022 19:15:48 GMT):
Has joined the channel.

gaganbabber (Sun, 06 Feb 2022 01:01:24 GMT):
Has joined the channel.

Rafeu (Sun, 06 Feb 2022 08:03:42 GMT):
Has joined the channel.

siddhantprateek (Sun, 06 Feb 2022 09:28:57 GMT):
Has joined the channel.

degex (Sun, 06 Feb 2022 10:31:26 GMT):
Has joined the channel.

govindlahoti (Sun, 06 Feb 2022 12:42:13 GMT):
Has joined the channel.

YU.I (Sun, 06 Feb 2022 13:52:22 GMT):
Has joined the channel.

sanjichained (Sun, 06 Feb 2022 16:44:23 GMT):
in the indy workshop it was said there is community call on AnonCreds Standardisation, could you point me to the link where i can register and read more about it

png_savvy (Sun, 06 Feb 2022 19:40:48 GMT):
Has joined the channel.

PhilippThater (Sun, 06 Feb 2022 22:25:30 GMT):
Has joined the channel.

rramon (Sun, 06 Feb 2022 22:41:17 GMT):
Has joined the channel.

rramon (Sun, 06 Feb 2022 23:45:03 GMT):
Hello everyone! I am new to the chat, so please let me know if this is not the appropriate channel for this question. I am working through the [Indy Walkthrough](https://github.com/hyperledger/indy-sdk/blob/master/docs/getting-started/indy-walkthrough.md) utilizing the [Indy SDK NodeJS Wrapper](https://github.com/hyperledger/indy-sdk/blob/master/wrappers/nodejs/README.md) and have gotten to the point where I am creating the transcript credential for Alice from Faber using the `issuerCreateCredential` method. I am receiving a 112 Indy Error Code, which from [THIS](https://github.com/hyperledger/indy-sdk/blob/master/libindy/include/indy_mod.h) doc appears to be a "Invalid library state was detected in runtime. It signals library bug" error. *Is there anything I can do to enable more verbose logging?* I have tried setting the `RUST_BACKTRACE=1` environment variable and calling the `setRuntimeConfig({ "collect_backtrace": true })` at the beginning of my code, but I still receive the 112 Indy Error code with no additional information. Any guidance on how to debug this issue would be greatly appreciated!

abwh2 (Mon, 07 Feb 2022 03:19:20 GMT):
Has joined the channel.

cacaocal (Mon, 07 Feb 2022 09:28:28 GMT):
Has joined the channel.

AntonYamkovoy (Mon, 07 Feb 2022 12:16:33 GMT):
Has joined the channel.

AhmadKh (Mon, 07 Feb 2022 14:08:44 GMT):
Has joined the channel.

jonathan1204 (Mon, 07 Feb 2022 15:02:12 GMT):
Has joined the channel.

jonathan1204 (Mon, 07 Feb 2022 15:05:58 GMT):
Hello. How are they? I am creating a mobile application and I want to connect to a mediator agent, I launch the faber agent with the --mediation tag, however I cannot connect it, is there an endpoint to which I should point my request in the faber agent? Thank you.

TelegramSam (Mon, 07 Feb 2022 16:20:26 GMT):
Still very early. @swcurran is a ringleader in that effort. Proposed Charter here: https://docs.google.com/document/d/1W7dTnVVM_oXbe7g_jyA_FhkfmyBKGQV-iRTKj9hoLpM/edit#

salikzquidz (Mon, 07 Feb 2022 16:54:17 GMT):
Has joined the channel.

gabrielsdev (Mon, 07 Feb 2022 17:17:01 GMT):
Has joined the channel.

swcurran (Mon, 07 Feb 2022 17:26:26 GMT):
There is a public test mediator run by Indicio that you can use. To use a mediator with a mobile agent, you wouldn't need faber to have a mediator (which is what the --mediator argument does).

swcurran (Mon, 07 Feb 2022 17:27:08 GMT):
https://indicio-tech.github.io/mediator/

swcurran (Mon, 07 Feb 2022 17:28:19 GMT):
That said there is a mention on #aries-bifold that the mediator might be having trouble this morningh. @lynn.bendixsen -- is all well with it?

lynn.bendixsen (Mon, 07 Feb 2022 19:29:04 GMT):
It was having troubles, but I believe all is back up and running now :)

khushalkunjir (Tue, 08 Feb 2022 05:51:44 GMT):
Has joined the channel.

SwapnilDeshmukh (Tue, 08 Feb 2022 11:17:38 GMT):
Has joined the channel.

Cai 238 (Tue, 08 Feb 2022 11:32:34 GMT):
Has joined the channel.

bruno_evaristo (Tue, 08 Feb 2022 19:58:25 GMT):
Has joined the channel.

