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.

I realize that the community is obviously more cautious after what happened with Wikipedia:Requests for adminship/Eostrix, but is there any actual precedent for people openly speculating about someone being a sock at RfA? To me, this seems like it's kind of a personal attack – we're literally comparing a good-faith editor whose given us no reason to indicate otherwise in my opinion to a WMF-banned user that was ArbCom blocked. Openly speculating about someone being a sockpuppet (I'm under the understanding that actual WP:SPIs require more evidence) is basically saying "yeah, this person is totally a malicious actor out to dupe all of us by being an admin"! Am I the only one kind of in disbelief about all of this? Like it's not entirely unexpected, but it's unfortunate to see my suscipions confirmed. It's the first thing in awhile to make me seriously question the community, so I thought I'd try to open up an actual conversation here.

I looked up what the ArbCom noticeboard said about this at the time [1], it seems like there's a strong emphasis on the part private evidence played and how this could actually influence future RfAs and lead to a chilling effect. Clovermoss (talk) 08:18, 14 September 2022 (UTC)

No, you're not the only one. I've been irritated by those accusations as well, and have hinted as such. As for whether it can change the course of an RfA, that's a bit more of an interesting question as RfA is kind of a vote, regardless of what anyone says. If the total percentage of support is above 75%, or probably even above 70%, it's probably not an issue. It gets a bit more interesting if the support is below 65%, as at that point the 'crats have to decide whether or not to pass the RfA anyway, or conclude that the accusations of sock puppetry gained sufficient consensus among the community to not promote. Neither of those is going to be popular. Ritchie333 ^{(talk) (cont)} 10:58, 14 September 2022 (UTC)

I edited quite a bit in October 2021 but wasn't involved enough in the inner workings of Wikipedia as to know what went down in Eostrix case. Can someone please help me with a brief summary of Eostrix/Icewhiz? Thanks! — CX Zoom[he/him] ^{( let's talk • { C• X})} 11:52, 14 September 2022 (UTC)

My understanding is that Icewhiz is a globally WMF-banned user that caused harm to real people. Icewhiz is an LTA that also has a history of sockpuppetry (see Wikipedia:Sockpuppet investigations/Icewhiz/Archive). It seems like all of this happened after this 2019 ArbCom case. I've never been someone who was super into following all the stuff that goes on at ArbCom, so someone else could probably give you a much better summary. Clovermoss (talk) 12:01, 14 September 2022 (UTC)

No, you're not the only one. It's disappointing to say the least. Levivich ( talk) 13:06, 14 September 2022 (UTC)

You're far from the only one. Unfortunately I'm hardly in a place to scold people about it. Vanamonde ( Talk) 13:33, 14 September 2022 (UTC)

WP:NPA#WHATIS notes "Accusations about personal behavior that lack evidence. Serious accusations require serious evidence..." Making an accusation of sockpuppetry is a serious accusation. When I commented in the neutral section regarding the candidate, I was very careful to layout evidence and ask for it to be reconciled with their statements about no prior accounts. I think that's a legitimate question. I think it needed to be asked. I was careful not to say sockpuppet. I also placed my comment in the neutral section, because doing it in support or opposition would imply that I felt one way or the other that it was true or false. Accusing someone of being a sockpuppet does require evidence, and past disputes about this have noted that such accusations require evidence. Accusing someone of being a sockpuppet without evidence is a personal attack. While not making excuses for RfA it is very common for the gloves to come off, so to speak, during RfAs. More abstractly (and NOT referring to SFR here...let me be crystal clear), there have been attempts of bad actors getting an account to admin status. Given how close Eostrix came to working (4 days shy, after two years of effort), it's likely some have succeeded without being detected. I've even heard scuttlebutt of admin accounts being sold in the past. Getting back to SFR; given the past, it's not unreasonable to ask the question. It is unreasonable to make accusations of sockpuppetry. I don't wish to fault SFR's nominators, but I do think that nominators should strive to uncover any potential hiccoughs a candidate might encounter. SFR's substantial knowledge on their first day of editing should have been addressed in the opening nominations and/or acceptance of nomination. -- Hammersoft ( talk) 14:25, 14 September 2022 (UTC)

As I understand it, Eostrix was already under investigation by Arbcom anyway when the RfA turned up; at the very worst they'd have passed RfA and got level 1 desysopped and blocked shortly afterwards. Ritchie333 ^{(talk) (cont)} 14:30, 14 September 2022 (UTC)

Perhaps. We'll never know, due to the private nature of the evidence (which is fine, certainly, in this case). Still Eostrix went at least most of 20k edits and 2 years without being detected. -- Hammersoft ( talk) 14:33, 14 September 2022 (UTC)

Actually we've stated that Eostrix was already on our radar as an IW sock before the RfA started so I can confirm that. The RfA just forced us to complete the investigation in a way it might have lingered otherwise. I don't think the nominators for Eostrix could have reasonably known that socking was an active concern and indeed I don't think anyone at the RfA brought it up. Barkeep49 ( talk) 15:06, 14 September 2022 (UTC)

If it's true that some sock puppeteer has already passed RFA, then... where's the damage? Which rogue admin is it? What bad blocks and bad deletions have they made? Clearly if there are already sock admins, they're not doing any harm. Let that thought sink in a minute as you consider how important it is to ferret out sockpuppet RFA candidates. And let's remember Ed, the admin who made socks and used them to vote multiple times in an arbcom election. Sometimes socking comes after RFA.

It's completely bogus that you (and others) think there is something suspicious about SFR's first edits. It's not there. He's not doing anything difficult, complicated, or--most importantly--undocumented. What needs to happen here is for certain very longstanding members of the community to "come to Jesus" that just because something was difficult for them doesn't mean it's difficult. One of the great quotes, from SFR's talk page in March, is an admin saying that even after 14 years, they still have trouble with some of the tasks SFR did in his first year. Well here's an unpleasant truth: if you've been using software for 14 years and you still find it difficult, you're technologically illiterate. There's nothing wrong with that, a lot of people have difficulty with computers, but don't cast as suspicious that someone knows how to use wiki markup language, which is the easiest markup language in existence (and was designed to be the easiest). It's like you're amazed a child can build Legos on their first day playing with them. Wikipedia, like Legos, is designed from top to bottom to be easy to pick up. Here's another thought to sink in: 10-year-old children pick this up no problem. If you really think it's hard to file an AIV, and that someone doing that is somehow suspicious or has something to answer for: it's you. It's just you, unable to process the notion that someone is better at this than you are.

Having been on the receiving end of these same types of accusations four years ago, I'm a little bitter about it, if you can't tell. But we really can't let this place become an idiotocracy, where "too good at the software to be new!" is considered a legitimate concern. Levivich ( talk) 14:44, 14 September 2022 (UTC)

Helpful link for people who might not know: Edgar181 is "Ed". Deli nk, one of his sockpuppets, made almost 80,000+ edits and was created in 2005. To some extent it's impossible to predict whether someone is going to shatter the community's trust like that.

As for people can pick up our markup language... I started editing wikiHow (which uses MediaWiki) as a 12 year old. I have no other computer science-related experience. I recognize that my experience is not universal, but I do think it is kind of ridiculous that people often get accused of being sockpuppets simply because they're competent and "know too much" like Levivich. It gets to the point where actual sockpuppets might know to avoid looking competent "too soon" while actual newbies won't be examining their edits through the lens of "will this make people think I'm a sock"? Clovermoss (talk) 14:55, 14 September 2022 (UTC)

I think that the rapid and prominent engagement with various drama boards is as much, if not more, of a sockpuppetry red flag as wiki-markup know-how. I was ultimately swayed by further explanations of SFR's editing history and various other arguments vouching for them, but I think it's worth noting that the original objections were not merely about a higher-than-baseline level of competence: it was higher than baseline competence, plus disproportionate engagement with dramaboards that are like catnip for sockpuppeteers. _signed, Rosguill ^talk 15:57, 14 September 2022 (UTC)

I can see why this might make someone concerned but aren't we suppossed to assume good faith? No one brought this up in the Eostrix RfA and their entire thing was hunting sockpuppets, ironically! It was private evidence that nobody could have known that sealed the deal. As has been brought up, WP:SPI has higher standards for evidence and connection to a specific sockpuppeter (not vague wave-of-hand "we know they must be a sock we just don't know who" – I think in a lot of those cases WP:NOTHERE block would make more sense). I don't think it's okay to be blatantly speculating about whether or not someone could be a sock, even in an a RfA. Policies like no personal attacks shouldn't just be thrown out the window. Clovermoss (talk) 16:08, 14 September 2022 (UTC)

Eh, I'm not sure it's fair to call sock-related speculation a personal attack if succinctly raised in a civil and non-bludgeoning fashion. We'll see how this RfA turns out, but at least as far as my own voting process was concerned, being able to see early opposes raise concerns and questions that were then addressed satisfactorily seems like the system working as intended. _signed, Rosguill ^talk 16:51, 14 September 2022 (UTC)

I disagree that dramaboards are like catnip for sockpuppeteers. Most dramaboard participants are not socks, and most socks do not participate in the dramaboards. There are a few LTAs who like to troll, but the type of sockpuppeteer that would run for RFA is not the type to be stupid enough to prominently engage with various drama boards, especially early in their career. The idea that dramaboard engagement is a sign of sockpuppetry is a myth, and a harmful one that encourages false, empty accusations. Levivich ( talk) 18:52, 14 September 2022 (UTC)

Levivich, your response here clearly exemplifies why some people, myself included, strongly veer away from RfA. This is the sort of toxic response one now has to expect if they do anything other than support a candidate. The concern I raised was a legitimate concern. I wasn't the only one who saw the potential issue. I didn't accuse SFR of being a sockpuppet nor imply any malicious behavior on their part. I wanted the disparate points reconciled. I specifically placed my comment in the neutral section to avoid implying anything. Yet, here we are with you taking offense at my comments. Referring to my concern as "bogus" and further impugning me by stating I'm acting as if a child putting together Legos is amazing to me is right off the rails a personal attack. My concerns, as expressed in my comments on the RfA, were legitimate. It wasn't an accusation. -- Hammersoft ( talk) 15:51, 14 September 2022 (UTC)

I do think that people shouldn't be afraid to oppose a candidate for what they feel are good reasons. I disagree with a lot of the oppose !votes, but apart from the ones that imply socking or malicious intent from the start on SFR's part (which I definitely take issue with), I think reasonable people can disagree about what makes a good admin. I think SFR's experiences as a new editor are actually more likely to make them more sympathetic to the "average joe" so-to-speak, but maybe it's just me. We need more editors with from a newer wikigeneration with differing perspectives, in my opinion.

I think a lot of Levivich's frustration stems from their experience (some of which is even fairly recent!) of people accussing/implying that they are a sock. Two wrongs don't make a right in regards to the lego comment, but again, why is one borderline personal attack it considered socially acceptable on-wiki to imply someone might be a sock while someone calling that logic out isn't? Clovermoss (talk) 16:02, 14 September 2022 (UTC), edited 16:52, 14 September 2022 (UTC)

"Hi, my name is Bob."

"Hi, I think you're lying about that."

Only on Wikipedia is this considered normal. Only on Wikipedia would someone justify that by saying "well I'm just raising concerns" or "I have suspicions" (variations of the old lark, "I'm just asking questions!"), as if that was a justification for accusing someone of lying. Just because we call it "sockpuppetry" doesn't mean it isn't an accusation of deceit. And really serious deceit. And it also doesn't matter if you use the word "sockpuppetry" or the words "had another account". Those who have "suspicions" or "concerns" that an RFA candidate has or had another undisclosed account, in the face of the candidate denying it, are saying, in public:

1. That the editor is lying about who they are
2. That they are in fact someone who was kicked out of the community or otherwise had such a poor reputation that they cannot disclose who they are
3. That they are engaged in a sophisticated, long-term scheme to deceive the community into giving them advanced privileges on this website by making them an admin

These are some damn serious accusations! Here some actual opposes in this RFA:

• "This editor had a very unusual start, highly suggestive of a returning editor, and set about from the outset appearing to be checking all the right boxes towards RFA, which is where it always appeared they were aiming." - making all three of the above points
• "I've reviewed a number of the candidate's early edits, and it's pretty clear this isn't their first account. I would like to see ScottishFinnishRadish be more forthcoming about their past editing history." - a bald accusation that the editor is lying to us, with no evidence beyond the vague assertion about having "reviewed a number of the candidate's early edits"
• "I am concerned that this user has edited under another account." - an accusation of lying phrased as a "concern"
• "...per socking concerns raised by eminent colleagues"
• "It's really hard to believe this is his first account..."
• "Their start on Wikipedia looks suspicious..."
• "There seems to be something fishy about this editor..."

That's seven accusations that the candidate is a liar. And there's more "per" those !votes. Of course there are also plenty of other oppose !votes for other reasons that do not involve any accusations of lying. But accusations like the ones quoted above shouldn't be tolerated. It's smear, it's not reasoned argument. Only on the internet would people say these things; none of these people would say this out loud without any evidence if we were all in the same physical room together, even if they might be thinking it.

The concern you raised was not a legitimate concern. You either believe an editor when they deny socking, or you're calling them a liar. There is no respectable place in between those two options where you can publicly doubt that someone is telling the truth but not also be calling that person dishonest.

You did imply malicious behavior on their part. Here is what you said:

I'd like to point out that on this editor's very first day of editing they had already learned about wikilinking [12], knew about WP:CSD#G11 (admins can see deleted history of this userpage), knew about WP:AGF [13], knew about how to request semi-protection [14], and knew about WP:AIV [15] (and despite saying they didn't know about template warnings, they were using {{uw-vandalism2}} less than 3 hours later [16]. That's a rather astonishing level of knowledge for someone on their first day of editing. I would like to see this reconciled with their statement that they have never had another account.

You implied it was not their first day of editing. The examples you pointed to are ridiculous. Wikilinking? Knowing about CSD, AGF, and AIV? Learning how to use a template in less than 3 hours?! How many hours do you think it should take, ffs? You asserted that these things contradict "their statement that they have never had another account" and needed to be, in your words, "reconciled". You cast doubt on the veracity of their denying having another account, based on their having a basic knowledge of common parts of Wikipedia like wikilinking, CSD, AGF, and AIV, and on the ability to learn how to use a template in 3 hours. (I can't let that go, it's so... wow.) You implied SFR was lying to us, or appeared to be lying to us (the most charitable interpretation I can summon), based on nothing beyond their following directions in our documentation, and all the polite phraseology and rhetoric doesn't make that any better, and it doesn't matter if you put that in the neutral section or the oppose section, or on his user talk page as another admin did last year, or anywhere else.

And no, my calling you out for baselessly questioning whether one of our colleagues was lying will not have a chilling effect on RFA. Hell, if it is chilling these sorts of serious unfounded accusations, then that's a good thing. But in reality, I'm not calling out anyone for opposing (there are plenty of well-reasoned oppose !votes in this RFA), I'm calling people out for making baseless accusations of dishonesty. It doesn't even matter if it's during an RFA or anywhere/anytime else. It's just that it's particularly egregious when it happens at an RFA, and as you mentioned, when there are a dozen people engaged in it.

Funny you should say comments like mine are why you generally avoid RFA. Comments like yours are why I generally avoid it. It was not legitimate and I am not sorry for calling you out on it. This sort of behavior needs to stop being tolerated. I know there are some long-term editors who have spent over a decade casually accusing their colleagues of being liars by raising "suspicions" or "concerns" about having another account, and by asserting that competence in new editors is somehow suspect. They've been doing it for so long that they've lost touch with what they're actually doing: accusing their fellow volunteers of lying. I am glad to volunteer some of my time to help reconnect these editors with reality by reminding them to stop questioning the honesty of our colleagues without damn strong evidence of dishonesty, and reminding them that early competence is not damn strong evidence. Levivich ( talk) 19:33, 14 September 2022 (UTC)

Levivich, I'm not going to read the above (and if you're offended by that, tough). Frankly, it's too long (north of 1200 words). It appears to be trying to prove what a despicable person I am. If that's the case, my talk page is that way, Alternatively, WP:AN/I. Further discussion about me is not appropriate to this thread at WT:RFA, which is about RfA. And no, I'm not trying to squelch conversation about me; it's just not appropriate here. Thanks. I hope you have a great day. -- Hammersoft ( talk) 20:17, 14 September 2022 (UTC)

I'm concerned Hammersoft is lying about not reading Levivich's paragraph. Should potential liars be admins? Hey, that rhetorical technique is easy! Thanks for showing me. -- Floquenbeam ( talk) 20:25, 14 September 2022 (UTC)

I understand why, statistically speaking, certain patterns of behaviour appear suspicious. I think the community, though, should strive to limit the number of situations that trigger this reaction. Ideally new editors would familiarize themselves with the environment and be ready to contribute, and the community should welcome them. Expecting everyone to go through a floundering phase can become a self-fulfilling prophesy if everyone who doesn't is treated with doubt. isaacl ( talk) 15:52, 14 September 2022 (UTC)

I agree 100% with this. Clovermoss (talk) 16:09, 14 September 2022 (UTC)

We haven't had a bad-faith admin do anything truly horrendous yet. That's mostly a matter of luck - there are plenty of things that +sysop allows that are irreparable, either in a PR sense, or unrepairable without a truly stupefying amount of work, or in a few cases literally so. Most of them we don't talk about. One of the most obvious, and one that we can talk about because it's already happened (before we had cascading protection, so it didn't require being an admin at the time), is to put a shock image on the main page disguised in such a way that it takes a long time to remove. "Bad blocks and bad deletions" are about the least disruptive things an admin can do. — Cryptic 16:20, 14 September 2022 (UTC)

I wasn't suggesting that bad blocks and bad deletions were the most disruptive things an admin can do. In fact I chose those examples so as not to spill WP:BEANS about worse things admins could do. :-) But the point is, as you say, we haven't had a bad-faith admin do anything truly horrendous yet. After 20 years, I don't think there is any evidence to support the conclusion that it's mostly a matter of luck, although perhaps there is evidence and I'm just not privy to it. Levivich ( talk) 19:37, 14 September 2022 (UTC)

I don't usually read this page, but find myself here because of BK49's ping to me in the section above, where I also had the pleasure of reading this misguided discussion filled with curious assumptions. Clovermoss, this might answer some of your questions. SandyGeorgia ( Talk) 07:31, 15 September 2022 (UTC)

I really dont see how its a problem. You can be suspicious that somebody is a sockpuppet of a banned user without knowing which banned user it is, or being able to prove it with or without checkuser. And the potential for a banned user to shift our articles as an admin increases exponentially over that of an editor, especially in one of the DS topics, so I dont particularly find it a problem that the standard of suspicion to vote no in RFA to be exponentially lower than to block an editor as a sock. And that is irrespective of the current RFA, for which I trust those nominator's judgement and I personally give approximately a 0% chance of being a sock of IW (no idea on anybody else, though I find the explanation on how one finds their way to understanding the dark side of WP totally reasonable, since it mirrors my own experience), but I dont see why somebody cannot say this feels too suspicious for me to be willing to have this person be given the admin bit. This isnt a ban discussion or anything other than does the community feel comfortable giving this person access to the rights an admin has? And being suspicious of sockpuppetry can certainly factor in to that level of comfort. nableezy - 16:31, 14 September 2022 (UTC)

• the problem with bad-faith admin is not blocks or deletions. It is quite the opposite. They might intentionally not block other vandals. Thats negligible problem though. The major problem is undeletion. A lot of editors had exposed their real identity in the past, then got it rev-del'ed later. Either photo, or real full name/address/social media, or any information which can identify an editor. Imagine some bad-faith admin leaking that info elsewhere. There are more concerns, but there are WP:BEANS as well. —usernamekiran (talk) 19:12, 14 September 2022 (UTC)

All very true, and the histrionics above attacking somebody for not having sufficient confidence to grant somebody those rights is silly. Adminship is not some badge of honor, it is not something anybody is entitled to. It is something the community grants iff we feel we trust you enough to give it to you, and you do not need damn strong evidence of dishonesty to not feel that the level of trust you have in a person is sufficient to grant them admin. And it is not accusing somebody of lying by saying Im not convinced. I know of at least one redlinked RFA I have in my watchlist in which my oppose will be I just dont buy that this is your first account. And I dont buy it because of x, y and z, and as such I just dont have the requisite amount of trust to be in favor of you having these rights. nableezy - 19:40, 14 September 2022 (UTC)

• Pastor Theo happened. Eostrix happened. Edgar181 happened. Fæ happened. There are several other ancient ones I'm vaguely aware of that caused significant community division at the time but I'm not going to dig through archives to find.
  All of these caused significant damage to the project by causing divisions in the community and creating distrust, which is a real problem on a project based on trust.
  In my RfA I was asked if I had a different account before. The answer was yes. It's a completely legitimate question. The community is also able to decide whether or not they believe the answer provided.
  As an aside (I'm not commenting on the current RfA), but the typical reasons people look like they've edited Wikipedia before are because they've edited Wikipedia before. There aren't all that many people who pick up stuff quickly. There aren't all that many people who are wikiaficienados who have had a love of wikis for decades and just decided to join the most successful wiki of all time in 2022. I looked like a not-new account when I joined because I was a not-new account when I joined. Most people who look like not-new accounts are not-new accounts. That can be completely legitimate, but because of the attraction adminship has to the type of people who would be obsessive enough about a project to sock after being told they aren't wanted here, the damage that can cause, and because of the history here with actually damaging sock admins, the community certainly has the right to ask questions, and if they aren't satisfied, to oppose on it. TonyBallioni ( talk) 05:42, 15 September 2022 (UTC)

