The W3C Credentials Community Group

Verifiable Claims and Digital Verification

Go Back

Credentials CG Telecon

Minutes for 2019-06-25

Amy Guy: Voip 429 is rhiaro
Vaughan Emery: Present +
Jeff Orgel: JeffO bystanding a tech call...sorry can not
Dave Longley is scribing.

Topic: Introductions and Re-introductions

Christopher Allen: We want to welcome anyone new to the calls today, anyone new?
Ajones_DB: I've been on a couple of calls, I'm the person writing the DID test suite. From Digital Bazaar.
Christopher Allen: Thank you -- we appreciate that work.
Christopher Allen: We also do reintroductions -- let's new people hear about existing members and reconnect with the community.
Drummond Reed: Note that I have to attend another call today, but I am monitoring the chat here.
Bill Barnhill: Btw, I have scribed once but am not on the scribe rotation list. Please feel free to add me, as Bill Barnhill (bbarnhill).
Joe Andrieu: I'm Joe Andrieu, I know a consulting company Legendary Requirements for understanding the human requirements for identity. I'm co-chair of this group.
Joe Andrieu: Thank you, BIll
Kim Hamilton Duffy: Updated just now; will take a sec for github pages to reflect them
Christopher Allen: We're trying to keep more of our materials and details online -- so if you want to see dates, see announcements or want to announce, do a PR to this announcements repo.
Christopher Allen: Or do an issue and we'll add it if you're not comfortable with PRs.

