Wikipedia talk:Requests for comment/Arbitration Committee Elections December 2024

v t e 2024 Arbitration Committee Elections 2024 ArbCom pre-election RfC Contact the coordinators Discuss the elections Quick guide Candidates: Candidate guide Candidate statements Questions for the candidates Discuss the candidates Personal voter guides These guides represent the thoughts of their authors. All individually written voter guides are eligible for inclusion.

Discussion at Wikipedia talk:Requests for comment/Arbitration Committee Elections December 2023/Electoral Commission#Qualifications for the commission highlighted that the current requirements (Open to anyone who is over 18, meets Foundation's Access to nonpublic personal data policy, and otherwise be eligible to vote.) are not ideal. I suggest we replace it with:

This is stricter than the requirements to vote, but the role is mostly about solving problems and we really want experienced editors for that. An editor blocked from the Wikipedia or Wikipedia talk namespaces would not in practice be able to undertake the role. Requiring extended confirmed automatically imposes an edit count requirement without regard to election-based deadlines, what counts as sufficient activity can be judged by those commenting on the nomination. Thryduulf (talk) 17:53, 7 October 2023 (UTC)[reply]

Works for me, thanks. RoySmith (talk) 17:59, 7 October 2023 (UTC)[reply]

To better reflect the effect of not choosing support or oppose, the third option is proposed to be "abstain". This aligns with the consensus decision in 2013, Wikipedia:Requests for comment/Arbitration Committee Elections December 2013 § Voting procedure: proposing change "No vote" to "Abstain".

Note in 2014, Wikipedia:Arbitration Committee Elections December 2014 stated that there would be an abstain option, but reported the results with a "neutral" option. Since then, every election has used "neutral", although there has been no explicit consensus to change the 2013 decision. (In 2017, Wikipedia:Requests for comment/Arbitration Committee Elections December 2017 § Should oppose votes be abolished? failed, and the summary statement assumed that the existing choices in effect were support, oppose, and neutral.) isaacl (talk) 00:34, 12 November 2023 (UTC)[reply]

For reference, I have mocked up what the ballot looks like for someone who voted already:

Welcome <user>!

Note: You have voted in this election before. You may change your vote by submitting the form below. Note that if you do this, your original vote will be discarded.

Election main page • Report problems or issues

Instructions

• Use the radio buttons. Please use the radio buttons below to indicate your preference for each candidate with "Support", "Oppose", or “Neutral”. A “Neutral” vote does not affect the outcome in any way.
• Vote in a single sitting. Voting must be done in a single sitting.
• You may change your vote by starting over. After your vote has been accepted, you may change your vote any time before the close of voting. To do so, reuse the voting interface and a fresh default ballot page will be displayed. You will need to complete the process again from scratch. For this reason, consider keeping a private record of your vote. Your new ballot page will override the old one.
• Discuss the candidates. You may read and engage in discussion of the candidates here.
• Give feedback on the election. Please give feedback on the election here.

2023 English Wikipedia Arbitration Committee candidates

isaacl (talk) 16:47, 2 December 2023 (UTC)[reply]

The eligibility rules at WP:ACE2023 talk about how many edits you've made, when you registered your account, etc, but don't actually specify that these mean "on enwiki". It's particularly ambiguous what "has a registered account" means in the context of Special:CentralAuth; If you register your account on another project and haven't attached to enwiki yet, have you met the requirement? Let's just say "on enwiki" in the appropriate places and avoid the ambiguity. RoySmith (talk) 01:06, 2 December 2023 (UTC)[reply]

I think it was clear in the proposal at Wikipedia:Requests for comment/Arbitration Committee Elections December 2019 § Voter activity requirements that "mainspace" and "any namespace" referred to English Wikpedia, and so I think that can be updated now. It's not as clear that everyone had the same interpretation of having a registered account, unfortunately. isaacl (talk) 03:58, 2 December 2023 (UTC)[reply]

I agree regarding edits, both in terms of meaning and updating now. I suppose that theoretically means that an IP editor who has an account on another Wikimedia wiki with the requisite qualifications there, but who does not use it to edit en.wp for some reason would be eligible, but in practice they wouldn't be on the voter rolls and so couldn't vote. Someone who isn't a CU cannot verify the link between an account on another wiki and an IP address here, a global CU presumably could but I don't know whether someone who is only a local CU could? I can't imagine that this is a common occurrence though. Thryduulf (talk) 11:33, 2 December 2023 (UTC)[reply]