HamidTariq (Tue, 08 Feb 2022 23:50:34 GMT):
Has joined the channel.

ankitkamra (Wed, 09 Feb 2022 04:41:05 GMT):
Has joined the channel.

sanjichained (Wed, 09 Feb 2022 07:20:28 GMT):
thank you for the info, I would like to contribute to it and learn from it, so I will follow its progress and see where i can add value

MuteMan (Wed, 09 Feb 2022 07:59:41 GMT):
Has joined the channel.

harikishantk (Wed, 09 Feb 2022 09:22:36 GMT):
Has joined the channel.

theNatassa (Wed, 09 Feb 2022 12:34:36 GMT):
Has joined the channel.

swcurran (Wed, 09 Feb 2022 15:17:33 GMT):
Good stuff -- we'll post here when we get a formal group going.

SwapnilPakhare (Wed, 09 Feb 2022 16:55:54 GMT):
Has joined the channel.

LiXu 17 (Wed, 09 Feb 2022 18:06:50 GMT):
Has joined the channel.

mccown (Wed, 09 Feb 2022 21:31:09 GMT):
The Identity Implementer’s WG meeting for tomorrow (Feb 10th) has been cancelled. Please join us in our next meeting on Feb 26th (9am MT / 4pm UTC).

joseucarvajal (Wed, 09 Feb 2022 23:16:07 GMT):
Has joined the channel.

lynn.bendixsen (Wed, 09 Feb 2022 23:59:57 GMT):
Apologies for the issues in the guide that you are using. Steps 8 and 9 need to be moved in front of Step 5 because the keys need to be created in the directory of the new network. @pSchlarb I suggest to re-run step 5, this time adding on the seed created the first time you ran it to the end of each init_indy_node command (so that you don't have to recreate your genesis files) then re-try step 9. init_indy_node Steward1 0.0.0.0 9701 0.0.0.0 9702 C378fAAAcE9Ba636eABaAAea39c0aD08 init_indy_node Steward2 0.0.0.0 9703 0.0.0.0 9704 2dDAc3EDBdEd3D44aE5ab0E51D597afE init_indy_node Steward3 0.0.0.0 9705 0.0.0.0 9706 3f80C620Be341aFFdC74c7ef2D757677 init_indy_node Steward4 0.0.0.0 9707 0.0.0.0 9708 8a7A2eaaA3aEF0912AdC8DD0A12AcEA3

lynn.bendixsen (Wed, 09 Feb 2022 23:59:57 GMT):
Apologies for the issues in the guide that you are using. Steps 8 and 9 need to be moved in front of Step 5 because the keys need to be created in the directory of the new network. @pSchlarb @iamchirag45 I suggest that you re-run step 5, this time adding on the seed created the first time you ran it to the end of each init_indy_node command (so that you don't have to recreate your genesis files) then re-try step 9. init_indy_node Steward1 0.0.0.0 9701 0.0.0.0 9702 C378fAAAcE9Ba636eABaAAea39c0aD08 init_indy_node Steward2 0.0.0.0 9703 0.0.0.0 9704 2dDAc3EDBdEd3D44aE5ab0E51D597afE init_indy_node Steward3 0.0.0.0 9705 0.0.0.0 9706 3f80C620Be341aFFdC74c7ef2D757677 init_indy_node Steward4 0.0.0.0 9707 0.0.0.0 9708 8a7A2eaaA3aEF0912AdC8DD0A12AcEA3

lynn.bendixsen (Wed, 09 Feb 2022 23:59:57 GMT):
Apologies for the issues in the guide that you are using. Steps 8 and 9 need to be moved in front of Step 5 because the keys need to be created in the directory of the new network. @pSchlarb @iamchirag45 I suggest that you re-run step 5, this time adding on the seed created the first time you ran it to the end of each init_indy_node command (so that you don't have to recreate your genesis files) then re-try step 9. Here are the adjusted commands with the seeds that were generated the first time you ran them (I got these from the output you shared with us): init_indy_node Steward1 0.0.0.0 9701 0.0.0.0 9702 C378fAAAcE9Ba636eABaAAea39c0aD08 init_indy_node Steward2 0.0.0.0 9703 0.0.0.0 9704 2dDAc3EDBdEd3D44aE5ab0E51D597afE init_indy_node Steward3 0.0.0.0 9705 0.0.0.0 9706 3f80C620Be341aFFdC74c7ef2D757677 init_indy_node Steward4 0.0.0.0 9707 0.0.0.0 9708 8a7A2eaaA3aEF0912AdC8DD0A12AcEA3

kaushikkumarbora (Thu, 10 Feb 2022 06:25:59 GMT):
Has joined the channel.

eeeeegik (Thu, 10 Feb 2022 07:38:30 GMT):
Has joined the channel.

Koraycill (Thu, 10 Feb 2022 11:49:56 GMT):
Has joined the channel.

jann2005 (Thu, 10 Feb 2022 12:07:43 GMT):
Has joined the channel.

