DID Working Group Telco — Minutes

Date: 2020-04-14

See also the Agenda and the IRC Log


Present: Daniel Burnett, Ivan Herman, Brent Zundel, Dave Longley, Justin Richer, Drummond Reed, Manu Sporny, Eugeniu Rusu, Amy Guy, Daniel Buchner, Orie Steele, Tim Cappalli, Joel Hartshorn, Yancy Ribbens, Kaliya Young, Michael Jones, Chris Winczewski, Dmitri Zagidulin, Ganesh Annan

Regrets: Markus Sabadello


Chair: Daniel Burnett

Scribe(s): Brent Zundel


1. Agenda Review, Introductions, Re-introductions

2. Agenda review and introductions

Daniel Burnett: we’ll get started with a straw poll, plus some other business, then spend most of our time on issues.
… any requested changes?
… anyone here for the first time?
… is there someone who hasn’t introduced themselves in a long time and would like to do so?

Drummond Reed: Drummond Reed from Evernym, trustee of the Sovrin Foundation.

3. Next Topic Call

Drummond Reed: worked on the DID spec from the beginning and really want it to be a standard, I’m one of the editors on the spec

Daniel Burnett: trying to alternate, but this week will also be Thursday at 12pm ET
… the topic is the resolution contract, which is blocking a number of other issues.
… there are a lot of issues to consider in it.
… brent already sent out a reminder

Daniel Burnett: For topic calls, will use #did-topic

Daniel Burnett: for our special topic calls, we will be using a different irc chat room
… the #did-topic room
… this is so we can cleanly separate the notes for different calls which fall on the same day.
… this also lets us have ac onsistent Topic for each call
… in the chat

4. DID Method Registry

Daniel Burnett: https://github.com/w3c-ccg/did-method-registry/issues/82

Daniel Burnett: I’m going to ask Manu to give us a summary of this issue.
… the chairs are after a straw poll to get some indication of what the group is okay with
… we are timeboxing this for 10 minutes and will stop at that time. No discussion today.
… Are you willing to to that Manu?

Manu Sporny: issue 82 is about moving the DID Method registry into the did-core registries document
… we are trying to consolidate where an implementor will have to look for things.
… currently we have all extension methods in that repo except the did method registry
… the simplest way to look at this is to decide if this is something we want to do.
… then we need to decide details.
… is that alright with the chairs?

Daniel Burnett: yes

Orie Steele: +1

Proposed resolution: Integrate the DID Method Registry into the DID Core Registries document. (Manu Sporny)

Manu Sporny: any objections to putting that proposal?

Amy Guy: +1

Drummond Reed: +1

Eugeniu Rusu: +1

Orie Steele: +1

Dave Longley: +1

Ivan Herman: +1

Manu Sporny: +1

Justin Richer: +1

Yancy Ribbens: +1

Daniel Buchner: +1

Daniel Burnett: +0

Brent Zundel: 0

Drummond Reed: we still in the spec are going to have a separate discussion about normative statements about specs, which could adjudicate about several did methods that are in there. Maybe DID methods will need to re-qualify

Michael Jones: +1

Ganesh Annan: +1

Dmitri Zagidulin: +1

Resolution #1: Integrate the DID Method Registry into the DID Core Registries document.

Daniel Burnett: that’s getting into details. Let’s resolve it
… we had fewer people responding than normal. we have double that number on the call
… we just resolved to bring the did method registry into the did-core registry document.
… that’s fair, it holds.
… this will go next to the CCG, but now we know what the working group would like to do
… anything else critical here to spend time on?

Manu Sporny: just outline some things we’ll need to do.

Daniel Burnett: let’s do that, the Orie can re-queue himself.

Manu Sporny: what the chairs and staff need to contemplate is if there might be IPR issues moving the did method registry into the did-core registries document
… usually when we move things from the CCG, we get everyone who contributed to sign an FSA
… this one has more than 50 people who contributed, which is a lot.
… bit there is also an argument that says, there’s no IPR, the contributions are just links
… but if there is a concern, the FSA sign-off could delay things fr quite awhile.
… and if they don’t sign off, they may be removed from the did-method registry.

Drummond Reed: +1 to dropping DID methods who do not sign FSA

Manu Sporny: that is a weird outcome that I am not expecting to happen, but that is the most complex thing that could happen.
… next the Chairs, staff , and editors will decide on the plan and we will begin executing it

Orie Steele: the did method registry takes a lot of PRs from a lot of people. If the document is not conformant, the person must make the necessary statements to make it conformant.
… if this is in did-core registries, we will get a lot of PRs that will need to be addressed.

Manu Sporny: yaay, yes please, we need more volunteers to help w/ DID Method review.

Amy Guy: orie, manu, I can also help with that I think

Michael Jones: to the IPR question: certainly, taking notes from IETF, there is never an IPR question for references to things.

