The Verifiable Claims Task Force

A Task Force of the Web Payments Interest Group

Verifiable Claims Telecon

Minutes for 2016-06-21

Agenda
https://lists.w3.org/Archives/Public/public-webpayments-ig/2016Jun/0096.html
Topics
  1. Verifiable Claims Working Group Proposal
  2. Terminology Results
  3. Data Model Specification
  4. Use Cases
  5. Next Steps
Action Items
  1. Manu to put together a 9 month schedule/roadmap and publish to CG/IG.
Organizer
Manu Sporny
Scribe
Shane McCarron
Present
Shane McCarron, Manu Sporny, Adam Lake, Dave Longley, Christopher Allen, David Ezell, Dan Burnett, Eric Korb, Gregg Kellogg, Matt Stone, Drummond Reed, Nate Otto, Dave Crocker, Stuart Sutton, Carla Casilli, Richard Varn, David I. Lehn, Rob Trainer, Les Chasen, Rebecca Simmons, Matthew Larson
Audio Log
Shane McCarron is scribing.

Topic: Verifiable Claims Working Group Proposal

Manu Sporny: Thanks to everyone. we got the proposal together and handed over to the IG yesterday.
... we hope people are reading that now. Will discuss at the f2f on 1 July.
... all documents are review ready, but we might want to make some refinements over the next couple of weeks.
Adam Lake: Thanks CarlaCasilli for your input on the primer
Manu Sporny: Some new implementor committments that we need to add. Data model and use cases are being edited.
... Joe had provided some ideas for how to restructure the use cases.
... terminology is consistent. Everything hangs together. We have demonstrated that people want to do it, and there are large organizations that will implement it.
Dave Longley: What's the next steps?
Manu Sporny: The reason we have all these documents is to create a proposal to create a working group.
... a working group is mandated to create a technical recommendation for ratificaiton by the membership.
Christopher Allen: When will the vote be?
Manu Sporny: At the f2f. But there might not be critical mass at that meeting. So if in the room we decide, we will still need to take it to the mailing list for a call for consensus.
... after that we need to get over a hurdle of W3M to put it forward to the membership.
... W3C representatives will be in the room at the f2f and will ask questions.
... might go to a vote end of July, early August. Vote goes out for a month. Some back and forth after that.
... at the end of August hopefully an approved charter. September is the W3C technical plenary meeting in Lisbon. That would be the first opportunity to meet face to face.
... We would need a two day agenda for that meeting.
Shane McCarron: TPAC fills up quickly, if we want a face-to-face - we need to allocate space. [scribe assist by Manu Sporny]
Christopher Allen: Does anyone have tpac reg URL?
Manu Sporny: As a CG we cannot ask for space. people have been reluctant to assume that this will be successful. dezell is there a way to ask for space?
David Ezell: I am not sure that between now and the f2f it will make a big difference. W3M will say it is premature today. But it is a fair question on 1 July depending on how the conversation goes.
... not sure how much attention we can get between now and then anyway.
Dan Burnett: They may say it is premature, but if we don't prep them for it they will later claim we should have asked earlier, so we need to prep W3M anyway
Manu Sporny: We could plan to meet there anyway. They are allowing CGs to have space and time at TPAC. We could plan to have a f2f at TPAC using that mechanism.
... if a working group is formed, we could just shift it.
... thoughts from the group?"
Shane McCarron: That's clever. [scribe assist by Dave Longley]
David Ezell: I think you should go ahead and do that.
Dan Burnett: +1 To planning to meet anyway
Manu Sporny: There are people who are going to be split between VC, IG, and Web Payuments. This is a problem.
Shane McCarron: This is a classic problem.
David Ezell: It is getting worse and worse
... it is a good / bad situation
Dan Burnett: Yes, it's a good problem to have
Shane McCarron: Let's put this roadmap out on the Web. [scribe assist by Dave Longley]
ACTION: Manu to put together a 9 month schedule/roadmap and publish to CG/IG.
Christopher Allen: I heard a little more from Dave's question than just the schedule. I would like to hear who is committing development resources etc. Who can we cooperate with?
... are there P2P things we can do to verify claims?
... are there ways we can discuss coding on the agenda at some point?
Manu Sporny: There is a list of who is going to implement. These are committments to implement if the work is successful. Having said that, there are organizations that have been doing this for years already.
... Digital Bazaar is one such. I am not going to call out others that are doing things in the next three months. They can speak for themselves.
Eric Korb: Accreditrust is

