Wikipedia:Requests for adminship/2024 review/Phase I

2024 Requests for adminship review

Status as of 14:43 (UTC), Monday, 4 November 2024 ()

Note: this RfC was created with several already existing discussions that were initiated at the village pump; the section headings for proposals 1–4 have been edited for consistency. theleekycauldron (talk • she/her) 08:15, 20 February 2024 (UTC)

Proposal 1: Impose community sanctions on RfA

The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

---

Should the requests for adminship process be placed under community sanctions (GS)? theleekycauldron (talk • she/her) 20:05, 14 February 2024 (UTC)

• The structure of this GS area would be modeled after the contentious topics procedure. It covers all participation in the requests for adminship (RfA) process, broadly construed. It does not cover mainspace editing about the request for adminship process, nor does it cover discussion of the same.
• Uninvolved administrators and bureaucrats are held responsible for enforcing civility and all other behavioral policies and guidelines through all remedies available under the contentious topics model, as well as the ability to strike or remove comments (up to and include the vote itself) that are judged to be disruptive, corrosive, or otherwise in breach of conduct policy. They are not empowered under this topic area to institute "enforced BRD" or "consensus required" restrictions on a page, or to institute revert restrictions on a single editor. No part of this GS can be made to empower administrators or bureaucrats to enforce any policy they are not already tasked with upholding.
• Requests for enforcement should be placed at the arbitration enforcement noticeboard (AE). An enforcement action can be overturned on appeal only with: a clear uninvolved editors at the administrators' noticeboard (AN); a clear consensus of uninvolved admins at AE; or (in more extreme cases) a clear consensus of uninvolved bureaucrats at the bureaucrats' noticeboard (BN). Standards of review for an appeal to uninvolved editors or administrators are listed here, and the same for uninvolved bureaucrats are listed here.
• Wikipedia:General sanctions/Requests for adminship is created as the log of enforcement actions and list of procedures. The appropriate awareness and sanction templates can also be created.

Survey (proposal 1)

• Support: Community sanctions are intended for topic areas where standard enforcement procedures are not enough to maintain decorum on the project, and, well, RfA and its chronic hostility problem certainly meet that definition. Not once, but twice in the past month or so have we had a situation where the bureaucrats failed to act on serious and baseless aspersions cast about a candidate in front of hundreds of other editors.
  This, despite the fact that as far back as WP:RFA2015 (nine years ago!), we as a community came to the conclusion that incivility and battleground behavior at RfA is out of control and tasked the bureaucrats with improving it. But unfortunately, bureaucrat enforcement of civility norms has been scattershot and ineffective, and nearly all bureaucrats avoid the task entirely. We tried to fix this again in 2021, but no substantive changes were made. Thanks to a systematic failure of enforcement, RfA remains toxic and discouraging. This is a chance to do real good for future candidates who don't want to get hazed and future !voters who don't want to get badgered.
  There is not going to be a perfect solution to all of RfA that waltzes down the aisle in a month or two if we procrastinate on this problem again – it's going to be another three to six years before we get another opportunity, and in the meantime, another crop of qualified candidates will be beaten down, discouraged, or scared off by the RfA process, not to mention the drama and time sink for RfA participants. It is time to fix what we all have already agreed, many times, is broken. I support, and implore the community to do the same. theleekycauldron (talk • she/her) 20:05, 14 February 2024 (UTC)
• Oppose. I'm sympathetic to the intent here, I really am. But I think this proposal will make things worse instead of better. Sometimes editors take controversial positions in RfA discussions. Not too long ago, I did, in fact, and one admin left a message at my user talk, calling my comment "shitty". It would be too easy to end up blocking someone for not having a rationale that was "good enough". We already allow administrators to enact sanctions for violations of the WP:CIVIL policy, and the fact that there have been a few recent examples when no admin elected to do so does not mean that we need to enact a new category of community sanctions. Wikipedia just isn't very good at dealing with civility, unless it's very black-and-white. RfA is no more toxic than ANI (yeah, that's a ridiculously low bar!), so successful RfA candidates will just have to learn to deal with it. --Tryptofish (talk) 20:40, 14 February 2024 (UTC)

  @Tryptofish: If civility is something the admins are already empowered to enforce, and the assertion is that admins being empowered to enforce civility would have them tend to abuse that power, why isn't that a trend we see already at RfA? The opposite is true: admins aren't using their current ability to enforce civility enough, maybe in part because of all the process questions around it. It's not the case that GS and CTOP have a reputation for admins who go around blocking people simply for having comments that aren't good enough, and I don't think RfA would be an exception to that – I think admins would be more cautious at RfA than they would be at ARBPIA or another similarly contentious topic area. theleekycauldron (talk • she/her) 21:02, 14 February 2024 (UTC)

  I didn't mean to imply that I thought that admins would intentionally engage in any sort of abuse of the admin policy. Sorry if it sounded like I did. --Tryptofish (talk) 21:27, 14 February 2024 (UTC)

• Oppose For most editors, that designation means that it's some type of a dangerous minefield and to not edit there. Not good to turn RFA into such. Having 1 or 2 outlandish posts isn't the thing that makes RFA something that most good potential admins don't want to go through. RFA needs some other changes. Sincerely, North8000 (talk) 21:18, 14 February 2024 (UTC)

  that designation means that it's some type of a dangerous minefield and to not edit there As I understand it, for a very, very long time the community has considered RfA to be a toxic and broken system. I know several candidates who have described the process as one of the most stressful things in their life, in no small part because the toxicity of some RfA participants. The only thing we as a community haven't been able to agree upon is how to reform it and address these issues. RfA is already a dangerous minefield, the only difference in this proposal is that maybe we'd acknowledge that in a way that would see some improvement in the process' moderation (or lack thereof). Sideswipe9th (talk) 22:58, 14 February 2024 (UTC)

• Oppose per wot Tryptofish and North800 just said. I just don't think this particular enforcement style is going to be compatible with the way RFAs work. Something is required, but GS is probably not it. Sohom (talk) 21:25, 14 February 2024 (UTC)
• Oppose - This proposal is certainly well-intentioned, but it ultimately will just add more pointless bureaucracy. Honestly, all we need is for the bureaucrats to deal with the occasional bullshit accusation. Alternatively, an uninvolved administrator should be allowed to if the bureaucrats won't. Cheers! Reaper Eternal (talk) 21:57, 14 February 2024 (UTC)
• Support. If this were an encyclopedic subject, it would have been designated a CTOP eons ago. Leeky perfectly explains why this might be beneficial, so I am going to take a look at the three reasons (that I can think of, at least) one might oppose this.
  First, one might argue that there is no problem at RfA. I'm not going to say anything other than, really?
  Second, one might say there is a problem but GS won't solve it. At the very least, I think we should try this. We will never know if we don't try. But I would also argue that CTOP usually works to ease the editing environment. I think one of the strongest endorsement of CTOP that I can give is to mention we haven't needed AP3 even with the Trump era of US politics.
  The final one I can think of is that it is undemocratic: we shouldn't give admins the ability to deny people the right to !vote for/against a candidate. First, WP:NOTDEMO. Second, we choose admins for their good judgement: I don't think they will abuse GS to block a !voter for having The Wrong Opinion. Even if one did, that is not the real problem; the real problem is that such a person is an admin. But the purpose of RfA is to produce suitable admins to help Wikipedia function, not to allow the community to judge a contributor. If you can't respectfully disagree about abortion, you get tbanned, no matter your stance on the topic. If you can't respectfully disagree about climate change, you get tbanned, no matter your stance on the topic. And so on. It is perfectly possible to oppose a candidate without attacking them. If you can't do so, you shouldn't be allowed to participate. HouseBlaster (talk · he/him) 22:39, 14 February 2024 (UTC)
• Oppose I'm sure this is well-intentioned, but it will probably do more harm than good. I do not believe RfA is actually as bad as it used to be, nor do I think the current problems are beyond the capacity of our admins to handle in a normal fashion. In practice, enacting GS would make it less safe to oppose a popular candidate, and I don't think that would be a good thing. I've seen a number of proposed solutions for RfA over the years, and I have come to the conclusion that RfA, to the extent that it is broken, is simply unfixable. LEPRICAVARK (talk) 22:52, 14 February 2024 (UTC)
• I don't know if I feel strongly enough about this to oppose/support, but I can't really see this being helpful, and I think it's more likely to have negative unintended consequences. The problem isn't that no one can enforce civility on RFAs; the problem is that the rules we already have aren't being enforced to the degree that some (most?) people would like. That problem can easily continue under GS. Meanwhile, it seems pretty likely to me that with GS involved fewer people would feel comfortable opposing, whether there's any real risk there or not. -- asilvering (talk) 22:53, 14 February 2024 (UTC)
• Oppose Like others, I believe this proposal was made in good faith. I think the simpler solution is that when these unfounded Oppose !voters or serial opposers comment, we just ignore them or follow WP:TYFYV. RfA is way less toxic than it used to be, and bad opposers are much less common than in the "too many admins"/"self nominations are prima facie evidence of power hunger" days. And yet, each bad opposer gets way more of a reaction than they used to. The better way is to trust Crats to strike or discount the !vote, which is the job we entrust them with. Having one or two opposers with dumb reasons isn't going to tank an RfA. The Wordsmith^{Talk to me} 23:11, 14 February 2024 (UTC)
• Support This strikes me as a good idea. For a very long time, the community as considered RfA to be a toxic and broken process. I know several candidates who have described the process as one of the most stressful things to go through in their life. As someone who primarily edits in one of our CTOP areas, I know from experience that the CTOP procedures far from perfect. However they do work in addressing some of the baseline problematic issues that frequently come up when writing content about controversial topics. Now this isn't going to fix all of RfA's problems, at the end of the day people are still going to be people, but it will help prevent some of the baseline problems that contribute to RfA toxicity.
  In recent months, there have been at least three RfAs that I'm aware of where the crats have been asked to intervene in relation to problematic contributions from editors. Leeky mentions two of these in her !vote, the other I'm aware of was at Clovermoss' Rfa (BN discussion). I'm sure if I went looking, I could find other recent examples of this. Now I understand the catch-22 situation the crats find themselves in when getting requests like this. If they start taking moderation actions when requested, some of them feel as they would start being perceived as non-neutral at close time or cratchat. And there is an argument that maybe if we had more crats (we currently have only 19), some of them might feel more free to take clerking actions where necessary. However this proposal sidesteps that issue, by empowering non-crat admins to act in the exact situations that the crats are currently hesitant to act upon.
  To the folks who are saying that this would make it, as Lepricavark said less safe to oppose a popular candidate, I respectfully disagree. If you want to oppose a popular candidate, you will still be able to oppose a popular candidate. If you have a good faith reason to oppose a candidate, and if you can evidence that and present it in a way that complies with the relevant conduct policies, I really don't think you have anything to worry about. If you feel a candidate is fundamentally unfit to be an admin, you will still be able to put forward a case to try and convince the community of this. What you won't be able to do however is make contributions that would almost instantly be reverted or redacted in any other venue, or be able to flagrantly contravene our civility policy and our policy against personal attacks.
  So yes, I fully support this proposal, and I would strongly urge other editors to do the same. Sideswipe9th (talk) 23:41, 14 February 2024 (UTC)
• Oppose as adding bureaucracy to something that can already be dealt with by reverting trolling, personal attacks or excessive rudeness. RFA is the place to be critical of a candidate, and extra rules should not be inhibiting that. Graeme Bartlett (talk) 23:45, 14 February 2024 (UTC)
• Oppose I'm all for being nice but a candidate should be able to tolerate a couple of trolls. The problem is the attention they get and GS would just encourage more civil off-topic bickering. Johnuniq (talk) 23:54, 14 February 2024 (UTC)
• Oppose. The standard conduct enforcement procedures don't seem to be inadequate here, rather they just aren't used as much as they could/should be. RfA is a regular projectspace page, and the comments made there should be held to the same standard as they would on any projectspace page; personal attacks should not be tolerated. However, the lack of enforcement is not a failing of it, and I'd prefer we try to be more proactive with standard enforcement before adding bureaucracy. Giraffer (talk) 00:23, 15 February 2024 (UTC)
• Oppose - More bureaucratic creep. Carrite (talk) 02:27, 15 February 2024 (UTC)
• Oppose. Administrators and bureaucrats (in their admin capacity) already have the ability to block someone if they are being particularly uncivil. Adding these specific proposed general sanctions doesn't enable admins to do anything new; it merely adds a level of bureaucratic friction should any block be challenged. But if the problem is that people aren't doing anything here, then merely adding more bureaucratic friction isn't a solution—we should try actually using our existing policies and guidelines before we declare that we need something extraordinary here. And if nobody is willing to enforce our ordinary policies and guidelines, then we should elect someone who will. — Red-tailed hawk _(nest) 02:31, 15 February 2024 (UTC)
• Oppose If I'm understanding the proposal correctly, it would enable any individual administrator to unilaterally decide that someone should be indefinitely topic bannned from participating in all future RfAs. I don't like that – it's overreaching and if that kind of outcome is needed, it should result from a community discussion. DanCherek (talk) 02:45, 15 February 2024 (UTC)

The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

Discussion (proposal 1)

• wrt [i]t does not cover mainspace editing about the request for adminship process – is RfA a topic that should even be discussed in mainspace in the first place? M Imtiaz (talk · contribs) 20:35, 14 February 2024 (UTC)
  • Well, I will say that the mainspace page Wikipedia administrators#Requests for adminship currently exists and covers exactly what it says on the tin. theleekycauldron (talk • she/her) 20:38, 14 February 2024 (UTC)
• Whilst I agree that RfA needs more guidance as to what can and can't be done - is GS the way to go? For instance, how would we notify users such sanctions are in place? Presumably this wouldn't include editing restrictions like on most other GS pages (such as 30/500). Lee Vilenski ^{(talk • contribs)} 20:40, 14 February 2024 (UTC)

  @Lee Vilenski: GSes that do 30/500 automatically often aren't full discretionary sanctions procedures like WP:GS/UYGHUR or this one. For this, no, I don't think we'd have any kind of need for an automatic 30/500. As to notification, that's something we take care of without much problem for other GSes and CTOPs – maybe we'd even consider an editnotice warning or other hatnote sufficient. theleekycauldron (talk • she/her) 20:48, 14 February 2024 (UTC)

  To clarify, general sanctions is an umbrella term for ... sanctions that apply to all editors working in a particular topic area. Examples include a one-revert rule, an extended-confirmed restriction, community authorization for discretionary sanctions, and arbitration committee designation of contentious topics. If I understand your proposal correctly, you're proposing a new set of rules for Wikipedia:Requests for adminship and its talk page (and possibly some other associated pages). This would a be new restriction that falls in the category of general sanctions. isaacl (talk) 23:36, 14 February 2024 (UTC)

• If the proposed set of rules explicitly disallows admins from enacting any remedies that they aren't already empowered to enact via existing policy and guidelines, then I think the comparison to contentious topics isn't quite apt. It's true that the contentious topics process allows admins to enact a remedy which cannot be reversed until there is a consensus to do so, which is similar to this proposal. However it also provides admins with a standard set of additional remedies. I suggest leaving out the part about modelling the proposed process after contentious topics. isaacl (talk) 23:47, 14 February 2024 (UTC)
• Is this yet another knee-jerk reaction to a recent thing? I see no discussion on long-term behavioral problems plainly evident in most RFAs. It's been enough to make regular RFA participation not worth my time. RadioKAOS / Talk to me, Billy / Transmissions 07:06, 15 February 2024 (UTC)

  I must have erred when I put the most recent examples up top, because no, this is about the systemic problem, not a collection of isolated incidents, RadioKAOS. theleekycauldron (talk • she/her) 07:09, 15 February 2024 (UTC)

• I've thought about this some more, and I thought of something that I would have added to my comment in the now-closed discussion, so I'll say it here, instead. The discussions here grow out of a perception that RfA is an unpleasant process that discourages good candidates from coming forward. I share that perception, and I agree that it's a problem that would be good to solve. But the problem isn't with plainly inappropriate reasons for opposes, or personal attacks. Those things don't sink RfAs that should have been successful. They get refuted, or ignored, and the process works, albeit with a lot of users justifiably rolling their eyes at the refuted or inconsequential stuff. The problem, fundamentally, isn't even badgering of opposes or inappropriate questions to candidates, even though those things really are problems. None of these things are likely to discourage a qualified candidate from coming forward. No, the problem is that little things, like a bad day or a one-off mistake, get blown up out of proportion. A strong candidate is found to have, just once or twice, posted something that was a bad judgment call, and all of a sudden, there is a flood of "Oppose, per the problem cited by [name]." A lot of good editors know perfectly well, that if they became candidates, this would happen to them, and decide that it wouldn't be worth the trouble. That, not any of the other things, is the real reason good candidates decide not to have an RfA. And none of the proposals being discussed would fix that problem. You cannot sanction an editor for opposing, with a diff or two that indicates what that editor thinks, in all sincerity, is a problem. And you cannot sanction other editors for agreeing with it. But that's the thing that makes RfA an unpleasant place. If editors really want to fix that, there's no easy fix, but the best way to start is to make a practice of thinking carefully before registering a pile-on oppose, and to explain support reasons clearly, including where appropriate, why an oppose does not demonstrate unsuitability. --Tryptofish (talk) 18:23, 15 February 2024 (UTC)

  Having a two-phase approach with a discussion phase and an anonymous voting phase (such as the proposal made during the 2021 review of the RfA process) would help mitigate the morale-draining effect that opposing votes can have on candidates. The rationales would still get discussed, but they wouldn't get repeated by each person opposing (which currently can lead to multiple threads about the same concern). isaacl (talk) 18:39, 15 February 2024 (UTC)

Proposal 2: Add a reminder of civility norms at RfA

This is not a formal change in rules (and I do not see it as mutually exclusive with the above), but an encouragement for admins to use the tools available to them. Add something like this to WP:RFA (i.e. literally add it to the RfA page; wordsmithing more than welcome):

Editors are reminded that the policies on civility and personal attacks apply at RfA. Editors may not make allegations of improper conduct without evidence.

Uninvolved administrators and bureaucrats are encouraged to enforce conduct policies and guidelines, including—when necessary—with blocks.

Survey (proposal 2)

• Support as proposer. If we are not ready for GS, I think we should have a formal acknowledgement (importantly, one placed there by community consensus) that editors should be respectful, and disagree without being disagreeable. I will also note that if an admin uses the tools to silence civil comments that would be egregious WP:TOOLMISUSE; I have much more confidence in our ability to deal with tool abuse than with civility RfA. (Note that this is intended to apply equally to all participants, including opposers as well as those who badger opposers.) HouseBlaster (talk · he/him) 02:19, 15 February 2024 (UTC)
• Support. This strikes me as a good idea. While a lot of the stress of RfA is caused by things other than outright incivility, a simple encouragement is unlikely to hurt. Eluchil404 (talk) 02:46, 15 February 2024 (UTC)
• Oppose I want to know (a) if someone has some dirt; and (b) how the candidate reacts to trolling. Re (a): obviously any aspersions would be closely examined and dealt with appropriately if unsubstantiated. We can't put a header on every noticeboard with this information. Johnuniq (talk) 03:04, 15 February 2024 (UTC)

  With all due respect, are you really arguing that trolling is a good thing at RfA because we can see how the candidate reacts? HouseBlaster (talk · he/him) 05:21, 15 February 2024 (UTC)

  I haven't been following the details of recent RfAs but I have looked and the trolling I've seen has been very minor. The current fuss is not trolling in the sense of something outrageous: it's one oppose that might be misguided but does not warrant all the discussion IMHO. Are there any links showing trolling that really should be removed but hasn't been? Johnuniq (talk) 06:12, 15 February 2024 (UTC)

  Off the top of my head, since December we have one RfA where more than 50 revisions had to be suppressed because it took 48 hours for a wildly inappropriate comment to be removed and one neutral which literally cited the candidate's religion as a reason they would be uncomfortable if the candidate got the mop (see also Q6 and Q7 by the same editor). HouseBlaster (talk · he/him) 12:53, 15 February 2024 (UTC)

• Oppose Johnuniq makes a good point. Until the FRA vote is private we will probably have this handwringing. Lightburst (talk) 04:55, 15 February 2024 (UTC)
• Support. The opposes above suggesting that we shouldn't discourage people from leaving personal attacks are disappointing. 0xDeadbeef→∞ (talk to me) 06:57, 15 February 2024 (UTC)
• Support: I would've thought this would be common sense, but apparently people need to be reminded. — Callitropsis🌲[talk · contribs] 07:02, 15 February 2024 (UTC)
• Oppose a lot of these concerns could be alleviated if someone would just p-block Lightburst from the RfA. We don't need to make community-wide statements in response to poor behavior from one editor (I realize there were two editors misbehaving at the current RfA, but the one !vote was already struck as a crat action, so evidently our current approach is not completely broken). LEPRICAVARK (talk) 07:05, 15 February 2024 (UTC)

  Plenty of other editors have recently engaged in problematic conduct at RfA. It's a systemic issue and admins and 'crats are way too reluctant to enforce community norms at RfA in my opinion. I'm not optimistic that this notice will solve all of RfA's problems, but hopefully it'll serve as a reminder to everyone that personal attacks and aspersions are just as unacceptable there as they are everywhere else and should be met with appropriate warnings or sanctions. — Callitropsis🌲[talk · contribs] 07:19, 15 February 2024 (UTC)

@Lepricavark: it is true that if you block the minority voters or those who defend them, you can get a 100% result. In my experience, whoever dares to oppose the majority gets in trouble at RFA. The process is broken because it is public. Imagine going to the polls in the US, and they say who are you voting for? You tell them and they ask why? Now imagine having to declare your reasons to everyone in the polling center. And if they disagree with your reasons they strike your vote. If part of the Homeostasis07 rationale was an aspersion, strike it [[WP:RPA]]. But that is not what happened, first they poked at it, then they became outraged. Then they moved the entire rationale to the talk page. Then someone erased the vote entirely. I am just trying to stick up for the minority even though I voted with the majority. I guess FRAM struck my vote as well and that did not surprise me. Lightburst (talk) 14:45, 15 February 2024 (UTC)

If you are going to ping me, then please have the courtesy to write something that it is worth my time to read. No, the process is not broken because it is public. The process is broken because of editors, such as yourself in this and other instances, who provoke needless disruption and drama instead of evaluating the candidate in a policy-compliant manner. To address one of your red herrings, this is not about getting a 100% result. The first oppose was removed because, despite your seeming wishes to the contrary, RfA is not a safe haven for innuendos against the candidate. Your !vote should be struck because it is admittedly irrelevant to the candidate. If someone were to subsequently draft a relevant oppose that did not violate basic policies, it should and would be allowed to stand. LEPRICAVARK (talk) 17:10, 15 February 2024 (UTC)

In light of the community's failure to do anything about the latter two pointy opposes even though it was blindingly obvious that they were pointy, perhaps this is actually a good idea. Therefore, I change my position to support. LEPRICAVARK (talk) 00:45, 18 February 2024 (UTC)

• Support :) I think that, one way or another, what admins and 'crats have lacked is a clear mandate. This seems like a good first step towards a more healthy community. (Maybe another good first step would be disabusing ourself of the notion that a little trolling is good for the soul. A world in which no troll ever enters Wikipedia again is not a world where everyone on it grows soft, there's a real world too and people learn how to deal with hardship.) theleekycauldron (talk • she/her) 07:19, 15 February 2024 (UTC)
• Support as plain common sense. Those saying "but but but but trolling an RfA lets us show how a candidate reacts!" miss the point that it also discourages candidates from standing in the first place. Ritchie333 ^{(talk) (cont)} 11:27, 15 February 2024 (UTC)
• Support as a reminder of our existing conventions and rules rather than extra bureaucracy. RfA is not a welcoming environment, and I'm sure it deters many potential candidates who would make good admins. Certes (talk) 11:44, 15 February 2024 (UTC)
• Support - The fact that we have to even vote on this and that it's not unanimous really is quite telling of how differently RfA is viewed across Wikipedia. But we need more civility, please. Not less. Particularly at a venue such as RfA. Believe it or not, we'd still like to encourage people to become admins. _{Duly signed,} ⛵ WaltClipper -(talk) 13:33, 15 February 2024 (UTC)
• Support - Proven as necessary, even if it is redundant. Certes explains it well. — Ixtal ^{( T / C )} ⁂ Non nobis solum. 13:48, 15 February 2024 (UTC)
• Moral support, but... I think we all know that we could already be doing this, admin or not. The problem with 'moderating' Wikipedia, compared to other places, is is that everything is out in the open and that everyone is on a level footing. Elsewhere, a mod sees a problematic comment and they remove it. Maybe the author can appeal, but it's going to be handled by other mods, and there certainly won't be a peanut gallery. If you remove a comment at RfA (or anywhere really), the author is very likely to publicly kick up a fuss, and that fuss is going to attract others to weigh in on both sides. There's a not-insigificant chance that, even if you're really sure that the comment was out of line, it's going to escalate to the point where you end up at ANI or ArbCom or wherever. I think admins are more aware of this than most, which is why you don't see us intervening unless the conduct is very obviously very bad. Encouragement or not. – Joe (talk) 14:05, 15 February 2024 (UTC)
• Oppose I know we're all trying to make RfA better, and I'm not for personal attacks. I'm just worried this will make it harder to oppose candidates who need to be opposed, since the "casting aspersions" line seems like it could cause problems if someone opposes an RfA without presenting evidence, and as Joe Roe notes, who is the arbiter here? For instance is opposing someone on religious grounds a personal attack or aspersions, or does it just say a lot about the person opposing? I think a better idea might be for there to be explicit moderators for RfAs instead of putting up a notice of something we all aspire to, but is in a situation where it's difficult to enforce. SportingFlyer T·C 14:29, 15 February 2024 (UTC)

  I'm really not seeing the connection between putting this notice and make it harder to oppose candidates who need to be opposed. If someone says I had off-wiki encounters with the user that do not make me believe they should be an admin or Some off-wiki observations that I cannot show have made me lost my trust in this candidate they would not be casting aspersions. People would definitely be curious at what it is, but if you can't provide any proof or evidence for your claim, you can't really specify anything other than how it made you feel without making an unsubstantiated claim.

