Wikipedia talk:Arbitration Committee/Noticeboard

Behaviour on this page: This page is for discussing announcements relating to the Arbitration Committee. Editors commenting here are required to act with appropriate decorum. While grievances, complaints, or criticism of arbitration decisions are frequently posted here, you are expected to present them without being rude or hostile. Comments that are uncivil may be removed without warning. Personal attacks against other users, including arbitrators or the clerks, will be met with sanctions.

Wikipedia:Arbitration/Requests/Case/Geschichte closed

Original announcement

Five-word summary: Geschichte's administrator privileges are revoked. —Jéské Couriano v^_^v ^{a little blue Bori} 21:34, 31 July 2022 (UTC)

Sad but necessary. Or sadly necessary. Deepfriedokra (talk) 22:59, 31 July 2022 (UTC)

Sadnecessary. Trainsandotherthings (talk) 23:00, 31 July 2022 (UTC)

On a related note, I'm thoroughly disappointed and grateful to all the ex-admins who decided to go for a 172-styled exit - disappointed because they would rather not face their mistakes, and grateful because it means ArbCom doesn't have to juggle too many cases at once, given the marathon pace of most Arbitrations nowadays. —Jéské Couriano v^_^v ^{a little blue Bori} 02:43, 1 August 2022 (UTC)

5 recusals! Izno (talk) 03:17, 1 August 2022 (UTC)

That case is the first time something like this was done, and that's what you latch onto? —Jéské Couriano v^_^v ^{a little blue Bori} 03:38, 1 August 2022 (UTC)

• What's sad is that it could have been avoided with so little effort but they were unwilling. Dennis Brown - 2¢ 15:11, 4 August 2022 (UTC)

  Indeed, I think we'd all be happy if the admins dragged to ArbCom over misuse of tools didn't just go the 172 route and quit Wikipedia in an effort to dodge Arbitration. An admin who's hesitant to defend their actions with the tools is a red flag in itself. —Jéské Couriano v^_^v ^{a little blue Bori} 16:29, 7 August 2022 (UTC)

  It does have real "you can't fire me, I quit" energy whenever this sort of situation happens. Personally I think everyone should be willing to defend and be held accountable to* the actions they make, both on and off-wiki, but alas that doesn't seem to be a universally held view.

  *some exceptions apply, eg health related Sideswipe9th (talk) 16:36, 7 August 2022 (UTC)

• Overall, I think this approach was the right fit. It gave the subject more than adequate time to respond, while also insuring they could not just ignore it and retain the tools. The aspect which didn't work all that well is that March to August is more than three months. We have another admin conduct case that is going to hit its expiration date later this month, hopefully we'll get that dealt with in a more timely fashion. --Beeblebrox (talk) 16:39, 7 August 2022 (UTC)

Wikipedia:Arbitration/Requests/Case/Conduct in deletion-related editing closed

Original announcement

• Ummm, what? How is the remedy of warning Lugnuts about behavior in deletion discussions going to be useful, when you have also banned him from them, and then also site banned him? An un-site ban request could always include conditions if they were to be useful later. — xaosflux ^Talk 22:26, 2 August 2022 (UTC)

  That assumes that Lugnuts' blowup yesterday - which looks more like a suicide-by-administrator action in my eyes - didn't burn any bridges with the community. In fact, I'm pretty certain that blowup was sparked by the voting on the siteban for him, as it makes direct references to some of the arbitrators' rationales. —Jéské Couriano v^_^v ^{a little blue Bori} 22:30, 2 August 2022 (UTC)

  This might have gone very differently. The total ban was only barely passing when he detonated, and at least one of the arbs who'd voted in favor of it hadn't edited the page since before the middle-of-the-road topic ban was proposed. —Cryptic 22:45, 2 August 2022 (UTC)

  If the ban is appealed (however unlikely that now seems) the warning and topic ban will presumably remain in place for some time before becoming successfully appealable. The warning isn't entirely redundant with the topic ban - "disruptive deletion behavior" is broader than deletion discussions and deprodding - and I'd expect it would remain in place after an appeal of the topic ban besides. —Cryptic 22:45, 2 August 2022 (UTC)

  In real life a person can be served up to multiple life imprisonment but I think a site-ban supersedes the topic ban imposed on Lugnuts. Thingofme (talk) 01:05, 3 August 2022 (UTC)

  If one wanted to nitpick, one could say that a topic ban's scope isn't entirely subsumed by a siteban's, because in most cases (although not this one) a sitebanned user can edit their talkpage, but would still have to comply with the TBAN on that page. Setting that aside, though, yes, a siteban supersedes a topic ban while the siteban is in effect. If the user is un-sitebanned, the topic ban remains in effect. And there have been a lot of users over the years who've gotten a TBAN+siteban at once, with the latter but not the former later repealed. -- Tamzin^{[cetacean needed]} (she|they|xe) 01:11, 3 August 2022 (UTC)

  The block (is) the enforcement of a official ban and they may include topic bans but after a site ban is appealed, the topic ban remains in effect and they have to appeal again the topic ban... Thingofme (talk) 01:52, 3 August 2022 (UTC)

  I admit I was a bit surprised by the warning, in part because of the triple action against Lugnuts and in part because no one else was warned and banned although I definitely recall a number of cases in the past where someone was both warned and banned in some way. And by the same token, I'm sure I recall a number cases in the past where someone was both sitebanned and topic banned, with the assumption if they want to come back the siteban will be lifted, but they will need to prove via editing for their topic ban to be lifted. While unban conditions could be imposed which include a topic ban, IMO it's fine for arbitrators who've viewed all the evidence hopefully in depth and considered everything carefully to decide if they feel if the editor ever comes back, these are some conditions that will be imposed. It also helps clarify to the editor some of what they should expect and may make the appeal less stressful. While more conditions could still be imposed on appeal of the siteban, I suspect it's less likely when there are already conditions ready to be imposed. The only thing which makes this case slightly unique (although I seem to recall this happening at least once before), is what the editor did which made a siteban appeal far more difficult. I guess this gets to the life sentence bit, on wikipedia there isn't really supposed to be a life sentence except perhaps some of those imposed by the WMF. (I know the WMF has an appeal mechanism now, but there seems to be some recognition some bans explicitly those done for childprotect reasons may never be lifted.) An indefinite block or ban is explicitly supposed to be something an editor may eventually successfully appeal. This case may have unfortunately gone into one of those where it's likely to be far harder, however I still can't fault arbcom for applying the principle that it's expected an appeal may one day be successful. Nil Einne (talk) 09:28, 3 August 2022 (UTC)

• (edit conflict) Wow. Sorry it came to this. I liked and will miss @Lugnuts:. Hopefully, they will eventually regain their perspective and return. There is hope in redefining oneself and one's role on Wikipedia. I returned after a three year hiatus, adapted to the changes, and now do things I never considered before. Redefining oneself may or may not come more easily for those who are not blocked and site banned. These are long term users who had been net positives before this debacle. I have only hope and best wishes for them all.

@Xaosflux: I thought the same, perhaps Lugnuts can take this all in for the eventual return-- Deepfriedokra (talk) 22:38, 2 August 2022 (UTC)

I can't imagine how hurt and humiliated they feel. Particularly Lugnuts. So the explosive departure is probably a manifestation of humiliation and hurt. We sometimes forget we are dealing with people with feelings. -- Deepfriedokra (talk) 22:42, 2 August 2022 (UTC)

• In light of the full text of the remedy about requesting community comment, it would be good if ArbCom makes some kind of announcement that includes those details and/or gets the process under way. --Tryptofish (talk) 22:43, 2 August 2022 (UTC)

  Thanks for following up on this Trypto. I will make sure this stays on our agenda and will hope to have a public update soon, though for ArbCom soon means closer to weeks than days. Best, KevinL (aka L235 · t · c) 23:37, 2 August 2022 (UTC)

  I will be disappointed if our next announcement isn't closer to days than weeks. Barkeep49 (talk) 23:49, 2 August 2022 (UTC)

Is the banned user template for editors banned by ARBCOM supposed to be hidden? It was apparently caused by this change [1] in 2016. suppress the banner on user pages of users banned by arbcom, and instead add them to a hidden category starship.paint (exalt) 06:29, 3 August 2022 (UTC)

@Starship.paint: apparently, yes, but it really shouldn't be. Still shows up in the category of users banned by ARBCOM -- Rockstone^{Send me a message!} 06:41, 3 August 2022 (UTC)

I checked the history of my userpage, for when I was banned from Apr 2013 to May 2014. The template can't be seen, during that time period. GoodDay (talk) 06:50, 3 August 2022 (UTC)