  You named only a few of the contemporary abusive admins. But well before any of them, Archtransit happened. I knew Archtransit was a sock, but I felt too new to speak up, anywhere, especially with a cabal of "we admins" after me already (not one of them still a sysop today-- two desysopped by Arbcom, one desysopped for socking, and one warned by the arbs and now deceased). So Archtransit got to have their way for a while. I see some editors who presumably aren't as aged as I am insinuating that none of the many desysopped and socking admins ever caused any damage; I disagree. Perhaps we have different yardsticks for measuring damage, and damaging editors is as bad as the proverbial erasing the mainpage. There are very good reasons we examine editors who submit to RFA, and when they seem to be less than forthcoming about their own editing history, they just make things harder on themselves. Since there has been no accusation of sockpuppetry on the current RFA, I don't know what Clovermoss is even talking about; that supporters twisted a valid oppose into that as a rallying cry is another matter. RFA is a place where we most certainly should and can examine editors, even if others are later going to twist and distort opposes into statements they never were and attach words to them that were never used. SandyGeorgia ( Talk) 07:45, 15 September 2022 (UTC)

  "Since there has been no accusation of sockpuppetry on the current RFA". I interpret the following comments as aspersions of being a sockpuppet:

  • "Of the main space edits, the editor started out principally by undoing edits...a very strange pattern for an obstensively novice editor" [2]
  • "This editor had a very unusual start, highly suggestive of a returning editor, and set about from the outset appearing to be checking all the right boxes towards RFA, which is where it always appeared they were aiming." [3]
  • "But having read the comments above, as well as the observations by [other editor], I am 'concerned that this user has edited under another account. [4]

  Also consider "I don't endorse the sock accusations above" [5] which implies they exist. In fact, having reviewed these comments, I find the suggestion that there have been "no accusation of sockpuppetry" to be disingenuous; at best people did not intend to fire off sockpuppet accessions, but that's how they were interpreted by others. Ritchie333 ^{(talk) (cont)} 08:14, 15 September 2022 (UTC)

  Funny that, I interpreted the reference to "sock accusations above" as relating to the original thread with Spartaz, who hasn't weighed in on the RFA. We're all responsible for what we write, not for what other people read that wasn't written. A returning editor is not the same thing as a sock; maybe Tony's not-new is more apt. As SFR has acknowledged, they were in fact not new (rather had been following Wikipedia extensively for years, by their account). Had they not failed to disclose something they appeared to have known they should have disclosed in Q3, and explained that before they had to be asked, none of those questions would have needed to be asked. Considering they decided to stay mum, having now stated they had years of following Wikipedia, why would they or anyone expect the non-disclosure to be ignored at RFA, where everyone will examine your editing history, and then by hiding something, it becomes a red flag? This is a problem of the nomination and the opposer who mentioned "barreling ahead" seems to have pegged it the best; bad cases make bad law, and using this kind of messed up RFA nomination to imply excessive scrutiny at RFA can't lead good places for the process overall. SandyGeorgia ( Talk) 08:41, 15 September 2022 (UTC)

I have now weighed in and have posted my support on the RFA. When I am active I spend a lot of time at AFD and it's an absolute Magnet for socks. I absolutely abhor the impact of socking on discussions so am generally quick to follow up a precocious new account. I used to do it off wiki through email but these days generally raise questions at account talk pages. The main reason is to create a permanent record of the denial if the user turns out to be a sock later on as I ended up having to provide off wiki evidence in an arbcom case involving an admin who had edited with a previous account. Anyway, enough ancient history. The situation here is that SFR was very quickly clueful and this raised serious doubts. I therefore challenged him. I also spent a considerable amount of time doing textural analysis on their edits and did not find any correlation with any other editors active or banned. Finally, I approached arbcom off wiki and had a conversation about my concerns. I have no personal knowledge of what background check arbcom conducted but the concern raised was not new to them and I was entirely reassured at the end of the discussion. Hopefully this will be useful. I am very comfortable in my support. If SFR is a sock, he is the most sophisticated one I have ever encountered. Spartaz ^Humbug! 13:37, 15 September 2022 (UTC)

It is helpful to hear from you Spartaz (apologies for not pinging you when I mentioned you here, but I was concerned that could be construed as canvassing). I am glad you were reassured, but your concerns are a bit different than mine, and by linking to your (old) concern, it now appears I gave a misimpression of what my concerns are. Had the candidate and the nominators not "barreled ahead" and instead put out this old concern, with your reassurance, from the start, we'd have a different RFA. Anyway ... It is interesting you bring up textual analysis in relation to ArbCom, as it has been my impression that some of the arbs ignore it, even when it is quite conclusive, but that's another topic. SandyGeorgia ( Talk) 18:01, 15 September 2022 (UTC)

The Textual analysis was for myself and my own concerns and was not shared during my conversation with the arb, indeed I have never disclosed before that I do this if I encounter a suspected sock at AFD. This would also be why the noms and candidate didn’t bring it up because I didn’t made SFR or anyone else aware of the extent of my concerns. To be clear also I was reassured by my own assessment & not just because I spoke to Arbcom. Spartaz ^Humbug! 19:27, 15 September 2022 (UTC)

Yes, I understood you to be saying that; I brought up the textual analysis re a very different situation ... one that leads me to wonder if you'll run for ArbCom at some point, so we might get an editor on the committee more conversant in textual analysis than what we appear to have now. Completely unrelated to this situation ... SandyGeorgia ( Talk) 19:36, 15 September 2022 (UTC)

There is no realistic likelihood that the community would entrust any further powers or permissions to me and frankly I wouldn’t either. Too many flakey decisions, barely active and I’m too prone to being unacceptably abrasive. I’m definitely getting grumpier and less patient the older I get. Spartaz ^Humbug! 22:11, 15 September 2022 (UTC)

Growin' old ain't for sissies, but there is a sense of freedom that comes with it :) Thx for the discussion, and hopefully the textual analysis situation will improve somehow. SandyGeorgia ( Talk) 22:22, 15 September 2022 (UTC)

I don't agree with the assertion that "All of these [troublesome admins] caused significant damage to the project by causing divisions in the community and creating distrust, which is a real problem on a project based on trust." I think the impact of these editors is minimal and maybe affects a couple of hundred people at the absolute worst; I reckon the effect on the tens of thousands of people who pop onto Wikipedia just to make a small edit, or the millions of people reading an article, is near zero. I think Andy Mabbett said it best during the WP:FRAMGATE saga with admins jumping ship left, right and centre : " I've just taken a look at Special:RecentChanges and the project seems to be getting on with business as usual, just fine." This is what I would expect. The most disruptive thing I can think of in recent memory is Wikipedia:List of hoaxes on Wikipedia/Alan MacMasters and I don't think any admins were involved in that (aside from me closing the AfD as "snow delete"). Ritchie333 ^{(talk) (cont)} 09:43, 15 September 2022 (UTC)

Ritchie, I suppose that's like mortality arguments ... it's meaningless that X disease only kills one in a thousand people, until you're one of those people. Then it's very meaningful to you and your family. On Wikipedia, it's not only people-- processes are also damaged. This RFA is an example; we have people seemingly advocating that we can't raise legitimate concerns at RFA because others interpret them as accusations of sockpuppeting. It would be most unfortunate, for the process, if that thinking took hold, because I should have spoken up at Archtransit in 2008. (I've never heard of Esotrix by the way ... I don't read RFAs usually unless I have some familiarity with the editor, or unless I see unfounded claims of content creation in the nom statement, along the lines discussed above in the thread with Kudpung, so I missed the whole Eostrix thing, although I was aware of Edgar because of the FAs). SandyGeorgia ( Talk) 18:08, 15 September 2022 (UTC)

I also looked into Pastor Theo. The link didn't say what articles or topics they were vandalising / disrupting / pushing a POV (delete as applicable, I don't know which) so I don't know about the disruption. I then note they instantly retired (no comment on that) and their talk page was protected to stop them filing an unblock request, despite not obviously doing so. I really do not get why on earth people care about this nonsense from thirteen years ago when we've got more important issues like hoaxes and extreme POV pushing on topics like WP:ARBIPA that has been going on for decades. Ritchie333 ^{(talk) (cont)} 11:09, 15 September 2022 (UTC)

• I think that's a very apt summary by TonyBallioni. One does indeed have a right to be suspicious and it would be perfectly natural to be so. The problem is that some people are so desperate for a bit of drama that they will look under every rock for an excuse to claim something is controversial. Ironically, serendipitous detection of a corrupt admin and his (or her) socks and other antithetical exploits can have devastating consequences for the discoverer and that's why many of them remain schtum about what they found. Kudpung กุดผึ้ง ( talk) 11:35, 15 September 2022 (UTC)

  There are other reasons; I've learned the hard way not to give away the tells of a long-term abuser. I've even withheld tells from the CUs, as I simply don't trust the BEANS won't be spilled. Curiously, ArchTransit and Mattisse had a shared tell (not saying they were the same sock). SandyGeorgia ( Talk) 18:20, 15 September 2022 (UTC)

  So I've read the comments since I last responded here. I wanted to say a few things:

  • I do think trust is an important part of any RfA candidate. [6] I'm aware that people have different standards for RfA and I don't think people should be afraid to oppose a candidate. I've had some very tramautic real life experiences that have taught me that forcing everyone into an environment where they have to agree can have some devastating consquences. I have bad memories associated with these IRL experiences and thus tend to avoid conflict in general. However, I also try very hard to see other people's perspectives and I never want to feel like I'm exhibiting an us vs them mentality ever again. I was sincere about starting an actual conversation here.
  • The comments implying that SFR has used a different account before still leave me with unease. My conscience screams at me that it's simply just not right and I felt obligated to say something because of that. If I had to give words of what causes that, Levivich's comments about lying express a lot of how I personally feel. I do think it's important that these discussions have more faith in new editors. I think it is harmful that new editors who've done nothing wrong and are simply competent get people asking questions. I do not think this is the healthy, welcoming atmosphere we should have to new editors. New editors can be aware of what goes on behind the scenes, even without previous experience here or on other wikis. It's not impossible that someone could click "Community portal" from the main page, see the policy pages where we discuss things, find it intriguing and want to get involved in that. Same goes for anyone who clicks the Wikipedia:Task Center mentioned in a lot of welcome templates. There's instructions everywhere and most things on-wiki are fairly transparent. If a new editor decided they wanted to read my entire talk page right now, they'd be able to understand that the Wikimedia Foundation exists and maybe form an opinion on it. How well would that go for them in an environment like this?
  • Obviously, an actual LTA becoming an admin has the potential to do serious harm to the project. I do think an interesting point has been raised about how someone can not feel comfortable supporting or have some sort of private evidence that they wouldn't nessecarily want to blast on-wiki. I also agree with the concept that no one "deserves" to be an admin. I do think that looking at people who run because they have a sincere desire to help the community and then getting this is part of what makes the RfA environment called "toxic", though.

  I'm still going to give all of this some thought, but in general I think I should be stepping back here. I think I need some time away from all this for my own wellbeing. Clovermoss (talk) 22:17, 15 September 2022 (UTC)

  Thx for the discussion, Clovermoss. The "lying" posts were the most offensive part for me, and hyperbolic, and they so distort what the concerns are. Perhaps we can conclude that mistakes were made all round. It is apparent that my initial oppose was muddy (my bad, my responsibility, and not entirely atypical of my writing). I'm fairly certain at this point the candidate realizes mistakes were made in the approach. I hope the nominators learned something about how not to position an RFA vis-a-vis past issues. It's also time for me to step back; I've taken more than my share of the discussion. Best regards, SandyGeorgia ( Talk) 22:28, 15 September 2022 (UTC)

  Hi SandyGeorgia. I wanted to be a bit more precise about certain things after some time and reflection so you could try and understand my train of thought better, but I understand if you've reached a point where you're no longer interested.

  The main reason I wrote the first item in the list above was because I wanted to recognize that the opposes aren't a monolith. I think the reason supporters are focusing on the !votes that imply socking/previous editing (and there are quite a few, as described by Ritchie333 in this thread) is because they take issue with that mentality. I started this discussion on WT:RFA because of the general trend, not to focus on your oppose specifically. I mentioned Eostrix because at the time I started this discussion, one of the opposes explicitly mentioned it and said something to the effect of "because we were almost burned" by said RfA. But what I was trying to get across in bullet point one was some sort of response to this: Since there has been no accusation of sockpuppetry on the current RFA, I don't know what Clovermoss is even talking about; that supporters twisted a valid oppose into that as a rallying cry is another matter. I guess what I'm trying to say is I think the reason that supporters are focusing on this is because they don't inherently have an issue with the other opposes even if they disagree with that accessment. There are opposes where I go "yeah, I can see why someone could think that and not feel comfortable supporting the candidate".

  Which leads to my second bullet point. I don't think Levivich's response that mentions the part about lying is nessecarily hyperbolic. Personally, I would've toned it down a little, but I think they have a valid argument. Accusing someone of sockpuppetry or having some sort of undeclared editing history is essentially saying "yeah, I think you're lying about who you are". Words like that have impact and I think it's important to consider that. I think almost everyone putting themselves forward as an RfA candidate cares deeply about the project and wants to contribute in another way they can, so accussations like this shouldn't be taken lightly. It's important to remember there is another person behind the screen. Not to say you're taking this lightly but it was something I wanted to emphasize, again, given the general trends seen in this RfA.

  As for the third bullet point, I guess what I'm trying to say is that even though an LTA becoming an admin could do very serious damage, I don't think accessing every candidate through the lens of "is this someone's malicious plan to fool the community?" is nessecarily the most productive way to view RfAs. I think it contributes to a more toxic environment. Going into the new editors side of this (because SFR has had people with concerns about their imo perfectly normal edit history from the start) is that people like us who've been around in the community awhile typically find it easier to make their opinions known. But new editors might feel pressured to self-censor if that goes against some established norm. Everyone knows that new editors don't do x... except the people who don't feel that way don't feel comfortable expressing that opinion. The overall harm of if we're wrong has a very real risk, too. It's not just about mitigating the effects of abusive LTAs. It's the difference between "hey, I've seen you doing x, y, z well. Keep it up!" vs "Hey, I've seen you doing x, y, z well. What's the past editing history you're hiding". I've seen more than one example of the latter and I think it's harmful and bite-y, when the very real possibility is that we have an actual newbie whose just interested in doing the best that they can. Clovermoss (talk) 12:09, 16 September 2022 (UTC)

  Thx, Clovermoss.

  • Accusing someone of sockpuppetry or having some sort of undeclared editing history is essentially saying "yeah, I think you're lying about who you are" I strongly disagree, and levivich did seem to be pointing specific fingers here. Thinking something is so insignificant that you don't even need to raise it in an RFA nomination is not the same as lying. Which leads in to:
  • I don't think accessing every candidate through the lens of "is this someone's malicious plan to fool the community?" is nessecarily the most productive way to view RfAs which is why this RFA has been so frustrating. If you still think that's what's happening, the reason I opposed has been entirely missed. I will never support a candidate whose trajectory suggests they are barreling ahead and ticking the boxes on the path to RFA. That is the basis of my oppose. Which opposer has viewed a candidate through the lens you describe ?

  Had the candidate and the nominators consulted Spartaz first, and brought that past issue forward in the nomination statement, we would not be where we are (a contentious RFA). If an issue was significant enough that you had communication with the arbs about it, how can one think it's not significant to the broader community? Had this issue been covered in the nomination statement, I'd still be an oppose based on the editing history, but it would have been my standard mild oppose. We have a TOOSOON candidate who, with the years of observing Wikipedia they say they have, did not understand that Wikipedia is not like a personal Wiki you run where no matter how many participants you have, there's still more or less a group in charge. The arbs are not "in charge" here; the entire community is. The "lying" framing is offensive, and the "socking" rallying cry is a strawman.

  And finally, I tossed the candidate a "get out of jail free card" with my question about Mally. A clueful answer would have caused me to soften my oppose; an exemplary answer could have led me to strike it. A really bad answer would have caused me to double down on the strong oppose. The answer left me where I started: the candidate had not absorbed the full dynamic that led to Eric's demise, and did not recognize the power admins hold over content creators. SandyGeorgia ( Talk) 12:56, 16 September 2022 (UTC)

  @ SandyGeorgia: Since our opinions start to diverge after the whole "does this RfA include allegations of sockpuppetry", we should probably focus on that? Again, I was talking about the general trends, but personally I could see how someone would intrepret your oppose as implying that:

  Strong oppose. Creating their account on 22 January 2021, this editor has been editing only a bit over a year-and-a-half. When they started editing on 11 February, one of their first edits was to blank the welcome template—an odd way to start editing. My concerns are in line with those raised early on at the editor's talk page: "You must be the most precocious editor we have ever had". This editor had a very unusual start, highly suggestive of a returning editor, and set about from the outset appearing to be checking all the right boxes towards RFA, which is where it always appeared they were aiming. Content creation was never primary for this editor, who nonetheless indicated considerable knowledge of Wiki-processes. An 800-word and an 1,110 word GA—passed by editors I'm unfamiliar with—do not convince me, and in fact, look like another box to be ticked on the much-too-quick route to RFA. I can't recall recently seeing an editor with only 18% of their edits in mainspace. I'm additionally concerned that neither the nominators nor the RFA candidate addressed this editor's odd history in their nomination statements. Considering this editor's early history, there is nothing that can be said to convince me that it is not much too soon to trust this "new" editor with the tools. This is my strongest possible oppose. [7]

  What was it you were trying to go for in this oppose, if not an implication of socking or some prior undisclosed editing history? I feel like I might understand your perspective better if we can find some sort of common ground. Then continue the discussion from there, if you're still interested. Clovermoss (talk) 01:44, 18 September 2022 (UTC)

  Why does a new editor blank the welcome template? We don't know that answer, but I've never seen that before. Considering they started by intensely fulfilling requests from unregistered (typically new) editors, to remove the welcoming template (which shows that they too are new) seems like something that could mislead other new editors whose edit requests were filled. But why? Confusing.

  Most editors enter Wikipedia via a certain content area, and they branch out from there to other (eg governance, content review, noticeboards, copyright) specialty areas, while this editing trajectory went the opposite direction. Their start that way was so unusual that another editor called it out, and yet this wasn't raised in the nomination. I don't trust the tools to editors who show up wanting to be admins rather than showing up wanting to edit content and gradually discovering their strengths lie elsewhere; whether they edited as an IP, or are a cleanstart, RTV, legitsock, anything else, I still don't trust that trajectory. If I had stated that first, and then backfilled the other concerns, including that another editor had raised these queries, perhaps my construction of the paragraph would have been clearer. A new account moves straight into governance, but removes welcome template to (??) maybe not appear new is odd. But at this point, probably everyone is bored silly of the topic, and maybe we should take it to my talk if you still want more clarification. Best regards, SandyGeorgia ( Talk) 03:28, 18 September 2022 (UTC)

@ SandyGeorgia: I'd prefer to keep the discussion here if you don't mind just so it isn't scattered. As for why a new editor would blank the welcome template, I've never seen that specifically, but I have seen many people blank messages they've read. So that's the assumption I would typically go with. Based off their answer to Q6, maybe they didn't think they needed the welcome message. As for your trajectory comment, I understand where you're going with that. My personal opinion is that just because it's different than what's typically expected, doesn't mean it's an inherently wrong path. I will say that it's a factor that results in me thinking this RfA wasn't going to be unanimous regardless of how the nom went. People in general don't like it when people focus on the "dramaboards". A lot of RfA-related advice mentions this specifically. Another reason I'd think this isn't a returning account, actually, is because wouldn't someone whose intent is to become an admin from the start know to avoid that? I think I expressed this line of thought a bit better in my earlier comments, though, which you've already read.

Temperment is something people mention in RfAs all the time. I wonder if maybe you'd focus on the latter concern more if we could rewind time? I see you kind of go a bit further into your reasoning on replies in the oppose section now. As I said, there are opposes that I find relatively convincing. I guess what I'm saying is that my issue truly is with the broader implication going on in this RfA is that SFR is hiding something. The nom could've been handled differently, but I think JBW made an interesting point in the RfA itself with I did some IP editing before creating an account. I have never made any secret about that, but nor, as far as I remember did I mention it in my RFA. That wasn't because I wanted to hide it, it was because I didn't think it in any way significant or relevant. I'm not sure how I feel about how people closing the RfA should take this part, though.

Anyways, this conversation has been slightly one-sided even if we've both been participating. Anything you want me to clarify in regards to my thoughts? Clovermoss (talk) 12:39, 18 September 2022 (UTC), edited 15:39, 18 September 2022 (UTC)

No clarification needed; I understand your thoughts (I just hesitate to keep putting this page through more of this, but neither do I want to ignore your questions).

Re Temperment is something people mention in RfAs all the time. I wonder if maybe you'd focus on the latter concern more if we could rewind time? hindsight is 20–20, but no, I don't think that would be the focus of how I would frame this particular oppose even if I could rewind. (I would compose my oppose more carefully and not write it from my phone while engaged in a labor-intensive fundraiser, but then, I knew I'd have no break for at least a full week.) RFA ultimately boils down to trust. If I pondered long enough, I could probably come up with an example of somehow who once or twice upon a time blew it in the temperament department, and yet I still trust not to use the admin tools in a huff, personal dispute, or when involved. Repeating, the focus on IP edits is misplaced (I edited for months as an IP, and only finally registered an account to initiate a mediation with a dynamic IP who was a COI/POV editor biasing an article using their own website as a source). IP editing is not the issue for me in this RFA; an editing trajectory suggestive of an editor who is here to seek adminship over content creation was, and is. SandyGeorgia ( Talk) 16:16, 18 September 2022 (UTC)

@ SandyGeorgia: I should have been more precise with what I meant in regards to temperament. I guess my off-the-top mental definition is a bit more broad. But I think of it as a "does what this person show me through their actions make me trust them" as a factor. One thing leads to the next. I will say I'm impressed in general with SFR in regards to how this RfA is going (even if you disagree). I'm not sure how I would react if I was in their shoes. I hope they're doing okay behind the screen, too.