I'm not sure I understand your hypothetical scenario: are you considering someone with a registered account who doesn't use it to edit English Wikipedia, and has the requisite number of edits as a non-logged in editor in English Wikipedia mainspace and any namespace? I do not believe that the 2019 proposal intended to enfranchise editors meeting this scenario. I feel that the proposal to have a registered account combined with the number of edits requirements implied, by context, that the edits were made by the registered account. (Yes, there are some long-time non-logged in editors, but they're already disenfranchised by the need to ensure that editors only vote once.) isaacl (talk) 16:08, 2 December 2023 (UTC)[reply]

I'm not proposing such people were intended to be enfranchised, nor am I proposing they should be. I believe the clear intent of the proposal was to require all the requirements be met on the English Wikipedia. That's just the only scenario I can think of where someone could has a registered account, meets the editing requirements on en.wp but doesn't meet the registered account requirements on en.wp (which is what I understand your first comment in this section to be referring to). Thryduulf (talk) 22:31, 2 December 2023 (UTC)[reply]

Yes, I understood this wasn't something you were proposing. As you wrote "with the requisite qualifications there", it sounded like you were referring to the edit requirements being met on another Wikimedia site, but since this didn't seem to fit, I asked for clarification about the scenario you were considering. Thanks for the response! isaacl (talk) 04:02, 3 December 2023 (UTC)[reply]

The ElectCom selection process/possibility of having more specific guidelines for the ElectCom RfC closer could use some additional thought during ACERFC2024; see this discussion for some additional context. There isn't really a specific way of signaling disapproval of a candidate other than by endorsing other candidates, which I think provides relatively little guidance for the RfC closer. Best, KevinL (aka L235 · t · c) 20:44, 4 December 2023 (UTC)[reply]

As background context, see Wikipedia:Requests for comment/Arbitration Committee Elections December 2022 § Proposal 4: Election commission selection. There was very little support for having a formal way to express opposition. isaacl (talk) 21:08, 4 December 2023 (UTC)[reply]

My personal view is that the old system was functioning fine, until this year. I also don't necessarily think that adding opposes is necessary, but something needs to change — for example, an explicit rule that those who receive less than one-third the support of the person with the most endorsements cannot be selected for the commission or be a reservist. (Not an actual proposal! I'd have to think through where the line should be, etc., before actually proposing that. That's just there to give an example of what such a change could look like.) Best, KevinL (aka L235 · t · c) 21:20, 4 December 2023 (UTC)[reply]

Certainly sentiment could have shifted, and the experience of this year's selection process is a new factor. Just providing the context that many support the current system for reasons they listed. isaacl (talk) 21:30, 4 December 2023 (UTC)[reply]

It sounds like maybe more a call for a "Guide on how to close the ElectCom RfC" than a need to overhaul. Barkeep49 (talk) 21:48, 4 December 2023 (UTC)[reply]

There's been a perhaps inevitable trend towards greater formalization of governance... At first the electoral commission was put in place to address any issues for which the community was unable to make a timely decision. Then the community started to expect the commission to co-ordinate the election, and gave it authority over moderating the candidate questions and guides. The whole election RfC used to be a fairly quiet affair (well, during the time I paid attention to it); it's now become more sprawling with additional participants, in spite of (or maybe due to?) the election running fairly smoothly for a few years. isaacl (talk) 23:27, 4 December 2023 (UTC)[reply]

I argue all things considered it still is a lowkey affair and I see that as a big positive that I want to retain. "Something worked less than optimal this year" given what the something is doesn't strike me as a reason for creating specific guidelines for the closer. Having some sort of guide - that explains decisions past closers made - does feel reasonable. Best, Barkeep49 (talk) 23:54, 4 December 2023 (UTC)[reply]

If you don't want it to be considered as a guideline, then I suggest not labelling it a guide... Sounds like you are thinking of a compendium of the rationales of those who evaluated consensus for the past election commissioner RfCs, so future evaluators can consider past precedents? isaacl (talk) 01:06, 5 December 2023 (UTC)[reply]

I'm open to a range of options. I'm 100% on board with "there's no problem" as the justification not to change things, until there is a problem. This year, there was a pretty significant problem. Best, KevinL (aka L235 · t · c) 02:00, 5 December 2023 (UTC)[reply]

The problem might be more "we need a pool of editors experienced in determining English Wikipedia consensus who are willing to evaluate the outcome of discussions", though, rather than trying to craft algorithms to determine the outcome. isaacl (talk) 02:14, 5 December 2023 (UTC)[reply]

I have no idea how one would go about crafting a proposal to limit who can close formal discussions considering that the community has not really supported any substantive limits on closing discussions that don't involve the sysop tools. Right now, the only way to reverse a closer who doesn't reverse themselves—even from someone who is not experienced in determining English Wikipedia consensus who are willing to evaluate the outcome of discussions, as you say—requires community consensus at a formal AN review.
Maybe a good alternative is limiting the set of people who can close the commissioner RfC to bureaucrats? KevinL (aka L235 · t · c) 02:24, 5 December 2023 (UTC)[reply]

I wouldn't propose a rule on who can close formal discussions. The problem is broader than the election commission RfC: we need to try to foster an environment where more editors are willing to build up their experience in evaluating consensus. Part of that is being understanding of when evaluators make missteps, so they'll be encouraged to continue to improve. It feels like the participants in some categories of discussions are becoming less patient, which unfortunately makes it harder to find suitable volunteers to determine discussion outcomes.