Yes. It was a good change. Best, KevinL (aka L235 · t · c) 06:51, 3 August 2022 (UTC)

@L235: Out of curiosity, why was it changed? -- Rockstone^{Send me a message!} 08:20, 3 August 2022 (UTC)

@Rockstone35 I believe this was covered in this discussion which I see you've already been a participant in. Johnuniq covers it well at that discussion. Best, KevinL (aka L235 · t · c) 08:25, 3 August 2022 (UTC)

@L235: Huh, that does answer my question. Somehow, I barely remember participating in that! See you when I ask the same thing in 2025! -- Rockstone^{Send me a message!} 08:28, 3 August 2022 (UTC)

Related to this, these edits have shown up on my watchlist: [2], [3]. I have a gut feeling that this should not have been done, and it was flawed reasoning to conclude it from the Lugnuts example. But I don't want to act unilaterally, as it's above my paygrade, so I'm asking for clarification here. Courtesy ping to @Yleventa2:. --Tryptofish (talk) 19:58, 3 August 2022 (UTC)

This is now being discussed at WT:RFAR#Edit request: banned user talkpage. --Tryptofish (talk) 20:18, 3 August 2022 (UTC)

• Meanwhile, the whole deletion-related area is still a toxic dumpster fire. Black Kite (talk) 09:29, 3 August 2022 (UTC)

  I agree that the community is continuing to struggle with how to handle the large changes to the Sports SNG. That is in scope of the RfC. If you have other ideas of how we should address it I am open to those thoughts, because I too recognize the issues that are present. Barkeep49 (talk) 14:24, 3 August 2022 (UTC)

  I've been in a 'few' Sports-related AfDs. My response to any questions on what position I take is simply "I'll let the closer decide on whether my 'post' should be ignored or not". I never let myself get bludgeoned. -- GoodDay (talk) 15:11, 3 August 2022 (UTC)

  @Black Kite: And I don't have a hazmat suit. I abandoned all hope for that part of W long ago. Some things never change. I remember reading of an ArbCom case because an admin pointily deleted what was then Wikipedia:Votes for deletion. (Before my time.) -- Deepfriedokra (talk) 20:24, 3 August 2022 (UTC)

• I have not got much time recently, and I try to avoid AfD when possible, so I did not follow the case closely, and I do not have any opinion on its merits. What I would like to say, however, is that back in 2018, Lugnuts for some reason unclear to me was following me around, commenting in the same discussions I commented, trolling, and trying to keep as close to the line as possible without formally crossing it. At some point, I overreacted, and I got a warning of an administrator that I would be blocked soon, and she was really serious. (Lugnuts did not receive a warning). This was the only serious block warning which I have got on any Wikimedia project in 15 years. I stopped editing, stopped checking my watchlist, but at the end decided to come back and continue editing. I am writing this to say that Lugnuts had the guts, completely unprovoked, a year later to come to my talk page and apologize. Most of the users who behaved in a similar way never did, most of them doubled down when they had a chance. I hope Lugnuts will find some way to appeal the ban and to become a constructive user.--Ymblanter (talk) 18:45, 3 August 2022 (UTC)

  It is nearly always my hope that when someone is banned, they see it as an opportunity to take a break, and maybe return in the future after re-assessing what was motivating them to act in such a way that it led to them being banned. So, even though I was the first one to vote in support of banning them, I hope the same. It's unfortunate that they chose to go out the way they did, that will certainly make it more difficult to return, but this community can be very forgiving if a person is patient, honest, and sincere when they do appeal. Beeblebrox (talk) 20:08, 3 August 2022 (UTC)

  There's a reason that bans can be indefinite, but never lifetime or forever - we always have some hope that the user will understand the error in their ways and make a successful appeal. Most don't, and for some it will be harder than others, but the option is certainly there. 93.172.250.2 (talk) 00:32, 4 August 2022 (UTC)

  Global bans are forever. Ymblanter (talk) 05:49, 4 August 2022 (UTC)

  @Ymblanter The foundation established a Case Review Committee a couple of years back to handle appeals for certain types of foundation bans, and community imposed global bans can be overturned via an RFC on meta. 163.1.15.238 (talk) 12:13, 4 August 2022 (UTC)

  Some foundation bans are not forever and maybe in one project, like Fram case in 2019. Thingofme (talk) 15:32, 4 August 2022 (UTC)

  @Thingofme The foundation stopped using time limited or individual project bans a few years back, after the fallout from the Fram ban, see the note at the bottom of Meta:Office actions#Primary office actions. 163.1.15.238 (talk) 15:37, 4 August 2022 (UTC)

  Here's an interesting thing. During my 1-year vacation, many of the items I wash pushing for at some British & Irish political pages (which partially led to my ban), eventually got passed without my participation. So sometimes, the message is correct, but the messenger just isn't arguing for it properly. GoodDay (talk) 13:03, 4 August 2022 (UTC)

• Lugnuts "blowup" was relatively mild as such things go, so I wouldn't worry about it. Maybe more worth mentioning: at bottom, at least regarding Lugnuts, this is a bot or meatbot case. The proposed rfc about mass deletion is also about how to deal with certain [meat]bot issues. That doesn't seem to have been remarked in the case itself.

  The suggestion of requiring Lugnuts to fix his own mess imho reflects naivety or inexperience of whoever made it. It's an attractive and well-meaning idea, but Beeblebrox got it right: it never works. Anyone who has seen enough of these cases should understand that by now. There is also supposedly a Korean saying, that if a doctor makes you sick, don't go to the same doctor to ask him or her to make you better. I don't see much new here. Fast and repetitive editing (typically automated even if the person doesn't admit this) makes people crazy. This is just another example.

  Maybe I've missed a few incidents but the mass deletion RFC doesn't sound that promising. Mass deletion proposals arise from different kinds of situations, and treating them case by case has been ok as far as I know. Before launching an RFC, I'd suggest doing a study of past situations to see if there are enough observable common patterns to support generalization. I wouldn't support proposals that aren't based on observation of that sort. 2601:648:8201:5DD0:0:0:0:34C5 (talk) 20:11, 14 August 2022 (UTC)

  • That was a blowup? Drmies (talk) 21:16, 16 August 2022 (UTC)

    Krakatoa -- Deepfriedokra (talk) 21:19, 16 August 2022 (UTC)

Community Comment and RFC

One of the next steps listed is the development of an RFC on how to handle mass nominations for deletion, and possibly how to deal with mass creations. Two arbitrators, User:L235 and User:Barkeep49, have referred to the fact that this is a next step. I think that there should be a preliminary period of discussion before the publication of the RFC, intended to refine the scope and wording of the RFC. I am willing to moderate the preliminary period of discussion, which will include asking the community for specific neutrally worded questions to include in the RFC. I think that at least some editors agree that some RFCs fail to determine consensus because they are not properly worded, and that it will be useful to work on an RFC of this importance before it is published.

Moderated discussion in the development of the RFC will not be the same as the role of a moderator after the RFC is published. Post-publication moderation will consist largely of collapsing off-topic or tangential discussion, while pre-publication moderation has to do largely with determining the wording of the RFC. Robert McClenon (talk) 05:43, 4 August 2022 (UTC)

Above all, we'd have to be extra determined not to allow bludgeoning, in such an RFC. GoodDay (talk) 13:06, 4 August 2022 (UTC)

Given that there's an explicit ask for the mod(s) to consult the community there will definitely be a pre RfC phase. Barkeep49 (talk) 14:29, 4 August 2022 (UTC)

I agree with User:GoodDay that it is necessary to prevent bludgeoning. How I would try to prevent it as moderator would be to provide a section for back-and-forth discussion and to disallow back-and-forth discussion or responses to other editors except in the section for the purpose. Pre-RFC, the moderator asks questions, and the editors address their replies to the moderator and the community. Back-and-forth discussion is permitted only in the section for that purpose. In the RFC, answers to the questions with supporting statements and back-and-forth discussion are again in separate sections. If back-and-forth discussion gets out of hand, or one editor comments on everything, the discussion will be collapsed; but any out-of-place replies will immediately be collapsed with a warning. Robert McClenon (talk) 19:22, 4 August 2022 (UTC)

When is it going to open? –LaundryPizza03 (dc̄) 23:16, 4 August 2022 (UTC)

The arbs are actively discussing possible mods and closers. I'm hoping we can announce those (or at least the mod) soon so that the community consultation phase can begin. Barkeep49 (talk) 23:45, 4 August 2022 (UTC)