Also, thank you for engaging with me. I haven't really changed my mind but I often have a hard time feeling comfortable disagreeing with people. While I definitely had to take some long breaks in this discussion, it's nice to know that it is possible for me to still defend a position I hold strongly without going over the top in a "nothing can ever change my mind" way. Clovermoss (talk) 18:48, 18 September 2022 (UTC)

After watching ScottishFinnishRadish's RfA unfold over the past week, I have two things to say:

1. This had better become the go-to link for every instance of someone asking why we have so few RfAs these days. I don't know when 18 months, nearly 30,000 edits, and 2 GAs became insufficient experience for an RfA candidate, but I've been around long enough to remember a time when that would be referred to as "overqualified" by the peanut gallery at large. Also, why are we using namespace distribution as a measure of an editor's value? There are so many different variables at play that a simple pie chart could never account for.
2. A returning editor with malicious intent who wants to become an administrator will likely try to avoid giving off any indication that they might have edited before. If they're especially smart, their contributions will be indistinguishable from those of a genuine newcomer. It is literally impossible to create a completely foolproof system to prevent a single "bad apple" from ever getting their hands on the tools. Any benefit we'd derive from doing so would be vastly outweighed by the fact that we would promote virtually nobody.

I have nothing more to add. I do respect a lot of the people in the oppose column, but I fundamentally disagree with the standards to which we hold would-be administrators. Kurtis ^(talk) 00:13, 20 September 2022 (UTC)

I agree entirely with point 1. For every candidate I've nominated at RFA, I've had approximately two with similar qualifications decline a nomination specifically because they were worried about the climate, and about unreasonable standards. I suspect that ratio is going to get worse. Note that at the time of writing all five of my earlier nominees were successful, and SFR's RFA has just entered a crat-chat. Vanamonde ( Talk) 07:18, 20 September 2022 (UTC)

"So few RFAs"

Kurtis wrote that " This had better become the go-to link for every instance of someone asking why we have so few RfAs these days."

1. You can scroll to the top of this page to see the results of the most recent (last 5) RFAs. On the last five RFAs, we have:

   two very heavy content creators (highly involved in the GA/FA and peer review processes) passing easily with essentially no opposes (Z1720 194–0 and Femke 200–2);

   a copyright patroller who also has an FA passing easily with no opposes (DanCherek 281–0);

   an editor who created a bot passing with some opposition (DatGuy 158–16);

   and one self-nom that seemed not ready for RFA, no real content creation (Shushugah withdrawn at 59–46).

2. You can see Wikipedia:Requests for adminship by year#2022 for more data, with similar trends.

   the next candidate was not a top content creator, and ended up in a 'crat chat.

   and scrolling on down the list for FA names I recognize, I see the prolific Modussiccandi at 196–0, among others who passed easily, too many to name.

I'm looking at three out of the last five passed with flying colors, and they all created top content. You ask "I don't know when 18 months, nearly 30,000 edits, and 2 GAs became insufficient experience for an RfA candidate"; I don't know when 1,900 total words of GAs became sufficient to consider one had engaged the content side of Wikipedia, but whenever I saw an RFA candidate nominated on content creation who did not actually fit that description, I opposed. If they presented with other strengths, not misleading, I was unlikely to engage. SFR was nominated as having a "solid portfolio of content work"; they do not.

For all the people I know who don't want to seek adminship, the factors that often come up are things like: distraction from content work; view admins as only here to block other editors; don't feel adminship is necessarily prestigious (the high rate of abuse and power-seeking); and don't want to give a full week of their life and endure 30 questions in an RFA.

I am not saying this was a desirable RFA, or that no mistakes were made, but they were made by all: in their incomplete presentation by both the candidate and the nominators; and by me as well in not more carefully composing my oppose for clarity, such that a rallying cry/strawman of socking took hold. Not being forthcoming about an early editing history that led to communication with ArbCom, and an editing trajectory suggestive of admin-seeking rather than content building are not the same as socking. But you seek to blame a poorly presented RFA that became contentious for all of RFA's problems, when there are many more, and more complex, factors that account for the declining number of admins, along with the equally declining participation in content review processes. Judging by the candidates who are passing easily, I see a trend. Judging by the desysops I'm aware of, I hope that RFA continues to do just what it is supposed to be doing when it screens a candidate we don't yet know well enough or have questions about. Well-prepared candidates with engagement of top content pass easily; that many of them don't want to be admins has often to do with the perception of adminship as not worth a week of their lives to answer an endless stream of questions. A well-prepared candidate will not face an RFA like this one, but they will still have to endure the 30 silly questions. SandyGeorgia ( Talk) 08:49, 20 September 2022 (UTC)

So then, what would you define as being a "solid portfolio of content work"? Do editors need to have an array of GAs and/or FAs before fitting the bill? What if someone finds articles, expands upon them by adding new content and references, but never bothers to nominate anything they contribute for GA-status, let alone FA? Would it be dishonest for them to describe themselves as "content contributors"? Kurtis ^(talk) 09:19, 20 September 2022 (UTC)

Kurtis "Finds articles, expands upon them by adding new content and references, but never bothers to nominate anything they contribute for GA-status, let alone FA" is a fairly good description of my content work around the time of WP:Requests for adminship/ONUnicorn. I did have several DYKs, but no GAs or FAs. ~ ONUnicorn^{( Talk| Contribs)} problem solving 17:17, 21 September 2022 (UTC)

That's a hypothetical, and one of those things that "you know it when you see it"; there are plenty of ways to demonstrate content engagement without GAs or FAs (and we discussed that already in the lengthy thread above-- and I can think of many others on top of those you list). Focusing on this case, the candidate demonstrated essentially no other significant content building of the type you describe. Adding together the two factors (misleading on content, and not forthcoming on past editing history) doesn't help towards the main factor we are evaluating at RFA: trust. That the examples of poor temperament came in late is another factor; had they come in earlier, I'd wager there would have been even more opposition. SandyGeorgia ( Talk) 09:29, 20 September 2022 (UTC)

In the case of people doing expansion and adding references, the concern is that there's no understanding of the process of taking an article from creation to peer review, and that if you haven't done that, you don't understand why it's so hard and what those editors face. It's not that the work isn't valuable, it absolutely is. But an admin needs to understand what actual content creators face, and the only way to prove that is by the peer review process. Valereee ( talk) 09:36, 20 September 2022 (UTC)

And one can be a very good reviewer (at PR, GAN, FAC or FAR) even if they feel shaky about their own ability to write! Or they can work at places like WP:3O; there are so many ways to engage content without producing a little star. SandyGeorgia ( Talk) 09:40, 20 September 2022 (UTC)

I will say Sandy, though, that there are very wide gaps between some GA reviews and others. To give an example of the difference, Here is my review of a Spanish vote of no-confidence article which I considered a detailed review, while This is a review of a similar article by the same nominee which I consider grossly incomplete. If a nominee had created 30 or 40 of the former I wouldn't see them as inexperienced with content, but the latter make a candidate look seriously unsuited for adminship at that stage. Of course, FAC and FAR are much much better. — Ixtal ^{( T / C )} ⁂ Non nobis solum. 10:45, 20 September 2022 (UTC)

As an aside, I do think Dispute Resolution Noticeboard moderation is something more future RFA nominees/candidates should do. The implicit goal of constructive conflict resolution without the use of tools, the strong understanding of PAGs needed to moderate the harder cases, and the conduct and character displayed in the successful moderation of cases is in my eyes one of the strongest indicators of admin suitability. Sadly there's only 2 or 3 active moderators there as well as issues with our internal guidance that encourages new editors to go to ANI before DRN in content disputes (as can be seen by the many wrongly filed ANI cases we see each week) resulting in few cases being requested. — Ixtal ^{( T / C )} ⁂ Non nobis solum. 10:53, 20 September 2022 (UTC)

Probably over 70% of my edits are in the article namespace, and I would define myself as a content creator because I am adding content to the articles other editors can not or can with some difficulties, but I do not have any GA/FAs and I am not planning to have any (since English is not my mothertongue, I believe that I better leave this last step to native speakers). For the record, as an academic, I am dealing with peer review every day in my life. Ymblanter ( talk) 10:41, 20 September 2022 (UTC)

And you gave a very plausible explanation for why you, primarily a creator of content, had no GAs or FAs, and no one opposed your candidacy for lack of content creation. :) Valereee ( talk) 12:39, 20 September 2022 (UTC)

Oh, I absolutely recognize how difficult the process is—that's one of the main reasons as to why I've never actually gone through the trouble of bringing anything I've written to either GAN or FAC. The notion of creating a nomination page for something I felt that I worked hard at, only for someone to ask me why I think that what I've written is even worthy of consideration for GA or FA-status, is discouraging.

Speaking only for myself, I am extremely perfectionistic in how I approach virtually every aspect of my life, to the point where I hold myself back from doing things because I fear the consequences of doing them incorrectly. Editing Wikipedia is no exception, and it's the main reason as to why I've never been a particularly prolific content contributor. When I attempt to write an article, it's not just a matter of making it legible and having a few external links here and there—the prose needs to be grammatically immaculate, there needs to be no close paraphrasing whatsoever (even unintentionally so), the references need to be thoroughly detailed, external links need to be from reliable outlets, the sources need to support the exact thing that they are referencing, the formatting needs to be clear, etc. In my view, you can't just submit content first, and then add references later, or revise the text so that it fits the sources; whatever you put out onto Wikipedia needs to be well-referenced at the time it is submitted. When I attempt to write an article, the very first thing I do is gather as many references as I possibly can from a variety of well-accredited sources, and then I attempt to summarize whatever they say in my own words. Writing articles is hard work, and being a content creator requires a very serious commitment. That involves maintaining a sustained focus on writing and copyediting a single article over periods that will often span weeks, months, or even years.

This is personal, but I feel it is relevant; I recently (i.e. the past few years) realized that I have ADHD-PI. One of the ways it manifests in my case is that I will often begin projects with a high degree of enthusiasm, invest a tremendous amount of effort into them from the outset, and then after a couple weeks or so I lose interest and move on to something else. This has made it incredibly difficult for me to live my life; I lack the type of sustained focus that is essential to maintaining a long-term, serious commitment to anything. The good news is that I'm seeking treatment for this, and said treatment seems to be more within my grasp now than it has ever been before, but whether it transforms my life or not remains to be seen.

This is why the “lack of content creation” argument hits somewhat close to home for me; it feels like my own contributions are being denigrated as falling below an ideal standard. The insinuation is that my value as an editor, regardless of what other good I might do for Wikipedia, is intrinsically tied to the number of high-quality articles I've written—which, in my case, is relatively few. I am the principle writer of forced evictions in Baku and capital punishment in Yemen, neither of which are "bad" articles per se, but I don't feel that either one of them is worthy of the green plus sign in their current state. Other than that, I am primarily a WikiGnome and I also like to engage in Wikipedia-space discussions, particularly on WP:ITNC and RfA. I put a lot of thought and consideration into everything I type, and I like to believe that everything I do is for the benefit of Wikipedia. I don’t want to subject myself to a week’s worth of having my entire contribution history reduced to “just over a quarter of all his edits have been to article space”. And if I don't want to deal with that, a lot of other editors probably don't want to go through it either. Kurtis ^(talk) 11:54, 20 September 2022 (UTC)

Again, it's not that the work isn't valued and important. It's that in order to convince people you understood content creation well enough to be trusted as an admin you'd need to find some other way to show what a high mainspace count + a few GAs provide basic proof of. People use those as a sort of shorthand, but there are other ways to show it. Some editors have hard-and-fast rules about supporting anyone without peer reviewed content, but most would listen to an explanation. Valereee ( talk) 12:31, 20 September 2022 (UTC)

So, just to be clear, my comment was not specifically intended to convince people of my own understanding of content creation and its tribulations. What I'm trying to do is provide my perspective as to why so many editors are hesitant to apply for RfA, speaking as someone who is not an administrator and has no inclination to submit an RfA at this time. I went through an RfA of my own just over a decade ago, which I'm loathe to link to because I withdrew it a few hours prior to its scheduled closure with a political statement that I find slightly embarrassing in retrospect; I was lucky that what I went through had turned out to be an overall positive, constructive experience relative to what others have had to endure. There is no guarantee that I would be so fortunate were I to run again. For me, the benefit of having sysop permissions is hardly worth the high possibility of having my entire history as an editor so heavily scrutinized. Kurtis ^(talk) 22:33, 20 September 2022 (UTC)

Kurtis, it sounds like you'd make a good admin. Self-awareness if worth a hundred GAs. SandyGeorgia ( Talk) 22:14, 20 September 2022 (UTC)

Thanks, but I'm a bit too sporadic in my editing for me to apply at the moment. I'll give it some more thought if I ever do get back into active editing (which I'd like to, I just... have to take care of the aforementioned ADHD-PI thing first 😏). Kurtis ^(talk) 22:38, 20 September 2022 (UTC)

Best of luck to you ... at least the science behind treatment has improved considerably over the decades since I wrote Tourette syndrome (often comorbid), and I'm confident you'll do fine :) And I meant what I said about the value of self-awareness relative to content creation. SandyGeorgia ( Talk) 22:40, 20 September 2022 (UTC)

Thank you. There aren't too many people in my life who understand the things that I go through, so it actually means a lot to me that someone who is familiar with neurodiversity wishes me well. Kurtis ^(talk) 07:39, 23 September 2022 (UTC)

• Not looking for an extended back-and-forth here, so further engagement from me will be minimal, but two points need to be raised in response.
  1) In 2022 we've had a net loss of 32 admins at the time of writing (44 desysoppings, 3 restorations, 9 successful RFAs). When people complain of the number of RFAs, that number is typically what they worry about. It doesn't matter how well-qualified the successful candidates are if there's too few of them.
  2) By your own admission, Sandy, you look for admin candidates primarily (exclusively?) among those engaged with FA-related processes. That's less than a hundred editors, of the thousands of active ones we have. By definition, that's a highly non-representative sample, and isn't going to tell you about why eligible candidates aren't running; particularly so because if content writers want to focus on content, they're not going to burn through the backlog even if they are persuaded to run. Vanamonde ( Talk) 09:28, 20 September 2022 (UTC)

  Aren't most of those in your number 1 due to inactivity? Not a factor in the control of RFA. On your number 2, I look for candidates to nominate myself, which is a separate matter from what I support or oppose or with whom I converse about adminship or the number of times I'm asked about candidates. But to a degree, yes, there is some skew towards content people in my sample. Having admins who are primarily content creators has potential to do more than "burn through the backlog"; it might lower it by helping not create some of the kinds of problems that fester and boil over to begin with. SandyGeorgia ( Talk) 09:35, 20 September 2022 (UTC)

  Inactive admins are still admins we don't have, and much as I like having candidates from within the FA contributors, they simply are not representative of the larger editor body, nor do admins with that background frequently end up making large dents in the backlog. We can't rely on editors with one sort of background providing all our admins and admin work; it's plain impossible. Vanamonde ( Talk) 10:43, 20 September 2022 (UTC)

  Yeah, it's a bit of a conundrum. We want admins to be content creators, but not primarily content creators. :D Before I ran RfA I routinely averaged 60%+ mainspace edits. As soon as I became an admin the numbers dropped into the 40s. I consider myself primarily a content creator, but this year my cumulative average dropped below 50% for the first time, which I am a little sad about. But here I am on a WT page not improving that. :D Valereee ( talk) 09:56, 20 September 2022 (UTC)

  Although there has been much discussion about the rate of attrition of administrators, it seems to be happening very slowly. Even if we lost 50 administrators a year, and assuming we need 500 regular admins to keep things moving along, it would still take ten years for us to be in trouble, and that assumes nobody ever stood for RfA again. So I've never been convinced that things have been that bad. Ritchie333 ^{(talk) (cont)} 15:45, 20 September 2022 (UTC)

  The number of administrators needed to keep the project going is effectively impossible to determine. We don't and likely can't know. As the project ages, and we continue to lose administrators, we will sometimes lose key administrators (defined in this case as highly active administrators or those in key roles). We'll note the loss of such admins, but the project itself will continue. The number of active administrators is already below 500, and slowly falling. Over the last 10 complete years, we've lost a net average of 82 admins (active and otherwise) per year ( WP:DBM). Thus, in theory, we have 12.5 years left before the last admin packs up and moves on. In practice, various functions of the project will begin to decay, become unsupportable, or otherwise fall apart. We're seeing this now on Commons. Commons is effectively a failed project at this point. So, 12.5 years is a highly optimistic view of when the project will just fail. Reality is likely considerably shorter. I raised the spectre of how we handle a post-admin era almost 6 years ago ( thread). Nothing's changed since I noted that. 6 years from now, we could be looking at a failed project if nothing is done. Is the sky falling? No. Is the atmosphere becoming increasingly difficult to breathe? You bet. -- Hammersoft ( talk) 16:35, 20 September 2022 (UTC)

  The shorter-term problem is when the pipeline slows to a trickle, we aren't getting new cohorts of admins which can lead to fatigue in existing admins and less diversity in discussions on administrative matters. isaacl ( talk) 16:42, 20 September 2022 (UTC)

• Trying to recruit admins from amongst people who mainly write high quality content is not a viable or sensible strategy. People who mainly write or review high quality content probably want to spend their time writing or reviewing high quality content, and there's very little intersection between that and admin work. It's not remotely surprising if they think adminship is a distraction from content work. Z1720, for example, said that his main focus as an admin would be DYK, and indeed he doesn't seem to have done any other admin work since becoming an admin three weeks ago. He's been writing and reviewing content instead. Even if we could draft our FA writers to do typical admin work, and make them delete corporate spam, block vandals and try to stop edit wars - why would we? I suspect the FA work they would have done otherwise would have been a good deal more valuable.
  The people who would be most interested in adminship, and the people who would make the most difference to admin backlogs, are the people who do admin-type work as a non-admin. An experienced NP patroller, for example, would find the admin toolkit directly useful in their everyday work. However someone like that would probably get a bunch of opposes for lack of content creation, unless they happen to write FAs in their spare time. Even if they did write some content to prove they can that's evidently not considered enough. I suspect that eventually the number of admins will dwindle to the point where the community will have to fix this, but for now it's one of the biggest faults in RfA. Hut 8.5 19:31, 20 September 2022 (UTC)

  By your own admission, those people who don't produce high quality content would be a lot more valuable if they did. FA work is not like dealing with backlogs because you are capped at four or five FAs per year. So more people equals more FAs. I have written 98 but that is only because I have been around a very long time. Like most FA writers I spend most of the time on NPP, DYK, AfD, backlogs, bot maintenance and administrative tasks. The admin toolkit would be far more useful to me than your experienced NP patroller. Hawkeye7 (discuss) 20:29, 20 September 2022 (UTC)

  Hut 8.5 Hawkeye addressed part of your post but re Trying to recruit admins from amongst people who mainly write high quality content is not a viable or sensible strategy if anyone in this discussion had advocated for that, I missed it. I have stated (somewhere) that I only nominate editors I know very VERY well, and those are the FA people. I (and others) have repeatedly stated that there are many ways to evaluate whether an editor has sufficiently engaged content creation sans the little stars. My intent when I started this section was to illustrate that, contrary to the hand-wringing about the state of RFA (which is similar to that state of all of Wikipedia), most content contributors sail through RFA. The numbers show us where the !voters stand, and they seem to agree that content engagement is key. SandyGeorgia ( Talk) 22:23, 20 September 2022 (UTC)

  contrary to the hand-wringing about the state of RFA... most content contributors sail through RFA. OK, how do we use that observation to increase the number of successful RfAs? There are only two options: recruit new admins from amongst people you consider "content contributors", or try to force people interested in adminship to become "content contributors". As I've explained, the former is not a practical option. The latter isn't very realistic either. While some RfA candidates are willing to write some content to demonstrate they can, your oppose in the ScottishFinnishRadish RfA said that Content creation was never primary for this editor, suggesting you're expecting potential RfA candidates to change the entire focus of their editing. And no, it's not correct to say that the decline in the admin corps is reflected in Wikipedia generally. The number of active editors has been fairly stable for the last decade or so, as has the rate of edits. Hut 8.5 12:20, 21 September 2022 (UTC)

  The number of active editors has been stable, but participation in content review processes has declined (even to the point of WP:PR being practically defunct). And that's the answer to the first part of your response; you don't have to "force people" to engage content. By seeing the trends, they can be encouraged (without force) to help out in more ways than 800-word GAs. You can engage content without GAs or FAs. Re the wording in my oppose, I've already acknowledged several times it was not carefully composed, as I was editing from my phone while working on a fundraiser. It's fairly clear that if I had written it more coherently, the RFA would have unfolded differently. SandyGeorgia ( Talk) 12:47, 21 September 2022 (UTC)

  The number of active editors has been stable, but participation in content review processes has declined - that suggests that people don't want to participate in those processes. Trying to figure out why that is and fixing it would be more effective than applying unrealistic standards at RfA. Hut 8.5 17:48, 21 September 2022 (UTC)

  Participation in writing content (you know, actual sentences in articles) has greatly declined, as nowadays very large numbers of editors never do anything except deletion & other talk discussions, templates, navboxes, geolocations, infoboxes, gnomish reference fixing, categories, quality ratings, short descriptions, images and so on. Johnbod ( talk) 18:00, 21 September 2022 (UTC)

  Hut 8.5 both content review and RFA may be suffering from something that is related. Editors aren't engaging top content because, unlike the earlier days where there was more reward/respect attached to building content, now there is more focus on the governance end (that is, we have truly moved to where adminship is a big deal and content generation is not). Whack-a-mole is now valued over content. I gave my answer to that dilemma in the delusional parasitosis example. If engaging content were held in the same regard as it was a decade ago, then more people fix articles, less time spent in unproductive talk page fights, more people pass RFA, an upward cycle results for both content and governance, we all win. I'm quite convinced I saved hundreds of editor-dispute hours by just rewriting delusional parasitosis-- hours that could have been misspent dealing with disruption, going to ANI, going to noticeboards, getting POV pushers blocked, etc. Editors who just do it don't seem to have problems passing RFA. The solutions to the two problems are related: the lowering of standards at FAC, GA and RFA lead to less people wanting to be part of the process. SandyGeorgia ( Talk) 08:06, 22 September 2022 (UTC)

• I have 2 FAs and 17 GAs. I would still fail at RfA. There's more to it than writing articles. Trainsandotherthings ( talk) 20:41, 20 September 2022 (UTC)

  Content is one of several criteria, for example if you ran before July next year you would get some opposes from those who think that admins should have two years experience. But your content contributions would get you a lot of respect at RFA. Ϣere SpielChequers 16:04, 23 September 2022 (UTC)

  I ran with less than 2 years experience and received no opposes for lack of experience and have nominated several candidates with less than 2 years experience who received no opposition for lack of experience (and some of whom passed unanimously). So I don't think we should take at all as a given that someone running with less than 2 years experience will attract opposes for that element. Best, Barkeep49 ( talk) 16:27, 23 September 2022 (UTC)

  I agree with Barkeep; I supported Wikipedia:Requests for adminship/Hog Farm (with perhaps less tenure than Barkeep49) because I had worked closely enough with him on multiple articles to know his character, and he was a bonafide content creator, well engaged in the trenches and I saw how he responded to pressure and criticism and he had been named by his peers as a MilHist Coord. The idea that tenure or content creation alone are deciding factors is a false meme that took hold in the last RFA, which was about five factors: a) presentation as having a content "portfolio" that was not accurate, combined with b) somewhat short (but not critically) short tenure, rather, not enough of that tenure in content building to be able to observe); and c) a curious early history that d) was not disclosed in the nomination, and some e) temperament concerns. It was not a typical RFA; potential candidates should stop wringing hands about possibilities that aren't likely to be problems. Do what you do best, engage content enough for us to know your work, and present an honest appraisal of your activity, strengths and weaknesses, and you're likely to pass. SandyGeorgia ( Talk) 17:05, 23 September 2022 (UTC)

  Please stop talking about this RFA, and specifically stop posting analyses of the candidate and his supposed faults. Levivich ( talk) 17:44, 23 September 2022 (UTC)

  ( edit conflict) For what it's worth, I'd support you if you ran for RfA, Trainsandotherthings. Clovermoss (talk) 17:08, 23 September 2022 (UTC)

  Hear, hear. Curbon7 ( talk) 17:54, 23 September 2022 (UTC)

  Trains, the only "negative" I know about you is that when I go to your user page I get that image popping up from the side that always creeps me out; not sure I can turn that into a judgment oppose though :) SandyGeorgia ( Talk) 17:59, 23 September 2022 (UTC)

  I'm quite surprised (and flattered) to see people actually think I could be an admin. I still think I need more experience and people would have concerns about my temperament, but this has inspired me to remove the "this user is nowhere near being an admin" userbox from my userpage. Trainsandotherthings ( talk) 14:39, 24 September 2022 (UTC)

  This may be atypical advice but don't feel pressured to run just because there's a lack of candidates. Do it when you're ready and I'll still be here :) I've felt that I needed more experience to feel confident in certain things for awhile and I know where my own weaknesses are so I understand where that perspective is coming from. Sometimes we can be our own worst critics. I feel like I have no idea what I'm doing here half the time but maybe that's because my self-confidence is kind of like a seesaw. Clovermoss (talk) 14:47, 24 September 2022 (UTC)

• I feel that my skills and lifestyle make me more valuable to the project as a gnome than a content creator. I think it's equally or more valuable to the project at this point in time to make our existing articles exemplary than to make a bunch of new ones just so I can tick off a box of "created new article".

