This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

Two proposals.

1. Right now the headers for the table aren't explained anywhere. Also I have no clue what a "dup" is.

Current:

RfA candidate S O N S% Ending (UTC) Time left Dups? Report

TJMSmith 118 1 2 99 13:53, 22 February 2021 4 days, 1 hour no report

RfB candidate S O N S% Ending (UTC) Time left Dups? Report

Proposed:

RfA candidate Support Oppose Neutral % Support Deadline (UTC) Time left [whatever "dups" is] Report

TJMSmith 118 1 2 99 13:53, 22 February 2021 4 days, 1 hour no report

RfB candidate S O N S% Ending (UTC) Time left Dups? Report

[blank row]

Or, if space is an issue, put a note underneath:

• S = support, N =neutral, etc.

2. There's a table that says "Recently closed RfXs"; can this be changed to "Recently closed RfAs and RfBs" to make it less confusing?

Cheers, Fredlesaltique ( talk) 12:51, 18 February 2021 (UTC)

"dup" means "duplicate" ie: one more editor has voted twice or more in that RfA. Ritchie333 ^{(talk) (cont)} 12:54, 18 February 2021 (UTC)

Ritchie333 Just to clarify, that means that certain editors can cast more than one vote? Fredlesaltique ( talk) 03:04, 21 February 2021 (UTC)

Fredlesaltique, Technically, they can (because it's only a wiki-page, which is the root of a lot of communication issues on here), but obviously per the procedure they can't vote twice, no matter what people tell you. Ritchie333 ^{(talk) (cont)} 10:17, 21 February 2021 (UTC)

I think clarifying RfX as suggested makes sense to make it a bit less "insidery". Best, Barkeep49 ( talk) 18:14, 18 February 2021 (UTC)

@ Barkeep49, I thought about that, too, but is RfX any more insidery than RfA and especially RfB? Anyone who knows what RfB is probably has no problem interpreting RfX? That is, we could rehead the table to "Recently closed RfXs (RfAs and RfBs)" which would help newer editors understand our lingo, but why would we expect anyone who didn't understand RfX to understand RfB? So is it really helpful except to someone in that minute space between understanding RfB and not yet understanding RfX? —valereee ( talk) 19:13, 18 February 2021 (UTC)

@ Valereee for me the answer would be because RfA and RfB are defined on Wikipedia:Requests for adminship and RfX is not. Best, Barkeep49 ( talk) 19:30, 18 February 2021 (UTC)

@ Barkeep49 and Valereee: I agree with Barkeep, and I think the Manual of Style ( MOS:JARGON) puts it pretty well: "Do not introduce new and specialized words simply to teach them to the reader when more common alternatives will do." Defining RfX in the article is good for now until the title can be changed. Fredlesaltique ( talk) 13:53, 20 February 2021 (UTC)

Is now. :D —valereee ( talk) 19:54, 18 February 2021 (UTC)

I suggest just titling the table "Recently closed requests" as there's no need to introduce unnecessary jargon. I know some people think jargon is something that forms a common culture for a community. I just like it to also provide greater concision in communication, so it has a purpose beyond being a shibboleth. (In isolation I'd also propose replacing RfA and RfB in the table with something like "Admin" and "Bureaucrat", but I imagine anyone reading this page or choosing to transclude the table elsewhere will have learned the terms, plus I assume some people won't want the table to grow wider.) A legend for the current status table would be nice to have (I never thought much about the "Dups?" column so not sure I knew what it was), though some of the transcluders might complain (it could be displayed only when an appropriate parameter is displayed, I suppose). However I imagine most people can infer what the other headings mean. isaacl ( talk) 20:23, 18 February 2021 (UTC)

@ Isaacl: Agree, but disagree for headers: took a couple minutes of scrolling to figure out that "N" is "Neutral" - Fredlesaltique ( talk) 13:53, 20 February 2021 (UTC)

I'm just extrapolating, but after all, you did figure it out, so you still fall under "most". Just explaining why I said I think a legend would be nice, versus is essential. Maintaining backward compatibility for those currently transcluding it without any parameters is the best way to try to avoid complaints. isaacl ( talk) 19:20, 20 February 2021 (UTC)

The bot operator for the bot that performs the task of keeping that table up to date has no time for even the most trivial of changes (one such has been requested some 3-5 times on his talk page over the past year and a half). Good luck team getting it changed. -- Izno ( talk) 06:13, 19 February 2021 (UTC)

What, CyberbotII? —— Serial 10:20, 19 February 2021 (UTC)

Cyberpower678 has always been responsive to change such things in the past - I find it hard to believe he'd not have time for a trivial change simply due to time. Worm^TT( talk) 12:33, 19 February 2021 (UTC)

Thanks for saying what I didn't want to, WTT; that's what I thought. Cyberpower is always helpful, and, well—if they're not responsive there's probably a good reason for it. Izno, were you by any chance thinking of the Citationbot, perhaps? (Which, IIRC, at some point at least had a notice on its talk telling would-be reporters that the owner would only talk to them on Github!) —— Serial 12:42, 19 February 2021 (UTC)

No, decidedly not. In this very task too, as it happens. -- Izno ( talk) 17:15, 19 February 2021 (UTC)

And in fact contrary to Anom's count of "3", the issue has been reported several times in the past few years. It amounts to a 2 or 3 line change and the addition of 2 character per line. -- Izno ( talk) 17:24, 19 February 2021 (UTC)

Worm That Turned, you rang? — CYBERPOWER ( Around) 12:53, 19 February 2021 (UTC)

The fact that Cyberbot I hasn't been updated after over a month proves Izno right. * Pppery * _{it has begun...} 17:04, 24 March 2021 (UTC)

The solution to this is to disable the bot task and switch to Module:RFX report, a reimplementation of the same thing using Lua that was created in 2013 but never quite reached the threshold of being ready for use. Courtesy pinging Mr._Stradivarius, the primary author of the current version of the module. * Pppery * _{it has begun...} 02:15, 20 February 2021 (UTC)

So what's the status on headers and changing "RfXs" to "RfAs and RfBs"? I also agree with Isaacl's proposals above (RfXs to Requests; RfA and RfB to Admin and Bureaucrat respectively).

I would edit it myself but as I understand it's a proprietary template? Cheers, Fredlesaltique ( talk) 13:53, 20 February 2021 (UTC)

The "recently closed RfXs" table can be edited manually at Wikipedia:Requests for adminship/Recent/Top and Template:Recent RfX , but it's not clear to me what exact change should be made right now. It's only the table at the top of the page that is maintained by a bot and cannot be updated by anyone other than the bot operator. (For what it's worth, proprietary is the wrong word, since the bot's source code is publically available at https://github.com/cyberpower678/Cyberbot_I/blob/master/SoxBot/rfx-report.php). * Pppery * _{it has begun...} 15:53, 20 February 2021 (UTC)

If a change can be made, since I think we should have an interest in keeping these templates small, use {{ abbr}} as intended. -- Izno ( talk) 18:57, 20 February 2021 (UTC)

Having pop-up tool tips is fine for those who can hover. A legend is better for those using mobile devices, which often don't support hovering or support it poorly, and those who have difficulties with fine motor control, such as those with tremors. (If the bot-generated report continues to be used, I agree the first priority is fixing the closing italics.) isaacl ( talk) 19:13, 20 February 2021 (UTC)

To keep current transcluders happy, I suggest leaving the default, parameter-less output the same, and only alter the output based on specific parameters being passed in. isaacl ( talk) 19:23, 20 February 2021 (UTC)

Seems like a good compromise. If space is constantly an issue, then a small-font legend at the bottom would work (S = support, O = oppose, N= neutral, etc.). (Also Isaacl could you do me a favor and edit out the RfX explanation in the prose, since it's not needed anymore? Would myself but can't.) Cheers, Fredlesaltique ( talk) 03:04, 21 February 2021 (UTC)

Are you referring to this edit that was made very recently? I suggest seeing what people think first; I don't what edits that were made will stick around or be reverted.

Small fonts are not great for legibility, so I'd advise being cautious about that approach. isaacl ( talk) 05:39, 21 February 2021 (UTC)

@ Pppery: Thanks for the ping. I'll have another look at the module today. Should be a nice Sunday project. — Mr. Stradivarius ^{♪ talk ♪} 00:33, 21 February 2021 (UTC)

@ Mr. Stradivarius: Any update on this? * Pppery * _{it has begun...} 21:07, 4 March 2021 (UTC)

Not as yet, but I'm working on it. — Mr. Stradivarius ^{♪ talk ♪} 08:51, 7 March 2021 (UTC)

@ Mr. Stradivarius: Any progress? * Pppery * _{it has begun...} 17:04, 24 March 2021 (UTC)

The situation in which this should-be-trivial edit gets stalled for two months for technical reasons is clearly undesirable. * Pppery * _{it has begun...} 13:46, 17 April 2021 (UTC)

@ Mr. Stradivarius: Any update on this? * Pppery * _{it has begun...} 13:58, 17 May 2021 (UTC)

Personally I think the new module-based template seems to work well, and we should just try using it (and making the relevant tweaks as above). If there are catastrophic bugs we can always resume using CyberBot's table. I think it's time we just do it. ƒirefly ( t · c ) 14:13, 17 May 2021 (UTC)

I see that "will you be open to recall" is now becoming a de facto standard question at RFA. Here's a fun new wrinkle: Pages related to the recall process were recently edited to say that if you say you will be open to recall during your RFA, it is binding. This was the text:

Administrators who have pledged to be open to recall during their successful Request for Adminship are assumed to not be able to withdraw this pledge or change the terms set then (absent a community consensus to so allow), since this is then a condition of election. No admin candidate is in any way required, during their Request for Adminship, to pledge to be open to recall. But if they do, it is binding.

This was added to both Wikipedia:Administrators open to recall and Wikipedia:Administrators open to recall/Admin criteria. Pinging @ Herostratus: as they were the one who added it. I understand the temptation to just back-door a community desysop process when every attempt to tdo it in a straightforward way has failed, but this is actually a major policy change and can't just be made up by one user and expected to stand. Beeblebrox ( talk) 19:32, 20 May 2021 (UTC)

As of now, Wikipedia:Administrators § Administrator recall remains in effect. If an administrator fails to follow through on a pledge, an arbitration case would have to be filed to establish that the administrator has lost the trust of the community. Being open to recall was a long-time favourite question at RfA; one commenter would oppose candidates who answered either yes or no, rather than instead supporting the establishment of an administrator recall policy. isaacl ( talk) 19:48, 20 May 2021 (UTC)

I know that, but it seems someone has been imagining it is otherwise. Recall hasn't been used in so long exactly because it is not binding. That's why I found the resurgence of this as a question at every RFA was somehwat silly, but this, were it remotely true, would've changed all that. Beeblebrox ( talk) 20:00, 20 May 2021 (UTC)

My apologies; I did assume you were aware of the appropriate section in the administrators policy, and was providing the link for the benefit of others. I'm a bit confused by your last sentence, though. If by "this" you are referring to the changes you reverted, then as you noted in your original post, unilateral changes to a non-policy page wouldn't take precedence over the administrators policy and established consensus. isaacl ( talk) 20:16, 20 May 2021 (UTC)

I'm confused. I have been an editor and admin for a long time and I cannot remember any time at which recalls were binding like that. Recalls have always been something voluntary. Wikipedia:Administrators open to recall was created with the wording "The process is entirely voluntary. Administrators are free to change their recall criteria at any time, to decline participation in the process despite having previously agreed to be open to recall, or to disregard the outcome of any recall proceedings." and this has not changed since four days ago when Herostratus added the language Beeblebrox mentioned. Even the subsection you linked to says so and has for years. As such, I don't think failing to follow through on a pledge cannot be the reason for an ArbCom case until and unless there is actual consensus to change this (and then only from this point forward). Regards So Why 20:10, 20 May 2021 (UTC)

Yes, as stated in the section I linked to, the optional recall processes are not binding. As I mentioned, an arbitration case would have to be opened to establish that the administrator had lost the trust of the community. It would not be based solely on whether or not a pledge was broken, although commenters could refer to this as part of their viewpoint. Presumably the bulk of the argument would be based on the same reasons why the recall process reached the decision to recall the administrator. The arbitration committee would not be bound though by the specific parameters of the recall process, and would evaluate all presented evidence. isaacl ( talk) 20:24, 20 May 2021 (UTC)

But how can an admin be sanctioned for doing what the policy explicitly allows them to do? Voluntary recalls after all often do not require any actual malpractice, so any case that is based solely on failing to step down after a recall would imho be without merit. On the other hand, if the admin really violated policies, people would not resort to a voluntary recall mechanism but bring them to ArbCom directly. Regards So Why 06:26, 21 May 2021 (UTC)

The community can try to reach a consensus in many ways: it can hold a discussion held in a venue such as the incidents noticeboard, it can follow an administrator's recall process, or it can use some other method. If a consensus is reached that the community doesn't trust the admin and the admin doesn't choose to voluntarily relinquish administrative privileges (perhaps they contest that a true consensus view has been determined), then the community can choose to file a request for arbitration to let the arbitration committee evaluate the merits of the evidence. Given the lack of experience with recall procedures being used, it's hard to characterize the results they might produce. The arbitrators can work out the underlying issues, though, and then evaluate the situation in its totality. isaacl ( talk) 15:40, 21 May 2021 (UTC)

I actually was thinking we should just start advising RfA candidates to add the statement "I am open to recall" to their statements about other accounts and paid editing in their acceptance. —valereee ( talk) 20:11, 20 May 2021 (UTC)

I feel the exact opposite. Recall is not binding, and never has been. It seems nobody can even remember when the last time was it was actually used successfully, as this question was asked six years ago at Wikipedia talk:Administrators open to recall/Past requests and never got a reply. I asked again yesterday but so far, no answer. People who want to pressure admins to be open to recall should read Wikipedia:Administrators open to recall/Past requests and see if they still feel that way. Even when it was used, it almost never succeeded. Beeblebrox ( talk) 20:22, 20 May 2021 (UTC)

"Open to recall" stands in opposition to the idea that as an admin you should always be responsive to feedback about your actions and how you use the tools. I know people who are pressured to commit to that in RFAs don't intend it to be a declaration that they will be unresponsive to the common folk, but that's what it is. I think a better commitment would be a promise to always be responsive to feedback about how they use the tools, no matter who it comes from. Guettarda ( talk) 20:35, 20 May 2021 (UTC)

I disagree that having a specified recall procedure means the administrator can't also be responsive to feedback provided in other ways, including relinquishing administrative privileges if they feel that is the best route forward. I agree it may be helpful to discuss a general approach to being receptive to feedback. isaacl ( talk) 20:45, 20 May 2021 (UTC)

That's already ensconced in policy at WP:ADMINACCT. The assumption is anyone who runs for RfA is committing to that policy, since they are bound by it if they become administrators. -- Hammersoft ( talk) 20:55, 20 May 2021 (UTC)

I had this question on my RFA, answered I am not open to recall, explained why, and passed. I do not see anything wrong with this. I do not think "open for recall" should become a standard.-- Ymblanter ( talk) 20:37, 20 May 2021 (UTC)

It was pretty fashionable around the time of my RfA in 2009 but the question wasn't asked of me, which was nice. I've never been a fan of the voluntary process as I've always felt that if I became sufficiently deranged that I needed desysopping, I'd probably also be sufficiently deranged that I wouldn't trust my own recall system. I genuinely suspect that if several admins I trust told me I would be better off resigning, I probably would - but I can't exactly be sure that that would be the case at the time. ~ mazca ^talk 21:22, 20 May 2021 (UTC)

• Beeblebrox, thanks for undoing that dramatic change to WP:RECALL. Aside from the obvious problems with attempting to backdoor a community desysop process, the reality is the recall process, as it says at the top of that page, "is not a formal policy or guideline and does not have any formal status". To change it to what Herostratus wanted creates a binding process that...paradoxically...then would be binding but not be formal policy, not be a formal guideline, and not have any formal process. On the face of it, Herostratus should have clearly understood this would never fly. @ Herostratus: If you want to create a de-adminship process, you can't do it by fiat. Further, you would be well advised to carefully review the more than two dozen other proposed processes that failed to gain consensus at Wikipedia:Requests_for_de-adminship#Proposed_processes. -- Hammersoft ( talk) 21:14, 20 May 2021 (UTC)
• Some people might be interested in seeing Wikipedia:Requests for adminship/Herostratus 2. I found it while looking around in regards to this. It's a voluntary recall for Herostratus from 2010 that was run as a second RfA, and failed. Herostratus' first RfA ran in 2006, and was successful. -- Hammersoft ( talk) 21:33, 20 May 2021 (UTC)
• My own personal experience with recall went like this: the admin in question had a list of people who's opinions they valued. Only those on the list could initiate the process. I was on the list, and saw them acting irresponsibly for like the billionth time. So I initiated the process. Their response was to remove my name from the list, thus making my submission invalid under their self-established criteria. And so,eventually, it fell to ArbCom to do what recall could not. That seems fairly typical of how effective recall was when it was actually even used. You could also pledge to be open to recall, and then set conditions like "100 admins must agree that I shouldn't be an admin anymore" thus making the whole thing pointless. It's a terribly broken system, which is why nobody uses it and why we should stop pestering admin candidates about it. Beeblebrox ( talk) 22:13, 20 May 2021 (UTC)

