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 propose that we should have an anonymous voting system and replace our RFAs board with a discussion page, similar to WMF's board elections. This would eliminate a lot of existing problems with RfA: toxic behaviors that induce signficant stress to candidates (see User:Tamzin/340/112/16:_An_RfA_debrief for a personal recount by an editor about this), pile-on opposes, burgeoning, etc. Our current RfA system will need to be completely rewamped if we want to vote this way though and I don't know whether that is going to be worth it or not. CactiStaccingCrane ( talk) 04:43, 28 October 2023 (UTC)

Using an anonymous vote is currently being discussed in the preceding section, "Banning replies at RfA and otherwise moving RfA reform forward". isaacl ( talk) 05:13, 28 October 2023 (UTC)

Yes, people will discuss other issues about RfA to death and nothing will get done like it has always been a thousand times. I think that me proposing a solution and other people critiquing the solution is a much more productive use of our time. CactiStaccingCrane ( talk) 14:09, 28 October 2023 (UTC)

@ CactiStaccingCrane: are you aware of Wikipedia:Requests for adminship/2021 review and Wikipedia:Requests for adminship/2021 review/Proposals#Closed: 8B Admin elections? A way forward is for somebody, preferably somebody with experience in these type of large difficult discussions, to workshop amendments to that proposal and put it to an RfC. This won't be changed by a discussion on this talk page. I've been tempted to volunteer when health allows (which might be another year.. ), but hope somebody else will volunteer before me. —Femke 🐦 ( talk) 14:41, 28 October 2023 (UTC)

FWIW I slightly disagree with this @ Femke. I think the way forward is for someone experienced in these type of large difficult discussions to (privately) workshop changes and to put it up to an RfC. I think large scale group writing of it will cause it to never move forward. Best, Barkeep49 ( talk) 16:14, 30 October 2023 (UTC)

When you say privately, do you mean a single person, or two/three editors? I was thinking the latter, even though I fully know that these small groups risk becoming quite large groups. —Femke 🐦 ( talk) 17:03, 30 October 2023 (UTC)

This has the content of something that sounds like sarcasm, but the tone that suggests it isn't... ~~ AirshipJungleman29 ( talk) 16:49, 28 October 2023 (UTC)

No sarcasm intended here :). What makes you say so? —Femke 🐦 ( talk) 16:52, 28 October 2023 (UTC)

Review the indentation, the comment was not in response to you but to CactiStaccingCrane. _{Duly signed,} ⛵ WaltClipper -( talk) 17:50, 28 October 2023 (UTC)

It's not sarcasm that I say so. We love to discuss about reforming RFA but never actually reforming it. For example, at Wikipedia:Requests for adminship/Nominate I have to remove insanely BITEY and obnoxious formatting for self-nominator instructions. Now I know why people do not want to become an admin.... CactiStaccingCrane ( talk) 04:22, 29 October 2023 (UTC)

@ CactiStaccingCrane I just saw what that page used to look like. Egads. And some of the names of the contributors involved in building that make me very much surprised. Unfortunately, that's pretty much what RfA has become nowadays, so it reflected common consensus if nothing else. _{Duly signed,} ⛵ WaltClipper -( talk) 20:03, 30 October 2023 (UTC)

"I think that me proposing a solution and other people critiquing the solution is a much more productive use of our time." what do you think has been happening for the last decade, CactiStaccingCrane? Do you really think you're the only person who's ever thought of a solution? ~~ AirshipJungleman29 ( talk) 21:41, 30 October 2023 (UTC)

That's what's happening in the other thread: a way to proceed was raised and it's being discussed. Your proposal has fewer details than the previous one and is ambivalent about whether or not it should proceed. isaacl ( talk) 18:34, 28 October 2023 (UTC)

Then why don't we make a proper RfC for this? CactiStaccingCrane ( talk) 04:24, 29 October 2023 (UTC)

For a similar reason that your initial post wasn't an RfC: editors are discussing the idea and trying to figure out the best way to craft a proposal that can attain consensus support. The result from the 2021 review was indicative of strong support from the editors who like to weigh in on these matters, so it's likely that a new proposal will be made at some point. There should be more editors now who have familiarized themselves with the SecurePoll extension, which should help make a future RfC discussion more fruitful. isaacl ( talk) 05:10, 29 October 2023 (UTC)

• Note, as we've said before this isn't currently technically feasible unless we want to lump a bunch of rfa's together and only do them once or twice a year -- if someone wants to dedicate the likely enormous amount of time needed to implement phab:T301180, go for it. — xaosflux ^Talk 20:15, 30 October 2023 (UTC)

  "isn't currently technically feasible" doesn't seem entirely accurate to me. We can't have encrypted local securepoll votes because something something GPG. But why can't we have elections on votewiki, the way zhwiki does? Whatever the advantages and disadvantages of doing that may be, it is currently technically feasible, is it not?

  Also, can we have non-encrypted local elections? I see that discussed at phab:T342774 for zhwiki. I'm not sure what the effects of non-encrypted elections means exactly... that everyone would be able to see how everyone else voted? So, the way RFA is now? I'm probably not understanding the technicalities of it all, but reading the phab tasks, it seems one particular technical implementation is not currently possible (encrypted local securepolls), but that other implementations may still be possible (votewiki securepolls, or unencrypted local securepolls). Am I right or wrong? Levivich ( talk) 20:24, 30 October 2023 (UTC)

  zhwiki doesn't just do this on votewiki when they want, they had BATCH election (all rfas are bundled, and done in bulk twice a year); which is what I suggested above is a possibility (and why I supported it in the last RFA omnibus RFC, as an option for candidates). Currently everything on votewiki requires WMF staff to be involved.

  If we wanted non-secret ballot elections, we could do that lots of ways locally (think of some of the public votes on things like picture of the year, steward elections, etc - they are script aided but underneath they are just page edits). A benefit of that is that the current examination voters stays in the existing processes.

  There are many considerations about using securepoll, including that currently if you want to be able to examine for socking it means you basically automatically checkuser every voter (in our arbcom election viewing of this is limited to the scrutineers).

  All that being said, I'm in favor of enabling local securepoll capability - but the support for that is certainly lacking (if "page editing" is broken, we know there is lots of support, if some local adhoc securepoll malfunctions - good luck getting it repaired in a timely manner -- that is unless you and a large enough collection of other volunteers are ready to become developers and commit to supporting it). — xaosflux ^Talk 21:33, 30 October 2023 (UTC)

  Votes are encrypted so no one with access to the underlying database (either directly or I suppose via a MediaWiki vulnerability) will be able to determine how people voted. I'm not very confident in the SecurePoll documentation (for example, wikitech:SecurePoll#Encryption starts with "This section has not been updated for the new situation after 2013"), but since encryption is said to be optional, I think it's a reasonable requirement for the extension to be installable in an environment without the required encryption library. Perhaps someone who's installed it before can say if that's already possible? Reaching a consensus on what the community wants will help determine what work items need to be done, and how to schedule them. isaacl ( talk) 22:42, 30 October 2023 (UTC)

I was busy when the "record low Admins" thread went up at AN and so I missed @ Ritchie333's proposal to ban replies at RfA. I am surprised it got so much support as it seems to identify the largest issue at RfA is that opposers are badgered too much. Some how people getting pushback for opposing candidates is really unappealing for candidates and letting unchallenged ridiculousness happen will make RfA more appealing for people to run? This doesn't track for me. Related ideas that do seem interesting:

• Have a straight vote section (oppose or support only) and all discussion goes in its own area. This feels like a half-way point to just doing an election.
• So maybe just do a secret ballot election. This keeps getting trapped in the "we shouldn't propose it until it's technically feasible" "it won't be technically feasible until there's a reason for it to be" dilemma and just deciding that there is consensus for a secret ballot election feels like a huge step forward.

But in the end this will only happen if someone actually is willing to put in the time to shepherd it forward. I am skeptical that any idea on its own will find consensus without an accompanying consensus that RfA needs to change, but that process while successful in 2015 was ultimately a failure in 2021, which was also my attempt to do that work. So I could be wrong. But I'm definitely not wrong that people just talking about it aren't going to change anything and we are already seeing the impacts of a too small admin core in places that de facto require Adminship like ArbCom and CUOS appointment. Fixing this problem before it gets even worse feels of real value to the project and those who love it (like I do). Best, Barkeep49 ( talk) 13:50, 26 October 2023 (UTC)

Don't we also have record-low active editors in general? I think this is simply a symptom of the organizational lifecycle. The exciting novelty of the wild-'n'-wooly founding days is long over, and what there is now is mostly rather tedious work, all the "sexy" articles already having been written years ago. If the editorial pool in general is shrinking, then the number of admin hopefuls within in it would also shrink. Has anyone done any analysis on the proportion of active admins to active non-admin registered users over time? I would think that would be more indicative of whether we actually have an admin shortage than doing a straight head-count of admins as if the editorial population wasn't changing. Anyway, as for the above, "banning replies" would probably do nothing but give a whole lot of new weight to opposes (few supports are ever challenged, while most opposes are, either directly or by questioning the original opposes that others are copycats of). So, this would just worsen the [alleged] admin shortage by resulting in fewer nominees passing RfA. Maybe that would be mitgated somewhat by permitting such challenges but in only a discussion section, but then again that would kind of defeat the "reform" purpose of the notion; if people say hurtful things, it doesn't matter whether they did so at line 75 or line 358 on the page, nor whether there's a "==Discussion==" line above the comment. As for a traditional secret-ballot vote, I think that is more practical, and the ArbCom elections already prove that it works fine for an analogous process here. (Though it is marred by a "double-voting" problem, where you get to vote against who you don't like not just vote for who you do like, with the result that average, productive, temperate editors vote for who they like and aren't apt to vote against anyone unless they have a really serious concern about that individual, while axe-grinders and PoV-pushers looking to manipulate the system calculatedly play their anti-votes and for-votes together to push something of a "party" platform in favor of their axe. It effectively gives more voting power to those with an agenda. I proposed changing this several years ago but do not seem to have at that time been able to clearly enough explain the issue or something, since the reform idea gained no traction at all. Not a problem for an RfA voting system, since it would not be a competitive race between multiple candidates, just an aye/nay vote.)  —  SMcCandlish ☏ ¢ 😼  14:48, 26 October 2023 (UTC)

@ SMcCandlish: FYI: active editors (5+ edits per month) have been flat for almost 10 years now, with a minor uptake during the height of the pandemic. If you look at user edits, there has been a rise since 2014, but I can imagine that's because of semi-automated editing rather than increased activity.

I'm not sure I understand the double voting. Does this relate to the concept of cascading bias by people following each others votes, rather than voting "independently"? I've been thinking about this through the lens of Noise: A Flaw in Human Judgment, which argues convincingly for deferred decision-making. First discuss and get all the information on the table. This ensures people don't get stuck in their knee-jerk reaction. Only after discussion would one want to let people make decisions, independent of each other in that phase. —Femke 🐦 ( talk) 16:21, 27 October 2023 (UTC)

No, and I'm not sure why I seem to have such a hard time explaining this; it's some kind of failure on my part despite a lot of experience doing things like writing documentation. Our ArbCom voting system permits you to simultaneously cast a vote in favor of one or more parties you actively support, and cast a vote actively against one or more parties you strongly disfavor. On average, an editor isn't steaming mad at any particular candidates, nor is looking to politically engineer ArbCom in a way that favors the biases of that voter. So the typical editor just votes for who they support, without casting any "hate" votes. A PoV pusher, on the other hand, has a vested interest in statistically skewing the power of their vote by voting for only the one or two candidates who support whatever bone-to-pick the editor has, and voting against every single other candidate even when they have no substantive issue with that candidate. In effect, it is casting two votes for the candidate[s] they support by recording a down-vote for every other candidate. The solution is to elminate the against-votes entirely, so that everyone just votes for who they support. As in, well, every other voting system in the entire world. But getting into any further detail about that on this page is probably pointless, since it's off topic. The problem doesn't arise for any sort of direct-voting for admins system, because the admin candidates are not running against each other but are each just a simple support/oppose.  —  SMcCandlish ☏ ¢ 😼  01:34, 28 October 2023 (UTC)

Does it have to be the biggest problem for us to try and solve it? Opposers at RfAs shouldn't be badgered because nobody should be badgered, regardless of whether it changes RfA outcomes. –  Joe ( talk) 15:15, 26 October 2023 (UTC)

As long as the community views RfA as a discussion, at least in part, I think it's not actually badgering to expect discussion. I think as with many discussions there are helpful and unhelpful versions of this and we should try to limit the unhelpful versions. But that applies just as much to personal attacks masquerading as opposes. As for Does it have to be the biggest problem for us to try and solve it? I think even if you accept we have a problem to solve that it sends a bad message to potential candidates that this is the one worth doing. Best, Barkeep49 ( talk) 15:28, 26 October 2023 (UTC)

I don't remember the last time I saw productive threaded discussion at RfA beyond perhaps a first reply. I suppose that's why I liked Ritchie's proposal. –  Joe ( talk) 05:08, 27 October 2023 (UTC)

I don't know if we need to ban replies, but @ Lourdes isn't helping things with the current one: GiantSnowman, my friend, for all the support in the past, do please reconsider. Lourdes 05:31, 24 October 2023 (UTC) and other similar. People have valid reasons to oppose, they don't need to change because you've supported them in the past. That's just straight badgering. (Uninvolved in that RfA) Star Mississippi 15:18, 26 October 2023 (UTC)

+1 why is this acceptable? Imagine if I placed a note saying "if you respect all the good work I've done in the past, please oppose" under every support !vote in the current RfA's. ~~ AirshipJungleman29 ( talk) 15:25, 26 October 2023 (UTC)

There are two instances of this in the Oppose section, and three in the Neutral section on the part of the mentioned user. It doesn't seem very appropriate, and might even be considered arm twisting. It's one thing to question a vote, but quite another to go about trying to persuade someone to change their vote in this way (at least in my view). Intothat darkness 17:14, 27 October 2023 (UTC)

I agree, especially as it's not based on merit in the slightest. I hope @ Lourdes will join this discussion or clarify their pOV. Star Mississippi 17:31, 27 October 2023 (UTC)

Just what RFA needed: a whip. Levivich ( talk) 02:20, 28 October 2023 (UTC)

Hi all: as you’ll notice on your TPs I started an AN thread about the matter which you can see here. It was quickly archived by myself after a quick discussion on request from Lourdes. Given that she has admitted to wrongdoing and GS themselves considers an AN thread unnecessary, I think there may not be much point in further discussion unless the pattern of behaviour is repeated (by the same person, or otherwise). I’d still be glad to hear your thoughts on the points I raised in the post, if anyone has anything further to say. Fermiboson ( talk) 19:32, 31 October 2023 (UTC)

While the AN closed on the early side, I think it's fair game to continue discussing it here since it's part of a pattern of issues at RFA, although not necessarily a pattern on her end that I have noticed. I'm more worried that she doesn't seem to understand why it was an issue, per my read of her "apology". Star Mississippi 01:21, 1 November 2023 (UTC)

My read on what she subsequently said is that she got emotional in the moment and now understands that she should not have brought up past interactions in the RfA comment. Otherwise, I agree, having these kinds of pathos based (or not even pathos based but entirely irrelevant pathos based) arguments in an RfA seems to be an issue that still continues and I stand by the two points I raised in the AN. Fermiboson ( talk) 02:39, 1 November 2023 (UTC)

Peeps in this thread may be interested in Wikipedia talk:Gaming the system#Proposal for an additional point.  —  SMcCandlish ☏ ¢ 😼  08:36, 1 November 2023 (UTC)

Replies to opposes are often some of the most helpful posts at RfA, right after the opposes themselves. One way I often decide how to opine at RfA is to skim the nomination, questions, and support section — enough to convince myself of the candidate's general competence, experience, and good faith — and then spend most of my time in the oppose section. I think responses to opposes in this sense are quite helpful because they define the conversation and the issues I should be spending time on. I think I'd be unhappy about scrapping them. Best, KevinL (aka L235 · t · c) 15:19, 26 October 2023 (UTC)

I agree with this entirely. Vanamonde ( Talk) 21:06, 26 October 2023 (UTC)

That's about my view of it too. ScottishFinnishRadish ( talk) 02:27, 28 October 2023 (UTC)

One idea I've been toying with proposing is, specifically, banning replies-to-replies-to-replies-to-opposes. Because this is a common enough, and pretty useful pattern:

Oppose, no GAs.

@Opposer, FYI candidate has a GA, just not listed on their userpage.

That's still a bit less content creation than I'd like to see, but on balance enough for me to switch neutral.

but if the opposer instead argues with the reply, you can bet that things are only going downhill from there. Continuing discussion will strongly tend to bring more heat than light.

