14:44:39 <RRSAgent> RRSAgent has joined #vcwg
14:44:39 <RRSAgent> logging to http://www.w3.org/2017/10/03-vcwg-irc
14:44:45 <burn> rrsagent, draft minutes
14:44:45 <RRSAgent> I have made the request to generate http://www.w3.org/2017/10/03-vcwg-minutes.html burn
14:44:52 <burn> rrsagent, make logs public
14:46:01 <burn> burn has changed the topic to: 2017-10-03 Agenda: https://lists.w3.org/Archives/Public/public-vc-wg/2017Oct/0000.html
14:46:11 <burn> Agenda: https://lists.w3.org/Archives/Public/public-vc-wg/2017Oct/0000.html
14:46:21 <burn> Meeting: Verifiable Claims Working Group
14:46:35 <burn> Chair: Matt_Stone, Dan_Burnett, Richard_Varn
14:56:15 <burn> present+ Dan_Burnett
14:57:40 <stonematt> stonematt has joined #vcwg
14:57:55 <burn> present+ Tzviya_Siegman
14:59:48 <reto> reto has joined #vcwg
15:00:24 <gkellogg> gkellogg has joined #vcwg
15:00:31 <burn> present+ Nathan_George
15:00:53 <dlongley> present+ Dave_Longley
15:01:24 <burn> present+ Matt_Stone
15:02:08 <burn> present+ Reto_Gmür
15:02:18 <nage> nage has joined #vcwg
15:03:00 <nage> present+ Nathan_George
15:04:26 <burn> present+ David_Chadwick
15:04:31 <DavidC> DavidC has joined #vcwg
15:04:52 <DavidC> I have just circulated my first go at answering the privacy questions to the list
15:05:03 <gkellogg> present+ Gregg_Kellogg
15:05:44 <burn> scribeNick: burn
15:06:02 <stonematt> Agenda: https://lists.w3.org/Archives/Public/public-vc-wg/2017Oct/0000.html
15:06:15 <burn> Topic: Agenda review, Introductions
15:06:30 <TallTed> TallTed has joined #vcwg
15:07:11 <bigbluehat> present+
15:07:16 <bigbluehat> present+ Benjamin_Young
15:07:19 <burn> matt: any mods to agenda?
15:07:21 <burn> (silence)
15:08:08 <burn> reto: (unintelligible)
15:09:14 <dlongley> reto: I'm with a company in Switzerland, work with Facts Mission.
15:09:19 <dlongley> reto: Working with Linked Data.
15:10:06 <reto> I'm connected via browser
15:10:25 <burn> Topic: TPAC topic review & TPAC Planning
15:10:40 <burn> link is https://goo.gl/8voHZS
15:10:51 <reto> The company is called FactsMission which wants to use Linked Data to help distinguish Facts from noise
15:12:20 <TallTed> present+ Ted_Thibodeau
15:12:37 <burn> matt: a few items we should discuss.  Several unprioritized issues at the bottom of the list.  We reduced the open-ended time and included some of the specific items we need for our charter.  Some other items were assigned to us such as privacy and uprove.  We need to make progress on those as well.
15:13:23 <burn> ... there may be others as well.
15:13:33 <stonematt> q?
15:13:36 <dlongley> q+
15:14:21 <burn> chadwick: Re integration with SAML.  Their model is very different from ours.  What are we proposing to do?  In our model the user controls VCs, but not in SAML.
15:14:40 <burn> ... what is it we're trying to acheive?
15:15:27 <burn> matt: agreed. But AC members have brought this up as a question, and thus it ended up in the charter.
15:16:00 <burn> ... we just need something that addresses this concern.
15:16:02 <dlongley> q?
15:16:12 <gkellogg> gkellogg has joined #vcwg
15:16:13 <burn> chadwick:  i already have text.
15:16:26 <burn> matt:  will add you as discussion leader for this topic, David.  Thanks.
15:16:40 <nage> +1 to use DavidC's text as a starting point and getting it out to the list for review
15:16:49 <dlehn> present+ David_Lehn
15:18:21 <stonematt> q?
15:18:27 <stonematt> ack dlongley
15:19:12 <burn> dlongley: the concern that came up was how this work was different from SAML and OpenID connect.  We can answer that and explain how our model can be encapsulated in SAML if you want.
