DID Working Group Telco — Minutes

Date: 2020-02-18

See also the Agenda and the IRC Log


Present: Daniel Burnett, Ivan Herman, Justin Richer, Joel Hartshorn, Chris Winczewski, Markus Sabadello, Manu Sporny, Dave Longley, Jonathan Holt, Orie Steele, Yancy Ribbens, Adrian Gropper, Kenneth Ebert, Dmitri Zagidulin, Joe Andrieu, Brent Zundel, Drummond Reed, Michael Jones, Sumita T. Jonak, Kim Duffy, Ganesh Annan, Kaliya Young, Andrei Sambra, Thomas Schwarz

Regrets: David Ezell


Chair: Daniel Burnett

Scribe(s): Kenneth Ebert


1. Agenda Review, Introductions, Re-introductions

Daniel Burnett: Agenda for today is simple.
… Working mode, and then issues.
… Questions or changes?

Manu Sporny: Updates on registry doc.
… A lot of changes and we would feedback to editors on where we should be working.

Daniel Burnett: Introductions?

2. Editorial team for new docs

Daniel Burnett: Several new documents were created as holding places for new discussions.
… Implementation guide, registry. We need editors
… In the case of the registry doc, Orie volunteered along with Manu.
… If others volunteer we could make adjustments.

Michael Jones: There are likely to be more than one registries.

Daniel Burnett: There has been some disagreement on the number of registries. When I say registry I mean registry or registries or registry mechanisms.

Michael Jones: Terminology is important.

Daniel Burnett: Yes. We should discuss the mechanism.
… We need a place for the discussion.
… We are looking for people to be editors on the other documents. If you are interested, please contact the chairs.
… I see people making nominations.

Orie Steele: I nominate daniel b from Microsoft and sam smith

3. Group work mode

Daniel Burnett: We have some big decisions to make. We need to get into more detail.
… We need to switch into a more detailed work mode.
… Please write pull requests.
… Editors will merge PRs when they believe there is consensus.
… Please object within two days if you disagree.
… If there is not consensus, the editors will back out changes.
… Don’t quit, or be upset. Let the chairs know if there is not consensus.
… We want to move faster, but not try to push things through.
… On the calls, we will review status, not do the detailed work of each issue.
… Issues are assigned to someone to push the work forward.
… If an issue sits with no progress, we may suggest closing the issue.
… Work needs to move forward between calls.
… Watch the repos that you are interested in.

Drummond Reed: +1 to Orie’s point

Daniel Burnett: Anything with discussion will remain in discussion until resolution.
… Make PRs to get your changes in.

Michael Jones: I request that weekends be excluded from the two days.

Joe Andrieu: +1 to business days

Ganesh Annan: +1 for excluding weekends

Drummond Reed: +1 to 2 business days

Kim Duffy: that seems reasonable

Daniel Burnett: I agree. Two business days is the guideline.
… Editors should ensure that those who have concerns should be sought out for feedback.
… Don’t try to sneak things in.
… We don’t want to overrun people’s concerns.
… If you will be gone, let the editors know so that your concerns can be addressed.

Drummond Reed: -1 to ever “sneaking anything past anyone”

Dave Longley: +1 to no sneaking

Dave Longley: +1 to benefit of the doubt on good faith of editors

Manu Sporny: (no audio)

Jonathan Holt: I appreciate the need to move forward. I raised issues a long time ago in Barcelona. As an invited expert, I have added thoughts, but this is not my full-time gig.
… I am trying to guide the discussion, but don’t have all the time I need to contribute.

Drummond Reed: This is my life!!!

Daniel Burnett: All of us have differing time constraints. There are companies who need to implement this, and need to make progress.
… The editors have to review all the issues and PRs every week.
… Please do your best. We are all busy and we need to do the best we can.

Brent Zundel: Every in the group, please create PRs. Even if they are not perfect, the editorial team will help.
… Be the change you want to see.

Manu Sporny: We are auto-assigning issues. If you raise the issue and you are in the WG, you will likely be assigned.
… For those outside the group, the editors will use best judgement to make assignments.

Brent Zundel: if someone WANTS to work on an issue to which they are not assigned, they can also be assigned

4. Issue status check

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

Daniel Burnett: Here is a link to the did-core repo.
… It is sorted by progress.
… The goal is to move issues forward.

Daniel Burnett: https://github.com/w3c/did-core/issues/33

Daniel Burnett: dmitriz?
… What needs to happen?

Dmitri Zagidulin: I think this is out of scope but migration is in scope.
… Chaining for delegation should be discussed.
… I think the issue should be rephrased or a more narrow issue created.

Daniel Burnett: I recommend that you write what you said into the issue.
… Orie has jumped in.

Daniel Burnett: https://github.com/w3c/did-core/issues/35

Drummond Reed: I feel that we should be settling on matrix parameters first.
… Are we keeping the current syntax or query parameter?
… I’ll add a comment.