  So my question is, if my understanding of what it means to cast aspersions is correct, in what way do we benefit from giving consideration to people making unsubstantiated claims where no evidence were sent to anyone? 0xDeadbeef→∞ (talk to me) 14:38, 15 February 2024 (UTC)

  Doesn't your question confirm my concern, though? If I say something like "I've interacted with RfA candidate in the past and over the course of numerous encounters don't feel they have the capacity to be an administrator," is that a validly held opinion/oppose, is that an aspersion, is that an unsubstantiated claim that should be disregarded? I think it's the former, but don't want it to be confused with an aspersion. SportingFlyer T·C 20:39, 15 February 2024 (UTC)

• Support. This issue is not merely something that happened recently. Repeated violations of WP:NPA and WP:CIVIL in RFAs have been problematic for a while - off the top of my head, I can name at least three users who were specifically topic banned from RFA in the last several years due to civility issues. If someone has a legitimate reason to oppose an RFA, there is no reason why it cannot be done civilly. Epicgenius (talk) 15:04, 15 February 2024 (UTC)
• Support I'm ok with most of this. I don't think we need Editors may not make allegations of improper conduct without evidence. I can only see people using this to stop people from having votes and asking for evidence. Otherwise, it's a big win to explicitly remind people to be nice at RfA. Lee Vilenski ^{(talk • contribs)} 15:10, 15 February 2024 (UTC)
• Support There's no way to prove it would help, but it certainly wouldn't hurt. Would it make sense to add something about how the 'crats are empowered to clerk? Sincerely, Novo Tape^{My Talk Page} 17:16, 15 February 2024 (UTC)
• Oppose. As I mentioned in my oppose comment in the now-withdrawn proposal, I got a lot of heat for a comment I made in a fairly recent RfA. I put myself at neutral, and based it on something that I said I did not feel comfortable posting in public. I also said that I did not think my reasons were solid enough for me to oppose, which is why I chose to be neutral. The problem I have with this new proposal is that it seems to say that, without evidence, what I did would have been block-worthy. --Tryptofish (talk) 18:06, 15 February 2024 (UTC)
  • Please see also my comment in #Discussion (RfA civility), above. This proposal targets conduct that isn't really the problem that prevents good candidates from applying. --Tryptofish (talk) 18:27, 15 February 2024 (UTC)
• Concern - I am concerned about how this will be applied when an editor leaves a good faith but bluntly worded comment at an RFA… situations where Incivility isn’t intended, but may perceived. Blueboar (talk) 18:24, 15 February 2024 (UTC)
  • It would be handled the same way we handle civility issues in the rest of the encyclopedia. That is, we WP:AGF and politely ask that the editor reword their statement. — HouseBlaster (talk · he/him) 19:51, 15 February 2024 (UTC)
    • Don't be so sure. Look at all the badgering of oppose comments that we have now. --Tryptofish (talk) 20:46, 15 February 2024 (UTC)
      • This proposal would apply to oppose-badgering, too (see WP:SAUCE). — HouseBlaster (talk · he/him) 13:58, 16 February 2024 (UTC)

    There is no one single way that we handle civility issues in the rest of the encyclopedia. It varies wildly depending on the particulars of any given situation. LEPRICAVARK (talk) 23:31, 16 February 2024 (UTC)

• Support I will continue to support any little steps that will improve RFA until we get an RFA that isn't broken. If there's sufficiently detailed bigger steps, I will happily support them too. The community both needs a reminder of our civility policies and stronger enforcement of them in RFAs. Soni (talk) 06:01, 16 February 2024 (UTC)
• Support (and yes, I know). A reminder never hurt. Plenty to gain (editors like Homeostasis07 will in future know they will be blocked for trolling before in advance, and admins will be reminded that their normal tools can still be used alongside any specific crat actions that may also be taken, and crats will hopefully be encouraged to keep a gimlet eye on proceedings safe in the knowledge that they have codified backing), and nothing to lose, except some of the toxicity which is an inevitable by-product of treating RfA where NPA, ASPERSIONs, etc, does not run (as a worrying number of admins and even crats seem to believe). ——Serial 19:55, 17 February 2024 (UTC)
• Oppose As long as RFA comments are coupled with voting, I will have to oppose anything that could be used as a restriction on how people vote. Anyone should be able to vote how they want for any reason bar outright trolling. Now if voting and comments became uncoupled (see #Anonymous voting) and not everyone was forced to give a rationale, it would be a good idea. Pinguinn 🐧 19:56, 17 February 2024 (UTC)
• Support RfA should at least have the same standards of user conduct as any other area of Wikipedia (not that our standards are that high). Maybe with an addition that explicitly states that votes that are aspersions/personal attacks can be struck/removed, since I think the often the issue is people being reluctant to enforce policies against oppose voters since they don't want to impugn people's "right to suffrage". Galobtter (talk) 06:52, 19 February 2024 (UTC)
• Weak Support If there are admins and crats who feel as though they will be willing to enforce CIV and NPA, then this is a good idea. I'd consider it a pre-warning that editors should not make the type of comments they cannot make anywhere else on the project without sanction or summary removal. However if there aren't admins or crats willing to enforce those policies, then at best this is a fig-leaf sign that some can point to when telling others that their contributions are unacceptable per policy, and at worst this will just be some boilerplate text that editors ignore while violating the relevant conduct policies. Sideswipe9th (talk) 22:56, 19 February 2024 (UTC)
• Strong oppose with apologies to HouseBlaster because I'd love to believe this would make a difference. One way it won't is regarding disruptive editors because as we've seen recently, they will never back down, not one inch, from their assertion that their behavior is completely fine, they only went on the attack with great reluctance because of something someone else did, or whatever.
  If this has a positive effect, it'll be because it motivated editors who can actually enforce these policies to finally end their infuriatingly hands-off approach to disruption at RFA. And that won't happen, either. On the most recent request for adminship's talk page, two bureaucrats, User:Primefac and User:28bytes, said that they wouldn't remove two oppose votes that were blatant policy violations from the final account because the RFA was going to easily pass and besides, the community can rest assured that 'crats don't take such votes into account. That, to me, is incredible: two bureaucrats are on the record saying they flat-out won't enforce policy at RFA. (Everybody knows they didn't affect the vote! Everybody knows you wouldn't take them seriously! Toss them anyways!) No administrator or bureaucrat needs to be told that they're "encouraged to enforce conduct policies and guidelines, including—when necessary—with blocks." They know. The problem is that they still almost never enforce anything. (Change the word "encouraged" to "required" and I'd switch to strongest support possible.) City of Silver 00:01, 20 February 2024 (UTC)

  I think the reason that admins and bureaucrats are reluctant to enforce the rules at RfA is the fact that it feels like election interference. The reason I think we need a note like this is to make it clear that the community does not see enforcing behavioral policies/guidelines as election interference. (In other words, I think that it will have a positive effect because it motivated editors who can actually enforce these policies to finally end their infuriatingly hands-off approach to disruption at RFA.) HouseBlaster (talk · he/him) 00:29, 20 February 2024 (UTC)

  Yeah, the "election interference" angle is important here. And this becomes especially worse since most admins who see policy-violating behaviour have probably already voted in the RfA and so are involved. Galobtter (talk) 00:31, 20 February 2024 (UTC)

  You both might be right but then the reasons those two 'crats gave for sitting on their hands would be flat-out lies, wouldn't they? And what kind of enforcement can we expect from people who allow policy-violating disruption to stand because doing something about it means they'll get called out by the rulebreakers? City of Silver 04:17, 20 February 2024 (UTC)

  I don't see how those 'crats were lying: it didn't affect the outcome, and had it gone to a 'crat chat they would have been discarded. And I don't think admins are concerned about being called out by rulebreakers (if they were, vandals would never get blocked). I think admins are concerned they will be called out by "the community" for "interfering" with an "election". I see this as the community telling admins that enforcing behavioral guidelines will not be seen as interfering with an election. HouseBlaster (talk · he/him) 05:19, 20 February 2024 (UTC)

  @HouseBlaster: No, no. From the parenthetical in my message: "Everybody knows they didn't affect the vote! Everybody knows you wouldn't take them seriously!" I believe the lies came when those 'crats claimed these facts were why they wouldn't do anything when the real reason was that they were afraid of being accused of election interference. City of Silver 02:44, 21 February 2024 (UTC)

  If you're going to attempt to throw me under a bus, at least avoid grossly misrepresenting what I said (in this discussion for reference). The two !votes in question were POINTY, but were not "blatant policy violations", and as such did not require striking. Primefac (talk) 07:31, 20 February 2024 (UTC)

  With all due respect to you, as you do do the thankless task of clerking, Primefac – how does one square a vote being an example of "disrupting Wikipedia (to make a point)" and not a violation of policy? theleekycauldron (talk • she/her) 07:35, 20 February 2024 (UTC)

  WP:POINT and WP:DE are behavioural guidelines, not policies, so mischaracterising my statement in such a hyperbolic manner is why I felt compelled to comment. Primefac (talk) 10:33, 20 February 2024 (UTC)

  @Primefac: Regarding "The two !votes in question were POINTY, but were not "blatant policy violations", may I say? The two votes in question were lies. No gray area: they were lies. (Both! Voters! Supported! The! RFA! Where! They! Voted! Oppose!) You won't answer but I'll ask anyway: is lying not forbidden by policy? City of Silver 02:38, 21 February 2024 (UTC)

*Support: however, I'd just want Editors are reminded that the policies on civility and personal attacks apply at RfA.

Uninvolved administrators and bureaucrats are encouraged to enforce conduct policies and guidelines, including—when necessary—with blocks.

We don't need to specifically state what civility means, not do I think we should be making claims about people even with evidence. Those claims can be gamed. Lee Vilenski ^{(talk • contribs)} 11:57, 20 February 2024 (UTC)

• @Lee Vilenski: it appears you !voted twice :) HouseBlaster (talk · he/him) 17:42, 20 February 2024 (UTC)

• Support in principle: the wording should be workshopped, and bureaucrats should be encouraged to enforce this more aggressively. Z1720 (talk) 14:41, 20 February 2024 (UTC)
• Support This shouldn't be necessary, but it can't hurt. In my RfA, the problem wasn't specific uncivil comments, but instead it being difficult to watch tens of people each bringing all of your faults to bear, even if they do it reasonably civilly. * Pppery * _{it has begun...} 17:11, 20 February 2024 (UTC)
• Support, because I don't see how it can hurt. My assessment of the problem we're trying to solve here, insofar as it is a problem, is there there's a handful of !voters falling somewhere on the spectrum of contrarian to troll; and another lot of participants who get worked up about these !voters and are just as nasty, or nastier, in response. We can, and should, do more to curb the worst of this behavior with the tools we already have: the rest of it, we need to learn to live with or ignore. Enforcing WP:ASPERSIONS would help, but that's not easy to do on a page that's inherently about discussing another editor. You cannot legislate against contrarianism, much as we would all like to. Vanamonde93 (talk) 17:52, 20 February 2024 (UTC)
• Strong support. We ALL need reminders on civility and how to conduct ourselves with proper decorum on Wikipedia every now and then. In my opinion, this can only help, and all voters--myself included--would be prompted to "check" themselves before casting a vote or asking a question/posting a comment that may come off as caustic. I am with WaltClipper - we need to be encouraging our best editors to seek the mop and view RfA as a process of POSITIVE constructive criticism. Breaking down candidates to the point of withdrawal, regret, a Wikibreak, or any combination thereof is not helpful to this project. — That Coptic Guy^{ping me!} (talk) (contribs) 02:22, 21 February 2024 (UTC)
• Support, as merely a reminder of the rules, not a new rule. Queen of Hearts (talk • stalk • she/they) 03:33, 21 February 2024 (UTC)
• Support my car reminds me to put my seat belt on when I start it. I don't see that as a restriction on my choice of direction. Regards, --Goldsztajn (talk) 07:28, 22 February 2024 (UTC)
• Oppose It's too subjective and vague about what is meant by improper conduct and evidence. Consider some recent opposes such as "I have some concerns about experience and maturity."; "Pretty damning concerns about judgement."; "Insufficient emotional maturity for an admin."; "User takes themselves way too seriously...". No specific evidence was given for these comments but demanding this would result in more toxic drama rather than less. See also leopards. Andrew🐉(talk) 12:55, 22 February 2024 (UTC)

Discussion (proposal 2)

Let's say an editor makes an oppose comment like this:

• Oppose. I don't have any specific incidents in mind, but my interactions with the candidate leave me with the gut feeling that they will be too quick with the block button.

Supporters of the RfA candidate might believe sincerely that this is an aspersion made with the express statement that no specific evidence is going to be presented. It's possible to read the alternative proposal literally, as indicating that the community would authorize an administrator to sanction the editor who made the oppose. And yet it seems to me that this should be an allowable argument to make at an RfA. --Tryptofish (talk) 18:56, 15 February 2024 (UTC)

That's an absolutely reasonable thing to say at RfA. I almost never comment at RfA except when it's somebody I've known for a while because that's the only way to really get to know a person. If somebody who has interacted with the candidate over a period of time has a gut feeling one way or another, that's something I want to know. RoySmith (talk) 20:05, 15 February 2024 (UTC)

Yes @RoySmith: that makes sense to me! Lightburst (talk) 04:33, 16 February 2024 (UTC)

Yes, indeed it is. The alternative proposal says, however, Editors may not make allegations of improper conduct without evidence. I suppose that one could wikilawyer that "they will be too quick with the block button" is speculation about future conduct instead of an allegation of improper conduct that has already happened. And one would hope that admins would be clueful enough to know that the oppose isn't really what this proposal is trying to prevent. But the practical reality is that this proposal would put a chill on such opposes, and could well lead to attempts to game the language to make a mountain out of a molehill. (If anyone doubts that, just consider how some opposes get badgered.) And if admins are clueful enough to know not to block the opposer in this example, then they are clueful enough to deal with real problems without the need for the proposed language. I think this alternative proposal would end up being a net negative. --Tryptofish (talk) 20:45, 15 February 2024 (UTC)

The precise wording can be tweaked. In general, there is a difference between accusing a user of bad behavior in the past, accusing them of bad intentions for the future, and trying to project their future behavior. The first 2 necessarily need evidence, while the third can certainly be "gut feeling". I certainly would be unaffected by this third type of reasoning, but the voter had at least given some indication of why he/she is opposing the candidate. Animal lover |666| 00:05, 21 February 2024 (UTC)

• IMO what's actually wrong with RFA is the whole idea of it. There is no fixing it, because it's a stupid idea from the get-go. It's the same stupid idea as WP:RFC/U was. And for some reason that I don't understand, this community figured out that RFC/U was a stupid idea years ago and got rid of it, but maintains the same damn system for admin rights.

  What's stupid about RFA and RFC/U, fundamentally, is the very idea that we should publicly evaluate one of us. Has anyone reading this ever, in any other aspect of their life, seen anything like this happen? Ever gone to a school where the entire faculty and student body gets together and talks about you? Or had a job where an all-staff meeting is called and the subject of discussion is the performance of an employee? The closest thing I can think of is an intervention (counseling), and even then, that's close family and friends, not hundreds of strangers. (And that's without getting into real world elections, which are done by secret ballot.)

  Why do we do this? Because years ago WMF Legal said "community vetting" had to happen in order to give someone the viewdeleted permission. More recently, at one of the rounds of RFA reform discussion, WMF Legal backed off of that and said they'd consider alternatives. I think the community should tell WMF Legal tough cookies, we are not going to engage in public discussions of the job performance of editors, and WMF Legal is just going to have to figure out some other way. Because public performance reviews of editors are not healthy. It's way too easy, as we have seen time and time again, for such discussions to be derailed or to devolve into accusations, arguments, and so forth.

  You can't control hundreds of people... the more participation, the more likelihood that someone, somehow, will cause a problem. That's why in the real world we don't have public performance reviews in which hundreds of people participate. It's foolish to even attempt such a thing, even moreso on a website with hundreds of strangers, where almost literally anyone can participate. Yet Wikipedia continues the tradition. It's time for everyone to stop pretending that public performance evaluations are a normal thing to do. It's unhealthy. End it. Levivich (talk) 21:09, 15 February 2024 (UTC)

  All that WMF has ever said is that there should be some form of community process. It never specified what form that should take. A proposal to have an Arbcom style election had good support but failed only due to technical issues. Hawkeye7 (discuss) 21:25, 15 February 2024 (UTC)

  I have a draft to reignite that proposal somewhere, but I'm just a bit tired at the moment. theleekycauldron (talk • she/her) 23:14, 15 February 2024 (UTC)

  The supports were close to double the number of opposes, so I'm not sure "failed" is the right term. There isn't a technology-based technical issue at the frequency level that was being discussed. The concern is WMF staffing to support the vote. isaacl (talk) 05:16, 16 February 2024 (UTC)

  That RFC result was very unexpected. In my opinion, it should have been similar to the bot approval process: you get consensus first, then worry about technical details later in a separate step. That rfc should have been the consensus statement. It should have closed as "consensus to try some kind of RFA voting system, with technical details to be determined later". If someone were to rerun that rfc, with the correct closers, I'm pretty sure it would pass. –Novem Linguae (talk) 05:42, 16 February 2024 (UTC)

  +1. Wikipedia:Requests for adminship/2021 review/Proposals#Closed: 8B Admin elections actually passed by a vote of 72-39, nearly 2:1 in favor. Levivich (talk) 18:23, 16 February 2024 (UTC)

  English Wikipedia has a history of being influenced by libertarian views that underlie its adherence to consensus decision-making for almost everything. But consensus doesn't scale up to larger groups; it stalemates action. And yes, on top of that, evaluating people in a large, public group meeting isn't done outside of very specific professions. On a sidenote, RFC/U didn't end because of a consensus view that it was a bad idea to have public evaluation of editors. It ended because the commenters couldn't impose any sanctions through that process, so they thought it was ineffective. WMF Legal is not to blame for this. Those who like to participate in discussions about decision-making are loathe to give up the outsized-influence a small number of people can have to block changes they disagree with. So the community has been unable to agree on following approaches used by other organizations to make decisions more effectively. I understand and appreciate the advantages of consensus decision-making and letting everyone weigh in, but they come with an inherent cost that can't be avoided by tinkering with rules of behaviour. The community will have to shift towards other options for at least some things, which can include voting on certain decisions, delegating primary responsibility for certain tasks to designated subsets of the community, adopting some community hierarchy for interpreting policy, or some other commonly used option in the real world. Alternatively, the community can shrink down drastically back to a level where consensus might be more effective, but that's probably only going to happen if the site is abandoned by most of the existing community (and likely overrun by promotional editing at that point). isaacl (talk) 22:56, 15 February 2024 (UTC)

  @Levivich during WP:RFA2021, legal confirmed what Hawkeye says. There are any number of community processes which can work from their perspective. What wouldn't work is something like we do for ECR where we would give sysop to everyone with at least X years and Y edits or whatever. Best, Barkeep49 (talk) 23:00, 15 February 2024 (UTC)

  Since the discussion has broadened to RfA in general, I'd like to make a general comment that does get back to the proposal here, as well. If I try to think of two things where the community keeps trying to come up with proposals for improvement, and the proposals never get consensus, the top two might very well be civility and RfA. Everyone (including me) agrees that we ought to do better with civility, and we ought to do better with the RfA process. And year after year, proposals are made, and fail. Alas, the proposals here have hit the jackpot, by trying to deal with both civility and RfA. (What could possibly go wrong?) I've also been editing here for long enough that I well remember when the widespread concern in the community related to RfA was how to have a sort-of reverse RfA, where the community could have a mechanism to de-sysop admins who were acting like jerks. Gradually, over time, ArbCom got to be good at dealing with that. So now, the pendulum has swung back the other way, with concerns over RfA being unable to get enough new applicants. As I've said earlier in this discussion, the solution to that problem won't be found in the proposals here. The community just needs to decide not to pile on when reasons for oppose at RfA are not part of a pattern. Editors don't decline to be candidates because someone at the RfA is going to be incivil. It's because real people, including those who would make excellent admins, don't want to get hassled over the one or two times they did something genuinely regrettable. --Tryptofish (talk) 23:31, 15 February 2024 (UTC)

  Can we please not waste more words and time on these proposals? The community in its current state has shown itself that it wants RfA reform but cannot and will likely never agree with itself on how carry out that reform. If the community can't agree to come up with a system that is better than forcing people into public performances in front of hundreds of strangers, then so be it. What should be a cordial and basic exercise of governance can evolve into a poorly moderated jeering mass at a moment's notice. Not to mention the public nature of !votes means that people expressing their opinions are subject to badgering and drama. This discussion is particularly disappointing for me, a poor guy had his religion belittled amidst drama and attention from other editors he did not ask for. If we had an actual secret ballot both candidates and !voters would be spared of ridicule, but obviously nothing will be done because of technical infeasibility or people liking the current system more for their own personal reasons. Better to let them continue their talk of the lack of admins and just move on. Alright, that's enough words from a rando like me wasted on this topic. Time to step away from this and get back to cleaning up spam. The Night Watch (talk) 03:34, 16 February 2024 (UTC)

  This ^ Lightburst (talk) 04:19, 16 February 2024 (UTC)

  I think there are structural reasons for the pile-ons too. Notionally opposes at RfA carry more weight because success requires a 2:1 ratio of supports to opposes. But the reality is the reverse, because unless an RfA fails immediately as WP:NOTNOW, there's usually at least 100 or so supports before the first substantial oppose vote is made. So opposers feel like they have to overcome the momentum of the candidate's friends and those who just vote support for every RfA (as they're perfectly entitled to), leading to lengthy and repetitive rationales, which provokes badgering, which leads to more verbiage and repetition, and so on.

  The only ways I can think of addressing this involve making RfA either a secret ballot and/or having some sort of selections committee. As Levivich says, holding public elections-cum-performance-reviews is just a fundamentally odd thing to do. It produces all sorts of weird dynamics that can't be fixed by exhorting people to just be nice or not pile on or whatever, because individually they're already trying to. – Joe (talk) 08:53, 16 February 2024 (UTC)

  Alternatively, an RfA could start with some period of discussion before voting begins (and this could be combined with one of your other methods, as indeed was proposed in the elections one that almost passed in 2021). Barkeep49 (talk) 17:56, 16 February 2024 (UTC)

  Yeah, I still think an open discussion period followed by a closed ballot is the best model all-round, and still think the 2021 discussion showed a solid consensus to at least try that... – Joe (talk) 21:41, 16 February 2024 (UTC)

  The "community vetting" thing is an extremely low bar, and I think it only precludes things like automatic promotion based on edit count. I am mystified why this WMF comment has been used so successfully to argue for the broken and harmful status quo. —Kusma (talk) 09:25, 16 February 2024 (UTC)

  Yes, exactly. As I noted above (but which might have gotten lost in subsequent discussion) in 2021 WMF Legal commented (in part) The key point, per our previous commentary on the issue is to ensure that the process is one that can make sure that the selected candidates are overall trustworthy and responsible.. Best, Barkeep49 (talk) 19:09, 16 February 2024 (UTC)

  Perhaps it has been used, but offhand Levivich's comment above is the only one I can recall right now, so I don't agree that this concern has been used successfully to support the current RfA process. My recollection is that commenters understood the rationale from WMF Legal, as both Barkeep49 and Hawkeye have expressed, that in order to meet the responsibility of removing legally prohibited content, the privilege of viewing deleted content can't be handed out automatically through criteria that any editor can achieve. Other selection processes still meet the need of choosing "trustworthy and responsible" candidates. isaacl (talk) 19:44, 16 February 2024 (UTC)

  Not really an offhand comment :-) WP:RFA2021 is what I was referring to when I wrote "at one of the rounds of RFA reform discussion, WMF Legal backed off of that." One can read the Brainstorming, Phase 1, and Phase 2 pages to see the discussion surrounding WMF Legal (CTRL+F for "legal," for Phase 2 you have to uncollapse all collapsed boxes), and in Brainstorming is the discussion about BK reaching out to WMF Legal which results in WMF Legal "backing off" as it were, saying they'd be open to RFA alternatives.