Also, I'm in the U.S. but tend to keep late hours, and I'd love to have more mop-equipped U.S. people online at midnight on the Pacific coast to block spammers and trolls whose contextual inappropriateness isn't necessarily apparent to people in the part of the world who are waking up at that time of day.

But what I don't think has been addressed sufficiently in this community discussion is the reason I don't want to do an RfA: I am deeply anxious about personal security and privacy, and I have zero trust in the WMF to provide supportive resources in this regard. I don't feel physically or professionally safe becoming a target of children who are capable of getting so wound up about being blocked for edit-warring the assertion that Chicago pizza is "a casserole" that they'll lose their minds and try to get me fired and/or arrested in real life for the lulz. I'm probably just some fat bitch who needs to die anyway.

Tell me how I'm being unreasonable. - Julietdeltalima (talk) 05:15, 21 September 2022 (UTC)

You aren't being unreasonable. Valereee ( talk) 05:19, 21 September 2022 (UTC)

Yep ... one of several reasons I've never wanted to be an admin; Wikipedia is dangerous enough as it is. SandyGeorgia ( Talk) 12:49, 21 September 2022 (UTC)

I consider improving existing articles to be content creation, and I can't imagine why anyone wouldn't. Most of my GAs came from existing articles I improved. Trainsandotherthings ( talk) 00:53, 22 September 2022 (UTC)

• Hammersoft above articulates my feelings almost exactly. If the attrition continues, we shouldn't be surprised when we're eventually unable to handle the flood of socks, spam, vandalism, and drive-by POV-pushing. I wouldn't be surprised if many editors were to apply unrealistic standards at RFA even under those circumstances, but for Wikipedia's sake I hope most of us don't. Vanamonde ( Talk) 16:28, 21 September 2022 (UTC)

  I think we're about to see a real impact about the diminishing capacity of the admin corp when this year's CUOS candidates are presented to the community. Best, Barkeep49 ( talk) 16:50, 21 September 2022 (UTC)

  On a strangely related note, my experience applying for CU in 2020 burned me out of Wikipedia much more than any other experience including my RfA. The combination of a high-effort application, followed by a lengthy public consultation with virtually no feedback, then a polite, vague "no" several weeks later after a secret discussion. For all the flaws of recent RfA's, I'd have taken that in preference! ~ mazca ^talk 23:20, 21 September 2022 (UTC)

  A perspective that we often forget as desktop editors is that of mobile access. While the actions of socks, spammers, vandals, and POV-pushers are possible with mobile phones, many admin actions are not as accessible on the phone, at least to my knowledge. As the internet population continues shifting increasingly towards phones and those of us that have grown up or primarily use desktop become ever more the minority, our ability to recruit desktop editors will decrease (in my opinion). How that affects the rate of active admins I do not know. But I do expect the context of Wikipedia editing to change. — Ixtal ^{( T / C )} ⁂ Non nobis solum. 16:54, 21 September 2022 (UTC)

  Ixtal, I am an active madministrator administrator who does the vast majority of my editing using the desktop site on Android smartphones. When I became an administrator five years ago, I wondered if I could carry out administrative tasks on my phone. The answer is thst the administrators toolkit works just fine on my phone. Cullen328 ( talk) 05:10, 23 September 2022 (UTC)

  "madministrator" — finally he admits it! Firefangledfeathers ( talk / contribs) 05:11, 23 September 2022 (UTC)

  Firefangledfeathers, if I was ever going to create a sockpuppet account, that would be a great name. But I am not going to. Cullen328 ( talk) 06:34, 23 September 2022 (UTC)

  Cullen328, I am familiar with your smartphone editing as I read your essay when I first joined Wikipedia. About 30% of my own edits are from mobile, split between mobile desktop and mobile browser. The issue with the demographic I am talking about is that they tend to not like using browsers (e.g. people my age poking fun at my use of reddit on the browser rather than the app) or if they do, many new mobile editors will not know that there is a big difference between the tools accessible for mobile and desktop editors. If they do, they might not know how to turn the site into desktop mode. Keep in mind as well that some of these mobile editors have never used a computer before so they may be unwilling to perform the workarounds you and I find alright. I remember at a recent discussion to help WMF improve its mobile editing editors mentioned this made recruiting these users to WikiProjects, for example, more difficult as some templates/notices in talk pages do not display and so they must be reached out to individually to guide them to the meta aspects of the wiki. Thus I still think the recruitment problem I mentioned above exists. — Ixtal ^{( T / C )} ⁂ Non nobis solum. 06:12, 23 September 2022 (UTC)

  Ixtal, without straying too far afield, I belive that, two plus decades into this project, all methods of editing Wikipedia from mobile devices should be fully fuctional without exception, or shut down promptly. I have spent a decade or more proving almost every day that the desktop site works just fine on Android devices, for reading, editing and administrative work. Cullen328 ( talk) 06:34, 23 September 2022 (UTC)

  Cullen328 I am speaking not about the desktop site (which I use as well) but the mobile site and the app. — Ixtal ^{( T / C )} ⁂ Non nobis solum. 06:47, 23 September 2022 (UTC)

  Ixtal, I understand what you are saying and I use a lot of apps unrelated to Wikipedia almost every day. Those are apps that are fully functional. Wikipedia apps are not fully functional. I am not a techie, and readily admit that my deep skills are limited. I am a user - a power user on Wikipedia to use the pejorative. I started using personal computing devices as tools to get things done back in 1979, and I do not care at all whether I am working in a browser or an app. All I care about is whether or not the software gets the job done in an efficient, fully functional fashion. Getting hung up on the word "app" seems silly to me. Apps that work well are wonderful. I remove apps that suck from my smartphone as soon as I have evaluated them. Cullen328 ( talk) 07:10, 23 September 2022 (UTC)

  If anyone is interested in what using the mobile app is like, I've been experimenting with it for a few months now. User:Clovermoss/Mobile editing gives a decent overview. Clovermoss (talk) 07:23, 23 September 2022 (UTC)

With fewer active admins, the demands on the remaining admins increase, which may enforce the view some hold that adminship is a big deal. I don't see standards going down anytime soon.

I just wish we could apply some of the accumulated wisdom of countless other organisations: when (!)voting on content, it's public, when (!)voting on people, it's private. I believe that the oppose rationales + knowing who would vote against / in support are a way larger deterrent to run for RfA than the opposes themselves. Femke ( talk) 18:47, 21 September 2022 (UTC)

That's a fair point. I used to be staunchly in favor of the !vote system over the vote system, as I worried the latter may be too easily game-able by socks and editors with past beef. At this point, though, I'd support an anonymous vote, provided the activity threshold for eligibility was high enough. Vanamonde ( Talk) 19:03, 21 September 2022 (UTC)

As with admins with probationary terms, admin elections was also closed with a result of no consensus. If reformulated, it could pass. 0x Deadbeef 03:29, 22 September 2022 (UTC)

Maybe that's my next RfC project car :) (although I better finish the current one first). Enterprisey ( talk!) 05:40, 22 September 2022 (UTC)

Do we have any empirical evidence that the shrinking number of admins is having negative effect? Are administrative backlogs increasing, for example? Because there's an alternative narrative, which is that the steady stream of inactivity desysops is just us getting rid of the substantial dead weight we have from the project's early days (there were over 1000 successful RfAs before 2010), and the slowing rate of new admins simply reflects the slowing rate of new editors. I've honestly never seen any data that would say whether one or the other narrative is true. –  Joe ( talk) 07:30, 22 September 2022 (UTC)

I did start collecting data that would give a measurement on this at one stage, but dropped it because of the ethics process that was potentially required due to working with information about living people. But the methodology I was trialing was to look at response times on secondary noticeboards (initially RFPP) to see how they change over time. It might be worth revisiting. - Bilby ( talk) 07:43, 22 September 2022 (UTC)

As I said above, the number of active editors and the edit rate has not declined and has been stable for the last decade or so. The number of active admins has declined as well as the total number of admins, so this can't be just down to inactive admins being desysopped (the edit history of Wikipedia:List of administrators/Active has a record of this, although it uses a rather generous definition of "active"). The amount of admin work doesn't seem to have declined either, the number of AfDs has again been stable for the last decade or so. Hut 8.5 07:59, 22 September 2022 (UTC)

That doesn't really answer the question, though. It could be that we had too many admins before, or that the efficiency of the average admin has increased (with scripts etc.) so that having less admins-per-jobs is not a problem. It's interesting that even in the heyday of 2007-2009, we had at best 50% more active editors than we do now, but were appointing something like 5000% more admins. –  Joe ( talk) 08:13, 22 September 2022 (UTC)

And many of these admins didn't do much admin work. What was different back then was that becoming an admin was a fairly natural step for any decent editor, not something unattainable for all but the very elite. I think this has had an effect on the social standing of admins, but I don't have data to back this up. — Kusma ( talk) 10:41, 22 September 2022 (UTC)

Although most admins don't do much admin work, we still have far more admins doing admin work than we can replace. Here is a table listing admins by number of logged actions in the past year. Counting admins by logged actions has the same pitfalls as sorting editors by edit count, but it's fair to say that someone who makes few logged actions probably isn't doing much admin work. There are 324 admins who averaged at least one logged action a week over that period. In the period 2017-21 (five years), we had 77 successful RfAs, and obviously not all those people will be highly active now. Consequently we're relying on people who've been admins for a long time (often a very long time) to do most of the admin work. As those people leave we will have to either learn to cope with having far fewer active admins than we do today or make radical changes to our process of promoting admins. Hut 8.5 12:23, 22 September 2022 (UTC)

As that table shows, a tiny handful of admins are performing nearly all the logged actions. Losing just one of those will have an immediate effect; the long tail, not so much. Espresso Addict ( talk) 00:39, 23 September 2022 (UTC)

• On the deletion front, ACPERM sharply reduced the number of articles in mainspace that need immediate deletion, and the semi-automated deletion of abandoned drafts by G13 has also reduced the deletion backlogs. With three or so highly active deletion-focused admins, I can't remember when I last saw more than a handful of speedy deletions awaiting processing, except for clearing out a sockmaster's contributions or someone blanking their userspace pages, neither of which is exactly urgent. Also, as far as I recall, back in the days when RfAs were common, even rollback was a bundled admin-only privilege. Espresso Addict ( talk) 00:31, 23 September 2022 (UTC)

  The issue isn't only that backlogs will increase over time, but that admins pay too high a cost to keep them low currently. I doubt there is a single admin at the perennially backlogged venues (SPI, DYK, probably CCI, PERM) who would not rather be spending less time doing admin work there. It leads to burnout and drops in activity, and over the long run jaded admins are almost more of a problem than inactive ones. Vanamonde ( Talk) 03:56, 23 September 2022 (UTC)

  SPI and CCI are essential; is DYK worth all the admin-hours it chews up ? SandyGeorgia ( Talk) 03:59, 23 September 2022 (UTC)

  The admins who work there are for the most part content specialists or at least DYK specialists. They're unlikely to move to other areas if DYK were scrapped; indeed many of them ran only because there was no other way to fix the DYK backlog. Those who work there feel it's worth it, else they wouldn't. Vanamonde ( Talk) 04:06, 23 September 2022 (UTC)

  I used to do a lot of work at DYK when I was first made an admin (indeed, it's why I ran), but the current set up is just too stressful. I don't think a DYK backlog generated by having made the system insanely stressful is a good reason to cry admin shortage. (For clarity, I am not saying there is no admin shortage; there are certainly areas where one exists but they are mainly rather specialised.) Espresso Addict ( talk) 04:19, 23 September 2022 (UTC)

  @ Espresso Addict: I'd be the first to admit there's room for improvement at DYK; indeed, that's implied in my post above. But my point is also that DYK is generating its own admins much more than it is drawing admins away from other work, and so its elimination won't exactly make life easier elsewhere. Vanamonde ( Talk) 04:32, 23 September 2022 (UTC)

  If DYK is suffering from its own design, maybe it doesn't need to be included as a perennially backlogged venue as the others that are urgent to the functioning of Wikipedia. PERM doesn't require urgency. CCI is demoralizing work (by sheer volume), but the needs there are not always urgent; some of it can be picked away at over time. The SPI backlog is quite impressive; it's an area where the growth over time has seemed impressive. One used to get pretty quick action at SPI; now looking at the list is frightful. But other than that, I recall days of yore where there were huge backlogs requiring admins everywhere one looked, but it strikes me that most other processes are better off than in years past. AFD, for example, is nowhere near as bad as it once was (I'm speaking more than a decade ago). Similar for CSDs. 3RR still gets quick attention. So ... if SPI is one of the biggest problems, that's a tough one, because CU is beyond adminship. Where else are there urgent problems? SandyGeorgia ( Talk) 04:48, 23 September 2022 (UTC)

  There's a lot for non-CU admins to do at SPI too. RFD, unblock requests, and UAA always seem to be backlogged. Pages tagged for copyvio have always been in there when I've looked recently. Closure requests is always backlogged, though non-admins can help out there occasionally. I haven't done historical analysis of whether these backlogs were larger in years past; I'd rather spend my time on the actual backlog. But if we have an abundance of admins, it isn't evident at all. Vanamonde ( Talk) 04:59, 23 September 2022 (UTC)

  @ Vanamonde93: Where are you seeing copyvio-tagged pages? The speedy category is usually sparse. Espresso Addict ( talk) 05:07, 23 September 2022 (UTC)

  Well, it was in there the last ~5 times that I've looked, and it wasn't a small number either. Vanamonde ( Talk) 13:51, 23 September 2022 (UTC)

  [To SandyGeorgia] And a lot of the work at CCI can be done by anyone. The deletion/revision deletion is the easy bit. Espresso Addict ( talk) 05:02, 23 September 2022 (UTC)

  Easy ... for the admin to fulfill once requested? When helping out at the latest huge CCI, getting the request for revdel right was the hardest part for me! I kept fearing I was creating more work for the admins who had to fulfill the revdel request in case I got the versions wrong in the template. VM93, you mentioned WP:RFD, and it looks to me similar to WP:AFD; perhaps my memory fails, but years ago, one would look at those pages and find entries going back months. Now they're both at weeks, so it seems like something has changed for the better. Anyway, maybe we should be asking how all of these processes can get more help from non-admins; the copyvio revdel request script as an example helped non-admins to participate more effectively in copyvio work, and I was able to make a dent in the latest CCI. Where else can non-admins help? Why not hold a discussion focused on what kind of help is needed where, and see if there is more work non-admins can be doing? Many like me who have no desire ever to be an admin may still be willing to help if we are pointed at where we are able to help. SandyGeorgia ( Talk) 05:40, 23 September 2022 (UTC)

  And a discussion on where semi-active or intermittently active admins such as myself might meaningfully help out might also be fruitful. It's easy both to look at unbacklogged places and get the impression there's next to nothing to be done, and to look at highly specialist places like CCI and not bother to try and work out how to get involved. Something in time to send round to all the soon-to-be-forcibly-retired admins (under the new activity rules) might be particularly handy. Espresso Addict ( talk) 06:03, 23 September 2022 (UTC)

  Maybe Barkeep49 knows someone who would spearhead such an effort. SandyGeorgia ( Talk) 14:00, 23 September 2022 (UTC)

  Category:Administrative_backlog feels like a natural starting point given that different admin will have different levels of comfort/competence in various areas (i.e. there's no one size fits all answer). Best, Barkeep49 ( talk) 15:04, 23 September 2022 (UTC)

  Barkeep49 I was referring to spearheading an effort of writing a page to explain to non-admins how and where they might help out with the admin backlog ... CCI is one example that shows that careful guidance is needed, as I've seen editors new to CCI work come in and mark as clear articles that actually had copyvio ... are there other places where experienced editors might be helping out ? SandyGeorgia ( Talk) 15:11, 23 September 2022 (UTC)

  Oh I see. Well that does seem useful. Not sure who might get excited by it. Best, Barkeep49 ( talk) 15:13, 23 September 2022 (UTC)

  Well, looking at Category:Administrative backlog:

  • Call attention to Category:AfD debates relisted 3 or more times; anyone can help out there by entering a declaration. Education/advertising (Signpost) might help, since some "regulars" like me might not even know of that category (now I do).
  • Similar at Category:Wikipedia pages tagged for copyright problems; now I know to follow that cat and do what I can to speed those along (I know what has to be done in those cases, don't need education, and don't suggest it's a good area for advertising, as inexperienced editors might make it worse).
  • Two of those where non-admins can't help are requests for unblock, SPI (as large as I feared).
  • Can non-admins, with a well-written instruction page, start approaching editors at usernames with possible policy issues ? This is an example where a guide might help non-admins help out.

