Generalizing AC Appeals and using this procedure for recall. #888
Conversation
frivoal
force-pushed
the
w3c-vote
branch
2 times, most recently
from
4866220 to
d6b102a
Compare
|
An alternative mid-ground would be stating a "supermajority" threshold:
twice => > 2/3 or 67%, thrice => > 3/4 or 75%
|
|
@chaals, I'd rather not phrase it this way, because when you just say "supermajority of 2/3" or some such phrasing, it's ambiguous how you treat abstain ballots. You can make it clear, but that usually make the phrasing longer and clunkier, which is why I think "x times as many ballots for as against" or that sort of phrasing is better. |
frivoal
force-pushed
the
w3c-vote
branch
2 times, most recently
from
d77299a to
e2edd24
Compare
|
I've updated this draft PR to reflect my current take on this issue, as expressed in https://github.com/w3c/AB-memberonly/issues/237#issuecomment-2354297741. |
frivoal
removed
the
Agenda+
|
The AB has not reached a conclusion on this topic. Temporarily removing agenda+ |
index.bs
Outdated
| An [=Advisory Committee representative=] initiates a [=vote of no confidence=] | ||
| by sending a request to the Team, and <em class=rfc2119>should</em> also share this request with the Advisory Committee. | ||
| The request <em class=rfc2119>must</em> identify which of the [=AB=] or [=TAG=] is targeted, | ||
| and <em class=rfc2119>should</em> also include the rationale. |
There was a problem hiding this comment.
I think that you want a three member threshold for this too. Otherwise, this is open to trolling and DoS.
I'd be OK with a higher threshold than three, but not a lower one.
If that takes the form of one AC member initiating an override that has to be seconded by two other members in all cases, that would be ideal. I know that this mechanism hasn't been activated, but it's an organizational vulnerability.
There was a problem hiding this comment.
It already is a 3 stage thing:
- someone calls for a vote of no confidence
- we check if >=5% of the membership agree that we should run a vote of no confidence (within a time limit of 1 week)
- if so, we run the actual vote of no confidence
Adding an "at least two people need to agree" seems redundant with step two.
There was a problem hiding this comment.
I think that the second step is not feasible (you allow more time for the final vote, which only needs to clear the same threshold). So I am suggesting a replacement for that stage.
Addresses w3c#888 (comment) Related to w3c#882.
|
I think this PR needs some updating; and we should clean up the discussion to focus on the current proposal at hand, rather than on previous variations of trying to address this problem. @frivoal Would it make sense to do the clean-up here, or to summarize the open points of discussion into the issue and open a new PR? |
Addresses w3c#888 (comment) Related to w3c#882.
Addresses #888 (comment) Related to #882.
This extracts the 5% confirmation vote, followed by the actual vote into a separate procedure, invoked by the AC Appeal, making it reusable. Co-authored-by: Ted Thibodeau Jr <[email protected]>
See w3c#882 Co-authored-by: Ted Thibodeau Jr <[email protected]> Co-authored-by: Martin Thomson <[email protected]>
The vote of no confidence phrasing seemed clunky, and didn't mesh well with the section's title. Use the word "recall" instead.
This comment was marked as spam.
This comment was marked as spam.
Sorry, something went wrong.
Use the same word (invoke) to declare the start and to talk about when we count 6 months from.
Co-authored-by: Jeffrey Yasskin <[email protected]>
| all [[#AB-TAG-elections|elected]] and [[#TAG-appointments|appointed]] participants | ||
| from the [=Advisory Board=] or the [=Technical Architecture Group=]. | ||
|
|
||
| An [=Advisory Committee representative=] invokes a [=recall=] |
There was a problem hiding this comment.
It seems like the bar for triggering this should be much higher than a single AC representative request.
There was a problem hiding this comment.
read on. This starts with a single person (how else), but then you need to meet the same criteria as an AC appeal, which means finding 5% needs to support having the vote, or it stops right there.
| The [=Advisory Committee=] <em class=rfc2119>may</em> | ||
| <dfn export>recall</dfn>, | ||
| as a whole, | ||
| all [[#AB-TAG-elections|elected]] and [[#TAG-appointments|appointed]] participants |
There was a problem hiding this comment.
Why can the AC recall appointed seats?
Elected seats, I get. The people in those seats were put there by the membership and are accountable to it.
But part of the rationale I've heard for retaining appointed seats on the TAG is so the Team can fill gaps that the membership left when it elected the rest of the TAG, which may or may not be gaps the membership values or prioritizes. It seems to me, therefore, that the appointed seats aren't accountable to the membership, which didn't pick them, but instead to the Team, which did.
There was a problem hiding this comment.
That said, it seems like the Team ought to have the power to recall people from appointed seats. It put them there for reasons. If it turns out their presence is not conducive to advancing those reasons, the Team ought to be able to remove them and appoint someone else.
There was a problem hiding this comment.
Why can the AC recall appointed seats?
Individually, it should not (and in this proposal, doesn't). As a whole, that's indeed debatable. Here, the idea is that if specific individuals, appointed or elected, need to be removed the group itself should deal with that, but the AC finds the group as a whole to be dysfunctional / harmful beyond redemption, it can call for a clean slate. Why kick the appointed people out too, even though they're not elected by the membership? My idea would be that (a) if the group is a whole is broken, they're part of that, and (b) they're appointed by the Team in order to complement the elected positions. If we're starting from scratch on that side, they need reevaluation too.
That said, while it makes sense to me to do it this way, I don't think I'd object to the recall only affecting the elected seats.
As for the Team having the power to recall the people it put it place, I'd need to think about it. First, it's not purely a Team appointment: the team proposes, but the TAG+AB decides. TAG members are supposed to act indepentently, as individuals, not in the interests or on behalf of any particular organization, but for the common good of the web. If the Team can give you the job and take it away, I think that might be a bit too direct of a power dynamic for my taste, making the appointees feel that they work for the Team, which I don't think would be appropriate. Also, "out their presence is not conducive to advancing those reasons" seems too low a bar to eject people. Removal from office is a pretty drastic measure, and I don't think being mediocre, or being capable but not in the expected way, is a justifiable reason.
| @@ -1356,7 +1381,7 @@ Elected Groups Vacated Seats</h5> | |||
| the participant resigns, or | |||
|
|
|||
| <li> | |||
| the participant is [=removed=], or | |||
| the participant is [=removed=] or [=recalled=], oor | |||
There was a problem hiding this comment.
| the participant is [=removed=] or [=recalled=], oor | |
| the participant is [=removed=] or [=recalled=], or |
typo
This includes the following commits 1c9f1a7 Enable AB/TAG to remove one of their members by super-majority See #882 ~~~~ 207955b Update Changelog ~~~~ 6553a80 Remove 'normally' which is unnecessarily judgemental and unnecessary. ~~~~ 12eef64 Notify the AC when an AB or TAG member is removed. Addresses #888 (comment) Related to #882. ~~~~ a3bbb15 Removal from AB/TAG does not impact Council membership #1006 ~~~~ 193b820 Removal hearings are held by the group, not the chair. Addresses part of #1007
This PR is a first draft attempting to address
#886 and#882.Neither haveIt has not been resolved on at this point, but this shows what adoptingthemit could look like.It can be reviewed as a whole, or commit by commit
, to distinguish the effects of #886 from those of #882.update: #886 has been handled separately, removing discussion of it from this pull request.
Preview | Diff