Verifiable Credentials Working Group Telco — Minutes

Date: 2023-05-03

See also the Agenda and the IRC Log

Attendees

Present: Ivan Herman, Brent Zundel, Ted Thibodeau Jr., Hiroyuki Sano, Shigeya Suzuki, Paul Dietrich, Kristina Yasuda, Andres Uribe, Przemek Praszczalek, Dmitri Zagidulin, Andrew Whitehead, Michael Jones, PDL ASU, Orie Steele, Oliver Terbu, Dave Longley, Joe Andrieu, David Waite, Will Abramson, Gregory Natran, Gabe Cohen, Steve McCown, Mircea Nistor

Regrets: Kevin Griffin

Guests:

Chair: Brent Zundel

Scribe(s): Oliver Terbu, Orie Steele

Content:

Brent Zundel: agenda for today.
… agenda review.
… PR primarily in VCDM + work item status updates and with whatever time is remaining we will discuss issues.
… we also had one proposal for FPWD.
… any other topics folks have for today?
… intros?
… we will move into proposal first.

1. FPWD Proposal.

Brent Zundel: Orie, do you have text for proposal?

Orie Steele: i do not, we should construct.
… but can do intro.
… we have adopted work item called VC Data Integrity BBS.
… will share link.

Orie Steele: See draft on https://github.com/w3c/vc-di-bbs.

Orie Steele: had several rounds of cleanup since adopted.
… and issue grooming.
… good discussions in key areas where we want to do work and risks.
… tobias and orie both believe it is important that this is worked on.
… ask for FPWD.
… to start clock and continue work that is already happening.

Brent Zundel: i’m constructing text for proposal but need get formal name of spec.
… is there an FPWD version yet?

Orie Steele: is the expectations to prepare a FPWD before asking group?

Ivan Herman: no rule for that.
… process is acceptable.
… should work.

Proposed resolution: Publish the Securing VCs with BBS specification as a First Public Working Draft with a short name of vc-di-bbs, and with a target publication date of May 18th 2023. (Brent Zundel)

Brent Zundel: +1.

Orie Steele: +1.

Brent Zundel: not anybody jumping on queue.

Dave Longley: +1.

Oliver Terbu: +1.

Ivan Herman: +1.

Shigeya Suzuki: +1.

Will Abramson: +1.

PDL ASU: +1.

Joe Andrieu: +1.

Gabe Cohen: +1.

Andrew Whitehead: +1.

Hiroyuki Sano: +1.

greg natran: +1.

Dmitri Zagidulin: +1.

Brent Zundel: no objections.

Andres Uribe: +1.

Resolution #1: Publish the Securing VCs with BBS specification as a First Public Working Draft with a short name of vc-di-bbs, and with a target publication date of May 18th 2023.

Brent Zundel: hearing no objections, resolved.
… next steps, orie will work with ivan to get it out.

Orie Steele: Thank you, I will prepare the PR.

Ivan Herman: was referring to editor’s draft. want to have it done before the AC meeting.
… can wait until tomorrow. it should be on issue list by friday.
… you will run by usual checkers?

Orie Steele: yes.

2. Work Item status updates/PRs.

Brent Zundel: brief status update on each work items.
… most attention today on backlog of PRs in VCDM.
… just had an FPWD on bbs.
… anything else on bbs?

2.1. Full Pass (pr vc-di-bbs#86)

See github pull request vc-di-bbs#86.

Orie Steele: open PR that has a full pass on. covers substantial amount of updates.
… is it ok to merge the PR before running through doc checker?
… i want to direct attention to PR. a lot of stale content in the PR.
… should be advantageous to merge it.
… if merge PR then can easily issues that FPWD might surface.

Ivan Herman: if you cannot finish doc by today then tomorrow we cannot run whole thing and this might be problem because next week there won’t be a meeting. that might put the publishing date at risk.
… unfortunate because AC meeting next week.

Orie Steele: PR addresses several issues. fixes linking scenarios. that is the part to ask for permissions to merge.

Dave Longley: +1 to merge this PR.

Brent Zundel: would accelerate ability to create FPWD draft?

Orie Steele: correct.

Brent Zundel: as long as sign off by co-editor, then go for it. since FPWD we are very early on anyways.

PDL ASU: no objections here.

Brent Zundel: any objections to that course of actions?

Orie Steele: i doubt tobias (co-editor) can review this on time.

Brent Zundel: no objections noted.
… tobias’ approval does not have to be super formal as long as it exists.

Orie Steele: if objections then i will revert it.
… but unlikely there will be any.

Brent Zundel: comfortable with that approach.

2.2. JSON Schema (https://github.com/w3c/vc-json-schema).

Brent Zundel: json schema updates?

See github pull request vc-json-schema#132.

Gabe Cohen: we have a PR up, the draft before FPWD.
… incorporate a number of issues that were outstanding but it would be good to have another look.

Brent Zundel: any comments on json schema?