But I haven't proposed that, for a few reasons, primarily that I don't think the community actually cares. The community has had ample opportunity to fix RfA and at every turn failed to do so. And I guess it's kind of been proven right in that? New adminships have slowed to a crawl and yet nothing bad has really come of that. (The only backlog that's massively worsened is WP:CCI, but most admins care about that as little as everyone else.) The community enjoys its bloodsport, and no policy proposal is going to change that. I think we all just need to accept that, if a successful RfA reform proposal ever comes, it's not going to come from anyone who watchlists WT:RFA. -- Tamzin^{[ cetacean needed} (they|xe|she) 15:34, 26 October 2023 (UTC)

I think the time to fix a capacity problem is before backlogs are everywhere and I will keep trying to do that both on s micro level (nominating people) and macro (trying to find consensus). Best, Barkeep49 ( talk) 15:39, 26 October 2023 (UTC)

Sure, that's the rational time to do it. But pragmatically, the community does not want RfA reform, and that isn't going to change until there's a problem to fix that affects the average community member (rather than just affecting candidates and the occasional nom or voter). -- Tamzin^{[ cetacean needed} (they|xe|she) 15:45, 26 October 2023 (UTC)

What makes you say the community does not want RfA reform? I think there is a difference between nobody making time to try and get the community on board with a proposal, and there being no appetite to engage in that discussion. The Wikipedia:Requests for adminship/2021 review had quite strong consensus for statements in the vein of "RfA is broken", and a few solutions came close to consensus. They might reach consensus if we can workshop smart amendments to these proposals, possibly after securepoll is available locally. —Femke 🐦 ( talk) 20:10, 26 October 2023 (UTC)

There are backlogs everywhere. Not all of them quantified and categorised, and not all of them admin-related. I've been worried recently about the number of revert-on-sight edits I've been finding 6 days down in my watchlist, where I would have expected 5 years ago to have been beaten to the punch if I'd reverted it at the 10 minute mark. More mops helps all anti-vandalism backlogs as it allows experienced editors to save time elsewhere in their workflow (you don't have to ask someone else to push a button and then come back to check they've pushed it and continue the next steps). I can only hope my small piece of the puzzle is unrepresentative. — Bilorv ( talk) 23:33, 27 October 2023 (UTC)

@ Tamzin I personally think the backlog at CCI, even if it the largest it's ever been, is in one of the better state's it's been in; in the mid 2010s, it became neglected after several admins stopped editing in the area and several long term cases and editors with several warnings continued unabated. Now and days, at least people are aware about it and care about it and running on a copyright platform is a legitimate one. Hopefully things can change in the area and the backlog can shrink with an RfC that I've been working with on and off. Moneytrees🏝️ (Talk) 15:49, 26 October 2023 (UTC)

I've been in favour for a long time of aggregating discussion in one place, in order to make the discussion more efficient (no need to track discussions on the same underlying issue across multiple threads). As discussed earlier this year, though, I appreciate that so far, amongst those who like to weigh in on these proposals, more people prefer reading inline threads, in spite of the redundancy.

To provide context for those who did not participate in a related thread in August, the WMF is working on enabling SecurePoll votes to be administered locally, which is the primary bottleneck in having such votes run more often. However the WMF first plans to update SecurePoll to no longer rely on an older version of GPG, to facilitate deployment on the local wiki installations. I agree that establishing the community's intent to run local SecurePoll votes once possible would help the developers decide on when to schedule its work. isaacl ( talk) 16:57, 26 October 2023 (UTC)

The RFA SecurePoll thing is pretty stuck. The last 72 supporting, 39 opposing RFC on it was closed as no consensus. And the phab ticket hasn't had a post in 5 months. If we feel confident that secret RFA voting is the future, the next step would probably be a new RFC. Not sure how I feel about it nowadays though. The current system is tough but transparent. – Novem Linguae ( talk) 20:11, 26 October 2023 (UTC)

Yes, I agreed that establishing the community's intent to use SecurePoll would help in driving progress. As I mentioned in the August thread, the Phabricator ticket is blocked on phab:T209892, though that one too has not been updated since May. isaacl ( talk) 21:25, 26 October 2023 (UTC)

@ JSutherland (WMF): not sure who's ultimately responsible for this phab ticket, but are you perhaps able to give us an update or point to the correct person? Is there work planned on this or would we need to ask at the Community Wishlist or show consensus for need at an RFC? Thanks :). —Femke 🐦 ( talk) 16:25, 27 October 2023 (UTC)

We're working on a way forward with that - updates will be on Phab :) Joe Sutherland (WMF) ( talk) 16:49, 1 November 2023 (UTC)

• As with BK49 above, I don't see how giving opposers more free rein is in any way going to make RFA more friendly to candidates; if anything, it will have the opposite effect. If the goal is to encourage good candidates who are wary of RFA, we need to be able to call out the not-infrequent oppose votes that are unmoored from reality. I recognize that a large part of the community finds the cycle of "oppose, no FAs" "but the candidate has 2 GAs" "please stop badgering me" "I'm not badgering, I'm providing necessary context" etc. very annoying, and I welcome attempts to pre-empty this repetitive conversation, but we ought to recognize that it's quite separate from the broader issue of making RFA less hostile to candidates. Vanamonde ( Talk) 21:15, 26 October 2023 (UTC)

  It's reasonable for commenters to discuss their viewpoints on a candidate's characteristics. It would be more efficient, though, to have that conversation once, rather than in multiple threads in response to multiple people. I also think it's less draining on the candidate to have consolidated threads, instead of seeing the same conversation play out in several locations. isaacl ( talk) 21:30, 26 October 2023 (UTC)

  Efficient only if we entirely segregate votes from discussion; that is, disallow vote rationales as well (which, to be clear, I don't support). Otherwise, we're splitting statements and replies; I don't see how that's more efficient. Vanamonde ( Talk) 22:38, 26 October 2023 (UTC)

  It's not necessary to disallow vote rationales or to separate them. Aggregated conversation about a concern raised by multiple participants can take place separately. Today, someone trying to catch up on a request for administrative privileges must revisit all of the viewpoint statements to look for updates (either through a diff view or on the page itself), in addition to the discussion section, and the conversation will often be repetitive. By separating discussion, they would only have to read the latest viewpoint statements, then see the updates to the discussion section.

  It's not unusual for those making decisions to identify considerations, and then explore each consideration, one at a time. This helps provide an overview of relevant factors, followed up by a more through examination of each. In the context of RfAs, extended discussion is analogous to the followup exploration of relevant factors. isaacl ( talk) 23:02, 26 October 2023 (UTC)

  If anything's unmoored from reality, it's the assumption that the average oppose vote is frivolous and the voter deserves to be harangued for it (to the candidate's benefit, somehow?) Let's just look at 0xDEADBEEF's ongoing RfA (which for the record I haven't voted in). Over a third of the oppose votes have been replied to, compared to 5 of the 166 supports – and several of those five are indirectly mocking opposers. The size of the threaded discussion (2,497 words, 14,823 characters) exceeds that of the all the oppose votes (2,164 words, 12,831 characters). One thread has been moved to the talk page. There is an admin begging multiple oppose voters to change their votes as a personal favour and another admin accusing an oppose voter of making a personal attack. And all this despite the fact that the RfA is comfortably in the automatic pass range!

  This is not productive discussion. It's circling the wagons, punishing people for wrongthink and turning the invitation at the top of the page, voice your opinion on this candidate, into a trap. We do not need more "calling out" around here, not at RfA, not anywhere. –  Joe ( talk) 06:07, 27 October 2023 (UTC)

  I tend to agree with all of that, Joe Roe. While L235, further above, has a good point with "Replies to opposes are often some of the most helpful posts at RfA, right after the opposes themselves", I feel this has been less and less true of late, with too many of the replies to opposes being rather "content-free" and just emotive browbeating.  —  SMcCandlish ☏ ¢ 😼  07:08, 27 October 2023 (UTC)

  As are some oppose rationales. It is unfortunate that we ask for those; people who oppose because they don't like the candidate or the candidate's nominator should just vote and not tell us their personal reason, which is unlikely to contribute to a discussion about the candidate's merits. — Kusma ( talk) 07:23, 27 October 2023 (UTC)

  Where do I make that assumption, Joe? I said it was frequent, and it is. Plenty of oppose votes, including hundreds I've disagreed with, are helpful contributions to a discussion of the candidate's skills and judgement. Plenty of others are violations of WP:ASPERSIONS that we permit only at RFA: I've seen comments made there that we'd consider sanctionable conduct in a CT area. And it is this harshness is one of two major reasons why dozens of candidates I've approached decline to run at RFA, the other being disinterest in admin work. Vanamonde ( Talk) 16:00, 27 October 2023 (UTC)

I believe we should have an RfC on the following question:

Should the following question be asked at RfA?

Do you have the intention of starting a paid consulting business after your RfA advertising yourself as an administrator? Assuming you do not now have such intentions, are you willing to have a confirmation RfA explicitly mentioning your intentions if you decide to do so later?

I do not wish to add to anyone's stress by asking it of the three individuals currently running. However, given the precedent recently set, and the fact that the admin has not agreed to a confirmation RfA... perhaps it would be wise to ask from now on? -- SashiRolls ^{🌿 · 🍥} 18:47, 28 October 2023 (UTC)

That is so overly-specific as to be basically useless and is firmly in the realm of closing the barn door after the horse has bilted. If you want to disallow that sort of behavior, change the policies around adminship, don't shoehorn it in with (non-binding) questions at RfA. GeneralNotability ( talk) 20:09, 28 October 2023 (UTC)

I appreciate the response. My fear, of course, is that bringing new horses into the stable after the trick to open the barndoor was shown wouldn't be wise. Thankfully, admin candidates are not horses, they are sophisticated language users who can agree to reconfirm before starting a business. But yes, I suppose it would be more straightforward to try to modify WP:ADMIN to say that an admin cannot advertise their adminship to attract customers for paid consulting unless they disclosed their intention to do so at their RfA. Does this wording seem fair? -- SashiRolls ^{🌿 · 🍥} 20:30, 28 October 2023 (UTC)

There's a ridiculously long RfC on this subject at VPP which currently has more opposition than support. Please don't relitigate it here. At some point, we either trust our admins to do the right thing or we don't have admins. HJ Mitchell | Penny for your thoughts? 20:47, 28 October 2023 (UTC)

The RfC at the Village Pump asks a totally different question about disclosing clients, it did not ask about transparently disclosing one's intentions to advertise at RfA . -- SashiRolls ^{🌿 · 🍥} 20:50, 28 October 2023 (UTC)

I am the administrator that the SashiRolls alludes to. As for the admin has not agreed to a confirmation RfA, I have repeatedly said that I will fully comply with any new policy, so I consider a "confirmation RFA" to be a red herring. Cullen328 ( talk) 20:56, 28 October 2023 (UTC)

I was going to say moral panic. And forum shopping. HJ Mitchell | Penny for your thoughts? 21:00, 28 October 2023 (UTC)

Also, I did not begin doing any Wikipedia related consulting work until more than five years after my RfA. I was working full time doing strenuous construction work back then. I am now semi-retired and do only a few hours of office work these days. Unless the policy changes, I have the right to engage in off-Wikpedia consulting work that violates no policy. As a sign of good faith, I stopped accepting new clients when the discussion at WP:VPP began six weeks ago and I have no active clients. Cullen328 ( talk) 21:06, 28 October 2023 (UTC)

Yeah, someone should really close that as no-consensus, it's been open forever. It's not fair to you. That said, a more straightforward RfC should be opened concerning disclosure. This isn't about you, but about what you pointed out. Not everyone admin candidate is Cullen328 (or Eostrix). ps: no need for a definite article before my username, of course. You can just call me Sashi. -- SashiRolls ^{🌿 · 🍥} 21:24, 28 October 2023 (UTC)

At some point, we either trust our admins to do the right thing or we don't have admins. " Trust, but verify", to quote a Russian proverb. Levivich ( talk) 16:21, 2 November 2023 (UTC)

Just a typo, Sashi. Cullen328 ( talk) 21:28, 28 October 2023 (UTC)

I echo the thoughts of GeneralNotability above about this being an unnecessarily granular question. That in itself would be forgivable if the question were neutral. Frankly, if one is predisposed to think that paid-anything behavior is undesirable in an administrator, then a simpler version of the above question would be: "Do you promise to renounce your adminship if you take money for any reason?" _{Duly signed,} ⛵ WaltClipper -( talk) 13:53, 29 October 2023 (UTC)

I agree with HJ Mitchell. The proposal was brought at VPP, after considerable discussion, it has (to my eyes anyway) failed. Bringing it here does not respect that result. Sometimes you have to respect you've received no for an answer and resist the temptation to find another forum to shop.-- Wehwalt ( talk) 14:05, 29 October 2023 (UTC)

You admins are absolutely kidding yourselves if you think Barkeep's RfC is going to be the one and only time we talk about this. It's not forum shopping for another editor to propose something else after a no consensus result. Levivich ( talk) 14:11, 29 October 2023 (UTC)

Well, in this case, unless I've missed something, you're not even waiting for the no-consensus close to occur. I guess forum shopping is just one of those things, to quote Potter Stewart, where " I know it when I see it". Wehwalt ( talk) 14:14, 29 October 2023 (UTC)

Walt, stop it. Levivich ( talk) 14:18, 29 October 2023 (UTC)

Pardon? I didn't say anything. _{Duly signed,} ⛵ WaltClipper -( talk) 16:39, 29 October 2023 (UTC)

I assume I am the person meant. Wehwalt ( talk) 16:52, 29 October 2023 (UTC)

Wehwalt's point is valid, and I would have thought that Levivich would have learned by now that this tiresome "[Username], stop it" posturing doesn't work. Pretending one has been unfairly maligned and that someone else is doing a wrong by being critical, when one has actually been entirely fairly and reasonbly criticized, fools no one at all. Ordering other users to stop raising concerns is never going to do anything but backfire.  —  SMcCandlish ☏ ¢ 😼  17:03, 29 October 2023 (UTC)

Ordering users to stop raising concerns is what's happening here, with multiple admins calling the raising of concerns "forum shopping."

Let's get real, folks. The RFC, which AFAIK had no RFCBEFORE discussion, asked a question about disclosing clients of paid admin consultants. That's a pretty narrow question and has led to a split decision, almost evenly split, with a number of editors saying it goes too far, or not far enough, that it shouldn't be about disclosure but about the paid advising, or that it should apply to everyone and not just admins. Lots of opinions from lots of different angles.

Obviously, there will be continued discussion of this issue. It is not forum shopping for someone to suggest something somewhere. It is bad for admins to suggest that editors continuing this discussion are forum shopping. That's trying to stop people from talking about it, and that's bad.

So stop it, walts and non-walts alike. :-P Levivich ( talk) 17:13, 29 October 2023 (UTC)

Indeed. In no sense is it "forum shopping." And it is poor conduct by administrators to try to shut the conversation down, since they are so obviously compromised by the act of selling adminship on the open market, without disclosure. Alanscottwalker ( talk) 17:25, 29 October 2023 (UTC)

Which admins in this discussion are compromised "by the act of selling adminship on the open market, without disclosure?" Wehwalt ( talk) 17:26, 29 October 2023 (UTC)

You. And the others. The effect of your opposition and attempt to shut down this discussion is to preserve for admins the "right"! its been called above to advertise their advanced permissions in sales pitches, without disclosure. Alanscottwalker ( talk) 17:33, 29 October 2023 (UTC)

I'm all for discussion towards crystallising consensus and finding a better question to ask, but the question originally asked here was an end run around the RfC, even if it wasn't meant to be. I still think we're missing the point; problematic edits should be dealt with as such, renumerated or otherwise. HJ Mitchell | Penny for your thoughts? 17:27, 29 October 2023 (UTC)

It's not an end run around the RFC, or even close to it. The RFC asked about disclosing clients. Sashi's question asks about engaging in paid consulting at all -- it does not ask about disclosing clients. When I say "you're kidding yourselves," and "let's get real," I'm talking specifically about this: if you think an RFC about disclosing clients is going to be the end of this discussion -- and we're not going to talk about whether anyone can do this paid advising at all, and if so, under what restrictions/disclosure requirements/whatever, whether admin or non-admin -- you're totally being unrealistic. Of course we're going to talk about that. We're going to talk about how WP:PAID has a loophole where if you don't actually make the edits, you aren't technically covered by it. We're going to talk about whether that's a loophole at all -- an unintentional gap -- or whether we intend to allow paid advising. We're going to talk about whether the rules should be different for permissions holders vs. everyone else. We probably won't talk about disclosing clients, but the rest of it? Still very much an open question, I am sure I'm not the only editor who wants to see it resolved. This controversy isn't about disclosing clients, it's about the activity itself. Levivich ( talk) 17:37, 29 October 2023 (UTC)

Which rather begs the question, if it's not disclosed, how will we ever know it took place? A cynic would suggest there isn't a problem (speaking as an admin who has blocked literally thousands of spammers and promotional editors, disclosed or undisclosed). HJ Mitchell | Penny for your thoughts? 17:44, 29 October 2023 (UTC)

How do we know what takes place on the internet, takes place? That should be obvious. -- Alanscottwalker ( talk) 17:54, 29 October 2023 (UTC)

@ HJ Mitchell, (I hope I'm not misunderstanding you), but that's true of any paid editing. The difference is that in the case of paid editing, it's often obvious. In the case of paid advising, not so much. And paid advising by a highly experienced editor, maybe even less so. We need to have disclosure so we know which articles to keep an eye on. Valereee ( talk) 20:24, 31 October 2023 (UTC)

Yes, of course it's true of any paid editing. The whole concept of UPE was another moral panic whipped up by a handful of editors. All it's led to is an unhealthy focus on editors' motives and identities. I was blocking spammers years before disclosure was required and I still block spammers now; I don't care why they're spamming, whether they're paid to do it, or whether they've disclosed that fact. Requiring disclosure has done absolutely nothing for Wikipedia except create another bureaucratic burden. This proposal will do the same but only increase suspicion and encourage people to dig into admins' off-wiki activities. I'd like to see more Cullens and fewer Orangemoodys but these proposals will have the opposite effect. HJ Mitchell | Penny for your thoughts? 20:52, 31 October 2023 (UTC)