  That's actually why I think we should just have candidates add it to their standard acceptance: it means nothing (although I think it should), someone's going to ask, and why waste the time? —valereee ( talk) 22:19, 20 May 2021 (UTC)

  I think a better approach would be to admit the obvious: it's a dead process that hasn't been used in a decade and should just be marked historical and deprecated. Seems like there's a small group of people trying to do the opposite though. Beeblebrox ( talk) 23:03, 20 May 2021 (UTC)

  I concur with Beeblebrox. It is added to about 63% of successful RfAs. But, in practice, it hasn't been used in nearly 10 years. It's a vestigial process that has no point. It isn't used. At all. Why waste time, indeed. It takes a lot more time to set up a personal recall page than it does to simply not have the question asked at all. -- Hammersoft ( talk) 23:06, 20 May 2021 (UTC)

  I don't think we should recommend to candidates to include a pro forma statement on recall. I think they should choose for themselves if they want to say anything about it, or on being receptive to feedback generally. isaacl ( talk) 00:25, 21 May 2021 (UTC)

• It surprises me that a lack of recall is described as a failure of recall. That does not compute with me. Levivich  ^harass/ _hound 06:01, 21 May 2021 (UTC)
• The premise at the start of this thread is flawed. There are admins who have passed RFA in recent months without any discussion of recall, I don't recall a recent incidence of someone being asked the question and saying that they don't believe in recall, but the time to start assuming that recall is now de facto mandatory would be if a candidate failed RFA solely because of the way that they answered the question. My guess is that if someone said "my recall Criteria is the community elected desysop process known as ARBCOM" that alone would not derail an RFA. As for the idea that recall criteria should be fossilised in time, I'm pretty sure that people can currently pass RFA before defining their recall criteria. Ϣere SpielChequers 09:11, 21 May 2021 (UTC)

Yes you're entirely right about that. Any candidate can answer the question however they like, and fine, and -- if you're right, which probably -- it won't matter, as only a few people will care, maybe. So it's not really worth worrying about very much, right? Herostratus ( talk) 16:13, 21 May 2021 (UTC)

In my experience as both a candidate and nominator, even candidates who are passing easily feel a lot of pressure at RfA. I know I was hoping I didn't get the recall question because my honest answer was no but I was going to feel pressured to say yes. I have since decided that having a recall method is the right thing to do but I am glad I got there on my own and not because I felt backed into it at RfA. Best, Barkeep49 ( talk) 16:26, 21 May 2021 (UTC)

I just feel like it's one less question to have to answer, like those about having previous accounts, ever editing for pay, etc. —valereee ( talk) 16:36, 21 May 2021 (UTC)

The difference is that previous accounts and editing for pay have community consensus where as this (unfortunately in my opinion) does not. Best, Barkeep49 ( talk) 17:07, 21 May 2021 (UTC)

Well, I don't know about that. While we have never been able to come to consensus on what a desysopping process should be, multiple RfCs have shown most editors want there to be one. Please don't ask me for the diffs —valereee ( talk) 01:03, 22 May 2021 (UTC)

Although there's no consensus on how to get one, most editors want a pony. But until everyone knows the actual process to get a pony, it's not really accurate to say there's a consensus for everyone to have a pony. isaacl ( talk) 02:15, 22 May 2021 (UTC)

Better yet, John Bolton, Vladimir Putin and the Pope all agree that there should be permanent world peace. However, there's no consensus on any route thereto, and so there's no consensus on anything, really.-- Wehwalt ( talk) 12:29, 22 May 2021 (UTC)

The problem is that it creates the appearance of trying to create policy (that is, a system of recall), that we have never been able to reach a consensus on, without such a consensus, by compelling people to agree to it or they'll be voted down. Note that since RFA requires a consensus to pass (and is more strict about that than most of our policies) this would allow a determined minority of editors to make any random demand they wanted a requirement at RFA, even if (as with recall efforts) it plainly lacked consensus. This is a broader issue than just recalls, even if they're the most prominent example of where that strategy has been used - to fix this, I would suggest we try to reach consensus on a more strongly-worded guidance to bureaucrats to disregard any votes at RFA that are plainly based on an effort to change or create new policy, with recall being specifically spelled out as something that it is unacceptable to base an RFA vote on. RFA is the place to judge individual editors for adminship - it is absolutely not the place to create new policy, and anyone trying to do that ought to be shut down hard. -- Aquillion ( talk) 03:26, 22 May 2021 (UTC)

Arbitrary break

Alright. All this is a pretty minor procedural, so no need overly worry about anything. I agree with you all that all this 99% theoretical. 99% of admins are never going to have to deal with any of this. And of course it doesn't affect any of you personally -- you're grandfathered out!. Nevertheless it's worthwhile on the margins IMO. Also, of course it's legit for me to have made WP:BOLD edits to that page, particularly since its not a rule page, and legit for somebody to roll them back and say "that was a terrible edit, what were you thinking?" And, I suppose we all agree that "will you be open to recall?" is a perfectly legit question at an RfA, not disruptive or something, been asked many times, and of course the candidate can answer however she likes -- Yes, no, or any other way. So far we're all on the same page I think.

Now, on to the sticking points, the basic being this (I'll box it for emphasis):

So maybe put yourself in my shoes for a minute. Suppose I (or another editor) asks the question. If you say "No", I won't vote for you. If you "well I'll consider it" or any other thing I consider unsatisfactory I also won't vote for you. That's my prerogative. Is it not? But if you say "Yes", I probably will vote for you.

So, if it turns out that admining is not an area of strength for you, and enough people want to have a second look at your qualifications and record and have a reconfirmation, should you be able to say "I was lying just to get your vote" (or some more mealy version -- "well circumstances have changed" or "I didn't really realize what I was signing on to" or "Come now, nobody takes campaign promises literally" or whatever)?

That's the key question: would it be OK for you say "I was lying just to get your vote", and there being no further formal recourse (not counting ArbCom, which I don't)? It's really a simple yes or no question.

Describe how people like me, who voted "Support, candidate has pledged to be open to recall, which shows a proper approach for the position and, more importantly, gives us a safety valve in the 100-to-1 chance that she goes off the rails" would feel. Can you? Describe how editors feeling that that would be helpful for community cohesion and moving the project forward.

OK? That's the main thing. There're other things to say, other reasons why this change would be helpful and an example of how it would have helpful in in earlier case, and more, but thats all for right now. Oh, and many of you had interesting and useful things to say, which I mostly don't agree with at this time, and thank you! but I'm of space for now. Herostratus ( talk) 16:13, 21 May 2021 (UTC)

• @ WereSpielChequers: I didn't say recall was de facto mandatory, just that it seems "are you open to recall" is becoming a de facto standard question. The ultimate goal of doing so, however, would seem to be making recall de facto mandatory even though nobody uses it. I don't personally care for it when anyone asks an identical question at every RFA, regardless of what it is, but I find this one particularly obnoxious because there is obviously only one right answer as far as the person asking it is concerned. It seems intended to pressure the candidate into publicly saying they will open to using an outdated, unused, toothless process that didn't work to begin with. So, you can imagine my alarm when I realized there was, at the same time, an attempt to turn that semi-forced answer into an ironclad unbreakable commitment to permanently be subject to recall, and that if they had not set their own criteria for one during their RFA they would be unable to change that later. This, if it had any basis in reality, is not a "pretty minor procedural" as described above, it would be a major policy shift.

@ Levivich: Perhaps "irrelevant" would be a better descriptor than "failure." As far as anyone seems to be aware it has not been used at all in a decade, while in that same time there have been numerous ArbCom desysops. The community moved on from recall, and it isn't hard to understand why if you look at Wikipedia:Administrators open to recall/Past requests. It was not effective as a deterrent against admin misbehavior in all but a very few cases, and was clearly subject to numerous spurious filings. So again, trying to back-door it into being a real desyssopping process that you irrevocably swear to abide by under public pressure at RFA is something I am just really not ok with. Beeblebrox ( talk) 18:31, 21 May 2021 (UTC)

To me, a lack of recall doesn't suggest that recall is irrelevant, doesn't have consensus, or that there is any other problem with it. The community of ten+ years ago isn't the community of today; 90%+ of today's editors weren't here ten years ago. That entirely new generations of editors and admins still ask the recall question and still add themselves to the category, after ten years, is what suggests to me the voluntary recall system we have now still has consensus. I don't really agree with the characterization of "back-door" but then you're kind of arguing a straw man there, as no one, not even Hero, is taking issue with your reversions. (In fact I'm not really sure why you started this thread.) But, when it comes to recall, I think everything is fine the way it is, and I don't see the lack of recalls as a problem itself or as an indication of a problem with the recall system. I mean, there is an obvious reason why there haven't been recalls in ten years: admins who would be recalled don't add themselves to the category. Levivich  ^harass/ _hound 18:47, 21 May 2021 (UTC)

Right, not taking issue with WP:BRD reverts on principle, but that doesn't mean I don't disagree on the merits. No, starting the thread was fine, that's part of the process... it's worth thinking these things out together, I think.

So, as to "I find this one [question] ...particularly obnoxious because there is obviously only one right answer as far as the person asking it is concerned. It seems intended to pressure the candidate into publicly saying they will open to using an outdated, unused, toothless process that didn't work to begin with"... Well let's see:

• Q: "Ma'am, if you're elected, will you promise not to embezzle millions from the state treasury? And if you do and are caught, will you resign?"
• A: "Well, now, see, this is one of the reasons people are fed up with you types. See, this is one of those obnoxious questions that has 'only one right answer'...

Heh. Well of course it only has one right answer. Lots of questions do! But OK, I get where you're coming from Beeblebrox. If I'm reading you right, you don't think it's a good question even if the person doesn't really have to follow thru if push comes to shove, right? OK, but I don't think too many people are the mind that very question itself is not so great. doesn't necessarily make you wrong, but it does put you in a minority (I think?) The question at hand is not should we keep asking it (we will!) but whether we're going to fix it. You yourself just said it's an "outdated, unused, toothless process". So we're on the same page, here. Let's make it updated, useful and... toothy? Is that the word? That's what we're trying to do here. Herostratus ( talk) 20:31, 21 May 2021 (UTC)

I also believe it is completely inappropriate to ask it in an RFA (and I am not at all sure it is such a minority; the question has clearly sparked severe controversy in the past - it is, in my memory, one of the most controversial questions that can be asked at RFA.) Policy should be changed, and procedures created, by broad measured consensus, not by pushing administrator candidates to agree to something that is not part of existing policy. As I mentioned below, this could lead to a situation where an organized minority of editors can push through policy changes by demanding that all new administrators accede to them. I should hope bureaucrats disregard any vote premised on the answer to such questions as being unrelated to policy or adminship criteria, but it is still disturbing; and, obviously, trying to retroactively make the agreements you extracted in that manner binding was a shocking overreach. If you want to create a recall system, you are an experienced enough editor to know the correct way to do so. These questions aren't that, and ought to stop at once. Personally, on reflection, I've come to the conclusion that anyone who would yield to such demands (presuming they were not previously involved in discussions over creating a recall effort) should not be an administrator, even before we get to the fact that (given the lack of any successful invocations of such pledges to date) it is extremely likely that they are actually lying. -- Aquillion ( talk) 21:50, 22 May 2021 (UTC)

I appreciate that everyone has the liberty of using whatever criteria that are important to them. At this time, there isn't a consensus to evaluate the outcome of a request for administrative privileges based on a limited set of criteria, as what can convince an editor to trust a candidate can vary greatly. I suggest, though, if you don't feel that a question on recall should be used as a litmus test, that it wouldn't be suitable to try to determine if the candidate yielded on the question (which is a somewhat similar scenario as the commenter who rejected candidates who gave either a yes or no answer). If they hadn't been involved in previous discussions, it would be hard to figure out if they were espousing a new viewpoint, and even if they did appear to change their mind, it would be hard to know if it was sincere or not—editors can change their minds upon reflection, and sometimes change them back, too. Personally, I think what's important is the reasoning expressed by candidates, and if it illustrates how they can put forth a reasoned argument in a clear manner. isaacl ( talk) 01:22, 23 May 2021 (UTC)

I get what you're saying, and certainly the fact that I've committed to opposing most editors who support RFA questions that try to bind their future actions in ways that don't reflect policy puts candidates in a terrible situation if eg. Herostratus continues to try and push them that way in the future, since regardless of their answer someone is going to oppose them based on it. But I do think it's a coherent and important criteria - having policy created via backdoor by trying to force administrators to accede to it during the heated debate at RFA (knowing nearly a supermajority is generally required to pass, which gives minorities more voice there than they have elsewhere) is completely inappropriate and would cause all sorts of problems if it kept happening. The only way I can see to prevent that at the moment is to push for and support administrators who are willing to stand up to such demands, and oppose administrators who seem to yield too flippantly. I'll grant that it is possible they are telling the truth and were suddenly convinced by Herostratus' logic or had just never thought about recalls before, rather than crassly saying whatever is necessary in order to get a vote; but the whole thing still feels improper to me and leaves a sour enough taste in my mouth that I'd prefer to oppose anyone who gives that appearance. -- Aquillion ( talk) 07:50, 23 May 2021 (UTC)

I personally am dubious that recall has suddenly become an important enough matter for enough single-issue commenters to appear at requests for administrative privileges to swing a decision either way (it wasn't in the past when the question was last commonly asked, and hasn't been recently). To the broader point, however: I think it would be better to model the desired behaviour of commenters. Encourage candidates to explain their viewpoints clearly, and evaluate candidates on their overall characteristics, with a detailed rationale in the support/oppose statement. Support statements tend to be thin; it is a good opportunity though to highlight why you are personally supporting, and how litmus-test questions are not key to your support, if that's the case. isaacl ( talk) 16:06, 23 May 2021 (UTC)

To know if opting into a recall procedure failed to be a deterrent, we'd have to look at all the admins who did so and see if they have acted in questionable ways. It's tricky to determine the reverse, because perhaps they didn't need a deterrent to start with, but I think we should count it as a positive if all went well. It might be irrelevant for the reason Levivich points out, but it can still help build trust with the general editing community for those who sign up.

Regarding the question being asked at requests for administrative privileges: I didn't particularly care for the question when it was in vogue before, and I'm not personally excited to see it return, but that's life. People can ask about topics that matter to them, and administrators ought to be able to communicate their viewpoints clearly. There are ways for a candidate to disclaim interest in a voting-based recall procedure while still emphasizing their commitment to be receptive to feedback, if that's what they really feel. isaacl ( talk) 20:35, 21 May 2021 (UTC)

• (multiple ECs later, this was directed as a reply to Lev's post)It's interesting to me that we seem to be in agreement about the basic facts but not what they mean. If a process isn't used in 10 years, that would seem to be to be evidence that it is not effective. Anyway, the reason I opnened this thread is because their seems to be a direct link between the edits on the little-watched-or-maintained pages on recall, and the resurgence of asking about recall at RFA. To me it looks like a trap, to get admin hopefuls to promise to permanently be bound to the recall process, and if they don't set their own criteria during the week of their RFA they can't change them later. It's troubling and I felt it needed more eyes on it. Beeblebrox ( talk) 20:37, 21 May 2021 (UTC)
• I imagine commitments made during the request for administrative privileges matter to most people. The community, however, has a long tradition of examining context for situations. There can be valid reasons to decide that the outcome of a recall process didn't adequately reflect community sentiment, such as a voting-based recall procedure getting vote-stacked by new editors. Admin candidates can pledge not to perform tasks in a given area, and later decide that the area is suitable for them after all. I'm not against the idea of having a binding recall procedure, but to-date the community doesn't support one without first reaching a consensus to update the administrators policy. isaacl ( talk) 20:48, 21 May 2021 (UTC)
• @ Isaacl: without a policy to empower involuntary removal of sysop access, us 'crats won't be doing it - so the "binding" nature is a bit hard without working on policy. (Yes, you could funnel the request via ARBCOM if they want to request the removal from us then we'd do it per the arbcom powers policies). Such a policy update would need to not just empower crats to take action, but also clearly identify what the prerequisites processes are, and should probably include verbiage requiring RFA to get access back in the future. — xaosflux ^Talk 00:31, 22 May 2021 (UTC)

  Yes, I agree that bureaucrats won't go beyond the current administrators policy for removing administrative privileges, which is why I responded to Herostratus saying that policy needs to be updated before changes are made to the recall procedure saying that it is binding. isaacl ( talk) 01:59, 22 May 2021 (UTC)