Topic: Accouncements

Kim Hamilton Duffy: The IEEE standards magazine calls for paper on identity, decentralized identity layer. The deadline for manuscript submission is July 6th, but it can be in draft form. The publication isn't until Dec 2019.
Kim Hamilton Duffy: There are a lot editors that can help you along.
Kim Hamilton Duffy: I recommend, if you're interested, an old rebooting paper from the past is a good starting point. I encourage you to submit and to reach out if you have any questions.
Christopher Allen: This group is also doing dedicated DID calls the last few weeks. This Thursday call has been used to work on the DID spec and DID resolution ideas.
Christopher Allen: I believe we're going to be moving towards the DID topic.
Kim Hamilton Duffy: We have a few DID WG charter issues. We had a breakout group that is trying to triage which are editorial. We did that and pushed through a bunch of PRs, half of the issues were addressed but we still have some more and some new ones.
Kim Hamilton Duffy: We want to wrap up DID WG charter issues soon, some will be rejected as won't fix.
Kim Hamilton Duffy: The ones that remain, we're pretty much at the point that we're waiting/pending specific wording around decentralization and how to craft language around dependencies and we've been nagging the past month to close out the remaining ones.
Kim Hamilton Duffy: We need to be all hands on deck now -- we should repurpose the Thursday DID meetings to address those issues.
Kim Hamilton Duffy: We had mentioned trying to set a deadline to close out the DID WG charter issues.
Kim Hamilton Duffy: It might take a pass from the Thursday group to get some feasibility about that. If you're bored with talking about matrix parameters that would be fun for you, please join.
Christopher Allen: There are some people that would like the concept of decentralized to be removed from this working group's charter. Being that that is fundamental to the work we need to resist that and have good arguments for that there. If you care about that, we have to have this WG charter approved in order to begin working on the spec as a standards track doc.
Kim Hamilton Duffy: Thanks Joe!
Joe Andrieu: I was going to ask about the word limit/length limit on the IEEE papers, and I found that, it's 4500 words or less. There's a policy on self-plagiarism, so they want original work you need to say something additional to build on what you've said before, so don't just resubmit rebooting papers.
Kaliya Young: Yes, it's important to pursue the W3C path but it's not the only standards body in the world and we can go to OASIS or elsewhere if they are being difficult. Also, who is going to the dweb summit/camp in July?
Christopher Allen: Yes, I agree that if W3C is recalcitrant on it then we'd explore other options.
Drummond Reed: Since Markus couldn't be here either -- I saw in IRC that we'd use the Thursday calls to concentrate on the DID WG charter until that's done. I would like to see ...
Kim Hamilton Duffy: We've gotten lots of support from W3C staff btw, so I'm thinking it won't come to that. We just need to get our duck in a row re closing out issues
Drummond Reed: We can return to the DID and DID resolution specs once the charter issues are done.
ChristopherA mentions announcements
Adrian Gropper: I would like to help with the "decentralized" charter issue. How?
Christopher Allen: There's a wonderful beach place off of SF and you bring your tents and campers and there's some space available for small cabins, etc. You hack and sing around the fire and dance and hack again.
Christopher Allen: If you're interested in that it's very family friendly as well with lots of opportunities there.
Christopher Allen: In Vienna on Sept 1st, there's a pre-rebooting meetup. They meet on the 1st and then take the train to Prague on the 2nd. Rebooting Web of Trust 9 is in Prague, Sept 3rd-6th.
Christopher Allen: There is an eventbrite for this.
Christopher Allen: When is the deadline for the discounted tickets and such?
Joe Andrieu: We settled the venue a little earlier this time so we have some time for folks to write things up. Early bird pricing ends July 19th, so just under a month to get a paper in. August 16th is next deadline after that.
Christopher Allen: There are discounts if you write advanced topic papers, 80%+ used those discounts last time.
Christopher Allen: Get your submissions in early.
Joe Andrieu: The papers are due on the 16th of August.
Joe Andrieu: That gives everyone 2 weeks to read your paper before the event.
Christopher Allen: We will not be giving discounts after that.
Christopher Allen: So get them in by August 16th.
Christopher Allen: Also in Sept in Europe, MyData 2019. I don't have details on their DID workshop but they were active last year with a number of people that went there.
Christopher Allen: Some of the usual suspects plus some new people coming. If you're still in Europe in Sept at the end of the month, MyData might be a good place.
Christopher Allen: For those of your involved in W3C through us or VC, etc. the yearly international AC meeting is in Fukouka, Japan.
Christopher Allen: We've requested time to discuss DID during that.
Christopher Allen: We hope to have the charter approved by then.
Christopher Allen: For DID WG.
Joe Andrieu: Correct, Chris. CCG does not have a time slot at this point.
Christopher Allen: Beyond that there will be VCWG meetings, etc. other meetings. We haven't requested a meeting for the CCG outside of discussions for the DID WG. If you are involved with W3C deeply and heading out there we'd love your support with those things that do get scheduled.
Christopher Allen: Anything I missed?
Adrian Gropper: I'd like to help with the charter decentralized issue, how do I get involved?
Christopher Allen: It would be on the Thursday calls. They are still working on the agenda, if not this Thursday, a week from then they'll be diving deep into it.
Christopher Allen: We do a lot of our planning via our community repos.