That can't be right. Readers should be entitled to be able to find out if the Greenpeace article is written by Greenpeace, or the Tobacco Industry writes the Tobacco article, or contributed to it, if that is what has happened. It is just plain ethics. Alanscottwalker ( talk) Alanscottwalker ( talk) 08:45, 1 November 2023 (UTC)

There is absolutely no way to make that happen. We do not ID-verify our editors, many of whom are anonymous. What you're posting here and in your rant below are fanatasies about a utopian pipe-dream world. Not constructive in a discussion like this about what to actually do and get done.  —  SMcCandlish ☏ ¢ 😼  09:26, 1 November 2023 (UTC)

Of course we make that happen, to the extent we can, it's one of the reasons why we ask for and expect disclosure. And no, ethics in publishing is no fantasy and neither is asking for it, not in the real world. Alanscottwalker ( talk) 09:41, 1 November 2023 (UTC)

"Asking for and expecting disclosure" does not make us "able to find out if the Greenpeace article is written by Greenpeace, or the Tobacco Industry writes the Tobacco article, or contributed to it". I may ask for and expect my car to not be broken into, but this has no effect whatsoever on whether a criminal will do it. You are totally confusing the idea "I wish something was a particular way" and "We can make something be a particular way". They are not the same.  —  SMcCandlish ☏ ¢ 😼  19:10, 1 November 2023 (UTC)

You seem totally confused about criminals, we are talking about publishing, not criminals. And you are totally confused about what writing Wikipedia policy is, "doing" in terms writing Wikipedia policy is almost entirely writing requests and guidance. And it is plain obvious that a COI disclosure on an article talk page which we ask to be done informs readers of the COI, so there is no doubt it informs the reader. You seem to be in a fantasy world that Wikipedia could be some kind of autocratic state, and only that's doing something, no it is not, on either count. -- Alanscottwalker ( talk) 19:31, 1 November 2023 (UTC)

If you are or are going to pretend to be incapable of understanding analogy, and if your habitual response to a criticism is to flip it around in a childish "No I'm not! You are!" manner, and make up weird bullshit about autocracy, then I have no further reason to continue reading a thing you say or taking anything you propose seriously, nor does anyone else. > plonk<  —  SMcCandlish ☏ ¢ 😼  20:35, 1 November 2023 (UTC)

Ah well, I'm certain that's not true, nor even remotely fair. -- Alanscottwalker ( talk) 20:50, 1 November 2023 (UTC)

I guess while I'm here I should address the OP question. Personally, I think it's a bit premature to propose any specific solution to the perceived problem, including the potential solution of an RFA question or RFA disclosure requirement. After reading the responses in the client-disclosure RFC, I'm not sure that this is an "admin issue" as opposed to a general editor issue, and hence I'm not sure that RFA questions are the place to make tweaks as opposed to elsewhere.

What I think would be helpful as a next step is a true "request for comments" -- not a support/oppose vote, but just asking people to give their opinions, on some discrete questions (which should be workshopped in an RFCBEFORE, after the current RFC is closed or archived without closure), such as whether paid advising should be allowed at all; whether paid advisors should have to disclose anything and if so, what; whether paid advisors should be under any restrictions and if so, what; whether the rules should be different for perm holders, and if so which perm holders (there are all the functionary hats to consider, some of whom signed NDAs, and does that matter?), and if so, different how. Only after getting the community's feedback on these basic questions can we start to think about whether we need to make any changes to policies (WP:PAID, WP:COI) or procedures (RFA questions). I have more questions than answers on these issues right now, and I think it's too soon to talk about solutions, since we don't yet know what we all think is, or is not, a problem.

I say after this RFC is closed because if someone uninvolved volunteers to write a good closing statement that summarizes the various arguments/counterarguments made in the current RFC (I hope someone does!), that may help narrow and clarify the issues, and thus inform an RFCBEFORE for another RFC, or whatever the next step may be. Levivich ( talk) 17:58, 29 October 2023 (UTC)

Every editor in good standing is permitted to ask questions at RfA; any editor who thinks that this would be a useful question to ask can always ask it even if we do not institute it as a standard question. I would be very surprised if it turns out to be a useful question: even if someone does answer no/yes and then begins a business advertising themselves as a Wikipedia admin, what recourse would the community even have? Would the precedent that administrators are not bound by their WP:RECALL pledges apply here? Caeciliusinhorto ( talk) 08:20, 30 October 2023 (UTC)

Part of the purpose of asking questions is to get an idea how someone thinks and how well they express their rationale -- should they change their mind, the community can do what it always does initially, ask more questions, and operate with open information. -- Alanscottwalker ( talk) 11:49, 30 October 2023 (UTC)

It also provides a vector for opposition if a candidate does not give the "ideal" answer of saying "No. I have no intention of starting a business", and perhaps it's just me, but personally I tend to shy away from any added steps or measures that would make RfA more difficult than it currently is. Giving the indication that Wikipedians will want to pry into someone's livelihood, private or not, before they will hand someone the "no big deal" mop is just going to make our admin recruitment problem worse. One might argue that this is a Good Thing, because it would deter unscrupulous admin candidates, but note also that there are people who are greatly concerned about their privacy on the Internet in this day and age, whether or not they have a side gig. _{Duly signed,} ⛵ WaltClipper -( talk) 12:35, 30 October 2023 (UTC)

A central purpose of RfA is to ask questions. It seems bizarre to criticize questions and answers because they inform the ivoters. What we want to know is what they will do with their admin-ship that they are asking us to give them, if they have a livelihood plan for it, or would consider one, they should tell us. And any potential candidate who does not already know they will get all kinds of scrutiny, welcome and not, and it is and will be ongoing, should stay away because they are neither well informed nor prepared (just as in arbcom candidates and members) . Also, it often seems that some who link 'no big deal' don't know that the purpose of that section is to inform that standards, expectations, scrutiny have increased overtime. -- Alanscottwalker ( talk) 13:51, 30 October 2023 (UTC)

I'm fully aware of that. In fact, my linking it is because of the fact that standards have increased exponentially, and in some ways the increase has been beneficial, but I think now it is truly becoming a detriment. Why are we continuing to add more hoops to jump through when we're already dealing with a retention issue? It seems to me that assessing the candidate's judgment specifically for their on-Wiki activities is the most important thing, and not what they do in their outside life.

I would also add that I have no problem with the question itself, but making it mandatory is unnecessary. The "are you open to recall" question isn't mandatory but it's essentially guaranteed and automatic anytime an RfA is opened. _{Duly signed,} ⛵ WaltClipper -( talk) 14:38, 30 October 2023 (UTC)

Informing others, and willingness to inform others, is an on wiki activity. Alanscottwalker ( talk) 14:44, 30 October 2023 (UTC)

And I guess one other thing I would add is this thread opened with the OP stating: I do not wish to add to anyone's stress by asking it of the three individuals currently running. It seems like an implicit recognition that the environment at RfA is about as stressful as going through an Arbcom case as the sole named party, and this does nothing to help that. _{Duly signed,} ⛵ WaltClipper -( talk) 14:41, 30 October 2023 (UTC)

For what it's worth, my views on the matter align with Levivich in that I'm open to any neutrally-worded RfC that addresses the larger question in the context of editing on Wikipedia as a whole, rather than just having a mop-and-globe icon on one's Talk Page. _{Duly signed,} ⛵ WaltClipper -( talk) 12:36, 30 October 2023 (UTC)

Sure. I think that yes/no questions tend to be a pretty ineffective way of getting a good idea how somebody thinks but as I say anyone who thinks it valuable can always ask Caeciliusinhorto ( talk) 18:50, 30 October 2023 (UTC)

Well, it seems obvious from the discussions that it is an admin issue, and that is true even if it is an 'other editor's too, issue', admins are the one's who are very visibly seeking and holding permissions from us. But personally I would not so much focus the questions on reconfirmation, I would focus on willingness to be up-front with us, now and in the future and probe disclosure and its extent. Alanscottwalker ( talk) 14:16, 30 October 2023 (UTC)

I'm not sure it should be closed as no consensus simply on numbers. Many of the opposes are arguing that because it doesn't include non-admins, it doesn't go far enough. Valereee ( talk) 19:09, 31 October 2023 (UTC)

Really? Which ones other than Oppose #14 (Cryptic) state that they are opposing only because the proposal was to apply to admins only and would support if the requirement was applied to all editors? The raw vote at present appears to be 42 in favor and 49 opposed, and I see only one oppose along the line you suggest. Perhaps I've read in haste. Wehwalt ( talk) 19:31, 31 October 2023 (UTC)

Sorry, not sure which opposes and discussion made me comment here about multiple such, it was five weeks ago and I don't recall. Valereee ( talk) 19:40, 31 October 2023 (UTC)

I do recall @ Roy Smith addressing it somewhere, but can't remember the exact. Valereee ( talk) 19:42, 31 October 2023 (UTC)

Roy is supporter #38. I do see a few opposers saying that any such restriction should not apply just to admins, but only Cryptic so far do I see saying they would support if it applied across the board. If we counted them as a supporter at heart, that would make it 43–48. Wehwalt ( talk) 19:50, 31 October 2023 (UTC)

I'm gonna bet you didn't see Cryptic saying he would support otherwise. — Cryptic 20:06, 31 October 2023 (UTC)

Sorry if I misunderstood. Wehwalt ( talk) 20:11, 31 October 2023 (UTC)

But wouldn't anyone saying that a restriction shouldn't only apply to admins are basically saying it should be applied to both? Valereee ( talk) 19:54, 31 October 2023 (UTC)

If they are otherwise supporters (Cryptic) but not if they oppose also for other reasons. Wehwalt ( talk) 19:57, 31 October 2023 (UTC)

Meh. Maybe? Valereee ( talk) 20:06, 31 October 2023 (UTC)

Maybe. I seem to remember somewhere in the volumes of discussion someone making a point similar to yours but I'm not sure it's supported. Wehwalt ( talk) 20:12, 31 October 2023 (UTC)

But I'm supporting it so beautifully. :D Valereee ( talk) 20:20, 31 October 2023 (UTC)

I can't argue with that! Wehwalt ( talk) 20:44, 31 October 2023 (UTC)

Might as well just ask, Do you have the intention of becoming an evil administrator after your RfA? Assuming you do not now have such intentions, are you willing to have a confirmation RfA if you decide to turn evil later?, if we're going to add very specific scenarios to the default questions or otherwise assume the worst of every candidate. Acalamari 06:53, 31 October 2023 (UTC)

@ Acalamari, I kind of feel like that's not fair. There is significant difference of opinion on whether or not paid-but-not-fully-disclosed advising is okay. Clearly Cullen, a highly-respected editor and admin, sees this as okay, not evil. So he wouldn't have answered, "Yes, I intend to become evil". And it seems like you don't see this as evil, either, or even problematic. So why would you think asking about it would mean asking if one intends to become evil?

To be clear, I don't think this is evil. I think it's problematic for an admin to be marketing their services, as an admin, to advise without disclosing the articles in question corporations/people who would like to exert control over the articles about them, because it doesn't let other editors know which articles someone should maybe keep an eye on. Which is why I supported the RfC. Valereee ( talk) 19:22, 31 October 2023 (UTC)

This comparison is unfair. Advice is not evil.

A fair comparison would be "Who are all of your past and present employers (and, if applicable, customers)? After all, you have a financial COI with them, so we have to make sure just in case you someday decide to edit their articles." — Cryptic 20:10, 31 October 2023 (UTC)

• These kinds of invasive questions based not on anything related to policy but on assumptions of bad faith (or sometimes just plain pet peeves), are IMO among the chief problems with RfA -- not badgering opposes, not pleading with opposes, not opposing because of something done on-wiki, not differing opinions about what on-wiki activity makes for a good admin, but the paranoid questions about real-life identities, real-life activities, etc. I won't repeat all my arguments from the disclosure RfC, but when such a question is posed here, it throws a "nothing to hide" wrench into the RfA works -- either you have to make commitments that you shouldn't have to commit or you leave people to assume the worst. — Rhododendrites ^talk \\ 20:02, 31 October 2023 (UTC)

  I'm not sure it's any more assuming bad faith to require paid advising be fully disclosed than to require paid editing be fully disclosed. Valereee ( talk) 20:10, 31 October 2023 (UTC)

  The analogy fails because paid editing is an on-wiki activity -- one for which we have a policy founded in a very strong consensus. It's asking an on-wiki admin about whether their on-wiki editing is influenced by money. It isn't asking them about off-wiki activities that do not affect their on-wiki editing. It can only conceivably influence their on-wiki activity if you assume bad faith, and if it doesn't influence their on-wiki activity it's as equally none of your business as asking who their employer, spouse, or friends are. But now I'm just rehashing the RfC, so this is my only response. — Rhododendrites ^talk \\ 20:42, 31 October 2023 (UTC)

  The analogy does not fail in the least. The private Wikipedia advisor is not advising about what's off Wikipedia, they are advising about what's on it, its content, its policies, it modes of behavior. Indeed, it's been billed by many and in some kind of triumphal manner as paid advice on how-to-act, write, participate on Wikipedia. That can be nothing but influencing Wikipedia. (Also, it was just this kind of parade of horribles argument you make that in part prevented PAID disclosure for so long, so your comment has an unintended irony in the way you talk about PAID) Alanscottwalker ( talk) 07:43, 1 November 2023 (UTC)

The questions I proposed for RfA and the proposed modification to WP:ADMIN are agnostic with regard to whether paid advising should be permitted or not. What I want is for the community to be able to vote knowing the intended use of the honorary title, and to know that the person involved is aware they need to ask at RfA before advertising the title. That's all. If the community supports a particular admin doing paid advising that's fine, that's consensus. Concerning the two caricatures of my question—Do you promise to renounce your adminship if you take money for any reason? and Do you have the intention of becoming an evil administrator after your RfA?—people can draw their own conclusions as to why the caricaturists have chosen to reframe Do you promise to ask us before you advertise your admin status?. Are the cartoonists revealing truths about RfA voters? assuming bad faith about them? Surely not. Again, assuming some business competence, none of this requires any disclosure whatsoever about identity. Finally, and parenthetically, as for this distinction between the "real world" and a "make-believe world", hmm... I know it's cherished wiki-wonkery not to think so, but en.wp exists in the RW :) -- SashiRolls ^{🌿 · 🍥} 21:23, 31 October 2023 (UTC)

Perhaps it is you who is assuming bad faith, but of the entire admin corps? We elect admins because we trust them to act in the best interests of the encyclopaedia. If an admin loses that trust, they shouldn't be an admin. Absent evidence that they have abused that trust, we should assume that they will continue to act in the best interests of the encyclopaedia. These are people who have invested thousands of hours into Wikipedia; they're not going to throw that away for a quick buck. People will continue to solicit the kind of advice Cullen has been giving, regardless of our policies, which just leaves the question of whether we want them to get it from genuine good-faith editors like Cullen or the likes of Orangemoody (who, by the way, will continue to pass themselves off as admins, because the public don't know the difference). HJ Mitchell | Penny for your thoughts? 21:54, 31 October 2023 (UTC)

Are there rules against accusing people of assuming bad faith of the entire admin corps without any evidence? I think there are. -- SashiRolls ^{🌿 · 🍥} 22:09, 31 October 2023 (UTC)

What I want is for the community to be able to vote knowing the intended use of the honorary title, and to know that the person involved is aware they need to ask at RfA before advertising the title. If that's the intent of your question, then the question should be so appropriate. Something such as: What sort of weight or status do you believe that the role of administrator carries outside of Wikipedia? is useful because, like a job interview, it's not intended as a "gotcha" question. Instead, it informs as to the administrator's ethics, morals, and general behaviors, which to me is more important than whether or not they answer "I have no intention of starting a paid business". That, while being the Most Desirable Answer, tells you nothing in the moment as to whether the admin candidate in question will be honest or trustworthy. It's only really usable as evidence to beat someone down after the fact if they do get caught lying.

Beyond that, the caricature exists because of the fact the question has... well, if not loaded language, then more some of the concepts being exercised. Such as a confirmation/reconfirmation RfA for example - has this ever really been considered on Wikipedia as a mechanism that can be exercised to reassess an admin's good standing? Generally, any reconfirmation process that exists is usually the result of an admin being de-sysopped by Arbcom and notified they can only regain adminship through RfA, so there is an inherently negative connotation there. It's hard not to view the question as anything other than adversarial. _{Duly signed,} ⛵ WaltClipper -( talk) 02:33, 1 November 2023 (UTC)

I see what you're getting at with the idea of posing a more open and less pointed question, but if someone asked me What sort of weight or status do you believe that the role of administrator carries outside of Wikipedia? I would have to ask them to restate the question, since it doesn't have clear/certain meaning. Strictly interpreted, I think the only proper answer is a bare "None.", but that is probably not the kind of answer someone asking such a question is looking for.  —  SMcCandlish ☏ ¢ 😼  07:04, 1 November 2023 (UTC)

And I think that would be reflected in the !voting and/or commentary. Rarely for these types of open-ended questions will you see a candidate confine themselves to a one-word or one-sentence response, because they understand that their behavior and answers will be scrutinized thoroughly, or in particular, certain !voters will hinge their support or opposition based on the answers, or lack thereof, to their optional questions. It's their Big Day. They want to provide as much of an opportunity as possible to market (for lack of a better word) themselves to Wikipedians, so they will be as detailed as they have to be.