  FWIW, there are good reasons not to have any kind auto-admin system (automatically given to all editors upon meeting some threshold like 2 yrs/20k edits), but WMF Legal's objections aren't among them. As I said in my post above, I think the community should devise whatever system works best for the community (discussion, secret vote, even auto-admin if that's what the community wants); any objections by WMF Legal, or technical challenges e.g. SecurePoll limitations, can be overcome. We shouldn't treat (as we have in the past) either an objection from WMF Legal or asserted technical limitations as if they created insurmountable obstacles. The obstacles are very surmountable. We should do what works best -- the technology and the lawyers will conform to the needs of the people creating the content, we don't have to do it the other way around. Levivich (talk) 20:35, 16 February 2024 (UTC)

  By offhand, I meant off the top of my head, I can't recall anyone successfully making the same argument you made. I've been following the discussions about this from long before the 2021 review; there hasn't been a consensus view that WMF Legal only supported keeping the RfA process as it existed at the time it issued its original opinion. But history doesn't matter for moving forward. I agree there seems to be an available consensus for anonymous voting that can be established. isaacl (talk) 22:53, 16 February 2024 (UTC)

  Levivich makes good points but there are repeated examples of such behaviour in history such as public humiliation, mutual criticism, and struggle sessions. On tech platforms, it's now online shaming. Of course, supporters want RfA to be more pleasant and so there's all this pressure to play nice but the purpose of the process requires it to be challenging. Andrew🐉(talk) 13:18, 22 February 2024 (UTC)

How to implement anonymous voting

As a practical matter, how would be implement anonymous voting? The available tool is meta:SecurePoll, but I think the community would find it rather onerous to use. It needs to be configured for each election, has some scheduling limitations, and requires WMF staff to get involved to set it up each time. RoySmith (talk) 23:28, 16 February 2024 (UTC)

I've talked to the WMF T&S team, and they say they'd be willing to do it on a limited scale. We couldn't hold monthlies, but it's doable. theleekycauldron (talk • she/her) 23:35, 16 February 2024 (UTC)

What do you mean monthlies, theleekycauldron? — ♠Ixtal ^{( T / C )} ⁂ Non nobis solum. ♠ 23:42, 16 February 2024 (UTC)

@Ixtal: We couldn't do monthly elections :) theleekycauldron (talk • she/her) 23:47, 16 February 2024 (UTC)

Oh, so we'd be batching RfA into quarterly (or whatever) tranches? I guess that would work, but wouldn't that mess with WP:BATON? RoySmith (talk) 23:42, 16 February 2024 (UTC)

We'd be leaving the old RfA system intact, so i guess the baton would live on for those who choose to do public RfAs. More to the point... I don't think it matters too much. theleekycauldron (talk • she/her) 23:47, 16 February 2024 (UTC)

Traditions are made for the participants, rather than the other way around, so I'm sure the community will find a way to adapt. isaacl (talk) 23:49, 16 February 2024 (UTC)

I'd say it goes in whatever order the 'crats implement the results. HouseBlaster (talk · he/him) 00:35, 18 February 2024 (UTC)

I don't think voters would find it unduly onerous. Staffing resources is a constraint. There is a Phabricator ticket open to track the task of allowing it be administered by local admins, so in the longer term, if the community is able to assume responsibility for configuration, it can ramp up usage. Initially I imagine that elections would be a supplementary path for selecting administrators. The community would be able to learn from the experience and decide how to proceed. isaacl (talk) 23:48, 16 February 2024 (UTC)

My favorite question when considering new and untried things is, "What's the worst that could happen?" In this case, the worst that could happen is we run an election and it goes badly. But we've got that already. So I say let's go for it. RoySmith (talk) 23:56, 16 February 2024 (UTC)

This 1000% — ♠Ixtal ^{( T / C )} ⁂ Non nobis solum. ♠ 00:38, 17 February 2024 (UTC)

"Allow local wikis to set up elections" is phab:T301180. SecurePoll is also very secure. Perhaps overly secure. Steps like scrutineering (checkusering every voter) can probably be dropped from the RFA voting workflow. The entire workflow of SecurePoll should be documented somewhere by someone knowledgeable, and then examined to see if other optimizations can be made to it. For example, is encryption overkill for RFA, and would turning that off help speed things up? etc. We should also consider if we want to institute minimum voting requirements such as extended confirmed, to prevent mass IP or throwaway account edits that could swing the vote. –Novem Linguae (talk) 00:01, 17 February 2024 (UTC)

I've worked up a lot of the details in a draft i've been writing over at Wikipedia:2024 administrative elections proposal, if we wanna kick that around? theleekycauldron (talk • she/her) 00:16, 17 February 2024 (UTC)

As I discussed previously on the Requests for adminship discussion page, votes are encrypted so no one with access to the underlying database (either directly or I suppose via a MediaWiki vulnerability) will be able to determine how people voted. Running an unencrypted election removes a bottleneck in administering the encryption, and would speed up the tallying process. I don't think the tallying process part is a big problem in the overall picture; scrutineering is the most significant delay. isaacl (talk) 00:23, 17 February 2024 (UTC)

• It is still quite an ordeal to run Secure Poll, but would be feasible if we wanted to do quarterly elections - perhaps in addition to having the existing RFA process for on-demand; candidates could choose which they prefer. The "big deal" would be that it is a vote. Something to decide would need to be if there is also an on-wiki "discussion" to go along with the vote or not. If no discussion is allowed, we could also have an on-wiki pure vote, and make a rule that the only contributions allowed are "support" or "oppose", without commentary. — xaosflux ^Talk 10:23, 17 February 2024 (UTC)

  I imagine there would still be discussion somewhere. Perhaps a week of q&a and discussion, and then a week of secret voting. Similar to how ACE is divided into the q&a phase and the voting phase. –Novem Linguae (talk) 11:47, 17 February 2024 (UTC)

Anonymous voting

Wifione and a couple of other very problematic contributors have demonstrated that RfA needs to be in the open. With anonymous voting, no one would notice the waves of sock and meatpuppets that had been carefully prepared to control who gets to be an admin. Once the word gets around the paid-editing community, they will also set up a hundred dependable voter accounts. They would vote oppose for anyone with a record of opposing paid editing, and support for their candidates. Johnuniq (talk) 01:34, 17 February 2024 (UTC)

I think someone with your technical expertise should know about ACE and its rather robust scrutineering procedures. Also, everyone in the core community (and everyone who RfAs) virulently opposes paid editing, it's a cliché. Even if that conspiracy theory were 1. true and 2. technically viable, there's no pro-paid-editing faction to even boost. theleekycauldron (talk • she/her) 01:45, 17 February 2024 (UTC)

There is no pro-paid-editing faction now because everything is in the open and someone would notice dubious RfA votes. Three scrutineers under pressure to quickly check 2 or 3 hundred votes would find it very hard to investigate dubious voters. Scrutineers would more likely be bound by prescriptive rules that prevent an investigation based on a vague suspicion. We know that Wifione was successful in socking as Lourdes and becoming an admin (diff). We know that an RfA that looked like it was going to be successful was closed when someone decided the candidate was a sock of an LTA (I would have to remind myself where that was). Johnuniq (talk) 03:24, 17 February 2024 (UTC)

Wikipedia:Requests for adminship/Eostrix. –Novem Linguae (talk) 03:53, 17 February 2024 (UTC)

The current process, though, didn't stop either of these situations. If the community really wants to reduce the probability of this happening again, it's going to have to be willing to tradeoff some of its other views that it currently holds by consensus. isaacl (talk) 04:01, 17 February 2024 (UTC)

Those two cases involved talented individuals working alone. There has been compelling off-wiki evidence of two cases showing people organizing off-wiki to set up teams of editors to push their favorite POV in contentious topics. In addition, several similar cases of paid editing teams have been found. They get noticed because of the open nature of editing—someone sees unfamiliar user names arguing in a meat kind of way and uses Google to find a website where it is being arranged. It would be easy for these kind of groups to organize underground and create 100 hard-to-detect socks with 500/30 status. They could then sway an anonymous RfA. That doesn't matter for Arbcom because there are a very large number of voters and they elect a committee where it wouldn't be an enormous problem if there were one or two bad apples. Johnuniq (talk) 04:25, 17 February 2024 (UTC)

Any group willing to put in the long-term effort to build up a coterie of reputable editors is going to be able to influence the current RfA process as well. I'm not saying we shouldn't be concerned about the possibility. But any remedies to address that type of concerted effort are going to involve changing something from the current setup, and probably something creating some kind of hierarchy of trust. isaacl (talk) 04:37, 17 February 2024 (UTC)

You don't know if those two cases involved talented individuals working alone or groups of people working together. What those two cases prove is that the current system isn't good for vetting because it's easy to beat.

Also, admins are a bigger pool than arbcom (400+ active admins) so there is even less damage a single bad admin can do, which is another thing the Wifione case proved. What harm did he do with the Lourdes account? The Wiki is still here. Frankly, bad admins can and have done damage, but it's always fixable, and the current system has passed plenty of bad candidates (and failed good ones).

Above all, we don't know if another system would work better or worse than the current system because we've never tried any other system. Levivich (talk) 16:31, 17 February 2024 (UTC)

I do not think it is prudent to dismiss something purely on the basis of containing a hypothetical conspiracy, since thousands of conspiracy theories are verifiably correct. "What if incandescent lightbulb manufacturers colluded to make shitty bulbs that burnt out prematurely", or "what if Bolsheviks conspired to depose the Tsar", for example, are things which actually happened.

More pointedly, "what if right-wing dudes took over the Croatian Wikipedia" is also a thing which actually happened, so the idea that conspiracies are simply impossible due to Wikipedians being too smart seems untrue. jp×g🗯️ 19:42, 17 February 2024 (UTC)

• With most configurations of the current Secure Poll setup, the "votes" are anonymous, but the "voters" are not. For example here is a list of everyone that voted in ACE last year, and if their vote was accepted. No one knows how they voted, but anyone can know if they voted. So traditional on-wiki methods can be used to investigate participants in secure polls. — xaosflux ^Talk 10:17, 17 February 2024 (UTC)
• Does this need to be anonymous, though? Would it be possible to set up a system exactly the same as the one we have now, but instead of !voting by posting in the article, you !vote by filling out a little form which gets revealed at the end of the RfA process, and the transcript of which gets sent to crat chat if it's close to the 66-70% threshold? The problem here is that we still need to discuss the RfA as a community, so there really may not be an easy fix. SportingFlyer T·C 10:38, 17 February 2024 (UTC)

  This is generally where I am at. I think there is great value in RfA !voters engaging in the discussion (whether reviewing the community questions or reviewing any general discussion) before casting a !vote (whether through a poll or in the current setup). Unlike others editors here, I think in a Secure Poll setup, there will be more "promote" votes because many editors won't notice any concerns that may exist. - Enos733 (talk) 17:46, 17 February 2024 (UTC)

  I think the community being able to assist in the scrutineering of votes in such a way would be positive. It might be a bit weird compared to know if we had a week of discussion followed by a week of votes and then untimed scrutineering, but it seems like a better alternative to now where you'll get a flood of 50-odd support votes before any concerns are even brought up followed by (usually) a shitshow for the 6 remaining days and a sudden closure. — ♠Ixtal ^{( T / C )} ⁂ Non nobis solum. ♠ 17:51, 17 February 2024 (UTC)

  As the arbitration committee elections demonstrate, discussion of the candidates can still occur while using an anonymous vote. Part of the confrontational nature of the current RfA process is that each person weighing in is potentially starting a new discussion thread. In addition to the candidate getting to witness the same debates play out repeatedly, those commenting have to consider their desire to participate in selecting an administrator versus their appetite for engaging in possibly a prolonged debate. I appreciate why some consider this to be a feature, but it also discourages certain personality types from wanting to participate, and I think the RfA process can benefit from getting input from those editors as well. isaacl (talk) 18:27, 17 February 2024 (UTC)

• I think it's perfectly possible to have our cake and eat it too per others in this thread. We can have a discussion section about the candidate where people can bring up potential concerns with a candidate combined with anonymous voting. If someone has a reason to vote one way or the other that they want to share with the community to convince others, they can put it in the discussion section. Otherwise everyone else would simply vote. Pinguinn 🐧 19:47, 17 February 2024 (UTC)
• I think the issue is that it is not anyone's business what a person's reason is for voting a certain way. It is none of the crat's business either. When we vote for ARBs we do not need to pontificate or satisfy someone else by providing a "valid" reason. If someone wants an issue or an experience that they had with a candidate to be known they can choose to share it on a talk page. This way everyone votes their conscience and nobody holds grudges over a vote. Primefac has said RFA is a consensus building, but that is not what it is, it is a vote. 75% is an automatic pass so where is the consensus in that? See this discussion where multiple editors said it is a straight vote. The cliff notes: In that discussion, HJ Mitchell said it was a straight vote, Vanamonde93, RickinBaltimore and Serial Number 54129 agreed. Lightburst (talk) 22:43, 17 February 2024 (UTC)

  The views of four editors can't be extrapolated to represent the views of the community at large. Many editors have more nuanced opinions on the matter. isaacl (talk) 23:13, 17 February 2024 (UTC)

Tangential discussion that involves a personal dispute Aaron Liu (talk) 21:31, 20 February 2024 (UTC)

Here is the broader context of HJ Mitchell's comment: If your vote (and it is a vote, RfA is about numbers, not discussion or consensus) is not based on the candidate's suitability for adminship, it absolutely should be struck. Given that you recently placed yourself at the center of controversy by casting a !vote that would be struck per HJ Mitchell's reasoning, it seems duplicitous for you to cite him as being in agreement with your position. I have taken the liberty of pinging him here since I'm sure you wouldn't want to misrepresent his position. LEPRICAVARK (talk) 00:41, 18 February 2024 (UTC) I am very concerned by your hostility toward me and I have left you a message on your talk page. I hope we can each discuss governance without sniping and parsing. The gist of the statement involved whether RFA is a vote or consensus. There seems to be this belief that we all have to agree on everything, including RFA candidates. And that leads to vote striking and other nonsense in an RFA. I want to ask you to please avoid following me as you did to Lilian's page, and avoid calling me out as you did here, and on other pages yesterday, and below. It is troubling and it feels like harassment. Thanks. Lightburst (talk) 02:55, 18 February 2024 (UTC) I like to call things what they are instead of what we think they should be (very un-Wikipedian of me!) so yes, it's clearly a vote. But I've always said that I'd prefer it to be a discussion of a candidate's suitability for adminship, like a (good) job interview and less like a popularity contest. I don't know if the outcomes would be different but I think it would make the experience more pleasant for all involved, which might attract a different kind of candidate and/or a different kind of voter. HJ Mitchell | Penny for your thoughts? 06:20, 18 February 2024 (UTC) and avoid calling me out as you did here, and on other pages yesterday, and below @Lightburst Unrelated to all your other comments, this feels unenforceable and a strong net negative. If you do not wish other people replying to you or calling you out, you should stop making statements. If you think someone is harassing you, please take it to ANI/another appropriate venue and get a formal IBAN. You assert bold claims that also require sufficient backing; undercutting any opposing arguments with harassment claims will not do any favours. Soni (talk) 04:29, 20 February 2024 (UTC) @Soni: Please AGF, you just accused me of hurling accusations to undercut opposition. Tsk tsk. You should read the details of the harassment. I do not ever want to go to ANI for anything and as you know discussion and conflict resolution should always come first. Anyway, read up in that link and see how the editor is following and harassing. I am trying to work it out with them rather than escalating. I asked the editor for an informal Iban and I think maybe they agreed. They snarled and deleted the message. Anyway, I did not want to let your bad faith accusation hang out there like a Matzah ball. Lightburst (talk) 15:08, 20 February 2024 (UTC) I had already read the details, found them lacking, and do not think your diffs supported your summary. But that's beside the point; this conversation, and 2-3 threads above, is already not for this venue. AGF is not a shield. And shaming others with a tsk tsk is usually not the way to prove your case. You seem to simultaneously believe there is an informal IBAN already agreed upon; and also need to reinforce it and request for it again in unrelated other venues. Either you both already have an informal IBAN, or you don't. Claiming harassment at unconnected conversations is the opposite of "dispute resolution", so I request you to take your diffs to an actually relevant venue please. Soni (talk) 19:56, 20 February 2024 (UTC) Thank much so much for enabling and shaming. I will always stick up for myself in all discussions. Now I have to ask you to leave me alone. Lightburst (talk) 21:21, 20 February 2024 (UTC)

• Consensus usually refers to general agreement among the members of a group or community. The community is in general agreement that any user receiving above 75% of the !votes at an RFA will be granted adminship, and any with 65-75% would be sent to a 'crat chat to find a general agreement amongst 'crats whether there was general agreement amongst the RFA participants whether the user should be an administrator. We do not need to go to a 'crat chat when a supermajority of participants finds general agreement to promote. So yes, it is a consensus-building exercise (see also the 2019 reaffirmation of this), and we use the votes to help determine that consensus. (please ping on reply) Primefac (talk) 08:56, 18 February 2024 (UTC)

  FWIW I was agreeing with what Harry said about the consequences of commenting at RFA. I don't think it's a straight vote. It's closer to a straight vote than many of our consensus-building discussions, and I would personally also agree with Harry that it should be more of a discussion, but if it was purely a vote then support percentage should predict whether someone gets the bit after a crat chat and it does not. Vanamonde93 (talk) 16:37, 20 February 2024 (UTC)

• My only thought here is that we should allow open community discussion so users can provide pieces of information (i.e. potentially problematic diffs). This way, some users that haven't seen a particular thing will be more likely to see it and will be more well-informed. And we should be able to ask the candidates questions, too. ‍ Relativity 01:28, 20 February 2024 (UTC)
• I agree entirely with Johnuniq above. Every behavior that we dislike in the current system would remain relevant in an anonymous vote; there would just be fewer consequences for being a jerk to your colleagues or for attempting to subvert the system. We have systems in place to stop people from being nasty to each other. We should use them. Vanamonde93 (talk) 16:50, 20 February 2024 (UTC)

Proposal 3: Add three days of discussion before voting (trial)

Following the passage of this RfC, for 5 Requests for Adminship (RFAs) which are not closed as SNOW or NOTNOW (or for six months, whichever happens first), RfAs will last 10 days. For the first 3 days (72 hours) no "Support", "Oppose", or "Neutral" comments/!votes may be made. Optional questions may still be asked and answered, and general comments may still be left. After 3 days, "Support", "Oppose", and "Neutral" !votes may be left for the remaining 7 days (same length as an RfA is now). This RfC does not change other RfA procedures/rules. 17:49, 17 February 2024 (UTC)

Support (proposal 3)

1. This doesn't fix RfA. But I think it has the potential to take some of the temperature of it down since people will be able to express concerns and respond to those concerns without the immediate stakes of having that discussion impact the support percentage. It might also allow for easier clerking because comments which cross the line may not be attached to a vote. The trial would also serve as a data point to understand how other proposed RfA reforms (such as anonymized voting with seperate discussion being discussed above) might work in reality. I suspect that this change might make increase the number of opposes because people who check an RfA once may not return to !vote or may !vote after seeing concerns and thus oppose or abstain from !voting where they might have supported before. But I do not think that will necessarily mean fewer people will pass RfA. It's just that support %s will return to historical norms, with fewer ending in the very high 90s. Importantly for me, I think it has the potential to make RfA less unpleasant for candidates. So I think this is worth trying. Barkeep49 (talk) 17:49, 17 February 2024 (UTC)
2. As I've previously discussed, I think a two-phase approach would help improve the effectiveness of discussion and provide the opportunity for additional information being available to those offering their support or opposition when the second phase begins. isaacl (talk) 18:01, 17 February 2024 (UTC)
3. Per Barkeep49. Having discussion first makes everything clearer and settles against inertia to make the votes much fairer. Aaron Liu (talk) 18:22, 17 February 2024 (UTC)
4. I think it's worth trying. It doesn't take away anything (there's still a 7-day !voting period), so it seems like a constructive experiment. Schazjmd (talk) 18:30, 17 February 2024 (UTC)
5. This seems like it could do quite a bit of good, and if it ends up going poorly, it's temporary. QuicoleJR (talk) 19:04, 17 February 2024 (UTC)
6. I support this trial being done. jp×g🗯️ 19:35, 17 February 2024 (UTC)
7. I like this as a first step to a secret ballot. Many of the concerns about a secret ballot are that voters would not be aware of past indiscretions without doing a significant amount of research, but if this process works to raise and vet concerns about the candidate first, it would open the door up to reconsidering a closed voting process. --Ahecht (TALK
   PAGE) 20:14, 17 February 2024 (UTC)
8. This seems to be a good trial. I would suggest that the trial last until 5 RfAs occur (without the time period), just to make sure we get a few RfAs to evaluate. --Enos733 (talk) 20:17, 17 February 2024 (UTC)
9. Per Special:Diff/1208285818 RoySmith (talk) 20:27, 17 February 2024 (UTC)
10. While I'm against ever turning RfA into a secret ballot, I think this is worth a trial run as a way of avoiding the premature pile-on supports that so often occur prior to proper vetting. I do, however, wonder if this will dissuade candidates from running during the trial period for fear that they will end up linked to an experiment gone wrong. Also, will there be any measures taken to prevent participants from declaring their intent to !vote one way or another during the discussion phase? LEPRICAVARK (talk) 20:34, 17 February 2024 (UTC)
11. I rather like this idea. If nothing else, it would allow people to express their thoughts, and receive feedback on those, before actually staking them to a position on supporting or opposing. It's always easier to change one's mind prior to publicly taking a stance than after having done so. But absolutely never would I support any "secret ballot", and I am certainly not supporting this as a way toward that. This should be a supplement to, not a replacement for, discussion during the actual "voting" phase. Seraphimblade ^{Talk to me} 20:46, 17 February 2024 (UTC)
12. Let's give it a shot.—S Marshall T/C 21:01, 17 February 2024 (UTC)

    And I would support the alternative proposal for 2+5 even more strongly.—S Marshall T/C 16:05, 19 February 2024 (UTC)

13. I think this is a very good idea, so don't mind my concerns too much. I'm concerned 10 days is a lot. I'm concerned people will comment and then forget to vote. And I'm concerned this will be a tool to increase self-nominations, since you won't immediately get jumped on with a bunch of opposes. The first concerns will be interesting to see if we approve this. The final concern is more pressing, I think we should exclude self-nominations from the trial process. But we should absolutely try this. SportingFlyer T·C 21:07, 17 February 2024 (UTC)
14. Worth a shot NW1223<Howl at me•My hunts> 21:11, 17 February 2024 (UTC)

    Support the ammended (5+2) duration. No need to add stress to what is already a stressful proccess NW1223<Howl at me•My hunts> 03:07, 20 February 2024 (UTC)

15. With the addendum that candidates be allowed to opt-out of this and in to the normal system. I would slightly prefer if votes were left on a seperate page entirely and not allowed to have attached rationales, but this is an excellent first step Mach61 (talk) 21:29, 17 February 2024 (UTC)
16. Yes, I support anything that gets us closer to a "secret ballot. As I said in discussion it is not anyone's business who another editor voted for. This is the way editors can be free to vote their conscience. But this is a step in the right direction by removing the badgering on the voting page. Lightburst (talk) 22:46, 17 February 2024 (UTC)
17. Hopefully something like this can make the process easier for both candidates and opposers. Thebiguglyalien (talk) 23:49, 17 February 2024 (UTC)
18. Let's try it. It can't possibly be worse... HouseBlaster (talk · he/him) 00:48, 18 February 2024 (UTC) UPDATE: second choice to 3B HouseBlaster (talk · he/him) 05:04, 22 February 2024 (UTC)
19. This'll be great because otherwise someone might support, but then a new piece of information comes up where they might oppose then, but may not strike their support. Or vice versa. Let's at least give this a try. ‍ Relativity 01:36, 18 February 2024 (UTC)
20. Worth a shot :) theleekycauldron (talk • she/her) 02:50, 18 February 2024 (UTC)

    I would also support a 2+5 – lengthening RfA isn't my favorite tack, but again, let's try something. I'm glad we're getting proposals off the ground :) theleekycauldron (talk • she/her) 18:24, 18 February 2024 (UTC)