Topic: Action Items

Christopher Allen: We wanted to talk about 75, 76, 77.
Christopher Allen: Any topics/concerns issues for how the CCG functions, this is where you add them and then each week we pick out a few to discuss.
Christopher Allen: We already talked about the DID WG charter issues, we'll be devoting Thursdays to that so I don't know if that needs a lot more review but it's tagged help wanted. There may be people that are less technical than others that can constructively participate in the discussion.
Christopher Allen: If you're feeling a little intimidated on the technical side for things like matrix params for DID URLs but want to help move things forward, that
Christopher Allen: There has been a discussion about tools that go from Google docs to Respec. Kim has found another tool that uses a markdown format.
Christopher Allen: It makes it easier than some of the other choices.
Bill Barnhill: +1 For Kim’s Bikeshed approach
Kim Hamilton Duffy: The last three comments, you'll see my proposal to use markdown.
Kim Hamilton Duffy: For many of the work items that we do in this group, this would be a fine solution. You can see an example because we used it for the btcr DID spec.
Christopher Allen: Rendered respec:
Kim Hamilton Duffy: There are a few tags that are used for generating the respec. You can look at the history of that file that the changes are very straightforward. If you're comfortable writing markdown you should be comfortable with editing this.
Kim Hamilton Duffy: You edit the bikeshed (.bs) file and you install bikeshed and it will convert the file and make the rendered result.
Kim Hamilton Duffy: In that comment I show that after you run that it looks like normal ReSpec.
Kim Hamilton Duffy: So you can edit it and diff it looking at plain old markdown which should be accessible to more people.
Kim Hamilton Duffy: So the concerns I raised for it, to use bikeshed the way I did it still requires installing and running a Python script.
Kim Hamilton Duffy: We could stand up a site or service that lets people paste in the bikeshed and returns ReSpec.
Kim Hamilton Duffy: As things progress I don't know how this will work, people might need to edit the HTML directly.
Kim Hamilton Duffy: I would imagine in some specifications that are more mature people might want to do that, but this is a good solution.
Dan Burnett: I don't know bikeshed, but people I respect who know both bikeshed and ReSpec seem to prefer ReSpec. And yes, ReSpec can take markdown as well (up to a point).
Christopher Allen: @Bigbluehat suggests we also look at at respec's support of markdown
Kim Hamilton Duffy: That didn't really solve the editing problem. I think this is a good approach, I think we should keep this issue open and then maybe have a site for people that don't want to install/use Python and then recommend that approach.
Mark down is fine
Google docs can export as markdown
Christopher Allen: We might want to compare/contrast bikeshed vs. ReSpec support. I think the key thing is that we can try to continue to make it easier for people that are less technical to contribute and comment on various documents that are in progress at the CCG.
Christopher Allen: We're increasing uncomfortable with having Google and the centrality and the privacy issues there being the greatest common denominator there, fine with unreleased docs there, but as we move through the process you really need to have tighter version control and authorship control.
Christopher Allen: Google docs just really doesn't give that to us.
Dmitri Zagidulin: I wanted to add to that -- I think Christopher you touched on some of it. We have two-three separate issues here.
Dmitri Zagidulin: We know that we want the end result to be in ReSpec, no problem. The discussion of ReSpec vs. bikeshed is an argument about markdown vs. html.
Dmitri Zagidulin: I'm for markdown. Editing HTML directly is silly.
Dmitri Zagidulin: A separate issue is decentralization. We can live with the decentralization of microsoft (things live on github) but we prefer not to also add the centralization of Google docs. Separately from that, but tied into Google Docs.
Amy Guy: Backend of github is decentralised
Dmitri Zagidulin: Github vs. some other collaborative tool. Some people have expressed concern with gitlab or microsoft's github -- it's too hard for people to use. This when we expressed, for unreleased drafts, Google docs.
Dmitri Zagidulin: It has comments in the sidebar and suggestion features that are easier than github, but that led to questions about centralization.
Kim Hamilton Duffy: In draft phase, people can use whatever they want, gdocs, etc
Christopher Allen:
Dmitri Zagidulin: I don't think there are any alternatives that are as good as Google docs.
Christopher Allen: Some people from Rebooting use hackmd.
Dmitri Zagidulin: That doesn't have annotations, that's a collaborative realtime editor I highly recommend but doesn't let you have meta discussions.
Kaliya Young: My company Wirelinw is working on decentralized applications - they are months away from
Adrian Gropper: Dropbox Paper?
Kaliya Young: Having a public thing...that we could maybe be an early adopter odd
Christopher Allen: My ideal would be something that was opensource, wasn't completely proprietary that allowed people to edit markdown but let the extra collaborative comments and things of that nature.
Christopher Allen: Yes, Microsoft owns github but git itself is an open protocol and open source itself, so we move to a number of different things.