Topic: Terminology Results

Manu Sporny: Picked "inspector" for what used to be called "consumer". That raised another question... So there is a new term "verifier" for the thing that really does the verification.
Gregg Kellogg: +1 For consistency
... there is a conflict between "identity profile" and "entity credential".
... this might cause some further discussion. But we don't need to have the debate now.
Dave Longley: +1 For consistency
Matt Stone: +1 On process
Drummond Reed: The conflict between the terms needs to be fixed.
Nate Otto: Hehe, wish I could have been on the call to push for consumer, but I'm prepared to adapt Badge Alliance terminology to "Inspector" to keep that in line. +1 for consistency. Identity is better, assuming we're clear that any entity can have an "identity"
Dave Longley: +1 For Identity as well.
Gregg Kellogg: +1 For identity too
Manu Sporny: We need to make a decision about whether we are talking about identities or entities. We don't need to do it today.
Manu Sporny: We are zeroing in on terminology that the community seems to be okay with.
Dave Crocker: Terminology: choices need to be driven less by personal preference and more by likely uptake. Who will use it and what are they most likely to be comfortable with and likely to use?
Stuart Sutton: +1 For consistency (prefer Identity)
Manu Sporny: It is arguable that this debate has been going on for 15 years.
Dave Longley: +1 For voting on the two names for inspector and verifier
... it was good to have a large sample when doing a poll like that.
Dave Longley: (So voting on pairs)

Topic: Data Model Specification

Dan Burnett: Did some major editorial terminology. The terms we use really effect this document
... the terms cause problems with references. But I think it hangs together much better than it did.
... there is a reference problem that we need to get sorted out. Shane and I are working on it.
Dave Longley: "Entity credentials are user-centric. This refers to an architecture where:" <-- maybe avoid "user-centric" and just say "Entity credentials are part of an architecture with the following properties:"
... there is a reference to the use cases document that needs to be updated.
Dan Burnett: I have one question for the group. Initially I included "entity credential" and the "identity profile". I included both so that there was a way to understand the overall context.
... the concern I have is that people will look at this will say "you are standarizing verifiable claims, but here is an identity profile as well".
... my question is should I remove it, should I push it to an appendix, or should I leave it alone?
Shane McCarron: I'm inclined to leave it in, it makes sense to have it in there, without it it isn't clear how this stuff maps together. If there is concern around exposing that concept, lets add notation to say it's just for context - wouldn't want to lose the example. [scribe assist by Manu Sporny]
Dan Burnett: Dlongley, missed the user-centric. Will either remove as you suggest or change to self-sovereign
Manu Sporny: To echo what Shane said... it is a part of the bigger vision. Being buried in the spec is probably deep enough.
... we dont want to distract from VC for the initial work, but we know that an identity profile is a goal and being able to express it is a requirement
Dan Burnett: Probably best to avoid a term at all there and just list the properties, -- we're already listing all that out so it's less controversial to avoid the term [scribe assist by Dave Longley]
... I don't think that the AC reps are going to read it in detail. And if they do then they are going to realize that this is a proposed spec and the WG will do whatever they do.;
Dan Burnett: @Dlongley okay
Dave Longley: +1 To keeping Identity Profile in there for now.
Nate Otto: +1 Keep it in the document
... an AC member will know that if they want to effect change they should staff the working group.
Stuart Sutton: +1 Keep
Gregg Kellogg: +1 To keep
Drummond Reed: +1 To keep
Matt Stone: +1 Keep - ultimately it's the key to a functional eco system.
Dan Burnett: Note that the doc does say that we will ONLY be working on VC stuff.
Manu Sporny: We talk about identity profile in the abstract. We may want to use some of the language from the architecture document in the abstract to ease the way into the document.
Dan Burnett: The abstract came from an older document. I will adjust it.
Dave Longley: +1
Manu Sporny: What are the next steps?
Dan Burnett: Actually... not clear given the schedule. I am hesitant to change anything other than editorial.
Manu Sporny: Good plan.

