Verifiable Credentials Working Group Telco — Minutes

Date: 2024-12-04

See also the Agenda and the IRC Log

Attendees

Present: Ivan Herman, Brent Zundel, Ted Thibodeau Jr., Joe Andrieu, Mandy Venables, Hiroyuki Sano, Wesley Smith, David Chadwick, Kevin Dean, Manu Sporny, Benjamin Young, Jennie Meier, Michael Jones, Will Abramson, Steve McCown, David Lehn, Nicholas Steele

Regrets:

Guests:

Chair: Brent Zundel

Scribe(s): Manu Sporny, Kevin Dean

Content:


Brent Zundel: Welcome, this is the VCWG meeting. We have our agenda – ask if anyone wants to talk about anything else, and then controller document rename.
… As we go into the topic, I’ll layout what I’m looking for there.

Ivan Herman: short question on something.

Brent Zundel: ok, that should be fine.

1. SM2 as Multikey.

Ivan Herman: Our Chinese colleagues have received a question/request from Chinese Expert who defines Multikeys for SM2, seems to be a Chinese alternative to elliptic curves, can it be added to the table for Multikeys?
… To be precise, question is around DID spec, but by now, the place is in the controller document. I don’t know what the official answer is.

Manu Sporny: All hats off, yes, it’s easy to add it and I was expecting the request to come in at some point.
… I can add it to the controller document. It’s widely recognized in China, it’s supported by their national standards body.

Ivan Herman: Raise an issue in the controller document repository. I will relay the answer.

2. Controller Document Rename.

Brent Zundel: ok, seeing no one else on the queue, let’s go into our topic for today.
… I’m seeing 14 people on the call today, currently.
… anyone is free to make a proposal, anyone is free to queue, anyone is free to talk about other proposals people have made.

Kevin Dean: Procedural question, so perhaps better to take me first.

Brent Zundel: my plan is to run queue, when a proposal is made, count the plusses, with 14 people, I expect to see at least 8 pluses and no minuses. If I see minuses, I’ll ask if they see it’s a formal objection, and if not, if there is a single or just a couple, we can move forward. That’s what I’m leaning toward as far as guidance on whether to determine we have consensus or not.

Kevin Dean: Quick procedural question, regarding pluses, 14 people in meeting, some companies represented by more than one person, how does that work? Since company is registered, not individual, how do pluses and minuses work in those circumstance.

Brent Zundel: Previously in the group, we’ve allowed people to indicate preference, if this were to turn into a vote, I’ll have to be clear that individual companies need to be represented, if that’s what people want to do today, happy to try to do the math to figure out how to do that.

Manu Sporny: I had proposed something about how we go about this today that is aligned with your proposal.
… The first being that I don’t think you can formally object to the title of a spec. The last time we asked for formal objections, but that’s not what we’re looking for today.

Proposed resolution: Close issue #100 with no further action on the renaming topic. (Michael Jones)

Manu Sporny: Today, we’re looking for the proposal that has the last number of -1s. Those who -1 will be asked to clarify their objection.
… I feel like formal objections for the title of the spec are totally off the table. The -1s indicate that you really don’t like it but there should be no formal objection.
… I’m fine with multiple people providing input and one vote per org.
… I think we pick the one with the least number of objections, but I don’t know if that’s aligned with Brent’s position.

Brent Zundel: All I’m trying to say is that if the group decides on a different name, that’s fine, in my view, I haven’t seen that yet.
… If the group all agree on a different name, that’s fine. So far, in my view, I haven’t seen that. Others say they have.

Joe Andrieu: I would like to follow process, when it gets to a point where we have a vote, we might do a poll first, ultimately, I think it meets process – one vote per member, IE counts as one vote.
… Other thing to push back to, you can’t formally object to something, you can formally object to anything under process – spoke w/ Florian and PlH about it – not aligned, but Florian says anyone can object to anything, PlH noted he doesn’t necessarily agree.

Brent Zundel: for my benefit, say who will indicate preferences.

Michael Jones: I’ll note that when we ran the polls, no poll for a name change gathered a majority of the recipients. Clear signal to not change name to anything in particular, I think we should be done. Proposal to that effect in the chat that I’d like to run.