21. I think this discussion-first proposal is worth trying, though I have concerns: 1) Extending the length of time someone is under scrutiny could increase the tension for the candidate - especially as they have to wait until after people start examining/criticising them before the voting begins; 2) It is possible that the same incivility will occur when people raise concerns, and others directly challenge those concerns, and heated discussions result, as this proposal doesn't address those issues, just delays the voting period; 3) This may increase the number of questions a candidate will have to answer both in terms of the extended time, and the lack of information being put forward by voters (so people may seek out that extra information by asking questions of the candidate), which may put off future candidates. However, my concerns may not materialise, so it's worth trying. My main thoughts, meanwhile, on toxicity in RfA is that it is caused by threaded discussions - people engaging in direct responses, sometimes emotionally, heated, and personal, which then escalate. I think it would be worth trialling an RfA in which people only comment in their own sections, as in ArbCom discussions. I won't formally suggest that now, as I feel that would take attention away from this proposal. I'll suggest it after this proposal has been trialled (as I hope and expect it will be). Thanks Barkeep49 for suggesting it. SilkTork (talk) 02:56, 18 February 2024 (UTC)

    Could prohibiting threaded replies in the support/oppose/neutral section work? - Enos733 (talk) 03:39, 18 February 2024 (UTC)

    That would be the idea, as that is where the toxicity occurs. Folks tend not to get so heated over comments as much as they do over votes, especially (almost exclusively) negative votes. I've seen people complain when someone raises a justifiable objection in an otherwise 100% support RfA as though that single negative vote somehow soils the RfA; or perhaps they may be concerned that the negative comment will have traction, and turn the course away from their favoured candidate. The nature of Wikipedia is that we invest a bit of ourselves into the project, this aids motivation, work ethic, pride in the job, etc, but can drive us to be too obsessive and to be lead by our emotions rather than our intelligence and judgement. SilkTork (talk) 11:00, 18 February 2024 (UTC)

22. Worth trying to see if it can improve our process. 0xDeadbeef→∞ (talk to me) 04:57, 18 February 2024 (UTC)
23. Frankly, I don't think this will accomplish much. Supporting in the hope that I'm just a little jaded when it comes to RfA, and in support of experimentation with fraught processes in general. — Rhododendrites ^talk \\ 04:58, 18 February 2024 (UTC)
24. Worth a try! Leijurv (talk) 07:21, 18 February 2024 (UTC)
25. Weak support. Making the period longer is a strong disadvantage for me, and I had hoped for a 3 days discussion and 5 days !voting proposal. That said, the current situation is untenable and this is a move towards a less cruel system where votes are blinded and discussion is an actual discussion rather than a vote. —Femke 🐦 (talk) 07:39, 18 February 2024 (UTC)
26. Support as a trial: almost anything is worth trying. Hopefully we have five knowing volunteers who are happy to test this out (or at least no less happy than they would be to try the current RfA system). Back when I thought it was possible I would ever run at RfA, I considered announcing it 2 days in advance and soliciting discussion and feedback so that constructive criticism wouldn't be so high-stakes or so argumentative as in a "vote". I don't like extending the total length of time, however: it should be 3 days of discussion and 3 days of !votes. — Bilorv (talk) 08:47, 18 February 2024 (UTC)
27. Definitely worth a shot, although I share @Usedtobecool's reservations below regarding both the increased length as well as the uncertainty for the candidate who is answering questions for three days without getting to see any confirmed support. Regardless, hard to know until we've seen it in effect—and better to try something different than to leave it alone when all agree it's currently in need of help. Retswerb (talk) 09:00, 18 February 2024 (UTC)
28. I think this needs to come with a prohibition of comments in the voting section, but it goes in a good direction. Femke is right about the length though. —Kusma (talk) 09:20, 18 February 2024 (UTC)
29. I am not yet convinced that this will make a meaningful difference, but I am fully in support of trying things out to see if they actually do. There is essentially no downside to testing this. firefly ( t · c ) 11:10, 18 February 2024 (UTC)
30. Per Firefly. No reservation re. length of time, as the period of most stress will be reduced from seven to three days. Just to call it a '10-day period' is too round; RfA is likely most broken now because everything is bundled together. ——Serial 13:49, 18 February 2024 (UTC)

    Serial Number 54129, there is nothing stopping whatever happens in the first three days from continuing the full 10-days. Or did I miss something? Best, — Usedtobecool ☎️ 03:12, 20 February 2024 (UTC)

31. We will never see if it works unless we try. I think that ideally, we should restrict the first three days to Q&A only, without discussion that may influence future votes (particularly since the candidate will be tempted or even obliged to reply to that discussion), but let good not become the enemy of perfection. We can also try the ARBCOM-style threaded discussions, where only the user and the candidate may speak to each other, but let's see first if this trial succeeds. Szmenderowiecki (talk) 15:14, 18 February 2024 (UTC)
32. Support per above. 🌺 Cremastra (talk) 15:57, 18 February 2024 (UTC)
33. Why not. ~~ AirshipJungleman29 (talk) 20:06, 18 February 2024 (UTC)

    I support this in principal. TEN consecutive days sounds like an ordeal. A couple days of Q&A followed by a break. Candidate should have the option to have a break in between Q&A and voting. Schierbecker (talk) 20:27, 18 February 2024 (UTC) (I now support the 5+2 proposal.) Schierbecker (talk) 18:21, 20 February 2024 (UTC)

34. Weak support. Mach61 (currently support vote #15) made a very good point that this should not be mandatory for the next five nominees and I'm disappointed that doesn't seem to have gotten traction. I also echo the concern expressed by Femke (currently support vote #25) and others that jeez, ten days is an awfully long time. City of Silver 20:32, 18 February 2024 (UTC)

    I don't love that it's mandatory, but I would less enjoy the perception that not waiving the discussion period is a sign of weakness. We need a mix of strong and weak candidates to test this, and allowing a self-selection bias throws off the results. theleekycauldron (talk • she/her) 23:15, 18 February 2024 (UTC)

    If I were ready to run for adminship but I didn't want to participate in this experiment, I'd opt out by, well, not running until the experimental period is over. Passively opting out like that would have no avoidable consequences for me. Either the experiment fails and I'd get my way or it passes and I'd be exactly where I would have been had I, against my will, been one of its five subjects. It's the community that would suffer here because we'd be missing out on up to six months of administrative contributions from someone who's good enough to get promoted.

    So no matter what, there will be an opt-out and the candidates who participate will have self-selected. I know I'm on kind of a weird side of AGF when I go on and on about an adminship-worthy candidate getting sneaky like this. I'm just pretty sure that someone who would opt out if they officially could would be awfully motivated to passively opt out if they officially couldn't. City of Silver 21:27, 19 February 2024 (UTC)

    Yes, this is exactly what will happen. Anyone who is hesitant will simply wait out the next five RFAs. Which may take a whole six months, if it makes enough people hesitant. -- asilvering (talk) 21:39, 19 February 2024 (UTC)

    Support only with Mach61's proposal that any candidate should be allowed to opt out of the trial; if consensus does not develop for that condition, my !vote should be counted as an oppose per Novem Linguae and Yngvadottir. Schierbecker also makes a good point regarding allowing the candidate to take a break. voorts (talk/contributions) 20:40, 18 February 2024 (UTC)

35. I'm guessing the benefits will be marginal at best... but same with the downsides. Why not try it? —Compassionate727 ^(T·C) 02:06, 19 February 2024 (UTC)
36. I think it is worth trying. Like a few others I think 10 days may be too long. But I am not sure we can make it perfect right away so I am willing to support. Bruxton (talk) 03:09, 19 February 2024 (UTC)
37. Worth a trial. — ♠Ixtal ^{( T / C )} ⁂ Non nobis solum. ♠ 11:53, 19 February 2024 (UTC)
38. We need to try something different, and we can always go back if this flops. _{Duly signed,} ⛵ WaltClipper -(talk) 13:32, 19 February 2024 (UTC)
39. Support I will continue to vote for all sufficiently logical RFA reform proposals. I wish this trial was longer, but 5 RFA should at least be a step forward, in case it's a net positive towards RFA's hostility. Having more days for RFA does feel like a negative, but not sufficiently enough for me to oppose. Soni (talk) 14:36, 19 February 2024 (UTC)
40. Support. Worth trying. It could mitigate some unfair RFA dynamics that happen even in the absence of incivility or trolling. MarioGom (talk) 18:16, 19 February 2024 (UTC)
41. Support. Although I like the 2+5 idea better. It should relieve stress either way. — Preceding unsigned comment added by Lulfas (talk • contribs) 20:23, 19 February 2024 (UTC)
42. Support I'd like to see this AND anonymous voting. That would eliminate the oppose-vote-badgering as well. But I'll take this as a sensible first step in the right direction.→StaniStani 01:25, 20 February 2024 (UTC)
43. Support. I think this is worth trying. In the interests of completeness, I note that there were at least two attempts at RfC-style RfAs in 2007: Wikipedia:Requests for adminship/Moralis and Wikipedia:Requests for adminship/Matt Britt. Neither was successful, either for the candidate or for changing how RfA works. It was also seventeen years ago; I doubt many people who participated are around now or if they are, remember the attempts. I think this idea avoids one of the worst pitfalls: it will still be obvious how to register support or opposition. Mackensen (talk) 01:47, 20 February 2024 (UTC)
44. Support – it's worth trying something new and this seems reasonable, though I think the 2+5 idea is better (I think the general tenor of most RfAs is clear well before 7 days, so no need to make a candidate stay focused on/tied to the process for 3 extra days). I also considered if candidates should be able to opt out but Barkeep49's comment below swayed me against it. RunningTiger123 (talk) 02:08, 20 February 2024 (UTC)
45. Support the trial period, let's see what happens. Z1720 (talk) 14:42, 20 February 2024 (UTC)
46. Weak support it's opt-in, if someone wants to try this then sure. I don't think it is going to solve any problem and may require some technical clerking of the timers. — xaosflux ^Talk 15:09, 20 February 2024 (UTC)
47. Support per Barkeep. Ritchie333 ^{(talk) (cont)} 16:37, 20 February 2024 (UTC)
48. Support. I'll have more to say when I write a debrief, but I would have found the first few days of my own recent RfA less stressful had this been in place, since it would have reduced the feeling that at any moment someone could leave a critical comment/!vote that I'd have to respond to as quickly as possible to try to stem a flood of immediate opposition. I oppose making it opt-in (I'll comment on that below). I'd prefer the 3+7, since for purposes of gaining information from a trial, I think it'll be easier for us to sense if the period is too long than if it is too short. Sdkb ^talk 19:35, 20 February 2024 (UTC)
49. Support (second choice) to 3b as 10 days in total seems quite long, but the discussion first model is good overall. Pinguinn 🐧 20:19, 20 February 2024 (UTC)
50. Support - I also think 3+5 (or even 2+5) could be interesting as well. - jc37 05:36, 21 February 2024 (UTC)
51. Nice idea; second choice to 3B. --JBL (talk) 22:33, 21 February 2024 (UTC)

Oppose (proposal 3)

1. IMO too complicated and won't do much good. RFA really wouldn't be that hard to 80% fix but IMO his isn't it. Sincerely, North8000 (talk) 20:34, 17 February 2024 (UTC)

   It's not complicated at all, the only difference is that we add 3 extra days to the start in which voting is disabled. Aaron Liu (talk) 22:12, 17 February 2024 (UTC)

2. Oppose. Seems to extend RFA from 7 days to 10 days, which will probably increase stress for candidates. –Novem Linguae (talk) 02:48, 18 February 2024 (UTC)

   @Novem Linguae: I agree that ten days is long, but I am hopeful that the actual vote may be less stressful. I think many of us would like to try a different way. (I hope this does not feel like badgering) :) Lightburst (talk) 03:01, 18 February 2024 (UTC)

3. per Novem Linguae. * Pppery * _{it has begun...} 02:49, 18 February 2024 (UTC)
4. Oppose per Novem Linguae, this would just serve to discourage more possible candidates. Lightoil (talk) 02:52, 18 February 2024 (UTC)
5. Oppose anything that makes the process longer isn't suitable. Lee Vilenski ^{(talk • contribs)} 09:38, 18 February 2024 (UTC)
6. Oppose basically per Novem Linguae. This would simply prolong the process for the candidate (and we already have participants being impatient with candidates who don't appear to have their RfA on speed dial for the entire 7 days). It would be difficult to address the merits of the candidacy without in effect casting a premature vote, so I expect the preliminary 3 days would be mostly questions, and there's already a tendency to ask quirky, sometimes repetitive questions; some RfAs in recent years have been a bit of a marathon of questions. I'd expect this change to exacerbate that, and also to front-load the voting once it opens with reactions to the candidate's responses; not only would this mean 3 more days of intensity for the candidate, but fielding questions isn't every candidate's strength, and it doesn't track perfectly with the kinds of communications skills we want from admins. Yngvadottir (talk) 11:27, 18 February 2024 (UTC)
7. Oppose. I'm sorry to be here, and I appreciate the effort BK49 and others are putting in to try to improve the climate at RFA. But I worry this would have the opposite effect than intended on the candidate's experience. First off, there's the length issue that Novem Linguae points out above. 7 days of hell are quite long enough. Second, there's the issue of morale. By their very nature, negative concerns are...louder, for want of a better word. I suspect that people who have concerns about a candidate are far more likely to post in those first few days than people who don't. As things stand, a candidate who is coming under fire can still see the supports piling up: here they're going to have to endure 3 days of hard questions without the emotional cushion of the supports. Unless of course you allow people who trust the candidate despite concerns to say so, in which case we're back at a !vote. Third, there's the issue of voter commitment. Having seen the RFA banner go up on your watchlist, you as a semi-engaged !voter would need to remember to check back in in 3 days time to express your opinion, assuming you don't have one at the outset. I suspect a fair number of editors are not going to be able or willing to do this. Finally, and taking a step back for a moment; even I've not been here long enough to remember RFC/U, but I did look up why it failed, and I believe a lot of the reasons it got to be an unproductive exercise will apply here as well: I think it does more to enable toxicity, not less (regardless of which "side" that toxicity is coming from). Vanamonde93 (talk) 23:04, 18 February 2024 (UTC)

   In terms of the RfA banner, I think that problem is partially solvable by changing the text after the initial discussion period. Starting with something like "questions are now open" to the normal watchlist notice later. My hope is that we can recreate the calmer environment of the Arbcom elections in this initial period; perhaps a follow-up proposal can mirror this process even further. —Femke 🐦 (talk) 08:27, 19 February 2024 (UTC)

   The banner problem probably can be fixed technically. But while this change may make the environment calmer, I think we'd be sacrificing the candidate's wellbeing for some decorum. I would like participation to be not unpleasant for every good-faith !voter (which, to be clear, is almost but not quite every !voter): but I would prioritize the candidate's wellbeing over that of the !voters, and over our desire for calm. Vanamonde93 (talk) 05:30, 20 February 2024 (UTC)

8. Oppose per Vanamonde93, Novem Linguae, and Yngvadottir. voorts (talk/contributions) 23:12, 18 February 2024 (UTC)
9. Oppose While I can see how RFA needs to be 7 days, I'm aware that a 7 day process is intimidating, and not something everyone can readily make time for. However most of the !voting is in the first three days, many RFAs are relatively quiet after the first 72 hours. A two stage RFA with an extra three days at the beginning makes one of the problems of RFA worse, as it is likely that the intense bit would also be extended as both the extra three days and the first three days of voting will be high pressure for the candidate. There's also the problem that RFA is already too focussed on the Q&A section at the expense of actually vetting the candidates edits. My fear is that adding an extra three days for increased question activity will exacerbate this problem and make RFA less attractive to potential candidates, as well as less effective at weeding out candidates who would make bad admins ϢereSpielChequers 01:05, 19 February 2024 (UTC)
10. Oppose. Any candidate wanting to extend the process from 7 to 10 days may already spend their first three days at Wikipedia:Requests for adminship/Optional RfA candidate poll. – wbm1058 (talk) 03:44, 19 February 2024 (UTC)

    OCRP is a meta-discussion about the candidates’ chances, it has nothing to do with what the participants personally want in an admin Mach61 (talk) 04:38, 19 February 2024 (UTC)

    Speaking as someone who would like to be an Admin to do more on WP:RM, and has already done OCRP, I honestly find the prospect of a ten-day process, the first three days of which won't have any !voting so you're essentially just twisting in the wind waiting for something to go bad, even more daunting than that of the present 7-day process. FOARP (talk) 13:02, 19 February 2024 (UTC)

11. Oppose per Vanamonde - as someone who had a controversial RfA, I think it would've been worse if the first 3 days had just been people discussing - i.e. most likely expressing concerns. The "emotional cushion" - as Vanamonde put it well - of knowing that 80+% of people did support me, was definitely nice and gave me confidence through my RfA. Also yeah, questions were definitely the most stressful part of RfA for me, and more time for those would not help imo. Galobtter (talk) 06:22, 19 February 2024 (UTC)
12. Oppose - I'm willing to be convinced otherwise, but in the one RFC where this was attempted that I've taken part in this seemed just to increase the drama rather than dialling it down. I think, even if it is obeyed, this just gives a three-day period to people for "debating" the significance of whatever reason they have for opposing without allowing supporters to register any real view, as their views will already be reflected in the nomination. However, I don't think it will be obeyed in practise, rather people will simply leave "comments" that are substitute votes. FOARP (talk) 12:55, 19 February 2024 (UTC)
13. Oppose There seem to be too many procedural problems. Trying to forbid !votes is impractical and impossible as there will be plenty such regardless in places like Wikipediocracy, Discord and elsewhere. And compressing the Q&A into a 3-day period will put more pressure on the candidate to be responsive during that time which might be difficult to schedule. And there will tend to be more questions because that's the main way for enthusiasts to participate initially. And they will tend to be loaded questions. And then, when the floodgates open for actual voting, there will be a big rush to get the groupthink bandwagon rolling. And so on. No – what RfA needs is a secret ballot like Arbcom. That seems to work well without so much intense drama, which is what's wanted. Andrew🐉(talk) 16:51, 19 February 2024 (UTC)

    I think you're misunderstanding.

    1. You can't vote for an RfA in Wikipediocarcy. Discussions can still contain opinions about the candidate.
    2. The Q&A and discussion are allowed for the entire period, not just the first 3 days.

    I do not see the benefit of secret ballots. These will strip reasons from !votes, which people can be persuaded by. Aaron Liu (talk) 16:55, 19 February 2024 (UTC)

    See How the Secret Ballot Changed Democracy. It's very clear from history that open voting results in the problems which we see at RfA and so secret ballots are needed. Andrew🐉(talk) 23:07, 19 February 2024 (UTC)

    Could you kindly explain how bribery, corruption and stuff have applied here? Aaron Liu (talk) 23:16, 19 February 2024 (UTC)

    RfA obviously has lots of "stuff". This discussion is now part of WP:RFA2024 which explains that "RfA is widely agreed by the community to be toxic and hostile to participants..." Andrew🐉(talk) 08:47, 20 February 2024 (UTC)

    I thought that meant hostility towards the person being nominated? Outright harassing people who oppose seems to be banned.
    In real life, the secret ballot works, also because of another factor: the election is announced a long time before it happens, we know who's up for selection, and most importantly, we have the time to discuss. Without a version of proposal 3 being enacted, a lot of votes can be blinder and misinformed, as there are probably less people participating in the discussion section than voting. Aaron Liu (talk) 00:36, 21 February 2024 (UTC)

    No, it means all participants. So, the starting point and reason for this RfC is is that RfA is broken and toxic. We know that a secret ballot will defuse this because it is used for Arbcom elections and that does not have a similar reputation. Andrew🐉(talk) 21:57, 21 February 2024 (UTC)

    The Arbcom election is a traditional election with candidates who are ranked and a ton of votes. It only has Q&A, and not other kinds of discussion. RfA needs quite a bit of discussion, so a secret ballot doesn't seem that good. Doing the scrutineer-"hiring" process for every single RfA would also be quite tedious. Aaron Liu (talk) 23:54, 21 February 2024 (UTC)

14. Weak oppose. I've held off on !voting, because I'm very reluctant to oppose just having a trial of something, to see if it works. And my oppose is a mild one, for that reason. But the comments of Vanamonde93, WereSpielChequers, and FOARP have persuaded me to oppose. In particular, WSC's point about extending the most stressful part while leaving decision-making for a stage where interest sometimes wanes, pushes me to here. --Tryptofish (talk) 19:52, 19 February 2024 (UTC)
15. Weak oppose. I initially started posting here since there isn't a formal "neutral" section. I thought I was leaning to support, but wouldn't want to do so unless a few suggestions already mentioned were formally added to the proposal: namely, I think 10 days is far too long, and I like Femke's solution to Vanamonde's issue #3. But in writing this out I've come around to think Vanamonde's issue #2 (echoed by FOARP) is the most important one. I don't see anyone in the Support section saying "I would run for RFA if it were like this", though Bilorv's comment comes closest. -- asilvering (talk) 21:12, 19 February 2024 (UTC)

    Add Sdkb's comment to Bilorv's and I'm back to equivocal about it. I do think it's important that the people we keep in mind are the potential candidates, not so much the feelings of other participants (I think Vanamonde said something like this somewhere else). The problem we're trying to fix isn't "RFA makes people upset in general" but "fewer and fewer people are running for RFA, likely due to a perceived toxicity in the process, leading to a dearth of new admins." -- asilvering (talk) 20:44, 20 February 2024 (UTC)

16. Oppose Maybe this is a failure of imagination on my part, but I fail to see how making an already stressful process longer is going to help. Maybe if after the three days, contributors could only vote without making any form of substantive comment, I could see that being an improvement. But as written this just seems like it's going to prolong the misery put on candidates. Sideswipe9th (talk) 22:45, 19 February 2024 (UTC)

    Like, this doesn't substantively change the environmental problems that are the cause of RfA toxicity towards the candidate. Editors will still be able to violate WP:CIV and WP:NPA because no-one is willing or able to moderate those violations when they occur. All this is doing is giving three extra days for that same type of vitriolic comment to be made. Sideswipe9th (talk) 22:50, 19 February 2024 (UTC)

17. Oppose: A longer RfA period will lead to fewer RfA candidates. I honestly don't see how this would lead to any type of benefit. Hey man im josh (talk) 01:22, 20 February 2024 (UTC)
18. Oppose per above. Overcomplicated and seems like a solution in search of a problem. Yeah RfA could use fixing, but this isn't the solution -Fastily 01:28, 20 February 2024 (UTC)
19. Oppose per above. Because future RFA candidates definitely need more stress. LilianaUwU ^{(talk / contributions)} 03:35, 20 February 2024 (UTC)
20. I do not see how adding three days of discussions before the RfA reduces the stress of the process; and I cannot understand how this fixes issues like oppose-badgering either. JavaHurricane 13:19, 20 February 2024 (UTC)
21. I like the idea in theory, but this just drags out a painful process even further. In practice this will not help and will cause more stress for the person who is nominated. Nemov (talk) 13:41, 20 February 2024 (UTC)
22. Overly long. 3B is better, but still not ideal. The process shouldn't be extended. Sincerely, Novo Tape^{My Talk Page} 16:36, 20 February 2024 (UTC)

Neutral (proposal 3)

1. Neutral. I don't like 10-day RfAs, but the concept of a discussion period before the !vote would reduce stress. Prefer 3b. Queen of Hearts (talk • stalk • she/they) 03:41, 21 February 2024 (UTC)

Discussion (proposal 3)

• What happens after the 5 RfA/6 month timeframe? Schazjmd (talk) 18:26, 17 February 2024 (UTC)

  It goes back to 7 days with immediate !voting. If the community wants it to continue a new RfC would need to establish that consensus. Best, Barkeep49 (talk) 18:30, 17 February 2024 (UTC)

• As a note, discussion-first has been tried before, see Wikipedia:Requests for adminship/Ironholds 2. Izno (talk) 19:03, 17 February 2024 (UTC)

  Two differences in that 2008 RfA is that the discussion phase was closed when the voting phase started, and unlike today's process, there was no limit on how many questions each person can ask. isaacl (talk) 19:09, 17 February 2024 (UTC)

  That was one of the four failed RFAs that that candidate has had, so I'm not sure they were the best guinea pig for that experiment. --Ahecht (TALK
  PAGE) 20:32, 17 February 2024 (UTC)

  I made only a comment about whether we have tried this before. Any other inferences are your own. ;) Izno (talk) 20:42, 17 February 2024 (UTC)

• Should candidates be allowed to opt out and use the regular process, if they prefer?—S Marshall T/C 21:19, 17 February 2024 (UTC)

