Credentials Community Group Telecon

Minutes for 2014-11-11

Agenda
http://lists.w3.org/Archives/Public/public-credentials/2014Nov/0002.html
Topics
  1. Review W3C TPAC Meeting
  2. Web Payments IG Meeting
  3. W3C TPAC Credentials Hallway Discussions
  4. Planning for next 2 months
Organizer
Manu Sporny
Scribe
Mark Leuba
Present
Mark Leuba, Manu Sporny, Sunny Lee, Mary Bold, Brian Sletten, Victoriano Giralt, Nate Otto, Eric Korb, Evgeny Vinogradov, Rob Trainer, David I. Lehn
Audio Log
Mark Leuba is scribing.
Manu Sporny: Reviewed agenda, the W3CTPAC meeting. Went really well. Reviewing for those that could not attend.
Manu Sporny: Review the TPAC Meeting, Discuss Priorities, Learned a lot at the meeting. A number of things will move quickly. A lot of groups interested.

Topic: Review W3C TPAC Meeting

Manu Sporny: Any updates or changes to agenda? <none>
Manu Sporny: These are public minutes. Gave parties a week for corrections. OK to share publicly now.
Sunny Lee: Sounds good
Manu Sporny: Everyone in Web Payments Interest Group were able to learn more about the CG. Turnout was great. Normally 12-15 persons, had between 38-48. Fantastic, a lot of interest.
Manu Sporny: The Credentials Meeting had 15 participants, really good turnout. Many more could not be there. Folks from Deutsch Telekom, Verisign, Federal Reserve, Accreditrust, Target, Mozilla (Briefly), W3C standards.
Manu Sporny: Very engaged discussion. Dialog went beyond agenda but it was good. Most resolved around Identity Proofing, how to tie credentials to proofed-persons. Went over high level use cases. Talked about identifier portability, Efforts to avoid vendor lock-in. Spoke about data rights.
Manu Sporny: Did not follow the agenda, because conversation as going well, discussion was well-aligned.
Manu Sporny: Details are in the minutes (link) above.
Mary Bold: The participants were knowledgeable, needed no instruction.
Mary Bold: This allowed fast moving discussion. Manu was very clear and expert in his explanation to the participants.
Brian Sletten: In addition, there was general consensus about the scope, where it would apply. More than I expected, There was a good sense that we are on the right track.
Manu Sporny: Agreed, I was surprised there was not arguing about minutia.
Manu Sporny: Anything else about the CB face to face meeting?
Participants at W3C TPAC had nothing else general to add.

Topic: Web Payments IG Meeting

Manu Sporny: Credentials CG was invited to present for one hour. Went very well. In the Web Payments IG meeting the credentials were not discussed much but the work of the CG will be folded back into the Web Payments group.
Manu Sporny: We spoke about the broad group of interests, ETS, Accreditrust. Health Care and government as well.
Manu Sporny: The Web Payments IG will discuss Credentials in a serious way. How can we make it productive? We need a set of deliverables to make that possible.

Topic: W3C TPAC Credentials Hallway Discussions

Manu Sporny: Other discussions in the hallway that myself and others had with W3C team around credentials. The preferred track has two specifications of interest, how to order JSON LD so it can be signed. Other, how to digitally sign JSON LD messagse so it can be verified by the recipient.
Manu Sporny: The Social Web working group has adopted JSON LD, they want to do digital signatures, they are interested in the work. Web Annotations is also interested in these specifications. The Data Activity Group, is also very interested in Graph Normalization and the other specification as well.
Manu Sporny: This is a good thing, the Web Payments IG will not have to do this, it can be offloaded to other groups.
Brian Sletten: Is that indicative of support? Is there a process where we track the work products of the groups?
Manu Sporny: If we want to take on the specification a group has to adopt the work items.
Manu Sporny: Better for existing group to adopt the specification, to avoid having to start a new group. This is especially true for Graph Normalization work, which is highly specialized. We expect a micro-working group to be created. The secure messaging spec may have the same process.
Manu Sporny: Did that answer your question Brian?
Brian Sletten: Yes, thanks.
Manu Sporny: This is a firehose for the new members, any questions or concerns?
No questions.