15:19:28 <burn> chadwick: I won't be at TPAC.
15:19:47 <dezell> dezell has joined #vcwg
15:19:48 <burn> dlongley: then at least having your paper could help the discussion.
15:20:07 <burn> ... maybe we need to spend time just on "items in the charter".
15:20:13 <stonematt> q?
15:20:16 <stonematt> q=
15:20:19 <stonematt> q+
15:20:20 <burn> ... as a group rather than as individual items.
15:20:34 <burn> chadwick:  we can address this topic during the meeting
15:21:07 <burn> matt: Agree with dlongley.  There may be other items we don't have listed.  The intention was to cover all of them.
15:21:23 <burn> chadwick:  include UProve and anonymous credentials as well.
15:21:37 <burn> matt: do you have insight into UProve as well?
15:21:38 <stonematt> q?
15:21:48 <stonematt> ack stonematt
15:21:53 <dezell> present+ dezell
15:22:01 <burn> chadwick:  I did a comparison of UProve and SAML, but I couldn't do the crypto comparison
15:22:16 <nage> I can help with UProve and Idemix crypto pieces
15:22:29 <burn> matt: it was widely referenced in the charter.  We have to decide in the group how much we want to speak to it.
15:22:41 <dlongley> nage: will you be at TPAC?
15:22:45 <nage> yes
15:22:48 <burn> chadwick:  have written a research proposal of how ?? could fit into VC
15:22:52 <stonematt> s/widely/lightly/
15:22:52 <dlongley> nage: ok great ...
15:23:09 <burn> ... won't know until 2018 whether that is approved
15:23:33 <reto> q+
15:23:39 <reto> q-
15:23:56 <burn> matt: if you know of other items please add them to the Google doc with a B next to them
15:24:43 <burn> matt: there are several items on here for brainstorming.  Asking for +1 for items to cover and in which order to cover them.
15:25:20 <burn> ... We are currently over budget for time, so need to figure out which are most important to cover.  Understand that many participants are absent today due to RWoT.
15:25:44 <burn> ... Also a -1 for anything we shouldn't cover.  Feel free to do that with your name directly in the Google Doc.
15:25:55 <burn> ... Will add a vote column.
15:26:34 <liam> present+ Liam
15:26:38 <burn> Doc link again: https://goo.gl/8voHZS
15:27:37 <burn> dlongley: another possible topic is how to represent credentials using a consistent UI.  Further down the road, but should consider in the data model work.
15:27:46 <burn> ... a visual representatiotn.
15:28:08 <stonematt> q?
15:28:16 <burn> ... would be great to figure out how to include this info in the claim itself so they can have a consistent look to them.
15:28:45 <burn> liam: how do you mark properties as translated or not translated into different languages.  may be related.
15:28:50 <burn> dlongley:  yes, it could.
15:28:59 <dlongley> https://github.com/w3c/vc-data-model/issues/42
15:29:19 <burn> ... not much discussion yet, but talks about integration with web components for consistent display.
15:29:24 <stonematt> q?
15:29:27 <stonematt> ack liam
15:29:27 <Zakim> liam, you wanted to ask if visual rep relates to i18n
15:29:46 <burn> matt: how related is that to what open badges does?
15:30:09 <burn> dlongley:  somewhat, this is more related to creating full web components that are shareable.
15:30:20 <tzviya> q+
15:30:26 <burn> ... how do we integrate with other work on web components.
15:30:59 <burn> tzviya: the RA21 group is focusing on a singular UX across the community, so they would be interested.
15:31:34 <burn> dlongley: Open Badges community has an idea of embedding into images.
15:31:43 <burn> ... this work could make that more consistent.
15:31:55 <burn> matt: would this be new metadata?
15:32:15 <burn> dlongley: that's an open issue.  Should look at other work in this area.
15:32:55 <burn> matt: another topic related to privacy is ODRL.  Would this be worthwhile to consider for TPAC discussion?
