Wikipedia:Village pump (policy)

"WP:VPP" redirects here. For proposals, see Wikipedia:Village pump (proposals).

 Policy Technical Proposals Idea lab WMF Miscellaneous 

• Table of contents
• First discussion
• End of page
• New post

The policy section of the village pump is used to discuss already proposed policies and guidelines and to discuss changes to existing policies and guidelines.

• If you want to propose something new that is not a policy or guideline, use Village pump (proposals).
• If you have a question about how to apply an existing policy or guideline, try one of the many Wikipedia:Noticeboards.
• If you want to ask what the policy is on something, try the Help desk or the Teahouse.
• This is not the place to resolve disputes over how a policy should be implemented. Please see Wikipedia:Dispute resolution for how to proceed in such cases.
• If you want to propose a new or amended speedy deletion criterion, use Wikipedia talk:Criteria for speedy deletion.

Please see this FAQ page for a list of frequently rejected or ignored proposals. Discussions are automatically archived after remaining inactive for two weeks.

« Archives, 172, 173, 174, 175, 176, 177, 178, 179, 180, 181, 182, 183, 184, 185, 186, 187, 188, 189, 190, 191, 192

Dispute Resolution and Limited English

I have a question about a situation that sometimes arises either at the Dispute Resolution Noticeboard or in other forums where dispute resolution is being attempted. The question is what can be done when an editor clearly wants to resolve a dispute in a collaborative manner and is trying to discuss the issues, but it is clear to a third party that the editor does not have enough of a command of English either to understand the details of the issues or to explain what they see as the issues. Robert McClenon (talk) 17:34, 9 August 2022 (UTC)

The immediate case in point is an article that was moved to draft space, and then its author requested discussion at DRN, which I declined because discussion could take place via the Articles for Creation draft process. I am not identifying the dispute because I am not asking for advice about the specific case. (I expect I will get advice about the specific case anyway.) The question is what should be done when the author or proponent clearly wants to discuss, but at the same time is clearly having difficulty understanding the English. This issue more generally extends to other cases where an editor is trying to discuss an article, but does not know enough English to understand explanations from volunteers. One guideline is Do Not Bite the Newbies, but how do we tell an editor that they do not know enough English, without biting them?

Standard advice includes trying to tell the editor to edit the Wikipedia in their first language, but some editors really really want to edit the English Wikipedia, even though they don't know as much English as they think that they know. Robert McClenon (talk) 17:34, 9 August 2022 (UTC)

I have an update that doesn't affect the underlying question. In the specific case in point, the author is an American. They seemed, over the Internet, to have a problem communicating in English. That problem is not due to lack of command of English. However, I have also previously encountered editors at DRN who have wanted to engage in moderated dispute resolution, and have not been able to communicate effectively in English. So I still have a question how to deal with editors who are having difficulty with English, in working on an electronic product that is in the English language. Robert McClenon (talk) 22:39, 9 August 2022 (UTC)

I've seen some disputes that sound similar. An editor with certain forms of severe dyslexia, for example, might find it difficult to communicate in writing. In my experience, discussions with people who cannot communicate with at least moderate fluency in writing are frustrating to all concerned, and eventually the editor leaves (voluntarily or otherwise). WhatamIdoing (talk) 18:14, 13 August 2022 (UTC)

It can be difficult. We all want Wikipedia to be a welcoming, inclusive place and I think we try hard to do that. But ultimately this whole project is about writing a comprehensive and comprehendible, detailed, balanced, accurate encyclopaedia in the English language. In addition, the project is 21 years old and in that time we've covered most of the easy stuff and are mostly left with the trickier, more intricate details or difficult subjects to cover. Now more than ever, competence is required. WaggersTALK 13:23, 22 August 2022 (UTC)

Came here to say the same thing. Putting that into words that aren't bitey can be challenging, though. I don't think I've had a situation quite like what the OP is talking about, but if I got into that position, I suppose I would try to lay it out for them, as gently but directly as possible, on their talk page. If it's not a non-native speaker, as seems to be the case here, it becomes even more delicate of a process. Matt Deres (talk) 16:55, 23 August 2022 (UTC)