But yes, @ SMcCandlish, I'd like to push us more in the direction of brainstorming open-ended questions that are searching or revealing. You could also make it a two-part question by asking them what sort of weight it carries both on and off Wikipedia, to get their overall take on adminship, and provide clues as to whether or not they may try to embellish or exploit it. If the concern is one of ethics and morals, which based on Alanscottwalker's statements such as because paying for what is free is a form of corruption I have reason to believe that it is, then the question needs to be revealing of the candidate's ethics and morals.

Now, this is probably going to be a highly cynical outlook, but frankly: a yes/no question about a specific troublesome scenario can so easily be gamed, because why on earth would any candidate who wants to become an administrator give an affirmative answer that could potentially garner opposition to a question which provides no room for nuance? Because that's exactly what will happen, there are people who have direct moral opposition to paid activities no matter who does it, based on the number of people who said they "lost respect for Cullen" during that RfC. And even if the candidate said no, how would we be able to determine if they were being sincere, without crossing the WP:OUTING threshold? _{Duly signed,} ⛵ WaltClipper -( talk) 13:44, 1 November 2023 (UTC)

I've mentioned my being an administrator on Wikipedia a few times. People don't really understand what that is, but I shine slightly by the reflected glory of Wikipedia. That's not for me. So I don't mention it. Wehwalt ( talk) 14:08, 1 November 2023 (UTC)

Someone above termed this a moral panic, and I have to agree. I advise people about editing all the time (primarily with two goals: reducing the "doing it all wrong" issues for the community of a noob not knowing anything, and reducing the "I'm gonna quit in disgust" frustration of meeting revert-happy resistance when doing well-meaning things that don't quite comply with our rules yet). No one's paid me for any of it, and none of them have been corporate entities. But if someone tipped me $20 for having helped them out I wouldn't suddenly feel "evil". Every time someone writes an article on their blog or social media about how to edit WP properly, they are "advising" a whole bunch of random strangers on how to edit Wikipedia. Every time we write explanatory essay material here on-wiki, we're doing likewise. Every instructor in a classroom-based, WMF-endorsed project to have students improve Wikipedia articles as part of their coursework is a "WP-editing advisor" (and paid). Everyone organizing editathons and showing noobs the ropes is such an advisor (and sometimes might be paid, especially under various affiliate programs that internally arrange their own minor stipends). If someone gets a grant from WMF to go to a wiki-conference and give a presentation on some aspect of editing, that's paid advising about WP editing. This panic here seems to be that someone might get remuneration to advise some particular celebrity or corporation or some other potential or actual article subject on how to edit here, but there is no definitional bright line. One of the inviduals I advised is a board member of or otherwise involved in various organizations, and working at a for-profit publisher, and concerned about lack of coverage of some of their professional collegues in their field; their editing desires are across a fairly discrete topic area, but all of those entities fall within that area's scope, and they're all potential CoIs. But CoI editing is not forbidden, and if it's going to happen there are ways to go about it that don't break things (especially: proposing new content or specific changes, with a pile of sources for it, and convincing non-CoI editors to do the actual writing, excercising a lot of NPoV judgment along the way). I was not in any way wrong for advising this person on how to do this properly. There seems to be a "folk belief" around here that advising some entity with a vested interest in something automatically means you are advising them on how to violate our rules instead of how to comply with them, and this is of course a terriblizing legend, a superstition.  —  SMcCandlish ☏ ¢ 😼  05:19, 1 November 2023 (UTC)

I've turned down Wikipedia-related consulting job offers several times in the last few years because I thought that would make me a WP:PAID editor, and I'd like to know what the rules are around this because maybe I won't turn it down next time. No moral panic here, just want clarity. Levivich ( talk) 05:53, 1 November 2023 (UTC)

These make-weight claims of moral panic and ABF are absurd, and merely extensions of seeking to shut down discussion. And that's when some of the ABF claims don't cross the line of personal attack and aspersions. There is nothing automatic, except the central value of transparency. Transparency in the how and the way of making Wikipedia is central to what Wikipedia is. If paying admins for private benefit is going into the making of the editorial decisions of Wikipedia (not publishing is as much an editorial decision as publishing), we should be up front about it as possible. If paying admins is going in to how editors and potential editors behave or act, we should be as up front as possible about it. If paying admins effects or influences policy choices (eg. if what's important is to protect the private profits they make off Wikipedia), we need to be upfront about it. (This too is how Wikipedia is made and unmade.) That the information they are paying for is otherwise free, all the more so, because paying for what is free is a form of corruption. And offering a public benefit to most, but a private benefit to the rich and lucky few is also a form of corruption. Moreover, ethical publications say who is paying who, and disclose other potential conflicts with their boards, executives, editors, and authors all the time. -- Alanscottwalker ( talk) 07:04, 1 November 2023 (UTC)

People pay for what they can also get for free all the time, most often when it provides them a convenience of some sort. That doesn't make it "corrupt", it's just how the world works. (And calling people who you disagree with, or what they are doing or thinking, "corrupt" is not exactly the AGF and civility you're intemperately accusing others of lacking.) Maybe admins being paid for consulting work that relates to Wikipedia should be something that is mandatorily disclosed like paid editing. But including "admins" instead of "editors" in that formulation doesn't really seem salient. I've been here 17 years, and deeply involved in shaping policy; if I were a bad-actor, I know way more than our "greener" admins and even many of our long-term ones about how someone could game the system in various ways to get things they want. Yet you want to impose something on 3-year editors running for RfA, but not on wikifossils like me. This seems like alarmist and misdirected worry aimed at a scapegoat, which are in fact characteristics of a moral panic or, on a more localized scale, a witch hunt. As for "And offering a public benefit to most, but a private benefit to the rich and lucky few is also a form of corruption" is certainly not something I agree with. If I offer "how to fix your credit rating" advice for free to the public on my Facebook page (I do), that in no way whatsoever obligates me to do hours or days of work advising someone in detail on what their credit woes' exact solutions are if they can't pay me well to do it. If I spend a year writing a book on how to mod the hell out of Skyrim (I didn't), I'm under no moral or other obligation to give that book away for free; the fact that someone who wants it might not be in a position to buy it isn't my problem, and the fact that they can do the work to get all that information for free themselves obviates that person's issue, to the full extent that I need to care about, and in no way makes me "corrupt". A WP user getting paid to advise on how to edit WP properly isn't "corruption", though advising people how to manipulate WP and get away with pushing a PoV or get coverage of something not really notable, or otherwise game the system, would be unethical. The quandary we have is how to reliably distinguish between these things, and we probably can't until after a line has already been provably crossed. Mandatory disclosure of paid WP-related consulting might be something we could write a rule about, but it wouldn't logically have anything to do with admins exclusively.  —  SMcCandlish ☏ ¢ 😼  07:58, 1 November 2023 (UTC)

First, there is no need for you to recite your experience to me, for goodness sakes. I'm not new here. Second, I am not wedded to admins alone, administration is the area where where the issue gained salience and prominence, and thus it is the place to start. Also that a system is corrupt (paying for what is free, charging for what is free, is an economic system -- and no its not economically rational, which means there is a corruption somewhere probably in the distribution system -- turning public benefit, into a private benefit is also an economic system) is a comment on that system, not a person. I of course think you are right to be open to disclosure (it is standard guardrails and if it's all to the good - as you guess - it especially should live in and embrace being in the open), even if I disagree with you about corrupt systems. Alanscottwalker ( talk) 08:26, 1 November 2023 (UTC)

You spend way too much time accusing other editors of harboring bad faith assumptions when you leap to the BFA that I mentioned my experience level here as some kind of comparison to you, an argument to authority. It's stark plain obvious that the entire point of mentioning it was to contrast myself with relatively new admins whose alleged "corruption" you so fear, and there is no rational read of what I wrote that has it otherwise. FFS.  —  SMcCandlish ☏ ¢ 😼  09:26, 1 November 2023 (UTC)

I did not leap to anything, and I was not referring to a comparison to me, I spoke that I am knowledgeable of you, so there is no reason for you to recite your experience. And that your experience was irrelevant to begin with because I am not wedded to admins alone, it's just the obvious and rational place to start, no panic, moral or otherwise about it. (I have no fear, I just have analysis.) And I still disagree with you about critiquing corrupt systems, not persons, so much of your comment makes little sense. FFS? - I'm sorry if you're angry. Alanscottwalker ( talk) 09:37, 1 November 2023 (UTC)

You can back-pedal and play CYA all you like, but we can all read your insulting insinuation that any provision of value on top of free sources must be free or is "corruption". "I just have analysis"? You are displaying great trouble (real or feigned) in parsing even the most basic reasoning in what is being said to you here, and then double-down on your incorrect interpretation even after the correct one has been explained to you in great detail. This is having a palpable degrading effect on the quality of the discussion.  —  SMcCandlish ☏ ¢ 😼  19:21, 1 November 2023 (UTC)

You've gone way off-topic with this post, so we'll just stop here. I of course disagree entirely. Alanscottwalker ( talk) 20:56, 1 November 2023 (UTC)

A key reason for people to pay for services that are being offered by others for free is because it brings contract law into force. For example, this allows the client to specify response time guarantees. Linux support is one example of many. isaacl ( talk) 14:21, 1 November 2023 (UTC)

Sure, they seek a private benefit including exclusive access not available to the general public, and in particular not available to those who can't pay. -- Alanscottwalker ( talk) 15:42, 1 November 2023 (UTC)

Yes, that goes for all purchases. It's not a corruption in the distribution system to charge for providing IT support, for instance. Both sides are exchanging something of value to them, and agreeing upon terms for that exchange. In the case of Linux, Linux consultancies help fund Linux development that returns value to the general public. isaacl ( talk) 15:52, 1 November 2023 (UTC)

We are not talking about a market for IT services, we are talking about a market for a free encyclopedia, that is meant to be freely transparent in its making. And no, when it comes to the exchange of money it by definition is not possible for those without it to participate in the exchange. Alanscottwalker ( talk) 16:02, 1 November 2023 (UTC)

Sure, there were some comments that appeared to question the idea of paying for any type of service that is also available freely through other sources, and I was addressing that. Regarding Wikipedia being freely transparent in its making: by design, Wikipedia isn't fully transparent as it supports anonymous editing (beyond just non-logged in editing), explicitly with the idea that it can help enable contributions that might be more difficult to obtain otherwise. For better or worse, this makes it difficult to evaluate edits in context of the editor. This can be good for eliminating potential biases against some editors, or avoiding potential harm they might encounter as a result, but is bad for understanding potential conflicts of interest. I have suggested that editors should be reminded of their ethical duty to place Wikipedia's interests ahead of any Wikipedia-related demands from other parties, much as some licensed professionals are required to place the interest of the public ahead of their employers when a conflict arises. This won't, of course, stop anyone from deliberately ignoring Wikipedia rules (no new rule would), but would give good-faith consultants something to point to when they tell their clients that they can't do something against the rules. isaacl ( talk) 16:35, 1 November 2023 (UTC)

Which is why it's a distribution problem: paying to address issues of delivery (including exclusive and legal claim on the time of the servicer) a choice and a service not available to those who cannot pay. -- Alanscottwalker ( talk) 16:55, 1 November 2023 (UTC)

As you said you are talking about Wikipedia and not commerce in general, could you perhaps describe your viewpoint of the distribution problem with respect to Wikipedia? For example, is the problem that Wikipedia volunteers are unable to provide help to everyone who requests it, with the legal responsibilities of a contractual agreement? (Exchange of consideration to bring contract law into force can be a miniscule token amount.) Sure, in an ideal world, there would be enough free help and it would all be high quality. I can't envision how to make this happen, though. isaacl ( talk) 17:32, 1 November 2023 (UTC)

The problem is not lowering the barrier of entry to information for the non-payor, or indeed not providing the service for free and to everyone. As Wikipedia is free encyclopedia publishing, that first part is something we try to do, and indeed need to continually work on, and the second part is more likely when the advice market moves to all free, instead of the advice being bought up. If that's an ideal, that's our purpose. Alanscottwalker ( talk) 19:20, 1 November 2023 (UTC)

You've referred a few times to a distribution problem. If I understand correctly, it's not a problem with making it easier to providing help freely, nor is it a problem with having paid support. Can you help clarify what the distribution problem is for Wikipedia? isaacl ( talk) 20:46, 1 November 2023 (UTC)

It's distribution of information, it how to edit and not edit Wikipedia. Alanscottwalker ( talk) 20:54, 1 November 2023 (UTC)

I appreciate the goal is to distribute information on editing Wikipedia in accordance with Wikipedia guidelines and norms. Do you have a specific problem statement on what needs to be improved? isaacl ( talk) 20:58, 1 November 2023 (UTC)

Disclosure of paid advising needs to be improved, that's that the distribution model, that has been brought to the fore. Alanscottwalker ( talk) 21:06, 1 November 2023 (UTC)

OK, so the distribution problem to which you are referring is to publicize the ways editing information is provided to users, and for everyone to disclose from whom they received advice? isaacl ( talk) 21:13, 1 November 2023 (UTC)

The discloser is the party paid. And the distribution problem is as I said above at 19:20.-- Alanscottwalker ( talk) 21:16, 1 November 2023 (UTC)

As long as the disclosure can be traced, does it matter who makes the disclosure? Since it only affects Wikipedia if the recipient causes an edit to be made on Wikipedia (directly or indirectly), it may be more suitable for the recipient to disclose the provenance of their knowledge. isaacl ( talk) 21:31, 1 November 2023 (UTC)

The disclosure request is to the advanced permission holder offering the service because the reason for the service is to effect Wikipedia, how it's viewed, how it's understood, etc. And as I mentioned somewhere above for example, deciding not to publish is a publishing decision, that effects our coverage. Alanscottwalker ( talk) 21:44, 1 November 2023 (UTC)

But as long as the relationship is reported by either side, then the end result is the same?

I'm not sure we really have a problem with advisors telling people not to add information that ought to be added. To me this feels a bit more out of scope; I don't know if it's helpful to know about all the situations where people decided not to add information to Wikipedia. isaacl ( talk) 21:52, 1 November 2023 (UTC)

People who are advisers better be telling their clients not to add to Wikipedia for the same reasons that are public, so yes it matters. And there is no reason to wait for anyone to disclose other than the advanced permission holder. Alanscottwalker ( talk) 21:57, 1 November 2023 (UTC)

To "give good-faith consultants something to point to when they tell their clients that they can't do something against the rules" is actually probably the most cogent argument for us to expand WP:PAID to cover WP-related consulting work.  —  SMcCandlish ☏ ¢ 😼  17:07, 1 November 2023 (UTC)

I would think that in an ethical consulting relationship, pointing to "the rules" would be ample and "the rules" exist. Wehwalt ( talk) 17:15, 1 November 2023 (UTC)

For instance, if an editor who had providing consulting services later detected some edits that appeared to be paid for by a client, they would remain duty-bound to deal with it as they would with any similar situation. Making this clear ahead of time with their clients would clarify that their consultancy does not forestall their involvement in all matters related to their clients. (Because unpaid editors are volunteers, there's no rule to point to saying they must enforce all rules.) isaacl ( talk) 17:39, 1 November 2023 (UTC)

What duties, as in the sense of obligations, do you see that administrators have, leaving aside their own compliance as users with the Terms of Use? Wehwalt ( talk) 17:54, 1 November 2023 (UTC)

As I stated, all editors (not just admins) have an ethical duty to place Wikipedia's interests ahead of any Wikipedia-related demands from other parties. isaacl ( talk) 20:51, 1 November 2023 (UTC)

Where does that derive from? Wehwalt ( talk) 21:22, 1 November 2023 (UTC)

Good-faith collaboration on any project requires that the participants work towards the common interests of the project. Of course no volunteer is compelled to take any action, so if one doesn't want to, say, copy edit an article related to a controversial area, they don't have to. If they don't want to copy edit an article on a friend, they don't have to either. If they want to copy edit an article on a friend to remove bias, but don't at the behest of the friend, then they are prioritizing a personal relationship above the project's interests. I imagine many editors will understand this circumstance, but I think they would be less understanding of someone prioritzing a financial relationship above the project's interests. isaacl ( talk) 21:44, 1 November 2023 (UTC)

Unintentional short circuit

I hadn't seen this discussion before I may have unintentionally short circuited it by asking @ JPxG: at his RfA whether he has done any paid Wikipedia advising. It seems appropriate as about half of the folks at the humungous WP:VPP discussion have indicated that they have serious questions about admins doing paid Wiki-advising. I do intend to ask the same type of question at future RfAs and do believe it will have the effect that any admin candidate will realize that they are not going to get 65% support to be an admin unless they answer the question. It's not a question that's asking them to admit to breaking any rules (just like the question about paid editing is just a declaration of whether they have done it - but in that case if they declare that they've done paid editing since June 2014 (without previously declaring it) it would presumably disqualify them). In short, the effect will be that all candidates will know ahead of time that paid Wiki-advising is a serious matter.

I'm not really saying anything about @ Cullen328:, but do thank him and Barkeep for bringing the matter up. My main concern about paid Wiki-advising is that most paid-editing websites solicit customers based on the premise that they are not a paid-editing service breaking our rules, but simply a consulting/advising firm, despite never being identified by paid editors as their client-employer-affiliate for their supposed extensive history of involvement in paid editing. Check out almost any paid editing website - they are transparently lying to their customers about being paid editors by saying they are consultants. Take a look around and it will be clear that they are writing musch of the material that the paid editors insert here, that they'll help find a paid editor to do the actual insertion, and that they will monitor and "correct" any unfavorable edits to the article. I don't want those types of advisors anywhere near Wikipedia. And if an admin candidate is anywhere close to that, I would want them to declare it.