Brent Zundel: I’ll run through queue and then run the proposal.

Ted Thibodeau Jr.: What Joe said, the process says that any decision is subject to FO, that includes a decision to title a spec. It does seem like a heavy lift for that decision, but it is a decision. If someone wants to make that heavy lift, they can make it happen.

Ivan Herman: something I noted on GitHub, when we come out of the call, is the decision final or do we have an implicit CfC via email until next meeting?
… we can decide that with this decision, we close no matter the outcome, but we have to agree on that first. Otherwise, we keep to the charter text on this.

Brent Zundel: As much as I’d like to proceed in that way, my recollection is that any decision made has a week of time where people can object.
… I am hesitant to act in a way that differs from our operating mode.

Ivan Herman: I agree.

Proposed resolution: Close issue #100 with no further action on the renaming topic. (Brent Zundel)

Michael Jones: +1.

Joe Andrieu: -1.

David Chadwick: -1.

Brent Zundel: +0.

Manu Sporny: -1.

Ivan Herman: 0.

Steve McCown: 0.

Hiroyuki Sano: 0.

Jennie Meier: 0.

Ted Thibodeau Jr.: +0.

Brent Zundel: I believe we have 11 organizations, we’d have to see six positive indications and no negative indications for a proposal to be clearly resolved.

Ivan Herman: I counted 11 as well.

Brent Zundel: This proposal doesn’t resolve, we have a single +1.

Michael Jones: What that result tells me, given that it was predominance of zeros, that a lot of people don’t care about this.

Brent Zundel: I did not mean to imply that only folks w/ proposals can run things… anyone can join conversation about title of the specification.
… We have three proposals to put forward, want to invite people to speak to proposal to add themselves to the queue.

Manu Sporny: I just want to highlight some of the thinking behind each proposal.
… The first is just to establish if we want to keep the current title.
… If we see a lot of +1s and few or no -1s or 0s, we should keep it.
… We may find ourselves in a situation where we don’t have a title and so we change the title to “Identifier document”.
… Third, we try to align with what Dave Longley and maybe Ted said, is to try to get alignment with the folks who want to keep identifier document, and add “controller” to the title. Keep both as a compromise.

Brent Zundel: First is to recap for those that have recently joined, today’s proposals… preferences are being made as organizations. Please coordinate with your company for single indication of preference or not. We can’t assume 11, we’re up to 13 or 12?
… A question to the group – if none of the proposals receive positive indications, of more than half of those organizations present, in my view, that’s an indication that none of the proposals have consensus. I would love to hear what people in the group feel the most appropriate action as chair should be in that event. Inclination as chair is to say there is no consensus and move on.

Michael Jones: If there is no consensus, we should move on.
… then move to CR.

Manu Sporny: If there’s no proposal that we agree on, we keep the existing title. That’s one of the arguments being made.
… If we can’t change the title to any other title, then we don’t have a title that we agree to.
… This is about more than the title. We need to do an IANA registration and we need a title as part of the registration.
… If we don’t have agreement on current or new terminology, my position is that we have no decision and we will sit here until we have one.
… That’s my current interpretation, I realize that’s not everybody else’s, but I don’t think there’s a clean argument for what the fallback position should be.
… The other way that other working groups deal with this is to proceed with the ones that have the least number of objections. If that turns into a formal objection, we deal with that then.
… If we go with the one with the least number of objections and the most support, we can move forward.

Joe Andrieu: There is good language about the fallback in the W3C Process document, I think those are good rules to do the proposal brent, thank you. If those rules don’t meet rules, then whole process where there is dissent, and we should move on (agree to move on). Groups should favor proposals that receive weakest objections.

Michael Jones: I strongly disagree with what Manu said about there being no decision. Controller document came from two documents that had WG approval, one was VC DI and one was VC JOSE COSE. Manu and I as editors consolidated it leaving the name in place, the name was already in WG approved documents. Unless there is consensus to change the name, we already have a consensus name.