Reviewing the Stubs

Another, unrelated next step is that there should be a process to review at least some of the stubs created by Lugnuts, in particular to selectively check them to sample whether they contain errors or copyvio (based on the parting comment that may have been trolling), as well as to answer (based on a sample) any other questions. Robert McClenon (talk) 05:43, 4 August 2022 (UTC)

I believe some people did/have been doing spot checks at a CCI and found no CV issues (most of the articles are too short to have any real CV in them). Not sure if they were checked for facts but IMO those comments were trolling and unlikely to be factual. ♠PMC♠ (talk) 06:41, 4 August 2022 (UTC)

That is what some of us thought, and is a relief. Trolling by an editor who is being sanctioned is very undesirable, but not nearly as bad as what was falsely described, which would have been long-term systemic vandalism. Lugnuts is a stubborn editor, not a vandal. Genseric was a Vandal. Robert McClenon (talk) 12:51, 4 August 2022 (UTC)

True enough, but I've lost what shred of sympathy I might have had for them when they went out like that. The display of utter contempt for community processes is indicative of what Lugnuts really thought about Wikipedia, that it was their show, the rest of us be damned. The above expressions of desire that Lugnuts may one day return in light of the years of ANI discussions and their reaction to this case mystify me. -Indy beetle (talk) 14:35, 4 August 2022 (UTC)

I politely disagree, and I have sympathy. To make 93,000 Wikipedia stubs is an extraordinary commitment to building an encyclopaedia, and I believe that a lot of Lugnuts' interactions were with editors who liked and admired him. He thought he was a rock star -- he clearly wasn't expecting to be sitebanned, and when he realized that was what was really happening, I think he reacted like an employee who sees security coming to escort him out of the building. He kicked over the wastebasket, spat on the floor and waved two fingers at his manager as he went. That's not really contempt for the community, it's stress and pain coming out. We can forget him: he's not going to start again from scratch and try to write another 93,000 stubs.

As for dealing with his 93,000 stubs, I would propose that we create WP:CSD#X3 -- a special speedy deletion criterion that applies only to stubs created by Lugnuts that have never contained more than 500 words and never cited any source that isn't a database.—S Marshall T/C 21:58, 4 August 2022 (UTC)

They are not enough to speedy deletion, they may be reconsidered by case-by-case basis. We don't create new CSD criteria to solve articles created by solely one editor, see WP:NEWCSD. Thingofme (talk) 01:19, 5 August 2022 (UTC)

Instead, we should open 100 AfDs a day for two and a half years to consider them on case-by-case. ScottishFinnishRadish (talk) 01:26, 5 August 2022 (UTC)

Yes, we do. WP:X1 * Pppery * _{it has begun...} 01:40, 5 August 2022 (UTC)

Pretty sure Neelix once had a special CSD to remove all of his unwarranted redirects, but that may have been too long ago. (EDIT: was ninja'd by Pppery.) — Ceso femmuin mbolgaig mbung, mellohi! (投稿) 02:13, 5 August 2022 (UTC)

Special CSDs for one editor can be represented as G5 but this is only a violation of his ban, however those stubs maybe contestable and not merit a CSD (we only consider very short stubs, less than 500 words) Thingofme (talk) 04:17, 5 August 2022 (UTC)

I think that discussing or organizing a review of his articles is probably better-suited for WP:VP or possibly WT:CSD or WP:AN. IznoPublic (talk) 04:52, 5 August 2022 (UTC)

I have been reviewing many of the women's cricket articles created by Lugnuts, which he conveniently listed in User:Lugnuts/Cricket#Women cricketers and, as a sample, I would say well over 50% need attention. Typically, they contain two or three short sentences with an information box. All the information has been lifted from a statistical database and converted to basic text. They are little use to anyone and must surely discourage readers, so they are a constraint on Wikipedia and not a benefit as the cricket project members apparently believe. I certainly think they should be removed somehow and the ideas raised above seem to me to be worth pursuing.

Lugnuts is not, however, the sole creator of large numbers of minimal info stubs in the cricket project and I think the community needs to include other members of that project in its deliberations. AssociateAffiliate, for example, claims to have made over 60k creations. I looked at these and the vast majority of them are stubs with a size of 3–5 kB, if that. I would propose a CSD#X3 for all cricket articles, not just those by Lugnuts, which do not meet the 500 word limit suggested by User:S Marshall above.

I am going to be away for a period after today but I will keep this in my watchlist and will try to help when I come back. Thank you.

Sistorian (talk) 13:02, 5 August 2022 (UTC)

The problem extends beyond cricket, and even beyond sports. I believe we need to do two things. First, we need to stop future mass creation of articles without community consensus. Second, we need to address the historic mass creation of articles.

For the first, we need to make it clearer that WP:MASSCREATE applies to large numbers of similar manually created articles, and we need to monitor article creation to identify when editors are violating this requirement. We also need a clear process for approving mass creation; approval should require a clear scope and an RfC.

For the second, I am looking to identify articles sourced only to database sources; unfortunately due to templates like Template:Sports links this list is extensive, and individual items will need to be discussed. Once identified, I suggest we group the articles in some manner - by creator, by creation month, or by category - and have a discussion proposing that they are all moved out of article space. If moved interested editors would be free to improve them or create new articles in their place, if they meet WP:GNG. BilledMammal (talk) 14:13, 5 August 2022 (UTC)

• No, we should not be speedy deleting all cricket stubs solely because they're stubs. BeanieFan11 (talk) 14:36, 5 August 2022 (UTC)
• We need to implement an efficient process to delete those. North8000 (talk) 14:54, 5 August 2022 (UTC)

  I thought we were here to build an encyclopedia. There is no "efficient process' to determine the notability of the subject of any article. Even a stub. A thorough WP:BEFORE and then a WP:ProD or WP:AfD -- Deepfriedokra (talk) 15:10, 5 August 2022 (UTC)

If it were up to me (but it isn't), I'd have all the stubs deleted. Re-start with a clean slate. GoodDay (talk) 15:23, 5 August 2022 (UTC)

You'd support demolishing more than a third of the entire website?!? BeanieFan11 (talk) 15:27, 5 August 2022 (UTC)

Lug's stubs, only. GoodDay (talk) 15:35, 5 August 2022 (UTC)

The way that that the big fuzzy wp:notability system actually works Wikipedia:How Wikipedia notability works is that topics that are clearly highly encyclopedic / have unusually strong compliance with wp:not get a more lenient notability sourcing standard applied. Regarding stubs, this applies to geographic places and species and maybe one more highly enclyclopedic category that didn't come to mid. IMO those should stay. Once you set those aside, the remaining stubs are are probably about 10% of Wikipedia's articles and roughly 0% of it's enclyclopedic content. And I'd say yes, delete all of those. A good place to start would be all "database only" stubs about individual people. Or Lug's stubs about individual people. North8000 (talk)

I agree with GoodDay and in part with North8000 relative to keeping the species stubs. As for the geographic places, there's no shortage of maps and tourist brochures, so delete them rather than committing other volunteers to spending countless hours verifying their notability. Atsme 💬 📧 16:14, 5 August 2022 (UTC)

I agree, in particular bios, living or not. Stubs don't really helps the encyclopedia with these, plus are not watched by anyone, adding to the potential for abuse. With plants and animals, those tend to get fleshed out over time, as the parent categories get worked on my knowledgeable individuals. Same with geography stubs, something I actually work on. Often, it is just taking one or two sentences and turning it into two paragraphs, but that is still useful for towns, counties, etc. Geo stubs or species can be deleted or kept, depending on if they have at least something credible as a source. Dennis Brown - 2¢ 16:18, 5 August 2022 (UTC)

Fwiw I'm fairly sure Lugnuts had the vast majority of their stubs on a watchlist - based on what happened whenever I edited one. I imagine my watchlist is 50% stubs at least, although nowhere near as extensive as theirs. Blue Square Thing (talk) 06:12, 6 August 2022 (UTC)

Far more than a third. Levivich 20:23, 5 August 2022 (UTC)

Modified proposal

When you're trying to eat an elephant, do it one plate at a time. Eating the elephant in individual tiny nibbles at AfD is impractical, but so is trying to swallow it whole. The best method for enacting a speedy deletion criterion is to define it clearly and strictly, then ask the community to authorise it at RfC.

I propose that we set up a series of subpages with names like Wikipedia:Articles for deletion/Projects/Lugnuts stubs, Wikipedia:Articles for deletion/Projects/BLP stubs sourced only to databases, etc. At each subpage we develop a proposal for a very specific and tightly-constrained temporary speedy deletion criterion. When each proposal's fully developed we move it to Wikipedia:Articles for deletion/Projects/Lugnuts stubs/RfC, etc., so there's a clean-sheet page where we hold a full RfC for the community to decide whether to enact it. These RfCs should be listed at WP:CENT. If passed, they should generate a fresh index number, so the first to pass is WP:CSD#X3, and others are CSD#X4 and on.