We can deal with the general issue of paid Wiki-advisor admins in due course - I don't think there are very many of them. But there's no reason to let admin candidates think that very many of our editors wouldn't oppose most paid-advisor admins. Smallbones_{( smalltalk)} 21:01, 1 November 2023 (UTC)

So we can close all these discussions and go on with the daily life of Wikipedia? Wehwalt ( talk) 21:31, 1 November 2023 (UTC)

Look at his answer and tell me how informative it actually is. I hope this draws some attention to the importance of asking revealing questions rather than ones more appropriate for a polygraph test. _{Duly signed,} ⛵ WaltClipper -( talk) 23:17, 1 November 2023 (UTC)

Note that the question did not ask if he would submit a reconfirmation request if he changed his mind. That's a big difference, such a question allows you to hold someone to their word, especially if it's "hell yeah!" -- SashiRolls ^{🌿 · 🍥} 23:41, 1 November 2023 (UTC)

You're right about that. I think if one were to rate questions against each other, yours is certainly more informative in the sense that it does provide evidence for later on (although I still contend it assumes we anticipate that the candidate will lie, which is not WP:AGF-friendly). Smallbones, if you intend on going forward with a question like you have, please consider modeling it on the one I crafted above, as I think it will be more informative as far as whether they may be susceptible to accepting money for Wiki-stuff.

Until we have a decent amount of evidence that administrators (note administrators plural, not singular) are actively operating in the "bad actor" manner that you're describing - they [...] monitor and "correct" any unfavorable edits to the article - then I think we need to select the question that provides the most benefit to readers of the RfA. Assessing an administrator's judgment, ethical or otherwise, is pretty important. I think the whole case with Lourdes truly reinforces that. _{Duly signed,} ⛵ WaltClipper -( talk) 13:28, 2 November 2023 (UTC)

So I just saw that whole recent case where a previously-blocked admin managed to get re-RFA'd. I don't know how to further crack down on this - maybe candidates should go through whatever the most extensive version of CheckUser is - but whatever it is should be done as there have been way too many cases where people who plainly shouldn't be admins either became admins or very nearly became admins. FOARP ( talk) 13:42, 8 November 2023 (UTC)

CheckUser isn't magic pixie dust that could find bad actors if used on every RfA candidate. The large-scale privacy violations wouldn't be justified given the very small chance of catching someone like that. Maxim ( talk) 14:02, 8 November 2023 (UTC)

A similar proposal was discussed at VPP recently: Wikipedia:Village pump (policy)#RFA process reformation. –  Joe ( talk) 14:07, 8 November 2023 (UTC)

Any "vetting" would either be easily circumventable by anyone smart and dedicated enough (Lourdes was checked after and it was found that they would not have been detectable), or be so invasive that it would not be worth it for most editors. We want admins who will do the right thing without fear and intimidation. That can only happen when they can be completely anonymous. Usedtobecool  ☎️ 14:34, 8 November 2023 (UTC)

Having made my earlier comment and thinking about how checkuserinq RfA, candidates is a perennial proposal, I thought it may be useful to explain how CheckUser works in the frame of such proposal. I don't think anything I'm going to share is a particular secret, but perhaps it hasn't been explained in a single comment, and I'm hoping it's something that can be referenced when such a proposal inevitably resurfaces.

CheckUser works broadly speaking, as a tool that shows IP addresses and user agents for a given account or accounts and user agents for a given IP addresses or range. A user agent is the operating system and browser details, and in the case of mobile devices, this generally gives the general brand, and most often, the specific brand, of a device. Information is stored when someone makes edits or certain logged actions. It is configured on WMF wikis such that data expires after 90 days (stales). This is the first pitfall of a systematic checkusering: ideally you need to have a second account or IP address/range that has been used within the past 90 days.

Say you have a particularly bad sockmaster. In that case, the results of previous CheckUser queries may have been manually saved, whether on the checkuser wiki, the arbcom wiki, or privately by an individual CheckUser. This is useful because when one checks an account suspected to be a given sockmaster, there is some reference information. Perhaps the behaviour is suspicious, but if there is a geographical connection, an ISP connection, or perhaps a distinctive mobile device known to be used in the past, that may well be enough for a block as it could be more likely than not that the new account is connected. Alternatively, because CheckUser actions are logged, it is possible to see when and why a given IP or account was previously checked, and that also may given some hints as to who a suspicious account belongs to. So, if one is going a fishing expedition, if one gets quite lucky, the IP address or range used by an account could have been checked previously. That said, please note the keyword "lucky".

Say there's been a fishing expedition—let's set the ethical problem aside fora second—and some IPs were found that have been used by a sockmaster. Does that mean that the new account belongs to the sockmaster? Maybe, maybe not. Users change ISPs and devices. IPs and ranges get reallocated. In some countries, that means someone can show up on multiple broad ranges in a single day. Unless there is temporal proximity, along with similarity in behaviour and user agents, the answer tends to fall under "maybe not", which is really, in a nutshell. why such fishing expeditions are not useful to find bad actors. But, in the context of RfA, let's consider the human factor too. If we have a sockmaster trying to get an admin account, then this person is already knowledgeable about how things work here. If said sockmaster used to edit off a PC, perhaps he tries a Mac. Depending on where the sockmaster lives then changing ISPs wouldn't be a bad idea at all. This means that someone searching for said theoretical sockmaster among a pool of RfA candidates with the checkuser tool will most likely find a candidate or two that lives in the same country as the sockmaster.

The moral of the story here is two-fold. First, circumventing CU is quite feasible and someone nefarious intent on passing RfA will absolutely be doing that. Second, even without trying to circumvent it, the chance of finding sockpuppetry by systematically checking all RfA candidates is low because of what information checkuser actually gives. My final comment is to loop back on the ethical and cultural aspect; we are leery of invading editors' privacy, and perhaps more significantly, the CheckUser culture on enwiki is to avoid checking established editors absent a very good reason. Maxim ( talk) 14:59, 8 November 2023 (UTC)

Having skimmed the Lourdes case a few times, and particularly the admin-action cleanup work afterward, it seems like the tangible effect of the years of deception was...essentially nil? On Wikipedia, no one knows that you're a dog, but the moment you start barking we're going to show you the door. _signed, Rosguill ^talk 15:08, 8 November 2023 (UTC)

But Lourdes isn't the only case, right? I forget the name but didn't we recently just have someone who was basically going to pass RFA right before they got caught? (Edit: it was this Wikipedia:Requests for adminship/Eostrix, so not quite as recent as I had thought, but not so long ago)

I think my main concern is this case flipped my assessment of "are there at least 1-2 admins who are essentially bad actors operating under the radar?" from "seems unlikely" to "probably, yeah". This isn't an accusation against anyone specific, just what seems likely given it apparently being possible to get through RFA as a sock and there being ~1000 active admins. FOARP ( talk) 16:03, 8 November 2023 (UTC)

Oh, you can count on it being way more than one or two. WP is one of the most influential websites in the world, so every major "agenda" out there is hot to inject someone from their PoV into our admin pool. All it takes is playing the game for a couple of years well enough to fool everyone.  —  SMcCandlish ☏ ¢ 😼  16:21, 8 November 2023 (UTC)

I think Eostrix is an example of the process working, not failing. From the ACN discussion, Eostrix was being investigated prior to running. When he started his RfA, ArbCom prioritized the investigation and ultimately blocked him before the RfA reached the halfway point. I'm not sure how that could have gone better, other than ArbCom prophetically intuiting that the Eostrix investigation was worthy of additional eyeballs before he got the chance to transclude a nomination? House Blaster ^talk 01:26, 10 November 2023 (UTC)

There have been a number of comments elsewhere about potentional admins being put off by the preceived trauma of going through an RfA. Now add to that a requirment that they undergo a background check. Requiring background checks will keep at least some qualified users from undergoing an RfA, while bad actors will be careful about hiding or obscuring their intentions and history. I am reminded of the old observation about requiring loyalty oaths. The real bad actors had no problem signing loyalty oaths. It was people worried about freedom of speech and thought who got punished. Donald Albury 17:08, 8 November 2023 (UTC)

I second what Donald Albury said. Bad actors will always slip through, but imposing "better vetting" of admin candidates will simply cause even fewer editors to apply. SouthernNights ( talk) 17:48, 8 November 2023 (UTC)

I'm entirely aware of this, but at the same time should there really be no formal background check at all? Only what people are willing to do ad hoc at RFA? FOARP ( talk) 19:37, 8 November 2023 (UTC)

No, there should be no formal background check b/c we're not recruiting people for the FBI here. Again, a checkuser and other tools available to us are unlikely to catch truly bad actors. But it will cause people either to not apply in the first place or give false negatives on people who are good editors. SouthernNights ( talk) 21:14, 8 November 2023 (UTC)

• It's hard to think what more we could do. CheckUser, as Maxim explains above, is not helpful. Having the WMF or some other third party verify candidate's real-life identity sounds like it could help, but to avoid being exclusionary it would have to be optional, and if it's optional (really optional, not technically optional but you won't pass RfA if you don't do it) then it's useless. –  Joe ( talk) 19:36, 8 November 2023 (UTC)

  If not for admins, then for Bureaucrats? The case over at CN Wiki was also pretty bad. FOARP ( talk) 19:39, 8 November 2023 (UTC)

  I agree, I think that if third party verification of candidates became an option it would effectively become non-optional. Not that I have anything to hide, but I'm not sure I would have ran for administrator if I had to disclose my real-life identity. Hey man im josh ( talk) 19:45, 8 November 2023 (UTC)

• I vaguely recall having to submit ID to the Foundation when I stood for admin, is that no longer a thing? Maybe it was for something else... Giant Snowman 19:41, 8 November 2023 (UTC)

  @ GiantSnowman: I didn't submit anything to anybody to confirm my identity, nor would I really have wanted to. Hey man im josh ( talk) 19:43, 8 November 2023 (UTC)

  You used to have to do it for CU/OS/Arbcom. The WMF have a scan of my passport somewhere. It's never been a requirement for admins. HJ Mitchell | Penny for your thoughts? 20:01, 8 November 2023 (UTC)

  Yes, that must have been it. Why can't we do the same for (new) admins? Giant Snowman 21:52, 8 November 2023 (UTC)

  You could but I think that would deter a number of people from running at RfA. I'd have given it more if that was a requirement. Hey man im josh ( talk) 22:34, 8 November 2023 (UTC)

  I'm also not sure that legal would want that responsibility as the storing of that personal information comes with obligations for the foundation. Best, Barkeep49 ( talk) 22:38, 8 November 2023 (UTC)

  I recall someone posting before that the WMF, not having any good way to validate scans of identity documents and wary of keeping private information indefinitely, dropped the requirement and destroyed previously submitted scans. isaacl ( talk) 22:42, 8 November 2023 (UTC)

  I'd imagine storing that sort of data would be somewhat of a security risk? Like if we did have a list of the real names of every admin on the website and a bad actor got a hold of it... I think that that would likely be a way worse outcome than a few bad actors slipping through the cracks? For example, if people had to provide scans of their ID to prove said identity (which typically has other sensitive information like someone's address), there could be the risk of large scale doxxing. Clovermoss🍀 (talk) 03:00, 9 November 2023 (UTC)

  Yes, as has been stated here and in other threads, thus the lack of desire to collect this info. isaacl ( talk) 04:16, 9 November 2023 (UTC)

  Noting here that the WMF got out of the business of requiring identification from anyone in 2014. There are certain exceptions that relate to NDAs from individuals who (due to the nature of their participation in a limited number of committees) may receive confidential or embargoed information. No admin, checkuser, oversighter, steward, interface admin, or volunteer in similar on-wiki positions is required to provide proof of identity. Even back in the day when some positions required it, the info was sent to a special email address, the data was printed out and/or put on a special USB, and the personal data was kept in a vault with the three or so T&S staff having access. As soon as the privacy policy was changed in 2014, that material was destroyed. There were many reasons for ceasing this policy, but amongst them was the fact that it's trivially easy to get false documents; and the fact that if the WMF had the data, they may find themselves in a position to have to release it for various reasons, and they did not want to put volunteers at risk. This has become increasingly important as the number of editors and functionaries in non-democratic countries has increased. Now, I realize that after all these years and all these committees I've been on, it is trivially easy to link my RL name to my username, but I went into these activities with the knowledge that this was the case. In other words, my RL and onwiki identities are easily linked because of my non-Wikipedia activities. If I had stayed just an ordinary admin all those years ago, and kept my activities strictly on-wiki, they probably wouldn't be connected, and I would have just been closing AfDs and blocking vandals and editing. There's no benefit whatsoever in asking for *anyone* to verify the identity of *anyone else* for the purpose of adminship. People who are bad actors can easily get around that. Risker ( talk) 05:47, 10 November 2023 (UTC)

• A different way to get at the core problem is better vetting of !voters. There are scores of !voters affecting every Wikipedia process (RFA, AFD, RFCs, etc) who have ever made less than 100 edits in article space, yet their opinions-- usually not well-developed or expressed opinions, just straight votes-- count as much as anyone else's, in spite of lack of meaningful engagement or a history demonstrating knowledge of core principles. Such is "democracy"-- but maybe at least 500 edits to article space before people can !vote anywhere would help RFA, RFC and everywhere else. I'm seeing the same serial !voters in many discussions, and they affect outcomes, including admin selection. SandyGeorgia ( Talk) 19:55, 8 November 2023 (UTC)

  Are bad actors getting in because of inexperienced voters? FOARP ( talk) 19:56, 8 November 2023 (UTC)

  Would need an analysis, but I suspect (unless I'm unusual-- normal is just a setting on the dryer that I don't aspire to) that more experienced editors only support at RFA editors they know well, where the inexperienced tend to run through places like RFA, AFD, and RFC supporting stuff just as a social game. I've certainly seen lots of uninformed and unhelpful !votes at AFD, RFC and the like, and many of those come from the same editors who enter here an unqualified straight vote, evidencing no knowledge of the candidate. At RFCs or AFDs, you can tell when the voter hasn't engaged policy and guideline; at RFA we have no measure of anything about their knowledge of the project or the candidate. SandyGeorgia ( Talk) 20:09, 8 November 2023 (UTC)

  Lourdes 2 only had three opposes and plenty of long-time users in support. Eostrix only had one oppose, and again, plenty of long-time users. Elli ( talk | contribs) 20:17, 8 November 2023 (UTC)

  Right. The chances of anyone - long time editor or not - being able to detect a bad-actor-admin at RFA, with the tools available, is pretty low. FOARP ( talk) 11:01, 9 November 2023 (UTC)

  Disagree (the "tools available" include engaging a brain that has extensive knowledge of and experience with a candidate, and not voting to support one if you don't have that); I haven't supported a bad RFA actor yet, as far as I know (and there have been far more of them than listed on this page-- a few in particular I knew in advance would end up right where they did). SandyGeorgia ( Talk) 12:54, 11 November 2023 (UTC)

• I honestly don't think that there's any more vetting that we should do, regardless of what we could do. I know there's a broad segment of the community who see adminship as a big deal, and I'm not saying that a banned user socking into advanced permissions isn't a significant breach of trust, but realistically there just isn't much that a rogue admin can do without someone noticing, and nothing that can't be repaired. As Rosguill said, the cumulative damage from years of Wifione/Lourdes adminsocking was basically nothing. There have been a few suggestions that we should do identity verification for prospective admins, but the potential invasion of privacy is significantly out of balance with the potential for damage. We don't even do that level of verification for functionaries, who really can do irreparable damage - since 2018 or so we've had to digitally sign some legal documents with our real name and have a verified email, but nobody is checking or confirming our real-world identity. It's completely unrealistic to expect that we could implement a level of background checking that would catch every bad actor without it being seriously invasive and a security risk in and of itself, and even then the vast majority of admin misconduct cases are the result of cluelessness, not malice. I honestly think the best response to the Wifione/Lourdes situation is to do nothing more than we already do; bad cases make bad policy, as the saying goes. Ivanvector ( ^Talk/ _Edits) 20:24, 8 November 2023 (UTC)

  Well said and total agreement! SouthernNights ( talk) 21:15, 8 November 2023 (UTC)

  • Agree with this. Both Lourdes/Wifione and Eostrix/Icewhiz would have found it trivial to falsify the documentation required for functionary vetting, as would anyone else with ill intent and a determination to create an admin account. If anything can be done, I'd suggest even more scrutiny by nominators, whose status as respected editors can lead others to support RfA's for candidates they don't otherwise know. But this would be a minor measure at best. The sad reality is there's no foolproof way of preventing the occasional situation like this one. -- Euryalus ( talk) 21:44, 8 November 2023 (UTC)

    And by "occasional", we need to remember that in 22 years we know of one adminsock like this one. Eostrix was caught before getting the permissions. I recall one other RFA where the candidate was revealed to have had an old abandoned account that they said they forgot about, which made them technically a sockpuppet and that torpedoed their RFA, but I decline to remember who that was. Ivanvector ( ^Talk/ _Edits) 22:46, 8 November 2023 (UTC)

    That would be Wikipedia:Requests for adminship/Ceradon 2. And there was also (from a quick glance through the WP:Former admins archives) Law and Pastor Theo and Archtransit who became admins while evading blocks, as well as Edgar181 who became an admin while actively socking (albeit not evading any specific block). Yes, all of those are very old, but it's it's not once in 22 years, and none of this IMO means any specific additional precautions are necessary. * Pppery * _{it has begun...} 00:57, 9 November 2023 (UTC)

    There was Cirt, whom I misremembered as an admin who also socked (i.e. the admin account was the master), but the SPI records that adminship was voluntarily resigned shortly before socking began. Adding this anyway in case anyone else has the same mismemory. Folly Mox ( talk) 12:53, 9 November 2023 (UTC)

    Actually, Cirt was desysopped for cause on enwiki in 2011, resigned adminship on other WMF wikis in early 2016, and then started socking in late 2016. * Pppery * _{it has begun...} 15:59, 9 November 2023 (UTC)

    The somewhat less stressful RfAs pre-2010 did not manage to weed out all bad actors or sockpuppeteers, and the current more rigorous regime does not seem to be significantly better, certainly not relative to the number of RfAs that take place. — Kusma ( talk) 16:44, 9 November 2023 (UTC)

  IvanVector, I suspect you're making those broad statements about "nothing they can do that can't be repaired" from a place of not having a bogus block log. They're indelible. Also, the list being discussed here of admins who got through, whether by squeaking through on marginal support via crat chats or via glowing support levels because they played the community, is incomplete. SandyGeorgia ( Talk) 13:00, 11 November 2023 (UTC)

@ FOARP: I'm not sure where to put this comment but I did figure it'd be worth saying something about further vetting of RfA candidates. I don't think more oppose sections like the one found at Wikipedia:Requests for adminship/ScottishFinnishRadish would be the best idea. That bothered me then and it bothers me now. The vast majority of people go through RfA with the best of intentions and because they care about Wikipedia. I can only imagine how disheartening responses like that would be. Also, the current number of active admins is less than 1,000. Clovermoss🍀 (talk) 10:44, 9 November 2023 (UTC)

I definitely do not defend the accusation that were made in that RFA, but this is also what will happen when people are aware that it is at least possible that some people entering RFA are bad actors and that they are the only ones who could possibly stop bad actors becoming admins at that point: we are essentially leaving it to RFA voters to "sniff them out". Essentially, if no formal vetting is done, then the job is left to individual !voters in the RFA to do ad hoc, and the result of that is always going to risk a potential witch-hunt. FOARP ( talk) 10:57, 9 November 2023 (UTC)

Putting this at the bottom for no real reason, but just noting I broadly agree with Ivanvector and others above. Eostrix was a special case, and I'm very grateful he's not adminning at AE right now, but in general I think this sort of situation is of the type better addressed by cure than prevention, due to the costs of the preventative measures.