  Probably yes, but that could lead to an unintended consequence of opposes being cast by voters who are angry at the candidate for not playing along with the experiment. LEPRICAVARK (talk) 22:09, 17 February 2024 (UTC)

  In my opinion, we need to let candidates opt out, or, alternatively, change the proposal from the first five, to the first five who opt in. It also seems to me that we simply don't know how this will affect the percentages for success or failure, and that could leave bureaucrats unsure how to determine consensus. --Tryptofish (talk) 22:50, 17 February 2024 (UTC)

  If a candidate wants to opt out they already have two ways to do so: go in the next 30 days or wait for 5 RfAs/6 months. Best, Barkeep49 (talk) 23:27, 17 February 2024 (UTC)

  I think what's likely to happen is that we'll figure out after the first RfA or two in the trial whether the discussion phase is a boon or hindrance to a candidate's chance of passing with high support. If it turns out to be a hindrance, we might have a lot of trouble getting enough candidates to use it to get meaningful information about how it affects RfA. I understand the reluctance to ask candidates to be guinea pigs on top of an already stressful process, but ultimately for the benefit of the community in the long term we to test out some reform. Someone will need to go first, and if a candidate really doesn't want to be the one, they can adjust the timing of their RfA accordingly as Barkeep said. Sdkb ^talk 19:48, 20 February 2024 (UTC)

• I'm also noticing that some editors are supporting the proposal, saying that they like the way it is a secret ballot, but the proposal as written is not that. --Tryptofish (talk) 22:52, 17 February 2024 (UTC)

  Searching on the word "secret", I'm not seeing any supports saying they like how the proposal is a secret ballot. At the moment, two supports say they think it is a good step towards having a secret ballot in future, and two say they support the proposal even though they do not support a secret ballot. isaacl (talk) 23:16, 17 February 2024 (UTC)

  Lightburst's !vote originally said I support any "secret ballot. That is the only instance I can see, though. Aaron Liu (talk) 23:18, 17 February 2024 (UTC)

  Lightburst also seems to believe that this proposal would remove the so-called badgering of opposes, but I see nothing in the proposal to support that perception. LEPRICAVARK (talk) 00:32, 18 February 2024 (UTC)

  I said removing the badgering on the voting page. So it will stop the badgering on the project page. I still do not like having to publicly declare, "Are you now or have you ever been..." I would like to just vote and move on to building the encyclopedia. When I vote here in my town it is drama free because I do not have to tell the whole town who I voted for and then pass their litmus test. Lightburst (talk) 01:39, 18 February 2024 (UTC)

  So it will stop the badgering on the project page. No, there is nothing in the proposal to support that conclusion. LEPRICAVARK (talk) 18:27, 18 February 2024 (UTC)

  I have real concerns about secret ballot/anonymous voting myself even as I feel it's the option likely to do be the largest change for positive in appointing adminship that has a reasonable level of community support right now. I think it will push the support %s down considerably - as we saw with the move from public to secret balloting for ArbCom - and I think @Johnuniq's concerns about opening us up to UPE has some validity. So I'm not sure we should adopt it despite its potential benefits. But I think the discussion first system could have merit on its own as an improvement at RfA and think it could be paired with some other mooted changes to RfA - including secret ballot - in ways that would complement each other. Best, Barkeep49 (talk) 23:33, 17 February 2024 (UTC)

  Keeping our current process is more likely to result in UPEs becoming admins than moving to a secret ballot. Our current process does not filter out all bad eggs (Eostrix, Wifione) and has horrible effects on the mental health of true volunteers, so hundreds of qualified editors refuse to run. Perhaps some people would find that acceptable if the process kept out bad eggs, but it demonstrably does not. So "but UPE" does not appear to be an argument against changing things, rather the opposite. —Kusma (talk) 10:25, 18 February 2024 (UTC)

  @Barkeep49, can you explain a bit more about the support %s decrease re: ArbCom? I think this predates me. -- asilvering (talk) 21:15, 19 February 2024 (UTC)

  @Asilvering just look at someone like NewYorkBrad. When voting was public here he is getting 97+% support. Where as in 2019 when he ran again he was by far the most popular candidate but got "only" a hair under 80% support. This is also reflected by the drop-off in % support the first year they did secret ballot voting but I'll leave the finding of that to someone else. Barkeep49 (talk) 01:14, 20 February 2024 (UTC)

  @Barkeep49 Hm, I'm not sure that NewYorkBrad's particular example is useful, since that first run is in 2007. Was 2008 the last election to be held without secret votes? I went looking, but I don't see a votes page for that one, just a log. If it's the last non-secret ballot, though, it's hard to say that the secret ballot itself made the %s go down. Here are the 2008 results: [1] and the 2009 (definitely secret ballot) ones: [2]. It's true that there are fewer extremely high % candidates. But there's also much higher voter turnout. Coren is on both, and their vote share increases in the secret ballot. It seems to me that this might reduce the %s for candidates who would otherwise be hitting 90%+ or so, but I'm not sure there's much effect below that. -- asilvering (talk) 01:37, 20 February 2024 (UTC)

• Find it hard to imagine how this would benefit the candidates. Seems like 10 days of stress instead of 7, especially while it's not yet clear what the Bureaucrats can or are willing to do. Was 2+5 considered? Is there something about the necessity to cast votes for 7 days? I think the real guns won't come out at RFA until the real numbers start to come in, which would be the voting period. The first few days may end up like ORCP. Or they maybe as bad as the typical days of current RFAs but without the comfort of a support percentage, making candidates who would have passed withdraw even before voting begins. Usedtobecool ☎️ 02:27, 18 February 2024 (UTC)

  In past discussions, some participants have expressed a desire to allow for editors who edit once a week to have their views considered towards the consensus outcome. This would require a 7-day minimum period where they can weigh in with their supporting or opposing opinion. isaacl (talk) 05:44, 18 February 2024 (UTC)

  I disagree with this argument as it views RfA as a Big Deal. The concept of everyone in the community being entitled and accommodated to vote frames it like a presidential election rather than a discussion to establish consensus. Discussions with much wider-ranging community impact on AN and ANI often last a couple of days only: editors often don't know about the discussion until after the fact, but it's still served its purpose of ascertaining consensus through a sample. — Bilorv (talk) 08:40, 18 February 2024 (UTC)

  I would agree with Bilorv. Furthermore, by engaging in the pre-!voting discussion, you also participate in the process, so people that edit less are included in the RfA by design. —Femke 🐦 (talk) 08:45, 18 February 2024 (UTC)

  Given how dynamic RfA is (not all votes are equal, providing a single diff that makes the candidate look bad has great effect within the first couple of days, not so much in the final hour), this is meaningless in practice. Also, the candidates' mental health is perhaps a bigger issue than the views of a few weekend only contributors. —Kusma (talk) 10:30, 18 February 2024 (UTC)

  Note that since RfAs can start on any day of the week, the potential effect is not limited to weekend contributors. isaacl (talk) 18:09, 18 February 2024 (UTC)

  Yeah I agree, to have a real impact on an RfA outcome diffs have to be dig up within the first few days. I'd be more supportive of a 2 + 5 proposal. Galobtter (talk) 06:28, 19 February 2024 (UTC)

  Barkeep49, given the above, do you think it would be possible to add a 2+5 proposal as an alternative and see if it receives a more enthusiastic support? I do strongly believe that changes that make the RFAs longer have a lesser chance of being successful. — Usedtobecool ☎️ 13:03, 19 February 2024 (UTC)

  Ultimately I don't control this RfC, of course, it's the community. That said I am not without thoughts about what 3 extra days could mean, but that for me is why doing this as a trial is important. It would not surprise me if an outcome of the trial is "discussion period first is good, but 10 days is too long" and so it gets permanently adopted as a 2+5. But in the spirit of an experiment I thought changing one variable (discussion period) rather than two (discussion period + !voting length) would yield more informative information. Best, Barkeep49 (talk) 01:08, 20 February 2024 (UTC)

• Per the concerns about prolonging, maybe we should trial both 3+7 and 2+5, per Usedtobecool and Bilorv. Aaron Liu (talk) 16:35, 18 February 2024 (UTC)
• What happens if the candidate withdraws before the vote starts? Is the RFA still listed at Wikipedia:Requests for adminship by year and other places? Do they have to put a "2" on a second attempt? Can they request courtesy blanking, or even deletion? Suffusion of Yellow (talk) 18:11, 18 February 2024 (UTC)
• The one time I've seen a discussion-first RFC on a contentious topic (the 2018 Daily Mail discussion) the whole thing devolved very quickly into round of "Oh my god HOW DARE YOU stop us from voting!"-style accusations. For better or worse, people come to a discussion with attitudes already formed, and things they want to say. This essentially limits the voting period to the last four days of the RFA or extends the RFA by 3 days, and will lead to people casting not-votes in the discussion as they don't have the time to come back to !vote. I see no sign that discussion would have avoided the FUBARs in recent RFAs - people put down their marker early either way. FOARP (talk) 12:49, 19 February 2024 (UTC)
• In my opinion, the 2+5 RFC should be in its own subsection instead of mixed in with the current RFC. The way things are signed, it now also looks like Usedtobecool wrote the 3+7 RFC instead of Barkeep49. I'd change it myself, but I think moving Usedtobecool's 50 pings down would mess up the pings and people wouldn't see the second RFC. –Novem Linguae (talk) 03:34, 20 February 2024 (UTC)
  • I don't really have much experience with RFCs. My first worry was whether it would be wholly inappropriate procedurally and with respect to identifying a consensus position on the question. Barkeeps's response above suggested to me that it would not be so, though he had other concerns. My priority then was to not taint the RFC, neutrality-wise. I am fine with whatever another editor or a group comes up with to improve upon it. Ultimately, I only care that we determine a best course of action that has some chance of achieving something. Best, — Usedtobecool ☎️ 03:48, 20 February 2024 (UTC)
  • I wonder if Barkeep unintentionally signed with the wrong number of tildes. Easily fixed, if so. — Usedtobecool ☎️ 03:52, 20 February 2024 (UTC)

    The introduction for RfCs is copied over by bot to various central pages (this one is present on Wikipedia:Requests for comment/Wikipedia proposals). The introduction is a short, neutral statement that ends with a timestamp, which the bot uses to determine the end of the text being copied. (There's a maximum character limit above which it won't copy the statement.) A signature ending in a timestamp will do, but so will a plain timestamp, which some might consider to be a more neutral presentation for those using the central pages. isaacl (talk) 04:23, 20 February 2024 (UTC)

    Basically Isaacl hits it on the head. Because the RfC statement is designed to be neutral I prefer to launch RfCs with 5 tildes. Best, Barkeep49 (talk) 05:08, 20 February 2024 (UTC)

  Yeah I noticed that when I got the ping but like you didn't want to double ping people. I believe it could get moved without Used's signature and then in a seperate edit the signature gets added back in. I believe that wouldn't double ping but wasn't certain enough to actually do it and ultimately it's not the biggest deal in the world. Best, Barkeep49 (talk) 05:11, 20 February 2024 (UTC)

  Tildes ping people, signatures don't. Or, we'd have pings going out every time a talk page is archived edited. — Usedtobecool ☎️ 06:03, 20 February 2024 (UTC)

• In order to encourage editors to run, I'd like to see this implemented in a way that allows those who sense that the !voting phase may not go as well as they'd hoped to bow out gracefully after the discussion phase with as little stigma attached as possible. One way to get at that would be to require candidates to affirmatively assent to proceed to the !voting phase as the discussion phase nears its end; if they do not, the default course of action would be to end the RfA with a neutral (not red) background and a note like This is an archived request for adminship that did not proceed beyond the discussion phase. Sdkb ^talk 19:52, 20 February 2024 (UTC)

Proposal 3b: Make the first two days discussion-only (trial)

Note I have just added an alternative, proposing a 2+5-day RfA trial instead of the 3+7-day trial originally proposed. Usedtobecool ☎️ 03:00, 20 February 2024 (UTC)

Pinging people who had commented/!voted already: @Barkeep49, Isaacl, Aaron Liu, Schazmjd, QuicoleJR, JPxG, Ahecht, Enos733, RoySmith, Lepricavark, Seraphimblade, S Marshall, SportingFlyer, NightWolf1223, Mach61, Lightburst, Thebiguglyalien, HouseBlaster, Relativity, Theleekycauldron, SilkTork, 0xDeadbeef, Rhododendrites, Leijurv, Femke, Bilorv, Retswerb, Kusma, Firefly, Serial Number 54129, Szmenderowiecki, Cremastra, AirshipJungleman29, Schierbecker, Voorts, WereSpielChequers, Compassionate727, Ixtal, WaltCip, Bruxton, Soni, MarioGom, Lulfas, Stanistani, Mackensen, and RunningTiger123: Usedtobecool ☎️ 03:00, 20 February 2024 (UTC) Also, @North8000, Novem Linguae, Lightoil, Lee Vilenski, Yngvadottir, Vanamonde93, Wbm1058, Galobtter, FOARP, Andrew Davidson, Tryptofish, Asilvering, Sideswipe9th, Hey man im josh, Fastily, and Schazjmd: Usedtobecool ☎️ 03:08, 20 February 2024 (UTC)

note: this was originally a subsection of the above section. I've moved it out here to match the format of the rest of this RfC :) theleekycauldron (talk • she/her) 08:15, 20 February 2024 (UTC)

Support (proposal 3b)

1. Support as this one does not extend the length of the RFA (per concerns expressed by many participants both supporting and opposing the original proposal and discussion in the general comments section). Usedtobecool ☎️ 03:00, 20 February 2024 (UTC)
2. Support (first choice) - no need to extend it. — Rhododendrites ^talk \\ 03:03, 20 February 2024 (UTC)
3. Support (first choice) 2 days for discussion and 5 days for voting seems to be the strictly superior version of the original proposal. It shortens the RFA's voting period as well, which I consider it to be a potential win. Soni (talk) 03:05, 20 February 2024 (UTC)
4. Support (first choice). Bruxton (talk) 03:08, 20 February 2024 (UTC)
5. Support (first choice) Leijurv (talk) 03:12, 20 February 2024 (UTC)
6. Strong support (first choice). If it's too little time then we can always adjust and extend. We can simply ask how grueling the candidates felt this process was. We don't have infinitely many RfA candidates to choose from in this year of experiments, and this doesn't introduce too many new factors either. Apaprently, my Waterfox browser really hates edit conflicts and basically hung when the edit conflict page appeared. Aaron Liu (talk) 03:18, 20 February 2024 (UTC)
7. Support (first choice) Lightburst (talk) 03:27, 20 February 2024 (UTC)
8. Support (first choice) Christ people, live a little. ~~ AirshipJungleman29 (talk) 03:31, 20 February 2024 (UTC)
9. Support per my initial !vote. I think this is arguably more similar to the current setup – yes, we're changing the voting window in addition to adding the discussion-only period, but 3+7 changes the overall window. RunningTiger123 (talk) 04:26, 20 February 2024 (UTC)

10. Support (first choice) per Aaron Liu. Namely, we can ask candidates how the process felt, so I don't see this as losing data.

    The first problem I see this solving is the rote +1 votes at the beginning. But more importantly, I see this as an opportunity for expressing concerns while allowing the candidate to respond to them. That way, there is no pile-on, and nobody feels wedded to a !vote they cast (i.e. you can easily express a concern without outright opposing). HouseBlaster (talk · he/him) 05:27, 20 February 2024 (UTC)

11. Support, mildly. I'm not sure if this is a good idea or not, but I really do want something to be tried in terms of RfA reform (I wasn't too happy about opposing the initial proposal, but the idea of extending RfA and the stress period was untenable for me), and I'm less concerned about additional stress for the candidate now that the total length of the period is the same. Galobtter (talk) 05:35, 20 February 2024 (UTC)
12. Support - striking my support above. Schierbecker (talk) 05:56, 20 February 2024 (UTC)
13. Support both – both are worth trying, I don't think we need to only pilot one :) theleekycauldron (talk • she/her) 07:33, 20 February 2024 (UTC)
14. Support either; i'm not sure it's worth trying both (one after the other? at the same time?), but it certainly is worth trying something new. Happy days, ~ Lindsay^Hello 07:36, 20 February 2024 (UTC)
15. Support (first choice): Not elongating the process is a big plus for me in terms of stress for candidates. Usually any concerns are expressed in the first 2/3 days, so that this gives the opportunity to respond without pile-on opposes. I'm also happy to try out both. —Femke 🐦 (talk) 08:13, 20 February 2024 (UTC)
16. First choice. NW1223<Howl at me•My hunts> 09:35, 20 February 2024 (UTC)
17. Weak Support (second choice) I am not sure about decreasing voting time, but both proposals are worth trying. QuicoleJR (talk) 14:21, 20 February 2024 (UTC)
18. Support, as I support the above. We can have a separate discussion on the exact time frame afterwards. Z1720 (talk) 14:44, 20 February 2024 (UTC)
19. Yup.—S Marshall T/C 14:45, 20 February 2024 (UTC)
20. Support Either of these options is fine by me. Ritchie333 ^{(talk) (cont)} 16:38, 20 February 2024 (UTC)
21. Works for me, possibly better than 3+7. —Kusma (talk) 17:28, 20 February 2024 (UTC)
22. Conditional support as second choice only if proposal 3 fails. Sdkb ^talk 19:49, 20 February 2024 (UTC)
23. Support (first choice) Either this or 3 would be good, this one is preferable as it does not increase the overall length. Pinguinn 🐧 20:17, 20 February 2024 (UTC)
24. Support per my neutral !vote on 3. Queen of Hearts (talk • stalk • she/they) 03:43, 21 February 2024 (UTC)
25. Support - I like the idea of discussion before the typical driveby voting starts. - jc37 05:36, 21 February 2024 (UTC)
26. Support for all the same reasons I supported 3a, although this would be a second choice. --Ahecht (TALK
    PAGE) 21:45, 21 February 2024 (UTC)
27. Excellent idea; first choice (over 3A). --JBL (talk) 22:34, 21 February 2024 (UTC)
28. Support first choice— don't need to stress valuable contributors out in their RfAs more than needed. ‍ Relativity 02:02, 22 February 2024 (UTC)

Oppose (proposal 3b)

1. Oppose as it makes it harder to tell if the discussion period is useful or not compared to the other factors this introduces. Best, Barkeep49 (talk) 03:09, 20 February 2024 (UTC)
2. Oppose. It is unclear to me what problem this is trying to solve. The only change is that voters have to wait to vote, correct? There will still be 7 days of questions and answers? Supports and opposes will still have rationales, correct? If the problem is opposers sometimes have rationales that the community doesn't like and some folks think these opposes aren't being policed enough and this makes all RFAs more toxic, I don't see how this proposal fixes that. I predict this proposal will just create impatience for support voters who are ready to vote but are not yet allowed to, and create a traffic jam of votes on day 3 of the RFA. In fact, it may even decrease RFA total votes because RFA is essentially only open for 5 days instead of 7. –Novem Linguae (talk) 03:26, 20 February 2024 (UTC)

   Novem Linguae, speaking for myself, I wanted to support the original proposal just because everyone says we need to do something, and this is something, but I could not bring myself to it because I see absolutely no potential benefit to making RFA even longer, because my reading is the same, things don't stop at the end of three days; if it is devolving, it will continue to do so the full 10 days. What brings me around to actually wondering that this might make things better is recollection that a discussion-first oppose was aired in a recent candidate's talk page and ended up not making it to the actual RFA, though I will not link it here as it is not a shining example of people coming together. — Usedtobecool ☎️ 03:42, 20 February 2024 (UTC)

3. Weak oppose. I believe users who are active on sporadic doses should be able to vote, and shortening the voting period disenfranchises them. Moreover, there's good reason to think the 7 days of the voting period under the proposed system will be less toxic than normal; most rationales will have already been discussed during the discussion period, and voters will be less likely to switch their positions because of that, lessening the need for persuasion. Mach61 (talk) 06:34, 20 February 2024 (UTC)

   As said above, adminship should be no big deal, or at least not much bigger than ANI. I agree with Bilorv's comment below. These sporadic users can also bring up points in discussion to influence votes. Aaron Liu (talk) 14:22, 20 February 2024 (UTC)

   If it wasn't a big deal we wouldn't have the watchlist notification, and would give out adminship at WP:PERM. Do not confuse aspirations with reality Mach61 (talk) 14:51, 20 February 2024 (UTC)

   WP:NOBIGDEAL is a policy. It's obviously a bigger deal than PERM as abuse would be much more catastrophic, but we're not electing a governor either. I see no reason for wanting everyone to vote. Five days plus arguments gives enough of a sample size. Aaron Liu (talk) 14:54, 20 February 2024 (UTC)

4. Weak oppose There is value in the seven day voting period. In many areas, seven days of discussion is standard. Removing two days of voting may mean that certain people are disenfranchised, because they only have time or ability to participate on particular days. I would rather have a full trial of 3+7 before considering changes to the length of the RfA.
   Secondly, and fundamentally, I don't think the community has a good grasp of why RfA is toxic. Certainly some of that could just be inherent in having 100-200+ editors scrutinize every edit and every interaction. If that is the case, then no reform based on community consensus will solve that issue. If the problem is how discussions devolve, then the only meaningful solution is likely more moderation. But, my two cents is that we need to start by understanding our values and our end goal when evaluating our processes and procedures. To me our values are consensus building and our goal is to have trusted administrators. --Enos733 (talk) 07:28, 20 February 2024 (UTC)
5. Oppose Seems too intense as the two days would be a vital campaigning period ahead of the formal voting. Candidates would agonise over their answers to every question and anxiety would make it difficult for them to sleep. Supporters would use the discussion to hold election rallies, declaring their support with endorsements which would make a farce of the voting embargo. Andrew🐉(talk) 09:11, 20 February 2024 (UTC)
6. Oppose this has the large presumption that the "general discussion" should not count towards determining the overall consensus (only having the enumerated discussion section be used for that purpose as is the general current practice), as such oppose limiting the time that editors can participate in the deciding discussion to only 5 days - this is not an urgent process and should allow for ample opportunity for community input - not all editors are active every day. — xaosflux ^Talk 11:30, 20 February 2024 (UTC)

   I am confused as to the point in the "discussion counts towards consensus" part. Wouldn't that mean that 2+5 gives equal time for community input, as the total time of evaluated input is 7 days either way? Aaron Liu (talk) 00:33, 21 February 2024 (UTC)

   @Aaron Liu when looking at results of RFA's in general the community is very focused on the S/(S+O) ratio - and this seems to be reinforcing that the discussion shouldn't count towards the conclusion (otherwise what is the point of splitting it?). This isn't trying to split QA period from feedback period - it is trying to split the feedback in to two classes (Day 1-2 discussion that isn't allowed to have numbered entries, and Day 3-7 discussion that is). — xaosflux ^Talk 10:20, 21 February 2024 (UTC)

   From my interpretation, RfAs have almost always never factored in discussion for cases above 75%, and have only factored them in when the ratio is between 60% and 75%. This change would allow the support ratio to better reflect community discussion by virtue of eyes. Aaron Liu (talk) 14:34, 21 February 2024 (UTC)

7. Opppose so all this does is prevent some users (who may only use Wikipedia on a single day per week) from being able to cast a !vote. In terms of closing, the discussion is worth as much as a !vote anyway. Lee Vilenski ^{(talk • contribs)} 11:50, 20 February 2024 (UTC)
8. Mild oppose, but let's not get stuck at no consensus closures. Let's just pass at least either of these options, OK? We will see if 3+7 is more stressful, that's why we have trials. But if 2+5 passes instead, it's worth a try too. My personal preference is 3+7 basically per Novem Linguae and Andrew Davidson, but let's just pass anything to see if it works, and not be mired in "no consensus" closures because no proposal will have 2/3 support that is by and large considered the threshold to rough consensus. If this proposal gains more traction, consider this a support. Szmenderowiecki (talk) 12:00, 20 February 2024 (UTC)

   That's a bit of a politician's syllogism; there certainly could be an even better idea. — xaosflux ^Talk 15:57, 20 February 2024 (UTC)

   My point is, if what is proposed for trial is not frivolous or obviously bad, it's worth investigating; otherwise we will never know if it's better. I prefer the 3+7 format, but if ultimately 2+5 prevails, I'm OK with it because the point here is testing if the !voting-free period makes sense and works. As a test, I'm all for it. If it's successful, I will support making it permanent, whether it's 2+5 or 3+7. Szmenderowiecki (talk) 19:03, 20 February 2024 (UTC)