• It is not at all a minor procedural question; this cuts to the heart of how we make policy. The core problem with trying to make the question binding is the same problem that makes asking the question itself highly inappropriate. RfA is intensely stressful and high-pressure. People are (obviously) more likely to agree to it there than they would be to support it elsewhere, just to make the people asking the question shut up. And yes, I know that seems unfair and sneaky and whatever, but consider the converse - the people asking the question know this. That's the entire reason they ask it at that precise moment; if there was clear support for a binding recall process they would just run an RFC for it and create it, without any need for individual admins to agree or not. Regardless of the disreputable nature of lying when answering questions at RfA, for the questioners, asking questions like these at RfA and then basing your votes on them in hopes of creating a cohort of admins bound to whatever system you're trying to create is is absolutely not the correct way to create policy, and should be strenuously discouraged. After all, drastic changes to policy are supposed to require a clear consensus - whereas in practice, since RfAs require more than a majority, an organized, determined group of editors could spoil the RfA for anyone who refused to accede to whatever random demand they wanted to add to policy, which is absolutely unacceptable and is honestly an accurate, if uncharitable, way to describe what some of the people who base their votes on open-to-recall questions are trying to do. Yes, I know it ultimately failed (although plainly this proposed change was trying to reverse that), but that's, in part, because I think that many people reached the same conclusion I did here and started to push back against the people who were trying to use RfA to make recall a universal standard. That's also why it has always, and should always, failed to ever gain teeth, a failure which made even its most fervent advocates realize that their efforts were moot (as they absolutely should be.) Therefore, I don't think any answers there should ever be considered binding. And I say this as someone who would be fine with a universal recall system - I just strenuously disagree with using this backdoor method to try and create it. It should be applicable to all admins or none of them, and should be created properly through our existing consensus-building mechanisms - it should not depend on people trying to use RfA to force through effective policy changes by indirect means. -- Aquillion ( talk) 03:13, 22 May 2021 (UTC)
• No, pledging to make yourself open to recall is not binding and never has been. Though, if I remember correctly, we've been through this trend of recall questions and pledges in the past. I think it was a popular question in the year or so before I ran for RfA in 2017, but had grown out of favour by October so I never answered it. I don't have a recall criteria because they're mostly for show and I think I'd just resign before a successful recall was triggered anyway, so why bother making people go through a process? Anyway, to Beeblebrox's original point, this was a pretty large change in the policy and not indicative of community consensus. TonyBallioni ( talk) 19:09, 22 May 2021 (UTC)
• I feel like the fundamental issue with any and all DRFA reforms is that the status quo doesn't have consensus, but no one proposed change can ever get consensus either, and there's really no way to change that unless somebody could somehow craft a proposal that everybody would agree on. Jackattack1597 ( talk) 01:23, 25 May 2021 (UTC)

As a tangential point to the above thread, the "Recent RfX" box now stretches back eight months. As well as not being particularly "recent", I think it also humiliates the two unsuccessful RfAs for far longer than should be necessary, to the point I think Wikipedia:Don't remind others of past misdeeds could be a reasonable argument for removing them.

With that in mind, should we change the criteria so that only actually recent RfXs, say in the last 90 days, get displayed instead? If that means the box is empty from time to time, so be it, but I don't think we've actually had a gap that long. Ritchie333 ^{(talk) (cont)} 11:49, 17 May 2021 (UTC)

• I'd support this, with maybe a link to the history of all RfAs. Best Wishes, Lee Vilenski ^{( talk • contribs)} 12:09, 17 May 2021 (UTC)

• Support moving the threshold of "recent" to 90 days for all the reasons Ritchie outlines.. ƒirefly ( t · c ) 12:29, 17 May 2021 (UTC)
• I agree it should be shortened, but I think we should always have a few (3+?) in there, so that those new to the process can see examples. Johnbod ( talk) 14:18, 17 May 2021 (UTC)
• Disagree that this should be shrunk based on if there are unsuccessful RfA's. Not being successful isn't a "misdeed" - it just wasn't a success. Being able to see the dates of the last RfA's at a glance is useful, as it shows how busy or idle the forum is as well. — xaosflux ^Talk 14:24, 17 May 2021 (UTC)

  I think NASCARfan0548 (who's NOTNOW RfA has sat linked to the top of the main page for four months) would be best placed to answer that. Ritchie333 ^{(talk) (cont)} 14:27, 17 May 2021 (UTC)

  I would also agree that the cut-off point shouldn't change depending on whether the recent RfXs were successful or not, but having it limited by count (currently 7) rather than time can stretch the meaning of "recent". ƒirefly ( t · c ) 15:18, 17 May 2021 (UTC)

  That's a nice ideal, xaosflux. If it were me, I'd be humiliated every time I looked at it. —valereee ( talk) 23:30, 17 May 2021 (UTC)

• I agree that 90 days is a reasonable threshold for deciding what is "recent". We will frequently have had no recent RfAs, an issue that has been discussed at great length, no reason this reality can't be presented in that template. Best, Barkeep49 ( talk) 15:57, 17 May 2021 (UTC)
• Agree with this. It really doesn't make sense to keep them posted up for the better part of a year, and doing so to unsuccessful candidates is unnecessarily cruel and humiliating. ~Swarm~ ^{sting} 18:07, 17 May 2021 (UTC)
• Support 90 days is enough. User:力 (power~enwiki, π, ν) 18:09, 17 May 2021 (UTC)
• +1 Levivich  ^harass/ _hound 19:02, 17 May 2021 (UTC)
• I disagree with the premise: we should be striving not to treat actions coming from inexperience as a misdeed, per se. Misunderstanding how to find a suitable role in a community is a fairly usual occurrence. Nonetheless, I agree there's no need to give additional prominence to the result of requests beyond a certain age. This is primarily a convenience to allow people to see what happened with requests that they remember seeing very recently (I'd support a much shorter threshold than 90 days). The regular archives are sufficient for people to look up past requests, and they are already organized showing counts of activity by year (successful RfAs have a summary table broken down by month). isaacl ( talk) 20:25, 17 May 2021 (UTC)

  Upon reflection, I think a considerably shorter period of say 7 days or so is enough. The archives are just a couple of clicks away (uncollapse navbox, select link to archive) and the state of affairs isn't changing that rapidly. isaacl ( talk) 05:29, 19 May 2021 (UTC)

• I think it is enough to rename the box to "most recent RfX requests" and to hang our head in shame whenever the oldest of the most recent seven is older than a month. — Kusma ( t· c) 20:40, 17 May 2021 (UTC)
  • @ Kusma: we will be hanging our heads for some considerable time if we need 84 candidates a year to stop Nosebagbear ( talk) 12:30, 20 May 2021 (UTC)
• I don't even know why this should be on the page. Is there any point? It might have been a fun ticker when it ticked over every couple weeks or so, but now it's just a wall of shame for the unsuccessful. Get rid of it altogether, or proposal: just include the successful RfAs. —valereee ( talk) 23:27, 17 May 2021 (UTC)

  I could be misremembering, but I believe it was introduced to help people see what happened in the most recently closed requests which have been cleared off the page. Of course people can look at the archives for this purpose. isaacl ( talk) 23:40, 17 May 2021 (UTC)

  I think it certainly serves a purpose for the community to be able to easily see recent admin promotions; not everyone is going to be able to see and follow every RfA, so I think it's a handy tool in that regard. But yeah I can't think of any particular reason we feel the need to showcase failed RfAs for months. We're currently still displaying a sockpuppet's failed RfA more than six months after the fact, I'm not sure how the community is supposed to be served by that. ~Swarm~ ^{sting} 03:08, 18 May 2021 (UTC)

• I've trimmed it to the 3 RFAs from 2021, mainly so people can see how it looks when it is shorter. Two of them are still older than 90 days. User:力 (power~enwiki, π, ν) 03:18, 18 May 2021 (UTC)

  That is fine, but in an ideal world 90 days would be enough to cover the last few... Giant Snowman 10:20, 18 May 2021 (UTC)

  I was just about to come in and give my opinion for pretty much the exact change you made. I support a "90 days or 3 RfAs, whichever is more" box. I oppose excluding failed RfAs, which are much more useful for taking the community's pulse than successful ones (admittedly, this applies moreso to non-SNOW/NOTNOW fails, although the borderlands of what's considered NOTNOW are still educational). I can't particularly think of a good case for a ticker-box that only shows successful RfAs, because then it transforms the box into some kind of weird bragging right rather than a quick way to read the most recent RfAs and see what quote-unquote 'standards' the community is currently holding, and a box with only successful RfAs is about as educational as no box at all. I concur that a 90-day box should ideally be holding a lot more than three RfAs, but such is such. Vaticidal prophet 04:25, 19 May 2021 (UTC)

  I actually think most failed RfAs aren't helpful to anyone. Of the past six fails, only two were reasonable candidates. The failed in the box now isn't helpful for anyone who listens to advice. More helpful than the box would be a link to Wikipedia:Requests_for_adminship_by_year —valereee ( talk) 14:44, 20 May 2021 (UTC)

  The link is in the navbox at the top. isaacl ( talk) 14:52, 20 May 2021 (UTC)

  So then why do we even need this box? Someone thought it was pretty and helpful at one time -- a time when it was a ticker, I assume? When there were so many RfAs that they aged out of the box in days or weeks? —valereee ( talk) 20:03, 20 May 2021 (UTC)

  Yes, I did say that people can just look at the archives for recent results. It was created in 2013, where there was a significantly greater number of RfAs compared with the last few years. isaacl ( talk) 20:38, 20 May 2021 (UTC)

  There were like 96 RfAs in 2013. Proposal: Let's just get rid of this as obsolete. —valereee ( talk) 22:22, 20 May 2021 (UTC)

  74; 96 was in 2012. isaacl ( talk) 00:08, 21 May 2021 (UTC)

• I've cut it back down to the last three. NASCARfan0548's unfortunate experience four and a half months ago has finally left the most "recent" RfX box. I'd be in support of any of the following outcomes: only the past 90 days, only the last three, or only the past 90 days with a minimum of the last three (regardless of age). I understand that the first one would result in an empty box sometimes, but so be it. Maybe it'll result in us debating the pros and cons of the RfA process for the nth time this week. Anarchyte ( talk) 14:48, 24 May 2021 (UTC)
• Agree with Valeree, lets celebrate the last few successes but not list the unsuccessful. Ϣere SpielChequers 21:22, 24 May 2021 (UTC)
• Listing only successful RfAs and keeping it at the last 7 seems reasonable to me. My second choice is 3 RfAs or 90 days, whichever is greater. KevinL (aka L235 · t · c) 01:38, 25 May 2021 (UTC)
• Disagree that unsuccessful RfAs should instantly disappear. At a minimum, they should stay for at least seven days, so people know what happened to that RfA that had appeared on their Watchlist and disappeared, or that they commented on or meant to. Anyone who started an RfA that failed should expect that people should easily be able to find it for at least a while and see what happened. For those that failed after a full seven days, it does no one any service to have them disappear immediately upon closure, especially if there was a 'crat chat that determined the result. Perhaps those should stay for a month. Also disagree that the table should go away: it's probably encouraging for potential candidates to see that people do make it through when they try. For the RfA table size, I'm fine with a max of 7, a min of 3, and a 90-day sunset so long as the min of 3 is retained. BlueMoonset ( talk) 04:08, 25 May 2021 (UTC)

I've moved an essay to Wikipedia:Should you ask a question at RfA? which I'd love input on, with an eye to perhaps adding it to the advice template. —valereee ( talk) 16:44, 26 May 2021 (UTC)

• Looks good to me; can't think of any improvements to make. What do you mean by "advice template"? My first thought was linking at Template:Editnotices/Group/Wikipedia:Requests for adminship, under the questions bullet point; and/or adding it to the question preamble in RfA pages themselves—a good proportion of the target audience might not see it anywhere else. — Bilorv ( talk) 18:36, 26 May 2021 (UTC)

  At the very top of this page and Wikipedia:Requests for adminship there's a nav box with links to various advice/info pages for RfX.

  I was thinking it might be helpful in the preamble, too. I don't know that it would make any clueless people more clueful, but it would give the rest of us the opportunity to act all righteous as we point out the shortcomings of others. That's always fun. —valereee ( talk) 19:02, 26 May 2021 (UTC)

  Valereee, you summed up the entire RfA process in those last ~20 words :D —— Serial 19:09, 26 May 2021 (UTC)

  Possibly all of Wikipedia space. —valereee ( talk) 19:30, 26 May 2021 (UTC)

  Ah of course, yeah it'd be good in that navbox too. — Bilorv ( talk) 19:12, 26 May 2021 (UTC)

• I made a few edits. It's a decent essay. - tucoxn\ ^talk 19:36, 26 May 2021 (UTC)

Wikipedia:Requests for adminship/Less Unless

Thank you to all of those who has devoted their time to participate in recent RfAs. I wanted to say how important this has been for me but also how stressful. The candidate’s position is very vulnerable from a number of perspectives and I wanted to talk about them.

The first and probably the most emotionally draining fact is that you are being scrutinized by such a diverse community and it’s impossible to be perfect. I fully support the idea that the community should pay careful attention to the candidate’s activity because adminship is a huge responsibility, but this doesn’t make it easier. Especially frustrating is the fact that you can’t explain yourself or clarify things in the discussion section or on the talk page (technically you can, but there's a strong feeling it's inappropriate). Meanwhile the majority of the concerns are voiced there. Thoughts and ideas evolving in those discussions are just to observe and this is very nerve wracking as some of them are just misunderstandings or misinterpretations. The only sphere of your influence is the answers to the questions, apart from that you feel quite helpless. It’s also worth mentioning that written communication has a number of draw backs – especially under pressure it’s hard to achieve perfect wording that would definitely be unambiguous. We all have different backgrounds, we come from different communities and countries – these things influence what we say, how we say it and what we mean.

I believe opposition is important – it can show a number of perspectives which could become a departure for improvement. However, knowing that you can potentially be criticized for every mistake you’ve made is another challenge to take.

Is has been very emotional for me and there was a point I was about to withdraw and I know I wasn’t the only one having those thoughts. Luckily, I had people who supported me. Right now, when the RfA is over, everything doesn’t look that bad, but that’s my exact point. I am grateful to be a part of this community. My little wish would be to make RfA process less stressful as many people just might not have the emotional resilience and access to emotional support to deal with it. Kindest regards, Less Unless ( talk) 19:48, 26 May 2021 (UTC)

Thank you, Less Unless.

We need a collection of recent debriefings. —valereee ( talk) 19:58, 26 May 2021 (UTC)

I've started a listing: Rfa debriefings —valereee ( talk) 20:05, 26 May 2021 (UTC)

On a related note, yesterday I went through 20 editors, who either had existing privileges like Page Mover, New Page Reviewer or Autopatrolled, or otherwise led me to believe there was potential to become an admin at some point. For each one, I found myself thinking one of; "no, too many people will bring up those AfDs where the !vote didn't match consensus", "no, only 4 AfDs", "no, they had an article deleted at AfD four months ago", "no, there's a warning template", "no project participation, they'll get 'no need for the tools' opposes", "no obvious major edits to an article, they'll get 'content creation' opposes", "there's a recent argument on a talk page, that'll come back to haunt them". And that's far from the first time I've done that. Having done a fair amount of nominations now, I've seen what people can oppose over, and what will get criticised, and it seems a whole bunch of people who might be able to contribute something to the maintenance side will get shut out. Not because the standards are higher overall, but because the standards are diverse, and can be almost anything.