If admins are making poor judgement calls, speak up. Take it to WP:XRV if you want, with a low-stakes "is this action endorsed / not endorsed" framing. Course correct people. Patterns of bad admin decisions are detectable and reversible in ways that constant suspicion towards each other isn't. Folly Mox ( talk) 13:15, 9 November 2023 (UTC)

I also agree with IvanVector et al. The main effect I can forsee of having admins identify themselves to the WMF is putting off otherwise-qualified candidates from running (or encouraging existing admins to resign, if we enforced it retrospectively). Doing that to make it harder for the very few people whose real-life identity is known to the WMF to sockpuppet their way to adminship seems like a terrible tradeoff to me, even if WMF would be willing to take on the responsibility. Caeciliusinhorto-public ( talk) 16:48, 9 November 2023 (UTC)

Adding to my previous, I see from the discussion above that five other admins have been identified who RfA-ed while socking; of those only one ( Law/ The undertow) would have been affected by this proposal. So of the 2,285 users who have ever been an administrator, we have had six known admin socks, of which only two would have been affected by this proposal. At a rate of one adminsock prevented for every 1,000 admins appointed, and at an average of 20.6 new admins per year since 2019, implementing this proposal would be expected to prevent one admin sock between now and ... about 2070, by my calculations Caeciliusinhorto-public ( talk) 17:08, 9 November 2023 (UTC)

Probably not even that, since someone intent on this could and likely would take countermeasures to get around such checks. Just watching admins for signs of PoV-pushing with the use of the admin tools (e.g. uniformly blocking "side A" in CTOP disputes to give the upper hand to "side B", etc.) is probably what we have to continue doing.  —  SMcCandlish ☏ ¢ 😼  17:52, 9 November 2023 (UTC)

RFA IS vetting. No more soul searching or time wasting bureaucracy is needed. Leaky caldron ( talk) 20:10, 9 November 2023 (UTC)

Agreed, and that is why RfA should never be a vote: there has to be an open discussion. It is also why comments from drive-by new accounts/IPs should be accepted. If someone has news about a candidate, it should be considered (and deleted if found to be trolling). Johnuniq ( talk) 00:01, 10 November 2023 (UTC)

OK, is it working against this specific threat? Is it potentially coming up with false positives on this issue? FOARP ( talk) 11:56, 10 November 2023 (UTC)

Not sure why several seem to think exceptional editors like Wifione & Icewhiz had malcious intent? Ice used to be a fellow ARS member, he was unusually gentle until he suffered severe provocation. He only sought adminship as a means to fight a certain evil. His play for adminship might have failed, but he simply kept taking things to a new level until his will proved too great for the community to resist. Much as I believe Ice was mistaken in seeing said evil as residing in certain edtiors or being tolerated by Wikipedia as a whole, one has to admire his ingenuity & determintation to achieve rightful victory (as he saw it). If y'all wanted to see a sockmin that might have had malicous intent, maybe check out someone like Eyrian; one of several abusive admins the legendary A Nobodoy helped take down until he finally met heroes end.

I'd agree with Ivanector's comment and go further in saying that it's regular editors who are best placed to indulge in arguably the worst corrupt practices. Much as our real world impact is sometimes exagerated, there are times when being able to present certain narratives in our articles would be worth 7 figure sums to certain parties. It's regular editors who would be best placed to do that as its easier for them to fly under the radar (obviously that does them little good if they're trying to POV push on hot topic, but sometimes the most effective time to change an article (from a corrupt PR persepctive) is shortly before the topic goes hot. In summary, no need to make RfA even harder to pass; the preceived sockmin "threat" is pretty minor in the wider scheme of things. FeydHuxtable ( talk) 12:32, 11 November 2023 (UTC)

Yes (re "regular editors" and "worst corrupt practices" not necessarily being sockmins); the definition of "bad actor" being used here, as well as the consequences of "bad admins" is limited artificially to those who have been found socking, when it might be expanded instead to those who have been desysopped for cause (who may have caused more damage than any found socking) or those who quietly walked away after satisfying their grudges and those who left to avoid an arbcase. And Clovermoss, I'm sorry it still bothers you now, but I obviously disagree that a thorough vetting of a candidate, as at SFR's RFA, is undesirable; it's the purpose of RFA as a discussion to bring forward concerns. I'd have brought forward a dozen more in the last few years if not for the badgering of opposers; we're desperate for admins, opposers are disparaged, and we're putting through substandard candidates, probably at a no different rate than we always have, with the advantage that arbcom seems to have become faster at desysopping. SandyGeorgia ( Talk) 13:30, 11 November 2023 (UTC)

Indeed. Personally I'd love to see RfA go back to pre Balloonman level of scrutiny – but your perspective is very valid, per the harm misjudged admin actions can do. Where I think we'd agree is it would be great if there was a process to get block log entries expunged. IMO a few bocks are almost a +ve on a veteran editor – they imply the person cares enough to put themselves on the line. But others see things differently so it would be good if the block log wasn't indelible – esp. as it sometimes comes to light that certain blocks were entirely unwarranted. RfA might help make RfA less high stakes & help us get more admins. FeydHuxtable ( talk) 13:34, 11 November 2023 (UTC)

Making bogus block logs fixable would help individual editors (and it's only been thrown in my face once, as most know the score on that block), but that won't address the more serious damage that can be done to content, by an admin supporting their personal POV via softer admin actions taken at ANI or AE or with semi-protection, for example. Admin POV is one of the many reasons I do consider who is nominating a candidate as a factor in my !votes. (It's always nice to dialogue with you, Feyd.) SandyGeorgia ( Talk) 13:43, 11 November 2023 (UTC)

I wouldn't call what happened at the oppose section at ScottishFinnishRadish's RfA "vetting". Unfounded socking accusations are not okay (and if they are credible, they belong at WP:SPI). RfA shouldn't be a free-for-all where other rules don't apply. I don't think I'm only one who continues to be bothered by what happened there. At least I hope not, because it definitely made me lose a lot of faith in our community processes.

Literal quotes from that RfA's oppose section include:

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.

I don't endorse the sock accusations above

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.

I am concerned that this user has edited under another account. I think we would be foolish to believe anyone would gain enough knowledge and experience to be an admin, by just spending the majority of their time on talk pages. No ... not enough experience ... and the above-raised questions about a previous account cause too much doubt

Oppose per socking concerns raised by eminent colleagues

any reasons, including: has said nothing about any previous account(s) despite questions about that having been repeatedly and reasonably raised

I am not convinced the editor did not previously edit using an undisclosed account.

I didn't start having nightmares about RfA until after this one. It's the one that crushed pretty much any motivation whatsoever I might have to run. Why in the world would anyone want to put themselves through something like that? I care about Wikipedia enough to not want to feel like the community at large hates me. I don't want to lose the magic of feeling like I matter as an editor to the project. It's a huge reason why I say no whenever people try to peer pressure me into running for RfA. I don't want to risk being thrown to the wolves because I didn't start editing thinking I was doing anything sketchy [1]. It's easy to not care about what passing vandals say but when it comes to people you edit with and respect? That all hits a lot harder. As I said at the time: 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"? I don't think these vague accusations help promote a welcoming environment at RfA and I don't think calling it out is somehow the problem. And hey, maybe all my concerns are overblown like everyone keeps telling me. That I wouldn't have issues. But I don't think it's crazy to see stuff like this and go "that could be me". Clovermoss🍀 (talk) 00:54, 12 November 2023 (UTC)

That's a reasonable concern, and that this took place is 2022 is troubling. Something more like what I would have expected in the late 2000s to early 2010s. One of several reasons I never ran for RfA again after that period is that one of my RfAs was largely torpedoed by a false claims that I "did a complete bashing on a new user [at] AFD" which others then dogpiled on with further false claims like that I had labeled their content/PoV position "vandalism", when the user was demonstrably a sockpuppet and vandal (blocked not long afterward), and my vandalism reports had nothing to do with their content viewpoint at AfD. But never mind the truth! The accusations just sound so juicy, especially combined! RfA participants have a mass-psychology bad habit of believing by default, and looking for ways to add to, any negative claim presented, without doing any verification. This problem has never gone away. [Other reasons I don't run for RfA again are that I don't have any adminstrative tasks I want to do any longer; the "wikicop" aspect of blocking and banning people doesn't appeal to me in any way; the tools I do have uses for have been unbundled from adminship; and it's become clear that anyone who is a regular MoS editor and blockades willy-nilly attempts to change it will build up an army of opponents, because everyone wants to change at least something in MoS to suit their personal pecadilloes and gets mad when they don't get their way – it's no accident that precisely zero of the MoS regulars are admins.] The ScottishFinnishRadish thing seems particularly dumb to me; haven't most of us figured out by now that lots of "new" accounts are people who were IP users for a long time first? I think it took me 6 months to bother creating an account.  —  SMcCandlish ☏ ¢ 😼  18:04, 12 November 2023 (UTC)

User:FeydHuxtable, Not sure why several seem to think exceptional editors like ... Icewhiz had malcious intent? .... He only sought adminship as a means to fight a certain evil. Anyone who remembers enough Latin is welcome to follow up below with the appropriate translation of "the question answers itself". Folly Mox ( talk) 15:28, 11 November 2023 (UTC)

And the idea that WMF-banned real life harasser Icewhiz, who went as far as getting his AE filings researched, expanded upon, and published by academics (kept taking things to a new level until his will proved too great for the community to resist) wouldn't have abused his on-wiki admin powers with his Eostrix account, is silly. Folly Mox ( talk) 15:34, 11 November 2023 (UTC)

Thank you FeydHuxtable for striking. Struck my response as it no longer applies. Folly Mox ( talk) 17:17, 13 November 2023 (UTC)

Agree with this. Wifione was a tireless POV-pusher for a commercial entity, and Icewhiz attempted to cause real-life harm to people he opposed on wiki. "Malicious intent" seems a fair descriptor for their conduct. -- Euryalus ( talk) 02:38, 12 November 2023 (UTC)

Agreed, that goes way beyond "intent", those are clearly malicious actions. That behaviour is not okay on many levels. Courtesy ping to Volunteer Marek. Clovermoss🍀 (talk) 22:46, 12 November 2023 (UTC)

• One thing I hope we can do to improve vetting is making RfA less adverserial. Having a few days of discussion before (!)voting starts. I think that helps in two ways: (1) you don't get the wave of early supports based on incomplete information. And (2) concerns don't have to be linked to opposes. Some members of the community may be reluctant to raise concerns in a high-stakes environment, knowing it can lead to early opposes before all information is on the table. Candidates can then respond to concerns without the time pressure of a RfA immediately open for !voting, allowing for further clarity and follow-up questions. —Femke 🐦 ( talk) 13:27, 11 November 2023 (UTC)

  This is what I think also - if you've already done some level of vetting first, then the chances of a witch-hunt RFA are lower. FOARP ( talk) 14:35, 13 November 2023 (UTC)

• The notion that desysopped admins are by definition bad-faith actors is nonsensical. Our list of admins desysopped by ARBCOM is dominated by two categories; long-term editors with a deep commitment to bettering Wikipedia, who were unable to show sufficient restraint, either with their language or (less often) with the revert button; and old-school editors who lost touch with community norms. They made unsuitable admins, but in a majority of cases they had the interests of Wikipedia at heart just as much as our long-term contributors who were never given the bit. Icewhiz and Wifione are in a different ballpark altogether, and Ivanvector and others have it exactly right with respect to the effects of additional scrutiny there. Giving the average candidate a hard(er) time does nothing to prevent the next Icewhiz, and indeed may make it easier given that we're making our need for admins more dire. Vanamonde ( Talk) 16:45, 11 November 2023 (UTC)

  Yep. There's little to do to prevent "badmins" other than paying closer attention, as RfA voters, to what their editing history shows about their behavior, and little to do to detect and deal with a "badmin" other than paying closer attention, as fellow editors (and in some cases fellow admins) to what they are doing with the tools and with the weight of the tools behind them. Making RfA more of a hellhole for people stepping up to the admin challenge isn't going to help any way and just cost us more good candidates.  —  SMcCandlish ☏ ¢ 😼  16:54, 11 November 2023 (UTC)

  To prevent "badmins", just don't promote any candidates who play badminton] ... I'll see myself out. Graham87 ( talk) 10:18, 12 November 2023 (UTC)

  Badminton is the clearly the town from which they come, though. If we knew whether it was the one in Gloucester or the one in Wales, we could finally take decisive action.  —  SMcCandlish ☏ ¢ 😼  18:04, 12 November 2023 (UTC)

  No comment on the Badminton connection, other than the safety factor - I'd far rather be hit by a stray shuttlecock than a cricket ball, polo ball, hockey ball or golf ball. Of the changes we could make, what do people think of the candidate meeting the nominator, either at a meetup or via zoom etc? Would that add confidence and or deter some badfaith operators? As for RFA being vetting, do we need better guides into how to review candidates? Are there things we could have spotted earlier re Lourdes/Wifione? Ϣere SpielChequers 18:14, 13 November 2023 (UTC)

  I'd imagine plenty of voters would feel more comfortable with the candidate being known to someone, although that could also be gamed by badfaith operators. I don't want to prompt WP:BEANS but there's a few different ways I can imagine this happening if one wants to email me. All that said, I think Wikipedians I've met IRL probably trust me more than others who haven't. Maybe that's more of a case of being able to put a face to the username on the screen, though. Clovermoss🍀 (talk) 18:56, 13 November 2023 (UTC)

  It's not that easy to meet Wikipedians in real life, though. I think there hasn't been a meetup less than one or two hours' drive from where I live for more than ten years. (The German Wikipedia seems to have a much stronger meetup culture). — Kusma ( talk) 19:08, 14 November 2023 (UTC)

  I agree with you. I literally went to my first meetup a few days ago and I've been a Wikipedian for more than 5 years. It's true I've showed up for some WMF/community meetings but it took me a really long time to even feel comfortable with that. I wouldn't blame anyone who never feels comfortable doing any of those things. Ivanvector's and ScottishFinnishRadish's thoughts below summarize my overall thoughts on the subject. While I think that a known quality might make some voters more reassured, it doesn't accomplish much, can easily be gamed, and to some extent could just give a false sense of security. The more barriers we have the more likely someone who genuinely does want to help out is less likely to do so and bad actors aren't really going to stop at anything if it accomplishes their goals. Clovermoss🍀 (talk) 19:27, 14 November 2023 (UTC)

  I've been to three meetups in 18 years, the last 11 years ago. I was a bit concerned after that last meetup when I realized that a couple of the attendees had created their user accounts after the meetup was proposed. I hate to say it, but good practice is to not reveal anything identifying about your non-Wikipedia self at a meetup. Donald Albury 23:50, 14 November 2023 (UTC)

  Candidate and nominator meetings won't bring us back to the "pre-Balloonman days" mentioned by Feydhuxtable; if someone doesn't know how to vet or review a candidate based on their written work here and how well they know the candidate and how closely they've worked with them, meeting them face-to-face isn't likely to suddenly confer common sense or good intuition or high standards.