After my experiences with WP:CSD#X2 I suggest that before each RfC is closed, we work up a complete list of the pages that would be affected, identify at least three named editors who're willing to work through each backlog, and at least two named sysops who're willing to actually carry out the speedy deletions when the stubs are appropriately reviewed and tagged. It's particularly important that we don't enact a rule that generates an endless string of queries and quibbles from the sysops who're asked to enforce it, so keep things very clear and rigorously-defined at all times.—S Marshall T/C 17:39, 5 August 2022 (UTC)

I must admit. I'm impressed with 'how many' stubs one editor can create. GoodDay (talk) 20:39, 5 August 2022 (UTC)

Like Ovinus (talk) 20:51, 5 August 2022 (UTC)

An issue you'll face will be identifying which articles are sourced only to databases. Some "database only" sources actually contain considerable prose. Take, for example, CricInfo, a source likely to be considered a "database". Well, the problem is that it isn't always just a database. For example, this CricInfo profile is clearly not simply a database entry, whereas this one is. I began looking through a list of over 1,000 stubs created by BlackJack which would probably be ones that are similar, if not worse. I'm tending to find that between 15 and 20% have some kind of prose profile in their CricInfo link - so in these cases it's not just a database. The same can be true of CricketArchive and of Olympedia I'm afraid: in some cases they're prose sources as well as database sources.

I've no idea how to resolve this - but there is plenty of evidence that with time and sourcing there are articles which can be developed into absolutely decent quality articles with plenty of sourcing - one nominated at AfD earlier this year now has in excess of 50 sources. Not every article will do that, but I can easily find another two in the same boat: all three Lugnuts articles with only database sourcing. We're in danger of throwing the baby out with the bath water here and losing really nice content - apparently in all three cases the nominator had done a thorough BEFORE - the three articles combined now have 135 sources... Blue Square Thing (talk) 22:55, 5 August 2022 (UTC)

I am genuinely thrilled that these articles are being developed. But if we delete a database stub, what is stopping people from recreating it in with more references? The issue is that a large number of these stubs are not notable. It takes less than five (leisurely) minutes to create a stub, but perhaps over an hour of editor time (often painful and stressful) to conduct an extensive search of sources and conclude a lack of notability. One could argue that these stubs are seed crystals for substantial articles, but the majority are not. There is simply no existing, standard process to clean up the mess.

Anyway, I think this debate has occurred ad nauseam but I think a well-moderated RfC should sort things out. Ovinus (talk) 23:58, 5 August 2022 (UTC)

Ovinus, see Brandolini's law. 😆 Atsme 💬 📧 18:10, 7 August 2022 (UTC)

I was a perhaps minor player in the X1 cleanup project and recreation of any deleted content by a user in good standing was explicitly allowed then, and probably should be now. The idea, then and now, is to streamline the process, and that includes streamlining objections to specific deletions. (it still took nearly three years to complete... hopefully we're not in that deep this time around) Beeblebrox (talk) 00:34, 6 August 2022 (UTC)

Worth noting that in the case of cricketers that there are very often suitable lists to redirect to - i.e. an ATD. That makes expansion much more likely in my view - I don't have to find the sources that act as the seed myself. It might take a little more time though - although, as below, categorisation would help. Blue Square Thing (talk) 06:08, 6 August 2022 (UTC)

Hm, a reasonable compromise would be to recommend eager use of redirect targets (per ATD, as you say), and only speedy delete (or speedy move to special draft space, etc.) those database stubs with no plausible target. One oft-cited concern of the database stubs is that they can "shadow" more notable subjects, but a redirect seems unproblematic: An editor would probably see that, no, 1923 British cricketer John Smith redirecting to XYZ team is not the bloke I want to write about. Either way, the friction for cricket enjoyers to create well-referenced articles is minor—just convert the redirect.

As an aside, thank you, Blue Square, for shaping up many of these stubs at AFD. I trust your comment below, that more of the stubs than I thought, perhaps more than half (!) in some classes of stubs, are candidates for substantial articles. Ovinus (talk) 07:27, 6 August 2022 (UTC) Revised 08:06, 6 August 2022 (UTC)

Just a quickie to reinforce, that’s a subset. Essentially all British county cricketers. That’s where using categories will help.

Redirect will mostly help though. Blue Square Thing (talk) 07:48, 6 August 2022 (UTC)

Thanks for the clarification. Fwiw, taking a look at a fairly random sample of Lugnuts cricket stubs, I haven't found any obvious factual errors beyond the odd typo. Similar sorts of proportions seem to apply in similar situations as well and it would be possible to also create a list of higher priority articles to work on (ones such as William Neale (cricketer) for example) - but redirecting will help this process so much. Blue Square Thing (talk) 10:05, 7 August 2022 (UTC)

Going back to the BlackJack list of a little over 1,000 - my gut feeling is that 50% are very, very clearly candidates for substantial articles. That's a specific sub-set of cricket stubs which might be partly identified using categorisation. If that can be done first then it might be possible to manually identify the articles that we think are definite seed crystals in a relatively short period Billed Mammal started a conversation at the cricket wiki project and I've managed to do this myself for over 300 of the 1000 BlackJack stubs in a couple of days - and, fwiw, I was being harsh in my judgement. Blue Square Thing (talk) 06:08, 6 August 2022 (UTC)

I would note that almost all prose in cricinfo entries is simply statistics in prose form, rather than anything approaching WP:SIGCOV.

However, this is why I prefer moving them out of article space instead of deleting them; though I generally agree with Ovinus, I believe there are a few benefits and no harm in keeping the articles around outside of article space. BilledMammal (talk) 02:43, 6 August 2022 (UTC)

Even if it were true that the prose in CricInfo profiles were simply stats, sports are generally statistical in nature. Of course there are stats in the prose - but that context is what makes then not a database. I thought the point here was that we were identifying the articles which are only linked to database sources: that's the objection, isn't it? In which case you need to filter out the articles that do have non-database sources, surely? And that prose often has plenty of context that would form the basis of a great article: try any of the articles linked from Template:Kent County Cricket Club squad, for example. Blue Square Thing (talk) 06:08, 6 August 2022 (UTC)

My objection is to the large numbers of mass created articles that don't demonstrate notability, are often on non-notable topics, and violate WP:NOTDATABASE. The reason for this is that they make the encyclopedia worse; they are detrimental to the reading experience, they lower the public perception of Wikipedia, and they make it harder to improve and maintain Wikipedia. In addition, the quantity of them makes it almost impossible to review them individually.

Identifying these articles at scale is difficult, which is why one proposed technique is to look at articles that are sourced solely to database sources. Rarely, this will include articles whose database source includes prose that provides significant context to the stats rather than merely repeating them, but I don't see that as an issue, because these articles still violate WP:NOTDATABASE. For example, this list articles created in February 2015 sourced solely to a partial list of database sources.

An alternative technique is to identify editors who have engaged in the mass creation of articles and produce a list of articles that they have created that are currently short, and have received minimal contributions from other editors. For example, see this list of articles created by Lugnuts.

Both of these can then be refined; for example, we can limit the lists to stub-class biographies on gymnasts, and I would note that neither of these techniques would result in any of the articles linked in that template being removed from article space.