I know how distressed Less Unless felt about the RfA, and I'm pleased they stuck it out and passed with a high level of support. Ritchie333 ^{(talk) (cont)} 20:02, 26 May 2021 (UTC)

I've had similar experiences -- looked at editors who would be good sysops, but there's something there that would raise opposes. It sucks. Not sure standards aren't higher. Sometimes it looks to me like anyone who has worked in contentious areas is just SOL. Which sucks. —valereee ( talk) 20:36, 26 May 2021 (UTC)

When I am looking at a group of editors who might be eligible, rather than some editor who has been recommended to me in some way, I too find more often than not that I rule out most editors. Mostly that's because I don't think they're ready but sometimes it's because I don't think they could pass. However, I think those of us who try to find RfA candidates need to be careful about ruling someone out just because I don't think they could pass. In effect then we become the gatekeepers of whatever norms we perceive at RfA. So a majority of the time if I think someone will be a good administrator I will approach them to see if they'd be willing to run while also being upfront about the challenges they face. That way they are making the choice and know at minimum that someone thinks highly of them, which is its own kind of barnstar. Most of those times they do decline to run but sometimes that turns into a plan to run down the road. If they do run, I also try to make explicit how their candidacy is going against a perceived norm. With Less Unless who I didn't find (that was all Ritchie) I was excited to co-nom and explicitly make reference to them not having GA/DYK/FA "bling" despite being a content creator. And they passed. So hopefully that can be used as an example of how maybe RfA isn't impossible the next time we find someone good like her (we should all be so lucky to have the confidence of 98% of the community). Best, Barkeep49 ( talk) 21:09, 26 May 2021 (UTC)

they had an article deleted at AfD four months ago -- perhaps we should be calling that a qualification for an otherwise strong candidate, considering how important empathy for people on the wrong end of AfD is? Vaticidal prophet 10:34, 27 May 2021 (UTC)

Hope my questions didn't cause you too much stress in particular - glad you stuck it out. Elli ( talk | contribs) 02:26, 27 May 2021 (UTC)

Hi, @ Elli:. Actually I felt quite confident answering the questions - I believe it was a useful procedure for both me and other users to see the extent of my knowledge and the ability to articulate it in an understandable and structured way. The biggest sources of stress were spheres out of my control and it's where good coping strategies are needed for those who are more anxious and less emotionally resilient. Generally, when people are dependent on something that feels uncontrollable, the anxiety rises and influences their health and the way they perceive situations. Often people who feel helpless about something they depend upon, characterize it as a traumatic experience; in contrast the level of anxiety drops once they gain control and feel liable. So the question I'm asking myself now - is it possible to somehow eliminate or at least decrease the influence of those stress factors? Less Unless ( talk) 08:41, 27 May 2021 (UTC)

@ Less Unless: glad I didn't cause stress :) as to your other point... yeah, that's a concern. Though, I do think it's good, in a way, that adminship candidates generally stay out of the discussion other than questions. I can see why that would lead to a sense of helplessness. Elli ( talk | contribs) 10:35, 27 May 2021 (UTC)

(really, I feel like this is kinda a social flaw of the wiki model in general - permanent records, and communication that isn't real-time. terrible for me, at least, which is why I was quite shy in approaching people until recently) Elli ( talk | contribs) 10:36, 27 May 2021 (UTC)

If I may ramble about this, I want to share a thought that has been bugging me since I actively returned to the project earlier this year. It feels like nowadays, if you want to pass RfA, you almost need to decide from nearly minute one (or edit one, in Wikipedia's case) that passing RfA is your goal, and "work" towards it. (There are some exceptions to this, and the two latest successful candidates may fall outside this, I am not familiar enough with them to know.) This is similar to a 'career politician' in some respects. I note Ritchie333's comment at 20:02, 26 May 2021 (UTC), which appears to be saying something similar.

Back in my day, it was totally different. I made tens, potentially hundreds of errors before running for RfA. That didn't stop it being successful. If I ran for RfA today, with the same experience, the same errors, the same edit set, as what I did when I ran over a decade ago, I'd have finished less than 50% and probably wouldn't have seen out the first day. Admittedly I am not a perfect administrator, but I'd like to think my involvement is a net positive and I haven't broken anything too major!

I think, in a nutshell, our standards for 'mistakes' are now too low. Someone who shows consistently rubbish judgement even after being here 12 months? Sure, no dramas. But someone who makes some dumb newbie mistakes (the time period that this defines also being an issue currently), the occasional minor lapse, potentially not the most perfect candidate in all areas but demonstrating reasonable judgement in the ones they do 'tick the box' in - this has to be enough for us, moving forward.

The way RfA is set up with the oppose section leads to really prejudicial rationales being given, with limited chance of a nuanced discussion occurring because a) it's frowned on to jump on opposers and b) the nature of an RfA means any emotion-heavy conversation is risky. My personal view with RfA is we need to throw our NOTAVOTE mantra - that applies through the rest of the project - out the window, and be realistic with RfA. It is a vote, and actually, being a straight vote is likely to make it less stressful for nominees. If an oppose vote wants to explain on the talk page, they can - and an extended conversation can happen there. If they don't want to explain, no dramas, no-one will question them. Supports can also do the same thing. But it's out of the way, not interfering with the key activity occurring (which is a vote, lets all be real), it isn't compulsory to explain, and it will hopefully make it less painful for all concerned. This idea may have been explored and consensus opposed it over the past few years, I'm not sure, but it feels like a potentially beneficial change from where I sit.

I disclaimed the above by calling it ramblings, and it is exactly that - I'm a bottle of vino into my Saturday evening stuck hidden away from the cold, so apologies if it doesn't flow particularly well or make sense in parts. Would love to hear perspectives on the various parts of the above, I am sure people will have diverging views on some, if not all, of it! Cheers, Daniel ( talk) 12:00, 29 May 2021 (UTC)

RfA should be a vote. The "discuss to find consensus" model just doesn't work when there is a choice to be made between exactly two possible outcomes. What we do (allowing people to campaign for their point of view while voting, and calling reasoned discussion of oppose votes "badgering") is a bad way to vote and a bad way to discuss. — Kusma ( talk) 16:08, 29 May 2021 (UTC)

+1. It would probably be better to just have a straight vote, similar to arbcom. Levivich 17:29, 29 May 2021 (UTC)

Stew elections are votes, so why not RfA? Java Hurricane 07:18, 30 May 2021 (UTC)

Are we sure the voting model would make opposes less stressful? The 'getting jumped on' seems to create a bar to opposing currently, especially if ones oppose rationale is spurious, which for many people probably means they'd only oppose when they have a substantive reason to (early on, anyway). I suppose a wall of signatures could be less stressful, but it's probably worth testing the hypothesis first; perhaps some of the regular nominators (@ Ritchie333 and Barkeep49 for example) could ask if this would change the mind of some of the candidates that rejected their offer to nominate? ProcrastinatingReader ( talk) 09:34, 30 May 2021 (UTC)

I am personally currently exercising restraint when looking at idiotic oppose rationales. If you ask me, spurious opposers are too well protected in the current RfA atmosphere, where pointing out minute errors of the candidate is acceptable, but pointing out idiotic nonsense statements by oppose voters is considered badgering. — Kusma ( talk) 09:52, 30 May 2021 (UTC)

I'm skeptical that a private vote would make things less stressful. I base this on she the fact that the experience LU writes about happened as she could see was passing with 98% support. So all the doubts couldn't change things. I also base it on the fact that having run twice on a secret vote, for ArbCom, I can tell you that is quite nerve inducing because you have no sense of how it's going in the middle of things. But Daniel isn't the first to suggest that this might help things so I wouldn't rule a secret vote out even if it isn't my preferred RfA remedy. Best, Barkeep49 ( talk) 17:30, 30 May 2021 (UTC)

Another option would be an open onwiki vote with comments prohibited in the vote (only support and oppose; not sure neutral would be useful). Everybody who wants to comment goes to the talk page.-- Ymblanter ( talk) 17:41, 30 May 2021 (UTC)

@ Barkeep49: Since you mention it, what is your preferred RfA remedy? ProcrastinatingReader ( talk) 17:49, 30 May 2021 (UTC)

Ymblanter's proposal is my thoughts exactly. I see it as the best of both worlds. Daniel ( talk) 21:39, 30 May 2021 (UTC)

I'm not sure of the net effect on the candidate, but I think having a straight vote would help streamline discussion, since it wouldn't be scattered across support/oppose comments from different people, as it is now. This would reduce redundancy in comments, and the candidate wouldn't have to read similar oppose rationales over and over (since in the current model, many people expect each oppose to provide a rationale). (Back in 2015 I proposed structuring requests for administrative privileges to determine a consensus of the pros and cons of a candidate and to weigh their relative importance, but it didn't get much support.) isaacl ( talk) 23:12, 30 May 2021 (UTC)

I preferred knowing what the opposes were based on for mine, as there were multiple distinct threads. I would oppose any attempt to change to an straight vote, especially since it would do away with the discretionary band that I also support. Finally, I think it would make it easier to oppose, which means all else being even, fewer RfAs would pass. Nosebagbear ( talk) 09:25, 31 May 2021 (UTC)

Many RfAs that fail have one or two opposers presenting Some Problem and then a bandwagon follows. Discretion might be good, but I'd like to hear from victims of bureaucrat chats how they feel about extending the agony for another week before recommending that we continue in this way. — Kusma ( talk) 09:33, 31 May 2021 (UTC)

A straight vote isn't my first choice, but I imagine there would still be consolidated discussion threads (as opposed to per-vote threads), and thus candidates will still know what are the key concerns of those who are opposed. isaacl ( talk) 14:20, 31 May 2021 (UTC)

Tommorow it will be 3 months since the last new Rfa, and nearly as long since the last "why aren't there more Rfa's" discussion here. Just saying. Johnbod ( talk) 15:28, 11 May 2021 (UTC)

Let's skip ahead and pick up from the most recent discussion in February, Wikipedia talk:Requests for adminship/Archive 258 § Nobody wants to run. Barkeep49's compendium of issues and proposals for changes to the process may also be useful to review. isaacl ( talk) 16:38, 11 May 2021 (UTC)

I'll just say you're not the only who has noticed that it's been a while since we've had a candidate. Barkeep49 ( talk) 16:45, 11 May 2021 (UTC)

I have been looking at a run but since I am currently under a few editing restrictions, I am hesitant to put myself forward because I fear that will just get jumped on. The C of E God Save the Queen! ( talk) 16:50, 11 May 2021 (UTC)

• Prior to this latest dry spell, the longest we've gone without a successful RfA was 76 days, from Wikipedia:Requests for adminship/Philg88 (21 August 2014) to Wikipedia:Requests for adminship/Jackmcbarn (5 November 2014). This latest dry spell is 78 days and counting, and that doesn't include a further 7 days that would have to happen if a successful RfA started today. So, the record has been beaten by 9 days and counting. As seen in WP:RBM, we've never had three months in a row without a successful RfA. This is only the second time (the other being the aforementioned record dry spell) we've had two consecutive months of no successful RfAs. If we don't have an ultimately successful RfA start within the next two weeks, it will be three months. Ok, the sky isn't falling. But, as I've said before, we need to start planning for a post-admin era. In some ways, we're already in it in terms of the current admin corps being able to keep up with required maintenance. -- Hammersoft ( talk) 19:51, 11 May 2021 (UTC)

  I would suggest rather than a post-admin era, I think we need to be planning for an era of insufficient editor maintenance hours, of which admin is but one element. I would love to say we should plan for this (or even plan for a post-admin era) but I don't think that's going to be possible at the moment. Instead I think we're going to need to experience more pain from these shortfalls before we'll be able to get consensus to enact the kinds of reforms that would address the issues. Best, Barkeep49 ( talk) 20:04, 11 May 2021 (UTC)

• The pain won't be sufficient to cause action to happen. Commons bears sad testimony to this. Commons is becoming a failed project, and is doing so rather silently. -- Hammersoft ( talk) 20:06, 11 May 2021 (UTC)
• Editing volumes are significantly higher now than during the late 2014 minima per Wikipedia:Time_Between_Edits. Adminship, and especially our ability to recruit new admins is the problem. I think it would be helpful to get the WMF to do a marketing survey - survey the potential admins and ask them why they haven't run and whether they might in the future. If this was a marketing survey rather than a research survey it might encourage a bunch of people to run, and at least give us an idea as to what the barriers are that dissuade people from running. Ϣere SpielChequers 20:16, 11 May 2021 (UTC)

  WereSpielChequers, As i said in the thread linked above, I was invited to a workshop hosted by Monika Sengul-Jones ( Shameran81) about helping to counteract systemic bias, examine our reliable source guidelines, and how we could ensure communities aren't marginalised. Towards the end of our discussion, I mentioned that I'd spent years trying to find suitable admin candidates, found it difficult, and wondered why so many decline. There wasn't a clear and obvious answer, but a common theme from those participating is that RfA wasn't a "safe space" (their words, not mine) and that it wasn't worth the hassle. Ritchie333 ^{(talk) (cont)} 20:59, 11 May 2021 (UTC)

I can think of at least two people who are not admins for reasons that appear to involve being afraid of the extra challenges and criticism that can come with acting with the mop, but who work in areas where I strongly suspect they would get much, much less criticism acting as admins than they do currently, without changing anything else at all about their behaviour. I think broadly there's a sense amongst a significant proportion of the "established newer editor" base (who would otherwise be becoming RfA-eligible) that being an admin is a terrible burden and RfA is the worst thing to have ever happened in all humanity, which...stems RfA numbers. There's also a bit of obsession about "need for the tools" discouraging people; I can think of at least two people who would be brilliant admins and work in an area constantly begging for "we need more admins" who can't see any reason they would need a mop, let alone people who may or may not strictly 'need' one but would make the project a better place if they had it. Vaticidal prophet 21:15, 11 May 2021 (UTC)

@Ritchie333 I get that RFA is going to deter some people because of not being a safe space, and an overlapping group who see it as not worth the hassle. Both of those would need to be in a survey. But are either of those any different than in 2008? Yes, our requirements for content contributions are higher than in that era, but are we less a safe space or is adminship more of a hassle than in 2008? The only change I can think of has been the desysops of the last year or so, the flipside of judging admins by a higher standard could be that some people now see adminship as too risky for them. Ϣere SpielChequers 21:40, 11 May 2021 (UTC)

• Stay tuned. – ♠Vami _IV†♠ 21:00, 11 May 2021 (UTC)
  • OK I have something personal to contribute to this conversation now. Within the hour after making the above comment, I was contacted in confidence by an admin who told me that an unnamed user gave them guff about me canvassing. I did think about whether posting that I was going to run, like what C of E did above, would be canvassing. Because I'm taking this seriously. But I did anyways, to say that the RfA drought would be ending soon. And I was immediately suspected of canvassing. Additionally, rather than contacting me, they went to an admin who is not even my nominator. They complained about me saying I'm going to run for admin to a completely unrelated person to me. Small wonder to me now why RfA seems so scary, why even ORCP is marked by many users, sysop or otherwise, "HERE BE DRAGONS". – ♠Vami _IV†♠ 22:49, 12 May 2021 (UTC)
    • I have zero idea how a comment like the one above had any connection to our policy on WP:CANVASSING. Seems bizarre. Best Wishes, Lee Vilenski ^{( talk • contribs)} 10:18, 13 May 2021 (UTC)
    • With respect to whomever that complained, I completely disagree and that should not deter Vami IV. Sdrqaz ( talk) 16:40, 13 May 2021 (UTC)
      • Indeed, if anything should deter Vami IV, it's threads like these, particularly if one intend to work that area. —— Serial 10:36, 31 May 2021 (UTC)

        I see him overshooting the mark, being called out on it and graciously accepting the correction. Looking at that thread in isolation, I don't see a problem. — Kusma ( talk) 11:10, 31 May 2021 (UTC)

        • I would not have expected you to? —— Serial 11:28, 31 May 2021 (UTC)

    • I echo Lee Vilenski and Sdrqaz. I do not see how that was canvassing. Scorpions13256 ( talk) 21:03, 13 May 2021 (UTC)

    Vami IV, I think if you ran for RfA next week, and somebody brought forward this as "canvassing", they would get short shrift very quickly and result in a bunch of pile-on support to counteract it. I don't think you've got anything to worry about. Ritchie333 ^{(talk) (cont)} 14:04, 14 May 2021 (UTC)

• Most Wikipedia editors are still suffering from a sort of mass delusion that being an admin is some sort of superior position, while believing that any complete idiot can be an editor. This is the equivalent to having extreme vetting for the office cleaners but letting anybody off the street take the position of chief scientist. That's what really need changing. Nigej ( talk) 21:36, 11 May 2021 (UTC)

• You don't fancy getting the dustpan and brush out Nigej? :P I think you'd be fantastic Best Wishes, Lee Vilenski ^{( talk • contribs)} 21:43, 11 May 2021 (UTC)

• Perhaps after weighing the pros and cons of it all - most folks just don't see the effort/stress as being worth the hassle. — Ched ( talk) 00:02, 12 May 2021 (UTC)
• FWIW, I had an admin approach me about possibly running last year, I believe. We had a candid conversation about it, and ultimately decided not to move forward. I think some of the factors were:

1. Having run once before, I'm aware that it can feel like the slightest slip can sink your entire nomination. RfA can feel a bit like an exercise in sadomasochism.
2. I'm not sure I have the desire or ability to invest further in Wikipedia in terms of time/effort.
3. I'm not sure what being an admin would do for me in terms of enhancing my enjoyment of editing.
4. I'm not sure what being an admin would do in terms of enabling me to help WP more effectively than I already do. Why apply for admin if I can't easily form a good argument as to why I "need" the tools?

In the end obviously nothing came of it. Maybe I wouldn't pass RfA, maybe I shouldn't be running in the first place, but I think #4 may merit some consideration in terms of whether the admins are effectively conveying areas where they need more help in the form of additional admins. Maybe there is something else I could do, that I would enjoy, and that wouldn't significantly alter the amount of time/effort I put into editing that I'm simply unaware of? At least in my case, learning about the nuts-and-bolts of WP has largely involved self-discovery, and years later there are still things I know virtually nothing about simply because I've never had occasion to research them.

Hope this is helpful! DonIago ( talk) 02:59, 12 May 2021 (UTC)

• I think I can echo WereSpielChequers' comment: The only change I can think of has been the desysops of the last year or so, the flipside of judging admins by a higher standard could be that some people now see adminship as too risky for them. Kudpung กุดผึ้ง ( talk) 16:04, 12 May 2021 (UTC)
• A few points:
  1. The standard has gone up. We have (as of now-ish) 309 users with >7500 edits who signed up in 2018 or 2019, and 6 of them are admins. I am sure that a much higher percentage of editors with that experience became admins in 2006. The list of non-admins is at https://quarry.wmflabs.org/query/55033 .
  2. The WP:AFC backlog is as bad as anything, and we don't need more admins to fix that. On the other hand, WP:UAA is often backlogged.
  3. For myself only, I would be more likely to run if there were precedent or consensus at a recent RFC that binding recall pledges are binding, and if there was a standardized procedure based in site policies for how to conduct them.

