Closing Groups

[mnot]

The director-free branch only empowers the Team to initiate an AC review to close a group before its charter is fulfilled. Should any other party (e.g., the TAG) have this power?

First raised in #316.

[frivoal]

I'd be OK with allowing the TAG (and AB?) as well to make such proposals. In practice, I'd expect AB/TAG/Team to collaborate on such things and be roughly in alignment, and the proposal to have a hard time succeeding in AC review when they are in disagreement, but I don't think broadening the ability to propose this AC review would be a source of problems. The AC decides anyway.

[mnot]

Tend to agree re AC. There are other parts of the process that specify multiple ways to initate an action, e.g., 6.3.13.4. Process for Rescinding, Obsoleting, Superseding, Restoring a Recommendation. It'd be good to consider where alignment makes sense.

[mnot]

#660 Looks good enough for 2022. A future effort might look at aligning this with other mechanisms as per above. Thanks!

[nigelmegitt]

I don't understand what problem this is addressing: it seems premature to have an open pull request for this, and I will comment this on #660. This issue is raised as a question, and there haven't been any comments in favour or against, just one "I'd be okay". We should have more motivation for the change.

Right now, if anyone can persuade the team to open an AC review, the team can do it. The idea that other parties can initiate an AC review, without even consulting the team, seems to carry risk unnecessarily, albeit low risk practically. I'm particularly thinking about any requirements that can put different sub-sets of the W3C community at odds with each other.

For example, if Team looks after strategy and the existence of a WG fits the overall strategy, but there is for whatever reason non-alignment between AC or AB and the Team, that's a problem that needs to be addressed first, rather than allowing any of the above sub-sets to take unilateral action.

[frivoal]

Opening a PR doesn't necessarily mean it's ready for adoption. I made this just so we could be clear about what the proposal is. If we like it, it's easy to pull in, but it can just as easily be rejected altogether.

[css-meeting-bot]

The Revising W3C Process CG just discussed Closing Working Groups, and agreed to the following:

• RESOLVED: Defer 653 for 2023, unless mnot pushes back

The full IRC log of that discussion

<fantasai> Subtopic: Closing Working Groups
<fantasai> github: https://github.com//issues/653
<plh> Github: https://github.com//issues/653
<fantasai> florian: mnot noticed that there are some things that can be initiated by Team and some other people, but closing groups can only be proposed by the Team
<fantasai> ... either way there's an AC Review, so can't force it through
<fantasai> ... but who can initiate that?
<fantasai> florian: I opened a minimalist PR to extend it to Team or AB or TAG
<fantasai> ... Nigel noticed that he didn't think it was a good idea
<fantasai> ... I think he prefers to funnel through the Team, to avoid e.g. AB wants to close a group while TAG has it open or vice versa
<plh> q+
<fantasai> ... I think the PR is simple, but whether we want to adopt or not is open question
<fantasai> plh: Right now the Director can propose to close a group
<fantasai> ... your proposal doesn't do that in spirit of Director-free, says TAG *or* AB not TAG *and* AB
<fantasai> ... Council is both
<tzviya> q+
<fantasai> florian: Council is a lot more complicated than TAG+AB
<plh> ack plh
<fantasai> ... if they were doing this on their own, I'd be concerned, but they have to start an AC Review
<plh> ack tzviya
<fantasai> plh: Team doesn't get a choice, would have to start an AC Review once requested
<fantasai> tzviya: There's some concern that the TAG would take on more work and AB would have less work, and for things like this as well as Charter Review, room for AB and TAG to work together
<fantasai> ... we're already working together for Councils
<fantasai> ... extraordinarily hard to coordinate, coordination is half the work
<fantasai> ... confluence of policy and tech work
<fantasai> ... I think it makes sense to work together
<fantasai> ... I think we should do something about this, how can we get the best of both groups
<fantasai> florian: personally, even if it's a good idea, it's non-urgent
<fantasai> ... in normal circumstances, we can tell the Team and they'll listen and there's no blocking thing here
<fantasai> ... so in theory finding a broader way might be better, but we can defer this to next year
<fantasai> ... Mark raised it as part of Director-free
<fantasai> ... so if we're not doing this now, we need to argue with Mark
<florian> q+
<fantasai> fantasai: [...]
<plh> ack florian
<fantasai> fantasai: Could say TAG+AB, would only really be relevant when the Team disagrees
<tzviya> q+
<fantasai> florian: Situation is about forcibly closing a group that otherwise wouldn't stop on its own
<plh> ack tz
<fantasai> tzviya: I can easily see a scenario where this could happen
<fantasai> ... let's say a subgroup within W3C who want to specify a technology, but W3C at large wants to disassociate itself from that work
<fantasai> florian: Not saying there isn't a case for that, but that it would be politically interesting
<fantasai> plh: We always have some WGs that we wish we didn't have
<fantasai> ... can make case to close several groups as of today
<fantasai> ... but not going to
<tzviya> q-
<fantasai> florian: Back to the issue, we can do several things
<fantasai> florian: a) defer
<fantasai> florian: b) switch to AB + TAG, if in agreement
<fantasai> plh: c) ask AB for feedback
<fantasai> plh: If we do that kind of change, we're adding more responsibility
<fantasai> florian: I don't expect this to be routine
<fantasai> ... this has been extremely rare or never happened, and I expect that to remain the case
<fantasai> tzviya: I would say defer
<fantasai> florian: Maybe defer and if mnot isn't okay with it, we work on it more?
<fantasai> plh: I would be happy for us to defer, not to merge
<fantasai> plh: objections to defer? If mnot comes back we can come back.
<fantasai> RESOLVED: Defer 653 for 2023, unless mnot pushes back

[frivoal]

Will link to the minutes once they're published, but until then, here's the relevant AB resolution on this topic:

RESOLVED: Not only the Team, but also the AB or TAG, can propose closing a working group.

[nigelmegitt]

@frivoal many thanks for the early heads up. Was any proposal made to change the criteria for such a closure, i.e. from the two in the current Process (i.e. lack of work or early completion)?

Edit: I see that those criteria were removed in #585, without any apparent discussion, and have opened #685 requesting reintroduction of that wording.

[css-meeting-bot]

The Revising W3C Process CG just discussed Allow AB or TAG to propose closing a group.

The full IRC log of that discussion

<fantasai> topic: Allow AB or TAG to propose closing a group
<fantasai> github: https://github.com//issues/653
<fantasai> florian: Needs a PR, should be the same PR as for 685
<fantasai> s/topic:/subtopic:/

[frivoal]

Closed by #696