The W3C Credentials Community Group

Meeting Transcriptions and Audio Recordings (2014-today)

Go Back

Verifiable Traceability Task Force

Transcript for 2024-04-02

Agenda
https://www.w3.org/Search/Mail/Public/advanced_search?hdr-1-name=subject&hdr-1-query=%5BAGENDA&period_month=Apr&period_year=2024&index-grp=Public__FULL&index-type=t&type-index=public-credentials&resultsperpage=20&sortby=date
Organizer
Orie Steele, Mike Prorock, Mahmoud Alkhraishi
Scribe
Our Robot Overlords and Nis Jespersen and Our Robot Overlords and Nis Jespersen
Present
Mahmoud Alkhraishi, Patrick St-Louis, TallTed // Ted Thibodeau (he/him) (OpenLinkSw.com), Nis Jespersen , Chris Abernethy, Ted Thibodeau
Audio Log
<tallted_//_ted_thibodeau_(he/him)_(openlinksw.com)> that works so much better when I click the right meeting title...
Our Robot Overlords are scribing.
Patrick St-Louis: Announcement page you know they provide the link and the time of the meeting so I I don't know where else that sort of information is available.
Nis Jespersen is scribing.
Our Robot Overlords are scribing.
Nis Jespersen is scribing.
Chris Abernethy: Might not be able to make it, but Brent will take over anyway. Starting next week
Mahmoud Alkhraishi: Objections to the Thursday simeslot?
Next meeting is on Thursday April 11th
Patrick St-Louis: My general impression of implementing this specification. There is a conversation to be had regarding interoperaability. What does it mean to be an implementer? What is the purpose of being an implementer? To be addressed?
Mahmoud Alkhraishi: To an extend this will be addressed by PRs
Patrick St-Louis: I pass 99% but am non conformant. Frustrating. Requirements changes. There is no way for me to comply. BC Gov is a government interested in complying. Raised concerns for me.
Push back on PR. So, what does it mean to be an implementer of interop?
Nis Jespersen : We have been reflecting on having overstepped what is actually interoperability
Patrick St-Louis: Yes, but there needs to be a way to test. Or is it just status 200? There needs to be a way to test.
There needs to be a way for an implementation to be tested
Mahmoud Alkhraishi: We are saying the same. But we have made a mistake, focusing too much on what is actually interoperability. It should be VC data model compliant
The most valuable thign we can do is revisit our interop, focus on /presentations
Patrick St-Louis: But how can that be done without /issue and /verify. And revocation
Mahmoud Alkhraishi: It would be Neoflow issues a credential with their internal endpoints. Then call Mesur.
It then submits to every other implementer.
So, if Mavennet submits a SL VC, rejected by Transmute. Either side failed. In any case, not interoperable
Patrick St-Louis: That makes sense. But how can only /presentations do this?
Mahmoud Alkhraishi: You have a set of valid and invalid credentials. It is reflective of actual implementator's capabilities
Patrick St-Louis: How?
Mahmoud Alkhraishi: With the triggers we have running
Patrick St-Louis: How will it do that?
Mahmoud Alkhraishi: We have valid and invalid credentials on the repo.
The majority will show who is not passing
"Give me a valid implementation, I will submit it to Org A"
Mahmoud Alkhraishi: They can be static VCs
Patrick St-Louis: That is a good place to start. It is considerably different
It means removing the traceability API
Mahmoud Alkhraishi: That is the statement we are making. What I submit to you is what matters
Nis Jespersen : Yes, keep VC API and interop focused on what they are made for
Instead of synching them
Patrick St-Louis: This would test the core of what is to be tested
Mahmoud Alkhraishi: We need to think how to test every MUST statement
Patrick St-Louis: Still with postman?
Mahmoud Alkhraishi: Yes
Nis Jespersen : Yes
Mahmoud Alkhraishi: It should be much simpler, docs plus implementation
Patrick St-Louis: Here is our credential samples
Patrick St-Louis: You still environment variables
Securing mechanisms? sending a JWT. How does the credential format change the test?
Mahmoud Alkhraishi: Let's raise an issue and have the discussion there.
Patrick St-Louis: Satisfied with the discussion
Patrick St-Louis: The schema should still be changed. Need to look into where they are cahnged
Chris Abernethy: We have always done earliest first.
Ted Thibodeau: I debate that, we have had PRs with substantial changes running multiple meetings
Mahmoud Alkhraishi: We have not used DIDAuth for a long time
Nis Jespersen : Yes, should have been gone long ago
Mahmoud Alkhraishi: Adding VC DM to the references. Using respec magic
Mahmoud Alkhraishi: The same, but for DID core
Mahmoud Alkhraishi: Again the same, for trace vocab
Nis Jespersen : D'oh, deleted the wrong part.
Nis Jespersen : This was a dupe with my huge PR, that got taken out
Patrick St-Louis: This is the same as last time, three vendors bringing in requirements
Mahmoud Alkhraishi: That is why it got split up, take it on its own
Patrick St-Louis: As much as I like to believe that, I highly doubt that this came on its own. Discussions made out of band.
Mahmoud Alkhraishi: It only says switch to 256
We could make it both, what we want to support
Objections?
Changes the path to the collection
Patrick St-Louis: This is where the ext/int endpoints was discussed
I would still like this to be merged in
Still running tests
Supporting 2020 which is still the spec
Do anyone still run 2020?
Mahmoud Alkhraishi: I htink so, but they need to be on the calls to object
Objections to merging 629?
No, merging
Noting we are moving away from this kind of testing
As dicussed
Patrick St-Louis: Safely remove the tests
Propose closing issue 482
Mahmoud Alkhraishi: It is bitstring now, not status list
Nis Jespersen : The terms are a little odd. But the PR is a step towards the better
Mahmoud Alkhraishi: Let's not make perfect the enemy of good
Merging
Patrick St-Louis: Let me explain where I am coming from with these credentials. We might invest into UNTP spec rahter than trace vocab
BCGov have their own VCs using AnonCreds
The mines permit will be similar to petro/natural gas, looking at that
Patrick St-Louis: I added examples
With AnonCreds snakecase is used
Mahmoud Alkhraishi: Where is the credential which this is based on
Patrick St-Louis: This represents permit to mine at a given location
Issued the anoncred it is based on that DID
Publish the same public key through the DID, signing
Mahmoud Alkhraishi: We are over time
We would want to do LD the same way we do it on trace
Patrick St-Louis: I do have a version of that