  So, what I'm suggesting is someone write and advertise a "how non-admins can help" along the lines laid out above ... SandyGeorgia ( Talk) 15:22, 23 September 2022 (UTC)

  I don't think the backlog at SPI is caused by there not being enough people to push the block and/or CU buttons. Block requests and CU endorsements are actioned fairly quickly, and when SPI reached zero open cases in January of this year, there were two fewer admins who are active at SPI than there are now (I'm referring to Tamzin and Firefly). In my view, the issue is that many cases are poorly presented or not really actionable. I wasn't around for the old days, so I'm not sure what has changed to increase the backlog. Maybe there were just fewer socks back then. Spicy ( talk) 05:15, 23 September 2022 (UTC)

  Re "fewer socks back then", that is most likely the case, at least from my (limited) experience. And I wonder how much of that increase is being driven by paid editing and LTAs. SandyGeorgia ( Talk) 05:44, 23 September 2022 (UTC)

  Perhaps it's a natural process. As time goes by, the number of blocked users increases, and so does the number of users who evade a block. I call this the theory of sock entropy. Spicy ( talk) 06:43, 23 September 2022 (UTC)

  I like the idea of "sock entropy" - brings to mind the idea of sockpuppets trying to understand a thermodynamics quiz - I'd also argue that with a larger user base now the absolute number of unconstructive users who will sock will increase even if the proportion remains unchanged. I also entirely agree that the quality of the filings matters immensely - I am far more likely to action a concisely presented case with diffs over a 1000 word rant. firefly ( t · c ) 12:21, 23 September 2022 (UTC)

  I would suggest a concise guide to submitting an SPI ... except ... BEANS. We had one recent FAR that has suffered under an LTA for over a decade, so I wrote to the new editor who has to deal with it to explain in gory detail how to submit an SPI with precise diffs just enough to document the case, but holding back as much info as possible so as not to give away tells. The "not giving away tells" makes it hard to write a guide; that is, I wouldn't write on Wikipedia all the tips I gave this editor backchannel to help them prepare better SPIs. SandyGeorgia ( Talk) 13:58, 23 September 2022 (UTC)

  @ SandyGeorgia we already have an excellent one - User:Blablubbs/How to file a good SPI :) firefly ( t · c ) 20:19, 23 September 2022 (UTC)

  Firefly yes but unless I missed it it doesn't cover my main advice re the FAR sock, which relates to how much evidence you need to be convincing, while not giving away tells. In olden days, I would give away everything I knew in the SPI, and the sock would just adjust. You only have to give enough to show that a CU is warranted ... where that balance is, isn't something I'd like to see in a guide. My other piece of advice was to just follow the LTA as long as they are staying out of trouble, because once that account is blocked, you'll just have to find them all over again at their next account. One LTA I followed loved to flaunt herself right under the CU's noses or in discussions about herself; as long as she wasn't being disruptive, I just chuckled. Once accusations that another editor was her surfaced, I filed the SPI. SandyGeorgia ( Talk) 22:11, 23 September 2022 (UTC)

  DYK is one of the few areas of the project that has actual deadlines. There've been suggestions for how to fix that -- run it like ITN, for instance, a ticker rather than a deadline -- but non-worker DYK regulars far outnumber workers and changes that are primarily aimed at making the work less burdensome to the workers tend not to gain consensus. Valereee ( talk) 15:06, 23 September 2022 (UTC)

  Do not forget that there are processes such as AIV and RFPP which require quick action (AIV would even be archived if action is not taken within 4 hours), and these tend to be backlogged when American admins are sleeping (smth between 8 and 12 UTC). Though I have not seen RFPP backlogs recently, consequently I am spending less time there and will probably not notice when it gets backlogged again. Ymblanter ( talk) 21:15, 23 September 2022 (UTC)

  I'd be interested to see whether that's still the case; with the speedy backlog, it's my experience that there's no longer any time period when it bulges (which certainly was not true a few years ago). Espresso Addict ( talk) 23:49, 23 September 2022 (UTC)

This is the wrong way to look at it though. Many content contributors are not interested in admin tasks, and many people who want to do admin-type tasks are not interested in content creation. Wikipedia needs both and there's no reason we need to limit admins to the people who are interested in both aspects, or force the latter type of editor to spend lots of time creating content before being considered for an admin role. ( t · c) buidhe 21:05, 27 September 2022 (UTC)

@ Buidhe That's a bit of a strawman too though, the idea that everything that an admin does can be lumped into "admin tasks". One of the biggest problems with the admin toolset is that it's so all encompassing. While you may be completely competent in a specific area and the admin tools would help your work in that area, giving the tools might allow you to cause a lot of damage in other areas. That's where trust comes in, trust that you won't go in "gung ho" and smash things up from a technical level, but also trust that you will empathise in the disputes you might have to adjudicate. One of the biggest types of damage an admin can do is drive off users - and worse than that, it's rarely visible or quantifiable. Worm^TT( talk) 07:18, 28 September 2022 (UTC)

Regular editors can definitely also drive off Wikipedia contributors as well, no need for an admin toolset. There are so many articles and entire topic areas I have stopped editing in because of other editors who are more interested in promoting their own POV than following Wikipedia policies. A lot of people do say that content creation might help admins in their dispute resolution role, but I've never seen evidence that this is the case. ( t · c) buidhe 09:15, 28 September 2022 (UTC)

Evidence is difficult, but as I explained in a different section, my take is that the vast majority of our disputes come out of content, and being part of that process is essential. What's more, to be part of the process, you need to have had some sort of peer review - that ensures you experience how much effort is needed to meet Wikipedia's standards. Seeing why people feel OWNERSHIP and why it's difficult to overcome. Building that empathy. It's not about "admins need to be content creators because that's the only thing you can do", it's "admins need to understand what this major part of the community deals with". Empathy is good. Worm^TT( talk) 09:37, 28 September 2022 (UTC)

Why many people oppose based on the fact that the candidate do not open to recall? Maybe voters set too high expectations? Thingofme ( talk) 23:01, 26 September 2022 (UTC)

I wouldn't say it's "many" (it's two), but there have always been voters who have criteria that other editors feel are unrealistic, obscure, or otherwise unfit for purpose. However, there is nothing wrong with having an opinion that is contrary to what most people think; it's just an opinion after all, and if (as I suspect is the case here) that opinion is <1% of the total opinions expressed, it is effectively irrelevant. Primefac ( talk) 09:06, 27 September 2022 (UTC)

As how many admins have been desysopped by the "open to recall" standard? Or some candidates' pressure means that they should open to recall? Thingofme ( talk) 09:52, 27 September 2022 (UTC)

Recall has worked in the past. See the list of times it was used. Some have resulted in resignation, others not. The problem is that Recall is entirely voluntary and has been seen as toothless since (I think) Durova's arbcom case where her recall goalposts shifted. Worm^TT( talk) 12:49, 27 September 2022 (UTC)

true for values of "the past" that are more than a decade ago; we've had two recalls in the most recent 10 years, if I'm reading that list correctly. Jclemens ( talk) 03:28, 29 September 2022 (UTC)

Just wanted to note that I appreciate your levelheadedness. I've learned that I clearly have a very different view of what a commitment to the recall process means to most people, which (as you say) is fine! I thought it was sort of funny that such a clearly minority opinion has warranted so much discussion at all. ThadeusOfNazereth(he/they) ^{Talk to Me!} 04:00, 28 September 2022 (UTC)

• As per Primefac, but it's good for creating some juicy drama. Do I hear the rustling of popcorn already? 🍿 Kudpung กุดผึ้ง ( talk) 11:08, 27 September 2022 (UTC)
  • Arbcom desysops have 1000x more drama than recall. Levivich ( talk) 12:51, 27 September 2022 (UTC)

It is now crystal clear to me that if anyone is ever to run for RfA who is not open to recall, the best thing for them to do when asked if they are open to recall is to lie. 🌈WaltCip-( talk) 02:29, 29 September 2022 (UTC)

All admins are open to recall by a panel selected annually by the community. Espresso Addict ( talk) 03:03, 29 September 2022 (UTC)

But they can use recall criteria to make them resign the tools, so it's not an option. Thingofme ( talk) 10:41, 29 September 2022 (UTC)

Sorry, they can change their criteria so not good at all. I think we should have community-based desysoping procedures, but it's a perennial proposal. Thingofme ( talk) 10:43, 29 September 2022 (UTC)

I don't think we should ask people with zero experience as an admin about their hypothetical recall preferences. If we want a working voluntary recall system, it should be led by battle-hardened veterans, not by people who have never had to defend their admin actions. The alternative would be Total Recall -- with a single mandatory system with clear criteria. — Kusma ( talk) 10:54, 29 September 2022 (UTC)

I never got the point of "recall". If an admin is good/dormat, then everything is fine. If there are some issue, then there's the option of AN/ANI/arbcom. —usernamekiran (talk) 11:53, 29 September 2022 (UTC)

There are at present 7 opposes citing recall versus 178 supports, so it seems being honest is fine... Back when this question was in vogue, one editor wouldn't support anyone who answered either yes or no: the answer they wanted to hear was that there should be a common process that applies for all administrators. isaacl ( talk) 16:05, 29 September 2022 (UTC)

Does anyone know of a way to look at RFA !voting history that's still functional? Scottywong's tool has been down for a while, AFAIK. Vanamonde ( Talk) 06:16, 18 September 2022 (UTC)

@ Vanamonde93, [8] is what I use (there may be others too). Giraffer ^{( talk· contribs)} 08:00, 18 September 2022 (UTC)

@ Giraffer: Thanks for that; I should have been clearer, I meant the history of a single editor's !votes at RFA, akin to AfDstats. We had an RFAstats tool, which has been down for a year or so, IIRC. I was wondering if there are alternatives. Vanamonde ( Talk) 15:25, 18 September 2022 (UTC)

Vanamonde, I've been missing that, too. Valereee ( talk) 09:40, 20 September 2022 (UTC)

I've found it a very useful tool when researching candidates at RFA (not just nominees), and ACE; and also whether it's worth discussing things with !voters. I find it quite informative to learn whether !voters are typically entrenched in their positions, or are willing to amend their views following reasoned debate. I've also found it informative to learn when an editors expectations are wildly out of sync with the community's. In past years I would run that tool for every ARBCOM candidate I was considering, and several times I have been swayed by observing a pattern with this tool that wasn't visible from individual !votes. Vanamonde ( Talk) 09:52, 20 September 2022 (UTC)

I made a new tool. It mostly works in my limited testing, one known issue TNT found is that it can't detect votes if a user has been renamed. Legoktm ( talk) 00:32, 27 September 2022 (UTC)

Wow, @ Legoktm, super useful! Bookmarking this. Valereee ( talk) 00:43, 27 September 2022 (UTC)

Awesome tool! It would be a bit more useful to me if the results were sorted chronologically instead of alphabetically. Is that an option? Firefangledfeathers ( talk / contribs) 00:46, 27 September 2022 (UTC)

@ Firefangledfeathers: done, give it a try now. Legoktm ( talk) 01:11, 27 September 2022 (UTC)

Worked! Wizardry! Wonderful! Firefangledfeathers ( talk / contribs) 01:12, 27 September 2022 (UTC)

On mine, it's pulling up votes where people mentioned my name rather than my votes ?? Still looking ... SandyGeorgia ( Talk) 01:21, 27 September 2022 (UTC)

Thanks for doing this ... it is also missing a lot of my votes for reasons I can't decipher (they don't seem to be malformed). For example, Fvasconcellos, TimVickers, Newyorkbrad ... many many more. SandyGeorgia ( Talk) 01:35, 27 September 2022 (UTC)

PS, Like Espresso Addict, I've never been renamed. SandyGeorgia ( Talk) 01:38, 27 September 2022 (UTC)

I wonder if it's older ones? One of mine was Iridescent. Espresso Addict ( talk) 01:48, 27 September 2022 (UTC)

It's hard to tell, because every time I click on one to check and then try to click back, it has to reload ... too much waiting ... those three I spotted are quite old though ... but in my case, between counting votes that aren't mine, and missing those that are, not yet a working tool ... so glad someone is working on it though! SandyGeorgia ( Talk) 01:51, 27 September 2022 (UTC)

It misattributes a couple of other users' !votes to me as well; I reported those on Legoktm's talk page (like the tool says to) before seeing the similar reports here. I do want to comment here, though, that I think it's really good design that the tool provides a snippet of the !vote in question, which goes a long way toward mitigating this sort of error. — Cryptic 02:07, 27 September 2022 (UTC)

I re-did the signature detection so it should make less mistakes (erring on the side of not attributing a vote rather than misattributing it), though it's still not perfect. As Cryptic noted on my talk page, it currently will fail to parse old RfAs that didn't use real section headings to separate the Support/Oppose/Neutral sections. I'll think about how to solve that tomorrow. Legoktm ( talk) 02:38, 27 September 2022 (UTC)

Thanks! It still misses almost half of my votes though ... perhaps a solution will be found. SandyGeorgia ( Talk) 12:02, 27 September 2022 (UTC)

The GitLab link in the bottom of the tool appears to be broken. I clicked on it and it made me sign up for an account, then it displayed a 404. Perhaps the repo is set to private? https://gitlab.com/toolforge-repos/rfa-voting-history. – Novem Linguae ( talk) 02:09, 27 September 2022 (UTC)

@ Novem Linguae: My bad, fixed. Legoktm ( talk) 02:39, 27 September 2022 (UTC)

This is wonderful, Legoktm, thank you. The recent ones, or at ones post-2014, are what I care about most, though I'm sure your work on parsing the older ones is appreciated too. Vanamonde ( Talk) 02:41, 27 September 2022 (UTC)

• It seems to have trouble extracting votes for me; a spotcheck revealed several cases where I made an explicit bolded comment using "Support" or similar. I've never been renamed. Espresso Addict ( talk) 01:17, 27 September 2022 (UTC)

I've had a play with the tool; I've discovered that Eric Corbett / Malleus Fatuorum supported more RfAs than people might have remembered; not a majority but still a sizeable chunk. Ritchie333 ^{(talk) (cont)} 11:52, 27 September 2022 (UTC)

"People" ... ??? Mally supported lots of editors at RFA; don't know where "people" got the impression he didn't. Anyway, the tool misses most of Mally's votes, as mine, so not yet an indication for old-timers. SandyGeorgia ( Talk) 12:05, 27 September 2022 (UTC)

@ Legoktm: I'll be annoying and request a feature here so that others can chime in — it might be useful to also show if the RfA passed or failed? — TheresNoTime ( talk • they/them) 11:58, 27 September 2022 (UTC)

First the tool has to be made to work correctly ... SandyGeorgia ( Talk) 12:06, 27 September 2022 (UTC)

I checked the text of Wikipedia:Requests for adminship/Newyorkbrad, and the source code for this tool, and the problem is exactly as Legoktm describes. The solution is to either parse the entire text of the page and use a state machine to work out which section you are in, which would make the tool much slower, or get a bot to do a one-shot reformatting of old RfAs so they have proper headers, which would need approval to run.

TNT's feature request can be implemented by changing is_rfa() in parser.rs to return an enum (passed, failed, unknown, error) instead of true or false. Ritchie333 ^{(talk) (cont)} 12:39, 27 September 2022 (UTC)

I was going to second this request; my memory is quite good for the last few years, but it's nice to have the visual. Thanks again for making this work. Vanamonde ( Talk) 16:49, 27 September 2022 (UTC)

If the old RfAs can't readily be parsed, until approval to edit the text into a more bot-readable form is forthcoming, perhaps just exclude all the ones before the date the header changed and write a note explicitly stating that it only includes RfAs from [date]? Espresso Addict ( talk) 22:42, 27 September 2022 (UTC)

If someone's going to implement a bot, I'd suggest not modifying the original pages, incorporating the parsing logic into the bot, and writing the extracted data into a machine-parseable format, and perhaps also a wikitable format, saved onto new pages. Query tools can then be built on top of the extracted data, and will work much faster. isaacl ( talk) 21:02, 28 September 2022 (UTC)

+1 to this. — CX Zoom[he/him] ^{( let's talk • { C• X})} 00:05, 29 September 2022 (UTC)

Disagree; there is no reason to edit dozens (hundreds?) of old RFAs so that they will be in a format suitable for parsing by a modern script designed to give some statistics that likely only a few dozen editors will look at or care about. This is not to say that I think this is a waste of time, because clearly there is interest, I am just disagreeing with the idea of a bunch of unnecessary edits to historical pages. Primefac ( talk) 05:40, 30 September 2022 (UTC)

Yes, I agree with you, which is why I suggested not modifying the original pages. Writing out the data into a database as proposed by Legoktm is fine, too. isaacl ( talk) 05:51, 30 September 2022 (UTC)

Whoops! Totally missed the "not" in your original statement. Primefac ( talk) 07:20, 30 September 2022 (UTC)

Now that I've gone through all the trouble to write a parser for all of this, I would like to reimplement the tool as a cronjob that dumps everything into a database and the web tool just needing to query from that. And making the database open to other tools/people doing analysis. It'll also make it much much faster instead of parsing on the fly each time. Legoktm ( talk) 05:07, 29 September 2022 (UTC)

An update

See https://rfa-voting-history.toolforge.org/ SandyGeorgia ( Talk) 12:26, 28 September 2022 (UTC)

I deployed an update just now that 1) (mostly) handles old RfAs properly (tested with all the ones SandyGeorgia has voted in), 2) displays the result of the RfA, and 3) adds an input field for also matching against old usernames ( example for TheresNoTime).

The remaining cause of incorrect "Unknowns" that I'm aware of is when the vote has indented blocks or a list, because it's incorrectly discarded as replies to a vote. Legoktm ( talk) 06:16, 28 September 2022 (UTC)

Impressive ... in looking through those that the tool couldn't handle in my case, I found this oddity. Will keep looking as I have time, SandyGeorgia ( Talk) 12:12, 28 September 2022 (UTC)

Another of those at Wikipedia:Requests for adminship/Dihydrogen Monoxide 3. SandyGeorgia ( Talk) 12:22, 28 September 2022 (UTC)

And now a third at Wikipedia:Requests for adminship/ScienceApologist 2; so this number of courtesy blanked RFAs affects the stats. SandyGeorgia ( Talk) 12:28, 28 September 2022 (UTC)

@ SandyGeorgia: correct, the tool looks at the current revision of the page so it cannot parse blanked RfAs. I've documented this limitation and others at User:Legoktm/rfa-voting-history (feel free to edit/expand). Legoktm ( talk) 23:36, 29 September 2022 (UTC)

It's still picking up comments where someone mentioned and linked to another user in their vote and trying to analyze them, but putting them in "unable to find"; see the 71 "unable to find" at https://rfa-voting-history.toolforge.org/votes?username=Newyorkbrad&old= (Is that a record RFA supporter?) SandyGeorgia ( Talk) 12:45, 28 September 2022 (UTC)

In looking through NYB's unknowns, a lot of them are just comments without actually making votes. In one case, he didn't sign properly so it doesn't get detected as a signature, and this one is a list (already noted as not working). I also updated the tool's message to explain that the Unknown list is RfA pages they've edited, but where no vote could be parsed. Legoktm ( talk) 23:47, 29 September 2022 (UTC)

Also checking https://rfa-voting-history.toolforge.org/votes?username=Malleus+Fatuorum&old= b, it is still missing a large number of "good" votes (eg Mally supporting Wikipedia:Requests for adminship/Karanacs and many more). SandyGeorgia ( Talk) 14:12, 28 September 2022 (UTC)

In that case, it's because he supported as Malleus Fatuarum and then commented later after renaming to Malleus Fatuorum. Ritchie333 ^{(talk) (cont)} 14:21, 28 September 2022 (UTC)

Yup. Works for me then. — Cryptic 14:23, 28 September 2022 (UTC)

I can't make it work; what am I doing wrong here? SandyGeorgia ( Talk) 14:57, 28 September 2022 (UTC)

If you look at the history of User:Malleus Fatuorum, you'll see that he wasn't actually renamed. The tool uses the account in the Username field to get the initial list of RFAs to parse; he edited Wikipedia:Requests for adminship/Karanacs with the Malleus Fatuorum account, not the Eric Corbett one. — Cryptic 15:03, 28 September 2022 (UTC)

Ah ha ... thx! SandyGeorgia ( Talk) 15:19, 28 September 2022 (UTC)

• Thanks, Legoktm, that's finding a great deal more of my old ones. Still oddities at Pengo, Oscarthecat 2 & Philippe, for some reason. Espresso Addict ( talk) 20:48, 28 September 2022 (UTC)

  @ Espresso Addict: all of those should be fixed now. Legoktm ( talk) 23:48, 29 September 2022 (UTC)

  Nice work on this tool! Seems to be working smoothly for me, though I have a pretty limited RfA history. —Ganesha811 ( talk) 15:47, 3 October 2022 (UTC)

It's long since we have 3 RFAs in process. Do you think the last time it happened? Thingofme ( talk) 09:31, 11 August 2022 (UTC)

I think the last time was September 2020. PhantomTech[ talk 09:34, 11 August 2022 (UTC)

March 2022 had 2 successful/1 unsuccessful, but the lat time we had 3 successful was September 2020. Worth noting there are 4 RfA’s in 2022 August, since DanCherek's RfA already closed as successful. ~ 🦝 Shushugah (he/him •  talk) 09:37, 11 August 2022 (UTC)

However we have no RfAs for 3 months until DanCherek succeed. Thingofme ( talk) 09:38, 11 August 2022 (UTC)

Four in a month is pretty exceedingly rare in modern Wikipedia. Although it does show that there are still suitable candidates out there. Lee Vilenski ^{( talk • contribs)} 09:39, 11 August 2022 (UTC)

Absolutely! A yearly average of 17 RfAs since 2018 means even two RfAs in one month is uncommon. The direction has been continuous, despite a momentarily uptick in wiki editing during onset of COVID. ~ 🦝 Shushugah (he/him •  talk) 09:41, 11 August 2022 (UTC)

As Wikipedia:RFA by month, in January 2020 we have four successful RfAs, September 2020 we have 3 successful RfAs. Thingofme ( talk) 09:45, 11 August 2022 (UTC)

WP:DBM, by comparison is a scary read. Lee Vilenski ^{( talk • contribs)} 09:58, 11 August 2022 (UTC)

Mostly removals now are due to inactivity, sometimes are ArbCom cases or emergency. Next year we will have new inactivity policy so Jan. 2023 or Feb. 2023 we will have loads of desysops. Thingofme ( talk) 10:02, 11 August 2022 (UTC)

Timing and required notices actually means that the admins affected by the new rules will not be desysopped until April. Primefac ( talk) 10:35, 11 August 2022 (UTC)

Three months notice for the new requirement (5 years)? So the bot will start reporting inactive admins by the new policy from January 2023 or October 2022, 3 months before the start? Thingofme ( talk) 11:07, 11 August 2022 (UTC)

The former; the new policy doesn't come into effect until January, meaning we cannot send out notices until then. Primefac ( talk) 11:09, 11 August 2022 (UTC)

What the hell happened in July 2011?!?! casualdejekyll 19:12, 1 September 2022 (UTC)

Per the note, it was when we regularly started desysopping for inactivity. I suspect we will see a similarly large spike in January. Primefac ( talk) 19:18, 1 September 2022 (UTC)

The low-contrast colour scheme (blue on black) makes the footnote very difficult to see, so I don't blame casualdejekyll for not seeing it. (I just checked and almost didn't see it although I knew it was there). — Kusma ( talk) 19:41, 1 September 2022 (UTC)

No kidding; I only knew there was a note there because there had to be a note explaining why, so I hovered around until I lit up the note. My comment was not meant to belittle but to explain the provenance of my knowledge (though I can see how it could be interpreted that way). Primefac ( talk) 05:36, 2 September 2022 (UTC)

I only realized there must be a note because the number is off center in the cell. Valereee ( talk) 18:29, 2 September 2022 (UTC)

That's strange. Wonder what scared people off. -- Tamzin ^{cetacean needed} (she|they|xe) 10:07, 11 August 2022 (UTC)

