- From: <kim@learningmachine.com>
- Date: Fri, 20 Jul 2018 17:17:05 -0700
- To: Credentials CG <public-credentials@w3.org>
Thanks to Kulpreet Singh for scribing this week! The minutes for this week's Credentials CG telecon are now available: https://w3c-ccg.github.io/meetings/2018-07-17/ Full text of the discussion follows for W3C archival purposes. Audio from the meeting is available as well (link provided below). ---------------------------------------------------------------- Credentials CG Telecon Minutes for 2018-07-17 Agenda: https://lists.w3.org/Archives/Public/public-credentials/2018Jul/0038.html Topics: 1. Agenda Review and Introductions 2. Roadmap 3. DID WG Charter Organizer: Kim Hamilton Duffy and Joe Andrieu and Christopher Allen Scribe: Kulpreet Singh Present: Joe Andrieu, Ganesh Annan, Heather Vescent, Mike Lodder, Samantha Mathews Chase, Dave Longley, Manu Sporny, Dan Burnett, Chris Webber, Christopher Allen, Kulpreet Singh, Kim Hamilton Duffy, Joe Kaplan, Kayode Ezike, Moses Ma, Ryan Grant, Adrian Hope-Bailie, David Challener, Markus Sabadello, Adrian Gropper, Chris Boscolo, Drummond Reed, Dmitri Zagidulin, Jarlath O'Carroll, Yancy Ribbens, Jeff Orgel, Anthony Ronning, Tzviya Siegman Audio: https://w3c-ccg.github.io/meetings/2018-07-17/audio.ogg Kulpreet Singh is scribing. Topic: Agenda Review and Introductions Joe Kaplan: We aren't gonna get closure on the DID charter, but we work towards it. Joe Andrieu: Calling for Introductions and re-introductions. Richard can you introduce yourself? Kayode Ezike: My name is Kayode Ezike. Working in the AI group at MIT https://w3c-ccg.github.io/announcements/ Joe Andrieu: Announcements: Midst of summer BTCR hackathon. Kayode Ezike: Name is Kayode Ezike (not Carl Davis) in the Decentralized Information Group at MIT CSAIL Christopher Allen: Not too late to participate! Christopher Allen: After this call. Samantha Mathews Chase: Have we tried ryerson for toronto? Joe Andrieu: RWOT - 24-26 Toronto. Challenges getting a venue there. Potentially a hackathon for DIDs of various methods. Moses Ma: Reiterate call for support for DID virtual summit Christopher Allen: URL? Moses Ma: Free event, GSMI co-producing, spending 1000s of dollars. Plan to be a component of bigger conference. Moses Ma: This is the URL to share: https://businessofblockchain.com/web/virtual-summits/blockchain-id Moses Ma: Doing something on Tuesday to bump up traction rates Moses Ma: Free event. Please email me with questions Ryan Grant: Might be me Dan Burnett: Reintroductions.. got some news.. background, one of the co-chairs on the VC Working Group. Got involved with personal interest as it leads to world that the individuals of the world want Dan Burnett: Joined the standards team with consensys Dan Burnett: Will be more involved with ccg Christopher Allen: https://decentralizedweb.net/ Christopher Allen: End of July, first few days of Aug is the decentralised web summit of internet archive Christopher Allen: Kim and I agreed to a booth thing, showing off BTCR and VCs Christopher Allen: Love to have anyone join us there Joe Andrieu: Action Items, anyone has progress to report? ... Looking at the link Moses Ma: To expand GSMI is investing several thousand dollars in bandwidth and production services, and really believes in decentralized ID. They're already hosting DID panels at their Blockchain East and Blockchain Healthcare Summit conferences, which I'm moderating. This is a good thing for DIDs and I urge you to share the link to your feeds and participate in the event. Thanks! Again, the URL to share is: https://businessofblockchain.com/web/virtual-summits/bloc[CUT] Joe Andrieu: How are the specs coming Manu Sporny: Implementations are moving Christopher Allen: Hoping to get more people involved in the hackathon from jsonld community, in particular around 1.1 ... action item is we need to draw more people, not just in jsonld side, but also with credentials cipher suite ... had a few snafus, this time the only code base that support @graph (things in 1.1) that we need for proper VC appears to be in the js version included in the playground ... not doing a great job in separating the signature suite ... need more work to identify people who are working on this and rope them in Manu Sporny: Happy to support as digital bazaar ... you can make progress without 1.1. Sigs will be wrong, but more important bits are can you express VC as you want and if the context is capturing things properly ... might not take tremendous amount of work to bring software up to speed ... we have plenty of people that can update the libs ... need to understand what is working and what isn't Christopher Allen: I tried some tests https://github.com/w3c-ccg/did-hackathon-2018/blob/master/BTCR-DID-Tests.md ChirstiopherA: Just shared a URl that is all that is relevant Joe Andrieu: Amira was published, and we hoped this would be an outcome of