Topic: Planning for next 2 months

Manu Sporny: How will the CG group coordinate so outcomes meet our goals?
Manu Sporny: We can do a number of things, a) focus on use cases to hand to Web Payments IG, they had a use cases document that looks like the CG use cases. They have basically copied the CG's use cases. Uses cases are important for our work to transfer into official work.
Manu Sporny: We need to discuss the strategy of creating official work. Which technology stacks to push through? JSON LD, the JOSE stack? Something we have not looked at yet? We need to get the right organizations involved working on those stacks.
Manu Sporny: We need to identify which organizations will charter our work officially.
Manu Sporny: We need 20, have 13 already. Remainder should be not difficult.
Manu Sporny: How to coordinate with other groups fast but not too fast to impact the work. Also need to talk about Test Coordination.
Manu Sporny: Thoughts in addition?
Manu Sporny: Would like to hear group's priorities.
Sunny Lee: Can you elaborate on the level of conversation about the test organization?
Manu Sporny: I can't say at this time. I was asked not to but will shortly.
Sunny Lee: There has been a lot of interest in Open Badges. A conference in DC in October, I was invited to speak. I'm happy these efforts are coming together.
Manu Sporny: They seemed very interested in the general area.
Manu Sporny: Mark have you had dialog with PESC?
Mark Leuba: I haven't had a chance to talk w/ PESC yet. [scribe assist by Manu Sporny]
Victoriano Giralt: Groningen Declaration Network http://www.groningendeclaration.net/
Sunny Lee: There is a parallel effort, Victoriano is the technical mind behind the Groningen Organization above.
Victoriano Giralt: We are seeking to get rid of paper credentials in education at scale.
Victoriano Giralt: RS3G - http://www.rs3g.org/
Victoriano Giralt: Rome Student Systems and Standards Group
Manu Sporny: Digital Student Data Depositories
Victoriano Giralt: We had a technical meeting with another group RS3G Rome Student Systems and Standards Group, we also had conversations in Rome about Open Badge. We are very interested in the education use cases.
Victoriano Giralt: I have come to "Spy" :-) I am very interested. We need a technical means beyond current methods.
Victoriano Giralt: We are looking for more open standards and I am very interested here. Someone from PESC was in Rome talking about Open Badges.
Manu Sporny: Welcome Victoriano, we are happy for you to join. Everyone basically wants the same thing. This CG is focused on making sure the technology works, around the world. Free to use, not behind a pay wall.
Victoriano Giralt: We are looking for ways to create open standards. I will offer to present our activities. Sunny and Manu are invited to our meeting. I want to run a technical track in our upcoming meeting.
Manu Sporny: I think your presentation would be helpful for this group Victoriano.
Victoriano Giralt: We have global reach in our group.
Manu Sporny: Learn more about Groningen Declaration here - http://www.groningendeclaration.org/about-network
Manu Sporny: The link above provides information on the Groningen Declaration.
Manu Sporny: Let us know best schedule.
Victoriano Giralt: Will try for next week.
Manu Sporny: Questions for Victoriano?
Manu Sporny: Credentials CG use cases: http://opencreds.org/specs/source/use-cases/
Brian Sletten: Have you looked at the Use Cases?
Manu Sporny: We have yet to include the GD use cases in the CG set.
Victoriano Giralt: I will try to prepare some use cases.
Manu Sporny: Any other items to focus on?
Nate Otto: +1 For continuing the Normalization Spec and the Digital Signatures.
Eric Korb: A conversation with author of JSON LD. We spoke about validation. There is a need for JSON LD version 2.0, to include validation.
Manu Sporny: Nate has been working on tying JSON schema to JSON LD for Syntactic Validation. Is the structure sound? This is not semantic (meaning) validation.
Nate Otto: Manu explained that right about how Open Badges are using JSON-schema for syntactic validation of JSON-LD documents.
Manu Sporny: A number of organzations are planning to use syntactic validation. A couple of approaches to best practices, JSON LD Framing, SPARQL engines, other semantic web stuff.
Manu Sporny: This is a work item we can look at. I hesitate as this is a lot of work, so many different ways to do it.
Manu Sporny: People are still producing new ways to do it.
Eric Korb: By using JSON Schema as a methodology, will that make adoption more difficult? Be a barrier to adoption?