Personally, I am leaning towards the alternative technique for practical reasons; the queries take considerably less time to run, and producing a complete list of database sources will be difficult. BilledMammal (talk) 09:29, 6 August 2022 (UTC)

I understand that reviewing all of a large quantity of articles manually is problematic, but when you start talking in more general terms it gets to the stage where there needs to be a manual review of each article. Now, if you can give me a list of dead cricketers where the article is less than about 500 words and sourced only to either CricInfo or CricketArchive (or other database style profiles) I'll tell you how long it'll take to manually review each source and determine a redirect target, or whether there isn't one or whether there is prose sourcing in either of those sources. How long it'd take will depend on whether I can convince other people to contribute to that process. That process might then be applicable to other sports - there are way more football, baseball, basketball, gridiron etc... editors for example - assuming that there are editors who are willing to undertake the work and will accept the need to compromise for the benefit of speed. Blue Square Thing (talk) 10:05, 7 August 2022 (UTC)

• What's the best method to go about opposing all of this and to more easily and actively remove all prods and such added to any of Lugnuts' stub articles (prods that I can basically 100% guarantee will have involved 0% WP:BEFORE done by the nominators)? What's the easiest way to prevent this active attempt to harm Wikipedia by all those above? Because you know what the solution is to Lugnuts being banned? Doing absolutely nothing in response. CCI has already shown that his copyvio claim was boasting nonsense and there's no evidence whatsoever that his articles have general problems, especially those that are stubs and are short enough to not even have copyvio or other such issues in the first place. Silverseren^C 03:38, 6 August 2022 (UTC)

  Mass created microstubs are a problem on their own for various reasons, not least WP:NOTDATABASE. BilledMammal (talk) 03:47, 6 August 2022 (UTC)

Also the fact that (especially with Lugnuts' banning) we now have a ton of stubs that have nobody watching them at all. New articles get patrolled, at least, so a new article created with obvious BLP issues or the like will get caught; but completely-unwatched articles are ticking time bombs in the sense that if we don't have some way to either get eyes on them or delete them, eventually one of them is going to end up saying something awful on par with the incident that originally led us to create WP:BLP in the first place. It is not enough to simplycreate articles, we have to be able to maintain them. --Aquillion (talk) 08:27, 6 August 2022 (UTC)

Is there a way as an interim solution to create an effective watchlist for the stubs that the community could track of? PaleAqua (talk) 18:41, 6 August 2022 (UTC)

Silver seren, the "best" method is to convince concerned editors that the stubs are not a problem, or to propose alternative solutions or compromises which address their concerns. As a relative newcomer here, I can understand the passion people have over the sports stubs, and also the annoyance at their creation; from the latter perspective the stubs are essentially a giant fait accompli. But so far, those in favor in deletion generally seem more willing to compromise. Ovinus (talk) 18:55, 6 August 2022 (UTC)

I agree entirely that there needs to be compromise. Fwiw I've seen fairly fundamentalist behaviour from both sides of the debate and whenever I've suggested that there needs to be middle way seem to have been largely ignored much of the time. The ways in which we compromise needn't necessarily be the same on each type of article (BLPs, for example, are clearly in a different sort of category), but if compromise can make all of this a lot more manageable then that's certainly to the good. Blue Square Thing (talk) 10:05, 7 August 2022 (UTC)

• I think BLP stubs based only on a database are more of a serious problem than other database stubs as BLPs are abide to a more stricter rules and BLP cited only on a database can be WP:BLPPRODed. Thingofme (talk) 13:15, 6 August 2022 (UTC)

  BLPPROD only applies to BLPs with no sources in any form. I don't think that applies to any of these stubs, and mass nominating them under BLPPROD will probably be seem as disruptive behavior to make a point. (Not saying that you're going to do so, of course.) Ovinus (talk) 17:37, 7 August 2022 (UTC)

  Such databases shouldn't count as reliable sources. I've worked on data integrity, and the idea of relying solely on a database for prove that someone exist is just bad. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 15:04, 8 August 2022 (UTC)

Part of the point is that they're aren't just databases in all cases. I've literally just finished working on an article based, in part, on this "database entry". If anyone wants to quibble with the authority of Wisden in cricketing matters feel free to take it up with someone important... Blue Square Thing (talk) 16:31, 8 August 2022 (UTC)

If Wisden had a database with an entry of a player name, and details of one first class match that that player played in in 1935, and none of those details could be verified elsewhere (onus on creator), then yes I very well "quibble" with with its authenticity whether it's from Wisden or St. J. Anderson (long my he bowl). Everyone sane should as well. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 10:22, 9 August 2022 (UTC)

• Blue Square Thing - Regarding the prose-profiles on Olympedia and Cricinfo: it is not clear where this data comes from and I have seen instances of it being flat-out wrong (eg wrong death-dates, life-history etc.). It appears to have been written by volunteer, wiki-like projects, often on the basis of what appears to have been input from the families of the people in question, without rigorous fact-checking. Why should this data be treated as reliably sourced? FOARP (talk) 09:42, 8 August 2022 (UTC)

I suppose that depends: in this case, for example, I'd suggest that, despite the minor inaccuracy, that the source is generally considered reliable. Published sources like that will often be used on CricInfo - and published sources do contain errors occasionally, as do newspaper articles and websites; you'd hope they'd still be seen as reliable in general. I've used Olympedia less, so I know less about it, but the prose seems useful as a starting point for where to look for more detail. If you're telling me that we're at IMdB levels of problem, then, OK, sure. But we're not at anywhere near that level with CricInfo or, where it exists, the prose elements of CricketArchive for that matter. Blue Square Thing (talk) 10:36, 8 August 2022 (UTC)

Proposal for X3

On the one hand, I agree with User:S Marshall that we should create a special CSD category X3 to deal with the Lugnuts stubs. I disagree with those editors who say that we shouldn't create a special category to deal with one editor, and S Marshall is right that there has been a precedent of X1 for the Neelix redirects. On the other hand, we should not create that CSD category now, but should wait until after the RFC is worked out and published, because creating the CSD category can and should be one of the actions in the RFC, which will probably be multiple RFCs or a multi-part RFC. In other words, any discussion of the Lugnuts stubs now is pre-discussion, knowing that it will be re-discussed. Robert McClenon (talk) 18:20, 6 August 2022 (UTC)

I believe this would be something to discuss/raise at WT:CSD. Primefac (talk) 11:08, 7 August 2022 (UTC)

• I object to any speedy deletion of Lugnuts' stubs. They each should be judged on their own merit. Many (probably most) are notable, some are not. We should not throw the baby out with the bathwater. BeanieFan11 (talk) 14:30, 7 August 2022 (UTC)

  Which is it? You object to speedy deleting any of them, or they should be judged on their own merit? Dennis Brown - 2¢ 14:48, 7 August 2022 (UTC)

  I'm saying none should be speedy deleted. Some may need to be brought to AFD, PROD, but CSD is not a sensible option in my opinion. BeanieFan11 (talk) 15:42, 7 August 2022 (UTC)

  I agree. Unless a claim for the regular CSDs or BLPProd can be made for a given article, it should only be deleted via Prod or AFD. Nominate them at a rate of 5 per day at PROD, and if contested then they can go to AFD at a rate of maximum 5 per day. 80.230.59.139 (talk) 16:02, 7 August 2022 (UTC)

  40000 stubs / (5 stubs / day) = 8000 days ≈ 21 years. Ovinus (talk) 17:30, 7 August 2022 (UTC)

  Based on the number of 93,000 stubs stated above by S Marshall, at a rate of 5 stubs per day, it would take 18,600 days or 50.92 years to PROD all, assuming the worst case scenario that all are worthy of deletion. Similarly if all are contested at PROD, then it would take another 50.92 years to nominate all for AFD. So such a proposed cleanup could theoretically take over a century in the worst case scenario. While the number of bad stubs will hopefully be lower than that, you're still talking about multiple decades of cleanup via this mechanism, which based on similar multi-year cleanup operations at CCI is just not feasible. Sideswipe9th (talk) 18:06, 7 August 2022 (UTC)

  I don't care how many stubs he has created. Articles still should not be speedy deleted just for being stubs. That's completely nonsensical. BeanieFan11 (talk) 19:22, 7 August 2022 (UTC)

  At the risk of spoiling everyone's argument, there is the draftification option that was used for Sander.v.Ginkel: draftify all the stubs, and let interested individuals save/improve/restore the useful drafts. Anything that is not sufficiently improved will eventually be deleted via G13. It gives everyone plenty of time to save the useful pages without speedy, but also means that the pages that no one will ever touch still get deleted. Primefac (talk) 19:25, 7 August 2022 (UTC)

  With the exception of one editor above whose suggestion doesn't seem to be gathering much traction, I don't think anyone has said the stubs should be deleted purely because they are stubs. Most above have pointed out the significant sourcing issues with the stubs created by Lugnuts, which results in them failing guidelines like WP:GNG and WP:SNG.