Drummond Reed: +1 to no IPR check needed for registry entries

Manu Sporny: thx rhiaro !

Manu Sporny: (we’d need to go through the DID WG “special job” process, probably)

Daniel Burnett: +1 to likely no IPR check for registry entries

Manu Sporny: +1, my read is the same as Mike’s

Michael Jones: to Orie’s point about timely response, for each registry there is a set of designated experts who look at incoming requests, perhaps Orie and Drummond and others, who can do that without bringing it to the WG.

Daniel Burnett: several of us who have been doing this for awhile have an idea how this is going to go.
… Amy asked a question about whether this will be a living document. That is essentially what registries are, and the details are still being worked out.

Ivan Herman: it is not a big deal to update a note.

5. Core issue status check

Daniel Burnett: https://github.com/w3c/did-core/issues?q=is%3Aissue+is%3Aopen+sort%3Aupdated-asc

Daniel Burnett: https://github.com/w3c/did-core/issues/105

Daniel Burnett: I believe for these first two the answer will be quick

Ivan Herman: I don’t think there’s anything we can do before the doc is ready. What we started in Amsterdam should be properly folded in. For the time being let’s postpone these until the structure of the document is ready.

Daniel Burnett: can you make a comment to that effect in the issues?

Ivan Herman: how could I refuse? :-)

Manu Sporny: pushing back a bit. I’m concerned we’re past the point when the document should be “done” enough for that review. I’m wondering if it would be best to just send it out for review.
… it may have enough in there for people to do a first pass.
… usually for horizontal review they prefer early enough that changes can be made.
… I’m concerned it will be pushed back two months an then it will be too late.
… I don’t expect a lot to come out of the accessibility and internationalization reviews, but the TAG could be tricky.

Ivan Herman: let me be a little more precise. Currently the document starts with a note saying the spec is undergoing a significant structural revision. So as soon as the doc is to the point we can remove that, we will send it off.

Daniel Burnett: I hate late reviews, but I agree with ivan here. We’re still missing section 6.

Justin Richer: q

Daniel Burnett: There’s a lot we don’t need to wait for, but there are pieces we need in order for the doc to be readable.

Manu Sporny: The editors can make a big push to get 3 and 6 done sooner than later, we’ll make that a priority.

Ivan Herman: +1 to manu

Daniel Burnett: +1 to manu

Drummond Reed: section 6 is being revised now after comments from manu. I was waiting on 3 for 6 to be done, but that should be this week.

Daniel Burnett: the chairs are not being unreasonable and we are all in agreement on next steps.

Daniel Burnett: https://github.com/w3c/did-core/issues/75

Daniel Burnett: looks like conversation is still happening

Orie Steele: I tried a PR for this and failed…

Daniel Buchner: there is some discrepancy whether 75 and 14 are related. seems to be most people don’t want a way to formally mark revoked keys, but some do. not sure how to move this forward.

Manu Sporny: https://github.com/w3c/did-core/pull/224

Manu Sporny: there was an attempt in PR 224 that did not gain consensus. The conversation was supposed to jump back to the issue, but it didn’t. If someone could propose some language.

Orie Steele: the spec is super weak… and there is no consensus.

Daniel Buchner: I think is comes down to: some people want this formal in the spec and some people don’t. Hard to write spec language around a binary thing.

Manu Sporny: the third way is to let the spec stay vague, it allows it and doesn’t forbid it. This can help us move forward when the positions are so polar. that wat people who want to can, but people who don’t, don’t have to.

Daniel Burnett: that may work, if it does, great. The standard one in this case is when there are disagreements, the people involved need to get on a call and talk with the goal of moving the spec forward.
… if the people who disagree aren’t able to work it out we can have a more moderated call. I hope those individuals will reach out to each other.

Manu Sporny: I added this comment: https://github.com/w3c/did-core/issues/14#issuecomment-613514751

Daniel Burnett: sounds like there’s a proposal from manu on possible next steps, let’s see if that goes anywhere.
… next one, same issue, same status

Daniel Buchner: LOL, Orie already down-voted what we just said in GH

Daniel Buchner: Orrieally?

Daniel Burnett: https://github.com/w3c/did-core/issues/194

Daniel Burnett: if you want to down-vote, please recommend an alternative proposal

Manu Sporny: I don’t know what to do with this one.

Brent Zundel: there is a massive conversation that is hard to boil down to something actionable.

Manu Sporny: I’d like to close this, but I can’t
… figure out what concrete text needs to be changed.

Daniel Burnett: https://github.com/w3c/did-core/issues/5

Daniel Burnett: sometimes you need to get specific.
… a theoretically simple question.
… the versioning issue may be keeping this alive.

Orie Steele: some will live in did-core-registries, others will live on the internet….

Orie Steele: NONE will live in did-core.