Kaliya Young: I’m trying to make sure they use verifiable credentials and DIDs and wallets for their systems/network
Kim Hamilton Duffy: In rough draft phase, the editor can use anything they want including Google docs. When it goes past rough draft then it's assumed to be output in ReSpec form.
Kim Hamilton Duffy: The bulk of the revisions can be done in Google Docs anyway -- you can do all that, hackmd, whatever, when they think it is ready they can switch.
Kim Hamilton Duffy: If you're doing the Google docs to ReSpec directly every time it does get into a diff problem.
Kim Hamilton Duffy: I think we're just making it easier on the backend of that.
Dmitri Zagidulin: +1 To editor's choice (including GDocs) while in draft, then Github / ReSpec.
Kim Hamilton Duffy: We're not precluding Google Doc editing for things like community notes and things like that. But for things like the DID spec it's clear that would get haywire after some amount of time if you're trying to do tight revision on it.
Kim Hamilton Duffy: Hopefully that clarify things, we aren't excluding things right here.
Christopher Allen: We call that phase unreleased draft, not rough draft, just to be clear on terminology.
Kim Hamilton Duffy: Just want to be precise since we went to all the effort
Christopher Allen: When you're going for a community consent and publishing aspect, which can take as little as two weeks, that's the only part that has to be in some form of ReSpec.
Manu Sporny: Just to follow up on Dmitri and Kim mentioned. Where everyone's heart is on this issue is to get more people contributing that don't necessarily have a super technical background and that's great and should be the driving metric.
Manu Sporny: I think we can get too wrapped around decentralized/proprietary. We should pay attention to it but we should focus on getting it so more people can contribute.
Manu Sporny: I think there are ways for making proprietary software more palatable, Google docs has some of the best inline comment features, if we had a tool that would let us move the docs into another format I feel like we could get the best of both worlds.
Manu Sporny: We make it easy for people to work with Google Docs and contribute and make sure we aren't locked into a proprietary format ... it's a simple matter of time and code.
Drummond Reed: +1 To Manu's point. At this point in our collective evolution, tools like Google docs are much easier for collaboration.
Manu Sporny: I don't think we need to have a decentralized vs. proprietary discussion but rather try to leverage all of these tools.
Christopher Allen: Ok, we're making progress, we're exploring the various options and such.
Christopher Allen: On progress items, the registry process... next steps.
Christopher Allen: We've been talking about our updated process. But we really need to make it official. One of the missing pieces is how do we deal with registries.
Christopher Allen: What I have suggested with the chairs and they agreed, in additional with the comments from Manu's original process. For a new registry to be approved it needs to come from an editor from an already approved spec. It needs to have at least two editors to review the requirements.
Christopher Allen: There's a process requirement in the registry and if that process becomes compromised then we can close the registry.
Christopher Allen: As per other things there is no such thing as a final registry.
Christopher Allen: We need to move forward on that, Manu has done a good job on getting it started the chairs are feeling constrained on time so maybe another editor or maybe someone like Dan who wants us to do a registry.
Christopher Allen: Someone other than Manu who can help us integrate that into our overall work item process so we can share that and more formally approved as how we do business here.
Bill Barnhill: Apologies, but I need to drop for another meeting.
Manu Sporny: Apologies for missing the call last week, was traveling. A couple of people have mentioned requiring two editors for official work items.
Christopher Allen: That's for specs.
Manu Sporny: That's best practice to have two orgs, but sometimes it takes a year or two for you to hash things out. I want to make sure that we're making it such that people won't surface work as they discover it.
Manu Sporny: One of the downsides of making that requirement too early in the process is that we're not enabling specifications to be incubated properly before another organization decides they want to join the work.
Manu Sporny: I've seen that result in things happen behind the scenes for too long and then pop up out of the blue.
Christopher Allen: What we've said is that those are informal work items. We may list them somewhere but the CCG just doesn't have time.
Christopher Allen: We have to have a cut off, I agree that finding these is difficult that the couple that come up that don't have two editors, we ask you to continue keeping us up to date on the list but we can't do process time.
Christopher Allen: We're already 12 minutes to end of the call just getting through process.
Christopher Allen: Doing it for all the informal items whether it's just one company, we can't do it.
Christopher Allen: If you've got some ideas on how to surface these -- we'd love to have someone run that page.
Manu Sporny: I'll send something to the mailing list.
Christopher Allen: What other things would you like us to do on these calls.
Christopher Allen: Maybe see some demos and screenshots of what people are doing.