Four in a month is not that odd, we had four in a 30-day period earlier this year (Mar/Apr). Primefac ( talk) 10:41, 11 August 2022 (UTC)

It's like waiting for a bus, then three come along at once. Perfectly normal for a random distribution, and even more so as RfAs effect each other. In fact, there's probably some clever study that could be done comibining the psychology of the process with the accordion effect or bus bunching Worm^TT( talk) 09:47, 11 August 2022 (UTC)

Shushugah is highly inspired by DatGuy's RfA but I think he will probably fail. (But the accordion effect may encourage people to run RfA) Thingofme ( talk) 10:00, 11 August 2022 (UTC)

When I transcluded my RfA, 14 others were already running. Those were the good times. Or the bad times, depending on your perspective :) — Kusma ( talk) 10:16, 11 August 2022 (UTC)

I am newer so I didn't know of the time when 10-20 RfAs are happening everyday, like WP:PERM now. Thingofme ( talk) 10:19, 11 August 2022 (UTC)

I never knew that until now. In my opinion, that was definitely a good thing because there was less courage needed to run for an RfA, and there would be less scrutiny on an individual nominee. 0x Deadbeef 12:22, 11 August 2022 (UTC)

I believe that's why we tend to get spurts of 2-4 nominations all in a short span. Primefac ( talk) 12:44, 11 August 2022 (UTC)

@ Kusma: It was the best of times, it was the worst of times, it was the age of wisdom, it was the age of foolishness. :) -- Hammersoft ( talk) 13:22, 12 August 2022 (UTC)

At this rate we might break even on the ever shrinking number of admins. Dr vulpes ^{( 💬 • 📝)} 11:30, 11 August 2022 (UTC)

Number of admins may shrink down to below 1000 this year, and down to fewer than 900 after the new inactivity policy take effect. Thingofme ( talk) 12:51, 11 August 2022 (UTC)

There are less than 500 that are active, by one measure (defined as 30 or more edits during the last two months) [9]. 152 administrators have not edited in the last 3 months. -- Hammersoft ( talk) 13:44, 12 August 2022 (UTC)

Also, despite the shrinking number of RfA per year, we have now balanced the shrink and we have RfA records broken, like most number of support votes in a RfA/support votes in a unopposed RfA. Thingofme ( talk) 10:18, 12 August 2022 (UTC)

But that doesn't really mean much - a pass is a pass is a pass. Cullen328 has no more rights than Moneytrees. Ritchie333 ^{(talk) (cont)} 10:19, 12 August 2022 (UTC)

The support rate (average) of passed RfA are mostly 90% or higher nowadays. Thingofme ( talk) 13:26, 13 August 2022 (UTC)

It is worth noting that we will have had more RfAs by the end of this week in 2022 than we did for the entirety of 2021. I don't know if it was the pandemic, but we really had nothing last year (less than one a month on average) Lee Vilenski ^{( talk • contribs)} 11:11, 12 August 2022 (UTC)

Also I expect next year to experience another increase based on the editors who joined during COVID-19 quarantine or became active in that timeframe (like me) and have gained experience in the project. To clarify: I'm not saying I'll run just that there's a cohort of editors I've seen learn alongside me who I expect will be good candidates a year from now. — Ixtal ^{( T / C )} ⁂ Join WP:FINANCE! 11:19, 12 August 2022 (UTC)

As an editor that got active during the pandemic and that is finally developing some clue, I agree with this statement :) – Novem Linguae ( talk) 11:24, 12 August 2022 (UTC)

Forget it. Since you've showed an interest in being an administrator, you're obviously unqualified ;) -- Hammersoft ( talk) 13:27, 12 August 2022 (UTC)

Yes, you need at least two or more years of saying "no way" like me and Hammersoft did :-) Ritchie333 ^{(talk) (cont)} 13:38, 12 August 2022 (UTC)

Too true. If you're not dragged to the altar kicking and screaming, you're obviously incompetent. Anybody stupid enough to become an admin is too stupid to be an admin. :) -- Hammersoft ( talk) 13:40, 12 August 2022 (UTC)

This suggests that people who want to be admins must keep the thought as a secret (to have a higher chance). That advice is worse than just stopping being biased against these editors. Heck, even self-noms are frowned upon by some! Makes me wonder whether anonymous nominations would be an improvement, but I don't doubt that there would be people who would like to have self-nom or not as a part of their judgement on character. 0x Deadbeef 13:59, 12 August 2022 (UTC)

@ 0xDeadbeef Going back a few years, when we had dozens of RFAs, we had concerns of "hat collection" and need of the tools. We're in a very different wikipedia these days but unfortunately some of those old opinions have lasted. I was a self nom back in the day, but even I think twice about self noms when I look over a candidate. That was a big reason I made Request an RfA Nomination, which did help a bit. Worm^TT( talk) 14:29, 12 August 2022 (UTC)

You wink, but there's actually truth in what you say. More often than not, people who actively want to be admins are usually considered by the community to be the least qualified, absent all other qualifications present. It's a bit like being the Watergate special prosecutor; a thankless job no reasonable person should want, pressure from all sides of the aisle, and it pays like shit. 🌈WaltCip-( talk) 14:27, 12 August 2022 (UTC)

Well, the pay here has always been a problem, especially when you become an admin. It was explained to me though. Joking aside, I've seen oppose votes for RfAs where candidates had been expressing too much interest in being an administrator. -- Hammersoft ( talk) 14:33, 12 August 2022 (UTC)

So people should not express our will of becoming an admin, and if you are really qualified you should request a nomination it's better to pass... Thingofme ( talk) 16:10, 12 August 2022 (UTC)

There's differing schools of thought on that. One school of thought recommends avoiding WP:ORCP if someone has an interest in becoming an admin. Seems counterintuitive, but it's true. Another recommends using it as a tool to evaluate where you are and what you need to work on. Yet, the former can interpret that as trying to burnish your resume to be an admin and views it as a negative. Personally, I didn't use ORCP, but not because I subscribed to any school of thought, but rather that if what I had done over the years wasn't enough to be an admin, then it wasn't enough to be an admin. I wasn't going to attempt to draw votes by having X numbers of AfDs participated in, or N numbers of WP:AIV reports, etc. That all seemed pointless. Either you're qualified and capable, or you're not. -- Hammersoft ( talk) 16:33, 12 August 2022 (UTC)

I agree that it's not necessary (especially for those not planning to self-nom). One can get a similar idea of where one is pass/fail wise by reading RFAs from the past 2-3 years. Nonetheless, it still is one of many useful tools for future candidates. As with all tools, mileage may vary. — Ixtal ^{( T / C )} ⁂ Join WP:FINANCE! 19:02, 12 August 2022 (UTC)

Sorry to throw cold water on all of this, but for all the rosy estimations, a good week can just as easily be a dead cat bounce. The long term trends shown by WP:RBM and WP:DBM are not easily reversible. In fact, it's very unlikely. -- Hammersoft ( talk) 13:27, 12 August 2022 (UTC)

I wish there was a solution to this. I can name three dozen users that I believe are nearly/already at the qualified threshold (Including some involved in this conversation). But these old habits appear to have scared away plenty from stepping up to the podium. Personally, I think most of the general fear towards RFA is unfounded, seeing that the opposes in the recent discussions appear to be cordial enough. But hey, I have been here only seven months, so forgive me if my perception is too limited. CollectiveSolidarity ( talk) 01:11, 13 August 2022 (UTC)

Maybe requests for RfA nomination should be made private through off-wiki conversations, to reduce the number of opposes. Thingofme ( talk) 09:28, 13 August 2022 (UTC)

They usually are, Thingofme. Kudpung กุดผึ้ง ( talk) 10:28, 13 August 2022 (UTC)

Hitting the nail on the head again as he usually does so admirably, Hammersoft is dead right with his dead cat bounce. For a long time now, the sample size of RfA transclusions has been far too small to demonstrate any trends other than the continued disinterest in going through the process, and the perennial parlance here aren't going to change anything. Biblioworm's Dec 2015 didn't change much other than greatly increasing the voter participation, and I'm still trying to mentally unravel what Barkeep's massive programme of RfCs achieved other than the ultimate creation of yet another noticeboard for bullying admins by providing a dirt track in the forest with more rocks to turnover to look for something to complain about. The end effect of that will be to discourage even more editors of the right calibre from throwing their hat in the ring.

The vast majority of candidates at ORCP do not have the potential to become admins any time soon. Most of them can't/won't read the instructions and not reading instructions for anything is an immediate disqualification. Others seem to be just trolling. An occasional rare gem does score high and goes on to get the bit. Most editors who have read all the advice pages first don't bother with ORCP - they don't need to. ORCP nevertheless still has some shelf life life because it is probably largely responsible for the huge drop in the totally time wasting transclusions. Anyone who joins Wikipedia with the intention of becoming an admin (and plenty do) has joined for the wrong reasons; hence why those 'I wanna be an admin' UBX don't get much serious attention. I would advise any newcomers to this talk page to read this Has the wind gone out of the AdminShip's sails? in which Hammersoft is cited and where the huge number of reader comments are even more telling that the article itself. Kudpung กุดผึ้ง ( talk) 08:42, 13 August 2022 (UTC)

Maybe this month we will have 4 successful RfAs since January 2020. Thingofme ( talk) 04:26, 23 August 2022 (UTC)

Namely, DanCherek, DatGuy, Femke, and Z1720. GeoffreyT2000 ( talk) 23:15, 26 August 2022 (UTC)

But it seems like RFA are now remaining stable with high/low times, at 8-15 successfuls per year. Thingofme ( talk) 10:25, 27 August 2022 (UTC)

With just under a third of a year left, this year is likely to end in the 8-15 range. But if it does it will only be the second to do so - and the other one was way back in 2018. So I wouldn't describe it as stable or running at 8-15 a year. I'm still seeing a slow downward trend, not as precipitous as it was from 2008 to 2012, nor as consistent. 2021 is the lowest year on record, but 2018 is the second lowest complete year and we now only need one successful RFA in the next four months to beat it and two to surpass it. The pattern I'm seeing is that each month we get 0-4 new admins. In the last 100 months there have only been two above that range, But months with one or zero new admins have tended to become more common. Ϣere SpielChequers 17:44, 2 September 2022 (UTC)

The next RfA is Wikipedia:Requests for adminship/ScottishFinnishRadish, which could be either the 10th successful or the 5th unsuccessful RfA this year. If the former, then the count would be the same as for 2018, and one more successful RfA would bring the count to 11. Only time will tell, of course, how many RfAs we will have in September, October, November, and December 2022. GeoffreyT2000 ( talk) 14:30, 11 September 2022 (UTC)

Unfortunately, ScottishFinnishRadish's RfA will likely end up being the 5th unsuccessful RfA this year. So, now, we would need two more successful RfAs this year after that to bring the count to 11. GeoffreyT2000 ( talk) 14:20, 13 September 2022 (UTC)

For now it may be ending in the discretionary range so I think it's 50/50 or 30/70 chances that he will pass. Thingofme ( talk) 08:33, 14 September 2022 (UTC)

There is now currently a bureaucrat chat at Wikipedia:Requests for adminship/ScottishFinnishRadish/Bureaucrat chat. The last time we had a no consensus RfA with a bureaucrat chat was for Jbhunley in August 2018. Let's see how the current bureaucrat chat will go. GeoffreyT2000 ( talk) 03:03, 20 September 2022 (UTC)

I believe Wikipedia:Requests for adminship/Tamzin/Bureaucrat chat is more recent, May 2022. Bureaucrat chats can be a lot of stress; best of luck to the candidate. – Novem Linguae ( talk) 07:08, 20 September 2022 (UTC)

Tamzin was promoted. Ymblanter ( talk) 10:16, 20 September 2022 (UTC)

Ah good point, I misread, apologies. – Novem Linguae ( talk) 21:53, 20 September 2022 (UTC)

Now, SFR has passed RFA, making the number of successful RfAs in 2022 so far the same as the total count (10) in 2018. Let's see whether there will be an 11th or 12th successful RfA this year. GeoffreyT2000 ( talk) 05:26, 22 September 2022 (UTC)

There are three months left in the year, but after Tamzin's RfA there weren't any RfAs for three months afterwards. Thingofme ( talk) 10:04, 23 September 2022 (UTC)

Now, there is Wikipedia:Requests for adminship/Whpq. If that RfA is successful, then that would make the count for 2022 surpass the one for 2018. A three-month gap after that would mean that that is the last RfA in 2022, so the next RfA would then be in 2023. Again, only time will tell, of course. GeoffreyT2000 ( talk) 04:42, 25 September 2022 (UTC)

Maybe we have the 6th successful RfA for 2 months but time will tell... 2021 has the last successful in September. Thingofme ( talk) 07:32, 25 September 2022 (UTC)

And now Whpq's RFA is successful, so the successful RfA count for 2022 (currently 11) finally surpasses the one for 2018. But we still do not know whether there will be another successful RfA this year. GeoffreyT2000 ( talk) 04:49, 2 October 2022 (UTC)

Another one Isabelle Belato, maybe another successful one. Thingofme ( talk) 11:11, 8 October 2022 (UTC)

In a Signpost interview, ScottishFinnishRadish noted that one easy way to make RfA better for candidates would be to close RfAs automatically after one week. Obviously, they would not be closed as successful or unsuccessful. Instead, they would be marked with something like {{ Rfah}} (without the note that it was placed by a 'crat). 'Crats would still evaluate and then close the RfA (or RfB) as they currently do. SFR had a great rationale in the interview, so I am just going to post it verbatim:

If you've been sitting through a week of contentious RFA, that end point is a light at the end of the tunnel. Let's have a bit of sympathy on someone who's endured a stressful week-long experience and just cut it off. Having an end to look forward to helps, and watching it go by without actually ending does not help at all.

I will note that when {{ Rfah}} was created, there was discussion that determined it was only for use by 'crats. Then again, that discussion was in 2007. Should this be implemented? Do we really need an RfC to make this change, or is this informal discussion sufficient? And if this should be implemented, what is the best way to do so? Is there some wikitext/module magic that can be used to achieve this, or is a bot (that might only make an edit once a month, but must check continuously for "overdue" RfAs to be useful) the best way forward? House Blaster ^talk 16:39, 1 October 2022 (UTC) Note: I am raising this for future RfAs, not for Whpq's.

• I would support such a change. Given that a bureaucrat could close the RFA at any time after 7 full days, I see no systematic benefit to the community by not doing so, and the considerable benefit of reducing stress in candidates. Vanamonde ( Talk) 16:47, 1 October 2022 (UTC)

  Noting for the record my complete agreement with Valereee and Levivich below, and also that as far as I am concerned, no criticism is intended of the bureaucrats. Of course it takes time to close an RFA properly, and the crats aren't expected to babysit. The candidate is, however, and as we can reduce the burden of that expectation by a little bit, we should. Vanamonde ( Talk) 20:28, 2 October 2022 (UTC)

  Completely agreed that this is not about bureaucrats somehow not moving fast enough. We're all volunteers. Valereee ( talk) 15:08, 3 October 2022 (UTC)

Traditionally we use a six-week two-stage RfC to implement things like this. I don't believe this would be as helpful as claimed, though: should someone come with a truly damning observation shortly before the official closure, bureaucrats could currently decide to wait for the reactions of others before closing. Under the proposal, they might consider reopening the RfA instead. As long as we view RfA as a consensus-building discussion, it shouldn't have hard cutoffs for the discussion. (I personally would prefer a far more vote-like system, where a hard cutoff makes more sense to me). — Kusma ( talk) 16:51, 1 October 2022 (UTC)

Is this an issue that needs solving? Looking at the last five successful RFAs, SFR's was closed 4.5 hours after it was supposed to end, but it was always going to go to a 'crat taking at least the better part of a day. The others were closed 4 minutes after "time", 12 minutes early, bang on time, and an hour early (due to time zone confusion). This is a solution in search of a problem. Primefac ( talk) 17:04, 1 October 2022 (UTC)

Femke's RFA closure diff ID is so cool: 1105055050 I can memorise it. — CX Zoom[he/him] ^{( let's talk • { C• X})} 17:22, 1 October 2022 (UTC)

I think it's an issue that is worth solving. It doesn't affect a lot of people, but for those whom it does affect it can add one more bit of stress to an already stressful situation. And it costs us very little to solve it: we just need to decide here, "Yeah, good idea, let any admin close the discussion with "Consensus to be determined." Valereee ( talk) 16:57, 2 October 2022 (UTC)

• I'd be interested to hear from others who had a contentious RfA closed a little after the cutoff point about if an automatic close would've made the experience better for them. Given that bureaucrats are likely closely watching the end of an RfA, this issue might also be solved just by crats changing their behavior and choosing not to wait, even if it means using a placeholder close. {{u| Sdkb}}  ^talk 18:46, 1 October 2022 (UTC)

  Um... who says 'crats are waiting? I've already indicated that this is not true. Additionally, the idea of edit-conflicting with an on-the-second bot that puts the RFA "on hold" is extremely problematic to me. Primefac ( talk) 18:54, 1 October 2022 (UTC)

  ( edit conflict) I think Sdkb is suggesting that 'crats (in non-obvious cases) 1) put the discussion on hold 2) evaluate consensus 2.5) possible 'crat chat 3) close discussion with summary of consensus instead of 1) evaluate consensus 1.5) possible 'crat chat, at which point the discussion is put on hold 2) close discussion with summary of consensus. If SFR was a one-off, I would agree that a bot/other technical solution is likely to be unnecessary. Maybe a gentle request that 'crats put the discussion on hold, even if you do not intend to be the closer, is all that is needed.
  W.r.t. solution looking for a problem, candidate stress at RfA is definitely a problem. SFR has said that closing the discussion "on time" would have reduced his stress. If something can possibly decrease RfA stress for the candidate, even if just slightly, I would consider that a good thing. W.r.t. edit conflicts, there is already the chance you will edit-conflict with a closing 'crat. I do not see how conflicting with a bot would be any different. House Blaster ^talk 19:14, 1 October 2022 (UTC)

  There is zero difference between your two suggestions, in that any 'crat that is even remotely following the RFA will know the outcome well before the 168 hours are up; there is no reason to put it on hold (barring due to a 'crat chat) unless it's just on the borderline of 75% and could still be unilaterally closed, and even then I would argue that the closing 'crat will already have a consensus determined.

  With all due respect to SFR, it was clear to anyone watching the discussion (from about day 3) that it was going to a 'crat chat; it would take a literal last-minute unprecedented wave of opposition to drop it even remotely close to 65%. Expecting it to "all be over" exactly after that time is unrealistic because there is still the inevitable delay between opening the 'crat chat and having us all respond.

  candidate stress at RfA is definitely a problem - sure, but this doesn't solve it. Primefac ( talk) 20:24, 1 October 2022 (UTC)

  If it even only slightly reduces it, why not? I babysat my RfA. Like literally sat and watched it. It was a very stressful period. Every time there was a new oppose, I had to deal with the things people were saying about me: that I was unaware, apathetic, exploiting recent events, or a strikebreaker crossing a picket line. It was not fun waiting for the next comment, and the fact those late comments would be removed or discounted wasn't a comfort. Valereee ( talk) 16:58, 2 October 2022 (UTC)

  I was well aware it was going to crat chat, which is all the more reason to not leave it open any extra time. There is a sharp difference between waiting out the crat chat, and actively monitoring your RFA. ScottishFinnishRadish ( talk) 18:06, 2 October 2022 (UTC)

  I don't think technical implementation concerns are relevant yet, it could just as easily be implemented with a template that automatically "closes" the RfA when time is up. Legoktm ( talk) 19:31, 1 October 2022 (UTC)

  @ Sdkb: To answer your implied question: There was one post-168-hour-mark vote at my RfA. It was an oppose, and so I didn't like that it was cast. If it had been a support, I'm sure I would have been thrilled that it was. :D But it was a bit surreal overall. I had an alarm on my phone for the minute it was to end, and it was strange having that alarm go off and then still being in limbo. -- Tamzin ^{cetacean needed} (she|they|xe) 11:09, 3 October 2022 (UTC)

For what it's worth, (and I am the newest crat, so it might not be worth much), I put a timer in my phone for about ten minutes before the RfA should close to remind me to take a look at things and put a closing template together. However, we are talking about an item that could happen at any time of the day, and there is only around 20 of us. Something that closes at a good GMT time for me might be poor for others, and we all still have lives (unbelievable, I know).