9. Per Novem Linguae. * Pppery * _{it has begun...} 17:12, 20 February 2024 (UTC)
10. Weak oppose. I'm reluctant to oppose simply giving something a try. But the more that we discuss these things, and the more that I think about it, the more I'm convinced that the problem really is that qualified candidates don't want to have their every uncharacteristic mistake pored over and amplified, making qualified candidates less inclined to step forward. It's not incivility or personal attacks. It's that normal people occasionally do things that are legitimately things to bring up in RfA opposes. Actual mistakes. And the real issue then becomes whether the mistake was part of a pattern, or was so bad as to be disqualifying – or whether it was a one-off and should not be disqualifying. We can't make rules about this, because such opposes are not inherently disruptive conduct, and the only solution is discussion, and seeing whether or not the oppose rationale gets traction. The kinds of things being proposed here cannot address that; maybe individual editors being convinced not to reflexively pile-on out of laziness to closely evaluate the situation, or out of virtue-signaling, might. I also share, at least to some extent, the concern expressed by others that not everyone can participate seven days a week, and I'm also not persuaded that we can temporally isolate the difficult parts of RfA to just some days of the process. --Tryptofish (talk) 20:26, 20 February 2024 (UTC)
11. 'Weak oppose per @Xaosflux. voorts (talk/contributions) 22:27, 20 February 2024 (UTC)
12. Per xaosflux and Lee Vilenski. —Compassionate727 ^(T·C) 23:11, 20 February 2024 (UTC)
13. Oppose I'm not in favor of shortening the voting period. LEPRICAVARK (talk) 03:07, 21 February 2024 (UTC)

Proposal 4: Prohibit threaded discussion (trial)

Following the trial of the discussion-first RfA (provided the proposal is enacted), a second trial of threaded-discussion RfA will run for six months or 5 Requests for Adminship (RFAs) which are not closed as SNOW or NOTNOW, whichever happens sooner. A RfC to then be set up to see if elements of either or both trials should be made permanent. In the no threaded-discussion RfA trial, the rules and procedures are to be followed normally, except that, other than 'Crats, nobody is allowed to comment below another user's vote or comment. Each user, including the candidate, may address issues or respond to other comments either after their own vote comment, or by creating their own sub-section in the General comments area. SilkTork (talk) 11:33, 18 February 2024 (UTC)

Support (proposal 4)

1. I was going to wait until the above Discussion-first trial had concluded before suggesting this, but I think it would be more helpful to set it up now to see if there is traction for the idea, so that if there is traction, the trials can be set up to follow one another, and then we have a RfC to see what, if any, ideas can be taken forward. SilkTork (talk) 11:36, 18 February 2024 (UTC)
2. Support this as well, though I think it would be better for the two trials to run concurrently. The endless bludgeoning of opposes and neutrals has got to stop and this seems like the only way to stop that. NW1223<Howl at me•My hunts> 13:17, 18 February 2024 (UTC)
3. Something needs to be done to make RfA less toxic, and I think this is a good start. — Ingenuity (talk • contribs) 14:44, 18 February 2024 (UTC)
4. This is an interesting option. My initial reaction to this idea was a bit meh, but upon reflection I think this might help a bit and moves us closer to the standard democratic process organisations have for selecting individuals for tasks (which is closed voting with a discussion about the candidate, rather than about a vote). There is a social dynamics when there are two options (support or oppose), which pits people against each other. By moving the discussion to the discussion section, you make that dynamics less prominent. —Femke 🐦 (talk) 15:20, 18 February 2024 (UTC)
5. Support per the idea behind WP:Thank you for your vote. I'd also be fine with this running concurrently with the previous trial. Thebiguglyalien (talk) 17:22, 18 February 2024 (UTC)
6. Consolidating discussion about a given concern will reduce repetition, making it easier for new commenters to join in or for previous commenters to catch up. It will also mitigate the demoralizing effect on candidates from seeing the same criticisms being discussed repeatedly in separate threads. I think there remains social pressure against oppose voters, as they can still be challenged and engaged, and it also makes it easier for multiple challengers to engage. The convenience of the challenged and an initial challenger is traded off for the greater convenience of everyone else following the discussion about a common concern. isaacl (talk) 18:07, 18 February 2024 (UTC)
7. Support. As an implementation procedure, there could be language on the page or in the section to remind editors to respond in the general comments section or the talk page. --Enos733 (talk) 18:49, 18 February 2024 (UTC)
8. Worth a shot :) I would encourage future RfA candidates to mess with the template on their RfA – if you can add something interesting to your RfA's structure that doesn't impair the community's ability to find consensus, ask for forgiveness, not permission. People have done it before! I wish I'd thought to do so before my second RfA. theleekycauldron (talk • she/her) 23:01, 18 February 2024 (UTC)
9. Support as a trial: almost anything is worth trying. It's possible that preventing threaded discussion will reduce the overall number of words wasted at RfA, as one could argue applies at various ArbCom pages. RfA has a tendency to create ludicrous overemphasis of minor events or details, completely unrelated to whether the candidate will be a net positive as an admin. I'm not sure whether this would exacerbate or alleviate this but we could find out by testing.
   My concern would be that false information or BLP and civility violations may remain unchallenged under this format, as bureaucrats and oversighters typically refuse to enforce such policies and no-one else feels qualified to (it took six days for a very prominent BLP-violating comment to be removed at a recent RfA). But such things should be challenged with removal rather than badgering anyway. — Bilorv (talk) 23:08, 18 February 2024 (UTC)
10. Weak support, basically per Bilorv and Femke. If the 'crats and oversighters enforce civility and keep BLP-violatey stuff out, this will work well. If they don't, it won't. -- asilvering (talk) 21:22, 19 February 2024 (UTC)
11. Middling support - I really find the bashing on oppose !votes unproductive and serves to heighten and draw attention to drama rather than reducing it in any way. I'd be happy to see discussions without it. FOARP (talk) 22:03, 19 February 2024 (UTC)
12. Yup.—S Marshall T/C 08:24, 20 February 2024 (UTC)
13. Support. I'm going to put this in bold because some people opposing seem to have missed it - this is not shutting down discussion. It's just that I've found discussion about a candidate's suitability can happen in the General Comments area, or the RfA's talk page, or at WT:RFA, or on a user talk page, or (if appropriate at ANI), there are many, many, many areas where discussion is still open and welcome. The only thing that is being shut down is the worst possible place to put it. Ritchie333 ^{(talk) (cont)} 08:54, 20 February 2024 (UTC)
14. Until we can get to a private vote - this should be relegated to the talk page. Lightburst (talk) 22:38, 20 February 2024 (UTC)
15. Support only if 3/3b passes. Threaded discussion before the !vote. But oppose if neither 3 nor 3b pass. Queen of Hearts (talk • stalk • she/they) 03:47, 21 February 2024 (UTC)

Oppose (proposal 4)

1. This just makes the discussion impossible to follow. If replies to votes are outlawed, then voting comments should be restricted to seven letters instead of allowing the posting of out-of-context character assassination diffs without a right to highlight a reply. —Kusma (talk) 14:19, 18 February 2024 (UTC)

   This is the procedure used in ArbCom cases. It works there. SilkTork (talk) 15:20, 18 February 2024 (UTC)

   This is not at all comparable to an Arbcom case. Voters will not be added as parties whose behaviour is scrutinised and will not be subject to sanctions if they misbehave. —Kusma (talk) 16:03, 18 February 2024 (UTC)

2. I think this will make it easier for people who oppose while not offrring any benefit to candidates. I think the social pressure against opposing is, on the whole, a positive. For me, fixing RfA isn't an ends of its own but a means to the ends of getting more admin. Since I think this gets us farther from that I am oppoosed. Best, Barkeep49 (talk) 15:25, 18 February 2024 (UTC)
3. Per Barkeep. This will result in an increase in non-substantive opposes for reasons not relevant to the candidate's suitability for adminship. We should not be rewriting our procedures in a way that will primarily protect disruptive editors. LEPRICAVARK (talk) 18:29, 18 February 2024 (UTC)
4. Per above. This does much more than consolidating comments; it makes them harder to follow. Aaron Liu (talk) 19:58, 18 February 2024 (UTC)
5. Per Aaron and Kusma. Regarding the Arbcom comparison, Arbcom also has much lower participation than RfA. voorts (talk/contributions) 20:34, 18 February 2024 (UTC)
6. Agree with Kusma. * Pppery * _{it has begun...} 20:53, 18 February 2024 (UTC)
7. No, this just makes discussions harder to follow, and breaks anyone trying to use discussion subscriptions. — xaosflux ^Talk 21:43, 18 February 2024 (UTC)

   Does it break subscriptions? I'm under the impression that since only level-2 headings can be subscribed to, you can't subscribe in the first place. Aaron Liu (talk) 21:50, 18 February 2024 (UTC)

   @Aaron Liu thanks for note, so not yet then - but it is being worked on (phab:T275943). — xaosflux ^Talk 11:38, 19 February 2024 (UTC)

   At present the request is to implement subsection-specific subscriptions, though, and not individual numbered items in a list. While I imagine the same underlying implementation can be re-used for lists, I'm not sure the resulting tradeoff of additional markup would be worthwhile given the lesser frequency where subscribing to list items would be a preferable option. isaacl (talk) 18:35, 19 February 2024 (UTC)

   I've stricken the subsection related part, my primary concern is that I still think it makes following the overall discussion difficult. — xaosflux ^Talk 10:55, 20 February 2024 (UTC)

8. I agree entirely with Kusma. I think the core idea here is a good one: why not prohibited threaded discussion altogether? Any discussion consisting of more than one reply to a bolded !vote must happen on the talk page (or elsewhere). Vanamonde93 (talk) 23:07, 18 February 2024 (UTC)
9. I think threaded opposes are what RfA voters perceive as toxicity. But as someone who had a controversial RfA I don't think the endless RfA back and forths matter much to the candidate - the hard part really is having 200+ people scrutinizing you (and every response to every question asked) and many people writing negative things about you (and it seems often, being able to cast aspersions and make personal attacks without anyone doing anything about it), and this doesn't really help with that. In fact I mostly appreciated the people who defended me in the oppose section of my RfA.
   I wish honestly that people removed bad opposes if they are personal attacks/aspersions, rather than endlessly engaging with them, but I don't know if a rule can be made to solve this (other than encouraging more RfA clerking and enforcement of civility norms at RfA). Galobtter (talk) 06:42, 19 February 2024 (UTC)

   +1. I think many of these proposals, while well-intentioned, are conflating RFA heatedness with candidate stress, and that's not a valid equivalence. If there's a conduct problem with the opposition, we should deal with that directly; and if there's a conduct issue with responses to opposition, we should also deal with that directly. Vanamonde93 (talk) 06:05, 20 February 2024 (UTC)

10. Let's just focus on implementing one idea at a time. The better is the enemy of the good. _{Duly signed,} ⛵ WaltClipper -(talk) 13:33, 19 February 2024 (UTC)
11. Per Waltclipper and Vanamonde93. I think this proposal has "an" idea, but it's not sufficiently baked. It does not go far enough to remove threaded discussion (regardless of if those are positive or not) but also risks making following discussions impossible. Soni (talk) 14:36, 19 February 2024 (UTC)
12. Mainly per Kusma. It might accomplish what it's supposed to accomplish at ArbCom, but it absolutely makes many arb cases impossible for a passerby to follow. In an arbitration case, there are far fewer people talking and most of the discussion is for the sake of an even smaller number of people who have to judge what's said. At RfA, it's in every participant's interest to be able to understand what's being said, and most people aren't following the whole thing closely. — Rhododendrites ^talk \\ 16:29, 19 February 2024 (UTC)
13. As mentioned above, following a conversation would be near-impossible. This would probably decrease stress on opposers since they'll likely be badgered less but increase stress on the candidates by increasing the number of petty, POINTy, and poorly thought-out opposes. Sincerely, Novo Tape^{My Talk Page} 19:39, 19 February 2024 (UTC)
14. Maybe something like this could be made to work, but I really believe that we should go in the direction of more discussion, not less. (Yes, I know that may sound strange, in the context of feeling that RfA discussions veer towards unpleasant piling on, as well as there being far too many "optional" questions.) If we want to curtail spurious arguments, we need to give the crats a basis to weigh different comments differently. --Tryptofish (talk) 19:56, 19 February 2024 (UTC)
15. Oppose as premature. Let's let the other discussion play out first please -Fastily 01:29, 20 February 2024 (UTC)
16. Oppose. I feel this RFC is too early. I think it would make sense to use data from the first RFC and data from the first RFC's 5 RFAs to inform our decisions for the second RFC. –Novem Linguae (talk) 03:36, 20 February 2024 (UTC)
17. Oppose. RfAs should be more discussion and less vote, not the other way around. Seraphimblade ^{Talk to me} 07:17, 20 February 2024 (UTC)
18. Oppose. If you are willing to make a criticism of a candidate, you ought to be sure enough to stand some criticism yourself. I like to see opposes (and supports) discussed. JMCHutchinson (talk) 07:18, 20 February 2024 (UTC)

    As I said in my comment above, you can discuss in the General Comments area, or the RFA's talk page, or even the talk page of the editor who opposed. Lots of places to do it. Ritchie333 ^{(talk) (cont)} 09:07, 20 February 2024 (UTC)

19. ArbCom discussions are already in a pretty unweildy format, but at least those have the justification of making it less likely people will go over the word limit. There is no word limit in RfA; all that's happened is making discussion harder to follow. I don't think this would make participants more civil, either. Mach61 (talk) 13:57, 20 February 2024 (UTC)
20. Oppose Any critique should be able to be rebutted if inaccurate or biased, though ideally not in a badgering way. The candidate themselves should also be able to respond, as seeing how a candidate responds to criticism could be potentially important in judging their temperament. Pinguinn 🐧 20:24, 20 February 2024 (UTC)
21. Per Kusma. —Compassionate727 ^(T·C) 22:59, 20 February 2024 (UTC)
22. Strong oppose - it's a discussion. - jc37 05:36, 21 February 2024 (UTC)
23. Oppose an unthreaded discussion is just !voting without a bold word at the front. --Ahecht (TALK
    PAGE) 21:47, 21 February 2024 (UTC)
24. Oppose - that sounds like a bad idea. If someone says "oppose because he is a serial killer", I think everyone would be fine with simply striking or removing that blatant falsehood. But what if someone says "oppose I don't like his attitude because he was mean to me and a bunch of other people"? We need a way that context can be provided. Maybe being "mean" just meant nominating their advertisement articles for deletion. Or maybe the candidate really is mean and some diffs need to be provided. But either way, context needs to be provided. --B (talk) 13:27, 22 February 2024 (UTC)

Neutral (proposal 4)

1. On the one hand, I agree with Bilorv that almost anything is worth trying, and since it is a non-frivolous idea, I can't oppose it. But the idea itself I think has little possibility of success. RfA is not ArbCom. This proposal does not see any restrictions on how much one can say in discussions, so I imagine there will be something like "Oppose, see diff" and a long tirade about all ills that that candidate allegedly caused.

It may be a matter of bad taste to reply to that criticism, but the candidate should have that possibility regardless of customs around RfA. When there are 20-25 opposes to handle, it becomes too difficult to manage it in one section, plus, as commenters above said, the flow will be severely impaired. The structure is also incompatible with RfA's aims. It is important in ArbCom because it's easier to police individual users against excesses and it lets ArbCom understand what are the parties', amici and arb positions. RfA is more about discussing the way candidates would handle admin issues and not about the !voters or the positions of power users/accused parties/whatever. Structuring that by editors increases the risk of certain users "stealing the show" when what we should be doing is minimising it and making sure it's about assessing candidates on their merits. Szmenderowiecki (talk) 01:29, 19 February 2024 (UTC)

1. I would support a ban on threaded discussion under supports/opposes/neutrals, and ask that the discussion take place in a dedicated discussions section or on the RfA talk page. Z1720 (talk) 14:38, 20 February 2024 (UTC)

General discussion (proposal 4)

Two quick questions: this will be a 7-day RfA vote with immediate voting but with comments grouped by editor or it will be a ten-day RfA? Secondly, will the candidate be able to reply to a user's remarks in the user's section? Because unlike ArbCom cases, RfAs are dynamic, particularly if the outcome is close, and realistically even if 10% of the usual 200-ish users start their own discussion subsections, replies to each of the users will be very hard to track. Szmenderowiecki (talk) 15:28, 18 February 2024 (UTC)

The intention is that this trial will utilise the usual seven days. If both these proposals are trialled, then the intention is that there would be a discussion on how each went, and the positive aspects utilised permanently, so we could end up with a 10 day RfA without threaded-discussions.

Candidates would comment in their own section. While candidates are currently not discouraged from responding directly to vote comments in their RfA, it is generally regarded as not a good idea. When it is seen that a candidate might benefit from responding to a negative criticism, the usual thing is that someone will write a question inviting the candidate to respond. I would expect that practise to continue, and for candidates to confine themselves to answering questions in the question section. SilkTork (talk) 16:38, 18 February 2024 (UTC)

Also, it may very well be that one change doesn't make much difference but two in conjunction will, so if the discussion-only trial is a success, maybe let's not revert back and just see if adding threaded discussions further improves RfA? Szmenderowiecki (talk) 15:32, 18 February 2024 (UTC)

I think it would be acceptable to set this trial up after the RfC following Barkeep's trial. I did consider that initially, but what gave me pause was that people can become RfC weary when it involves the same topic. I think there may be some benefit in having just the one RfC after both trials, rather than having two RfCs - one after each trial. But it could work either way, and people may indicate their preference when they comment. SilkTork (talk) 16:38, 18 February 2024 (UTC)

The Spanish Wikipedia does something similar: votes can only have a short comment (15 words max), and all questions to the candidate and extended discussion happen on the talk page. I am still undecided on its effectiveness (after all, there was only one successful RFA in all of 2023), but it's a point to consider, and I have certainly noticed that the discussion focuses more on the candidate than the voters. –FlyingAce^✈hello 16:02, 18 February 2024 (UTC)

Let's say that, in my own comment in an RfA of this sort, I want to say something about what another editor has said. If I understand the proposal correctly, I would say that within my own comment, as opposed to saying it in a threaded reply to the other editor. If I think about how this format has worked at WP:AE, where it is also used, it often leads to walls of text that end up being limited when admins enforce a word limit, but it doesn't reduce the back-and-forth sniping. At ArbCom, the format works better because, in part, editors are required to address their comments to Arbs or clerks, not one another, which would not be workable at RfA. Also, both ArbCom and AE use individual sections for each editor, whereas there would have to be a hundred-plus sections if we were to do actual sections at RfA. --Tryptofish (talk) 22:05, 18 February 2024 (UTC)

Conversation among editors can be held in the general comments section, much like is being done in this discussion. isaacl (talk) 22:34, 18 February 2024 (UTC)

Seems to me these proposals are all too complicated. The principle should be to discuss the candidate first and then vote after the discussion -- not during the discussion.

A sequence of events might be: (1) A nomination and seconds of the candidate to become an administrator followed by a statement by the candidate in which they tell us why they should be an administrator (with a maximum length of, say, 500 words). (2) Discussion of the candidate for, say, 5 days. Commentators may present pro or con evidence in favor of the candidate. "Evidence" is the key word here. You can't just say I don't like this candidate because they were mean to me; you have to present specific examples of said incivility or mistakes. Nor can you just say that the candidate is a good person; you have to present specific examples of what they have done that was good. The candidate and other people may respond, if they wish to address any of the comments. (3) Vote, for say 3 days. Support or oppose only. No comments allowed during the vote.

What that system would do would be to ensure that people have more information about the candidate before the vote begins. It would also encourage the candidate to participate directly in the discussion about their candidacy. Smallchief (talk) 00:14, 19 February 2024 (UTC)

I get the concerns about organization. I'm bothered by the arguments that it should be harder to oppose an RfA or that we should maintain a chilling effect on potential opposers. Thebiguglyalien (talk) 17:47, 20 February 2024 (UTC)

Proposal 5: Add option for header to support limited-time adminship (trial)

Following the passage of this proposal, candidates will have the option to include a header between Support and Oppose that will read Support adminship for a year. When closing the RfA, bureaucrats are to grant a year-long term of adminship only when there is no consensus to make a candidate an admin indefinitely, but the combined strength of both support sections does achieve consensus. After this has been tested on five RfAs that are not closed as SNOW or NOTNOW – or six months, whichever is sooner – this trial will end. No candidate may use this option twice.

Support (proposal 5)

1. An opportunity to encourage candidates to approach RfA and succeed much more easily. This nearly gained consensus two and a half years ago, but was instead closed as "no consensus" primarily due to objections that the option might become the de facto standard: admins would have to seek a limited adminship or else not get elected. This proposal addresses this in two ways: first, the community has the option to elect a candidate to an indefinite term should they feel that a probationary term is unnecessary; the candidates who pull 99% at RfA aren't gonna get dragged into having to run again in a year. Having a section for each option is a much better system than having to make the candidate choose at the outset, but the candidate does still get to run a traditional RfA if they want to. Second, it's a trial run, so the potential risks are inherently limited. As I've said for the other fine proposals made thus far, this is worth a shot. theleekycauldron (talk • she/her) 09:01, 20 February 2024 (UTC)
2. Support - it will give editors that we, as a community, are on the fence about a chance to demonstrate that they are suitable admin material. However, I'm not convinced by No candidate may use this option twice - if an admin is rejected narrowly once with this option, I don't see why it wouldn't be appropriate for them to run a year later with this option again. BilledMammal (talk) 10:57, 20 February 2024 (UTC)
3. Sure, but would want to revisit the one-time-limit conditions if this goes past trial (timing and limit things get odd when stretched to indefinite). — xaosflux ^Talk 11:04, 20 February 2024 (UTC)
4. Support it similarly to the 2021 proposal similar to this. I expect some opposes to the proposal will be concerned about the gaming of it. — ♠Ixtal ^{( T / C )} ⁂ Non nobis solum. ♠ 11:40, 20 February 2024 (UTC)
5. Seems good' for !voters who support but are a bit wary. This could lead to vote-splitting making the RfA's more difficult for 'crats to close, but we'll see once the trial ends. 🌺 Cremastra (talk) 13:22, 20 February 2024 (UTC)
6. Support if this means the reconfirmation RfA after one year is limited to Support/Oppose, and not that the candidate may opt-in for a "Support for one year" option at their RfA but may never do that on their next try. Because the last sentence of this proposal can be read both ways. I think whether a trial admin is qualified should already be apparent after a year. This is better-worded than the 2021 proposal. Also, bureaucrats will start to be at least somewhat meaningful and not simply rubber stamps. Szmenderowiecki (talk) 13:26, 20 February 2024 (UTC)
7. I don't think this will actually work, but if other RfA regulars do there's no harm in letting them try it. * Pppery * _{it has begun...} 17:15, 20 February 2024 (UTC)
8. Support for the same reasons we do this for, by my count, two other userrights (NPP and PM, which for disclosure, I went through the trial periods before being granted both). Grant them the right, see how they do, and reevaluate after a certain time. Queen of Hearts (talk • stalk • she/they) 03:53, 21 February 2024 (UTC)
9. I think it would be good to have a trial where candidates can volunteer for a fixed-length term to perform administrative tasks. In a world where the number of choices for online pastimes is increasing, the time between the start and end of being an active contributor is shorter for new editors than before. Getting people to sign up for thankless, tedious administrative work is hard. I think new volunteers might be encouraged by knowing they are only committing to helping out for a fixed period, after which they can resume spending that time on other pursuits. isaacl (talk) 05:05, 21 February 2024 (UTC)
10. Strong support. When we look at an admin candidate we look for their commitment into contributing to the behind-the-scenes parts of Wikipedia. Sometimes candidates get opposes simply for doing not enough work, or more specifically not doing enough work in the areas that they want to work in. This happened for my RfA, and I can imagine it happening to future RfAs quite often. It is perfectly normal for admin candidates to do completely different things after they pass RfA, because the admin tasks are simply different from the non-admin tasks. (I've recently picked up clerking at SPI, something that I did not do for almost a year leading to my RfA) Giving these candidates temporary adminship would give them an opportunity to show that they would be helpful with the tools. It would be helpful for us to have a position that says "maybe not yet for getting the toolset indefinitely, but the candidate definitely has my trust and I don't think they will abuse the tools". Also, since this is a trial, so why not? 0xDeadbeef→∞ (talk to me) 12:24, 21 February 2024 (UTC)
11. Support. I was on the fence about this, but IMO the opposition to this so far has not been very convinving. Mach61 (talk) 19:47, 21 February 2024 (UTC)
12. An RfA that was based on actual track-record would be much, much, much better than the present mess; any process that allows developing a track-record before going through an RfA is good. --JBL (talk) 22:35, 21 February 2024 (UTC)
13. I've always thought this would be a good idea. How many people have we seen be on their best behavior, then get the bit, then turn into petty tyrants? Sure, nothing is stopping them from turning into petty tyrants after the year is up, but this would be something. --B (talk) 13:31, 22 February 2024 (UTC)

Oppose (proposal 5)

1. I am oppose right now for a couple reasons. One is that it will create a two tiers admin: those that are controversial in their RfA (and have a one-year limit) and those that have the full trust right away. I would rather this be implemented for all admin (like Stewards, who have to go through re-confirmation) or not at all (though I do not recommend reconfirmations for admin).