Brent Zundel: If there is not clear consensus on a proposal, I’ll declare as resolved the proposal with highest number of indications, least number of negative indications. I think that’s in line with what people have said and W3C Process. I’d like to move to the proposals.
… If it was clear we had consensus on the name, we would not be having this discussion.

Proposed resolution: Keep the “Controller Documents” title. (Brent Zundel)

Brent Zundel: +0.

Manu Sporny: -1.

Joe Andrieu: -1.

Hiroyuki Sano: 0.

Ivan Herman: 0.

David Chadwick: -1.

Jennie Meier: -1.

Ted Thibodeau Jr.: +0.

Steve McCown: 0.

Michael Jones: +1.

Brent Zundel: This proposal doesn’t have consensus.

Proposed resolution: Change the title of the “Controller Documents” to “Identifier Documents”. (Brent Zundel)

Michael Jones: -1.

Brent Zundel: +0.

Joe Andrieu: +1.

Manu Sporny: -1.

Jennie Meier: +1.

David Chadwick: +1.

Hiroyuki Sano: +1.

Steve McCown: +1.

Ivan Herman: 0.

Ted Thibodeau Jr.: -0.

Brent Zundel: five +1s, 3 0s, and 2 -1s – my suggestion is we run the other proposal to see how it compares before talking w/ people w/ indication of -1s.

Proposed resolution: Change the title of the “Controller Documents” to “Controllable Identifier Documents”. (Brent Zundel)

Manu Sporny: +1.

Brent Zundel: 0.

David Chadwick: +1.

Hiroyuki Sano: 0.

Ivan Herman: 0.

Joe Andrieu: +1.

Steve McCown: 0.

Ted Thibodeau Jr.: +0.

Jennie Meier: +1.

Michael Jones: -1.

Brent Zundel: the first proposal has more support and more objection. The first proposal is meeting qualification at beginning of call wrt. abstentions… obstentions do matter as they are not in support.

Ivan Herman: I voted zero everywhere, customary for company-based vote, W3C abstains. That’s the reason I voted 0.

Manu Sporny: Yes, ours is non-blocking.
… I know that Dave feels very strongly about “Identifier Document” in that it’s not very descriptive.

Brent Zundel: MikeJ, can you speak to your -1s?

Michael Jones: Identifier Document as a name is pretty meaningless, it’s not evocative as a name as what it does – titles should indicate what it does. This does not, the second name “Controllable Identifier Documents” as english is also strange, it raises the mental question of if there are “uncontrollable identifier documents”.
… In what sense are they controllable?
… The english result is non-sensical, I would object to having my name on either of those.

Brent Zundel: Is either of your objections weaker than the other?

Ted Thibodeau Jr.: “Identifier Controller Document”.

Michael Jones: They’re both non-sensical names, neither of those help.

Joe Andrieu: I’d look at those two -1s as more objections, and would support that.

David Chadwick: selfissued your arguments equally apply to the name Controller Documents.

Joe Andrieu: Also want to push back on what you’re saying MikeJ, it’s one of the hardest problems – that’s what we’ve been doing with DIDs – whole point is through proofs, you show control over an identifier. That is, in fact, quite semantically accurate that describes controllable identifiers and how you establish/prove control.

Proposed resolution: Change the title of the document to “Identifier Controller Document”. (Brent Zundel)

Brent Zundel: I’ve drafted as a proposal, TallTed’s comment.

Ivan Herman: 0.

Manu Sporny: -1.

Brent Zundel: 0.

Joe Andrieu: -1.

Michael Jones: -1.

David Chadwick: 0.

Jennie Meier: -1.

Steve McCown: -1.

Hiroyuki Sano: 0.

Ted Thibodeau Jr.: +0 … it was mostly an attempt to address mike’s objection to the other names.

Michael Jones: PROPOSED PROPOSAL: Change the title to Verifiably Controlled Identifiers.

Brent Zundel: I’m going to run proposal as just the title.

Ted Thibodeau Jr.: When resolved, do the full string.

Michael Jones: The problem with Controllable, in terms of the English, it raises the question of “uncontrollable identifiers”.

Joe Andrieu: rogue identifiers are identifiers you don’t control.

David Chadwick: q=.

Michael Jones: It has same linguistic problem, hope we don’t go with that, the “controllable” vs. “uncontrollable” distinction is real.