Daniel Burnett: it is assigned to chairs and staff, it was assumed we just needed to do it, but that’s not the case

Manu Sporny: there has been progress.
… did-core contexts live here now.
… third-party did extension contexts is what remains
… then there’s the versioning strategy.
… that’s what is holding this up, we need a concrete proposal.

Daniel Burnett: is there somebody other than you that can propose a versioning strategy?

Drummond Reed: Can we be specific about what the “versioning strategy” entails?

Manu Sporny: anyone else who is comfortable managing contexts on a 10 year timeline?

Dmitri Zagidulin: I have a concrete proposal that I linked to in the issue. I’d like to hear counter arguments against.
… it would solve a lot of our problems, both while we are rapidly moving during the spec writing, and give stability after we are done.
… where should I put the text?

Orie Steele: lets take it to the did core registries issue dmitriz…I can help

Orie Steele: IMO, this issue is where this discussion should be had: https://github.com/w3c/did-core-registries/issues/17

Amy Guy: wouldn’t this go in the … yeah what orie just said ^

Orie Steele: See the issue in did core registries

Dmitri Zagidulin: what directory and what file should I make the proposal against?

Manu Sporny: did core registries is where we should put this, but maybe something in an issue first

Daniel Burnett: if he’s got text somewhere that may solve this, it should be copied here somewhere.

Manu Sporny: it would be nice to discuss this in the issue in did-core-registries and then go to PR if there isn’t violent disagreement.

Daniel Burnett: https://github.com/w3c/did-core/issues/119

Daniel Burnett: TAG review, we know the status of this one. still the same.

Daniel Burnett: https://github.com/w3c/did-core/issues/195

Manu Sporny: Added some ideas on what we need to do here dmitriz – https://github.com/w3c/did-core/issues/5#issuecomment-613519860

Daniel Burnett: says blocked by did-core registries. Tobias is not on.
… Orie?
… is it still blocked?

Orie Steele: yeah. We’ve made progress on this. a lot of these properties have been added to the did-core registry. But I think this may be did method specific. I will add that to the issue.

Daniel Burnett: https://github.com/w3c/did-core/issues/10

Daniel Burnett: Mike, can you give us a status on this one?

Michael Jones: this is one of the JSON-LD signature identifiers. Have we removed the dependency on the community group’s registry for algorithm identifiers?

Manu Sporny: I believe that everything we need has been moved into, or is in the process of being moved into the did-core-registries context.
… so there’s no external dependencies. there is a deeper question around whether the did core spec should say anything about digital signature protection of docuemtns beyond saying there is cryptographic proofing using public keys.

Brent Zundel: I think proof should be taken out and made part of the extensions rather than in the core spec. that would remove this issue.

Michael Jones: could you link to the part of the did core registry where this is?

Orie Steele: the proofing issue doesn’t solve the issue fully, the way these suites define a key representation and signature representation together
… I will add some comments about this. This really affects interoperability, so I will add that to the issue.

Michael Jones: I agree with that analysis

Daniel Burnett: https://github.com/w3c/did-core/issues/8

Daniel Burnett: I think this is you Mike, oh wait, it says the chairs will schedule a discussion.

Michael Jones: to the extent we support JWK key representations, they can have algorithms which come from this registry. I will comment to that effect.
… I think that’s all we need to do for right now.

Daniel Burnett: https://github.com/w3c/did-core/issues/92

Daniel Burnett: CBOR

Manu Sporny: we’re waiting on Jonathan to give us the CBOR stuff.

Drummond Reed: Should we assign the issue to Jonathan?

Daniel Burnett: right, this is the one that used to be IPLD. He has responded, and we’re not far enough to say there is a deadline.
… the only thing that would be good is to ad an update ping in the issue.

Justin Richer: I wanted to point out that since this issue has been raised and discussed there is a spot in the spec for it to go.
… along with enough structure to allow them to define how to get into and out of CBOR, whether that uses IPLD or another method.

Daniel Burnett: https://github.com/w3c/did-core/issues/160

Michael Jones: I have to do the PR. I will try to do that soon.

Daniel Burnett: we are all doing the best we can right now.

Daniel Burnett: https://github.com/w3c/did-core/issues/85

Daniel Burnett: this is a metadata one. I don’t think there’s any new status

Michael Jones: that is correct. this depends on the whole metadata discussion. Is there anything else we should link to from this?

Daniel Burnett: and Justin’s point is good. If the actions that need to be tracked are elsewhere, we could point to them from this issue and close it.

Michael Jones: I’m not seeing this as a duplicate.

Justin Richer: The work we are doing on resolution contracts does begin to address this, but the PR hasn’t been accepted yet.
… the issue at its core is the difference in representing metadata and document data, which the PR addresses.

Daniel Burnett: It is good to link to the PR even before it is accepted.
… we’ll err on the side of linking to the PR when it is created.

Justin Richer: the link has been added. I request Mike takes a look at the PR.

Daniel Burnett: thanks everyone for joining a great productive discussion.

6. Resolutions