  User:力 (power~enwiki, π, ν) 18:36, 12 May 2021 (UTC)

  • @力 Thanks for the stat on 309 editors from 2018/19 of whom only 6 are admins. Good to hear there are so many editors from that wiki generation. I don't think our problem is the standards having gone up as I doubt if many perhaps any of those 309 have failed an RFA. OK some of them will have recent blocks or other things that don't go down well with the RFA crowd. But an active editor whose been around 18 months, can show a need for the tols and some referenced content they have contributed, they'd likely have an easy run at RFA. Ϣere SpielChequers 20:12, 12 May 2021 (UTC)
• I noticed. I have been in discussion with a couple of candidates, and while adminship isn't supposed to be a big deal, RfA is definitely seen as one. I have no suggestions for how to fix that. @ 力, I am not sure I'm understanding your comment "I would be more likely to run if there were precedent or consensus at a recent RFC that binding recall pledges are binding, and if there was a standardized procedure based in site policies for how to conduct them." I think you're saying you'd be more likely to run if there were a better community desysop procedure? I agree that we need such a procedure, but why does that make you more likely to run? —valereee ( talk) 19:12, 12 May 2021 (UTC)
  • The short answer is that I don't want debate over the technical details of recall during an RFA. User:力 (power~enwiki, π, ν) 19:46, 13 May 2021 (UTC)

I have to assume this is due to more people interested in contributing to build the encyclopedia then seeking extra tools which is a good thing. I'd be happy to try the RFA but I've been on a long hiatus and would be heavily opposed on. NYC Guru ( talk) 10:08, 13 May 2021 (UTC)

There's a reasoning error here. Specifically that which implies those with extra tools cannot 'build an encyclopaedia'. Nobody is forced to edit in any particular way, regardless of how many technical restrictions they have or don't have. (Or the alternative interpretation, that the time spent seeking RfA is wasted time, is the Lincoln fallacy [good chance this was not said by Lincoln, of course]) ProcrastinatingReader ( talk) 10:16, 13 May 2021 (UTC)

The actual time sink into RfA is more around worry and planning than the actual RfA itself, which takes less time than you'd think - there's just a couple questions to answer each day, and by day three it's pretty plain where it's going. Best Wishes, Lee Vilenski ^{( talk • contribs)} 10:26, 13 May 2021 (UTC)

• Lay out why you-all think admin is important, draft a help wanted add that convincingly says 'we need you to do x'. Write an op-ed for the Signpost. You often get emphatic but opaque comments in Rfa such as, 'we need more admins', but it's presented as if it's a self-evident truth -- it is not self-evident, it needs a convincing rationale, better, several convincing rationales. I've asked before on this page 'how many admins do we need, and why?'-- Alanscottwalker ( talk) 10:43, 13 May 2021 (UTC)
  • The truth is we don't need more admins, but we would benefit from an additional amount. The thing about not having RfAs is that we are having more and more desysopings, so we aren't even living in a status quo, we are losing net admins all the time. The more admins we have, the more the backlogs go down, the less time it takes for things like revision deletion/protection/CSD actually take and things like AfDs get closed closer to when they are supposed to. I think you'd be a fantastic candidate, drop me an email if you fancy talking it through Alanscottwalker. Best Wishes, Lee Vilenski ^{( talk • contribs)} 10:57, 13 May 2021 (UTC)

    Lee Vilenski, So, 'we need more admins to help . . . . If we don't have more admins, then . . .'. Thanks, but, as for me, I'm in the long - long - standing, 'just, no interest in the position, camp.' -- Alanscottwalker ( talk) 11:25, 13 May 2021 (UTC)

    • I mean if you can't realise that most of our important jobs (specifically legal for copyright and personal attacks, and filtering to stop the place being overrun by vandals and trolls) are undertaken by admins. If we continue to lose them at the net loss we have now, the place will grind to a halt. Best Wishes, Lee Vilenski ^{( talk • contribs)} 11:29, 13 May 2021 (UTC)

      If you can't realize that you making people guess what you think is important for new admins to do, than that's on you. There may be some people who may want to do 'legal' so tell them, advertise we need them to do 'legal', if you think that is what is important. -- Alanscottwalker ( talk) 16:48, 13 May 2021 (UTC)

  • I think it is quite self evident. What +sysop does is remove technical restrictions. Less technical restrictions generally cannot hurt, even if one does not use their new buttons. There are people complaining about backlogs at AN frequently; RFPP is at AN weekly, there are currently 60 ANI threads (many of which could do with an admin moving them forward), AIV is frequently backlogged (see User:Enterprisey/AIV analysis), etc. Are any of these going to kill the project? No. But it would run smoother without them. Those backlogs become less either by: (a) existing users who have technical access investing more of their time to process them; (b) more users given technical access. As for how many more: there is no upper limit on the number of admins. Many of our backlogs are artificial, by use of CSD tagging or a person compiling instructions requesting an admin carry them out, with many editors submitting hundreds of such requests frequently. That unnecessarily takes up double the time. ProcrastinatingReader ( talk) 11:06, 13 May 2021 (UTC)

    Well, that took quite a few words, would you need so many words, were it self-evident? Alanscottwalker ( talk) 11:12, 13 May 2021 (UTC)

    No, but you felt it wasn't self-evident, so I felt it worth explaining from my POV some rationale as to why I feel it is. I think it's self evident mostly because of the third sentence. However, self-evident does not mean a lengthier argument is non-existent or impossible. ProcrastinatingReader ( talk) 11:24, 13 May 2021 (UTC)

    I think that sentence is an answer to "why should I accept additional privileges if they are offered to me?", but not to the question that was posed: what is the rationale for the statement "we need more admins". To me, the question is not so much that we need more admins, but we need new admins, to support turnover of admins. This alleviates pressure on long-time admins and helps with ensuring that administrator views are as diverse as the editing community as a whole. isaacl ( talk) 16:13, 13 May 2021 (UTC)

    That's fair. You're right; the two questions are slightly distinct. ProcrastinatingReader ( talk) 16:18, 13 May 2021 (UTC)

    • Not all backlogs are equal: some are important and others are not. The important maintenance tasks are not backlogged. The unimportant maintenance tasks (like CfD) are backlogged, but no one cares, and rightfully so. ANI, AIV and RFPP get backlogged because of a ton of bad reports. I don't think it will help the project run more smoothly if CFD, AIV, RFPP, or ANI (or AN or AE or many many other queues) were clear of backlogs. I doubt anyone would even notice. When we have over 500 active admins and like almost none of them spend any time on those perennially backlogged queues (because everyone has more important things to volunteer their time for), it's really hard to argue that the encyclopedia's fate depends on more people getting the bit. Unimportant maintenance tasks being generally ignored is not a problem that needs solving IMO, and it's actually a good thing to let those unattended areas die a natural death. That's why I've never found "we need more admins because of backlogs" to be persuasive. Levivich  ^harass/ _hound 13:54, 13 May 2021 (UTC)
      • Levivich, An often-forgotten task of admins is to reply to an AIV report saying "that's good faith, not vandalism", close an ANEW report as a "storm in a teacup", or decide that an ANI thread is a time sink and not get involved. These are best coming from a variety of different admins. Ritchie333 ^{(talk) (cont)} 14:27, 13 May 2021 (UTC)
      • But this seems like an argument about whether certain areas "matter"? I mean for the most part I don't disagree with you. As I say none of them are going to kill the site (but then again, it's difficult to think of any kind of neglect that is). Taking your example, if we did get CfD down to nil, presumably our categories would be better structured. Would it make a difference in the big picture? Maybe not, but that seems more an argument for why the category system might be make-work.

      I think at the end of the day any person can disappear, and any area can become backlogged, and the effect is not noticeable in the big picture (or people just adjust to the aftermath). For example: Bbb23 is no longer a CU, and he dealt with a large amount of socks. Consequently, the number of sock blocks has substantially decreased, and Bbb23 was responsible for about half the sock blocks done. But do you get the feeling there is a lot more abuse on the project as a result? I don't. But I'm not sure I'd equate that with SPI is useless. Same reasoning extends to any area I think. More GA reviews => better articles. But only 10 GA reviews last month? Nobody would really notice in 'the big picture'. The readers of those articles probably will, though.

      While there may be justification for why a certain backlog is useless, until such time the community decides to get rid of it I don't see any reason why it shouldn't be efficiently processed. And since admins don't cost anything, there's no real barrier to achieving that efficient processing. It just requires more permissions be doled out. ProcrastinatingReader ( talk) 14:41, 13 May 2021 (UTC)

      • I think there's some truth to what you write Levivich but also it becomes self-fulfilling arguing that if something isn't being done it's ipso facto not vital enough to need doing. At the moment we've got quite a few editors investing time at CCI which had, for years, experienced huge backlogs. I would suggest that was work that always needed doing and having editors to address that backlog is a good thing. However it is an example of where something that is important might not have been urgent. So you're likely right that the urgent tasks are getting done but I see some things at ANI where admin intervention early on (perhaps before it reached ANI) could have addressed a problem that then spiraled into something that just became too much to address. For me backlogs are just one of the reasons I want to see more admin but to avoid us becoming Commons it remains a reason. Best, Barkeep49 ( talk) 14:42, 13 May 2021 (UTC)
  • Only addressing the need issue, not the RfA issue: Yes, we do need more admins. Roughly, we have two kinds of processes which require admin work. Type I are processes with a lot of traffic and a lot of admins responding. Examples are RFPP, AfD, and AIV. These typically need to be acted on quickly, and quickly here means different things - minutes to hours for AIV and weeks for AfD. Here, a lot of admins are working, but backlogs are clearly growing. I am sure I would be able to bring statistics, but for now I will go with anecdotal evidence. I am one of the most active RFPP admins, and I see bad backlogs (say more than 30 pages or longer than 36 hours) now much more often than say two years ago. We have an AN announcement that RFPP is backlogged almost every week. I am not doing AfD for years now, only occasionally, but I was one of the most active AfD closers since after I got the flag, in 2013? for a couple of years. I have never seen a more than three weeks backlog there. Even three weeks was a catastrophe and was considered as something requiring an immediate attention. Now apparently five weeks is not unheard of. The backlogs will be growing and growing, and if you want to see how it is going to look like look at Commons where there are backlogs of seven years or longer. FfD on Commons (the core process) could easily take half a year, and this is considered normal. Furthermore, Type II are areas maintained by one or two admins. MfD is one example, I have heard that the main page is another one, but do not hold me to this one. When the admin is alive and well, we do not see backlogs, but if they become inactive or desysopped, we for some time have zero admins, and the tasks just pile up without being addressed, and eventually some admin takes the task, but stops doing something else. Most of these tasks are much less important than Type I tasks, but they exist and need to be done. There are four general solutions to these backlogs: (i) more admins; (ii) unbundling; (iii) bots; (iv) abolish or ignore the tasks. Bots, I am afraid, are already doing whatever they can so, for unbundling, one would need to come up with some good ideas, and this does not happen very often, and there is some reason why these tasks exist. It looks like more admins is the easiest solution - to this very real problem.-- Ymblanter ( talk) 18:25, 13 May 2021 (UTC)

    Bots can do more. For example Wikipedia:Bots/Requests for approval/QEDKbot to handle WP:C1 deletions. I believe QEDK shut the bot down after some criticism but its approval and premise remains valid. There are other processes that follow a formula and afaik don't really require human judgement. ProcrastinatingReader ( talk) 18:54, 13 May 2021 (UTC)

    If this is the case, then buts must do it, otherwise it is waste of admin time.-- Ymblanter ( talk) 19:07, 13 May 2021 (UTC)

    Bots can do more, but an admin bot has to be controlled by admins. It makes no sense for a bot to have greater privileges that the user who controls it. Which brings us back to a need for admins. Hawkeye7 (discuss) 19:40, 13 May 2021 (UTC)

    @ Ymblanter: The argument some admins made is that they enjoy doing the work and it only takes 15-30 mins each day. IIRC this was the main criticism against QEDK bot. That's fine, I guess, but if those admins later find a different hobby or change areas of interests for some reason, then there's no guarantee QEDK will be around to make and iron out another bot. ProcrastinatingReader ( talk) 13:44, 14 May 2021 (UTC)

I mentioned this elsewhere on-wiki a few months back, but essentially the reason I look for admin candidates is:

• There is no limit on administrators, and it shows people that the admin corps is not "us" vs "them"; it is possible to join
• It reduces any possible backlogs; while people won't rush off and start clearing them off, spreading the workload results in the old adage "many hands make light work" as many admin tasks can be done in isolation from others
• It's good to get new people coming in, with potential new and fresh thoughts and ideas; the admin corps is not just tired old faces of "the old guard" who've been there since time immemorial
• A popular editor passing RfA makes the community feel good, and gives a brief rest from the regular sense of deflation as seen in threads like these

Ritchie333 ^{(talk) (cont)} 11:29, 13 May 2021 (UTC)

( edit conflict) I think the argument about "need" for the tools is an interesting one and has always struck me as a slightly puzzling reason when I see people oppose RFAs on those grounds ... it usually seems to me to be covering for another reason (to oppose). To some extent, I think it can be hard for prospective admins to know where they will end up using the tools until they have them and see what they find doable/not draining/worth doing/whatever makes them tick. The conventional path / answer to the question seems to be some combination of RPP, AIV, AFD, and CSD, and so I wonder whether—given that those are the most visible areas where admins work—those who do not have a lot of experience there are discouraged from running because they feel an expectation that that is where admins are "supposed" to work or need to have experience in order to pass an RFA. Of course, I think most of us agree that experience in those areas is mostly a proxy for good judgment. Anyone who has worked in content spaces and is a generally trustworthy/competent editor for a long time should have a decent enough handle (or be able to develop a decent handle) on those areas to be successful there. All of this is to say, I suppose, I wonder whether one barrier might be not feeling competent in all of the "basic admin areas" and thus deciding one does not have "reason to use the tools." Maybe a better training program—formal or, perhaps especially, informal—would help ease those fears and encourage more content-oriented folks to give it a go? I would endorse a survey trying to learn more about what holds people back from running. I suspect if all you see of adminship is those few areas above plus the drama boards, it wouldn't be particularly appealing for someone who mostly enjoys writing articles. But perhaps if we could convince those folks to pick one of those areas and contribute an hour a week to working in them, that would chip away at these backlogs a bit. I dunno. Three months seems like a long time, and I agree with Hammersoft above that we need look only to Commons to see what a dearth of admins can do. (Also, edit conflict: Ritchie333 said everything I wanted to more succinctly right above) Go Phightins ! 11:31, 13 May 2021 (UTC)

• Go Phightins! has their finger on a key issue. Bluntly put, nobody individually needs administrator tools; it is the community as a whole that needs people to have and use these tools. I can guarantee that I do very little admin work in the areas I said I would focus on in my RFA, and I suspect the same is true of many (if not most) administrators. As an individual editor, I don't need to be able to block vandals or delete edits or protect pages; however, the project needs people to do those things, and I and most other administrators have sufficient knowledge of the project and its policies in order to do them and support other editors in their continuing work. I'd like to see the end of questions that say "why do you need the tools?" because there is no good answer to it. Risker ( talk) 15:50, 13 May 2021 (UTC)

  True, in the sense that nobody needs to edit Wikipedia at all. But if you do, the work of content creation frequently demands the use of the toolkit. Hawkeye7 (discuss) 19:40, 13 May 2021 (UTC)

  I think that's an important point, albeit one that often when I approach content creators about running doesn't often seem to resonate ... it seems to me there is a perception that unless you have a desire to be a backlog buster in the conventional admin areas, either you shouldn't be an admin or you won't pass RFA. Sadly, the latter might be true, but I'm interested in ideas about overcoming the former. If someone has been around the block creating content for a few years and they have a clue and show generally good judgment, it seems to me, from the encyclopedia's standpoint, we should want that person to be an admin even if they seldom make use of the tools. Go Phightins ! 20:53, 13 May 2021 (UTC)

  Yeah, lately I've also come to dislike the "need for tools" argument because of how ill-defined the phrase is. Indeed, no one ever really needs the admin toolset in the sense that without it they will cease to be a productive editor. It seems like whenever people actually say that a candidate has a "need for the tools", what they really mean is that having the toolset would be helpful to the candidate given evidence of their participation in administrative areas of the project—on the other hand, there is an appeal to not having a "need for the tools" in one of the current RfAs that seems to disregard even that kind of evidence. With backlogs increasing daily at WP:AIV, WP:RFPP, and WP:UAA, it seems very much that we need more admins than the RfA candidates themselves "need" the tools. Mz7 ( talk) 19:12, 16 May 2021 (UTC)