2.3. Status list 2021 (https://github.com/w3c/vc-status-list-2021).

Brent Zundel: looking at status list 2021.
… can anybody talk to this work item?

Orie Steele: several open PRs.
… for some no reason why we cannot merge it.
… eg PR 46.
… requires some changes but not clear what folks are requesting to change.

Orie Steele: 1. https://github.com/w3c/vc-status-list-2021/pull/46.

Orie Steele: then PR 60.

Orie Steele: 2. https://github.com/w3c/vc-status-list-2021/pull/60.

Orie Steele: adding JSON Web Tokens to the spec.

Orie Steele: 3. https://github.com/w3c/vc-status-list-2021/pull/61.

Orie Steele: 4. https://github.com/w3c/vc-status-list-2021/pull/62.

Orie Steele: 5. https://github.com/w3c/vc-status-list-2021/pull/63.

Orie Steele: closing tags.
… it would be good to have some discussions of PRs with changes requested.

Brent Zundel: we want to have PRs merged out of call time.
… if folks who need to review are not responding, pls reach out but in general avoid the need for call time to merge PRs.
… if consensus cannot be found in the comments of 46 then we can have call time but ideally we won’t need it.

2.4. EDDSA (https://github.com/w3c/vc-di-eddsa).

Brent Zundel: next work item is data integrity eddsa.
… anyone can update on this work item?
… dmitri?

Dmitri Zagidulin: no updates at the moment.
… ongoing work, will keep folks posted.

Brent Zundel: moving to.

2.5. ECDSA (https://github.com/w3c/vc-di-ecdsa).

Brent Zundel: ecdsa for data integrity.
… in absence of editors we skip.
… and let’s go to data integrity now.

2.6. Data Integrity (https://github.com/w3c/vc-data-integrity).

Brent Zundel: any updates?

Dave Longley: PRs coming along nicely, a couple of deprecation things for vocab which are approved. so those might be getting merged if no objections.
… work continues on other issues.
… any questions on data integrity?

2.7. Feature/editorial expansions (pr vc-jwt#77)

See github pull request vc-jwt#77.

Brent Zundel: moving to vc jwt spec.

Orie Steele: can cover it.
… mike p has made a pass on vc jwt and there is a lot of content he is updating.
… large PR but very much needed.
… spec feels light and a lot of paragraphs build out the feeling of the spec.
… personally don’t like those sections but acknowledges that other readers prefer that.
… a few unresolved suggestions but my feedback is non-blocking.
… kristina has requested changes but not sure if mike addressed them?

Brent Zundel: kristina will re-revieww.
… anything else that is vc jwt related that editors want to cover?

3. VCDM PRs.

Brent Zundel: next up is core spec.

Brent Zundel: https://github.com/w3c/vc-data-model/pulls?q=is%3Apr+is%3Aopen+sort%3Aupdated-asc.

Brent Zundel: PRs.
… for some of these we are at the point whether consensus is possible.
… if not possible, then suggesting closing PRs.
… first PR we are going to look at today.
… PR 1035.

3.1. Add rendering property to VC Data Model (pr vc-data-model#1035)

See github pull request vc-data-model#1035.

Brent Zundel: rendering property.
… a few approvals, a copule of folks requesting changes.
… my read is that this will be addressed by rendering property being added to the table of reserved properties.
… so anticipate being resolved by that.
… so, i don’t assume we need conversation on that.
… if no conversation, we can move on.

Orie Steele: There is also a new CCG work item proposal for the rendering property https://github.com/w3c-ccg/community/issues/244.

3.2. Add language clarifying multiple data schemas (pr vc-data-model#1023)

See github pull request vc-data-model#1023.

Brent Zundel: nobody wanted to speak to that.
… next up is PR 1023.
… gabe, what needs to be done to move this forward?

Kristina Yasuda: PR 1035 should be closed when rendering gets added to the reserved table.

Gabe Cohen: some discussion whether there should be a single credential schema possible per credential.
… or something else.
… i want to have one defined in the VCDM.
… but need to discuss what people think.

Dave Longley: we already have mechanism for starting a set.
… for the case where we have more than one thing in the data model.
… if we are saying if we want to support either way, ability to have multiple schemas in a VC, i would expect to use that mechanism.

Orie Steele: The current JSON-LD context does not define credentialSchema as a container… https://github.com/w3c/vc-data-model/blob/main/contexts/credentials/v2#L18.

Dave Longley: if not, if different credential schema types can become multitype schemas then i would expect that somebody creates a multi purpose mechanism.
… i don’t know if we need to invent a new thing to do that.
… not sure about the arguments and what are tradeoffs?

Gabe Cohen: not exactly an accurate representation.
… arguing for keeping complexitiy in specific type specifications.
… that is the purpose of having types for defining processing logic.
… want to avoid complextiy in data model.