I understand that a lack of decent command of English or many severe forms of dyslexia may make it nearly impossible to work collaboratively on English Wikipedia. However, at the level Robert McClenon is asking about it seems to be getting at issues of a bit more complexity. There's obviously quite a bit of research into the interaction, and conflict, between individuals at different language competencies, and it might help to have someone more familiar with this work weigh in. I do want to note (as noted in the abstract of Cohen 2001) that plenty of editors may be coming in with perfectly workable communication skills in a technical field (such that they would be a valuable WP editor on such articles, even collaboratively), but as soon as the environment switches to something like dispute resolution which inevitably gets wordy and lawyery (even if we pretend we don't), the effectiveness of communication can quickly break down. So are such editors simply to be discarded, as several commenters above are suggesting? Is the heated lawyery dispute resolution really the only possible way in which such situations can be resolved, and those not capable of participating in the most nonproductive of transfer of bytes on WP servers are disqualified from the project?

Some effort would have to be made by a small group of editors, two or three maybe with the help of an admin, to work out some general procedure for notifying an admin if such a situation is suspected, positively identifying such a situation, then eventually a summary judgement with pre-arranged simple means of communicating, in a much more visual and concise manner than the walls of text that are our policy pages, the limitations set, should there be some. As I'd imagine it off the top of my head, a temporary partial topic ban would be the most likely outcome in practice if the judgement is positive, but I'm sure people figuring it out and testing it would create something more robust.

I laughed out loud at Waggers's final line. Maybe their watchlist needs more variety. Try hitting the "random article" button from time to time. SamuelRiv (talk) 05:12, 27 August 2022 (UTC)

Yup. If anyone ever wants some easy wiki work to do, hit me up. Absolutely no shortage of work, including time-consuming and mind-numbingly repetitive but extremely easy, 'round the Lepidopteran corner of the wiki... AddWittyNameHere 06:13, 27 August 2022 (UTC)

Doesn't matter if they really really want to edit English Wikipedia: competence in written English communication is an absolute requirement, and frankly in my view it's insulting when people try to edit articles in a language in which they do not have sufficient proficiency. I would feel I was being culturally offensive if I tried to edit another language project without having command of that language. If I don't understand the words on the page, it would be extremely egotistical of me to think I could possibly improve them. If I tried, I'd just be wasting the time of that project's volunteers, and risking misinforming the project's readers. It's real harm. And so if someone told me to stop editing in that other language wiki, I wouldn't think that was rude or uncalled for. At this point I think we have all the world's major languages covered, there's no need for anyone to edit in a language they don't speak fluently. Levivich 21:18, 27 August 2022 (UTC)

Thought experiment: in a CIR article, would you be more offended by an editor leaving behind any of these common ESL mistakes (that can be cleaned by any passing IP editor), or by making basic mistakes in the subject material itself?

In real cases, lots of our articles benefit (or potentially benefit) from native speakers of non-Romance languages, or who have access to non-Western offline libraries, and many editors in this pool could potentially fail your "absolute requirement". Since I've heard general agreement in other threads that more effort should be put into developing conflict management strategies on wiki, I can't see why language discrepancy can't be a part of what's developed, at the very least as part of a wider scope of alternative needs. Obviously there are plenty of cases where a particular user's edits to an article are generally counterproductive, and as those go beyond just language issues we're still really talking about conflict resolution in practice and the improvements needed, and not just some flat new policy (that would still in practice come down to a Talk page impasse and an admin intervention). SamuelRiv (talk) 17:06, 28 August 2022 (UTC)

I've made contributions to wikis whose language I don't speak. There are things that anyone competent in wikitext can do, at least in a left-to-right script. I'll bet that @Levivich could empty w:ht:Kategori:Paj ak lyen fichye kase without knowing a word of Haitian Creole. (It's an error category for articles that used to have pictures in them, but the image has since been deleted from Commons.)

Even if you don't know the language, you can sometimes make useful content edits, too. If the subject is Joe Film, and there's no image, you can add an image and then copy and paste the article name for the caption. You might not be able to write "Joe Film at the debut of his 2019 blockbuster film" but you can manage to copy and paste "Joe Film". There are editors who seem to do this as their main activity, and they end up improving dozens of Wikipedias. WhatamIdoing (talk) 22:43, 3 September 2022 (UTC)

Has Wikipedia policy ever shifted toward inclusion?