  To me, the expression is shorthand to describe an editor who often performs tasks for which one or more steps requires administrative privileges, and so has to request assistance in order to continue. At some point, if the editor has demonstrated sufficiently trustworthiness, it becomes more effective for that editor to become an administrator. Yes, they can of course continue to work as always without additional privileges, and it's not a criticism of their efforts if they fail to be granted admin privileges. Though it's a good route to become an administrator, I agree it shouldn't be the only one. isaacl ( talk) 19:37, 16 May 2021 (UTC)

  Prospective candidates with a strong edit history, are trustworthy, civil, collegial and all the rest, but who expect to be light users of the tools, are put off because they can't be bothered trying to justify having a need for tools they might use only occasionally. Continuing to ignore things they don't have the tools to fix is preferable, albeit mildly frustrating. Nurg ( talk) 09:48, 17 May 2021 (UTC)

• Having administrative privileges is basically signing up to do chores for the site. Some people like doing some chores; others will do them out of a sense of obligation. But typically it's a challenge to get people to do chores, especially in a volunteer community. In the recent RfC on adminship term limits, I discussed the advantages of a fixed admin term from a recruitment perspective, similar to what you're suggesting. I think there may be some editors willing to sign up for chores for a limited time, knowing they're not making a long-term commitment, with a future option of continuing for longer if they are willing and the community doesn't raise any show-stopping concerns. isaacl ( talk) 16:25, 13 May 2021 (UTC)

  Non-admin chiming in here: The "need" question is one which resonates with me. Looking at my record from at least some angles, I think I'd be a reasonable admin candidate--editing since 2007, over 30k edits, clean block log and a mostly drama-free existence. After early intensive activity in vandal-fighting and new page patrol, I've settled into a fair bit of light-duty "admin-adjacent" work like monitoring WP:BADAFD, cleaning up malformed AfDs and completing nominations on behalf of IPs; occasionally peeking over at the user creation log looking for potentially problematic usernames; accidentally becoming somewhat of an authority on the habits of the Teenage Fairytale Dropouts vandal; and random other bits and bobs along the way. I keep up with discussions on the various admin and Arbcom boards, but I haven't gotten involved in contentious areas either there or in article space. I'm sure I'd get dinged by a lot of !voters for an admittedly very weak record in content creation, but I'd like to think I've shown that I could be trusted to not do anything evilnastybad with the tools if they were given to me. Do I need them? Eh, if I thought I did and could make an argument for it, I'd have submitted an RfA long ago. And I could happily continue doing what I'm doing now indefinitely. Would I use the tools if I had them? I'm sure I would--there's always a backlog somewhere and I seem to have a nose for finding areas where there is a need. Do I want to go through the ordeal of having my record picked apart at an RfA? Having seen what it's turned into lately, I'm extremely hesitant. My sister Valfontis is an admin who didn't attract significant opposition at her RfA, but that was a decade ago in a much different environment (and besides, she's way more content-focused than I ever will be). So I'm continuing to stay on the sidelines for now. -- Finngall ^talk 17:29, 13 May 2021 (UTC)

Mmm, I forgot to mention in my earlier comment that I have very little experience in content creation and I'm reasonably sure there are editors who would oppose my becoming an admin solely on that basis, my clean block log and almost 90K edits in 13 years here notwithstanding. There are probably also editors who would oppose me because I say on my user page that I might like to be an admin someday. Gasp! I believe I once heard that the best way to help an RfA pass is to minimize one's editing as much as possible in the period leading up to and including the RfA...don't give people ammunition, essentially...and that's sad and seems to encourage potential admins to game the system. We're all humans, and humans make mistakes and lose their tempers, and aberrations shouldn't be confused with regular patterns. Is there really any admin on this site who's never exhibited behavior beyond reproach? I have my doubts. Anyway, thanks for reminding me that my lack of content creation would also be considered by some to be a black mark against my ever being an admin. Cheers! DonIago ( talk) 19:25, 13 May 2021 (UTC)

• I feel that the "chores" approach rings true for me (as a non-admin). I don't spend much time on the site on a daily or weekly basis, and when I am, I do some "admin-adjacent" work at AfD and Pending Changes. If I ever get my activity count up to community expectations, I think I wouldn't have much difficulty with the RfA process so I could do more work (in areas I am comfortable with and interested in) on some of these administrative areas needing attention. -- Enos733 ( talk) 20:43, 13 May 2021 (UTC)

@ Doniago: I passed with significantly fewer edits, an "I want admin" box on my user page, a clean block log, little content creation, and a similar tenure. If I can do it, you can too. Izno ( talk) 18:09, 1 June 2021 (UTC)

@ Izno: If you see my earlier comment in this thread, at this point it may be more a matter of whether I want to be an admin, and whether editors want me to be an admin, than the concerns I noted in the comment you responded to, though I do appreciate your insights. I'd run if there was someone out there who felt I might be a greater asset to the project if I had the tools, but I'm not even sure I can make that argument for myself, and at that point it doesn't seem realistic to expect that someone else might make the argument for me. If I've piqued your interest though, I'd be happy to discuss further. Cheers. DonIago ( talk) 18:58, 1 June 2021 (UTC)

@ Izno I worry that given that this would be RfA2 for DonIago, that it wouldn't be as straightforward as in your case. But Id be happy to be proven wrong as DonIago is an editor I would support at RfA. Best, Barkeep49 ( talk) 21:08, 1 June 2021 (UTC)

Oh, I skimmed things above, so I was focused on my issue: the portrayal that someone with those specific qualities can't or won't pass. Izno ( talk) 21:10, 1 June 2021 (UTC)

My apologies, I didn't mean to suggest that those qualities would in and of themselves prevent me from passing; I was merely anticipating objections that would arise if I did run. I'd welcome claims that those concerns are no longer valid, as that would be a pleasant surprise. Also, thanks Barkeep. :) DonIago ( talk) 03:17, 2 June 2021 (UTC)

Do I want to go through the ordeal of having my record picked apart at an RfA? Having seen what it's turned into lately, I'm extremely hesitant. So, here's my thoughts as a newer-generation editor with a habit of reading archives. I've read every RfA to run in 2016 or later, every crat chat RfA, and a significant chunk of "others" (nowhere near as large a percentage, but at least, I hope, a reasonably representative sample). I really do not see the idea that goes around routinely that RfA in the past few years, specifically, is a uniquely awful experience. If anything, my impression is someone running in the past couple years with a nontrivial tenure who isn't a personality case and appears vaguely aware of mainspace/projectspace (delete as applicable) will probably pass. That's not to say "miscarriage of justice" post-2015 RfAs don't happen -- I could namedrop if I wanted -- but when I'm reading RfAs that are really disastrous, they're almost never the past couple years. There's also the unspoken fact that tenure and edit count requirements are immensely flexible; GoldenRing is the obvious fantastic outlier, but there are far more normal RfAs to sail through for people who don't fit certain memes (Hog Farm passed at 95% at one year's tenure with a single tenure-related oppose, so clearly the people insisting on two or more don't actually care). The RfAs I've seen with significant opposition based on tenure, for someone editing actively for a year or more, have really been opposition based on inexperience -- where a user who's been editing for a significant period nonetheless doesn't seem to act in ways concordant with that nominal expertise. There are RfAs from the past few years with utterly bizarre tenure standards in the oppose column, but they're for people who had other serious concerns raised throughout. Vaticidal prophet 01:01, 15 May 2021 (UTC)

It's all about power

The "chores" angle and the desire to serve are almost certainly the main motivation for most admins. But this has next to nothing to do with the precipitous drop in candidates we've suffered this last decade. Take a look at WereSpielChequers' RFA stats. In the 5 years from 2007 – 2012, runs per month fell from from 34 to just over 2. Granted, social historians argue that over the past 3 centuries, individualism has generally rose at the expense of communal values. But none would think that folk were 1500% more likely to volunteer for community chores back in 2007 compared with 5 years later.

The explanation is power, and the fear of same. For its first decade or so, all kinds of power struggles played out across the pages of Wikipedia. Rival gangs generated countless 500KB+ hyper passionate threads where they warred over Inclusisms v Deletionism, real world politics, even just different technical preferences concerning wiki markup. Being an admin enabled one to participate more effectively in these power struggles. Now they've largely subsided, adminship is less attractive than it used to be. The isn't a major factor in the candidate shortage, but its more of a driver than declining willingness to do chores!

Who would be a leader, in this wicked world? The answer depends on context & how you frame the question, yet some studies suggest less than 3 in 10 prefer a leadership role, which ties in with actual behaviour: for example work places which try to pay supervisors only very slightly more than entry level workers, can find almost no one wants to step up, despite the supervisory role offering more autonomy. This said, almost everyone has the capability to enjoy having and weilding power, even if few want to do so most of the time. Adminship confers power without commitment to being a leader. It also confers status, and a potent form of social affirmation. ( I hope no admin gets annoyed reading this – as before, the main motivation for being an admin is it allows one to contribute more effectively to certain tasks. But human nature being what it is, power hunger / status seeking is going a significant motivator for many would be candidates. This isn't even a bad thing. Moderate power hunger is a good quality. People wielding power with fairness and a little benevolence are widely appreciated. The world would be a much happier place if there was a dom for every unfulfilled sub.)

The key change in 2007 wasn't any big decline in the attractiveness of adminship, but rather the rise of a countervailing force – the fear of power. Wikipedia began to become a big deal, not just one of the internets most visited sites, but one that could hugely influence the media (Allbeit mostly only for a limited time on certain hot issues. Controlling an article for a trending topic could shape literally thousands of media reports.) Along with other factors like the Essjay controversy, and the intentional standards raising efforts of Ballonman, the community swiftly became more wary about who we'd promote.

Rising standards got out of control, hence for most potential candidates the possible benefits of adminship seem outweighed by the risk of a painful rejection at RfA. This is all just my opinion – so I 3rd WSGs suggestion for a survey. We could just advertise a survey monkey as a Watchlist type notification, or better a SecurePoll. The framing could be something like "Please indicate 1-3 changes that would make you more likely to apply for adminship:"

1. If RfA was easier to pass.
2. If the RfA process could be made less unpleasant.
3. If Arbcom did not hold admins to a higher behavioural standard than regular editors.
4. If I didn't fear being an admin would adversely effect my relationships with fellow editors.
5. If adminship was made more attractive.
6. Nothing – I prefer being a regular editor.

Once we have some data, there'd be concrete steps we may be able to get consensus for. FeydHuxtable ( talk) 22:39, 14 May 2021 (UTC)

• Comment I've changed the formatting of the list a bit, hopefully that's acceptable. — Ched ( talk) 04:57, 15 May 2021 (UTC)
• Now, for my view: 2 and 4 would be two things I would find that would give me pause. — Ched ( talk) 04:59, 15 May 2021 (UTC)
• I know blaming ArbCom for stuff is a fun hobby for a lot of people, but I'd point out that the committee holds admins to higher standards of behavior because that's what the community wants them to do so to change that you'd need a very broad consensus that admins are held to standards no higher than anyone else. Once you have that the committee would certainly follow that standard. Good luck with that. Beeblebrox ( talk) 19:22, 20 May 2021 (UTC)

  Agreed. Though if we did launch a survey, the Arb effect question should probably should be an item as a it's been suggested above as a possible reason by two veteran and well respected editors. Even in the unexpected event we had a significant % of respondents ticking the box, I see no chance we'd ask the Arbs to stop holding admins to higher standards. But such a result could be flagged to the committee, together with a suggestion that for borderline cases, you consider erring more towards admonishments rather than desysops. FeydHuxtable ( talk) 20:30, 20 May 2021 (UTC)

Splitting admin position

I was an admin for several years, but I am now over 80 years old and my memory is going, so I resigned as an admin. I suggest that we split the admin position into 2 positions, something like "level A admins" and "level B admins" or "probationary admins" and "full admins". The names can be sorted out latter. "level A admins" would have limited tools and be easier to obtain, while "level B admins" would have the full tools. Details can be worked out later. This would perhaps encourage users to give it a go with limited tools. We clearly have to do something or we will not have any active admins. -- Bduke ( talk) 06:08, 15 May 2021 (UTC)

Hmm, I'm pretty skeptical about the "level A admins" idea primarily because the admin toolset isn't that large to begin with. Ultimately, the admin toolset can be boiled down to three "core" tools: delete, block, and protect. Everything else that comes with the admin toolset are sort of nice to have, but are not essential to the role and in many cases are given to non-admins on request, e.g. rollback and autopatrolled. In general, I think it would be a mistake to separate the three core admin tools, e.g. by giving a user protect without block or block without the ability to view deleted revisions, and I've written a broad summary of why at WP:COREADMIN (basically, they're interconnected and often need to be considered simultaneously). Mz7 ( talk) 06:27, 15 May 2021 (UTC)

Delete/protect/block should not be split, as having the choice between them is necessary for proper response to certain issues). If we want two levels, something like the ability to block extended confirmed editors could probably be reserved to the higher level admins. But is that really worth the effort? — Kusma ( t· c) 06:46, 15 May 2021 (UTC)

I might agree about block/protect. but delete? block and protect address behavioural concerns. Delete concerns content. A specific type of admin could function fine with block/protect and not have the tools to delete. it's a nice convenience for those who address behaviour-related concerns but aren't needed. "There's always another admin". So in those uncommon cases where the behavioural issues require a delete, a blocking style admin could always reach out to one with thhe ability to delete. No different than when a template editor or rollbacker needs to reach out to someone with different permissions/tools. So if the thought is to split the admin into two "sub-types", splitting by behavioural and by content could be a decent way to do it. - jc37 07:20, 15 May 2021 (UTC)

Block should come with unblock. Unblock without viewdeleted is nonsense. — Kusma ( t· c) 07:38, 15 May 2021 (UTC)

"Delete" and "View deleted" are separate user-rights. - jc37 07:41, 15 May 2021 (UTC)

And nobody should have "delete" without "viewdeleted". — Kusma ( t· c) 07:47, 15 May 2021 (UTC)

This needn't be an either/or situation. If it's determined that both the content admin and the behavioural admin "need" viewdeleted. then that's just a matter of adding it to each tool package. - jc37 07:58, 15 May 2021 (UTC)

( edit conflict) I think we tend to underweigh how serious it is to block new users and overweigh how serious it is to (temporarily) block established ones. In particular, a lot of people mistake content disputes with new users for vandalism. I'd probably want more vetting for a block-only-newbies position than for adminship, for the same reason admin candidates who are weak on content are scrutinized so heavily for temperament. On the object matter: I categorically oppose this sort of unbundling, because all it serves to do is widen, not narrow, the admin/non-admin gap. Probationary admins will be categorized as non-admins just as thoroughly as page movers, template editors, etc are, while the already-kind-of-ridiculous "need for the tools" opposition will be allowed to fester as the number of tools shrinks. Vaticidal prophet 07:22, 15 May 2021 (UTC)

@ Bduke, I just want to say thank you. Thank you for your service, and thank you for recognizing that maybe you weren't as effective as you once were and stepping away. Not everyone would do that. A ton of people who don't even edit any more have clung to their adminship hats. You are to be commended. —valereee ( talk) 19:57, 16 May 2021 (UTC)

Maybe you are both right, but you do not address the decrease in applications for admin. The lower level might just attract more applications. -- Bduke ( talk) 07:02, 15 May 2021 (UTC)

I've tried this approach several times with varying specifics. Maybe it's time to try again. I've been thinking about it lately and who knows, maybe WP:CCC and the community may support such an idea. - jc37 07:06, 15 May 2021 (UTC)

The lower levels attract lots of applications, but they don't become admins later. See rollbackers and page movers and template editors. — Kusma ( t· c) 07:11, 15 May 2021 (UTC)

Rather than two level of Admins by dividing the toolset, how about giving a choice of temporary and permanent Adminship? Those seeking permanent Adminship will have to go through the regular RFA. Those seeking temporary Adminship, say for 6 months, can go through a less strict RFtA. The toolset for both would be same, only term limit would be different. Something like this can encourage more people to try out. ಮಲ್ನಾಡಾಚ್ ಕೊಂಕ್ಣೊ ( talk) 07:32, 15 May 2021 (UTC)

The ability to "view deleted" would prevent that. My understanding is that, per WMF legal, granting a user group that has "view deleted", needs to go through the same process as RfA. - jc37 07:35, 15 May 2021 (UTC)

At least a "RfA-like" process in which the community decides to trust this user sufficiently to give them access to these rights. I don't think Legal requires RfA as it is, just some process that requires community participation. After all, they are fine with other wikis using basically straight up voting without much discussion (e.g. de-wiki). So theoretically a "RfA light" could work, we would just have to run the specifics by Legal.