15:32:56 <liam> [we'll need i11n and accessibility signoff for this spec at some point before leaving CR, so we need a clear story about UX before then]
15:32:58 <DavidC> -> Burn. ?? = anonymous credentials
15:33:04 <tzviya> https://www.w3.org/2016/poe/wiki/Main_Page
15:33:22 <burn> dlongley:  don't know who could lead, but could be part of vocabulary discussion.
15:33:34 <stonematt> q?
15:33:38 <burn> s/??/anonymous credentials/
15:33:44 <tzviya> q-
15:34:40 <burn> matt: Seeing lots of voting coming in.  Will review next week and order so we can assign timeslots.  Thanks all.
15:34:57 <stonematt> q?
15:35:16 <burn> Topic: Readiness for Privacy Group exposure
15:35:42 <burn> matt: we need to reach out to PING.
15:36:13 <burn> ... We will have them do an initial review of our data model doc.  Is it ready, or is there something else we need to do first?
15:36:21 <DavidC> q+
15:37:07 <burn> chadwick: I've started a review from that standpoint.  Have sent in several issues.  Until those issues are resolved it shouldn't go for review.
15:37:35 <dlongley> potentially related to rendering - custom elements: https://w3c.github.io/webcomponents/spec/custom/ and web packaging: https://www.w3.org/TR/web-packaging/
15:37:37 <burn> ... should also include answer to privacy questions.
15:37:46 <stonematt> q?
15:37:53 <stonematt> ack DavidC
15:38:12 <burn> matt: david, have you talked with Manu yet?
15:38:27 <burn> chadwick: no
15:39:04 <burn> matt: chairs will ask Manu to review
15:39:12 <burn> davidc: will finish reviewing the doc as well.
15:39:25 <stonematt> q?
15:39:31 <burn> ... and post any other issues from that review
15:39:31 <tzviya> note that Web Packaging work is shifted to https://github.com/WICG/webpackage
15:39:43 <burn> davidc: we will rely on you and Manu to tell us when the doc is ready to share with PING
15:39:53 <burn> .. we are on hold there until we hear from you
15:40:05 <stonematt> q?
15:40:07 <burn> s/davidc:/matt:
15:40:29 <burn> Topic: Data Model Spec curent milestone issues
15:40:40 <burn> link: https://github.com/w3c/vc-data-model/milestone/3
15:40:44 <stonematt> milestone issues: https://github.com/w3c/vc-data-model/milestone/3
15:41:11 <burn> matt: The chairs believe some can be closed
15:41:46 <stonematt> q?
15:42:04 <burn> dlongley: Issue ??? is just waiting on a PR.
15:42:28 <burn> davidc: the comments I had on vocab were addessing this issue
15:42:36 <stonematt> q?
15:42:39 <burn> dlongley:  agreed, will include in the PR
15:42:46 <burn> matt:  issue 59
15:43:04 <burn> ... manu said +1 to close.  any objections?
15:43:10 <burn> ... (none)
15:43:20 <burn> dlongley: will comment and close
15:43:26 <burn> matt: issue 33
15:44:00 <burn> ... active discussion still on this one.  anything needed to keep it moving forward?
15:44:09 <reto> q+
15:44:13 <burn> dlongley: can't close 59!!! someone else will need to
15:44:17 <stonematt> q?
15:45:38 <reto> typing then: RDFS vocabulary was the thing I missed most
15:45:47 <reto> a vocabulary independt of any serialization
15:45:56 <burn> reto: needs an RDF schema.  lacking that is a big issue
15:46:02 <reto> so I suggest to discuss it at TPAC
15:46:11 <stonematt> q?
15:46:15 <stonematt> ack reto
15:46:22 <dlongley> https://github.com/w3c/vc-data-model/issues/33#issuecomment-274268657
15:46:49 <burn> dlongley:  this is part of our charter.  issue 33 talks about this.  We plan to add a schema but haven't tried to build one yet .
15:47:05 <stonematt> q?
15:47:06 <burn> ... formal semantics are tribal but need to be nailed down per the charter.
15:47:31 <burn> matt: could we invite Reto to help with this?
15:47:37 <stonematt> q?
15:47:42 <burn> reto: yes would be glad to help
15:48:07 <burn> matt: dlongley, is the location for this accessible?
15:48:31 <burn> dlongley: once conferences are done, we can create a place for PRs to happen.