That said, the actual close is rarely the important part, as either the user knows they will pass/fail, or have to wait for a cratchat anyway. Having a cratchat close within a day to me is a good indicator that the work is being done in a timely manner. All an automated close is going to do is remove later !votes, which isn't something I've ever seen be an issue before. Lee Vilenski ^{( talk • contribs)} 19:13, 1 October 2022 (UTC)

The delay caused by a crat chat is inevitable, and candidates understand that. What is avoidable is waiting for yet another crappy oppose comment added late. Waiting for the close can feel pretty excruciating when you've been ticking down expecting for it to be over. Valereee ( talk) 17:30, 2 October 2022 (UTC)

With no disrespect intended towards SFR, the change to "on hold" of that RfA came a little less than four hours after it was scheduled to conclude. The counts changed from 232/87/5 [10] to 234/92/5 [11]. In terms of percent, that's a difference of less than 1%. The outcome of the RfA wasn't affected by those four hours. I don't see how creating a new procedure for RfA would have changed anything in this case. Creating new procedures for stopping something that had effectively no effect is not good business. Show me several candidates who were stressed over such closures, and then we might be on to something. For one case? No. Also, Kusma is correct; as long as RfA is run as a consensus gathering mechanism, an RfA that closes late isn't a problem. If some new revelation comes in late in an RfA that begins shifting people from support to oppose, that is a good thing, not a bad one that we should be shutting down by an arbitrary seven day cutoff. I could just as well see an argument that if N% of !votes come in on the last scheduled day of an RfA, that it will be automatically extended, as it would seem the community has become far more interested in the RfA. So, no, I oppose this change. -- Hammersoft ( talk) 19:29, 1 October 2022 (UTC)

If one person says they were stressed, I can guarantee you multiple others were and haven't said anything. My feeling is that if we can do something easy and simple to reduce RfA stress on someone, there is no good reason not to. When someone is going to a crat chat, there's gigantic stress, and that stress has been going on for perhaps the entire week and almost certainly most of the week. And honestly, we should be listening to the folks who had stressful RfAs, here. For those of you who had basically stress-free RfAs: listen to those of us who for whatever reason didn't. We are the ones whose RfAs people are looking at and thinking "JFC, I am not risking going through that." Valereee ( talk) 17:09, 2 October 2022 (UTC)

That ties into my advice to people thinking about RFA from the interview, Read contentious RFAs closed in the discretionary area and decide if that's how you want to spend eight or ten days of your life. ScottishFinnishRadish ( talk) 17:47, 2 October 2022 (UTC)

If extending the duration of an RfA by 4 hours (2.4% of the entire duration) is enough to dissuade people from ever running, it's probably a good moment for some self reflection. As I noted, if you can find others who felt an RfA going past its 7 days normal allotted time was stressful, then we might be on to something. Reacting to a single case (across thousands of RfAs) as proof of a problem is a very bad basis on which to make changes. RfAs are stressful period, even if they go well. -- Hammersoft ( talk) 17:57, 2 October 2022 (UTC)

Right. Yes. Exactly. I was actually about to make exactly this point, that even unanimously-supported RFAs, even ones from before it became the norm to ask more than maybe one "optional" question instead of the 30+ that are the norm now, are horrifically stressful. Why go out of our way to make them even worse in a way we could actually fix for a change? — Cryptic 18:02, 2 October 2022 (UTC)

Ok great, so show me others who experienced significant stress because their RfAs went some time after the 7 days normal period? -- Hammersoft ( talk) 19:42, 2 October 2022 (UTC)

Tamzin, SFR, and I are ALL telling you this. :D Valereee ( talk) 13:58, 3 October 2022 (UTC)

I think a bot that "closes" (puts on hold) at the closing time is a good idea. Anyone worried about an edit conflict with the bot can just wait until a second after the closing time. Another very easy solution is to just allow any editor to apply the "on hold" template at or after the closing time. Just crowdsource it; someone will be on and watching. Although a bot could also be used to close any discussion that editors wanted to auto close at a certain time, so there might be some broader use there. Levivich ( talk) 20:31, 1 October 2022 (UTC)

Could you elaborate on why you feel that this is a good idea? (please do not ping on reply) Primefac ( talk) 12:13, 2 October 2022 (UTC)

I can. It's because RfA is stressful, and recently someone who has gone through RfA has said having the damn thing go on for hours longer than they'd been counting down to added to that stress, and removing potential sources of stress may help encourage other people to run RfA. When removing those potential sources of stress costs us basically nothing, why wouldn't we think it was a good idea? Valereee ( talk) 17:13, 2 October 2022 (UTC)

Sure. 1/ First thing that comes to mind is that auto-closing at 168 hours would have avoided discussions like this and this (and two years later...). 2/ An extra 4 hours might not seem like a lot to me or you, but I bet it sucks if you're the RFA candidate going through it. Some expect candidates to be online and responsive constantly during their RFA (see this recent exchange for example). The candidate should have the peace of mind that comes with knowing that there is a definitive end time. During those 4 hours, the candidate didn't know it would be 4 hours. It could have been 14, who knows. 3/ During those 4 hours, BTW, 5 new oppose !votes came in. They didn't affect the outcome, but it still sucks to read five more oppose votes over the extra 4 hours, simply because there isn't a crat online right now. 4/ Deadlines provide motivation. If everyone knew that 168 hours was a strict drop-dead deadline, people who want to !vote will find time to !vote before then. That's a good thing, as it reduces the possibility of late developments in an RFA. 5/ Honestly, if the RFA candidates say it would be less stressful, then that's the end of the discussion for me. I can see no reason not to close it at 168 hours, and at least 5 reasons to close it. Levivich ( talk) 17:16, 2 October 2022 (UTC)

+5 Valereee ( talk) 17:46, 2 October 2022 (UTC)

6/ if we auto-close, crats won't have to worry about being online right when it closes. Levivich ( talk) 20:33, 2 October 2022 (UTC)

+6. I am going to expand slightly on point number 2: those 4 hours may have been 2.4% of the time, but they were a much larger part of the stress. Personal anecdote: when I was about five years old, my mother dropped me off at school for one of my first days of kindergarten, which I knew would last until noon. I was a mommy's boy; I hated that she was "abandoning" me for a couple of hours. I was nervous. I was scared. I was stressed. The knowledge that she would be back at 12 was comforting. She came to get me at around 12:15. Those last 15 minutes were the worst part of the day, by far. Even though they were negligible in terms of time, they had an outsized influence on my stress. I did not know when she was coming. After 12 came and went, there was no longer an "endpoint". There was no longer a light at the end of the tunnel. It was actually this memory that convinced me to bring this up. I also want to respond to Primefac's point above about how it was obvious that it was going to a 'crat chat since day 3. We know that candidates who are well above the discretionary zone still have a negative experience with RfA. An RfA that pass at 90% but has 10 opposes that hurt is still stressful, even though it is successful. I have never gone through RfA, nor have I considered doing so, but I would guess that the anxiety surrounding RfA is not entirely about whether you pass or not. I believe it is about the scrutiny you are put under, the expectation to justify any diff thrown at you, and reading the searing opposes. As further evidence, both SFR and Tamzin have said that the 'crat chat was the least stressful part of their experience. If it were entirely about passing, I would imagine a 'crat chat would be the most stressful part of RfA. House Blaster ^talk 20:59, 2 October 2022 (UTC)

Since this started because of my statement in the interview, maybe I should elaborate a bit on why I said it. Near the week-long highly stressful and contentious RFC I was strongly looking forward to the end on Monday evening, not of the entire process, but of the part that I was expected to follow, absorb, and possibly respond to. The crat chat wasn't as big a deal, because that was a process that I wasn't actually involved in. Monday evening rolled around, and right around dinner time it was slated to end. My wife saw that I was still focused on my phone, keeping an eye on my watchlist like I had been for the past week and asked when it was going to be over. I told her that that (at that point) it was going to end in about 20 minutes. We started eating, and when I finished up I saw that it was still open. My wife asked why, to which I responded, "well, no one closed it yet." She responded, as she does to most things I try and explain about Wikipedia, "that's dumb." I understand the delay was only four hours, but for me that worked out to an entire extra night, which was made worse by the continuation of comments. There was still discussion continuing as I retired for the evening, for another night of wondering what I would wake up to.

As has been pointed out above, the additional votes and statements made no difference to the outcome. They did, however, make a difference in the total time I was under that microscope. Like I said in the interview, knowing that there was an end point made the whole ordeal easier to deal with. Watching it go by was pretty disheartening, and heading to bed that night not knowing when it would actually be over was really not great.

Some of the objections above seem to me to be even more of an edge case than "contentious RFA doesn't get closed immediately and causes unnecessary stress." Sure, it's possible in the last couple hours something wild could happen, some sufficiently damning diff dug up, or the candidate could rip off a mask and show that they're actually Larry Sanger, but how often has that actually happened? That concern can be addressed the same way it is now, a crat says "we're keeping this open to see if the revelation that the candidate thinks that Adrian Paul is the best Highlander is going to sharply change the vote ratio." The end date could be adjusted at any time with a number of different solutions. As it stands it's making it less likely that someone who thinks there's a chance they may have a contentious RFA will stand at RFA. It's also adding stress to an already stressful experience for no actual gain. ScottishFinnishRadish ( talk) 17:39, 2 October 2022 (UTC)

Speaking of things that don't actually happen, can anyone remember when was the last time a bureaucrat deliberately extended an RFA? That's about the only potentially-valid excuse I can see not to do this. — Cryptic 17:55, 2 October 2022 (UTC)

• If we auto-close the community should step away from its long held stance that RfA is a discussion and not a vote. I obviously support the idea of autoclosing with a purely mathematical determination of passing — this has been my stance for years. The community as a whole, however, does not. If the community continues to insist that RfA is a discussion and not a vote, and that bureaucrats are to be given discretion on when to close, when to put on hold, when to go to crat chat, etc. the idea of auto closing at exactly 168 hours makes little sense.
  So I'd support the change if and only if it comes with larger RfA reform that sees us admitting that RfA is a pure vote and nothing more. Unless that happens, we I oppose this change. TonyBallioni ( talk) 18:54, 2 October 2022 (UTC)

  It makes sense to the stress level of the candidates. Why does it have to "make sense" in terms of whether we're pretending this isn't a vote? We can keep our pretensions and still make life easier for candidates and potential candidates. I think Emerson had something to say about that. :D Valereee ( talk) 14:01, 3 October 2022 (UTC)

  I think the idea is that incremental improvements could dampen the enthusiasm for a radical restructuring of the way RfA works. If RfA is made less awful, then there will be less of an appetite for a major overhaul. 28bytes ( talk) 15:06, 3 October 2022 (UTC)

  Perhaps if we set up an actual gantlet? :D Oh, wait...I think those did actually have a definite end point. Hm, so that won't work. Valereee ( talk) 15:23, 3 October 2022 (UTC)

• Having been through RfAs and an RfB myself, I agree with SFR that arbitrarily extending what many here call “hell week” simply due to the unavailability of a bureaucrat at the scheduled closing time is an added stress for no particular benefit to anyone. This seems like a low-risk change (just code it to show as closed in the RfA templates themselves, you don’t even need a bot) that might add a little touch of humanity and empathy to the process. (I would also support codifying a caveat that 'crats may extend or re-open an RfA due to significant last-minute revelations in case people are concerned about that.) 28bytes ( talk) 19:16, 2 October 2022 (UTC)
• I'm going to disagree with TonyBallioni. I don't see that closing but not deciding an RFA at the end of seven days changes it from a discussion to a vote. Rather it makes it a discussion that defaults to ending after 7 days. My experience of two RFAs and an RFB is that a week at 60% is stressful, a week clearly above the discretionary zone? Not what I call stress. So I don't think that this change would make much difference, but what change it would make seems purely positive (provided a crat could decide to extend an RFA). One last point. Wikipedia:Bureaucrats#Current_bureaucrats shows that we crats are rather concentrated geographically, with the majority of us in just two time zones. It might help timing and other things if we had a crat from the Indian subcontinent or Australasia, or if people starting RFAs avoided doing so at hours that meant it likely that everyone on the shores of the North Atlantic was likely to be in bed 7 days later. Ϣere SpielChequers 19:30, 2 October 2022 (UTC)

  We used to have at least one crat from India, but Nichalp has been inactive for well over a decade now. So perhaps it is time to search for some good candidates from Australia/New Zealand/India etc. Any admins from these areas interested? Graham87? @ Schwede66? Anyone? — Kusma ( talk) 20:09, 7 October 2022 (UTC)

  There's also Gadfium and Canley from these time zones. Schwede 66 20:15, 7 October 2022 (UTC)

  Thanks but no thanks; I've thought about this before (for reasons unrelated to time zones) but I'd prefer to stay on the sidelines. the whole RFA process is too crazy for me. Graham 87 04:11, 8 October 2022 (UTC)

I don't have very strong feelings on this, but would lean to leaving it be. RFA's are a discussion, and we set a minimum time on them so that there is an opportunity for many editors to contribute - keeping in mind that some editors may only contribute once a week, such as on weekends. This is similar to almost all other discussion we have, for basically the same reason. — xaosflux ^Talk 21:05, 2 October 2022 (UTC)

RfAs, unlike other discussions, publicly count down. In multiple places. RfAs, unlike other discussions, are a job interview conducted in public with up to two questions allowed by everyone, including your worst work enemy and the intern who started yesterday. RfAs affect real people in their real lives. So, no, not really similar to almost all other discussions we have. Valereee ( talk) 14:07, 3 October 2022 (UTC)

I agree with Tony and Xaosflux on flexibility being one of the characteristics of Wikipedia discussions. We shouldn't give incentive to editors to drop in last-minute comments before a hard deadline. I am sympathetic to the stresses of those who feel they are being evaluated in public for a week. To me it would be better to address this root cause. The admin elections proposal made last year (that had a significant majority in favour but was not enacted) is one possible approach. (On a side note, although I appreciate there was no negative intent, I don't think comparing those volunteering to perform administrative tasks to kindergartners is apt.) isaacl ( talk) 22:28, 2 October 2022 (UTC)

(Just responding to your side note) I would be happy to strike if others find it inappropriate, and I can assure you there was no ill intent. My sincere apologies to anyone I offended. I was not trying to compare RfA candidates to kindergartners, rather, I was trying to compare the closure of RfA to my mother coming to pick me up. I did so to explain why I believe a firm end time to a stressful event is a good thing, and that a tiny amount of additional time can amount to an outsized increase in stress. House Blaster ^talk 23:15, 2 October 2022 (UTC)

A kindergartner, though, will not have had enough life experience to learn how to deal with the stress of a sliding deadline. Someone seeking a grant of administrative privileges will have had more experience to help manage this situation. We can argue if the community should have lower expectations than they currently do on the degree of experience, but I don't think going to one extreme helps support your position. isaacl ( talk) 23:32, 2 October 2022 (UTC)

I can see merit to both sides, but I would be fine with it automatically locking at the deadline and awaiting an available bureaucrat to close it, start a bureaucrat chat, or (presumably rarely) unlocking the RFA for further comment. Useight ( talk) 23:36, 2 October 2022 (UTC)

• I'm sympathetic to anyone feeling stress for being involved in any aspect of Wikipedia editing. We are all volunteers, and nobody should feel that they have to be subjected to stress. An RfA runs for a minimum of seven days, but can run longer. We try to close them as close to that exact seven days as possible as a matter of curtesy, but as it stands there is no requirement for them to end on exactly seven days. And I kind of like leaving room for human judgement. It reminds us that there are humans involved in this project, and I am cautious of too many human actions and decisions being taken away from us. I hope that we can think of ways we can alleviate the stress of a contentious RfA (a non-contentious RfA is less stressful because it is clear soon enough if it will be successful or not) rather better than involving a bot. I think that making it understood that a candidate can withdraw from a contentious RfA and try again later, is one thing. I'd be interested in someone running the numbers on how many candidates who withdrew early from a contentious RfA then ran again successfully compared to how many who went to a crat chat and failed then ran again successfully. Another thing is Crats getting more involved in moderating RfAs. I don't think we do enough of that. In fact, I don't think there are enough of us to do that. I'd like more Crats, especially younger, more active users becoming Crats. I'm not sure allowing inline responses to opposes (or supports) is helpful - these can degenerate into unpleasant hot spots. It might be more helpful to allow everyone a space to comment, in which they can give their support or oppose along with any reflections they have on the candidate and/or on what other people have said - do away with the support and oppose column. Put more emphasis on the comments, and less on the voting. More like AfD or like an ArbCom case. Take away the daily vote counter with that dreadful percentage of doom (though, to be fair, this can alert people to when a RfA is turning, and encourage them to return to it to re-evaluate their !vote). Consider having RfA more like an ArbCom election - a period of questions and discussion followed by a secret ballot. Perhaps have a monthly RfA - candidates are invited to step forward (or are nominated) on the 1st of each month, followed by seven days of questions and discussion, and then seven days secret voting - Crats then look at the votes, announce the results, and flick the switches on the new admins. I'd prefer us to focus on reducing the overall stress of the RfA, than replacing humans with bots in order to save a few minutes (or - very rarely - hours) of stress. SilkTork ( talk) 00:50, 3 October 2022 (UTC)

  When the community weighed in on the topic of voting/discussion last year the consensus was the current balance was OK but if there was going to be a change it would be for more voting. This was then born out by an election proposal nearly passing during the second phase. Barkeep49 ( talk) 01:01, 3 October 2022 (UTC)

• I support automatic closure. If the status quo and proposal were reversed, I couldn't see "Let's shut down the bot that closes after 7 days and have a crat close it up to an hour early or five hours late" being a winning suggestion. I agree with WereSpielChequers that automatic closure doesn't affect the discussion/vote question. Firefangledfeathers ( talk / contribs) 01:14, 3 October 2022 (UTC)

  That gave me a solid chuckle. ScottishFinnishRadish ( talk) 01:20, 3 October 2022 (UTC)

  This was a convincing argument. / Julle ( talk) 11:09, 3 October 2022 (UTC)

• A couple of thoughts. Firstly, I have no problem with an RfA shutting itself down at the end of the time limit - just as we don't allow votes prior to going live, a hard stop works in my head unless the bureaucrats actively choose to extend. The concern that Primefac has of a bot edit conflicting with a closing 'crat is a valid one, but edit conflicts happens, so it's not that big a deal in my mind. However, cannot the RfA template simply have an end date included and automatically show itself as "on hold" after that end date? No bot needed, the page simply goes "yellow" and a note at the top says it is on hold pending closure.
  To those saying this is not worth fixing, time and again community members complain about the RfA process being awful - why would we not try to make it a bit better with clear goalposts? Worm^TT( talk) 08:59, 3 October 2022 (UTC)

  I don't think this is so much about the "how" but the "should" of the matter. — xaosflux ^Talk 13:25, 3 October 2022 (UTC)

  We have lots of discussions that run for at least 7 days, such as WP:AFD; if some editors write an article and it is up for deletion - it can be at AFD, where it certainly can run "long", while the authors hypotetically sit around "stressed" if their efforts are going to be wiped away. I haven't heard any arguments that we should shut down people contributing to those discussions because they show up past a redline timestamp - and in XFDs it is also routine for discussions that have evolved over their course to be extended. We generally only have hard "not after" constraints on actual "votes" - and generally that is related to the fact that the ending is holding up a team of volunteers from starting more work. — xaosflux ^Talk 13:25, 3 October 2022 (UTC)

  I don't think the stress of RFA is comparable to AFD. A "bad RFA" is one in which your colleagues publicly and repeatedly call you a liar, or biased, or fascist, or stupid, or a jerk. I've never been through it but I'm sure having your article deleted ain't the same. Levivich ( talk) 13:58, 3 October 2022 (UTC)

  Strangely we have no shortage of people willing to AfD. :D Valereee ( talk) 14:53, 3 October 2022 (UTC)

  Indeed. The comparison to AfD isn't reasonable; AfD isn't discussing your character. Vanamonde ( Talk) 15:27, 3 October 2022 (UTC)

  It is discussing your work, a team of people maybe deciding if what you created (and possibly even what a team of people created) is worthy to be read by everyone in the world or not. We have certainly had contentious AFD's before. I'm not saying they are the same thing - just that they have some similarities. — xaosflux ^Talk 16:29, 3 October 2022 (UTC)

  Except an AFD is not about people discussing the authors' work, nor about deciding whether that work is worthy to be read. The discussion at AFD is about whether or not the topic meets notability requirements, it's never about the quality of the article. It's not even close to an evaluation of an editor; there is no similarity whatsoever, other than both are discussions. Levivich ( talk) 16:35, 3 October 2022 (UTC)

  If we conclude that an article is not-notworthy, we are indeed saying this isn't something that we should publish for our readers. Articles don't materialize from the air, proposing that an article has poor quality is claiming the authors produced poor quality work; proposing that an article subject is "non-notable" is a reflection that the authors poorly picked a topic. There are certainly denials at WP:PERM/AP because the authors have been judged to not be sufficiently skilled at writing articles (oft citing their creation deletions); generally this comes with feedback on how to improve - just as many RFAs opposers identify things to work on. I wholly agree that AFD and RFA are different, and that RfA speaks more directly at the character of the candidate. My main point is that AFD can be stressful for an author as well, but no one seems to worry that we keep them stressed too long, especially with the vast difference of expectations that RFA candidates should have going in vs anyone else who clicks publish. — xaosflux ^Talk 17:39, 3 October 2022 (UTC)

  I've heard RFA compared to a lot of things - a job interview, a cross-examination, a public flogging. I'd think that the fact these comparisons are being made would mean we should err more towards caring for the human element rather than trying to prolong an already agonizing process. 🌈WaltCip-( talk) 16:36, 3 October 2022 (UTC)

  Yeah, I think the concern of edit-conflicting with a bot can be easily put to rest with a simple bit of conditional logic in the template. No bot required. 28bytes ( talk) 14:50, 3 October 2022 (UTC)

  Creative use of {{ show by}} is my (current) technical solution to the above; it will just be one more line of code the 'crat closing the discussion will need to replace. Primefac ( talk) 13:44, 5 October 2022 (UTC)

• Speaking as someone who has gotten seriously hammered at RFA, and as someone who once made the mistake of suggesting that his RFA could be closed a few hours early in the interest of not running into the beginning of the work week, I really don't see the need to auto-close RFAs. The discussion ends when it ends. -- SarekOfVulcan (talk) 15:42, 3 October 2022 (UTC)

  Things have hopefully changed for the better since then (I see some familiar faces in that thread who have now been blocked for ongoing incivility, for example). 🌈WaltCip-( talk) 15:50, 3 October 2022 (UTC)

• For what it's worth, I understand the stress candidates go through; I twice ran for adminship and twice ran for bureaucrat, so I get wanting an RfA to be closed as soon as possible once the scheduled end is reached. That being said, if we bureaucrats were taking days and days to get around to closing RfAs or if we had an overwhelming amount of candidacies up that was too much for us to handle, then this proposal would have some merit but even in those circumstances the problem would be slow or insufficient bureaucrats, not RfA closing procedures. While there was a short wait between the scheduled RFA closing time and when I placed ScottishFinnishRadish's candidacy on hold and started the bureaucrat chat, nothing that happened after the scheduled time changed the result. Overall, I think this is a solution looking for a problem. Acalamari 22:08, 3 October 2022 (UTC)

  No one is objecting to crats needing time. No one is accusing crats of being slow. They're saying it's quite likely not necessary that the grilling they thought would be over by the time they finished dinner still wasn't over by the time they went to bed several hours later, and that there may be a cheap simple fix. They're pointing out this unnecessary extra bit of stress is not only unkind to candidates but may even in some slight way be counterproductive to attracting candidates, and they're suggesting maybe this is one tiny way in which we could improve the process. Valereee ( talk) 22:19, 3 October 2022 (UTC)