  Better guides might help discourage the straight supports, by listing things one might check. But what good does it do to guide !voters or nominators about what things to check, if then those who do check are badgered, both at RFA and post-RFA? It won't do any good to educate !voters or nominators if they're still going to be maligned when they do raise concerns. SandyGeorgia ( Talk) 18:57, 13 November 2023 (UTC)

  Balloonman was my nominator, and I think RFA would really benefit from his return. Opposes do often trigger discussion, but rereading Wifione's RFA, I can't help but notice Xenon54 opposed partly over an article with NPOV concerns. My own oppose generated quite a bit of discussion and was mentioned by several other !voters, it garnered some reassurances that moved me into the support column. Xenon54's oppose generated no discussion at all, but in hindsight, they spotted the problem that lead to Wifione's desysopping. Perhaps one lesson from the Lourdes/Wifione saga is that we need to be more wary of spammers than we have been, and if a candidate's editing shows a lack of a neutral point of view, that needs to be a rather redder flag than it has been. Personally I've long been aware of spammers being a big threat to the project, but I was under the impression that they were more likely to try and stay below the radar. Is Wifione/Lourdes an isolated case, or do we need to ask ourselves at each RFA, does this candidate look like a subtle spammer? Ϣere SpielChequers 12:49, 14 November 2023 (UTC)

  I think editors who want to maintain their privacy should be able to RFA. Best Barkeep49 ( talk) 19:02, 13 November 2023 (UTC)

  It wouldn't need to be compulsory, but also, unless you are one of the thousand most famous people on Earth, a zoom chat or even turning up at a meetup doesn't have to completely void your privacy. We have some regulars at the London meetup who I assume live in South East England, but I have no idea where. Ϣere SpielChequers 12:49, 14 November 2023 (UTC)

  That's not good enough. Some people don't want to show their face on camera, period, and I can't blame them particularly considering how AI and deepfake technology is evolving so fast that it's impossible to even comprehend the consequences that may result from displaying your real life identity on the Internet. Even if we make something like this an optional choice, inevitably the question is going to come up during RfA - "Has the nominator met the candidate in person or over Zoom?" - which adds yet another wrinkle of difficulty to an RfA because obviously, if it's even remotely under consideration, people would prefer that the nominator know who they're nominating as much as possible. _{Duly signed,} ⛵ WaltClipper -( talk) 15:30, 14 November 2023 (UTC)

  And I'm not sure what's accomplished. Like how would Eostrix's nominators have known they were talking to Icewhiz even if they had done a video meeting? Best, Barkeep49 ( talk) 15:53, 14 November 2023 (UTC)

  And what do you do with admins who were interviewed by Wifione? ScottishFinnishRadish ( talk) 16:58, 14 November 2023 (UTC)

  Did either Wifione or Lourdes nominate anyone? I accept that not everyone would want to go down this route, but my experience from off wikipedia is that you get a measure of someone by actually talking to them. Ϣere SpielChequers 17:39, 14 November 2023 (UTC)

  Lourdes nominated Chetsford and Dreamy Jazz, both as the second nominator.

  Wifione nominated Anomie, GiantSnowman 2, Harrias, and Richwales 2. For all other than Richwales 2, they were the first nominator. Elli ( talk | contribs) 17:56, 14 November 2023 (UTC)

  I am very bothered @ WereSpielChequers with the guilt by association that you're heaping on a list of people with this question. Barkeep49 ( talk) 18:07, 14 November 2023 (UTC)

  I don't think any less of those admins listed above, but it's valid to ask whether who you accept as a nominator indicates something about your judgment. I've been approached by dozens, but I thought these odd even before we knew, partly because a) Lourdes and I didn't have that much interaction, and b) when we did, they never spoke Spanish to me (intuition, common sense red flags -- people who speak Spanish, speak Spanish to each other, because it's a warmer language). [2] [3] No face-to-face meeting would change my impression that something about the limited interaction we did have didn't align with those requests. SandyGeorgia ( Talk) 01:25, 15 November 2023 (UTC)

  Why is it valid to say "you, individual, should know more and do better than the community as a whole in knowing about your nominator, who has already passed community review. I don't think any less of you, but I think it indicates something about your judgement."? Barkeep49 ( talk) 01:33, 15 November 2023 (UTC)

  Because the community as a [voting] whole that passed RFA candidates (under percentages I have always thought too low) includes editors we don't expect to have the same judgment we expect an admin candidate to have. Alternately, viewed in terms of how FAC works, "drive-by votes" count equally at RFA, whereas at FAC, you can't become a featured article unless valid opposes are resolved. Or for yet another view on the nominator judgment issue from an entirely different angle; Spicy's RFA demonstrated good judgment, intuition, knowledge, common sense, experience, reading of the room, in nominator selection. And from yet another angle, take the diffs above and suppose for some weird reason I had accepted. I expect a long-standing editor like me to know better. SandyGeorgia ( Talk) 02:11, 15 November 2023 (UTC)

  I think it's a very different thing to say "I should have known better" and "these admins should have known better and not having known better indicates something about their judgement". Put another way, self-reflection is good, negatively implying something about fellow volunteers without evidence isn't good. Best, Barkeep49 ( talk) 15:44, 15 November 2023 (UTC)

  That's fair (there's a whole lotta "I should have known better" on this page, and lessons learned). SandyGeorgia ( Talk) 17:03, 15 November 2023 (UTC)

  Barkeep49 the example between Lourdes and me gives another way to look at the dysfunction of RFA. I always had red flags with Lourdes, because they never indicated knowledge of Spanish. Now, considering the oppose-badgering mindset at RFA, and the amount of "evidence" that is demanded of an opposer, let's contemplate what the reaction would have been, if I had tried swimming against the sea of support at the Lourdes RFA by saying "I don't trust 'em because they don't speak Spanish to me". I'd be laughed out of or badgered out of the room (and probably ignored by 'crats). People who speak Spanish, speak Spanish to each other, and they don't sign their posts with "love" when they don't even know each other (they sign them with cariño or something similar).

  Scores of drive-by supporters can simply !vote on trust, but an oppose on that basis would be/is ignored, discounted by 'crats, or badgered. An opposer can't oppose on the single most important factor at RFA: intuition is telling you not to trust this person with the tools. And badgering silences opposes of those who knew (as I was silenced with one of the first sockmins, Archtransit). So a lesson learned (that has come up before) is allow opposes to be simply trust-based just as supports are: don't demand that an opposer build an evidence-based case, and 'crats, don't discount trust-based opposes if you aren't going to also discount all the drive-by "support, 'trust em" declarations. The process already does enough to discourage opposes, as retribution is a possibility down the road; the inequity between how opposes and supports are treated worsens the chances of good vetting. SandyGeorgia ( Talk) 19:19, 15 November 2023 (UTC)

  I don't understand. When did Lourdes make the representation that they knew Spanish? Shells-shells ( talk) 20:21, 15 November 2023 (UTC)

  I wouldn't know where to find that this many years after the facts. SandyGeorgia ( Talk) 21:39, 15 November 2023 (UTC)

  Lourdes strongly implied that they were a semi-popular Spanish musician, that's probably where that belief comes from. ULPS ^{( talk • contribs)} 21:39, 15 November 2023 (UTC)

  Sandy, some !votes were rightfully given lower weight at SFR's RFA because they were allegations of socking, not because they were based on trust. As far as I can see you are the only editor insisting that no socking allegations were made, when there's literally dozens of experienced editors saying the opposite. Much as I respect you, you've lost perspective on that RFA. Vanamonde ( Talk) 20:22, 15 November 2023 (UTC)

  Well, it's sure not pleasant to be an opposer at RFA, and here we're seeing an example of why I didn't speak up at Archtransit, and why i wouldn't have spoken up at the sea of support at Lourdes -- the badgering endures for years, so people don't speak up when they have concerns. Maybe, yes, my perspective is skewed per that. SandyGeorgia ( Talk) 21:49, 15 November 2023 (UTC)

• I support in principle the idea of some useful technical scrutiny becoming part of the RfA process. I hear those who know the most about CU saying that it would not be useful. If that changes, or if someone has a bright idea, I'd be interested. Firefangledfeathers ( talk / contribs) 20:18, 13 November 2023 (UTC)

  It would not have been useful in the Wifione/Lourdes case. Without going into unnecessary detail it could be useful under different circumstances, and if the community developed consensus for checking every RFA candidate then there's probably a way that could work with the relevant policies. I personally would not support that: checkuser exposes sensitive private data with the aim of detecting abuse; checking every RFA candidate would be excessively invasive when weighed against the expected benefit, and it's rather simple to evade detection if you know you're going to be checked. Ivanvector ( ^Talk/ _Edits) 18:36, 14 November 2023 (UTC)

  I also explained during my rfa that I had been checkusered and revealed personally identifiable information to Arbcom, checkusers, UTRS, and oversighters. That didn't stop accusations of sockpuppetry. I went far beyond what is being proposed here, and it didn't seem to satisfy many editors. ScottishFinnishRadish ( talk) 19:10, 14 November 2023 (UTC)

  That's just it: you shouldn't have had to provide any of that, and even though you did it still didn't satisfy the mob. Some editors are convinced you're a sockpuppet just like birthers are convinced Obama is Kenyan; there is nothing you can show them and no vetting thorough enough to convince them otherwise. It is utterly pointless and an absolute waste of time to try anyway; see also WP:GREENCHEESE. We could force every blocked editor to submit a DNA sample and have every RFA candidate be tested against our biometric database of offenders, and for one it still wouldn't be enough for some people, and two it still wouldn't stop dedicated bad actors. Ivanvector ( ^Talk/ _Edits) 21:53, 14 November 2023 (UTC)

  I guess the takeaway from that is that at least one 'crat then said (paraphrasing) "The accusations of sock puppetry are without evidence and I'm discounting them completely. Consensus to promote ScottishFinnishRadish as administrator." So yeah, it's unpleasant and beyond the pale, but in this case, what goes around comes around. Ritchie333 ^{(talk) (cont)} 11:23, 15 November 2023 (UTC)

  It's interesting that we have that, along with big green blobs of requoted material re these alleged socking allegations (redundancy intended), which overlook and diminish the very issues this thread should attempt to address-- how to vet RFA candidates. If nominators and candidates alike don't anticipate and mention in their nom statement prior activity (which was, according to SFR, later revealed as editing as an IP), and then muddy that further by advancing a candidate as a content contributor based on little content contribs, they can and should anticipate that others will and should look more deeply. That's what RFA is for. "Returning editor" (as an IP) was apparently correct in this case, but badgering the opposers continues. That's not going to advance the case for how to better vet RFA candidates. SandyGeorgia ( Talk) 13:23, 15 November 2023 (UTC)

  On the other hand, we have things like Wikipedia:Requests for adminship/Ifnord, where I gave what I considered to be a reasonable and polite oppose, which drew hardly any comments, other than a single one that concluded that they understand what I was getting at .... and the editor withdrew the RfA, made a single comment about one opinion they thought hurtful, and has not edited Wikipedia since. So one editor's "due diligence" is another's "unnecessary harassment". Ritchie333 ^{(talk) (cont)} 14:57, 15 November 2023 (UTC)

  Right. By bringing the focus (forcefully -- others, not you) on this discussion around to re-litigating an old RFA, including repeated statements about alleged socking allegations, this discussion becomes less likely to bring forward any potential solutions. Of which I see plenty. Better vetting of candidates; better presentation of nom statements; perhaps clarify what is meant when asking about previous accounts, to include IP editing history; better !voter education; more awareness of things everyone should be looking at; less badgering of opposers; less drive-by !voting ... there's plenty. Continued badgering of opposers doesn't advance solutions, unless one thinks the solution is an easier ride for candidates where !voter concerns about trust, and how to evaluate it, are stifled. Stop making opposers jump through linguistic hoops (where any words they choose will be twisted anyway) to say "I don't trust this editor with the tools". There are differences regarding what "due diligence" looks like, and any time I see a "content contributor" advanced based on one GA, I'm going to do a deep dive. That has always been my criteria, I have supported non-content contributors when they are presented as such, and my criteria are not going to change because my concerns are misrepresented and badgered. SandyGeorgia ( Talk) 15:41, 15 November 2023 (UTC)

  I don't know how someone can read the quotes I posted earlier and only say there were "alleged" socking allegations. It's frustrating because even the discussion at the time had people saying that no one implied SFR was a sock. There's plenty of evidence that people did. I honestly encourage people to read that RfA whenever they wonder why people don't run. It's a great example of why people don't. The whole thing frustrates me and I wanted people to see for themselves what exactly I was going on about. I really don't usually care about opposes in RfAs, but what happened with some at SFR's is just so egregious that it's impossible for me to not say anything. People can have honest disagreements about what makes a good admin, but as I said earlier, RfAs shouldn't be a free-for-all where anything goes.

  It also saddens me that you haven't apologized to ScottishFinnishRadish since then. It really does. I hope that one day you do, truly. I do think all this is very relevant to this discussion because if that's what further vetting of candidates mean, we really are going about things the wrong way. It's easier for bad actors to become admins if genuine and sincere people don't want to put themselves through hell. It's honestly a miracle we still have SFR given their experiences, especially when they didn't really do anything all that suspicious anyways. Clovermoss🍀 (talk) 19:45, 15 November 2023 (UTC)

  ScottishFinnishRadish, I'm sorry for the stridency in my oppose at your RFA, for not posing my concerns first as questions to give you an opportunity to defend yourself, and for not choosing my words much more carefully and adopting a more measured tone. The way I opposed made the RFA more painful than need be. I acknowledge the discomfort my role in your RFA played, and regret that.

  I have not approached you sooner as some of the post-mortem from your RFA suggested to me you wouldn't be receptive to the part that follows, without the passage of more time. The part that follows is ...

  Clovermoss I'm NOT sorry for my criteria being what they are, and an opposer should not have to apologize for their criteria, or be responsible for comments from opposers that followed. Mine have softened over the years at RFA (I now endorse non-content contributors who work in technical areas and don't try to present themselves otherwise, because we need them, too), but other than that, my criteria have remained steady. If someone presents themselves as a content contributor, I'll check that for validity ( see Novem Lingaue's GA). If someone doesn't explain upfront their past editing experience, when it was raised early on at their talk page, I'll wonder why, or what else we aren't being told, and delve into that, too. If someone appears to be ticking the boxes en route to adminship, I'll oppose, particularly if they haven't shown up in the trenches of content creation. SFR's RFA could have been more pleasant if the candidate had dealt with early posts on their talk page from the outset of their RFA. In this case, the strong oppose was because there was not just one of these-- but three. Nominators know that !voters are going to look at the candidate's early editing history, and should be forthright in addressing anything that surfaces in the nomination statements.

  In other words, presented with the same situation today, I would still strongly oppose, but would do it with a hopefully more measured tone, after first asking the candidate why they didn't address their early history upfront, and how they think one GA represents content creation. SandyGeorgia ( Talk) 22:55, 15 November 2023 (UTC)

  SandyGeorgia, I appreciate that you finally apologized in some form to ScottishFinnishRadish. I hope more of the opposers that implied they were a sock back then do so as well. A late apology is better than one that never happens and actions always speak louder than words. :) As I said, I don't typically have issues with people opposing for what they think are good reasons, but those specific accusations were just really beyond the pale for me to not say something. I've spent too much of my short life bottling up emotions and how I think about things. My personal criteria is more lenient than some other voters, but I've never really had much of a problem with people who oppose for reasons like this. We disagree on what makes a good admin and that's fine.