Markus Sabadello: Depends on resolution of https://github.com/w3c/did-core/issues/159

Daniel Burnett: https://github.com/w3c/did-core/issues/64

Dmitri Zagidulin: I’ll add a comment recommending closing this issue with reasons.

Manu Sporny: Should issues be marked as pending close first?

Daniel Burnett: That would reflect an editor’s opinion.

Brent Zundel: dmitri opened this, so should he be able to just close it?

Drummond Reed: +1 to editors marking issues as “pending close”

Daniel Burnett: dmitriz is not an editor; it would be his opinion.

Ivan Herman: “Suggest close” label?

Daniel Burnett: We will discuss how to do this with the editors and chairs.
… Rules, etc.

Daniel Burnett: https://github.com/w3c/did-core/issues/20

Michael Jones: My question stands. I think we should close it.

Daniel Burnett: Michael is not a member of the WG, please add @name to let him know that he needs to respond.
… That will be a courtesy notice to him.

Michael Jones: I can do that.

Daniel Burnett: https://github.com/w3c/did-core/issues/25

Drummond Reed: This feels like it should be attended to as editorial.

Daniel Burnett: Please add a comment that says that and see if there is any disagreement.
… Everything needs a PR.

Drummond Reed: I will do this.

Daniel Burnett: https://github.com/w3c/did-core/issues/36

Ivan Herman: Github usage; if drummond writes “fixes #25” in the PR then it will be automatically close the issue. Is this a feature we want to use?

Joe Andrieu: -1 to auto-close

Daniel Burnett: Sometimes a PR submitter puts “fixes 25” but there is additional discussion in the issue, then the other work will be lost.

Justin Richer: -1 to auto-close. Issues can be re-opened but that’s a problem

Daniel Burnett: -1 to auto-close

Manu Sporny: Editors should scan for the this corner case to ensure that the PR really does fix all of the issue.
… I would encourage highly focused PRs so that we can truly address a single issue and the “fixes 25” is accurate.
… It is an extra burden for the editors, but useful functionality.

Daniel Burnett: Closed issues can be re=opened if there is a mistake.

Justin Richer: I have a lot of experience with github managing issues. We have used “addresses” because it creates a link without taking automatic action.
… Also works to reference PRs.

Orie Steele: +1 to linking…

Brent Zundel: +1 to “addresses” for linking

Drummond Reed: +1 to linking

Daniel Burnett: +1 to “addresses” for linking

Manu Sporny: -1 to addresses, but I won’t stand in the way of the group.

Justin Richer: The automatic actions are problematic.
… Narrow PRs are good, but often issues are broad and tend to expand.
… to be clear, there’s nothing special about “addresses”, it’s just readable

Manu Sporny: The Editor’s process is designed to work well with the Fix #XYZ mechanism.

Manu Sporny: (but again, I won’t stand in the way of the group – it creates more work for the Editors)

Daniel Burnett: https://github.com/w3c/did-core/issues/36

Daniel Burnett: Please don’t use “fixes” or “closes”. Please do link with “addresses” or something else.

Drummond Reed: this is the same status as dlongley has commented.

Brent Zundel: perhaps there could be a “Depends on Matrix Parameters” label?

Daniel Burnett: https://github.com/w3c/did-core/issues/55

Daniel Burnett: Please add content to issues, not just comments that will show an update without progress.

Michael Jones: I think we should remove all instances of ethereum addresses from the spec and close this issue.

Daniel Burnett: I think there will be some interesting discussion in this issue.

Orie Steele: There will be issues around “where is this term defined?”
… I think there will be discussion around where the definitions are stored.

Daniel Burnett: https://github.com/w3c/did-core/issues/14

Daniel Burnett: I think it is important to distinguish between comments as an editor and comments as an individual contributor.
… None of those assigned is present today.
… I’ll ping them all!

Jonathan Holt: I think this should be a revocation methodology, not just a list.

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

Daniel Burnett: Please add that to the issue.

Ivan Herman: We had a discussion before the FPWD publication. We arrived at a tentative decision, but Manu had some concerns.

Manu Sporny: I’m happy to close if there is no objection to the URL that has been selected.
… We need a versioning and caching strategy for the group.

Daniel Burnett: Ivan please add your opinion. Manu please add your concerns.

Ivan Herman: I am happy leaving this issue open.

Daniel Burnett: On the issues list, look at the sorting options so that you can look at the recently updated issues. This will help you stay up to date without having to review all of the issues,

Manu Sporny: I agree with what Ivan just said.

Ivan Herman: There are two issues that can be (104 105) addressed after the core changes settle.

Daniel Burnett: We will continue to review issues weekly.
… When there are PRs we will close related issues.
… Thanks all!
… See you next week.