Manu Sporny: I don't think it will stop us from moving forward. We can move forward without the syntactic validation but it's important. There should be tools to validate. The Badge Alliance has a tool for a couple of versions of the Badges.
Manu Sporny: Would organizations be ok moving forward without Syntactic validation?
Nate Otto: We are making Syntactic Validation an optional component.
Nate Otto: Processors of badges can also use Syntactic validation optionally. If they are comfortable with semantic validation.
Nate Otto: We will use this approach for now but are open to other approaches in the future. I will provide a link to a presentation we are making.
Sunny Lee: Info for the call today: http://etherpad.badgealliance.org/ba-standard-nov11
Manu Sporny: I cannot join the call. Good thoughts and opinions, the work Nate has done is along the same path other companies and groups are doing.
Manu Sporny: This is not an absolute requirements for version 1.0 of the spec. The specs have a very long "half life". A word of caution, everyone has a different way of doing validation. Best practices have not emerged. We can however make a recommendation.
Manu Sporny: In the JSON LD vocabulary we can state a field for validation to refer to a validation best practice as of 2014. As new and better validators emerge this can be updated without breaking the standard.
Nate Otto: Is it ok to have version context files? That can be updated to indicate which validation the version is "obeying".
Manu Sporny: That's how a lot of companies are doing it. You can always add to the context. You should never remove from the context.
Manu Sporny: There is an upgrade path, we want to do the minimal amount of work to get an operating standard. What is the bare minimum to lead to success for version 1.0?
Manu Sporny: Any other thoughts on this subject?
Brian Sletten: I second the initial proposed focus.
Nate Otto: Very few Open Badges in the real world are issued signed with JWT
Manu Sporny: There is our current spec stack. That is the last discussion we need to align the two technology stacks (with the Badge Alliance).
Manu Sporny: Is there any pushback to move off the JWT stack?
Nate Otto: There is very litte to no pushback
Nate Otto: Sunny there are only a few implementations of the JWT, correct?
Sunny Lee: Yes, there are very very few.
Manu Sporny: Because the tooling wasn't available or because it wasn't needed?
Nate Otto: I think the tooling weighs heavily on individual issuer's decisions not to use signed badges.
Sunny Lee: I think both. Security of signed badges was desirable and the tooling implementation barriers were quite high.
Manu Sporny: So there will be little pushback if we changed the digital signing mechanism?
Sunny Lee: There will be a little because not many people implemented it.
Manu Sporny: This is one of the key things I am concerned about. With this aligned, Badge Alliance and W3C CG will be aligned.
Nate Otto: Should we start the conversation in the Badge Alliance now suggesting this change?
Manu Sporny: Should we discuss next week? Too soon?
Nate Otto: I think the people who have been paying attention have seen the direction we're moving towards (JSON-LD, normalization, that world of signatures)
Sunny Lee: Let's put on agenda for next week.
Nate Otto: Some detail: The main difference cited in the use cases document refers to what element of the credential is being endorsed. The Open Badges endorsement supporters have been focused on the "badge class" as recipient of endorsement, where the use cases from the w3c would be focused on the "assertion" if expressed in Open Badges terms. Open Badges endorsement, while intellectually focused on the "badge class" so far, may also be applied to the "assertion" with no other technical changes.
Manu Sporny: Thank you Nate - that really helps clarify that in my mind. I agree w/ you now - not that different... you're just signing different things. They're both valid use cases.
Nate Otto: Yeah, I don't know how the other non-openbadge credentials used by this group will be structured (there probably won't be the same 3-part division as with Open Badges), but there may still be possibilities where endorsement of those credentials function in different ways.
Manu Sporny: We also need to align on "Endorsement" terminology. The only two things left.
Nate Otto: I will look at that but I don't think they are far apart, the W3C is more technical but it boils down to the same basic discussion.
Manu Sporny: Any last words?
Manu Sporny: If everyone can send thoughts on their specific priorities (e.g. 1, 2, ...) to the mailing list that will help move this discussion along. We will then try to execute on those.
Manu Sporny: Thanks all, same time next week.