  One thing that sticks out in my mind about how an unexpected delay can impact the process was Tamzin's crat chat. Several of the crats said that because it was a bank holiday weekend, that they would have to wait a couple extra days before they would have a chance to review the RfA and provide their opinion. Obviously no one intended to make Tamzin wait any longer than necessary, but real life gets in the way for all of us at times, and I think removing the risk of an unnecessary and unexpected prolongation of the process would indeed pay dividends. 🌈WaltCip-( talk) 16:35, 4 October 2022 (UTC)

Since the beginning of 2016 (when the discretionary range changed), there have been six RfAs (Money emoji, Floquenbeam 2, RexxS, Jbhunley, GoldenRing, Godsy) that have gone to crat chat (not including this year, and thus SFR). In those six years, there have been 93 successful RfAs and 161 unsuccessful. Crat chat RfAs comprise 3.7% of all RfAs over that time. The average time those RfAs have gone beyond their scheduled close is 3 hours and 26 minutes. On those RfAs, there have been an average of 4.3 votes cast after scheduled close. Curiously, the voting improved the support/oppose % in every case where it happened (RexxS, Jbhunley, GoldenRing, Godsy). Of those six RfAs, two (Money emoji, Floquenbeam 2) closed within five minutes of scheduled close. Of the other four, none of the editors in question have edited in the last two weeks. I was considering reaching out to those four to inquire as to their thoughts about their RfA, but it seems futile given their inactivity. We can't draw any conclusions based on six RfAs as a statistical sample. But, the results are nevertheless interesting. One thing I think we can reasonably conclude though is that crat chats happen so infrequently (once per year) and RfAs that go to crat chat going past their scheduled time significantly being even less common, it's doubtful it is any sort of deterrent to running. I'd rather we poll potential candidates perhaps by looking at Wikipedia:List of Wikipedians by number of edits to find out why they don't run. That would, I think, be considerably more productive. -- Hammersoft ( talk) 00:21, 4 October 2022 (UTC)

Well, I guess you've scientifically proven that the people who say it's stressful are just making it up? Levivich ( talk) 00:29, 4 October 2022 (UTC)

I haven't proven anything. I do note your unwarranted hostility though. -- Hammersoft ( talk) 00:57, 4 October 2022 (UTC)

Sorry, didn't mean to be hostile. Levivich ( talk) 03:04, 4 October 2022 (UTC)

Ritchie333, Barkeep49, TonyBallioni, and a few others including me, have spent a considerable amount of time talking to potential adminship candidates, drawn from that list and elsewhere. Of the people I've approached, at least two-thirds have declined outright, and the stress of RFA was by far the most common reason cited. Quite a few more have decided to wait on it. I've now nominated six candidates, which ought to give you some idea of the numbers. The other nominators (all more active than me) may have additional perspectives to offer, but I'm quite certain they've heard very many concerns about the atmosphere of RFA too. Also; why do you assume the RFA isn't going to be stressful if it's outside the discretionary zone? Mine was, and I would have appreciated a quick closure too. Vanamonde ( Talk) 04:42, 4 October 2022 (UTC)

The numbers I looked at were focused only on those that went to crat chat. I'm not presuming those that didn't weren't stressful. It's just that those that don't go to crat chat very likely have a clear outcome that extending past the scheduled end won't change. -- Hammersoft ( talk) 11:28, 4 October 2022 (UTC)

Mine had a clear outcome. Still quite stressful, and I really was counting the minutes until it was over and no more awful things would be said about me. :D There are multiple stories from recent candidates whose RfA did not go to crat chat at WP:Requests for adminship/Debriefs. Valereee ( talk) 12:47, 4 October 2022 (UTC)

I haven't done as much of it as in past years, but I literally have not had a single person who I approached for RfA take me up on it this year. Every candidate I've done (or am on track to do by the end of the year) approached me. I'm not sure stress of the RfA would be the #1 decline reason I get - undesired scrutiny would probably be the #1 reason I hear - but stress is something that does come up in basically any discussion that gets at all serious and it's a common theme in the debriefs even among people who pass relatively easily. Best, Barkeep49 ( talk) 15:46, 4 October 2022 (UTC)

I would say that the scrutiny is the same as the stress. The scrutiny is the primary driver of the mental toll, as well as (as my wife put it) answering for your crimes. ScottishFinnishRadish ( talk) 16:29, 4 October 2022 (UTC)

Same as BK40. I've approached multiple editors. Very few are willing to even think about it. Very few people I'd actually want as an admin look at the RfA process and think, "Oh, totally worth it!" That may partially be because in recent years one of the things we've required in an admin is "not a jerk."

Jerks don't care if they're pummeled as long as they get the mop in the end. Non-jerks -- those who tend to be capable of empathy -- also tend to take criticism more to heart. Which in my opinion is a good thing, no matter how many people think being willing to go through RfA's hell week is evidence you can stand up to the stresses of being an admin.

An admin should take criticism to heart. The ones who don't are the ones who end up refusing to adapt. Valereee ( talk) 16:56, 4 October 2022 (UTC)

I don't know that quite matches my experience/perspective. I think a good admin needs the wisdom to be able to (politely) brush off bad faith accusations, listen to good faith criticism, and to know when to admit mistakes. Knowing how to appropriately handle criticism is one of the questions I emphasize with people who I get serious about exploring RFA and it has been the deal breaker for some. As I not infrequently say, if you're happily editing wikipedia and there's a reasonable chance RfA will make you unhappy, don't do it. Best, Barkeep49 ( talk) 18:33, 4 October 2022 (UTC)

I agree that admins need to be able to brush off bad faith criticisms. It's the jerks who also brush off good faith criticisms, and if you can do that, you probably won't feel much stress in RfA. Valereee ( talk) 18:47, 4 October 2022 (UTC)

Are we able to differentiate between bad faith and good faith criticisms? In SFR's RFA, there were criticisms that some fools felt were made in good faith and others (correctly) felt were in bad faith. Should those be brushed off or taken on board? (The first one.) While I may have an opinion (you'll never guess what it is) if we collectively don't agree on what is a good/bad faith criticism, we're going to have a hard time agreeing on whether an admin's response to it was good/bad. That's why I think character-based evaluations of RFA candidates are kind of bunk. You can't evaluate whether someone knows when to admit mistakes unless you can objectively determine what is and what isn't a mistake... and the putative mistakes that actually matter to us are the ones where there isn't any objective right-or-wrong answer. For example, the unusual use of advanced permissions by a few editors during the WP:FRAM saga: were those mistakes, or were they proper uses of tools? Depends on who you ask. Can I fault those editors for not admitting their mistakes? Again, depends on who you ask. One person's "Support - he correctly brushed off bad faith criticism" is another person's "Oppose - he failed to admit mistakes". RFA is, in part, a test to see if a candidate's opinion of what is good and bad matches the opinion of at least three-quarters of a couple hundred RFA voters. In other words, "Are your values our values?" We call this "trust". If the answer comes back "no", it's hard not to take that as a rejection of your values. Hard to feel a connection to the community after that kind of rejection. I imagine the process of finding out is, and always would be, stressful. Levivich ( talk) 19:05, 4 October 2022 (UTC)

I'm not talking about this w/re making an RfA !vote, though. I'm talking about it w/re whether someone feels stress in/is willing to undergo RfA. If you can brush off any criticism, fair or unfair, you probably aren't going to feel as much stress at RfA. And in recent years we've been focussing on people who "aren't jerks". Which (if what I'm hypothesizing is true) might mean we're getting fewer people to run because the people who can brush it off are the people we aren't asking. We think, "Oh, X is very clueful and without fail civil and helpful" but the reason they're without fail civil and helpful is the same reason RfA looks awful to them. It's because they've got the capacity for empathy or however else we want to shorthand "not a jerk". Valereee ( talk) 19:27, 4 October 2022 (UTC)

This is a general problem that goes beyond RfA. I think for many collaborative editors just looking to help out a little, encountering someone who reverts their changes with an abrasive, caustic comment results in them thinking, this isn't very fun, and then silently leaving. Most of people who like to talk about these things will shrug it off, saying editors have to grow a thicker skin (that is, brush off criticism). The net effect is that the English Wikipedia environment selects for editors willing to brush off criticism, and those who are more abrasive. I appreciate there are new editors who ought to be discouraged from editing, and being abrasive towards them is an easy way to get a lot of casual, unsuitable editors to leave. We have to understand, though, that it affects the community's ability to get and keep editors involved. isaacl ( talk) 20:58, 4 October 2022 (UTC)

We can evaluate how someone handles a disagreement, whether or not we agree with their viewpoint. There are a lot of disagreements that are related to the different weights being given to different factors, so there can be honest disagreement. You can examine someone's communication skills, and whether or not their lines of reasoning unfold logically. That being said, giving and receiving critical feedback is very difficult. The best way to make it less stressful is to reduce the degree to which it is done in public. To accomplish this, the community needs to be convinced to support a different procedure where public feedback is given by skilled communicators who can do it sensitively and constructively. There are lots of ways to do this, with different pros and cons. For better or worse, most of the editors who like to comment on these matters seem to prefer having a big mass, public conversation. isaacl ( talk) 20:44, 4 October 2022 (UTC)

And we actively invite anyone who's discovered their watchlist. Advertising it on watchlists was considered 'reform' when it happened. I'm sure there's no argument that it increased participation, but did it improve the quality of the process in any way? Valereee ( talk) 20:59, 4 October 2022 (UTC)

For better or worse, there are lot of editors (amongst those who like to comment on these things) who are wary of delegating responsibility, so tend to support proposals that solicit discussion from more people. (I suspect in the larger population of all editors, many would be happy delegating to some sysop group.) Given that RfAs are already a poll of self-selected editors who happened to show up, I understand the argument that we might as well encourage more participation, to get a (slightly) broader sampling. isaacl ( talk) 21:12, 4 October 2022 (UTC)

Arbitrary break

• I am inclined to oppose any proposal that automatically forces a cessation of participant activity (!voting/commenting) precisely at the 168 hour mark. That said, I understand candidate stress and believe it's a factor worthy of mitigation. To that end, I would support a counterproposal that did not implement a default procedure until a prescribed amount of time (beyond 168 hours) elapsed (perhaps 2 or 3 hours) where the RfA was still pending closure at that time. Best regards.-- John Cline ( talk) 05:49, 4 October 2022 (UTC)

  There's a strategy used for silent auctions, which usually have a stated end time...a bid in the final five minutes increases the closing time by another five minutes; this continues until there are no more last-five-minutes bids. That makes sure someone isn't gaming the system by sneaking in a bid just under the wire. But in a case like RfA, it could allow for the possibility of a late-discovered reason for opposition. Would that be a solution for your concern? Valereee ( talk) 12:55, 4 October 2022 (UTC)

  If the goal is to allow everyone time to respond to the new part of the discussion, 5 mins would be way to short; that would need something like a day - and I don't think we'd want anything to force the "minimum" time to be pushed indefinitely (as people could keep replying). — xaosflux ^Talk 13:28, 4 October 2022 (UTC)

  I thought about this idea, but didn't think it would address the expressed concern over the effect of a sliding deadline on the requestor. Anyone who would be waiting anxiously for additional comments with the current procedure would still be doing so. Also, I suspect in practice it wouldn't shorten the duration of an RfA versus current practice, assuming that an extension is on the order of hours. isaacl ( talk) 15:38, 4 October 2022 (UTC)

  I appreciate your input Valereee. While it's interesting, and I can see its utility in concert with an auction, I agree with Xaosflux that it isn't well suited for RfA. I also agree with Isaacl that any process involving auto-extensions would do little or nothing to alleviate a candidates stress regarding the lack of a fixed time for the RfA discussion to end. For what it's worth, my rationale for intimating the counterproposal I suggested is basically that I do not wish to commingle my support of a fixed deadline for the RfA discussion to end with a proposal that fixes the deadline by faulted reason and wrongful assumptions (encroaching outright assumptions of bad faith). To elaborate, fixing the deadline at T-168 hours implies that the only way to end an RfA discussion at or reasonably proximate its scheduled end time is to effect an automated process that forces the discussion's end at precisely the earliest opportunity that would have afore been possible. In the context of "a fix", the thing broken is implied to be rampant bureaucrat lethargy resulting in an unacceptable prevalence of intolerably late RfA closures. Of course, the exact opposite has historically been the overwhelming norm regarding RfA closure and instead of implying otherwise, I prefer to embrace the track record while encouraging its continuation by fixing the implementation of any automated process sufficiently beyond the 168 hour mark (to allow the professionalism and unimpeachable discretion of our bureaucrat volunteers to selflessly emerge as we have come to expect) yet reasonably near the scheduled end time (perhaps 2 or 3 hours as has been suggested) to effectively mitigate the stress of anticipating an RfA's closure where currently no fixed ending exists while simultaneously precluding the fringe minority of RfA closures which one could reasonably feel had languished in its pending state. Similarly, an overriding concern with participant activity while an RfA's closure is pending is undoubtedly born of the wrongful assumption that latecomers are editors of bad faith and ill will whose pending closure participation is singularly purposed to negatively impact the RfA where again, the sum of pending closure activity at RfA does not bear out any such preponderance of negativity (as I recall of 10+ years of RfA observance and participation). I, therefore, see no reason for automatically disallowing respondent activity solely because an RfA is pending closure. These things said, I hope others consider the merits of the counterproposal and perhaps elivate it to become their own first choice. Thank you and be well.-- John Cline ( talk) 05:33, 5 October 2022 (UTC)À

  A couple crucial clarifications: No one, and I mean no one, is saying there's 'rampant bureaucrat lethargy' or anything close to it. No one here is saying bureaucrats aren't working fast enough. Nothing like that has been said. Likewise no one is saying latecomers are acting in bad faith. No one has said anything like that at all. In fact I don't believe anyone here has implied anyone is acting in any kind of bad faith whatsoever.

  The idea of closing when countdown ends has literally zero to do with preventing potential bad faith action on the part of anyone. Zero, nothing whatsoever, and no one here is saying that. Valereee ( talk) 11:44, 5 October 2022 (UTC)

  What difference would an extra 2-3 hours really make? Or an extra 5 minutes after every late comment? You had a full week to make your opinions known. If you couldn't adequately express your opinions in a week, than an extra 2 hours ain't gonna help. Besides, how many RfAs are really decided in the final hours of the RfA? Most are either closed after 12 hours, or already a foregone conclusion after 3-4 days. Any RfA that's really on a razor's edge by the end of a week will trigger a crat chat anyway. —⁠ScottyWong⁠— 19:38, 6 October 2022 (UTC)

  My preference for a 2-3 hour delay before implementing an automatic procedure to end participation is not about ensuring extra time for the discussion to continue but instead about trusting our crats and allowing them the continued opportunity to effect a timely close of the discussion as they have predominately done to date. As such, the discussion would end in concert with their decision to close it or place it on hold which would mostly occured well before the 2-3 hours elapsed. -- John Cline ( talk) 23:12, 9 October 2022 (UTC)

  ( edit conflict) Like Isaacl, I have thought about this idea. I think it would be an improvement over the status quo: there would be a known "light at the end of the tunnel". However, I consider it a distant second choice to putting the discussion on hold after 168 hours. Putting the RfA on hold after the minimum length is less arbitrary than 2-3 hours later. One reason that I support this change is that extensions would be done actively through 'crat action, as opposed to passively through 'crat inaction. The counter-proposal does not have this benefit. Finally, I would prefer to eliminate the extra time a candidate has to be actively monitoring their RfA, instead of putting an upper bound on it. House Blaster ^talk 00:58, 6 October 2022 (UTC)

• I'd support. Small step to make the process just that tiny bit more manageable for the candidate, but bigger steps are needed (discussion before !voting / anonomous (!)voting). Femke ( talk) 16:45, 4 October 2022 (UTC)
• My super scientific measurements say that we are at 61 KB of discussion, and I am getting the feeling that we are starting to run into heat:light issues. Both sides have non-negligible support, and this is a change to a highly contentious process. I do think semi-structured and wider input would be useful. To that end, unless anyone has a really good reason not to, I intend to start a RfC below within the next day or so. To be clear, it would just cover this proposal: I am not looking to start WP:RFA2022.
  I also want to clarify, as the guy who brought this up, that I am absolutely not criticizing the job that 'crats currently do. Despite real life, 'crats have a pretty darn good record of closing RfAs on time. Yes, there was a contentious RfA that lingered for 4 hours, but that is quite clearly the exception, rather than the norm. I did not raise this because I felt that the 'crats are "failing" to do their "job". I did so because someone who just went through a tough RfA said this would have improved the experience, and their reasoning really struck a chord with a deep memory. I figured I should just go ahead and start the discussion, rather than wait for someone else to do so. House Blaster ^talk 22:27, 4 October 2022 (UTC)

  I haven't the slightest doubt that your efforts are entirely motivated by good faith. I wish you the best and hope that the RfC culminates with a clear consensus and unambiguous answer to the question posed. To that end, I encourage you not to rush into an RfC and further encourage you not to squander the " RfC before" opportunity that is currently underway. In my opinion, this discussion has more to offer in terms of information that will be useful in a subsequent RfC provided you slow down enough to reap its full measure. Sincerely.`-- John Cline ( talk)

  ( edit conflict) I echo your first two sentences right back at you. I do believe there is a little more that could be said about your counter-proposal, and so I will hold of a bit on a RfC. That being said, comments have slowed somewhat (as I get double edit-conflicted... maybe not?). House Blaster ^talk 00:58, 6 October 2022 (UTC)

• I think that bureaucrats have generally appreciated the ability to allow a discussion to 'run just a bit longer': to allow folks to react to some late breaking revelation, or to see if a trend reversal holds; however in reading the above (particularly the experiences of candidates who have been held in that liminal space), I'm wondering if it would be better for bureaucrats to be actually clear that this is what they're doing, rather than quietly holding back like some kind of pocket veto. (That being said, bureaucrats have not always agreed on whether an RfA should be extended.) – xeno ^talk 00:46, 6 October 2022 (UTC)
• Compromise? I would only support automatic closure where the Rfa is clearly passing after 7 days - with say over 80 or 85% supports. Otherwise it is unfair to potential voters who only check in once a week, have exams, are on holiday or travelling, or are just busy. Johnbod ( talk) 00:54, 6 October 2022 (UTC)

I support ending all comments/"votes" on all RfAs after seven days unless the bureaucrats make an affirmative decision to extend discussion because of "late breaking news". I was fortunate enough to experience an easy RfA, but even for me, the process was moderately stressful. It must be orders of magnitude higher for those undergoing a contentious RfA that ends up in the discretionary zone. I empathize with ScottishFinnishRadish and Valereee and Tamzin. I am sorry that things were so stressful for you. Tamzin, you know that I opposed your RfA over one specific point but in my opinion, you have been a very good administrator so far. I regret adding to your stress level. So, I agree with Vanamonde93 and Levivich. Let's bring the "voting" to an end when seven days have passed in almost every circumstance. Let the candidates sleep. Cullen328 ( talk) 02:45, 6 October 2022 (UTC)

• Comment IMO a hard deadline for commenting should be set irrespective of how long it takes a crat to review the discussion. Seven days is long enough and especially in the case of contentious AfDs, I believe it is unfair to the candidate to subject them to further unnecessary stress. The crats can formally close the discussion whenever they get around to it. - Ad Orientem ( talk) 02:52, 6 October 2022 (UTC)
• I too believe that an auto-close could apply unless a Crat had specifically stated on day 7 that an extension would be occurring. I think @ Xeno: makes a good point that clarity on when a crat is doing so would be beneficial. I would suggest that when a Crat does so (which is likely to be really rare) they should automatically start a crat discussion, rather than risk another Crat coming in and undoing. This discussion would likely not be anything like a CRATCHAT - once it's clear enough it can be closed. Nosebagbear ( talk) 11:28, 6 October 2022 (UTC)
• I support an automatic closure by bot per Levivich. This system is already gives the appearance of being rigged and due deference to procedure is the least this community can provide for the appearance of impartiality. I realize many prefer a system where the RfA closes exactly when it is going "their way" and I warn against this impulse. Chris Troutman ( talk) 18:37, 6 October 2022 (UTC)
• I'd support as long as all RfA pages (going forward) have some kind of countdown timer to clearly let everyone know when voting ends, similar to the "Scheduled to end" text that already appears on an RfA, but with a live timer added. Something like "This discussion will be automatically closed at 13:55, 10 Oct 2022 (UTC), which is 3 days, 7 hours from now." This auto-closing change shouldn't force people to do calendar math to figure out how much time they have left to comment. As long as it's made clear up front, no one can reasonably complain. —⁠ScottyWong⁠— 19:43, 6 October 2022 (UTC)
• I am strongly in favor of at least a trial run. — Ixtal ^{( T / C )} ⁂ Non nobis solum. 20:11, 6 October 2022 (UTC)

Ixtal, A trial run would not demonstrate anything. The sample size would be too small unless you want the trial to run for several years. Kudpung กุดผึ้ง ( talk) 04:36, 8 October 2022 (UTC)

I prefer the minimum not maximum time intent for these discussions. Yes, it's stressful for the candidate but the community should expect full discussion up to and if necessarily beyond the minimum time allowed. The counts typically make the outcome clear but breaking news in the last second must be given time for discussion if bureaucrats think it could change the result. While not essential I do think that it would be useful but not mandatory if a 'crat who is initially reviewing for decision and sees late-breaking comments of that nature added a "minimum extension of n days due to possible effect of recent comments" observation. Not for the candidate's benefit but to make it clear to the community that consequent observations aren't going to be ignored just because the minimum time has passed. The absence of such an observation a few hours after the minimum time should be enough to reassure candidates that the usual response time will probably happen. Things taking time and the need for patience is one of the routine aspects of this place. Jamesday ( talk) 19:15, 9 October 2022 (UTC)