  I would ask you be slightly more mindful of comments like how they think one GA represents content creation? I understand that you have higher standards, but the overall community sentiment on what counts as content creation tends to be a bit more broad. I've had people describe me as a content creator which has always been a bit of a surprise because I don't consider myself to have done anything particularly impressive. But for a lot of people, a single GA is impressive in its own right. In my experience, a lot of people tend to think of a content creator as someone who fits in about here. I'm not asking you to change your own personal standards but to perhaps reconsider how certain things come across? I think that we all have a responsibility as experienced long-term editors to set an example for others. I hope you agree. Clovermoss🍀 (talk) 23:59, 15 November 2023 (UTC)

  Clovermoss, I can certainly be aware and take greater care with how I talk about GAs, particularly during an RFA, but being dishonest isn't one of my strengths, and finding a circumspect way to say "that GA is meaningless" will result in contorted words, complicating my already verbose prose. Having spent about three months of my life at WP:DCGAR-- where I saw even what most people would consider good reviewers passing complete rubbage to GA-- I have today even less of an understanding of the value of GA than I had during SFR's RFA. Sure, I can express that differently, but that won't change the fact that many GAs are meaningless as a measure of content contribution, and if they are going to be used as evidence of content creation at RFA, nominators and candidates alike would advance their case by being sure they are presenting strong GAs with strong reviewers (see my reference above to Novem Lingaue's GA at his RFA). By the way, SFR has pointed out to me on their talk that they now have three GAs, so that's a happy end to that issue. But the value of an FA with respect to RFA is that one's work is criticized before a much broader audience, so we get a measure of how the person responds under the kind of pressure and criticism that isn't present in the one-person-review GA process. SandyGeorgia ( Talk) 13:46, 16 November 2023 (UTC)

  It's also worth pointing out that a lot of editors (I would hazard a guess that it's a majority, given the number of editors we have, the few who participate in GAN/FAC, and the few GAs and FAs we have) are far more concerned with consistent high-quality contributions to the encyclopedia in general, than GA/FA work. For many if not most of us it is more important to improve crappy Stub and Start articles toward C and B class (and I guess A, to the extent that still exists) than to polish the chrome on articles that are already encyclopedic quality. RfA voters that have a wild hare about GAs and FAs make a certain amount of noise at RfA, but it's not clear that this really has an impact on the majority of the particants there. Especially since people get GA "credit" often for fairly minor contributions to an article, yet can work tirelessly on improving a particular article/category or just generally across a lot of weak articles, or both, without pursuing the GA and FA processes, which are bureaucratic and stressy for some people. The broader content work is usually of much more significance than a couple of GA or FA icons. No one would say I'm not a content contributor, but after 17+ years here I have zero FAs, and my GA work is rather incidental (either I made some incremental improvements to something that became a GA, like CornerShot and Jasmin Ouschan, or I wrote the original version, as at William A. Spinks and William Hoskins (inventor), which someone else later pushed through GA without me being much involved in that process). I'm not running for RfA, but the point is that lots of candidates are likely to be in the same class of conscientious content work that isn't related to GA/FA process-wonkery. In a sense, this really comes down to m:Immediatism being a practical reality and m:Eventualism being basically defunct as a wikiphilosophy since WP became one of the most-used websites in the world by the mid-2000s. It is ultimately more important for all of our articles to be at least basically encyclopedic than for a small subset of them to be near-perfect examples of encyclopedist literature.  —  SMcCandlish ☏ ¢ 😼  20:50, 16 November 2023 (UTC)

  This is something that has bothered me a great deal. When I started editing about the same time as you, there were crappy Stub and Start articles wherever I looked. I have limited time, and it the task of improving the thousands of crappy articles to an acceptable condition seemed too overwhelming. Whereas on some subjects I am the world expert. So I chose to bring a small number of articles up to the best condition I could. Hawkeye7 (discuss) 01:53, 17 November 2023 (UTC)

  I fully agree with this. Contrary to some suggestions I'm a strong believer that admins need to understand content work, but a demonstration that they do can happen in many ways. Bringing some start-class BLPs up to scratch is just as good as a GA, possibly better. And many many editors who have done invaluable content work don't have it in them to give something the degree of polish it needs for FA status. I like writing FAs, and even so only a tiny fraction of the pages I work on will end up there. I have no problem evaluating a candidate's understanding of content; I get very frustrated that we evaluate it based on arbitrary metrics and nitpicks with nom statements rather than actual assessments of their writing and policy application. Vanamonde ( Talk) 17:25, 17 November 2023 (UTC)

  Well, everyone should contribute in the way that works for them. We also have editors who do little but technical work (Lua modules, etc.), and that also is helpful. My point wasn't to denigrate FAs, just to point out that over-focus on them at RfA is kind of counterproductive, and dismissive of sometimes years of dedicated work by a volunteer. My own adminship criteria page has an expectation of significant content work, but not GAs and FAs in particular. More like 30% mainspace edits, unless there's a good reason otherwise, like lots of vandal-fighting resulting in many, many user-talk warning templates.  —  SMcCandlish ☏ ¢ 😼  07:46, 17 November 2023 (UTC)

  What I find interesting is that 3 of the 5 people to make WP:RFX300 had no GAs or FAs, unless you count my lists, then it's still 2 of 5 (Tamzin and Floquebeam being the other two). Hey man im josh ( talk) 13:28, 17 November 2023 (UTC)

• I nominated Lourdes for RfA, which was successful. I've gone back through my emails with Lourdes just before the RfA started, and there's a deep dive into contributions, including discussion about Van Diemen's Land v Port Phillip, 1851 which they passed through WP:FAC, and handling the stress they got from the first (Lourdes) RfA. There's absolutely no indication whatsoever from my deep dive of contributions to suggest anything amiss, and this seems to be endorsed by the very strong support the RfA actually had. While I have nominated RfA candidates that haven't been successful (typically because of some shooting themselves in the foot, or just not realising what the community wanted), I'm at a loss to know what else I could have done in this one. Ritchie333 ^{(talk) (cont)} 16:07, 14 November 2023 (UTC)

  For me the point in time that something could have been different was doing a username block when Lourdes was named Russian Red. That block would have either forced them to verify their identity or rename and essentially admit they weren't them. I'm not sure that would have actually stopped an RfA in the end but is, for me, the only place where I feel like our systems really let us down. Otherwise it's about accepting known trade-offs and, at least for me, solving the problem here isn't worth the cost of doing so. Best, Barkeep49 ( talk) 16:19, 14 November 2023 (UTC)

  I suppose the only "smoking gun" that's obvious is that I didn't actually seek Lourdes out for adminship, she asked me via email on 16 January 2018. Ritchie333 ^{(talk) (cont)} 16:23, 14 November 2023 (UTC)

  Which, of course, by itself shouldn't be disqualifying. Eddie891 ^Talk _Work 16:26, 14 November 2023 (UTC)

  I don't want to discourage reflection on what our vetting process should be, but if we begin from the assumption that something was wrong with the process for Lourdes, I think we're going to be barking up the wrong tree. Everything I've seen about the Lourdes situation so far suggests Lourdes was attempting to game the system entirely for their own amusement; there wasn't promotional intent, or other disruptive behavior, exhibited so far (it's possible, of course, that the end goal was to Bad Things as a member of ARBCOM). As such what Lourdes was trying to do was climb the rungs of power largely by being a good editor, and they had many years of experience as Wifione to help them determine exactly what that entailed. We can, should, and generally do examine an editor's record thoroughly before giving them advanced permissions; but no amount of scrutiny is going to reveal a bad-faith actor who has gotten to where they are by being a good editor. I'm not saying that Lourdes didn't do silly things, but the most ill-considered actions came late into their admin career, and even then I believe precious few of those actions have been reversed following community scrutiny at AN. Vanamonde ( Talk) 23:21, 14 November 2023 (UTC)

  Yes, I haven't of course read this immense thread in its entirety, but it doesn't seem that anyone has produced a very practical suggestion for what "better vetting of admins prior to RFA" would mean in practice. Maybe time to move on. Johnbod ( talk) 02:23, 15 November 2023 (UTC)

Arbitrary break

I know hindsight is not that useful a tool, but a couple of things spring to mind. When the person operating the Lourdes account claimed to be the well-known person in question, the account should have been blocked pending evidence of ID ( WP:IMPERSONATE only says "may be blocked as a precaution against impersonation until the user's proof of identity is provided", but is it time to make that a harder rule?). Secondly, the account's editing hours (see https://xtools.wmcloud.org/ec-timecard/en.wikipedia.org/Lourdes) fit someone living in India, not Spain or Los Angeles - at least they do now, but I don't know what it was like at RfA time.
I'm not sure if this gives us anything useful for the future. But, at least disallowing any account at RfA with a link to a famous name without proof of ID could have stopped the potential impersonation damage in this case. And check people's editing patterns fit in with anything they say about where they are or where they're from? Boing! said Zebedee ( talk) 09:56, 15 November 2023 (UTC)

All a timecard really tells you is when someone does or does not edit Wikipedia. People don't always work or sleep the same hours as other people. Clovermoss🍀 (talk) 12:32, 15 November 2023 (UTC)

Yes, but when the timecard shows activity spread across the typical working day in India, it's at least a hint. And I doubt the real Lourdes, having a fairly public life on the West Coast of the USA, would be mostly nocturnal. Boing! said Zebedee ( talk) 12:37, 15 November 2023 (UTC)

In this case it might have provided a hint. In general, it's not likely to tell you much. I used to work nights and at other times I've worked very late/very early shifts and kept strange hours closer to some Asian time zones or the US west coast than what you'd expect for someone in the UK. I'm sure many Wikipedians keep atypical hours, especially with the post-Covid rise in working from home. HJ Mitchell | Penny for your thoughts? 12:44, 15 November 2023 (UTC)

Yeah, I'm about to go to sleep myself and it's 7:45 am in my timezone. Clovermoss🍀 (talk) 12:45, 15 November 2023 (UTC)

Yes, I'm not suggesting there's anything conclusive at all, just that there was a hindsight hint here... I wasn't expecting some sort of Spanish Inquisition ;-) Boing! said Zebedee ( talk) 12:53, 15 November 2023 (UTC)

Nobody expects the Spanish Inquisition! ;) HJ Mitchell | Penny for your thoughts? 13:00, 15 November 2023 (UTC)

Thankfully nobody ever asked me for my ID, and I am not sure how I would react if they did. Ymblanter ( talk) 13:20, 15 November 2023 (UTC)

Do you think you would have objected to being asked to send an email from an academic email account to OTRS if asked, providing it was not retained after being seen? That's all that's generally required, and it wouldn't be revealing anything that you're not already open about. Boing! said Zebedee ( talk) 14:37, 15 November 2023 (UTC)

I have revealed relatively little information about myself onwiki so there isn't a simple "do x to verify X thing Barkeep49 said". So I'm not sure how it would work for people like me. Best, Barkeep49 ( talk) 16:36, 15 November 2023 (UTC)

If you're not claiming to be anybody notable, then I don't think it's such an issue. I'm really only thinking about how to stop WP:IMPERSONATE violators - because, for me, that's by far the worst thing our fake Lourdes did. Boing! said Zebedee ( talk) 17:31, 15 November 2023 (UTC)

I disclosed my employer to checkusers and UTRS, which I'm sure was verified to the IP address I was on. It wasn't terribly effective at RFA. ScottishFinnishRadish ( talk) 17:04, 15 November 2023 (UTC)

With that one I would likely be fine, though I guess many people would not. Ymblanter ( talk) 17:09, 15 November 2023 (UTC)

Question from someone who has been distracted with other stuff lately and has not been able to follow the Lourdes saga aside from passing mentions: Did this "rogue" admin actually use the admin bit for nefarious purposes before being caught? From what little I can tell, it's really more of the embarrassment factor of "how did we not catch this" than any actual nefarious actions. I've not heard anybody state that Lourdes blocked Jimbo, deleted the main page or unblocked hoards of LTA's. Where I'm going with this is asking is there an actual problem that needs to be solved by toughing up admin vetting? Or is it more like we were caught with our pants down and want to do something for the sake of telling people we did something? Honest question despite any snark. Dave ( talk) 15:49, 15 November 2023 (UTC)

Essentially, the full details are here, an Arbitration case was started after some editors thought Lourdes was badgering opposition at RfA excessively, and follows on from an incident last year where they unblocked an administrator who had been blocked for "hate speech" without discussion or anything resembling a consensus. Ritchie333 ^{(talk) (cont)} 16:19, 15 November 2023 (UTC)

I find it sickening that a responsible and kind user was brought through hell after they were admonished for a misuse of Checkuser after suspecting something amiss, and it turned out that the perpetrator actually was a sock. I think we should spend less time trying to add more hurdles to RFA, and almost certainly having no major changes done (per usual) and more time rehabilitating TheresNoTime for being the victim of a very treacherous long-term abuser. The Night Watch (talk) 17:32, 15 November 2023 (UTC)

This doesn't really make sense to me. Athaenara was the one who had the most participation in that imbroglio anyway, all Lourdes had done was unblock her. Athaenara wasn't, and isn't, a sock of anybody. The specific checkuser in this case was run against Lourdes on a hunch that she was Athaenara. It wouldn't have turned up anything about Wifione, and nobody suspected that anyway. jp× g 🗯️ 21:24, 15 November 2023 (UTC)

I guess moreso the irony is the part that was sickening. TNT used checkuser on Lourdes presuming that user was a sock, was admonished for misusing checkuser, and then almost a year later it turns out that Lourdes was indeed a sock (just not Athaenara's). The Night Watch (talk) 00:08, 16 November 2023 (UTC)

The most egregious behaviour was an admin impersonating a real-life person, and it led to threats against them. Boing! said Zebedee ( talk) 16:31, 15 November 2023 (UTC)

Two years ago we had determined through an RfA reform process that we all agreed RfA was broken, and in particular, potential admin candidates who would make great administrators did not want to become admins because they did not want to go through RfA. That being said, throughout this thread, semblances of the following suggestions _{(please note, this is purely rhetorical and for the purpose of argument, thus is not intended to be an exhaustive detailing of the positions or those who hold them)} have been tabled as possible optional or mandatory ideas for RFA to prevent another Lourdes situation:

• Full-blown CheckUser for RfA candidates to the furthest extent possible.
• RfA candidates required to submit real-life identification to the Foundation.
• Restricting RfA !voting across the board only to editors with at least 500 mainspace edits and/or a clear understanding of Wikipedia principles and policies.
• Recommending or requiring that the candidate meet the nominator through in-person or on a video call.
• Comparing the user's online editing hours with their stated location to determine that they are using Wikipedia at socially acceptable hours based on their timezone.

And maybe this is all worthwhile. But to me my views align with those said by a couple of others: We're assuming that we should always be able to catch bad actors at the RfA level, meaning that RfA can ideally become a perfect, bulletproof process where no bad actors can get to the point of becoming administrators, and that if we can't stop it at that point, it's somehow our fault as community members for not asking tougher questions or having higher standards. But really, human beings don't operate on a time-insensitive continuum of "good" and "bad". WP:AGF means we assume good until provided evidence after the fact to convince us of bad. After the fact, not before the fact. _{Duly signed,} ⛵ WaltClipper -( talk) 18:20, 15 November 2023 (UTC)

No the fuck there does not. Many people have already said a lot on this, so I will not reiterate every point that's been made, but the process already consists of a seven-day-long job interview, conducted in public, and preserved permanently for the world to see. That isn't enough vetting for you? Per my own nomination statements, I am also a sockpuppet. I haven't logged into said account in decades, but it is true, I had a Wikipedia account when I was eleven years old and I was stupid. Do you want my scalp too? A lot of people were asking me why I delayed so long in running: that's why. I was not particularly interested in having a bunch of Inspector Javert LARPers comb through my history to see if they could suss out any clues as to what sort of embarrassing stuff I said on here when I was eleven years old, go hunting for gems where I said the stupidest thing, and then post it publicly on a document bearing my real name (JPxG are my real initials) forever. By the grace of God, no such characters appeared. But now your proposal is -- what -- to add yet more humiliating ordeals to the gauntlet? Good grief. What we need is a selection process that "the candidate to stand naked in an amphitheatre for a week and the only way to oppose is to publicly throw rotting vegetables at them". jp× g 🗯️ 20:36, 15 November 2023 (UTC)

Thank you for saying what I felt afraid to say. When there are so many voices above - trusted voices - who are calling for more layers of difficulty and complexity, it's a breath of fresh air to hear such a response as yours. _{Duly signed,} ⛵ WaltClipper -( talk) 21:12, 15 November 2023 (UTC)

Thank you for saying this. I think some people are forgetting that perfect is the enemy of good. The truth is, there is no real way to 100% eliminate any would-be bad actors without eliminating many great editors. ULPS ^{( talk • contribs)} 21:45, 15 November 2023 (UTC)

This is exactly what I was afraid of and why I asked the question above. I'm of the opinion that changes to processes/procedures/policies should be to fix something that needs to be fixed. I'm not a fan of "OMG something bad happened, so let's make a ton of changes to make it look like we care about fixing the problem even know we know in truth they wouldn't have stopped this or similar incidents". This is why the governments of the world spend millions of dollars per year paying people to ensure your carry-on bottle of shampoo has 3 ounces max, not 3.1ounces, when you fly. That's among many, many, many other policies and laws where I would say the cure is worse than the disease, except the cure didn't even cure anything. Dave ( talk) 17:03, 16 November 2023 (UTC)

I largely agree, but I do think that "OMG something bad happened, are there any changes we can make that would be of net benefit?" is valid. It looks like the answer is probably no in this case, but I'd say the question was worth asking. Boing! said Zebedee ( talk) 10:49, 17 November 2023 (UTC)

I agree it was a worthy question and the discussion will lead to changes, albeit not major structural or process ones. I was tempted to engage SMcCandish's statements about how to view and evaluate content contributions in RFA when a candidate presents as a content contributor, but decided that's another distraction that would be TLDR, and going too far afield of what this discussion could have focused on (RFA is about trust, and opposes on trust are discounted, while supports are not), but unfortunately did not, so maybe that's a conversation for another day. SandyGeorgia ( Talk) 16:11, 17 November 2023 (UTC)

A very determined actor or group of actors plans a sophisticated attack that catches the public off-guard. Stunned, the public requests vastly heightened gatekeeping of access to the means of attack, leaping to tools that can easily be bypassed, and unethical tactics that chase away good actors while only mildly inconveniencing bad ones, but beget a false sense of security among the public. This feel like déjà vu to anyone else? -- Tamzin^{[ cetacean needed} (they|xe|she) 22:26, 15 November 2023 (UTC)

Someone might come up with the solution that actually seriously impacts the bad actors (e.g. locking cockpit doors). Firefangledfeathers ( talk / contribs) 22:30, 15 November 2023 (UTC)

Be careful what you wish for. — Kusma ( talk) 22:58, 15 November 2023 (UTC)

Is any Wikipedia discussion really complete if it doesn't include at least one absurdly hyperbolic analogy? ;D Shells-shells ( talk) 22:46, 15 November 2023 (UTC)

How about a bold prediction: Someday a user will pass RfA and later turn out to be a robot. (Maybe it's already happened.) You read it here first. 🤣 _{Duly signed,} ⛵ WaltClipper -( talk) 20:02, 16 November 2023 (UTC)

How insightful! I should write that down somewhere. Shells-shells ( talk) 21:48, 16 November 2023 (UTC)

Already discussed, back in early April 2010. User:WereSpielChequers/AI accounts might need updating at some point. Ϣere SpielChequers 01:24, 17 November 2023 (UTC)