I am curious if there are any points when PAGs that previously excluded content have changed to include it. For example, has a notability guideline ever been successfully amended to extend notability more broadly than it previously did? (My hypothesis is "no, never", but I'd be very interested in falsifying that.) Just curious if any examples come to mind. Many thanks in advance! -- Visviva (talk) 17:10, 10 August 2022 (UTC)

The general notability guideline has been stable over the years (and a quick note that guidelines and policy have different specific meanings on Wikipedia). I'd have to do some research to see how it's changed but since it's been stable I don't think it's shifted in any meaningful way either towards or against inclusion. But if I really understand your question, the recent changes to NSPORT were a shift away from inclusion, but likewise SNGs have been added (though admittedly none since 2011). Existing SNG criteria are also sometimes expanded, such as when more women athletes were presumed notable under NSPORT as the number of women's leagues increased (though obviously this is a bit complicated now with the overall change to it). Best, Barkeep49 (talk) 17:23, 10 August 2022 (UTC)

Thanks! I will need to add some nuance to my thoughts on what SNG creation signifies. -- Visviva (talk) 17:30, 10 August 2022 (UTC)

You really have to examine when a specific SNG was created. Some predate the GNG, but others were written as a reaction to GNG (laying out alternatives to GNG)… while a few were written to reinforce GNG. Some are more inclusionist in their criteria, others are more exclusionist. The most stable ones tend to aim for a practical middle road: too inclusionist for some editors, while at the same time being too exclusionist for others. Blueboar (talk) 20:10, 10 August 2022 (UTC)

• Another factor is that there have been several inclusion regimes that have never been supported by actual SNGs, just sort of unstated agreements. I think this was the case with the view that every winner of a US state level Miss America or Miss USA pageant was notable, it seems to have been presumed and implemented without actual support. The same seems to have been the case with all competitors in the Olympic Arts Competition, I have never seen anywhere where people were extending the same rules that apply to sports competitors to arts competitors, but in creation of articles there was the same assumption of default notability. The discussion of the Olympic notability guidelines assumes we are talking about sports competitors. The fact that the arts competition was ended in 1948, 74 years ago, in part because it was not reaching the level of acclaim that the organizers wanted, makes it likely that lots of people do not even know it ever existed. I didn't until I happned on some articles on competitors. There are probably lots of other examples. Some of this may come about because in addition to SNGs, which say various things and position themselves with regard to GNG in various ways, we also have other statements on inclusion that look sort of like SNGs, but are not quite SNGs.John Pack Lambert (talk) 12:53, 12 August 2022 (UTC)
• What is the best way to go through the full list of currently accepted SNGs?John Pack Lambert (talk) 13:00, 12 August 2022 (UTC)

  Category:Wikipedia notability guidelines Ductwork (talk) 14:24, 12 August 2022 (UTC)

• There was a time when many SNGs were written rather willy-nilly, and many of them seemed to endorse creating new articles for which no reliable source content could actually be found. In recent times, this has been started to be reeled back in, to ensure that we can actually have an article whose content is based on reliable sources, and not just mostly empty sub-stubs on subjects that meet some arbitrary criteria. The basic inclusion criteria has always been "Does enough reliable source content exist out there (waves vaguely at the entire universe of knowledge) to be used to help us write a reasonably well-written and comprehensive here at Wikipedia". SNGs were an attempt to circumvent that process, a way to start a new article about some subject for which no attempt needs to be made to determine if enough reliable source material even exists to support a reasonable article. Recent history has thankfully started to reel that in. --Jayron32 14:30, 12 August 2022 (UTC)
• For what it's worth, of course we are going to be more inclusive as we move forward in time. We don't have a deadline, like say aiming to be one by 2030 (except for new events and developments). We've done most of the really notable stuff, and the pretty notable stuff, and the somewhat notable stuff, and the barely notable stuff. Time to move on to the hardly notable stuff and the marginally notable stuff. That means lowering our standers, to include more material and make the internet suck that much less. Herostratus (talk) 02:57, 26 August 2022 (UTC)