15:48:49 <stonematt> q?
15:49:01 <burn> matt: issue 25
15:49:19 <burn> ... where are we on this one?
15:49:25 <stonematt> https://github.com/w3c/vc-data-model/issues/25
15:49:45 <burn> dlongley: this may be resolved by the PR we are working on for issue 66
15:49:49 <burn> s/???/66
15:50:15 <burn> ... we have been using different terms here and need to fix in that PR.
15:50:33 <stonematt> q?
15:50:35 <burn> matt:  can you please comment in this issue (25) so we can track their relationship?
15:50:40 <burn> dlongley: yes
15:51:29 <burn> matt: have we achieved the milestone when these are closed?
15:51:41 <burn> dlongley:  yes, as long as the test suite reflects that
15:52:00 <burn> matt:  we may be able to publish this milestone doc before TPAC, which would be good
15:52:14 <burn> dlongley:  pretty tight schedule, so might not happen.
15:52:18 <burn> Topic: Test Suite Progress
15:53:13 <burn> dlongley: both Chris and Manu are at RWoT, so probably not much progress, and definitely no feedback until the conference is done.
15:53:23 <stonematt> q?
15:53:23 <burn> matt:  will leave on the agenda each week as a standing item.
15:53:54 <burn> Topic: Future Agenda Topics?
15:54:07 <burn> matt: anything else to add in future weeks?
15:54:12 <burn> ... (silence)
15:54:38 <burn> rrsagent, draft minutes
15:54:38 <RRSAgent> I have made the request to generate http://www.w3.org/2017/10/03-vcwg-minutes.html burn
15:55:13 <burn> present- bigbluehat
15:55:19 <burn> present- dezell
15:55:23 <burn> present+ David_Ezell
15:55:35 <burn> present- Liam
15:55:40 <burn> present+ Liam_Quin
15:56:42 <burn> s/?? could fit/anonymous credentials could fit/
15:56:47 <burn> rrsagent, draft minutes
15:56:47 <RRSAgent> I have made the request to generate http://www.w3.org/2017/10/03-vcwg-minutes.html burn
18:01:43 <stonematt> stonematt has joined #vcwg
18:02:04 <stonematt> dlongley: are you still online?
18:02:16 <dlongley> stonematt: yes, what's up?
18:03:11 <stonematt> re: https://github.com/w3c/vc-data-model/issues/42#issuecomment-304921994 and our discussion this morning.  What milestone do you think this drops into?
18:03:29 <stonematt> I'm thinking we need to articulate what the n+2 milestone is...
18:03:41 <Zakim> Zakim has left #vcwg
18:04:25 <dlongley> i definitely think it's sometime after milestone 2
18:04:44 <stonematt> me too. (although I'm pretty excited about the discussion)...
18:04:49 <dlongley> i don't know that we need to tag those issues as any particular milestone if we don't have such a milestone yet ...
18:05:05 <dlongley> yeah, i figured it's a subject people would be interested in
18:05:41 <dlongley> there's nothing stopping people from engaging it early if there is energy to do so -- provided that their energy wasn't likely to be spent on the current milestones anyway
18:05:48 <stonematt> i'm thinking like a Product Owner now, maintiaing the backlog.  we have a MS2 for expiration/revocation, etc.  we need to start coalescing around what's in  MS3
18:06:04 <stonematt> +1 to your last comment.
18:06:43 <dlongley> yeah, milestones 1 and 2 are basic functionality ... milestones 3-N could focus on additional features to promote usage and adoption
18:07:13 <dlongley> where a consistent UI and addressing how one specifies how to display certain types of credentials would fit in
18:07:23 <dlongley> "interop" on display
18:07:45 <dlongley> we've also got to fit in terms of use somewhere
18:08:17 <stonematt> have you looked at the ORDL stuff?
18:08:50 <dlongley> yeah, in the past, i think we want to reuse -- we just need to specify integration points or examples for how it will work
18:09:10 <dlongley> we've been targeting reusing ODRL for rights expression for years now
18:09:21 <dlongley> just need to get into the details in a future milestone
18:10:25 <dlongley> i'd like future milestones to setup, as much as possible, future work for implementing protocols and so forth
18:11:01 <dlongley> so saying "here's how you display this information, here's how you express terms of use, ... " and so on, makes it much more obvious how you can put those things together into a protocol for future work in the group -- which would help complete VC use in an ecosystem.