Personally, I don't favour using bureaucrats as a pool of experienced discussion evaluators. I think it works against the idea of building up a broader pool of trusted evaluators. isaacl (talk) 02:55, 5 December 2023 (UTC)[reply]

My first thought was that if there is a desire to limit who can close the RFC (and I don't know at the moment if I would support that) was to require the closer to be someone who met the requirements I suggested above to be a member of the Commission. My second thought was that being over 18 and meeting the ANPD requirements aren't relevant to closing the discussion, and someone blocked from the Wikipedia namespace wouldn't be able to do it anyway. That just leaves being extended-confirmed, and if we're proposing that then we might as well add admins to the list of options, giving something like:

Who should be eligible to close the RFC for election commission selection:

• Anyone
• Extended confirmed editors
• Anyone with a track record of evaluating consensuses correctly
• Administrators
• Bureaucrats

Presented in that order because I don't think it's possible for anyone to establish a track record before they've become auto-confirmed. "Evaluating consensuses correctly" would probably need rephrasing but I mean at most a small proportion of their closes have been overturned. Thryduulf (talk) 03:40, 5 December 2023 (UTC)[reply]

I guess I just disagree that there was a significant problem this year. Best, Barkeep49 (talk) 02:21, 5 December 2023 (UTC)[reply]

I was in support in last year of a simple cut off of "Still only collect endorsements, but set a cut-off to those exceeding 50% of the third highest endorsement gatherer." the general consensus in the last RFC was that it didn't need fixing; feel free to propose though. — xaosflux ^Talk 00:59, 5 December 2023 (UTC)[reply]

Such a cut off this year would have produced the current slate (after the claw back occurred). — xaosflux ^Talk 01:01, 5 December 2023 (UTC)[reply]

We should consider adjusting the election timeline so that there is a larger gap between the end of the election and the end of the year. We could do this in a number of ways, but the obvious one is by reducing the gap between the end of the Electoral Commission selection and the start of the nomination period. As of when I am writing this, the scrutineers have not finished scrutinizing yet, and there are only 4 days left in the year. Even if the results are posted right now, this is still a very short amount of time for ArbCom to ensure a smooth transition between committees. Incoming arbitrators need to be subscribed to the requisite mailing lists, granted and trained on the use of checkuser and oversight permissions, and become familiarized with the more esoteric arbitration procedures, and ideally this should be done before they start their term. We can facilitate this by moving up the election a little bit for 2024. Mz7 (talk) 00:00, 28 December 2023 (UTC)[reply]

Thanks for getting this conversation started. The nomination period and the voting period could also be examined for their length. Also the idea of having local CU rather than stewards could be discussed because local CU are going to feel accountable to the community. I'm personally opposed to this last idea (because ArbCom oversee the CU and so there's some inherent tension there) but it feels like it belongs on the table. Best, Barkeep49 (talk) 00:55, 28 December 2023 (UTC)[reply]

Part of the challenge with that is the nature of the securepoll, it basically forces you to view CU data on every voter -- so there is a good argument that that is excessive local checking. — xaosflux ^Talk 00:57, 28 December 2023 (UTC)[reply]

This doesn't make sense to me so please make that good argument so I can understand better. It's the same invasion of privacy regardless of whether it's stews or local CUs doing it. Barkeep49 (talk) 01:04, 28 December 2023 (UTC)[reply]

@Barkeep49 the selected CU'sstewards should be more disinterested, in that they are specifically not part of the community of voters - additionally, they would not inadvertently overlap SP information with local CU investigations which wouldn't have reason to cu all of those people. — xaosflux ^Talk 02:59, 28 December 2023 (UTC)[reply]

I opposed your suggested adjustment in the last ACERFC and would do so again despite also being quite unhappy about where on the timeline we are this year. These are periods where I think it is both reasonable and important to give people the fullest opportunity to participate. The period Mz7 identifies above does not have the same importance from what I can see. I imagine commissioners do have some quantity of time in which the new ones need to onboard to what needs to be done, but 2.5 weeks to do seems in the realm of unnecessary (and that's even assuming the work they have to do is something that couldn't be done in parallel with nominations).

