Verifiable Claims Working Group Telecon -- 16 May 2017

<RichardVarn> joined

<gkellogg> https://lists.w3.org/Archives/Public/public-vc-wg/

<liam> [to subscribe to the public list (for receiving) you can mail "subscribe" to public-vc-wg-request@w3.org but WG participants should not need to do this ]

<burn> scribe: ChristopherA

<gkellogg> Scribe User Guide https://www.w3.org/2009/CommonScribe/manual.html

<burn> This is also useful: https://www.w3.org/dpub/docs/scribing-guidelines.xhtml

engelke: Charles Engelke. We are involved in public agencies submit credentials

liam: I am your official W3C contact
... My background is in CS and typography and I am in charge of the XML work at W3C.

planning for f2f

<burn> other conference dates: https://docs.google.com/spreadsheets/d/19Ndqc5pLsTu2ZmP4Wy7OlMOmskQFHPh28sMjW3ugsww/edit#gid=0

<manu> +1 to not having a face-to-face yet, late September would work, but then TPAC is two months after that... so, might be better to have first face-to-face at TPAC.

richardvarn: we would like to consider having a F2F before the next big W3C plenary

burn: I dropped off in IRC a spreadsheet for other events

<Zakim> manu, you wanted to mention RWoT5 in Boston.

RichardVarn: Ideally we'd like to have an co-loading event where a coordinator is part of the event,

manu: #rebootingweboftrust is useful, but is not necessarly good for F2F, as we may also need a more casual F2F. #RWoT is a great place for bringing up to speed,
... That way we don't have to spend too much time at TPAC bringing new people up to speed.

<dlongley> ChristopherA: Basically the same thing. I'm running RWoT, so I'm quite open to offering help there. As Manu said, RWoT is not necessarily a casual event, we work hard to move these agendas forward, it's not a traditional conference.

<Zakim> burn, you wanted to mention that more people are likely to join so we should be planning anyway and to say also that we will get more work done in our own meeting than at TPAC

christophera: #RWoT is not a traditional conference, not a great for informal discussions.
... but I welcome a deep dive at #RWoT

RichardVarn: We should consider getting started planning before TPAC, even if we don't have a set time or place yet.
... TPAC is not ideal for getting work done. better for cross-collaboration and discussion.

<Zakim> gkellogg, you wanted to ask about TPAC plans

gkellogg: I have been in productive groups at TPAC. It depends on how much people are moving in and out. If sufficient core people it can work.
... typically groups get two days at TPAC

liam: We have booked slots, but we don't know what days, for 30 people. We can revise number up to 50, not less than 12.

Open mic

<manu> ChristopherA: We're focused on data formats and such... but there are also @contexts that we need to look at. We're adopting concepts from other groups... are we supposed to look at specific contexts that are specific to this group? Needs of this group?

<dlongley> i think "vocabulary" may be closer to what we're talking about here not @context

<dlongley> i would say we want a core vocab -- common to all verifiable claims -- and then show how to extend it in specific market verticals

manu: vocabulary work, my read on charter that is "in scope". Yes we will reuse, but we do need to move on these.

gkellogg: if we are talking about formats, if we are layering on top of JSON-LD and we would need to do a schema.

dlongley: The main thing we want is a core common vocabulary, and talk about how to extend it into market verticals.

<burn> https://www.w3.org/2017/vc/charter.html#deliverables

johntib: Several of us presented at IIW (joe, nathan, drummon, myself) to familiarize people with our work. There was a real disconnect with some of the objections. They said we were too "verbose" for what they were interested.

<JoeAndrieu> +1 verbosity of meta-data was an issue with several parties at IIW

johntib: my work is for very rich verifiable cliams. They are talking about verifiable tokens. Something we are not realy doing. We need to write some domain specific vocab, for instance small fields, it might be a fruitful area.

<nage> The folks at Consensus working on uPort (on Ethereum) are using very small claims for introductions and authentication type cases, and the terseness or compactness of these claims is very important for embedding them into QR codes and introduction mechanisms

<dlongley> a "JWT" is pronounced "JOT"

<nage> They are introduction IDs plus claims about what information is going to be shared over the connection

<dlongley> and a "JWT" is a "JSON Web Token" (https://tools.ietf.org/html/rfc7519)

johntib: there is a tie between the utility of a domain and its granularity. So people who want JWT can do microclaims, using it for web login, with one or two fields. The overhead of VC (not just vocab and size) but also the processing at runtime, which are others are confused with, or can be processed differently.

<Zakim> manu, you wanted to ask what our first deliverable is... and when we want to do that...

johntib: there are some practical elements that allow us to support the community at large.

manu: Interested. Part of our work is to harmonize between JWT and this is worthy work over the next few months.
... we do have to be very careful to not fall into the auth and webauth work — if we enter that territory there are other groups.

<burn> WG deliverables: https://www.w3.org/2017/vc/charter.html#deliverables

manu: I would like to try to understand the first deliverable.
... we need to know what to deploy, what deliverables, when, etc. We have some use cases to address, and when do we adopt those from the CG into the WG.
... we need to buckle down and work on them.
... do we want to change the work in any way.