this group https://github.com/WebOfTrustInfo/rebooting-the-web-of-trust-fall2017/blob/master/final-documents/amira.pdf ... reviewing the paper in detail next week. Link shared. Quick read. Christopher Allen: .Md for markdown version ... interpretation of Christopher's original WoT user story ... diving into amira in detail next week ... question is what do we do with it ... what is the right amount of engagement for a work item like that ... manu, do you have a comment on that Manu Sporny: Haven't thought much about it Samantha Mathews Chase: Thinking about california consumer act, going through use cases ... lot of stuff on the law/bill about a VC. Do they know about VC or is it just legalise overlapping ... how can we connect that ground more together, as it seems importnat Manu Sporny: As far as next steps, most detailed use case i have seen at w3c ... uncharted territory. min we could do Samantha Mathews Chase: https://docs.google.com/document/d/1DvT5TlKHHTtnm43scpfAa1GxML3onx36mhhH2D1S89Y/edit ... we could put it in respec format ... put in as acommunity group ... at lease the CG is supportive of it ... up to the working group to decide if they want to publish it ... furthest it could get to is w3c NOTE ... publish it for final community group specification, we push it for as a w3c note and they will do it ... however, human heartfelt story, but some subset of engineers might not respond well to it ... that said, no one should be able to block it from being published ... depends on what we want to do with it Adrian Hope-Bailie: Can somone connect me Sam Joe Andrieu: Will make the intro Heather Vescent: There is some noise on the call. Joe Andrieu: No other work items... doing great in terms of time https://w3c-ccg.github.io/roadmap/diagram.html Topic: Roadmap ... manu: can you introduct the roadmap Manu Sporny: Has come up before.. we try to plan the future of the group ... some new to the group might be puzzled by community group and working group ... CG is a long running group ... WG is more focussed ... CG is formed around security, education credentials, and self sov. identity ... roadmap linked above is a view of what we _could_ do solely based on tech specs we could produce ... specifically doesn't point to charters, use cases, stories and those kinds of things ... i wanted to just put down a technical roadmap ... what techs are we working on ... and what we need to put in place . .Hopefully gives a fair idea to people ... on the left of the diagram, is time ... starts in the past JWS and JWK, signature mechaniss And then progresses to 2018... 2022 ... looks out to 2022 ... each thing beside year is when we are expecting XXX to be mature enough so we can start a specification ... in blue 2018, jsonld has active working groups, they are failrly mature ... then there are others that we might need to slap together, we don't know if they will happen or won't... they are in red ... it is politicial reasons we have things scheduled, or there are implementation delays ... each spec has a diff. reason for being where it is ... any questions on the diagram David Challener: +Q Christopher Allen: Am seeing sig suites on the cryptographic side, but we are encountering work on data minimisation, canonicalisation and sig suite issues ... example, the redaction sig suite, it is more about the canonicalisation ... basically every node in the graph is hashed and the sum of the graph is hashed, which is completely independent of which crypto suite is used ... where does all this fit in? ... requirements like that people don't over reveal or correlate Manu Sporny: If we don't have specs where it looks like people aren't working on it, then it doesn't get in there ... for example, the redaction sutie, no one is implementing it in production ... it is an important spec, but is it a nice to have? ... in general, if not being worked on, then it is not in the spec Heather Vescent: Looks like a roadmap from internal ccg how we are building tech ... i have xxx about how we can drive adoption, tech lifecycle adoption ... this will help to see how tech is rolled out to be adopted ... to change the market ... i have understanding about what happens when it goes out into market and is adopted ... if you want to have evidence or reasons for the need for such specs ... or standards ... will be interesting to map out the foresight about tech adoption Manu Sporny: Aboslutely, this roadmap is missing the user story or use cases ... which will be handy to decision makers Samantha Mathews Chase: I'd love to help with that Heather Heather Vescent: I do think there is gap here about people appreciating the inputs of a trained futurist Heather Vescent: Can result in a lot of delicious stuff and will be interesting to do it here ... will be great to apply this to a tech roadmap ... can we do it with Project or something Joe Andrieu: Will be a great work item ... samantha and manu chimed in to help you out..could be a RWOT paper David Challener: Haivng a little trouble understanding the chart ... presume red moves to green ... why aren't they all green? Manu Sporny: Was hard to pack info into this Dave Longley: Maybe the timeline should say "goes green" ... colours reflect maturity (as of today) ... should be updated every 3 to 6 months David Challener: Doesn't mean when it is expected to be done? Dave Longley: So blue Manu Sporny: Expected to be good enough to be put into an internation standard.. then it takes 2 years to be ready Manu Sporny: Example of XXX works starts in 2020 ... it will be difficult to start work in 2020 from a political perspective Dave Longley: So key for timeline should say "when a WG starts" ... if DIDs don't catch on, then we can't start on David Challener: Problem with being correct for a roadmap is that then you will be killed as a witch! Manu Sporny: Laughs Heather Vescent: LOL, no one ever gets the future "correct." Adrian Hope-Bailie: My work is around human agency ... manu mentions in the diagram.. because i depict the self. sov. id as human agency ... UMA, and others, non in standards track in ieee, but there are groups (3/4 standards) or 7000 series working on AI and XXX ... my project hiev1 is implementation of such stacks, but i want to know how we deal with authentication, delegation and xxx in this diagram Christopher Allen: Viewing this as a lens of who is coding things, is not enough ... our charter includes importance of privacy and deployment of stuff ... so there are all sorts of holes Dave Longley: Sounds like this roadmap is doing a good job of identifying gaps! :) ... authentication, for example, are we going to use UMAs auth receipts ... just identifying who is working on what will lead us to a hole Manu Sporny: This is just one lens There was another lens mentioned, and adrian pointed to another lens ... this diagram was intended to be a technical roadmap ... not how tech is difused or things tht you are talk about christopher ... this is a proposal for a tech roadmap and i hoped for just this sort of conversation ... ocap-lp hp ocap auth, dkms stuff.. i tried to link this stuff ... uma is missing ... we need to figure what arrow we are drawing from where ... great conversation for this group to have Adrian Hope-Bailie: We might want to include groups like DID foundation, who are working in parallel with things related to this roadmap ... would it make sense to engage them and mention them in this diagram Manu Sporny: I put in pointers in this diagram for DIF to contribute to this Joe Andrieu: We don't have a perspective on what needs to be built in terms of storage and consent https://w3c-ccg.github.io/did-wg-charter/ Topic: DID WG Charter ... DID charter.. this is the first full draft shared with the group Manu Sporny: Long standing charter item to create an official standard ... members end up voting on the charter . And if enough vote then you get to go ahead to work on the thing ... requirements: focussed, implementations in the real world and should be far along in knowing what is going to be standardised ... we have been following roughly the same specifications, that shape that w3c memberhship is interested in ... we have focussed only on the DID specification ... two things the wg should work on ... DID models and DID use cases ... we are working on both in this CG in the past months ... take that material and make standard out of it ... liam, from w3c, suggested we remove some groups we thought we wanted to co-operate with ... the really imporant stuff, mission of group, scope, not in scope and what will be delivered ... get DID spec standardised, out of scope, anything related to Auth, browser apis ... not standardising any DID methods ... in general, not going to solve identity on the web . This WG is just there to standarise DID specs https://github.com/w3c-ccg/did-wg-charter Joe Andrieu: Here is the url for that repo ... issues there are welcome ... if you see a typo, or an area you can improve Markus Sabadello: Asked the same question last time... did resolution, will it be in scope Manu Sporny: DID resolution is not in scope, doesn't mention it as a work item ... if there were two interoperable implementations and if they talked to each toehr ... you can always, put in optional work items ... we could say did resolutions could be an optional work items ... down side is sometimes companies object saying you are trying to do too much ... i think we should really incubate did resoltuion a bit more in the CG https://github.com/WebOfTrustInfo/rebooting-the-web-of-trust-fall2017/blob/master/final-documents/amira.pdf Markus Sabadello: I think did resolution is close to the core model, but i understand the options Ryan Grant: Will defer my question, we need to look at what DID XX is?
Received on Saturday, 21 July 2018 00:17:32 UTC