  Re-framing the problem, we have 93,000 stubs of which many fail our notability guidelines. A speed limit of 5 PRODs and 5 AFDs per day, would result in a cleanup taking up to a century in the worst case scenario of all of the stubs being bad. How do you propose we solve this problem in a timely manner, and that does not in the long term take significant editorial energy and time away from other important content areas? Sideswipe9th (talk) 19:33, 7 August 2022 (UTC)

  I don't think anything needs to be done about his stubs. If someone thinks he created an article failing GNG – nominate it for AFD or PROD. Also, I oppose any speed limit for nominating his articles at AFD. BeanieFan11 (talk) 19:39, 7 August 2022 (UTC)

• I'm afraid that taking them to AfD without a speed limit isn't very realistic. Editors complain, and they're right to do that, because there's a limited amount of time and attention available at AfD and we can't spend it all on Lugnuts for the next several years. The arbitration case that gave rise to this discussion is, to a substantial extent, about people overusing AfD. What we need to do is follow Wikipedia policy.
  In this case the relevant policy is set out in WP:BLP: Be very firm about the use of high-quality sources. These are mostly biographies of living people to which that rule applies with full force. We shouldn't use AfD, because it's a mass-cleanup job, and AfD is not for cleanup, so editors will rightly reject its use for that purpose. And as a further consideration it's not procedurally fair to use AfD. Lugnuts was creating these articles at the rate of one every two or three minutes, and a properly conducted WP:BEFORE takes at least ten or twelve, so it's disproportionate and unfair to expect a full WP:BEFORE on each one. Where sourced only to databases these articles never complied with policy, so the process should be check, tag, and where appropriate immediately delete without an individual discussion.—S Marshall T/C 22:24, 7 August 2022 (UTC)

Fwiw, it wouldn't surprise me in the least if this ends up being about a lot more than Lugnuts' stubs. From memory, the ArbCom suggestion was for an RfC regarding mass creation/deletion, yes? That seems like it's a lot broader than just however many of Lugnuts' articles are database-only stubs and, given what happened with the sport notability RfC, it seems likely that this RfC will end up much broader than these. I'm hopeful that some of the comments above about moderation will help, but given how a small number of users dominated the sports RfC I imagine we'll see the same sort of process here. Blue Square Thing (talk) 08:14, 8 August 2022 (UTC)

It will end up being about more than Lugnuts stubs, and that's why I suggested a series of RfCs.—S Marshall T/C 12:19, 8 August 2022 (UTC)

• But on the same coin of "anyone can take mass stub articles to AFD", anyone can recreate the articles if they could find good coverage other than Lugnuts' databases. I prefer we TNT all these "X player exists + infobox + database" articles in exchange for better manually created articles. — Ceso femmuin mbolgaig mbung, mellohi! (投稿) 22:47, 7 August 2022 (UTC)

To be honest, I think we might be able to adopt slightly different approaches depending on the subject matter. Right now I don't think we know too much about how many articles we're dealing with and what they cover; we suspect we have a lot of biographies of athletes, but those fall into different categories and not all of them are stubs and not all of those stubs simply rely on database sources. What works for cricketers (where, best guess, Lugnuts created perhaps 10,000 articles, almost all of which will have a sensible redirect target) might not work at all for cyclists or gymnasts. For me, with cricketers redirection works much, much better than deleting the articles - we'll just create a bunch of red links from the lists which creates all sorts of other problems.

Another factor is the time frame that people have in mind to deal with these. I don't know what the time frame was in the other cases people have referenced here. Obviously 50 years is far too long, but what sort of time frame do people have in mind? Blue Square Thing (talk) 08:09, 8 August 2022 (UTC)

WP:REDLINK is the relevant policy; we shouldn't be trying to avoid them if it is possible that the topic is notable. BilledMammal (talk) 10:38, 8 August 2022 (UTC)

I've found enough random incoming links that no one thought to check in articles to understand why red links to people's names are a really, really bad idea in stuff like this. Blue Square Thing (talk) 11:02, 8 August 2022 (UTC)

• Support - Violation of WP:MASSCREATE should be a stand-alone basis for PROD, and WP:BUNDLE and should be explicitly mentioned as so.

Blue Square Thing - "Sensible redirect target" is a bit of a red herring since in nearly every case the "sensible" redirect target will be a page where the subject isn't at all mentioned and in many cases will be only one of a number of clubs for which the player played, and that player is typically only one of many people with that name so it is in reality not a very sensible redirect target at all. Wikipedia:Articles for deletion/Harry Oppenheim is the classic example of this - the subject wasn't actually called "Harry" but Heinrich, we already have an article about a Heinrich Oppenheim, "Harry" Oppenheim was one of half-a-dozen people who were similarly (non) notable to have names that could be nick-named "Harry", the redirect target was a team they played for once and not the team they played for multiple times (though this also is a page where they are never going to be mentioned).

In reality if these redirects were being created from-fresh NPP and others would be quickly asking questions as to what the creator was doing. Instead they are created at AFD where no-one would be bothered to DELREV a redirect close into a delete close because ultimately it would be pointless and the trouts would come thick and fast for anyone who tried it. I'm OK with that, up to the point where the possibility of redirection becomes something that stops us clearing these notability-failing stubs up, at which point they are no longer simply WP:CHEAP. But for those cases where a redirect does make sense, deletion does not prevent re-creation as a redirect, so it is not clear why it should be a barrier to doing a thorough clean-up anyway. FOARP (talk) 10:21, 8 August 2022 (UTC)

I was specifically discussing cricketers, where there is an established precedent going back years that we redirect if there is not suitable sourcing to support a standalone article. We can usually find a sensible redirect target - the major side, a list by year etc... Personally I'd rather keep the sourcing we have as that's often the basis for establishing a well sourced article later on and avoids having to find the darned sources again. Blue Square Thing (talk) 10:29, 8 August 2022 (UTC)

I think it's a little rich that those who effectively made these stubs non-notable by changing the SNGs at WP:NFOOTY, WP:NCRIC etc. are now whining because it's going to take a long time to achieve the result that they wanted when they started the whole process. I am opposed to a speedy deletion criteria here. Black Kite (talk) 13:41, 8 August 2022 (UTC)

I agree with Primefac that this belongs at WT:CSD and Im not sure why it hasnt been clerked as off-topic here. You arent going to set CSD policy on WT:ACN. nableezy - 14:31, 8 August 2022 (UTC)

This page clearly isn't the venue to create a new CSD criterion. But the discussion is relevant to ArbCom: one open question for the RfC mandated by Remedy 11 is whether consideration of a new CSD or other measures for responding to Lugnuts's creations is in-scope for the mandated RfC. I see the RfC's role as supporting the community's resolution processes. So if the community decides to handle that discussion elsewhere, of course I would support that. But if the community wants it to be part of the ArbCom-mandated RfC, I will advocate for that too. Best, KevinL (aka L235 · t · c) 16:41, 8 August 2022 (UTC)

Special Circumstances Blocks

Original announcement

It's subtle, but ArbCom has significantly exceeded its authority here by turning what was once an optional process into a required one. It's important to note that there are two different but easily-confused policy provisions at play:

• WP:BLOCKEVIDENCE says that admins cannot unilaterally block based on information that will not be made available to all administrators.
• WP:ADMIN#Special situations says

  Rarely, in blocking an editor, an administrator will have to note that their block "should be lifted only by the Arbitration Committee" or that "any appeal from this block is to ArbCom only". Such a provision must only be made if the nature of the block demands that its circumstances not be further discussed on-wiki (and instead be considered further only in a confidential environment).

The latter aplies to a considerably broader category than the former. A sockmaster's tells can be shared with other admins, but should still be kept private. A COI editor's identity can, per WP:OUTING, be discussed privately by admins, but must be kept private. BLOCKEVIDENCE allows blocking unilaterally in these circumstances. ADMIN allows referring such cases to ArbCom, but does not outright require it. And, apparently, basically no one has exercised that option on a case that wasn't "highly sensitive" until I did a few weeks ago on a very complex behavioral block, I gather sparking this discussion.

So, that's fine, ArbCom can restrict how admins make "appeal to ArbCom" blocks, because there's no circumstance where the community requires admins to use that option.

What ArbCom cannot do is restrict how admins make non-ArbCom-related blocks that are allowed by community-made policy. And when ArbCom says Administrators should contact the appropriate group rather than issue a block covered above, they are prohibiting admins from doing what community consensus allows, and that substantially exceeds the Commmittee's constitutional authority as an arbitrating rather than governing entity. BLOCKEVIDENCE draws the line: The community's consensus is that admins may issue blocks based on private evidence, so long as it's private evidence they can discuss with other admins. ArbCom cannot move that line. ArbCom cannot overrule that provision of policy.

(That's setting aside the logistical concerns here of suddenly requiring rather than merely permitting the use of this process. Does the CU team really want to handle every single case of John DistinctiveName, who writes spammy things about Acme Corp., which has a director of marketing named John DistinctiveName? Does the CU team really want to decide every case of behavioral-evidence-only sockpuppetry that has some off-wiki component? CUs are, as several have acknowledged to me directly, not necessarily selected based on skill at reviewing behavioral evidence of sockpuppetry.)