Topic: Work Item Review, Fin

Christopher Allen: The chairs have gone through the work items that we've already approved and trying to figure out where they are in the process.
Christopher Allen: You'll see that there are updated categories and targets. We did have some questions.
Christopher Allen: The first one is -- the OCAP document? Is that intended to be a spec, a note, a commentary?
Manu Sporny: A spec.
Manu Sporny: Authorization Capabilities more likely, we have to a rename, and there's active implementation going on there and finishing that up before revising the spec.
Christopher Allen: I would appreciate if the title could be updated and the name if you know what that will be and we'll mark it down as on the spec track.
Christopher Allen: DID resolution, there has been a lot of work on that -- is DID resolution going to be moving into the DID WG?
Christopher Allen: Or is the plan for it to be another WG someday?
Christopher Allen: Updates and thoughts on that?
Drummond Reed: DID Resolution is not in the scope of the DID WG charter, so it will probably have to be an evolutionary step.
Christopher Allen: I will reach out to the people running that group to puzzle out the plan for that. I haven't heard that it will be part of the initial DID WG charter so it may be a CCG spec thing.
Kim Hamilton Duffy: Manu, dlongley, bigbluehat: to clarify, we've been talking about process during the call to flush out the open items as a result of our cleanup efforts (culmination of a year of effort). In general, we don't expect to be talking about process during the call except for briefly during the action item part. But yes, any changes you'd like to propose, let's iterate in the mailing list
Christopher Allen: Finally the people working on hashlink stuff -- the target is for IETF.
Christopher Allen: Have they been official submitted as IETF drafts?
Dave Longley: They've been submitted to IETF as drafts.
Manu Sporny: They are all independent drafts, same thing with HTTP signatures.
Joe Andrieu: @Drummond "evolutionary step" is an ambiguous destination. Do you expect it to become a specification somewhere or just be published by the CCG?
Manu Sporny: And it probably should be on the list.
Manu Sporny: It has its own life at IETF, it has something insane like 16 implementations and a test suite but Authorization Capabilities etc. build on top of it so we should list it here.
Christopher Allen: If you can help clarify these.
Christopher Allen: This document of work items generally expresses what we think is the current state of our work items. If there are things where "we're beyond rough draft and we've reached a level where we want it to have a version number and thus be a published draft stage" go ahead and publish a version number and let us know.
Manu Sporny: Awesome work kimhd, JoeAndrieu, and ChristopherA! Thank you for collecting all this stuff together... I know it's largely unacknowledged administrative work, but super important...
Manu Sporny: So, thank you! :)
Christopher Allen: We'll change it from a rough (unpublished) draft to a published draft.
Christopher Allen: Whomever works on registries should look at what the stages are. Having those in Google Doc and stages and version number should be considered so there's some organization.
Christopher Allen: If you have comments on the work items or we haven't categorized them properly please let us know.
Christopher Allen: Anything else about work items?
Christopher Allen: We will be working in the next couple of weeks with people who want to introduce new work items so if this is on your agenda to have your particular thing be a new work item, now is the time and make sure you get the issue in.
Christopher Allen: If you want something to be a formal CCG work item, go to the issues, use the work item template, you can do it without a second item, bring up the discussion on the list for at least the week and we'll schedule it in a meeting to make it an official work item.
Joe Andrieu: So this just came across my feed -- a prohibition on our national patient ID was struck -- that would be a very bad idea, if you care about decentralized ID and you care about this please write a letter to your Senator.
Joe Andrieu: We used to have a prohibition and it has been struck from the law.
Joe Andrieu: It has been passed by the US House and is moving to the Senate.
Joe Andrieu: We tend not to get political here but it speaks to the work being done here.
Joe Andrieu: The arguments for wanting a national patient ID are well intended, it's a complicated issue. I wanted to encourage folks to express their opinion either way directly with their Senators if applicable.
Christopher Allen: We don't have a formal plan for next week's agenda, if you'd like to get on, or talk about the status of your project, etc. we'd like to get at least one of those on there per month, let us know.
Moses Ma: Bye folks
Yeah a lot of people die because hospitals can not share records or identify previous diagnoses in a patient