Another idea would be to grant temporary adminship to users who gain less percentage of !votes than the threshold but still a majority. This would allow them to try and convince the community with their actions which could then decide whether to make the adminship permanent. Regards So Why 08:06, 15 May 2021 (UTC)

I basically like the idea of a more flexible threshold, but the thing is, it applies to a tiny minority of RfAs. Most successful RfAs are waved through with trivial opposition. Crat chats are about once a year. We expend a lot of energy at the borderlands of RfA, but it changes basically nothing because almost none are in those borderlands, while unbundling proposals of this type can only possibly make the situation worse and not better by continuing to make adminship a further and less reachable goal. What we need to do is make it clear it isn't, and that someone who's not a jerk, has a clue, and has a record has a much better shot than the reputation that precedes it. Vaticidal prophet 08:17, 15 May 2021 (UTC)

Agree with you on unbundling. On the flexible threshold point, that's a logical view. But for me our lack of borderline candidates results from the fact that anyone who's been passionate enough to get into occasional drama knows there's a significant chance it could tank their RfA, so they stay away. If we significantly reduced the threshold, my view is we'd get a flood of good candidates. Isaacl's temporary term idea might be the best way to achieve that. As SoWhy says, it's unlikely WMF would object. Long have they viewed the obstructed editor > admin pathway as a problem. I recall a 2010 London pub wiki meet when Sue Gardner told the Colonel & I that she feels it hurts community health - kind of like when long term reporters at her media orgs got stuck at street level roles, making them Grumpy. Here's explicit confirmation that WMF could support a much easier RfA process.

Rather than trying to agree on a specific mechanism for the easier temp promotion first, I'd suggest first having an RfC to get consensus for the idea in general, maybe with some minor conditions like the temp process could only be used at most 3 times by the same individual, before they'd need to pass a full RfA . Then folk could workshop various mechanisms, get the specifics approved by WMF legal, and then select the most popular option with something like Ranked voting. FeydHuxtable ( talk) 08:46, 15 May 2021 (UTC)

I think there is less number of borderline RFAs because currently people don't run unless they are fully confident they will pass. Failure at RFA would feel like their contributions have been judged and found to be not wanting. If there is an intermediate level of temporary Adminship, more people may come forward who are holding back in the current system.

To editor Jc37: WMF actually says nothing about going through RFA like process to become an Admin. From what I have seen in smaller wikis, they just require an on site record where the user has agreed to run for Adminship and there is community consensus for it. They treat it as the user accepting responsibility for their use of Admin tools, including viewing deleted edits. I am an Admin in Konkani language Wikipedia, which has very few active users so only temporary Adminship is granted. What I had to do was post in the local village pump that I am want to become an Admin and submit a steward request a week later. If there is no reasonable opposition, the Stewards will grant it. This is how Adminship is granted in most of the smaller wikis. Stewards will grant Adminship even if no one has !voted. ಮಲ್ನಾಡಾಚ್ ಕೊಂಕ್ಣೊ ( talk) 09:17, 15 May 2021 (UTC)

I tend to concur with Malnadach just above me on why we get few borderline RfAs -- more specifically, that people wait way, way longer than they need to to pass RfA. The corollary of this is that it fucks up the maintenance of one of the biggest things man has ever done (because that's what we're doing, here -- building a repository of human knowledge unimaginable but twenty-five years ago), because someone sits on their hands going "oh, I couldn't pass RfA" instead of becoming an admin and handling the administrative challenges that sum of human knowledge faces. In turn, of course, the big obstacle for things like what you (Feyd) link to make RfA less scary by turning it into another trip to the haberdashery fails on its face because it removes checks and balances; the community's varied standards for tenure, edit count, content creation, project maintenance, civility, etc manage to swirl into something most of it doesn't abjectly hate, whereas a PERM-based individual admin discretion situation turns into what that individual admin thinks, and boy do I expect that to be a matter of significant variance. At that point you end up with admins checking-and-balancing each other, and you just get basically a weird version of crat chats. It would only be able to isolate non-admins from such discussions further, not include them. Vaticidal prophet 10:35, 15 May 2021 (UTC)

Looks like we've mostly converged our thinking on why there are so few borderline candidates. I'd add when we are able to pass them, IMO borderline candidates often (not always) turn out to be more productive admins than some of the "shoe ins". I also agree about WP:Perm - that wasnt my idea, it was simply the least stringent of a number of proposals made at a village pump discussion on alternatives to RfA. My purpose in linking to Phillipe (WMF) saying he'd support even the WP:Perm method, is that it suggests there should be no WMF issue with a middling proposal between WP:Perm and what we have now. FeydHuxtable ( talk) 11:01, 15 May 2021 (UTC)

I think we agree on the facts and disagree on the methods, as it were. I don't think technical changes would be particularly useful for the most part. (I do have a sense the discretionary range is too narrow, but that's in both directions.) The problem I see is RfA's reputation precedes it. I'd also quibble with some of the standards, but those are nitpicking people's passing percentages; from my birds-eye view Wikipedia:Requests for adminship/Ergo Sum should've been 250/0/0, but that's just a matter of how poorly we prioritize writing an encyclopedia. As I noted above, I think most perceptions of RfA's standards are mirages, and ideas like "you need two years tenure to pass" are trying to backfill a combination of RfAs where people who were around for years but still bereft of clue were told they needed to wait a ludicrously long time and RfAs where people sailed through after a year. Vaticidal prophet 11:42, 15 May 2021 (UTC)

Yup, the ideal solution might be to change RfA's reputation &/or get voters to be more tolerant of folk who don't tick all their boxes. And you're right - we disagree about methods. I don't see any feasible method to effect said ideal solution. Folk having been saying voters have become too picky for over 10 years, several have claimed there is concensus for that view, but it changes nothing. It would make little difference if even the top 10 most respected editors wrote an essay attempting to reset RfA's reputation or voters attitudes. Much of the community would be like "Bah - cabal!" I guess someone could change the 2007 RfA add , which was written at a time when we arguably had too much distracting drama from 5-15 concurrent RfAs at once, and hence seems to try to make RfA seem as unpleasant as possible. I was thinking of suggesting this depending on the results of the survey. But it would likely have little impact.

Consider the work of Balloonman. In 2007 he decided that RfA standards were too low. He thought it ridiculous that many RfAs passed with just a few dozen ultra casual comments, & no sign of serious scrutiny. An individual of immense tactical ability, Balloonman knew it might be fruitless to openly argue for higher standards. Instead he used Gandi’s "be the change you wish to see" tactic. He started making increasingly serious & analytic votes in both support & oppose. He complimented others doing the same. It took years, but ineroxably this behaviour spread by social contagion. Balloonman didn't admit this had been his plan all along until it was too late. As mentioned above, by 2012 it was clear rising standards were out of control. Balloonman himself began to despair at the changes he wrought, realising that even he was powerless to reverse them. He quit later that year. Others tried to reverse his work using the same tactics. But some methods can only work in one direction. As Einstein said, any fool can make something more complex, it takes a Genius to simplify.

What we do have the power to do is make technical changes - as demonstrated by Admin Biblioworm's 2015 reform. Admittedly, the stats show it made little different to RFA numbers. IMO this is as the 10% discretionary range reduction was too modest. I've always said a reducing the lower bound to 50% would be more sensible. Coupled with voters being assured resulting promotions would only be temporary unless the candidate passed the full RfA process, this change should allow us to fully replenish the admin ranks. FeydHuxtable ( talk) 12:45, 15 May 2021 (UTC)

Note the proposal I outlined for fixed terms did not propose granting a limited set of privileges initially with a more complete set being granted after a more rigourous evaluation was made. The proposal was more along the lines of the fixed terms proposal on which I was commenting: the candidate would be fully vetted by the community to receive all administrative privileges for a fixed period, after which the new admin would decide if they wanted to keep serving (and for how long). This checkpoint would prompt the community to provide feedback. isaacl ( talk) 15:43, 15 May 2021 (UTC)

Well, then maybe they have changed their position since this. - jc37 09:22, 15 May 2021 (UTC)

I know it was 10 years ago, but I am pretty surprised everyone just accepts that requirement at face value without even so much as an explanation. Levivich  ^harass/ _hound 14:55, 15 May 2021 (UTC)

With a bit of digging I found this, which seems to be the explanation. Basically, apparently, a lot of complaints would happen and might result in corrective action by Congress. ProcrastinatingReader ( talk) 15:23, 15 May 2021 (UTC)

Also this, I guess. ProcrastinatingReader ( talk) 15:25, 15 May 2021 (UTC)

Yup, those are the explanations that I'm surprised the community accepted for so long. In a nutshell, the reason we have RFA (as opposed to some other method for selecting admins) is to reduce the chances the WMF gets sued. Volunteer my time to help reduce the WMF's liability? Hell no, they receive over $100 million a year, they can hire more lawyers if they need to (and not rely on volunteers to protect them from liability). Levivich  ^harass/ _hound 15:37, 15 May 2021 (UTC)

The process does not have to be RFA asit exists now, or as it did in 2008. There just has to be a process that the community accepts as valid. For example, were a non-admin to be somehow elected to ArbCom, the foundation has made it clear they would accept that as sufficient community vetting to grant all the user rights that come with committee membership, admin, checkuser, and oversight, (although those last two are contingent on them signing the Confidentiality agreement for nonpublic information). I have yet to see a design for a process that I am sure would work better than RFA but it isn't impossible to fundamentally change the process. Beeblebrox ( talk) 19:55, 20 May 2021 (UTC)

A fairly popular view is that RfA is a good stress test for real adminship, and that RfA stress helps candidates grow as Wikipedians, something that often is true in an admin's own wikibiography. I would like to challenge this point of view. Instead of attacking anyone's views in particular, I am going to distill them into some strawmen and attack those. Here are some thoughts.

1. The amount of harassment and being yelled at that admins get is not actually significantly different from what other Wikipedians have to endure. There is a huge amount of peacuful admin work to be done that does not turn you into a troll magnet.
2. For some people (non-admin closers of contentious RMs and RfCs), adminship might even reduce the amount of flames they get.
3. Admins also need to deal with the extra respect that comes with being an admin (I get far more unwarranted respect than unwarranted abuse these days). Maybe we should test in RfA whether they can deal with that?
4. Adminship isn't only for Real Men with Balls of Steel and a fireproof suit. Diversity and neurodiversity are good.
5. Many current admins have passed RfAs that were not trials of fire and nevertheless have boldly done the Right Thing in the face of criticism.
6. That current admins have overcome difficult RfAs does not mean that others need to endure the same (or worse). There are many ways to learn on here and many ways to grow. Professionals or subject experts with long distinguished careers may not appreciate any attempts to build their character.
7. Wikipedia isn't a fraternity where pledges need to pass the hazing ritual. Excluding people who don't enjoy being hazed seriously limits our candidate pool.
8. Given our low promotion rates, we can't actually afford to limit our candidate pool only to people who should have been promoted 5 years ago. We need to promote enough admins.

OK, I'll stop here. Please submit ideas how to fix things to Tony's thread above. Just don't pretend everything is fine. It isn't. — Kusma ( talk) 09:00, 4 June 2021 (UTC)

It really is not all that bad, it is about as bad as your average office job's performance review. A few tough questions and some silly oppositions. I would not describe it as a trial by fire. The actual job is far more difficult than the interview. HighInBC ^{Need help? Just ask.} 09:12, 4 June 2021 (UTC)

I like the"performance review" metaphor. My own RfA was like that, but that was 15 years ago. As evidenced on this very page, recent candidate experience is very different. — Kusma ( talk) 09:29, 4 June 2021 (UTC)

One small observation: I've never been to a 'Performance Review' which lasted a week and where the entire workforce were invited into the car park and could say whatever they liked about me, and I was discouraged from responding back to them. But I enjoyed my RfA - far more than any performance review I've ever had. Nick Moyes ( talk) 10:48, 4 June 2021 (UTC)

You have clearly never worked for Amazon as a software developer then. HighInBC ^{Need help? Just ask.} 11:00, 4 June 2021 (UTC)

And Amazon is famously bad at doing performance reviews, so I don't think we want to emulate them. —valereee ( talk) 13:18, 4 June 2021 (UTC)

HighInBC, In the case of Less Unless, I'd like some evidence that the average office job performance review includes speculative accusations of transphobia. Ritchie333 ^{(talk) (cont)} 10:57, 4 June 2021 (UTC)

I am not familiar with the details of that particular RfA, however if someone is engaging in personal attacks or otherwise acting in a manner inappropriate for the project then that should be dealt with by an uninvolved editor or admin. The same issues exist in every single part of Wikipedia. At no point did I suggest it should be a free for all. HighInBC ^{Need help? Just ask.} 11:48, 4 June 2021 (UTC)

Tell you what, if someone makes a proposal that we more strictly enforce our existing AGF and civility policies at RfA then I will support it with enthusiasm. I believe the main failure is that we have a culture of not challenging each other at RfA. The term "badgering" is tossed about to quash any suggestion that someone's point of view might not be entirely valid. HighInBC ^{Need help? Just ask.} 12:11, 4 June 2021 (UTC)

I don't think that's the problem. Nobody was incivil or assumed bad faith at Less Unless' RfA. In fact, in my experience the overall level of discourse at RfA is pretty good these days, and descending into name-calling or petty insults is pretty unusual. The worst it got in this RfA were two opposes right at the end about being available for recall which I considered to be disrupting Wikipedia to illustrate a point. The concerns raised by Steve Smith and Ineffablebookkeeper were perfectly valid opinions, expressed in a civil and collaborative manner, and in no way merit any sanctions whatsoever. Indeed, part of the problem is that candidates find RfA stressful despite everybody working in good faith. It's more the volume and intensity of the combined views and comments that is the issue. Ritchie333 ^{(talk) (cont)} 12:20, 4 June 2021 (UTC)

That's one of the reasons why in my RfA proposals I've suggested consolidating discussion of the candidate's characteristics in one place, rather than having discussion threads branching out from any commenter's support/oppose/neutral statement. (See my 2015 proposal and my proposed variant of a two-phase RfA.) I think acrimony would be reduced when each characteristic has a single thread discussing it, rather than having many threads covering the same territory repeatedly. The current RfA format emphasizes individual convenience, by having commenters explain and defend their rationale in one location, versus the community benefits gained by reducing redundancy. isaacl ( talk) 13:31, 4 June 2021 (UTC)

I support this as well (and have supported it when I happened to see it suggested in the past). It might also help get rid of token-reasonings, would direct the focus away from who said something to what was said, and it might even be easier on the !voters: contributors whose reasoning gets challenged wouldn't need to feel singled out or badgered. Plus, it would allow for more in-depth argumentation without the fear of cluttering up the votes section or turning it into a dramatic dialogue, or whatever other scruples might stop people from presenting their arguments and concerns. (In other words, what Isaacl said). --- Sluzzelin talk 14:02, 4 June 2021 (UTC)

WTF? @ HighInBC, have you actually read any recent RfAs, or the beginning of these discussions? The Less Unless RfA, which is the most recent RfA, was literally what started this entire discussion. I can see you haven't edited much in the past five years, maybe you simply aren't very familiar with Wikipedia space any more. —valereee ( talk) 13:11, 4 June 2021 (UTC)

I have read the RfA, I have not done a deep dive into the underlying issues and extensive discussion referenced by the RfA. HighInBC ^{Need help? Just ask.} 22:02, 4 June 2021 (UTC)

@ HighInBC, a performance review conducted 24/7 for a full week, in public, and your worst work enemy and the intern who started yesterday can both ask up to two questions. Anyone in the world who knows you IRL and can read English can watch as your least mistake from five years ago is dragged into the conversation. Yeah, that sounds like a great way to conduct a performance review or interview -- which btw is for a volunteer position. Let me know when you're ready to suggest that at your workplace. —valereee ( talk) 13:06, 4 June 2021 (UTC)

The performance review analogy doesn't really make much sense at all, not least because performance reviews to be all that useful have to be regular and in this case the only "performance review" you get is when you're asking for additional responsibilities, not before, and you will never get quizzed again unless you repeatedly and royally screw up. Der Wohltemperierte Fuchs ^talk 16:27, 4 June 2021 (UTC)

I agree it doesn't make sense. I was responding to a post saying it wasn't worse than the average performance review, pointing out that it's much worse on pretty much every aspect. No objection to thorough vetting of candidates. —valereee ( talk) 00:58, 5 June 2021 (UTC)

• I agree with all of those points, Kusma. Seeing the effects of #3 right here lol. —valereee ( talk) 13:30, 4 June 2021 (UTC)

My research isn't in-depth enough pre-mid-2010s to be confident on this, but I'm unconvinced from what I have seen that RfA of the last few years is best described as 'trial by fire'. There's a very obvious cooldown in the past few years; even the more recent horror-clusterfuck RfAs I've studied are clustered in the first couple years of the modern era. The admins who needed to pass a trial by fire are roughly the 2012-2016 cohort. (Been picking up some numbers to do some analysis; I'm planning to make a thread here when I've finished some of my comparisons, but it's a massive amount of data to go through and a lot of it (e.g. content creation) needs to be assessed subjectively, so it won't be soon.) Vaticidal prophet 01:06, 5 June 2021 (UTC)

@ Vaticidalprophet, we have to deal with the perception of and reality of what the people who have gone through it have felt. From the point of view of someone who has both gone through it recently (2019) and supported candidates through it since then, I can tell you the process does not encourage people to run. Most of the "it's not that bad" arguments I'm seeing here are from people who have not gone through this in the past five years. I don't know how to be clearer than that for those who went through it 15 years ago and those who have never gone through it. —valereee ( talk) 01:16, 5 June 2021 (UTC)

Oh, absolutely, I'm very aware that there's a gap between number-crunching and people's internal experiences. I'm not sure where, exactly, this circle is squared, and I'm unconvinced of the solutions most often proposed (e.g. the "make RfA more PERM-like" above, which is qualified there as a hot take but is held unironically plenty often). (I am particularly unsympathetic to "make RfA more quantitative" because I think that inasmuch as RfA is broken, it's because people try to make it quantitative.) Vaticidal prophet 01:19, 5 June 2021 (UTC)

2. For some people (non-admin closers of contentious RMs and RfCs), adminship might even reduce the amount of flames they get. Where did you find this picture of Buidhe? – ♠Vami _IV†♠ 07:48, 5 June 2021 (UTC)

Yeah, I'd consider running if it were a NomCom that decided on creation of admins rather than the current trial-by-fire setup. I think I've pissed off enough people that it could get ugly. ( t · c) buidhe 08:06, 5 June 2021 (UTC)

Guilty as charged: I was thinking of her. But I expect it applies more generally. — Kusma ( talk) 08:42, 5 June 2021 (UTC)

Perhaps I should preface this by noting that I am, as my user name indicates to those familiar with certain languages, female; I was granted adminship on a NOBIGDEAL basis; and I was desysopped by ArbCom after reversing an ArbCom block. And that I'm going to tiptoe around the elephant in the room, a relatively recent RfA on which I have expressed a vehement opinion; that editor was ultimately lost to the project, and in any case I would like to avoid personalising this.