RichardVarn: We have a start of a list (link above).

manu: The group is expected to do the first public working draft of the CG documents.

mattlarson: This is related to vocab, and tibbets in education, we would like to bring back to charter education cases for data model.

<burn> s/mattLarson/stonematt/

mattlarson: personally I've been wondering how does OBI spec overlays verifiable claims. Open Badges Initiative is suited for our work, prevents us from reinventing the wheel? Is is a blob into JSON-LD. How do these connect?

<manu> Yes, it's basically a blob that gets included in a Verifiable Claim, but the group should discuss the specifics.

<Zakim> JoeAndrieu, you wanted to bring up missing terminology

<RichardVarn> https://openbadgespec.org/

<JohnTib> +1 on discussing OBI / VC relationship

joeandrieu: Building on a discussion with JohnTib where the metadata is small for the data. If the claims what are issued, we haven't named that thing we are giving the inspector. In uprove these are called proofs.

<RichardVarn> we have used evidence as word for that in other contexts

<RichardVarn> and in this context

<manu> ChristopherA: We have been discussing selective disclosure... it's possible to have a very large Verifiable Claim Set, but only present a subset of that to another party.

<manu> ChristopherA: Where the signature can still be valid.

<dlongley> "IdentityCredential" or "EntityCredential" was a term used by the CG to refer to a container for claims that is digitally signed and presented to inspectors

<dlongley> another CG term for a collection of credentials was "IdentityProfile" -- a container for many "IdentityCredentials", each of which contains one or more claims.

<nage> The CL signature scheme Evernym has proposed follows this idea

<Zakim> burn, you wanted to talk about adopting CG drafts

<manu> ChristopherA: I'm very interested in this area, which is related to data minimization, where I can prove that I did something w/o exposing an entire transcript, for example.

<manu> ACTION: Manu to chase down IPR commitments. [recorded in http://www.w3.org/2017/05/16-vcwg-minutes.html#action01]

Link to the early partial JSON-LD sig: https://github.com/w3c-dvcg/lds-merkleproof2017

burn: The chairs have discussed first steps. First is to adopt the two CG docs once we have IPR commitments. Second is to work towards FPWG

<Zakim> manu, you wanted to note that we /do/ have a term for that... but one that may not have buy-in and to mention that it's hard to work on the docs/issues until we adopt the docs...

manu: We do have a term, but the term may not have buyin. We are missing some concepts, or at least people think we are missing concepts. The term for a set of claims: a claimset, or an entity credential.

<JoeAndrieu> there's a difference between claims issued as a set and those selected from multiple sources for presentation to an inspector.

manu: My suggestion those that care about terminilogy make a concrete proposal, Joe, dlongley, manu, christophera

<JoeAndrieu> yes. let's chat offline.

<nage> "attribute based credential" is a term that has been used in other places (relative to the selective disclosure variety of signature schemes)

(me too on terminology)

<nage> yes, lets chat about the terminology

<stonematt> +1 on preliminary adoption

scribe: we can't really add issues for the spec until group adopts the CG, "the WG will adopt the CG documents as soon as the IPR commitments are shaken out". Or do we have to wait.

+1 Yes, on chairs call.

<Zakim> manu, you wanted to argue against himself.

<dlongley> we can always adopt any other specs later and merge

manu: I put myself on the queue to argue with myself (laugh) We don't want to rush, in case we want some time for other orgs to join. One of them may have a spec proposal, that we may want to adopt. Counter to that is we are careful and say that we are not accepting the CG as working draft, but as an "under consideration" draft.

<burn> We want to move quickly, but we need to make sure that we are not viewed as trying to make decisions before others join

liam: The danger of moving ahead before IPR is that someone who owns patents might not make the statements. Thus you do need to wait for everyone to be clear.

<manu> ChristopherA: The group says that the Chairs may do this... if they think IPR have been sufficiently taken care of. We're not saying do it despite the IPR... just do it as soon as you've made the judgement call. I'm fine with it being considered as a "under consideration draft", so we can start processing issues, etc. In case there is another proposal that comes along.

<manu> ChristopherA: I'd like to keep it away from being an item that comes up in the next couple of weeks. The reality is that we've done the work... people may be fine w/ it being under consideration.

<burn> PROPOSAL: Chairs are approved to adopt the CG docs as starting points in the group once they believe IPR is clear.

<manu> +1

<gkellogg> +1

<dlongley> +1

<RichardVarn> +1

<JoeAndrieu> +1

<JohnTib> +1

<burn> +1

<stonematt> +1

<MattLarson> +1

<cwebber2> +1

<nage> +1

<engelke> +1

<burn> PROPOSAL ADOPTED

RESOLUTION: Chairs are approved to adopt the CG docs as starting points in the group once they believe IPR is clear.

- DRAFT -

Verifiable Claims Working Group Telecon

16 May 2017

Attendees

Contents

planning for f2f

Open mic

Summary of Action Items

Summary of Resolutions

Scribe.perl diagnostic output