I'd like to hear from previous commissioners whether the 2.5 weeks is being well-spent today. Izno (talk) 01:32, 28 December 2023 (UTC)[reply]

@Izno there is seldom any commissioner-specific tasks that have to occur at that point in the schedule (see to do list) the main tasks are recruiting scrutineers and scheduling WMF resources. — xaosflux ^Talk 03:04, 28 December 2023 (UTC)[reply]

One thing for which buffer time should be allocated is to allow for changes to the election process. Until any related RfCs are closed (currently scheduled by the end of September), the process isn't locked down and some adaptation may be required. (The introduction of an electoral commission RfC is a prominent example.) I think there should be some room, nonetheless, to trim down the current interval of two to three weeks between the end of the electoral commission RfC and the start of the nomination period. isaacl (talk) 03:56, 28 December 2023 (UTC)[reply]

@Izno at some point the gain in additional voters is not worth the additional time to administer the election. Given the surge in last day voting, the loss from 14 to 10 is would be far less than the 10% drop in voters (which is an even smaller % of the overall pool of people eligible to vote). But by your logic we should really be extending out to longer than 14 days because we would capture even more voters so why not, as is proposed below, start the whole process earlier so we could have a 30 day voting period? Barkeep49 (talk) 05:01, 28 December 2023 (UTC)[reply]

I mean, if we want to take your argument to its logical extreme, we can shorten it much more also - let's say 3 days, just to line it up at about a week less than yours.... Two weeks is a pleasant number for any of a variety of reasons while 10 days is not. As for your "we lose 10% of voters", that argument is folly on its face, which you brushed off with a "you didn't understand my argument" in the context of the RFC. You need to explain why you think "we lose 10% of voters" is an acceptable attrition - and you haven't. As for your suggested month, we do have to end it at some point otherwise we can't move onward, and clearly our ending date is too late in the year already to ensure prompt attention from scrutineers (having which has been the subject of some discussion also) as it is. Meanwhile, we have apparent dead space the most useful point of which is only possibly having to deal with changes to the election. (And I would argue that any actually significant changes will need much more than the month or so as currently scheduled from the closing of ACERFC to the transclusion of candidacies.)

As I've said elsewhere, we need a week, maybe two at most, to get people chugging along on the committee. There isn't a large win and a known and obvious loss from stealing from the election time (of your suggested half a week) compared to stealing the week and a half (or more) of currently-dead time. Izno (talk) 05:12, 28 December 2023 (UTC)[reply]

I agree for this problem that the electcom pause is the lowest hanging fruit, Barkeep49 (talk) 15:25, 28 December 2023 (UTC)[reply]

I don't buy all of your arguments, but some of them make sense - especially the "scrutineering may take a while" part - and moving it up seems reasonable. — xaosflux ^Talk 00:55, 28 December 2023 (UTC)[reply]

Has anyone thought about just moving the RfC start earlier instead of stealing space from one part of the process to make room for another? * Pppery * _{it has begun...} 03:13, 28 December 2023 (UTC)[reply]

I have previously suggested that the community should be encouraged to submit RfC questions early, so it can decide how to best pose them—it might want to break them down into a sequence of themes, for example, and hold RfCs consecutively where the outcome of one might affect the other. But this requires people to be interested and involved earlier in the calendar year. I suspect there isn't much wiggle room to start the RfC earlier and still get good participation. In the northern hemisphere, there tends to be a lot of people on vacation in August. isaacl (talk) 04:05, 28 December 2023 (UTC)[reply]

If we could make it so that questions had to have been discussed first on this page for at least a week, including specific wording, before being added to ACERFC, I would support that in a heartbeat.... I don't think it would necessarily be popular, but that's a change I'd make which is irrelevant to the question of timing. Potentially, a limit in the RFC proper to when questions might be asked, say, in the first week or two weeks, with all others shuffled off to the talk page for next year (if they are serious suggestions that must be answered, and ones which are new that haven't been asked before - I'm real tired of repeatedly asking questions about % of voters in support). I have been frustrated when proposals get thrown up late in the process. But this is rules making for rules making at best to a lot of people, so that's the other group of people who might oppose such a requirement. Izno (talk) 05:22, 28 December 2023 (UTC)[reply]

I've previously raised the question on the RfC talk page (prior to the RfC starting) about having a deadline for proposals (with the caveat that ignore-all-rules situations can still arise). Personally I think that since the community has all year to think of proposals, and the elections have been running smoothly for years, there's little risk of a showstopping proposal coming to light in the last two weeks of the RfC. However there was no support for a deadline. isaacl (talk) 06:01, 28 December 2023 (UTC)[reply]

Before we change the election timeline, we should first ask @Martin Urbanec, Mykola7, and Superpes15: whether starting earlier would have resulted in the results being available earlier? Thryduulf (talk) 18:41, 28 December 2023 (UTC)[reply]