  I don't think we've written articles about all the "really notable" subjects yet. Italian Renaissance sculpture was created only a few months ago. French Renaissance sculpture doesn't exist yet, nor does Spanish Renaissance sculpture, English Renaissance sculpture, Dutch Renaissance sculpture, or German Renaissance sculpture. These are all "really notable", if you measure that status in terms of something which whole books have been published about over the course of many decades. I think this is a pretty common situation. Once you get out of the "really popular" subjects (e.g., Star Wars, professional football), I think that many such holes exist. WhatamIdoing (talk) 22:31, 26 August 2022 (UTC)

Well, fair point. But still. Herostratus (talk) 22:08, 27 August 2022 (UTC)

I still think Herostratus makes a good point because the notable subjects are indeed strictly limited even if they are not all done yet, and so the shift must inevitably lead toward more inclusive since the stated goals of Wikipedia also include "...free access to the sum of all human knowledge." "...that contains information on all branches of knowledge." it seems inclusion is a foregone conclusion, and arguments that I always hear mainly from deletionists with needless worry about "endless this" or "endless that" make no sense at all to me because even when we run out of the notable stuff we will follow through to the logical conclusion that even less notable stuff is limited, and finite. There is no "endless this" or "endless that". It's just an imaginary "problem" to "solve". Some people wrongly identify me as an inclusionist by the way I talk, but I'm not. I'm just against hard core deletionism. I find it to be most harmful. Hardcore inclusionism is also harmful, but I only see it manifest on articles, where it is easy to dispense with, whereas I see deletionism manifest in policy, where it takes diligence, and an act of congress to sniff it out, and set it right so I think deletionism is far more harmful. Huggums537 (talk) 22:59, 27 August 2022 (UTC)

Our standards have raised considerably since I've been here. Especially through 2015ish, the running theme of FA after a few years is that they either improve with the times or they get delisted (which is why I find FA OWNers' typical interpretations of FAOWN hilarious, but that's for other threads). Bot-created stubs have been largely cleaned up, and although if I weren't dormant at the time I would have vehemently opposed AfC, the research and implementation have held up. There's a lot to worry about with the future of Wikipedia (for my part I think the editor retention issue and ingroup mentalities are hand-in-hand at the top of the list, though probably only the former matters, since more editors would make the ingroup people matter less), but a decline in standards is not one of them. A decline in quality by other metrics and definitions, however -- politicization/moralization, single-track options to editing and conflict, editorX over UX (not exactly new behaviors, but not acceptable with how ubiquitous a resource WP is) -- that I will definitely hear out. SamuelRiv (talk) 22:38, 27 August 2022 (UTC)

For what it's worth, of course we are going to be more inclusive as we move forward in time. I have to say that I really don't think this is true unless there's a major sea change. We are actually becoming far less inclusive, with a number of non-guideline but still widely applied (and even some that were written into guidelines) notability standards being quashed by RfCs. For example, once we considered pretty much all railway stations, all secondary schools, all degree-awarding institutions, all Olympians, all top-flight sportspeople and all generals, admirals and air marshals to be notable and pretty much every AfD on these topics was closed that way. That is no longer the case and the fact it is no longer the case is endlessly crowed by the deletionist lobby. So no, sadly this is not true at all. -- Necrothesp (talk) 15:07, 31 August 2022 (UTC)

It has been necessary to restrict what topics are included due to either on popular culture topics that can be endlessly sourced to primary material but not secondary works, excessively detailed news coverage, or due to those b trying to ask WP as a promotional source. Most other changes in notability have been a result or the mass article creation leave unexpandable stubs, we have a long way to go in covering more academic material since we are a volunteer project and academic subjects are lacking. We also know that they are minority groups underrepresented on AP that we are limited by sourcing that we are seeking the means to expand. So there are areas we want more inclusion, while we are still more deletions in others.Masem (t) 02:25, 28 August 2022 (UTC)

• 'Neither an inclusionist nor a deletionist be' is the basic precept for the approach to notability and quality of sources by New Page Reviewers - their task at least, is not hardcore one way or the other. Their work is significantly constrained however, by some SNGs that favour inclusion under the flimsiest of interpretations of sources that support notability such as by the massive fan base for the world's favourite field sport (outside the US) that will outvote any motion to get its SNG tightened up, and I'm not as convinced as tJayron32 that 'this has been started to be reeled back in'. The natural tendency of the clean-up workers is also to perceive, to their sorrow, the Foundation's policy/philosophy as being 'quantity is more important than quality' and hence the salaried devs' reluctance to service the tools that makes the reviewers' work less depressing.