Orie Steele: agree with gabe.
… dave is aruging that json-ld expansion can handle all of that.
… how to deal with polymorphism.
… let’s not do any json-ld processing.

Brent Zundel: if i understand correctly if multiple schemas are present and the processing rules that then apply that we limit schemas to a single schema differentiated by types, so processing rules for single schema can be defined by the type and we also recognize that multiple schemas can be developed by having a composed schema.
… fair summary?
… is this something we can come to some agreement?

Gabe Cohen: correct brent.

Dave Longley: i don’t know where i am at on this issue in particular.
… json-ld is sort of red herring.
… json-ld processors will find out a way to handle that no matter what we put into the spec.

Orie Steele: +1 dlongley.

Dave Longley: whether or not schemas are used is up to verifier.
… we can keep the data model simple and we can still put specific rules in specific type spec.
… and then up to verifier if follows rule or not.

Orie Steele: We are basically saying what is legal value for credentialSchema… object, array string or all of them.

Orie Steele: Sounds like we don’t need to say anything if the verifier just knows how to do JSON-LD processing.

Oliver Terbu: I am in favor of keeping complexity on the type side.

Dmitri Zagidulin: my initial thought was that multiple items in cred schema was initially confusing. but then realized that.
… easy to understand example is different schema languages where multiple schemas would make sense.

Dave Longley: +1 to dmitriz, multiple schema languages.

Dmitri Zagidulin: shacl schema or json schema or whatever schema language other come up with.

Gabe Cohen: dmitri, that can easily be handled within a type.

Dave Longley: decentralgabe: a new “multi” type will just need to be invented.

Dmitri Zagidulin: question is if we need to specify and or or semantic for that field.

Orie Steele: sure, so we say credentialSchema supports array, object or string… and let type handle multiple langauges?

Orie Steele: and the core data model does not comment on the complexity?

Gabe Cohen: yes, multi type -> complex rules in the data model.

Dmitri Zagidulin: Orie - that seems reasonable.

Dave Longley: data model doesn’t need those rules regardless.

Orie Steele: seems like we are really just saying “the core data model is JSON-LD” with extra words.

Brent Zundel: encourage folks to have that conversation.

Dmitri Zagidulin: decentralgabe - re “multiple schemas can be handled within a type” – yeah, that’s true too.

Brent Zundel: we would like to have this PR to move forward.

3.3. Add issuer dependent term handling (pr vc-data-model#1083)

See github pull request vc-data-model#1083.

Brent Zundel: next up is PR 1083.

Orie Steele: default vocab expansion in v2 context. it is there to make v2 context usage easier. this PR is about vocab expansion.
… we are trying to give good guidance.
… cleaning up PR.

Brent Zundel: feels like this PR is making progress. what are next steps?

Orie Steele: additional reviews would be helpful.
… want to remove JS from page.
… to prevent xss.
… no json-ld IRIs on the page.

3.4. (pr vc-data-model#1066)

See github pull request vc-data-model#1066.

Brent Zundel: next up is PR 1066.
… request for changes from manu but not here.

Orie Steele: goals.
… technical recommendation contains malformed JSON-LD.
… PR intends to remove 404.
… manu and dave and potentially others have a model of what example they’d like to see examples in the core spec.
… examples that contain multiple contexts.
… but i don’t like examples with multiple contexts.
… just having some conversation on different opinions on examples.

Dave Longley: just quick clarification.
… we should have examples of both.
… first examples should be issuer independent vocabs.

Orie Steele: I am also in favor of examples of both, as I stated.

Dave Longley: and then issuer dependent vocabs but we should also say that this could be a risk.

Orie Steele: Not sure I agree regarding the “market failures” statement.

Orie Steele: but i look forward to reviewing text that elaborates on that.

Dave Longley: we should fix context instead of deleting them from examples.

Brent Zundel: two paths possible.
… one PR gets merged and then issue gets created.

Orie Steele: There are already open issues that track handling examples.

Brent Zundel: agree that the links in the examples should work.
… the other option is to create another PR that fixes the broken links.

Dave Longley: .
… another PR somewhere.
… to define an examples v2 context.
… that has examples vocab in it.
… not looked into it recently.
… but if gets traction.
… then we can merge that.
… because fixes broken links.
… and then can iterate on context.

See github pull request vc-data-model#1110.

Dave Longley: to improve examples.

Orie Steele: ^ that is the PR he is refering too.

Brent Zundel: orie?

Orie Steele: haven’t had a chance to review it.

3.5. misc.

Brent Zundel: other PRs will be possibly addressed by reserved properties table.
… currently two PRs, 1097 and 1108.
… attempting to add properties to table.
… encouring to review those PRs.

4. Zoom reference change.

Ivan Herman: in calendar invite i’ve changed already to use new zoom url. please don’t put URL in public channel.
… refresh calendars if you are subscribing to W3C calendars.
… new URL is there.

Brent Zundel: please refresh invites to get the new zoom URL.

5. Resolutions