AngelEscudero (Thu, 10 Feb 2022 12:58:13 GMT):
Has joined the channel.

WadeBarnes (Thu, 10 Feb 2022 14:26:15 GMT):
@lynn.bendixsen is correct. @pSchlarb are you able to make the change to the guide please?

WadeBarnes (Thu, 10 Feb 2022 14:28:57 GMT):
The order is correct in the [Setting up a New Network](https://github.com/hyperledger/indy-node/blob/master/docs/source/NewNetwork/NewNetwork.md)

WadeBarnes (Thu, 10 Feb 2022 14:44:00 GMT):
Actually it's not crystal clean there either.

iamchirag45 (Thu, 10 Feb 2022 17:02:39 GMT):
Yes @lynn.bendixsen , it is working fine for me now. I was able to create this network also it will be beneficial to others as well if you can make these changes in that document. and thankyou so much all @pSchlarb @WadeBarnes @lynn.bendixsen , for the help. I may request for some more help in near future. :sweat_smile:

jayzhan (Fri, 11 Feb 2022 08:27:49 GMT):
If we need identity management in Fabric, do we need Indy ledger? Can we store DIDs / DID Documents in Fabric, so we only need to run a single ledger and utilize verifiable credentials to replace authentication/authorization that is mostly managed by Fabric CA.

mgmgwi (Fri, 11 Feb 2022 09:22:19 GMT):
Has joined the channel.

Koraycill (Fri, 11 Feb 2022 09:35:04 GMT):
Hello, I would like to ask if I want to create an Indy Network, do I have to install from sovrin repos and if I do that will I be connected to Sovrin? Since I want to initiate a closed network.

ethereal1m (Fri, 11 Feb 2022 09:38:56 GMT):
Has joined the channel.

WadeBarnes (Fri, 11 Feb 2022 13:15:06 GMT):
The short answer to your question is yes. If you want to run an officially released version of indy-node, you'll need to use the packages from the Sovrin repo. That does not mean you have to use the Sovrin packages. You can use the generic Node packages. `https://repo.sovrin.org/` is where all of the deb packages for the indy projects have been hosted. With the indy-node 20.04 upgrade we have been moving the indy package hosting over to the Hyperledger repository; https://hyperledger.jfrog.io/ui/repos/tree/General/indy. The Hyperledger package repository only contains new builds, there are no official releases contained in the Hyperledger repository yet. Guides to create a new network can be found here; https://github.com/hyperledger/indy-node/blob/master/docs/source/NewNetwork/NewNetwork.md

WadeBarnes (Fri, 11 Feb 2022 13:15:06 GMT):
The short answer to your question is yes. If you want to run an officially released version of indy-node, you'll need to use the packages from the Sovrin repo. That does not mean you have to use the Sovrin packages. You can use the generic Node packages. https://repo.sovrin.org/ is where all of the deb packages for the indy projects have been hosted. With the indy-node 20.04 upgrade we have been moving the indy package hosting over to the Hyperledger repository; https://hyperledger.jfrog.io/ui/repos/tree/General/indy. The Hyperledger package repository only contains new builds, there are no official releases contained in the Hyperledger repository yet. Guides to create a new network can be found here; https://github.com/hyperledger/indy-node/blob/master/docs/source/NewNetwork/NewNetwork.md

Koraycill (Fri, 11 Feb 2022 13:23:25 GMT):
Thank you very much.

WadeBarnes (Fri, 11 Feb 2022 13:59:50 GMT):
Oh, and even if you decide to use the Sovrin packages, that does not mean you'll be connecting to the Sovrin network, it just means you then have access to the token-plugin that is contained in the Sovrin package.

Koraycill (Fri, 11 Feb 2022 14:04:48 GMT):
As I see from genesis files node_ip and client_ip of some stewards are same. I looked at Indy Node documents for server requirements and it says 2 public ips for 2 NICs. I guess this is not mandatory for a test network but recommended.

WadeBarnes (Fri, 11 Feb 2022 14:06:32 GMT):
Correct. That is the current recommendation. Some stewards joined before that was implemented.

swcurran (Fri, 11 Feb 2022 14:43:43 GMT):
I don't know enough about Fabric to answer how hard that is / if it is a good idea. Perhaps @troyronda might know better -- he knows the SSI space in general, along with Fabric.

swcurran (Fri, 11 Feb 2022 15:00:16 GMT):
All the Aries and Indy channels from RocketChat have been created on Discord.

rjones (Sat, 12 Feb 2022 21:59:37 GMT):
Has joined the channel.

rjones (Sat, 12 Feb 2022 21:59:38 GMT):
[Please move to Discord](https://discord.com/channels/905194001349627914/905205711850594336)

rjones (Sat, 12 Feb 2022 21:59:38 GMT):
[Please get an account on the Hyperledger discord](https://discord.gg), then [join Indy](https://discord.com/channels/905194001349627914/905205711850594336)