   I also think that this creates more bureaucracy for the community, where after one year the admin will have to go through another RfA to make their one-year limit permanent, in which over 200 editors will comment. I think the community's time is better spent writing articles and making the site better, and the candidate's time is better spent doing admin tasks than defending themselves at a reconfirmation.

   Tagging on to the above, I am scared of this becoming the default for admin, where all admin candidates, even uncontroversial ones, will have this limit placed on them, increasing Wikipedia's bureaucracy.

   Lastly, if an admin has a one-year limit, it might prevent them from making controversial decisions because they do not want to endanger their reconfirmation. I want admin to feel confident in helping out in controversial areas when they are ready, and telling an editor that they will be scrutinised in a year will prevent them from helping. Z1720 (talk) 14:35, 20 February 2024 (UTC)

   No comment on the other concerns, but I find the "community time" argument uncompelling. It's metapedians who self-select their participation in RfA; only the candidate is required to pay attention during the entirety of its duration, and they can control when they want to run. Mach61 (talk) 14:48, 20 February 2024 (UTC)

   I agree that editors self-select their participation. However, I think it is a determent to the project for over 200 editors to self-select to participate in an initial RfA, and then self-select to participate in a reconfirmation (especially because I know some editors spend hours researching a candidate that could be used to improve the site.) Z1720 (talk) 15:08, 20 February 2024 (UTC)

   I mean, I guess my thought on that is that we're looking to encourage more candidates to run anyway – so if we're talking community time, an increase of 5 new candidates per year is probably a bigger time sink (a good one) than the one or two reconfirmation RfAs that stem from them. theleekycauldron (talk • she/her) 19:45, 20 February 2024 (UTC)

2. I fail to see the point. If I trust someone to have the tools, I trust them to have the tools. If I don't, I don't, be that for an hour or a year. Also at that point, we are asking the candidate to go through RfA twice, and if we're having a hard time getting people to do it even once, I certainly do not see that as an improvement. Seraphimblade ^{Talk to me} 18:12, 20 February 2024 (UTC)
3. Oppose per Z1720. Admins should feel empowered to take the actions that they deem necessary. An admin who was approved for a 1-year term only would be discouraged from taking any controversial actions knowing that they would come up again at the end of the year. When we elect admins, we elect them with no restrictions on what kinds of actions they can do, up to and including potentially controversial ones. I don't see a reason for that to change, and it certainly shouldn't change in this discussion about RFA. If we want to create a more limited kind of admin role, that's a different kind of discussion altogether. Pinguinn 🐧 20:33, 20 February 2024 (UTC)

   @Pinguinn: if an admin doesn't want to be held to a time limit, they can run their RfA without that section. I certainly imagine a lot of more headstrong candidates would rather not be held to this, and would simply opt out of the trial. theleekycauldron (talk • she/her) 20:57, 20 February 2024 (UTC)

4. The part of this I don't like is actually the part about having a separate section. I'd rather see editors given a choice between the existing three options (support, oppose, neutral), and for those who feel this way to explain it as part of their regular comments. However, I could support having the crats be able to decide to grant 1-year adminship in close cases (a little like ArbCom candidates who only get a one-year term), based on a borderline discussion, maybe as an option to be reached during a crat-chat. I also oppose the formulation of only being able to do it once. --Tryptofish (talk) 20:36, 20 February 2024 (UTC)
5. Oppose per @Seraphimblade and @Z1720. voorts (talk/contributions) 22:29, 20 February 2024 (UTC)
6. Oppose this seems like extra bureaucracy. LEPRICAVARK (talk) 03:10, 21 February 2024 (UTC)
7. Oppose – Z1720 and Seraphimblade both make very strong points. I particularly agree that 1-year admins may avoid controversial decisions (which are the areas admins should be able to help the most) and that if someone is not trusted with the admin tools, limiting their time with them won't change that (1 year is still plenty of time for "bad" decisions, whatever "bad" means). RunningTiger123 (talk) 03:37, 21 February 2024 (UTC)
8. Oppose - Consider that 1 year is all of only half an Arbcom term! And besides, the limit should be on what tools are given, not on length of service. - jc37 05:36, 21 February 2024 (UTC)
9. Oppose because it wouldn't really be optional (any candidate not including that header would be get oppose votes for not doing so), and I imagine that most voters would opt to vote for the 1 year option anyway for all but the most well-known candidates. Either made adminship always start with a 1-year probationary period (which I wouldn't necessarily oppose) or don't, but don't pretend that it's optional. --Ahecht (TALK
   PAGE) 21:51, 21 February 2024 (UTC)
10. Oppose creates a further cumbersome process; probationary period will not reveal anything not already understood, possibly likely to incenctivise limited engagement. Regards, --Goldsztajn (talk) 08:57, 22 February 2024 (UTC)

Discussion (proposal 5)

I have a feeling that this could make it harder for people working under controversial areas or would have a controversial RfA to get the bit without expiry. It is quite possible we'd get more single year admins from otherwise discretionary or less than the cut off, but in the other direction, this could push admins who would otherwise pass to be pushed towards single year. With all that said, what is proposed is a trial, so it would be many times better to see what actually happens than to speculate. 0xDeadbeef→∞ (talk to me) 12:13, 20 February 2024 (UTC)

Then it would work as intended. Candidates who only get adminiship for a year should better avoid controversial areas, since the community did not have full trust in them. — kashmīrī ^TALK 22:51, 20 February 2024 (UTC)

Additionally, the provisional administrators will also have relatively low experience (probably very few would be given the bit provisionally more than once, as the one year should help the community understand better if this user should really be an administrator). We want experienced administrators to handle the controversial areas. Animal lover |666| 23:29, 20 February 2024 (UTC)

Proposal 6: Provisional adminship via sortition

Following the passage of this RfC the position of "Provisional Admin" will be created, to be selected by sortition at monthly intervals.

Provisional adminship

Provisional admins serve for a period of one year. During this year they may perform any activity a full admin performs, with the exception of enforcing discretionary sanctions or participating as an admin at WP:AE. In addition, full admins are permitted to revert the actions of provisional admins even when they would otherwise be forbidden from doing so by WP:WHEEL.

A provisional admin may at any point elect to go through WP:RfA; should they pass the process they will be permanently granted full adminship. Should they fail their provisional adminship will be revoked.

Provisional adminship may also be revoked at any time by a consensus of full admins at WP:AN, or by WP:ARBCOM.

Selection

Provisional admins are selected by sortition, with an editor who meets the following criteria being drawn once a month by the bureaucrats:

1. At least 10,000 total edits, including at least 5,000 in main space
2. At least 1,000 edits in the past year, including at least 500 in main space
3. Account registered at least three years ago
4. No sanctions within the past five years
5. At least one featured article or three good articles
6. Have never lost adminship under a cloud
7. Have never previously held provisional adminship

Following the passage of this RfC ArbCom may, as a one-time action, unilaterally alter these criteria before the process goes into effect. Future modifications to these criteria would need to be done through standard processes.

Prior to the drawn editor being announced, the bureaucrats shall inform ArbCom of the editors identity. ArbCom may, if they deem it necessary, silently veto the editor, in which case a new drawing shall occur.

A drawn editor may decline provisional adminship, in which case a new drawing shall occur. 10:54, 20 February 2024 (UTC)

Support (Proposal 6)

1. This proposal is bold, but I believe it is worth considering as the sort of drastic action necessary to revamp the admin process:
   1. I believe it will result in us gaining excellent admins we would never have otherwise gained due to them having been unwilling to run for various reasons
   2. I believe that it will simplify the RfA process for those admins who do a good job as a provisional admin; I hope that we will be less critical of candidates who have already demonstrated they are competent with the tools.
   3. I believe it will provide a strong boost to the number of admins; even if only 50% convert from provisional to full that is still an extra six a year, and if sortition is discovered to be functional we can increase the number of editors selected each month.

   There is the risk that an unsuitable editor will gain the tools, but I believe there are sufficient safe guards in this proposal to prevent harm coming from that.

   One concern I anticipate is whether the WMF will accept this process as sufficient for granting admin rights. My hope is that ArbCom's veto will be sufficient to push the process over their line - but I would also hope that editors avoid opposing on that basis, as that decision is out of our hands and it makes little sense to try to guess at what it may be. BilledMammal (talk) 10:54, 20 February 2024 (UTC)

2. Hell yeah this is the type of bold proposal we need. I think the criteria is strict enough. — ♠Ixtal ^{( T / C )} ⁂ Non nobis solum. ♠ 11:41, 20 February 2024 (UTC)
3. Sounds good. The proposal are certainly strict enough, maybe even too stringent, but it's best to err on the side of caution. 🌺 Cremastra (talk) 13:19, 20 February 2024 (UTC)
4. While the criteria are pretty stringent (the yearly activity criterion is IMHO a bit too much), I absolutely support this idea in general and not only because I may stand to benefit from it in the future (I have 5.5K edits, but two or three years will maybe get me to 10K). I have reservations about criterion 5 because it excludes "technical" editors and those editors who write articles where not much sourcing exists but are notable, so those articles will fail the broadness GA/FA threshold but which otherwise capture the topic pretty well (yeah, and also we don't require perfection, just consistent movement towards betterment). Also, you may create a lot of good articles but still be an arse to others, or you can be an editor doing background stuff who is very helpful but who will not qualify under that criterion. Or you can rescue GAs/FAs from delistment. That doesn't mean that we shouldn't expect content creation (that's what we are for here), but I think we need some flexibility in enforcing criterion 5 in these cases (which will also make bureaucrats actually somewhat relevant). Regardless, it is a brilliant idea and is worth implementing. Szmenderowiecki (talk) 13:42, 20 February 2024 (UTC)
5. This or any model that doesn't involve unhealthy "public performance reviews." Sortition will make better choices than RFA voters. My support vote should be counted as supporting not just exactly what's proposed but also any similar model, so even with adjustments to the criteria, etc. Levivich (talk) 14:22, 20 February 2024 (UTC)
6. Weak support: I like the essence of this proposal. I was going for neutral due to concerns about gaming, but criteria #5 should do. I do have a query: Maybe there should be a bit more clarification on what kinds of contributions count as "their" good article. Is it putting it up for review? Writing 50% or so of the page? Aaron Liu (talk) 14:28, 20 February 2024 (UTC)

   I also echo Xaosflux's concerns about the technical implementation of selecting a user partially based on whether they have a good article. Aaron Liu (talk) 15:00, 20 February 2024 (UTC)

7. I've come across quite a few good editors who would make sensible admins but are remarkably unwilling to stand. The suggested process might help in making this a civic duty, rather like jury service. It would also give greater diversity in the admin pool with better representation of those who are focussed on building the encyclopedia. As the proposed procedure is just one a month and is provisional, the details can be tweaked in the light of experience. Andrew🐉(talk) 08:35, 21 February 2024 (UTC)
8. An RfA that was based on actual track-record would be much, much, much better than the present mess; any process that allows developing a track-record before going through an RfA is good. Quibbling about the precise details of the criteria is beside the point. --JBL (talk) 22:32, 21 February 2024 (UTC)

Oppose (Proposal 6)

1. I don't love the content creation being directly tied to it, and more importantly, WMF Legal said that there has to be some form of vetting process for adminship, so we aren't allowed to do this. QuicoleJR (talk) 13:59, 20 February 2024 (UTC)

   Surely we should wait for a response from WMF legal before acting like this proposal is certain to be shut down from them? Mach61 (talk) 14:18, 20 February 2024 (UTC)

   This is a myth, don't believe it. Levivich (talk) 14:27, 20 February 2024 (UTC)

   I still don't love removing the vetting aspect entirely, especially since the GA/FA process is directly tied to it. I still oppose. QuicoleJR (talk) 15:06, 20 February 2024 (UTC)

   For the record I completely respect that. Putting the WMF aside, there are perfectly valid reasons not to be in favor of this proposal. Levivich (talk) 15:24, 20 February 2024 (UTC)

2. Strong oppose Absolutely not, nope. RfA needs reform, but not this - yes, this will boost adminship numbers and stuff, but this strips the community of the voice to choose whom they deem appropriate to gain adminship or not. Besides, there can be other extenuating factors other than edit count, content creation etc such as communication, knowledge of policy and such that makes users qualified for adminship or not. Not all admin has GAs or FAs - some prefer working with scripts and bots while others focus on copyright. Prodraxis _(talk) 14:04, 20 February 2024 (UTC)

   I'd disagree with the claim that this strips the community of the voice to choose whom they deem appropriate to gain adminship. If consensus is found for this proposal (and WMF Legal is fine with it), the community has ipso facto voiced whom it considers suitable for the mop (i.e. anyone, or almost anyone, who meets the 7 points enumerated).Sincerely, Novo Tape^{My Talk Page} 17:25, 20 February 2024 (UTC)

3. The "safe guards" aren't just insufficient; they're so paper-thin they don't deserve the name. Some of the most problematic admin actions can't "just be reverted" (trivial example: no one can make you unsee deleted content), and the Arbitration Committee's past record in overturning community bans is enough to show they can't be relied on to be the only people vetting potential admins. —Cryptic 14:21, 20 February 2024 (UTC)
4. Strong Oppose primarily we should not be removing the need for affirmative community support, as is required for contributors that will gain access to restricted functions such as deleted content and soon-to-be ip masking data. Additionally, this seems like it is going to require some sort of lottery commission to run the first component - a process which should have been more clearly defined before trying to vote on it. Furthermore, this seems to require technical development (building a secure lottery management tool) that hasn't been started. — xaosflux ^Talk 14:51, 20 February 2024 (UTC)
5. Strong oppose there are some editors who are not the best fit for adminship, and would not support their RfA. Adminship can easily damage the site (deleting the Main Page, blocking admin and editors, etc.) and giving the power to editors so easily could lead to severe damage to our reputation. Z1720 (talk) 15:03, 20 February 2024 (UTC)

   Nit: admins technically can't delete the main page. I see why we need to worry about provisional admins going rogue. * Pppery * _{it has begun...} 17:24, 20 February 2024 (UTC)

6. Strong oppose: aside from the many problems laid out above, we should not create a tier-system for admins, as we would end up inflating the standards for regular admins (when we should be lowering them). Also, a large majority of the people who meet those criteria don't want adminship, and have no experience with admin tasks; those who do would likely be successful at RFA anyway. Vanamonde93 (talk) 16:32, 20 February 2024 (UTC)

   I don't think this will raise the standard for adminship, because the community has no ability to make someone a provisional admin. Oppose please wait an unknown and unknowable number of months for the crats to draw your name out of a hat. - really? * Pppery * _{it has begun...} 17:24, 20 February 2024 (UTC)

   That's a fair point, Pppery, but my concern is more basically about standard inflation. If there are temporary admins via this process, the standard permanent admins are expected to meet will be higher. This is rather similar to why a lot of people oppose further unbundling. Vanamonde93 (talk) 21:05, 20 February 2024 (UTC)

7. This is unfair to those admins who don't meet the sortition criteria like myself. * Pppery * _{it has begun...} 17:24, 20 February 2024 (UTC)
8. Certainly not with criteria 5 being in place; that's nothing at all to do with adminship. One does not need the tools to write articles, nor does the ability to write articles speak to any suitability to be an admin. The remainder of the criteria are insufficient to substitute for actual community review of prospective candidates. Seraphimblade ^{Talk to me} 18:44, 20 February 2024 (UTC)

   Personally, I feel like criteria 5 is not to do with adminship but rather to prevent people from gaming the system to get sortitioned as admin and then wreak havoc across enwiki. It's not like everyone with that amount of edits can be vetted as admin, and creating that content demonstrates a clear understanding of editing policy. Aaron Liu (talk) 21:24, 20 February 2024 (UTC)

9. I simply can't support a process that would produce considerable numbers of admins that I (and, I think, much of the rest of the community) would oppose at RfA. Extraordinary Writ (talk) 19:54, 20 February 2024 (UTC)
10. I've been wondering whether something like this could work, but I think that more thought needs to go into the criteria, and into whether this is a role the bureaucrats should have, before this would be ready for prime time. --Tryptofish (talk) 20:33, 20 February 2024 (UTC)
11. On second thought, this would probably be a net negative without formal recall critera. Mach61 (talk) 20:36, 20 February 2024 (UTC)
12. Oppose not everyone can be an admin, and that's fine. No matter how strict the criteria become you're never going to be able to isolate a hypothetical "admin class" of users of which 100% would be good admins. Besides, not everyone wants to be admin in the first place. Whether a user at RFA is nominated by others or nominates themselves, it's at least proof that they understand the requirements of adminship and want to be one. Pinguinn 🐧 20:40, 20 February 2024 (UTC)
13. No. ~ ToBeFree (talk) 20:44, 20 February 2024 (UTC)
14. One year is too short a period, the criteria are too stringent, and I don't think selecting people who haven't thought hard about whether they want to commit to the mop is a particularly great way to select new admins. voorts (talk/contributions) 22:31, 20 February 2024 (UTC)

    I feel like being short was the point. Aaron Liu (talk) 00:27, 21 February 2024 (UTC)

15. Qualified oppose. The concept of sortition is definitely worth exploring. However, the candidate pool should be identified using other criteria than mere edit count and tenure. It's been stressed again and again that adminship is as much about wiki experience as it is about temperament, communication skills, and technical aptitude. Please come up with better criteria and I may support. — kashmīrī ^TALK 22:49, 20 February 2024 (UTC)

    @Kashmiri I am confused, wouldn't criteria #5 which is about content creation which has a lot of communication satisfy the skills you've said? Aaron Liu (talk) 00:27, 21 February 2024 (UTC)

    @Aaron Liu: Yeah, #5 is a somewhat different proposal (no sortition). I'm yet to make up my mind about it. — kashmīrī ^TALK 00:36, 21 February 2024 (UTC)

    Sorry about the confusion, I meant criteria (criterium?) #5 of this proposal, At least one featured article or three good articles. Aaron Liu (talk) 00:59, 21 February 2024 (UTC)

    For what it's worth, the singular is criterion, like in mitochondria/mitochondrion. Sincerely, Novo Tape^{My Talk Page} 01:55, 21 February 2024 (UTC)

    To me, regular participation in AfD, AN/I, and ARV would be much more relevant than having drafted a GA. We aren't looking for copywriters, are we. Also, 10,000 edits is the minimum threshold for full vetting, not for automatic assignments. I'd make it 20k at least. Finally, candidates for the pool should not perhaps have been part of major controversies or displayed CIV problems. Overall, as I wrote, criteria for the candidate pool need further work IMO. — kashmīrī ^TALK 10:36, 21 February 2024 (UTC)

16. Sortition seems very promising to me, but not on the terms proposed. —Compassionate727 ^(T·C) 23:15, 20 February 2024 (UTC)
17. Weak oppose - I think this idea is very interesting and worth discussing more, but I'm not sure there will ever be a form of criteria #5 that I would like. I definitely don't like it in its current form. It favors featured content creators at the expense of other editors who might have more experience in technical or project maintenance areas (which are arguably more important for prospective administrators), or who have created a lot of "good" content but have not gone through any formal featured content process. MaterialsPsych (talk) 23:35, 20 February 2024 (UTC)
18. Oppose community trust should be a required component of adminship. LEPRICAVARK (talk) 03:12, 21 February 2024 (UTC)
19. Oppose per Extraordinary Writ, currently oppose vote #9. The concern that's getting raised the most is regarding criteria 5 but folks? If that were removed, I'd be eligible for adminship under this. RFAs have been righteously sunk because of considerations that an algorithm like this simply would not be able to deduce. Someone with my temperament should never qualify for adminship and that's just one of countless things that only human beings, voting at RFA, would be able to suss out. City of Silver 03:38, 21 February 2024 (UTC)
20. Oppose per lack of vetting, if nothing else. IMO, every admin should have the confidence of the community. Queen of Hearts (talk • stalk • she/they) 03:57, 21 February 2024 (UTC)
21. I disagree with drafting editors without prior discussion to see if they are interested in taking on administrative work. It's a thankless role with tedious tasks that makes you a target of criticism. Let's not put editors under this scrutiny unless they've volunteered to do so. isaacl (talk) 04:49, 21 February 2024 (UTC)
22. Oppose - As above: "Oppose - Consider that 1 year is all of only half an Arbcom term! And besides, the limit should be on what tools are given, not on length of service." And also noting that this does not remove all the ways that an admin can assess behaviour. So this is a no. - jc37 05:36, 21 February 2024 (UTC)
23. Oppose - moral support for the bold suggestion (and following through on Douglas Adams's assertion that "anyone who is capable of getting themselves made [admin] should on no account be allowed to do the job", but I just can't see this meeting any sort of standard for community vetting, which is important whether or not WMF legal requires it. --Ahecht (TALK
    PAGE) 21:56, 21 February 2024 (UTC)
24. Oppose Hello D.R.A.M.A. Regards,--Goldsztajn (talk) 09:01, 22 February 2024 (UTC)

Discussion (Proposal 6)

• Does this suggestion simply mean we have a small section of users (crats) electing users who may or may not want the bit? I can think of at least 20 users who meet the criteria above who aren't interested in the toolset. Lee Vilenski ^{(talk • contribs)} 11:53, 20 February 2024 (UTC)

  Not quite - the crats don't get to choose the candidates, they randomly select them from the pool of eligible editors. Their role would be comparable to court clerks or jury commissioners in the jury system.

  This proposal would permit editors to reject the bit if their name is drawn, but I would hope that at least some of the twenty you have in mind would, considering that there is no obligation for them to go through a strenuous process or even use the bit, decide to give it a go for a year and see if having it benefits the project. BilledMammal (talk) 12:42, 20 February 2024 (UTC)

• On the topic of whether WMF would accept this, they did release a statement around ten years ago that they'd want a process like RfA. I think this proposal is different from the proposal in that discussion, and a considerable amount of time has passed, so I think we'd need WMF Legal to weigh in again. 0xDeadbeef→∞ (talk to me) 12:07, 20 February 2024 (UTC)

  We already went over this in 2021. It's fine. Levivich (talk) 14:17, 20 February 2024 (UTC)

  {{cite}} please. — xaosflux ^Talk 14:52, 20 February 2024 (UTC)

  Somewhere in WP:RFA2021. (Also this was discussed somewhere in the VPR discussion before everything was moved to this page but I don't know where that's been moved to.) To the broader point, going back to their 2014 statement, if WMF Legal has a problem with admins viewing certain deleted content, they can suppress the deleted content they're worried about. Restricting the admin selection process isn't the only way, or even a good way, to address legal's 2014 concerns (which is as I recall what they acknowledged in 2021, when they said they'd be open to alternatives). As for an even broader point, we shouldn't tolerate WMF Legal putting any constraints on our admin selection process in order to protect WMF from liability. What they're doing is outsourcing WMF liability protection to the community, which is wrong. If there is something that admins selected by this or a similar process shouldn't see, then WMF Legal should use suppression (aka oversight) for that content, as that's what those perms are for -- the "legal" stuff. Admins shouldn't be involved in any of the "legal" stuff, leave that for OSes who signed NDAs and let WMF Legal vet OSes however they want. Admins are here to serve the community not the WMF. Levivich (talk) 15:01, 20 February 2024 (UTC)

• Hmmm. Interesting idea, but I'm not quite on board with criteria #5. This seems to exclude "content creators" who, for one reason or another, have not or choose not to participate in featured content processes, even if they have contributed to a lot of content creation in other ways. Its seems to heavily favor certain types of editors over others. I think I can understand (or at least attempt to venture an educated guess about) the reasoning behind it, but I'd have to think about what other good alternatives there could be to this. MaterialsPsych (talk) 13:11, 20 February 2024 (UTC)
• Fascinating idea. Do we know or can we find out how many meet conditions 1 to 4 (I'm guessing it could be as low as 1,000)? That seems amenable to processing similar to the X-tools edit-counts analysis. Condition 5 doesn't, unless it applied only to page creators even if they did little to bring an article to GA/FA status and not to those who did substantial work on it. #6 would also require some assessment as to whether the bit was given up under a cloud. The crats may not eagerly take up either burden; I imagine they'd accept the output of a verified automatic process applying 1-4 and 7, might be persuaded to then evaluate 6, but wouldn't want to assess 5. NebY (talk) 14:02, 20 February 2024 (UTC)
• Would it make sense to add a minimum number of non-semi-automated edits? For example, "8. Must have at least 3,000 mainspace edits without using semiautomated tools such as Autowikibrowser or Huggle". Sincerely, Novo Tape^{My Talk Page} 16:46, 20 February 2024 (UTC)

  3000 is way too low IMO but I agree with the sentiment to show proper content creation experience. _Dial^mayo 18:43, 20 February 2024 (UTC)

• Shouldn't the 'crats also voice some kind of opinion, in addition to arbcom, since they'll be among the ones actually implementing this (assuming it gains consensus)? Sincerely, Novo Tape^{My Talk Page} 23:53, 20 February 2024 (UTC)
• I think any sortition proposal using a pool delineated by objective criteria like edit count is going to be a non-starter: the necessary qualities for a good administrator (e.g., temperament, familiarity with policies and processes) are too difficult to quantify, which reintroduces the need for vetting. However, I wonder if there isn't a relatively simple way to draw up an acceptable pool subjectively; for example, existing administrators could add promising editors to a list somewhere, and we could periodically sortition from that. —Compassionate727 ^(T·C) 23:57, 20 February 2024 (UTC)

  There's Wikipedia:Administrators without tools, though it's informal Aaron Liu (talk) 00:30, 21 February 2024 (UTC)

• Can someone with a bit more technical know-how than me please generate a list of who would currently meet those criteria? I'm guessing it's very small. Bonus point if you can tell us how well these criteria would predict someone passing an RfA over the last several years. I imagine someone who meets these is almost certain to pass. Ajpolino (talk) 02:12, 21 February 2024 (UTC)

  Maybe try WP:QUARRY? Sincerely, Novo Tape^{My Talk Page} 02:14, 21 February 2024 (UTC)

  A partial algorithm could probably involve scrolling down the category of featured articles, identity the creator, and see if they meet the rest of the criteria. I'd expect there to be quite a bit. Aaron Liu (talk) 03:52, 21 February 2024 (UTC)

• I don't know why the first stage mentions crats, it looks more like a bot process where arbcom then get the chance to veto. You could do something similar where the crats get to vet the candidate instead of arbcom, but I think we'd need to give the community the chance to tweak the criteria. In particular, if the candidate has been active in deletion discussions/nominations they need to show they are broadly inline with community consensus. ϢereSpielChequers 07:07, 22 February 2024 (UTC)

Proposal 7: Threaded General Comments

Following this RfC, the General Comments section would become a threaded discussion

Threaded discussion

So, maybe a bit minor, but I do find the general comments section to be a bit weird. We have a discussion, which is almost always bulleted. These are almost always places to comment on either other people's !votes, individual questions or functionary issues with the RfC. Items that get too long get moved to the talk page.