Topic: Use Cases

Manu Sporny: It would be good for poeple to review again.
Nate Otto: I feel bad having my name on there as an Author given how little time I've been able to contribute to Use Cases during calendar 2016. Is there anything I could do to help next week or this coming weekend (not this week)?
Nate Otto: "Consumer" language still appears in use case. todo: change to Inspector?
Manu Sporny: We have an opportunity to create / modify use cases through september.
...once the working group starts, we can update the doc as well.
Nate Otto: Thumbsup, ShaneM

Topic: Next Steps

Manu Sporny: Thanks again everyone. I think this is one of the best proposals I have seen.
... what do we want to do?
... plan for success. Two day agenda for TPAC. Rely on the web payments working group schedule from their first f2f as a model.
... when a new wg starts 60% of the people in the room will have no background.
... We should continue to refine the spec and use cases.
... ChristopherA pointed out that we could start doing preliminary implementations.
... there are a number of other things that have come up. We have rebooting web of trust, ID 2020, refugee use cases.
... there are a bunch of components that will factor into success over the next 5 years. WebDHT, registries, etc.
Drummond Reed: Respect Network will be working on an implementation of DIDs
... what do people think we should be doing?
Shane McCarron: You said "group". Do you mean VCTF or something else?
Manu Sporny: I mean the credentials community group
Carla Casilli: Die ... more like evolve
... the VCTF is going to evolve within a month. Assuming the vote is successful.
... but there is still a Credentials Community Group that is around to stage work.
Nate Otto: I think the people on this call should focus on answering questions from people who are considering voting on the proposals, keep showing our interest and activity through discussing these documents & proposals.
... I am unfortunately going to continue to be vague. The group is amorphous.
Christopher Allen: We should try to get the MIT group that is looking at Badges to get them into the fold
Carla Casilli: Umm, not sure we can call them badges. they're certificates.
Manu Sporny: MIT Media Lab + Learning Machine
Manu Sporny: Chainpoint is Proof of Publication
Nate Otto: I'm working with them for making sure their 2.0 is standards-compatible, uses Open Badges data classes. Can introduce anybody who needs it to the MIT Media Lab & Learning Machine folks.
Christopher Allen: I would like to make a claim not just about a security reviewer, but also a source code repository...
... I want to do this soon because these are the things developers care about now and they are likely to be early adopters.
... work out the issues with a limited community before we go to broader deployment.
Nate Otto: Kerri Lemoie working with the BadgeChain group will be at that meeting, I believe.
Carla Casilli: This idea blockchain and badges is something we are working on too. The conversations we have had are in this area. So I am cautiously voting in this direction.
... we have not yet decided if blockchain is the best direction. We have a great amount of interest in picking apart what blockchain adds but we are not specifically interested in FinTech.
Christopher Allen: This group makes a big deal about using blockchain, but it is just really putting a has on the blockchain.
Carla Casilli: Right, we're a lot more investigative about exploring the future of badges related to the future of blockchain
Carla Casilli: Where we = BadgeChain
... this is just proof of existence.
Carla Casilli: Yep
Manu Sporny: There are digital signatures that we are working on with a number of companies about proof of publication. Put into various chains. Ethereum, bitcoin, ipfs.
Carla Casilli: Yes, BadgeChain is very interested about Flex Ledger. all of these things are interesting to us.
... flex ledger is designed to publish verifiable claims into the ledger. It is a more flexible way of publishing what people have learned, what they have accomplished, etc. That work is ongoing.
Manu Sporny: There is also the did stuff. Drummond?
Drummond Reed: We welcome the input. We are just finishing the first 4 milestones that we are doing for DHS. We are researching DIDs and what the best way is to format identifiers and objects.
... if there are people with strong feelings about this we would love to interview you.
... we are hoping to produce a recommendation that will move into a spec in the appropriate body.
Manu Sporny: As far as verifiable claims... what should we working on next?
... a number of us are on the road for the next two weeks, so cancel those meetings.
... we should be working on a F2F agenda. TPAC is a good opportunity to do that.
... the way this stuff happens is that people step up and do the work. People who proposed things should bring forward proposals.
Christopher Allen: The next #RebootingWebOfTrust is September 28,29,30th in SF, http://www.WebOfTrust.info
Manu Sporny: DID stuff is central to this group.
... there is no finalized mechanism way to allocate those.
Christopher Allen: We will be doing more with DIDs, smart signatures, proof of existence, peer claims, etc.
... We have references to them in our specs though, so it feels like it is next.
Christopher Allen: Rebooting web of trust is in late september. Right after TPAC.
Drummond Reed: It is definitely a high priority for Respect Network and several of our partners as well.
Christopher Allen: (Dave crocker - sent you email about scheduling phone call)
Nate Otto: Thanks all! Safe travels.
Manu Sporny: So - brief hiatus. next meeting second week of July.