-- Tamzin^{[cetacean needed]} (she|they|xe) 23:30, 8 August 2022 (UTC)

+1 I think at the very least, they should clarify their guidance because followed literally, it will lead to unnecessary WP:BURO in cases such as those you've mentioned. -- King of ♥ ♦ ♣ ♠ 23:48, 8 August 2022 (UTC)

Thanks @Tamzin. We have spent a little while discussing the announcement, largely looking back at what happened a decade ago and focussing on "Blocks that can only be appealed by Arbcom". I don't believe, as a committee, we were aiming to stop any form of off-wiki block. That's said, I do think that the community should possibly review those sorts of blocks - Wikipedia is not an island, but the level of information that is required to make a Wikipedia account means that it is not difficult to "Joe-job" based on off-wiki evidence. Worm^TT(talk) 09:38, 9 August 2022 (UTC)

• That does seem to be an overreach of authority. One of the traditional points of contention is whether Arb has the authority to create or change policy (for the record, no), versus establish procedure. As someone who worked SPI a great deal, I would echo the sentiment that CUs aren't chosen for the behavioral review skills, and in fact, tend to avoid it at SPI, instead providing technical information and almost always leaving behavioral analysis to other admin. CUs tend to be the slowest to respond as well, when compared to OS or Admin for their respective duties. CUs have the highest level of workload vs how many people have the bit, and it has always been this way, which may be part of the reason the tradition of seldom connecting behavioral evidence exists. So it does seem an overreach compounded by a requirement to use the slowest and least willing group of people here. Dennis Brown - 2¢ 10:38, 9 August 2022 (UTC)
• Hi @Tamzin. Thank you for writing. Let me try to clarify: blocks that are not based on private off-wiki evidence are not covered by this announcement. If I'm not mistaken, the block that you last marked as "appealable only to ArbCom" relied only on a behavioral analysis of evidence available on-wiki. You could have made it without marking it appealable only to ArbCom before, and you still can now. (Now, I do think it was wise for you to send it to ArbCom, and I think it would continue to be wise if you sent it to the checkuser queue, either for action or for information. But this statement doesn't prevent you from having the authority from enacting the block.) The only blocks covered above, as you write, are (1) off-wiki evidence of socking/UPE/spam, (2) oversight blocks, and (3) "Highly sensitive and private information".
  As to your other points, I think you and I interpret WP:BLOCKEVIDENCE differently. As I understand it, available to all administrators means information that all administrators have technical access to: deleted revisions, UTRS, etc. It does not cover "email me and I'll send you the off-wiki evidence": I don't have meaningful access to that information unless you send it to me, and if you're taking a wikibreak or you retired and I want that evidence, I'm out of luck and so is the appellant. (Am I supposed to unblock them? Or just assume the evidence was legit?) That's why this announcement tells you to send it to the checkuser queue, where the evidence will be permanently archived and available for review on appeal.
  This statement is ArbCom's attempt to issue guidance that is compliant with both BLOCKEVIDENCE and the ADMIN subsection – no easy feat, as you know, given the state of those policies . It was reviewed by the functionary team and was thoroughly revised, probably on at least half a dozen occasions, before it was voted on. It's not exactly the statement that I would write on a blank slate, but ArbCom – as you point out – is bound by policy, so I don't get to write on a blank slate. I'm happy to answer additional questions as they come up, and hope this clarification is helpful. If you think there are workload challenges that come up, I'd love to discuss those too, and talk about how to resolve them, whether by seeking community amendment of the underlying policy, or by appointing more CUs with different backgrounds. Best, KevinL (aka L235 · t · c) 15:13, 9 August 2022 (UTC)

I'm not sure I understand what is going on here. My interpretation is that any WP:DUCK blocks for sock puppetry or paid editing are still fine (as they are based on onwiki evidence), but for anything that requires sleuthing (say, looking at a Facebook post or some other social media stuff that requires me to log in), I should consider the question of blocking to be Someone Else's Problem? (I don't quite understand why I shouldn't just block and let a functionary take over the block if they want to make it irreversible). Or to put it differently: if I am not interested in making irreversible blocks, why do I need to change anything about my blocking habits? —Kusma (talk) 15:49, 9 August 2022 (UTC)

If it needs sleuthing, it should go to one of the CU email queues. @Kusma, if that's going to mess with your workflow, that's not what we meant to do and I will work to fix that. Is there a system that you would prefer more? Best, KevinL (aka L235 · t · c) 15:54, 9 August 2022 (UTC)

Given the way I currently work (mostly content, a little bit of CSD work and some related blocks) this won't mess with my workflow very much, other than to turn me off from doing even the most trivial sleuthing work (like, say, googling for the username + company they promote; don't want to bother the checkusers with that kind of stuff). The vast majority of my blocks are onwiki evidence only anyway (and are rather trivial blocks of fairly new accounts). —Kusma (talk) 16:05, 9 August 2022 (UTC)

Kusma, the "irreversible block" element was a big push behind this statement for me. So if you don't want to make those kinds of blocks and are otherwise in compliance with our blocking and admin policies you're fine from my point of view. Barkeep49 (talk) 16:01, 9 August 2022 (UTC)

That's great, but the statement Administrators should contact the appropriate group rather than issue a block covered above seems to apply to all blocks, not just to irreversible ones. If this is only meant to cover irreversible blocks, could you clarify that? —Kusma (talk) 16:08, 9 August 2022 (UTC)

Apologies if I'm misreading or asking a stupid question. Regarding off-wiki evidence (facebook sleuthing or even company website) - in non-unusual circumstances, Administrators should contact the appropriate group rather than issue a block covered above. which seems very odd. I can see why there is a desire for the evidence to be logged somewhere accessible should the admin become inaccessible, but I don't see why I couldn't issue the block and then send an email. Nosebagbear (talk) 16:25, 9 August 2022 (UTC)

Regarding @Nosebagbear's note specifically, I would be supportive of such a system. But I will note that's not what policy establishes now and would be a considerable expansion of the role of individual admins: to make blocks appealable only to CUs, on a routine basis. I therefore don't think ArbCom could establish that system under current policy but I would support a community RfC on that. Best, KevinL (aka L235 · t · c) 16:36, 9 August 2022 (UTC)

Why would Nosebagbear's block be appealable only to CUs? —Kusma (talk) 16:59, 9 August 2022 (UTC)

If you emailed your evidence into the CU queue, wouldn't it make sense that only CUs would be able to review or lift the block? Other admins wouldn't have access to the relevant evidence. Best, KevinL (aka L235 · t · c) 17:32, 9 August 2022 (UTC)

To put it bluntly: as the person who has handled over 80% of the ~275 paid queue tickets closed in the past year, I cannot begin to quantify how uninterested I am in handling emails from admins to the effect of "LinkedIn says this person's name matches the marketing manager of the company they're writing crap articles about, but that connection cannot be made using on-wiki evidence, block please," and any for-the-record emails made after such a block will receive even less attention. GeneralNotability (talk) 17:41, 9 August 2022 (UTC)

If CUs think the block should not be reviewed by non-CUs, they can reblock stating so. If they don't do that, any admin should be allowed to unblock under the usual rules. —Kusma (talk) 17:43, 9 August 2022 (UTC)

@L235 - the storage point for evidence could always be tweaked, so as to not clog the "actual" CU stuff. Additionally, that route will only be used when the blocking admin isn't accessible in the current fashion in a few days. And they'd always have the email as well, if they couldn't remember directly. CU time is one of the most finite resources we have - upping the CU-appeals (or CU-blocks if we use the current wording) in this fashion would be an unnecessary loss. Nosebagbear (talk) 19:05, 9 August 2022 (UTC)

• The concern about admins going inactive and evidence being lost to time is absolutely a valid one, and one I've had as well. And there's a number of ways ArbCom could fix it. ArbCom or an individual arb could propose amending BLOCKEVIDENCE to say what it's being construed here to say, which I think goes against its plain-text meaning. (will not be made available to all administrators seems to encompass material that is not, by default, available to all administrators. Otherwise it would say is not available to all administrators.) ArbCom could create a "dropbox" email account to which admins can send documentation of blocks based on off-wiki information without expectation of a response, and then propose that usage of that account be made mandatory for such blocks. ArbCom could push for an admin wiki like the Italian Wikipedia has, where this information could be logged. None of those things exceed ArbCom's authority. The approach taken here does.
  Don't get me wrong, I understand how this kind of thing happens. I'd consider myself friendly with more than half of the Committee, and I don't think any of y'all are attempting some power-grab. Groupthink is real. If JFK could land men at the Bay of Pigs with a "retreat" plan that required going through enemy lines and impassable terrain, so too can 13 volunteers, debating something for a few weeks, settle on an outcome that at one point goes beyond what they are allowed to mandate. But ArbCom should recognize that error and correct it. All that is necessary to bring this statement in line, I think, is prepending In cases where the reason for a block could not be shared with other administrators to the sentence at issue. If BLOCKEVIDENCE is at some point changed to restrict a broader category of unblocks, a new statement can be issued to match that. Or the policy-violating sentence could be edited to say Administrators should consult WP:BLOCKEVIDENCE to determine whether it is appropriate to block a user or instead to contact the appropriate group as described above. -- Tamzin^{[cetacean needed]} (she|they|xe) 20:15, 9 August 2022 (UTC)