I have seen users complaining about the talk page not being taken into account when closing (I certainly read it, but I can see why people have this reservation). My thoughts, use level 3 headers to thread the discussions in such a way to allow people to discuss items in a more controlled way. This would also run in parralel with deprecating replying to individual !votes. Lee Vilenski ^{(talk • contribs)} 12:13, 20 February 2024 (UTC)

Support (proposal 7)

1. Support, I guess? I don't think it really matters, but... Queen of Hearts (talk • stalk • she/they) 04:02, 21 February 2024 (UTC)
2. Ehhh, not sure where i should put this, but the proposal to add a simple edit notice to encourage section headings would be enough, without any actual rule changes. Aaron Liu (talk) 14:37, 21 February 2024 (UTC)

General discussion (proposal 7)

• Threaded means "sorted" according to what? By topics, users, timestamps? There is no technical problem preventing discussions to be threaded, and I see nothing that prohibits threaded discussions by topic (which I assume is what this proposal is about). I think it can be done spontaneously without asking for special permission, just nobody cared enough so far. Szmenderowiecki (talk) 13:48, 20 February 2024 (UTC)
• Huh? @Lee Vilenski: would you perhaps make a mock up what this is trying to achieve (Perhaps reformat an old RFA GD section on a sandbox for demonstration). The GD is already a h5 so I don't think putting h3 sections in to it is a great idea. Regarding the talk page, conceptually I prefer the talk page be reserved for meta-issues (admin/crat reqeusts, discussion about contributors or their statements, etc). — xaosflux ^Talk 15:02, 20 February 2024 (UTC)

  Sure, sorry, this read a bit more obvious to me when I wrote it than how it came out. I hadn't realised it was level 5 headers, which is a bit of a shame. I have, however, moved some comments to User:Lee Vilenski/sandbox for the jist of the idea. The thing about the general comments is that they are often just a series of non-related comments about the RfA. I'd rather a discussion style, where we'd have a location to hold discussions about other people's !votes. Lee Vilenski ^{(talk • contribs)} 19:16, 20 February 2024 (UTC)

  That helps some and may not need any rule changes - perhaps just encourage discussion creators to start h6 sections? — xaosflux ^Talk 19:19, 20 February 2024 (UTC)

  Sure. I don't think we need any rules to be changed, but as we are looking for proposals, it's a good opertunity to gauge opinion. Lee Vilenski ^{(talk • contribs)} 19:26, 20 February 2024 (UTC)

  Oh for sure, if that is what you are going after, this can be about if that's a good idea - and perhaps do it with a comment on the preload. — xaosflux ^Talk 21:50, 20 February 2024 (UTC)

• I have no idea what this is asking to change. * Pppery * _{it has begun...} 17:27, 20 February 2024 (UTC)

Proposal 8: Straight vote

Following the trial of the discussion-first RfA (provided the proposal is enacted), a second trial will run for six months or 5 Requests for Adminship (RFAs) which are not closed as SNOW or NOTNOW, whichever happens sooner. During the trial, the main RfA page will consist exclusively of a numbered "support" and "oppose" section. The only thing participants will be allowed to type in those sections is their signature. If a candidate receives 65% of the vote, they are automatically given adminship, and there will be no bureaucrat chats. The talk page of the RfA may be used for any relevant discussion. 00:52, 21 February 2024 (UTC)

Support (proposal 8)

1. I found it strange that previous proposals for this were shot down due to concerns about SecurePoll, when there's no reason it would need to be used at all. This low-tech implementation would work just as well. Mach61 (talk) 00:52, 21 February 2024 (UTC)

   I realize it's probably worth attaching a proper rationale. So, in no particular order;

   • Democratic systems are already successful in practice on other WMF projects
   • Direct democracy has strong, empircially backed benifits for large discussions (like RfA)
   • It will be easier to moderate discussions, as those who do so don't need to worry about affecting the vote count.
   • Total discussion will decrease, as voters will not feel the need to add rationales indicating their agreement with one already stated (no "Per x" votes), and, as participants will not see rationales on the same screen they vote, they will be less likely to argue or even be aware of what they see as faulty rationales.

   Mach61 (talk) 14:00, 21 February 2024 (UTC)

   Direct democracy has strong, empircially backed benifits for large discussions (like RfA)

   That has a catch: the probability of people voting for the "correct" option. Without discussion and informed opinions, the chance of that can drastically decrease.
   Attaching rationales to votes is still direct democracy, and allows for discussion.

   Total discussion will decrease

   Per above, I don't see how that is a win. Aaron Liu (talk) 14:46, 21 February 2024 (UTC)

   It's a win because it's less stressful on one aspect for the candidate (seeing themselves heavily scrutinized in public). I'm not saying there are no drawbacks for the candidate if this is tried, but I think they will be outweighed by this and other benifits. Mach61 (talk) 14:48, 21 February 2024 (UTC)

   Unfortunately, I see it somewhat as a necessary evil to prioritize ensuring that the candidate is good to a degree over their well-being. Aaron Liu (talk) 14:51, 21 February 2024 (UTC)

2. Seems harmless and worth a shot, I'd support trying this out. As I understand it, this proposal is the same as the current system, except with all discussion moved to the talk page. I find threaded discussions in the general comments section at the bottom of RFAs and on their talk pages to be easier to follow than back-and-forths in the support and oppose sections. So maybe discussions will be easier to follow, and perhaps improve, if they were all threaded discussion in one place on the talk page. I don't really see a downside to trying it out. RFA crat chats are so rare (and RFA crat chat failures even more rare) that if they were eliminated and the threshold lowered to 65%, it would make no practical difference whatsoever to the wiki. I'm not going to name names but think about it and I think you'll agree: we've had more demonstrable admin failures who got 99% support than who got 65-75%, at least while I've been here. Levivich (talk) 02:36, 21 February 2024 (UTC)
3. Forcing opposers to spell their out their reasons for opposing is what causes the most trouble. The support !votes are often quite perfunctory or formulaic and it is quite inequitable that opposition is penalised and pilloried. Andrew🐉(talk) 09:39, 21 February 2024 (UTC)

   IMO opposing without a reason would cause more trouble and confusion, as there's no discussion or persuasion without arguments. Aaron Liu (talk) 14:40, 21 February 2024 (UTC)

   Arguments are rarely persuasive. Mostly what you get is bludgeoning, bullying and drama. No thanks. Andrew🐉(talk) 23:56, 21 February 2024 (UTC)

   I'm talking about the arguments for the oppose. As opposes are usually the minority, opposes do need a rationale so others might share that opinion and the candidate would understand what they can do to earn approval. Having an oppose without a rationale anywhere is not going to help anyone. There's a reason people don't leave blank opposes, even though there does not seem to be any policy forbidding them. Aaron Liu (talk) 00:04, 22 February 2024 (UTC)

   There's a reason that people don't leave any kind of oppose and it's that they know that there will be reprisals. The intimidation factor is so huge that there's a big difference between Arbcom elections and RfA. You have exactly the same sort of candidates but, in the process with the secret ballot, opposition is common whereas, at RfA, you see the one-sided voting found in nasty dictatorships. It's disgraceful. Andrew🐉(talk) 11:26, 22 February 2024 (UTC)

4. Sure why not. --JBL (talk) 22:36, 21 February 2024 (UTC)
5. It is a good start. But anonymous is the way to avoid all drama. Editors have to be worried about grudges and future runs for RFA if they vote their conscience under any system where they have to declare. Lightburst (talk) 22:56, 21 February 2024 (UTC)
6. I would prefer the discussion to be on the main RfA page, but I agree it should not be mixed with votes. Drama-free oppose votes without forced "rationales" would be a good change though. —Kusma (talk) 09:02, 22 February 2024 (UTC)

Oppose (proposal 8)

1. I see this as the worst, no offense. It does not protect the anonymity like secret ballots do, removes the usefulness of having attributed rationales attached in our current system, without any apparent benefit. Aaron Liu (talk) 01:01, 21 February 2024 (UTC)
2. I was going to post this to talk and leave a pointer here till I realized it would be POINTy. WP:NOTDEMOCRACY is an obvious policy to cite, but argument from tradition feels like a poor approach (and arbcom is already a straight vote, setting a precedent). My reasons are:
   1. This proposal counts petty !!votes the same as reasoned !!votes. Even if I don't post to talk or leave an informative edit summary, my vote matters all the same.
   2. There's no guarantee this will make RfAs more civil. In a recent RfA (many, really, but typed this thinking of one in particular), discussions on talk got heated and violated #PA. Why would moving discussions decrease stress?
   3. This may extend the amount of words (without adding content) because you'd have people typing "Support" on one page and "I voted support per XYZ."
   4. No anonymity per Aaron Liu.
   I'm clearly missing something: What do you think this would solve? Sincerely, Novo Tape^{My Talk Page} 02:12, 21 February 2024 (UTC)
3. I find Novo Tape's refutation convincing. * Pppery * _{it has begun...} 03:03, 21 February 2024 (UTC)
4. LEPRICAVARK (talk) 03:14, 21 February 2024 (UTC)
5. Oppose per LiuAaron. Worst of both worlds, imo. Queen of Hearts (talk • stalk • she/they) 04:03, 21 February 2024 (UTC)

   call me Aaron :) not sure why a lot of people call me Liu Aaron Liu (talk) 14:39, 21 February 2024 (UTC)

6. Strong oppose - this is a discussion. - jc37 05:36, 21 February 2024 (UTC)
7. Oppose - firstly, knowing why each user made his or her vote is useful in helping users unfamiliar with the candidate to get a good idea of his/her advantages and issues, allowing such users to make a more informed decision. And I see no reason why moving the discussion to an other page would be useful (moving the discussion away from the vote could be, although it makes it harder for users to tell the difference between a relevant argument a completely nonsense one, as in the latter case someone who does know the user is likely to comment on this fact).
8. Oppose We lose the benefit of rationales for people's votes being helpful for other voters, and we do not gain the anonymity that SecurePoll would provide. Also, moving all discussions to the talk page will do nothing but make them a little bit harder to see. QuicoleJR (talk) 14:09, 21 February 2024 (UTC)
9. Oppose the opposite of what an RfA should be. Lee Vilenski ^{(talk • contribs)} 15:27, 21 February 2024 (UTC)
10. Strong Oppose. Absolutely not, this is indeed the very opposite of what we should be doing. --Tryptofish (talk) 21:26, 21 February 2024 (UTC)
11. Oppose Aaron Liu took the words right out of my mouth. --Ahecht (TALK
    PAGE) 22:13, 21 February 2024 (UTC)
12. Oppose per Aaron and Novo Tape. Throwing away rationale just to prevent replies to your !vote is way worse than the proposal of threaded discussions only. 0xDeadbeef→∞ (talk to me) 08:55, 22 February 2024 (UTC)
13. Oppose no net gains with this proposal. --Goldsztajn (talk) 09:13, 22 February 2024 (UTC)

General discussion (proposal 8)

I'm not sure what you mean by candidates automatically getting adminship when they get 65%. If the first voter votes support, that would be 100%. Would there be some minimum threshold of votes? I was going to propose a straight vote like this myself seeing as we don't seem to be ready to implement SecurePoll on a large scale just yet, but this part is tripping me up. Perhaps it would be better to just stick with the established method of counting votes after a specified end time. Pinguinn 🐧 00:58, 21 February 2024 (UTC)

@Pinguinn At the end of seven days, no minimum participant count Mach61 (talk) 01:00, 21 February 2024 (UTC)

Ok, so like the current system. I think I'd much rather have the discussion take place on the same page as the voting so people will at least read it before they vote. This could even be combined with the discussion-first proposals where there's a discussion-only period before the straight vote opens. Pinguinn 🐧 01:08, 21 February 2024 (UTC)

I'm confused by this proposal, especially when you say (RFAs) which are not closed as SNOW or NOTNOW. Is this suggesting a straight vote for already-closed RfAs? Further, the 65% support threshold isn't one I'm comfortable with. Curious to hear more from you on what you see this proposal changing. Thanks Mach. — That Coptic Guy^{ping me!} (talk) (contribs) 02:15, 21 February 2024 (UTC)

This means that SNOW/NOTNOW RfAs don't count toward the 5 RfAs that the trial runs for. They will still be done as pure votes, which may make it less obvious it's a SNOW/NOTNOW, though. * Pppery * _{it has begun...} 03:02, 21 February 2024 (UTC)

• So this is "make it a vote" - which is an easy enough proposal; but not sure about the rest. Will there still be Q&A, will there still be the ability to have a discussion somewhere? — xaosflux ^Talk 16:22, 21 February 2024 (UTC)

  There will still be Q&A, and as mentioned the RfA talk page is available. Mach61 (talk) 16:24, 21 February 2024 (UTC)

Proposal 9: Require diffs

The following discussion is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

---

Require all support or oppose comments to also add unique diffs relevant to, and/or supportive of, their comments.

You can still say "also per so-n-so", but you still have to provide unique diffs of your own first.

If you support or oppose a candidate, you should be able to show why. If they are ready for adminship, this should be an easy thing to do. - jc37 05:43, 21 February 2024 (UTC)

Support (proposal 9)

1. Support - as proposer. - jc37 05:43, 21 February 2024 (UTC)

Oppose (proposal 9)

1. This would be very onerous for supporters. The last three RfAs had 265, 215, and 207 supports. For a new commenter to review all past statements to ensure they didn't duplicate a diff would be very time-consuming. It also seems unnecessary to require unique diffs for every commenter (support or oppose). If we want to make the selection process a weighing of pros versus cons, then we should just do that directly: collect supporting information about the characteristics of the candidate, and consider how they balance against each other. isaacl (talk) 05:57, 21 February 2024 (UTC)

   Reducing "drive-by voting" would be a feature, not a bug. This isn't a popularity contest, or a competition to see how many people can support a nomination. It's a Request for comments concerning a request for administrative tools. If that means fewer but more substantive commenters. then as I said, that's a feature, not a bug. And on the oppose side, this can help address drive-by opposition as well. - jc37 06:19, 21 February 2024 (UTC)

   He didn't comment about drive-by voting. There's a huge gap between drive-by voting and needing to spend over half an hour looking for an appropriate diff. Animal lover |666| 06:40, 21 February 2024 (UTC)

   As you can see from my previous proposal to which I linked, I support making the process based on the relative strength of pros and cons rather than counting up people on each side. But just requiring unique diffs from each person makes this into a net diff competition, without weighing the importance of each diff. isaacl (talk) 07:06, 21 February 2024 (UTC)

2. This has multiple problems: firstly, as the votes in either direction pile up, it would discourage additional votes in that direction; the mere need to look back far enough to find a unique diff and compare it to all the already linked diffs. Low participation is a frequently cited problem with RFA, not a goal. Secondly, on the negative side, a single diff could convince many to oppose, and there won't be any additional diffs to use for opposers after the first. Thirdly, if you're talking about a pattern of behavior (this user has a 99% success rate at tagging speedy deletions), how would you find an appropriate diff? It would also tend to cause too much emphasis on the user's distant past (it's much easier for your diff to be unique if you look there, and while doing so you notice some outdated concern). Note also that if CSD tagging is a major part of what convinces some non-administrator voter, then there is no way for the voter to find the diff - it may even be difficult for an admin to do so. Animal lover |666| 06:25, 21 February 2024 (UTC)
3. Nope. The "unique diffs" is a showstopper for me - if you want to "Oppose per this recent example (1) of policy violation" - why shouldn't someone else also be able to share that concern? — xaosflux ^Talk 11:33, 21 February 2024 (UTC)
4. Oppose I would support a proposal to require diffs for specific statements about specific circumstances, but you can't provide a diff for "not enough experience" or "great success ratio at CSD" and those should not be banned as reasons. Also, this would make it very hard for later voters, since unused diffs will quickly become hard to find, ignoring the fact that compiling more than one in a single comment is already a hard task on mobile. QuicoleJR (talk) 14:15, 21 February 2024 (UTC)
5. Per all of the above. The unique diff requirement is an extra deal-breaker, but even if that was removed I would still oppose since some valid rationales are impossible to justify via diff. LEPRICAVARK (talk) 15:12, 21 February 2024 (UTC)

General discussion (proposal 9)

• It's good for !votes to be evidence-based but the proposal hasn't been thought through. For example, consider the common oppose reason of a lack of content creation or low edit count. Such a lack of relevant experience is not verified with a diff. Andrew🐉(talk) 08:11, 21 February 2024 (UTC)

The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

Proposal 10: Unbundling 90% of blocks

Identifying vandals and spammers is easy, that's why we hand out rollback like candy. Blocking for incivility and pretty much any block involving goodfaith editors is difficult, and we need to vet candidates for that right very seriously. Uncontentious blocks of IPs and newbies for vandalism are the vast majority of the blocks we currently only have admins to issue. So it would be logical to create a new userright with the block/unblock button and also view deleted. Provided that the block/unblock button only worked against IPs and editors who are not yet extended confirmed.

Support (proposal 10)

1. As proposer ϢereSpielChequers 08:05, 22 February 2024 (UTC)

Oppose (proposal 10)

1. Not with full view deleted, which is spectacularly easy to screw up with. Maybe just with deletedhistory (which is security through obscurity anyway; you can get everything it tells you from database dumps and sql queries) to facilitate unblocking. Or maybe don't unbundle unblock either; unblocks are much rarer than blocks, and easy enough to let CAT:UNBLOCK patrollers handle it. —Cryptic 08:27, 22 February 2024 (UTC)

   Though, hrm, there isn't currently a separate permission to unblock (or modify blocks, which amounts to the same - doesn't matter if you can't unblock someone if you can just reblock them with a one-second expiry). Though I guess we'd need a new permission for block-only-ip-and-newbie anyway. —Cryptic 08:44, 22 February 2024 (UTC)

2. I am generally opposed to unbundling any of b/p/d. Also, bad blocks of newbies are just as much a problem as bad blocks of non-newbies are. —Kusma (talk) 09:00, 22 February 2024 (UTC)
3. This technical functionality doesn't exist yet so this is super premature, if you want to work on the effort to build a block (but not for these type of targets) suggest you start the massive time it would take to prototype that capability first (even if enwiki didn't use it, someone may). — xaosflux ^Talk 10:54, 22 February 2024 (UTC)

   Actually, I've spoken with them in the past. And they said it would only take a few seconds to create this as a user-right group. The functionality has already been done. - jc37 11:10, 22 February 2024 (UTC)

General discussion (proposal 10)

If this was a group of: block/blockemail/unblock; I think I could support this unbundling (presuming that the separated bundle is grandfathered/added to all current admins). But "view deleted", (even just the history) brings along issues that I would rather see as part of a discussion closer's package of user-rights, than merely someone who is helping out by blocking users due to behavioural issues. Not opposing yet, because I think through discussion this could become a workable proposal. - jc37 08:46, 22 February 2024 (UTC)

Proposal 11: Set some of the Admin criteria

Much of the angst and disagreement at RFA isn't really about the candidate, it is about the criteria for adminship. Hence the critique of RFA as it being like taking your driving test whilst driving a minibus full of examiners who are arguing with each other about the criteria for the test. This also causes uncertainty among potential candidates as to what the criteria is. Part of the problem of RFA as opposed to Rollback, Autopatroller or other userrights is that we don't have defined criteria for adminship that potential candidates can look at. Some of those criteria, 12 months activity, no blocks in the last month, at least 3,000 edits and some experience could be set by consensus in individual RFCs per criteria.

Support (proposal 11)

1. As proposer ϢereSpielChequers 08:05, 22 February 2024 (UTC)

Oppose (proposal 11)

1. Adminship doesn't have defined criteria because it's a bundle of wildly different rights. Say I only want admin for, I don't know, granting userrights and blocking vandals. Should I really have to pass some arbitrary criteria like "5 Good articles in the last 3 months" or "25 XfD !votes in the last month" just so I can perform actions completely unrelated to them? – Hilst [talk] 12:10, 22 February 2024 (UTC)
2. Oppose per Hlist. 🌺 Cremastra (talk) 13:26, 22 February 2024 (UTC)

Proposal 12: Abolish the discretionary zone and crat chats

RfA has a discretionary zone as part of its "not a vote" tradition. At some point (when RfAs had become rare), individual bureaucrats no longer wanted to exercise this discretion so crat chats were born, leading to additional days of bureaucrat scrutiny of the voters' arguments. This was a nice idea when it started, but now the community anticipates the crat chats. This means that during voting, people are inclined to make more and more lengthy arguments to ensure their vote is counted, further increasing the contentiousness of the discussion and the stress for the candidate. A simple hard pass/fail boundary would help to calm things down instead of inflaming things further like the anticipation of crat chats.

Support (proposal 12)

1. As proposer. —Kusma (talk) 08:57, 22 February 2024 (UTC)

Oppose (proposal 12)

• If "Change RFA to a strict vote" is what you want, then just call it like it is. — xaosflux ^Talk 10:56, 22 February 2024 (UTC)
• Strong oppose - closer discretion is a fundamental foundation of Wikipedia. - jc37 11:12, 22 February 2024 (UTC)

General discussion (proposal 12)

Very few RFAs are anywhere near the discretionary zone, if people are being clearer in their RFA !Votes maybe that's about them wanting to have others pay attention to their !votes. Also there was one crat chat last year and maybe two the year before. The numbers are too few to justify the assertion that every RFA that ends in the discretionary zone is expected to go to a crat chat. For example, I assume that if there is an RFA just inside the discretionary zone, but clearly outside if you downweight the weak supports and opposes, I could close that uncontentiously. Equally if an RFA was in freefall due to something happening at the start of day 6, and the RFA had dropped from >99% to clearly in the discretionary zone, I could close as unsuccessful without people calling for a cratchat. ϢereSpielChequers 09:29, 22 February 2024 (UTC)

Have you seen Tamzin's RfA? Crat chats may be rare, but they do more harm than good, both to candidates and the overall health of RfA. —Kusma (talk) 09:54, 22 February 2024 (UTC)

Seen it? I was in the support column, and recused from the cratchat. But that's irrelevant to this proposal and to my comments on it. This proposal is saying that cratchats have become automatic for every RFA in the discretionary zone, I made the point that "there was one crat chat last year and maybe two the year before" that's three, and you've just named one of them. You could also name the other two and it wouldn't refute my comments. If Cratchats do more harm than good that's a different thing and more relevant to those proposals to make the RFA a simple up and down vote. There is an argument that we should speed up crat chats, but that would need its own proposal - and the easiest way to do it would be to recruit more crats and then set an expectation that any uninvolved crat could close the cratchat after x crats had opined. ϢereSpielChequers 10:44, 22 February 2024 (UTC)