18:11:25 <dlongley> all the while the CG is working on those sorts of protocols and so on.
18:13:40 <stonematt> do you tink the CG would follow the MS?  At some point, they probably come from the CG
18:13:54 <stonematt> our scope is so oddly narrow
18:14:46 <stonematt> also, after today's meeting, it seems like vocabulary should be a more visible discussion topic
18:14:59 <dlongley> i think the CG is going to forge ahead building out what is necessary to actualize the whole ecosystem, but we want the WG to be providing any data model related pieces as we go, informing the more cutting edge protocol work
18:15:13 <stonematt> +1
18:15:18 <dlongley> stonematt: well, i think the vocabulary falls out of what's in the data model spec ...
18:15:30 <dlongley> the data model spec more or less has, in prose, what the vocabulary is, and in examples, a specific syntax for it
18:15:43 <dlongley> the RDFS vocabulary that Reto is looking for, for example, formalizes that into RDF semantics
18:16:27 <dlongley> so we need it, but it's largely informed by the work we're doing in the data model spec now ... we just haven't written down specifics that Reto could pull in and do some validation or reasoning on with various tools
18:17:18 <dlongley> so, for example, we have the term "issuer" in our data model spec, which maps, via a JSON-LD context to: "https://w3id.org/credentials#issuer" ... but we don't have an RDF schema living at the end of that link
18:17:26 <stonematt> (you have to excuse my naivete) -- is the vocabulary a distinct separate deliverable, distinct and embedded in the DM doc, or infered in DM and referenced somewhat loosely in a repo like w3id?
18:17:28 <dlongley> that you can fetch that will tell you (or a machine) about that property
18:17:35 <dlongley> let me give you an example
18:17:53 <dlongley> https://w3id.org/security
18:18:00 <dlongley> that's an RDF vocabulary
18:18:35 <dlongley> and you can request that in JSON-LD format (instead of HTML) to get a machine readable version of it
18:18:50 <dlongley> so we use "signatureValue": https://web-payments.org/vocabs/security#signatureValue
18:18:57 <dlongley> in our credentials data model today
18:19:11 <dlongley> and for expiration: https://web-payments.org/vocabs/security#expires
18:19:29 <stonematt> i see.  (is that useful to peruse w/ human eyes? :) )  if so, how would I do that in a browser?
18:19:31 <dlongley> and these terms aren't specific to credentials, but are in a more generic "security" vocabulary that's out there ...
18:19:43 <dlongley> in the same way there's a "schema.org" vocab that's out there that we may also pull from
18:20:09 <dlongley> what we need to do is create a "credentials" vocab document just like the security one and include all of the new terms we're inventing in our data model
18:20:18 <dlongley> it's useful for human eyes ... if you're a developer, mostly.
18:20:24 <stonematt> got it.
18:20:45 <dlongley> if you're implementing this stuff, you want/need to know the definitions/constraints on each term
18:20:55 <stonematt> is that a milestone 1 deliverable?
18:21:15 <stonematt> MS1 is the core issue/verify process.
18:21:19 <dlongley> it's debatable, you can have the test suite pass without the formal schema written
18:21:39 <dlongley> ideally we'd get it done sooner than later but i think we've been punting because of some moving targets and discussions
18:21:47 <dlongley> once you put something out there, people start using it.
18:21:57 <dlongley> which is exactly, i think, what Reto wanted to do! :)
18:22:05 <dlongley> so then it's hard to change and you have to support backwards compat.
18:22:18 <dlongley> so we need to be really careful and sure about publishing that sort of thing.
18:22:28 <dlongley> because they'll hook it up to implementations
18:22:42 <dlongley> now, we want to start hooking this stuff up to implementations, certainly, so it is coming down soon that we want this piece.
18:22:53 <dlongley> and hooking up the implementations to the test suite and getting things to pass there is the first step
18:23:07 <stonematt> GOT IT!    can we make a draft in our repo to nail down what we know and expose more explicitly what we don't?