I came here from discussion that had spilled over into a current RfA, and I see some fairly nasty responses here to HighInBC's comments, some of which I consider valid. I am not convinced that we have such a terrible need of increasing the number of admins as to outweigh either the danger of promoting an abusive (or otherwise destructive) admin or the alienation of the community from reducing their input into the decision whether to promote someone. For this reason I vehemently disagree with the idea HIB puts forward above, of having admins themselves select new admins. There are enough fissures within the community already (one of them being the division caused by admins reflexively closing ranks); adding to these an emphasis on people's wiki-generations or admin generations is counter-productive divisiveness. If we need more admins (and I personally think the need is not so great as many vocal editors think, and could be reduced by dusting off WP:NOTBURO and revisiting the need for the last few years' worth of systematic restrictions on editing and template encrustation, but I am not the Dictator of Wikipedia), we don't particularly need admins who are relatively youthful or relatively new to Wikipedia—this is not Admin Bingo—we just need more admins, without bias or impediment, and that way some of those we get will be relatively youthful or relatively new to Wikipedia; and it's counter-productive, to say the least, to be hostile to admins who became admins a while ago simply for that reason.

I don't think comparisons to workplace reviews take us very far, for the same reason I don't think appealing to one's personal experience of norms of workplace discourse is useful in establishing civility boundaries. We—and potential admin candidates—live in a multitude of different countries and have a huge variety of work experiences (or no work experience); further, extreme examples can always be adduced, such as the one above concerning software development at a company that is often in the news for its treatment of workers, or for that matter boot camp or a Klingon honour gauntlet, and how useful is that, really? or how persuasive to those considering adminship? Like it or not, this is an on-line text-based community. RfA participants are drawn from among the most experienced and the best on-line searchers active at that time, with access to an intentionally almost complete record of the candidate's on-site activity. And they're choosing someone for a position that is not just button mashing but in large part decision-making about whether or not to mash a button, and which one; how to avoid button-mashing being necessary (many of the most important actions of the best admins are not logged; they are de-escalations, advice, and explanations) or follow-up on button-mashing by themselves or others. I concur with HIB that RfA should be at least in part a discussion, and that the candidate's comportment is an important part of evaluating whether to entrust them with the tools. And I've probably said enough so I'll stop there. Yngvadottir ( talk) 02:47, 6 June 2021 (UTC)

I did not suggest that admins themselves select new admins. Perhaps I am misunderstanding, or being misunderstood? Of course it is the choice of the community at large. HighInBC ^{Need help? Just ask.} 02:50, 6 June 2021 (UTC)

Oh I am sorry HighInBC; I finally tracked it down and that was a suggestion by Bduke, not you. Where we appear to disagree most is over "bludgeoning"; while this discussion is about not having RfA be unnecessarily daunting for the candidate, it takes courage to raise an issue about a candidate in front of their powerful friends and in the face of the argument that the need for more admins outweighs any concerns. But adminship shouldn't be based on the clout of the candidate's friends. Especially since the project is now so big, and the types of work we choose to do as well as the types of articles we work on can be so disparate, that how many "power users" know the candidate doesn't mean much, and I've been happy to see a trickle of unusual RfAs recently in terms of areas the candidate works in. I know that leading up to many RfAs today, a lot of more-or-less behind-the-scenes consultation goes on to find candidates, advise them on the process, and prepare nominations. I offer two small suggestions. One is already advised, but based on some RfAs I've observed, not followed enough for the candidate's sake: nominators should work really hard, both with the candidate and on their own, to turn up any skeletons in the candidate's closet or areas of disagreement between editors that the candidate has taken a strong position on, and help the candidate prepare to deal with those issues. That is far less stressful for the candidate than being suddenly confronted with a question or opposes based on something that they may not even have thought of as problematic, and the RfA will be more like a mutually respectful discussion if the candidate responds effectively to the issue (including saying they disagree with the criticism) than if a bunch of the candidate's friends jump all over the critic. Secondly, let's encourage non-admins to propose nominees. We have scads of excellent editors who many of us consider would make good admins, but who don't want to run. Fine, I understand. But could some of you look around and suggest someone else? Perhaps in association with some of the admins who are experienced nominators, like Ritchie333? Such collaborations might be useful in both finding candidates and helping them prepare. Yngvadottir ( talk) 06:28, 6 June 2021 (UTC)

I have a couple of editors who make suggestions to me. In fact Trialpears first came to my attention via a recommendation of a non admin editor. Best, Barkeep49 ( talk) 08:07, 6 June 2021 (UTC)

If we look at the two RfAs running currently, several editors supported with a comment along the lines of not knowing the candidate but trusting the nominators. Ultimately, most users who have widespread community trust in their judgement are administrators. So it makes sense why, proportionately speaking, most successful RfAs have admin nominators. ProcrastinatingReader ( talk) 16:48, 6 June 2021 (UTC)

Having seen a few RfAs from a distance quickly cured me of any notion of becoming an Admin. There is much to User:Kusma's statement. Have WP considered creating a lower level admin (Assistant Admin or such) with limited powers? 12 years here have convinced me that operating with a minimum of power and responsibility is the way to go, but there are probably others who would do well with some additional authority. It could be a path towards full adminship as well. Again, apologies if this has already been discussed to death several times.  Mr.choppers |  ✎  04:04, 6 June 2021 (UTC)

• RfA is an utterly horrible way to choose our decision-makers, and the actual problem is that adminship itself is stupid. The fact that someone needs and merits one advanced permission does not mean that they merit all others, and the honest truth is that we do have admins whose judgment I think is defective, and we don't have any effective way to limit them unless (a) they screw up badly and (b) they refuse to utter the mea culpa of automatic sysop exoneration. Our system is broken and unfixable. Only the fact that most admins are basically decent human beings who're trying their best makes it workable at all.— S Marshall  T/ C 16:26, 7 June 2021 (UTC)

The last sentence works just as well if you replace "admins" with "editors". I'm not suggesting what you're saying is not valid, but rather that Wikipedia as a whole is a broken and unfixable system...that somehow manages to work (mostly) anyway. -- Hammersoft ( talk) 20:31, 7 June 2021 (UTC)

Amen + ; we muddle along. Alanscottwalker ( talk) 22:13, 7 June 2021 (UTC)

I sometimes feel constrained about commenting in these discussions since my RfA went smoothly with 316 supports and the only two editors who opposed have since been indefinitely blocked (though not for opposing me). So, I realize that I am an outlier. But when I describe it as "smooth", I do not mean stress free. I was acutely aware of the ongoing process and paid very close attention to every question and comment. I do not really accept the performance review analogy, because RfA is not about continuing your current job with some feedback and hopefully a raise, but more like an application for a new job or asking for a major promotion. Actually, I think that it is more analogous to running for a local political office like city council. RfA lasts for one week but those kinds of campaigns last for months, and successful candidates need to deal with kooks, cranks and intelligent local citizens with serious concerns about policy issues. Being a Wikipedia administrator is somewhere between "no big deal" and becoming a neurosurgeon. It is not rocket science and it is not burger flipping. I have been either supporting mostly or opposing sometimes at most RfAs for quite a few years, and I think that the end result of each RfA is correct a large majority of the time. Of course, there are some candidates who pass who are later shown to be unfit but there is no selection process devised by humans that is 100% reliable. I very rarely see a case when an obviously qualified candidates is turned down. Just look at the Catholic priesthood with much more rigorous vetting that in the past regularly promoted pedophiles, and I say that as a Jew who went to a Catholic high school and graduated from a Catholic university. I've heard the "RfA is broken" rhetoric for well over a decade and do not accept that assessment. Stressful? Yes. Would we want every RfA to be a cakewalk? I don't think so. "Broken"? That's kind of a trendy word that I associate with New Age jargon, and I do not think that it really applies here. I make my comment in the context of a recent RfA that was withdrawn by the candidate because very significant issues were raised though not always artfully. Tough to be suave about fascist self-declarations, even if four years old. On the other hand, we have another RfA in progress, and so far, no closet skeletons have been uncovered. So it goes. Cullen³²⁸ Let's discuss it 02:33, 9 June 2021 (UTC)

Would metawiki's limited adminship work here? The requests look like this: [1] [2]. They're limited for a duration, and the tools can only be used for the purposes specified in the request. A possible advantage is that, since the period of limited adminship is limited to the scope specified in the request, there's no need to speculate how they'd handle other tasks. Many then go on to request full adminship, a renewal, or if the task is complete then don't renew. It also serves as an alternative to further debundling. ProcrastinatingReader ( talk) 10:01, 5 June 2021 (UTC)

If I trust someone enough to give them the delete button I trust them enough to block and protect. I'm not sure what benefit this would have. Anarchyte ( talk) 10:05, 5 June 2021 (UTC)

I suppose not everyone thinks like you. If page mover didn't exist as a technical role for example, a greater technical access with a social prohibition on using tools beyond the scope of that access would be another way of allowing a person to do the same tasks, and there are plenty of page movers who wouldn't pass or be suitable for full adminship. That role could be unbundled but I suppose there are others that can't. Evaluating competence at doing a specific task is substantially more narrow and focused than evaluating a person. ProcrastinatingReader ( talk) 10:10, 5 June 2021 (UTC)

@ Anarchyte: (see also next below on if there is actually a use case that needs this) - in general limited admins on meta have no "behavioral" administrator responsibilities, as such they don't run in to areas that are likely to have drama. It isn't used that often unless there is a specific need that doesn't have a better solution. One thing that could go along with something like that is a much easier removal criteria (for example allow any 'crat to remove access if what they deem to be misuse occurs - as well as by any showing of consensus for removal at a venue such as AN). — xaosflux ^Talk 10:59, 5 June 2021 (UTC)

@ ProcrastinatingReader: If there are actually enough editors that would make use of something like that I'd probably be in support of it, or something like the engineer or eliminator permissions found on some other projects. If there isn't going to be a critical mass of actual users though the overhead isn't really worth it. Where is the technical type backlog that has willing editors that are only held up by lack of sysop access today? — xaosflux ^Talk 10:54, 5 June 2021 (UTC)

About use cases: Backlog due to lack of necessary permissions is probably difficult to measure, especially if an area has historically had resource deficits. The admin toolset can briefly be summarised, in technical privileges, as: block, protect, page deletion, revision deletion, viewing deleted revisions, editing the interface, editing/moving sysop-protected pages, and hist merging. Translating these technical tools to social tasks we get a long list, to name a few: closing AfDs, deleting copyvios, processing page protections, blocking vandals, creating/editing gadgets.

These areas generally have non-admins working in them to the extent that they can. In some cases admins are rubber stamping non-admin actions (eg RfPP/AIV requests (see User:Enterprisey/AIV analysis) and {{ cv-revdel}}s (also see [3])), and in others active support by someone with admin access can be necessary (eg viewing deleted revisions to process some SPI reports, and creating edit filters that target deleted revs). At AIV, Enterprisey's analysis found several non-admin users with 95-100% reporting accuracy. ProcrastinatingReader ( talk) 12:27, 5 June 2021 (UTC)

@ ProcrastinatingReader: as a large amount of the things you listed above deal directly with "people" as opposed to just "things" I'd expect this to be a non-starter. I don't think that most of the hurdle in our RfA today is about "will nominee make bad technical choices" but more about if they will be problematic with content or editors. The meta-wiki LA group primarily only deals with completely uncontroversial actions. — xaosflux ^Talk 12:51, 5 June 2021 (UTC)

I always support such proposals, as I believe further unbundling would help by giving more tools to trusted users without requiring those users to put themselves through RfA, but that argument never seems to convince people. They just can't come down off of "they should just run RfA". They can't seem to wrap their heads around the idea some people don't want to run RfA/don't want to be admins. My feeling is that what Anarchyte is saying is what the majority feels, and that further unbundling is a non-starter right now. —valereee ( talk) 11:14, 5 June 2021 (UTC)

While I concede it's certainly possible for there to be some in the category Valereee notes, I also would say that I feel that it would risk becoming expected for anyone to do this before their full RfA, and quite possibly with as much of a abrasive process. After all, if someone was getting the delete right I'd have to do about as much investigation as I do for the full toolkit Nosebagbear ( talk) 11:17, 5 June 2021 (UTC)

Nosebagbear, but permissions are easy to remove. Something like this would just be another permission. The whole reason sysop gets people digging so deeply is that sysop is hard to remove. —valereee ( talk) 12:05, 5 June 2021 (UTC)

• While I almost always oppose unbundling proposals, this isn't really an unbundling proposal. As I understand it (correct me if I am wrong), limited adminship isn't limited technically but by the expectation of the area where tools should be applied. We could say, you can do speedy patrol for a month, and you're supposed to use viewdeleted only for G4 speedies, and only block creators of nasty attack pages, and only use the protection button for salting. Or we could say "you are only allowed to do copyvio revdels".
  Something like this would make sense for people who are trusted enough not to publicise deleted revisions that shouldn't see the light of day ever again but lack the AfD (or whatever) experience that they might need to pass RfA. I could imagine something like this (for some type of focus) working, but I am not convinced it is worth it. — Kusma ( talk) 20:14, 5 June 2021 (UTC)
• We should be able to select admins that we can trust enough to limit themsleves to areas where they are technically competent or emotionally suitable. We should expect that an admin candidate has enough knowledge to keep away from areas that they can't handle. For example I keep away from SPI as it is too bureaucratic, with precise templates having to be used. But what problems are we trying to avoid? One issue is admins being rude to others. In fact very minor offensive behaviour seems to be enough to get desysopped. And I suspect there are quite a few of us that have that potential, it is just that no one reported to arbcom yet. Sure there a few mistakes by admins, but gross incompetence does not seem to be a problem. Graeme Bartlett ( talk) 23:47, 6 June 2021 (UTC)
• I agree with Kusma above. Although, there is one more possible problem. We already say RfA is broken, harsh process and whatnot. If we implement this limited/partial adminship, then RfA for for a full adminship might become even harsher than it already is. Other thing is, most of the admin do utilise "limited adminship" except for time-period. Like Graeme Bartlett said above, almost every admin has a speciality area, and an area where they barely work. Very few admins are all-rounders. Anthony works primarily in hist-merge, one ex-admin was speedy specialist. One admin was excessively active in SPIs while they were CU. Its on rare occaissions when Anthony would comment in ANI. You know what I am trying to say. I think limited adminship would not be a good idea for the long run. —usernamekiran (talk) 12:34, 7 June 2021 (UTC)
• There's really no point for this at enwiki. Meta doesn't deal with content, so the expectations of what administrators do on it are much more limited, therefore a "character" test with using the tools via an RfA equivalent isn't necessary like it is here; Mr. Bartlett hits the nail on the head with his comments. Any "semi-admin" usergroup, technical or de facto, would just be another layer of bureaucracy. As for unbundling, all of the tools that arguably could be unbundled from the mop already have been, and the two that really can't be given to non-admins are block and delete (indeed, the WMF has explicitly vetoed attempts to allow non-admins to even view deleted material, for good reason). In other words, if you want to delete stuff or view deleted stuff, run for RfA; if you can't or won't, deletion isn't for you, full stop end of.  –  John M Wolfson ( talk •  contribs) 22:34, 9 June 2021 (UTC)