Revising W3C Process Community Group

Propose to Close Items

Brent: I have removed "Propose to Close" label to any issue where there was discussion since the previous call.

Florian: Thank you. When you do, please add agenda+ to such issues

Brent: Issues that are important that we close will certainly be re-raised moving forward.

florian: +1. I recently converted issue 948 from "propose to transfer" to "propose to close". It was a pubrules issue and was fixed.

florian: I'd rather we not close 903 yet until we've heard back from PSIG

PROPOSED: Close issues 1064, 1044, 1006, 963, 948, 946, 942, 933, 326

<florian> +1

<Ian> +1

<brent> +1

<tidoust> +1

SO RESOLVED

ACTION: Brent to close issues 1064, 1044, 1006, 963, 948, 946, 942, 933, 326 in GitHub

Agenda+

<brent> https://github.com/w3c/process/issues?q=state%3Aopen%20label%3AAgenda%2B

<brent> w3c/process#1047

Florian: Process today (in section 5.6.2.7) says: "In case of a vote, if two members of a Council who share the same affiliation cast an identical ballot, then their ballots count as a one vote, not two."
… however, in rare situations, there can be (temporarily due to changes on the AB or TAG) more than 2 people on council with same affiliation.
… note that in a WG, if this arrives then company hierarchy can help sort it out.
… but that's not an option in a Council because people are not representing their companies.
… proposal here is "if they can't agree their votes cancel each other."

Ian: Do votes happen secretly?

Florian: The vote happens in secret but there is some info shared about the results. The council report names who voted, and the vote totals.

Ian: Does cancellation affect quorum?

Florian: I don't think it affects quorum (and I don't think it should).
… there is no strict sense of a quorum in a council

Ian: +1 to merge 1047.

RESOLUTION: Merge pull request 1047

<brent> w3c/process#1040

Florian: The Process grants the CEO authority to take disciplinary action and the CEO delegates this in guidelines.
… the proposal is to update the guidelines to name the council explicitly as in scope for suspensions

PROPOSED: Transfer this issue to the Guide repo

<Ian> +1

<florian> +1

<brent> +1

<tidoust> +1

SO RESOLVED

ACTION: Brent to transfer issue 1040 to the Guide repo

<brent> w3c/process#1038

Florian: Regarding elected group vacated seats, the Process says "The group Chair should not request such an election if the next regularly scheduled election is fewer than three months away." An ambiguity arises because an election is not a point in time. But this can be best addressed in the Guide, so this is a proposal to transfer.

Ian: My preference would be to clarify the understood meaning: three months from the start of the election.

Florian: I don't object to merging, but it's still interesting to have discussion in the Guide about when a chair might or might not start an election.

Brent: +1 to merging

<Ian> +1 to merging

Ian: I don't think we have a lot of experience about starting/not starting election. I would hesitate to document good practice without much practice.

Francois: +1 to merging in process and waiting until a concrete problem arises before documenting good practice

RESOLUTION: Merge pull request 1038

ACTION: Brent to merge 1038

<brent> w3c/process#1029

Florian: A case arose where a council addressed an FO and started work on a report when another FO was registered. We could propose something to the AB or ask them to start thinking about it without a proposal from the CG.

Ian: I would suggest proposing an approach that favors continuity to the AB

Florian: It's reasonable to ask the AB but also I don't think we have to.
… we could just handle it; it's not controversial

Francois: What causes a council to be disbanded?

Florian: Right now, publishing a report.

ACTION: Florian to write up a proposal to address question of new FOs that arrive when a Council has nearly finished or finished recently.

<brent> w3c/process#914

Florian: The process used to mix "non-normative" and "informative" but now only says "non-normative" EXCEPT where "informative" is used intentionally to mean "carrying information"
… there's one remaining case of "informative" to mean "non-normative" but it's generated by bikeshed. I've filed a bug with bikeshed so I think we can close 914

PROPOSED: Close 914

<Ian> +1

<tidoust> +1

<brent> +1

So RESOLVED

<brent> w3c/process#904

Florian: The idea was floated to use a registry to manage a glossary. The process issue was whether the AB and/or TAG could publish registries, and now they can. So this is no longer a process issue so we can transfer this issue to the AB.

Francois: Terms are sometimes used in normative contexts (e.g., algorithms)

Florian: Things that are normative should be defined in a spec. The glossary should just be a way to find them.

Ian: We have had many glossaries. They are hard to maintain. Have these glossaries ever been widely used?

Francois: We have a glossary (Webdex) gleaned from specifications

(We discuss whether to transfer this issue to the AB or simply close.)

PROPOSED: Close issue

<Ian> +1

<brent> +1

<tidoust> +1

<florian> +1

Ian: Suggest link to webdex be included in the comment

Florian: We should also say "If people think a registry managed by the AB is the right thing to do, then we note that this is now possible."

ACTION: Brent to close issue 904