  @Tamzin: I understand your position that ArbCom's statement here violates policy. I've described why I don't think it does – I believe your interpretation of BLOCKEVIDENCE is colorable but I don't think it's the one we've adopted. I'm not here to litigate that disagreement, but I very much am open to ideas on how to rework the system more broadly. I don't know how old BLOCKEVIDENCE is, but I'm sure it was written well before e.g. UPE blocks based on off-wiki evidence became extremely common. Would I support a system that, for example, retains the "I'll email evidence to reviewing admin" but establishes an ArbCom evidence "drop box" email as a backstop? Very probably, yes. That has to be a decision the community's involved in, though – it's surprising to me that you think implementing such a system is within ArbCom's sole power but not a statement written specifically to comply with existing policy. Best, KevinL (aka L235 · t · c) 22:24, 9 August 2022 (UTC)

  Like I said, ArbCom could set up such a system, and then propose to the community to make it mandatory. Failing community support, it would be voluntary. -- Tamzin^{[cetacean needed]} (she|they|xe) 22:39, 9 August 2022 (UTC)

• I was curious how often this comes up, so I did a little Quarry hacking (https://quarry.wmcloud.org/query/66527). My SQL is not great, so I'm sure there's a better query that could be written, but this should at least give a reasonable upper bound of a few times a year. -- RoySmith (talk) 15:17, 10 August 2022 (UTC)

  @RoySmith the affected bubble is bigger than it might be on first glance, as it's applicable to any regular block issued by a regular admin, who got any of the evidence off-wiki (most commonly from the company's own website), and I don't believe that quarry could detect those (I'm not sure any could easily do so). Nosebagbear (talk) 08:57, 13 August 2022 (UTC)

  I don't think that's true. Per WP:BLOCKEVIDENCE, administrators should never be making unilateral blocks based fundamentally on off-wiki evidence, i.e. where there would not be enough evidence to make the block if not for the off-wiki evidence. On the flip side, administrators can and should continue to make unilateral blocks where there is enough on-wiki evidence to justify it, even if there may be off-wiki components that the administrator is privy to. To take Tamzin's example above of John DistinctiveName, who writes spammy things about Acme Corp., which has a director of marketing named John DistinctiveName—in such a case, administrators may unilaterally block the account because the act of writing "spammy things about Acme Corp." is inherently disruptive, and a block would be justifiable based solely on the on-wiki evidence. There would be no need to contact the CU team or ArbCom unless the editor is being sneaky about it (e.g. they're trying their best to write a neutral article and perhaps actually succeeding, but we found a posting on an online job board that confirms it's an undisclosed paid job and thus violates our policy). Mz7 (talk) 02:22, 17 August 2022 (UTC)

• I came here to see if others had noticed this part of the update. Not because I have a strong opinion on whether it would be a good change or not, but rather because it is an unexpected, significant change that seems to affect all future non-CU blocks based on off-wiki evidence. As Tamzin has described in detail, the resulting situation is unclear; the announcement needs clarification. ~ ToBeFree (talk) 20:51, 12 August 2022 (UTC)
• In my view, ArbCom has not exceeded its authority. The relevant policy is WP:ARBPOL#Policy and precedent, which states that The Committee's decisions may interpret existing policy and guidelines, recognise and call attention to standards of user conduct, or create procedures through which policy and guidelines may be enforced. In this case, ArbCom is interpreting existing policy to mean that administrators are prohibited from making any block that is based on information that not all administrators "have technical access to" (as L235 put it above; e.g. deleted revisions, UTRS appeals). WP:BLOCKEVIDENCE delegates the authority to make these sorts of blocks only to the Arbitration Committee or a checkuser or oversighter. This is the correct interpretation, and the procedure that ArbCom establishes here is in line with the existing policy.
  Tamzin claims that BLOCKEVIDENCE does allow administrators to make blocks based on private evidence (e.g. emails, Facebook posts, non-WMF websites) as long as they are willing to provide that evidence on request to other administrators. I'm afraid that is not correct. As L235 points out, that is not really making the evidence available to all administrators—especially in the case that the blocking administrator retires. The community has never agreed that admins may issue blocks based on private evidence, so long as it's private evidence they can discuss with other admins. ArbCom is not overstepping its authority by forbidding administrators from making these kinds of blocks because the community never allowed administrators to make these blocks in the first place. Mz7 (talk) 01:57, 17 August 2022 (UTC)

  @Mz7 I'm not sure about that either- see the first point at Wikipedia:Administrators#Special_situations: "Rarely, in blocking an editor, an administrator will have to note that their block "should be lifted only by the Arbitration Committee" or that "any appeal from this block is to ArbCom only". Such a provision must only be made if the nature of the block demands that its circumstances not be further discussed on-wiki (and instead be considered further only in a confidential environment)"? It basically condones these sort of blocks. It seems like that part of the problem is that the section at BLOCKEVIDENCE and the one at ADMIN contradict each other.

  I made an "appeal to Arbcom block" one time (Special:Contributions/LaceyUF) which was eventually upgraded to a regular Arbcom Block. In hindsight I probably didn't need to do that- I could have just blocked for disruptive editing and sent the related evidence to the committee. At the time I was under the impression that there was nothing wrong with how I made the block, and I wasn't told otherwise by any Arbs. This announcement confuses me a bit with regards to that block; it fell under the purview of the committee as far as I understood since it dealt with offsite disruption that was spilling over on-wiki, but that sort of offsite disruption doesn't appear to be explicitly covered by this announcement. It could be that I'm seriously misunderstanding, or the committee is implying but not directly making a statement about these sorts of offsite-behavior blocks with the "Highly sensitive and private information" section of the announcement; even then, that seems like a large stretch. But since they don't explicitly say anything about these sorts of offsite-behavior blocks, I'm guessing this announcement is really only outlining best practices for impersonation/socking/spam/UPE "private evidence blocks".

  The committee cannot override the sections at ADMIN or BLOCKEVIDENCE, right? So the question of whether offsite-behavior/private evidence blocks not related to impersonation issues, socking, or UPE are allowed is still a matter that needs to be decided by the community, since policy is unclear. This announcement isn't really taking a stance on that issue. That is my understanding of the issue- an arb or someone else in the know should feel free to correct me.

  If my understanding is correct, then I would like to start an RfC or something on the matter to figure out what the community currently thinks about these blocks. Related to that, I was wondering if any past/present Arbs could tell me if they ever found these sorts of "appeal to the committee" blocks in the vein I mention above useful. Moneytrees🏝️(Talk) 03:18, 17 August 2022 (UTC)

  Ah yes, I suppose I did forget to analyze how that paragraph at ADMIN fits into the picture. I don't necessarily think BLOCKEVIDENCE and that text at ADMIN are irreconcilable, but I will admit that their relationship is confusing. The paragraph at ADMIN was added boldly in August 2012 by AGK (then an ArbCom member) following this reminder from ArbCom about these kinds of blocks. (Based on that, it seems to me that the genesis of that paragraph at ADMIN was in fact the ArbCom motion that the current ArbCom is now amending with the current announcement... makes for a bit of a headache, to be sure.) The specific examples it highlights of the kinds of blocks that could be made in this manner are situations where discussion would reveal or emphasize information whose disclosure could jeopardize an editor's physical or mental well-being, where on-wiki discussion would identify an anonymous editor, or where the underlying block reason would be defamatory if the block was unjustified. I'm not sure whether this irreconcilably conflicts with BLOCKEVIDENCE's prohibition on private-evidence-only blocks—it could just allowing administrators to restrict appeals to ArbCom in cases where there are extenuating privacy-related factors that should be taken into consideration when reviewing the block, but also where the fundamental reason for the block would still be related to on-wiki behavior (BLOCKEVIDENCE only prohibits admins from making blocks solely based on private evidence). Definitely tricky though. Maybe you're right that an RfC may be necessary after all. Mz7 (talk) 05:16, 17 August 2022 (UTC)

Deletion RfC moderator appointments

Original announcement

• If the moderators have any questions about what worked or didn't work in the GMO RfC please let me know. --Tryptofish (talk) 18:44, 16 August 2022 (UTC)