Manu Sporny: Just to respond, Mike, an uncontrollable identifier example is a Social Security Number.
… There are many uncontrollable identifiers that we use: URNs, SSNs, tax ID numbers, these are all uncontrollable because you can’t prove that you control it.
… If you assert it to someone, you can’t prove that you control it.
… The name “controllable identifier” is the thing that makes these types identifiers provable of having control over it.
… That is what this specification establishes. It isn’t a decentralized version of that, because it doesn’t have the characteristics of a DID.

Brent Zundel: We are at the tail end of our call, can we run proposals before we’re done.

David Chadwick: Wanted to address Mike’s point of “rogue identifier” – equally controllable as non-controllable – think rogue identifier is spurious.

Joe Andrieu: Maybe shifting term around document might open up some options – could run Controllable Identifier Descriptor? How control is done there.

Proposed resolution: “Controllable Identifier”. (Brent Zundel)

Brent Zundel: Can you draft proposals for that? I’m going to run one Dave proposed, one Mike proposed, then we can try to get others in.

Michael Jones: -1.

Manu Sporny: +1.

Joe Andrieu: -1.

Brent Zundel: 0.

Ivan Herman: 0.

Jennie Meier: -1.

Hiroyuki Sano: 0.

Steve McCown: 0.

Ted Thibodeau Jr.: -0.

David Chadwick: 0.

Brent Zundel: PROPSAL: Verifiably Controlled Identifiers.

Michael Jones: +1.

Brent Zundel: 0.

Joe Andrieu: -1.

Ivan Herman: 0.

Manu Sporny: -1.

David Chadwick: +1.

Jennie Meier: -1.

Hiroyuki Sano: 0.

Ted Thibodeau Jr.: +0.

Steve McCown: 0.

Proposed resolution: Controllable Identifier Descriptor. (Brent Zundel)

Brent Zundel: 0.

David Chadwick: 0.

Ivan Herman: 0.

Joe Andrieu: +1.

Steve McCown: 0.

Michael Jones: 0.

Manu Sporny: -1 we don’t use descriptor anywhere else :(.

Jennie Meier: 0.

Ted Thibodeau Jr.: -0.

Joe Andrieu: lol.

Hiroyuki Sano: +0.

Brent Zundel: I’m not seeing consensus on any of the last 3 proposals.

Proposed resolution: Controlled Identifier Documents. (Brent Zundel)

Manu Sporny: +1.

Brent Zundel: 0.

David Chadwick: +1.

Ivan Herman: 0.

Joe Andrieu: 0.

Jennie Meier: 0.

Ted Thibodeau Jr.: +0.

Steve McCown: 0.

Hiroyuki Sano: 0.

Brent Zundel: least objections, but active support for it is poor. Not seeing everyone’s vote either.

Michael Jones: -1 because it should be “Document”.

Brent Zundel: MikeJ? Comments?

Ted Thibodeau Jr.: -1 “Controlled Identifiers Document”.

Michael Jones: Should it be controlled “Identifiers”?

Joe Andrieu: There is only one identifier in the document that is what the document is about.

Michael Jones: What about the keys?

Michael Jones: I would support Controlled Identifier Document.

Proposed resolution: Controlled Identifier Document. (Brent Zundel)

Brent Zundel: Other documents would hav ether own documents that proved control.

Brent Zundel: 0.

Michael Jones: +1.

Manu Sporny: +1.

Ivan Herman: 0.

Steve McCown: +1.

Joe Andrieu: 0.

Hiroyuki Sano: +1.

Jennie Meier: 0.

Ted Thibodeau Jr.: +1.

David Chadwick: +1.

Brent Zundel: I’m not seeing anyone on the queue, not seeing anyone else making other proposals, this has 6 +1s, no indications against, clearly indicates consensus, we are resolved.

David Chadwick: clap clap clap.

Resolution #1: Change the title of the”Controller Document” to “Controlled Identifier Document”.

Brent Zundel: We have consensus.
… I hope to participate next week, but might not work out, next weeks call might be an issue. Please look for calendar invite update.


3. Resolutions