18:23:09 <dlongley> once it's clear that we have interop via the test suite, i think it's more obvious that we could produce such a doc
18:23:35 <stonematt> is there a "not done yet" status that let's us be more fluid in the initial days/weeks?
18:23:40 <dlongley> making a draft exposes it to the world :) ... so we can do that, but there's no magic that tells people not to use things
18:23:45 <dlongley> nope, no such thing :)
18:23:51 <stonematt> hmmm
18:24:00 <dlongley> people will hook up their implementations and then the argument will be come "but we have to do what implementations are doing"
18:24:07 <dlongley> rather than what we think is necessarily the best thing to do
18:24:16 <dlongley> "now we must reflect the reality of implementations"
18:24:26 <dlongley> so we just have to tread carefully there
18:24:41 <stonematt> that's seems a bit limiting for early innovation...
18:24:49 <dlongley> i don't think it's a big issue -- i think once we have interop with the basics in the test suite we should put those terms used and tested into a vocab doc.
18:24:57 <dlongley> yeah, it's always a battle.
18:24:58 <stonematt> almost need an "alpha.schema.org"
18:25:19 <stonematt> and everyone knows "thar be dragons"
18:25:39 <dlongley> yeah, never been a real thing on the Web... it's too decentralized ...
18:25:47 <dlongley> someone will run off and implement something and put it into production
18:25:52 <stonematt> ha.
18:25:57 <dlongley> it just happened, for example, with a draft RSA signature spec
18:26:26 <dlongley> now the largest open standard social media network is using an RSA signature suite that is misimplemented because the spec text wasn't clear :)
18:26:51 <stonematt> ok.  you don't seem too worried about it so I'll compartmentalize my angst :)
18:26:58 <dlongley> people see these things and want to use them right away ... which is great, because there's a real need, but also difficult :)
18:27:02 <stonematt> we'll be careful about when to expose it.
18:27:07 <dlongley> yeah, i've just learned to accept the reality of it :)
18:27:25 <dlongley> i think: 1. test suite + interop implementations
18:27:32 <dlongley> 2. vocab doc for those terms that are in the test suite so far.
18:27:47 <dlongley> that gives us some semblance of control over it
18:27:52 <dlongley> before it leaks out into the wild and runs amok.
18:28:26 <stonematt> then expirations/revocation and update vocab w/ additions?
18:28:31 <dlongley> yup
18:28:32 <dlongley> exactly
18:28:43 <stonematt> k
18:28:59 <dlongley> so get the test suite pinned down and the test implementations working against each other ... that helps things settle into something workable.
18:29:08 <dlongley> and hopefully covers peoples' use cases.
18:29:19 <dlongley> then you formalize in the doc that people will start having their tools pull down and use.
18:29:43 <stonematt> that makes sense. almost need a MS1.5 for vocabulary.  or a vocabulary track in github somehow, so we keep it on the radar.
18:29:43 <dlongley> it's easier to change text that just humans are reading ... vs. machines are using to do things.
18:29:55 <stonematt> for sure.
18:30:06 <dlongley> yeah, not a bad idea to have a quick rep on getting the vocab out there after MS1
18:31:59 <stonematt> that sounds like an action item, is it?
18:32:44 <dlongley> yeah, i guess the action item is just to create a MS1.5, and that includes publishing a draft vocab doc that people can do PRs against
18:33:28 <stonematt> i think I can make the MS.  i'm assuming the vocab doc is a new file in the DM repo.
18:33:58 <dlongley> yes, that's my understanding, and i'll need to check with manu in case he had a different plan
18:34:10 <dlongley> he knows the buttons to push for that to make the spec tools do the right thing.
18:35:21 <dlongley> it's probably as simple as copying an empty doc over with the same ReSpec settings ... but i will check with him on that.
18:35:50 <stonemat_> stonemat_ has joined #vcwg
18:36:00 <stonemat_> ok.  that's a bit of an implmentation issue any way.  i'll make the MS.
18:36:07 <dlongley> ok, thanks
18:37:13 <stonemat_> thanks for helping me understand.
18:39:32 <dlongley> sure :)
18:41:02 <stonemat_> https://github.com/w3c/vc-data-model/milestone/5