While the sheer number of articles is something to boast about and may attract donations, the obverse is that the increasing complaints and jokes in the media about the reliability of Wikipedia may well be putting other donors off. Atsme, a lead NPP tutor, sums it up well with her “It's better for us to have 5,500,000 quality articles than 6,500,000 that include a million garbage articles. Funders donate because they are expecting some level of quality in what we publish.” I think NPPers and anyone else who is familiar with the content of the daily submissions of new articles will quickly concur that all advantages brought by the 2018 ACPERM policy have since been lost through the increase in the availability of broadband connectivity and the drop in prices of mobile devices.

Not all the traditional encyclopedic topics have been exhausted yet; as WhatamIdoing suggests, 'many such holes exist' but as basically a technology company for hosting the corpora, the Foundation is ostensibly developing what has become its main goal rather than developing the genuinely required software needs of its encyclopedia editors and attending to the community's appeals for tools for its quality supervisors. Kudpung กุดผึ้ง (talk) 03:54, 28 August 2022 (UTC)

• I'm of the mind that the WMF needs to make more quality sources readily available to editors, especially AfC and NPP reviewers, and content creators. We are expected to determine whether or not a topic is notable based on the cited sources and coverage via what means? A Google search? And what happens if we don't use the right combo of keywords, or the needed material is in a book, or it is in an archived news article, or paywalled journal? Using geographic feature as an example, what are our expectations of widespread coverage for named, obviously notable, geographic features? Should we automatically assume named geographic features are not notable if they fail SIGCOV, and swoosh them away at AfD? I think not, but at the same time, I don't think we should automatically include articles about every named sandbar around the globe. On the other hand, if a particular group of named sandbars on major rivers in different states are known to provide critical habitat for migratory endangered species, then should we only include an article about the endangered species and a little section about the sandbars, or would those named sandbars be notable as standalone articles? I think the latter, as long as those named sandbars are verifiable and can be cited to at least 1 or 2 sources upon article creation. NEXIST would then apply because there's a high likelihood that a group of biologists and researchers have documented important information about those named sandbars in their respective state resource publications, or perhaps USF&WS has published something about them, and what the public needs to know; it could even be something as simple as a USF&WS sign at the location that makes it noteworthy. There may also be a bit of coverage by local news, so where do we draw the line for "adequate coverage" and notability? Is it a general belief that all editors who focus on "cleaning up the pedia of non-expandable stubs" actually know what resources to access in order to expand a stub if they are not familiar with that particular topic? There are many notable topics that have not received widespread coverage as say a celebrity, or sports personality, a disaster, or a scientific break-thru, but that doesn't make them any less notable. We have a finite number of writers/reporters/journalists/researchers and publications in our global talent pool, and they cannot possibly cover everything on a global scale. We should also not overlook potential historic significance of a topic which means we need access to old newspapers, and the like.
• WP:TWL has been moving in the right direction relative to securing more free access to important sources for us, and deserve accolades for their efforts, but free access is still limited which means you may end up on a waiting list. It is time for WMF to shake loose of some funding or at least bargain for more access to paywalled resources so we can properly do our jobs. I have mentioned the paywall issue more than once on Jimmy's TP several years ago before it became the issue it is today. Not all NPP reviewers and serious content creators can access Nature, or PNAS, or JSTOR along with a host of other resources behind paywalls, so how can we properly determine WP:N if we cannot access the necessary sources? What exactly is needed to establish "adequate coverage" when common sense tells us a topic is indeed notable? Some of our guidelines actually work in that regard at AfD, including WP:CONTN, WP:NEXIST and SNG, but they need more clarity and clout, not the opposite as some have suggested. SIGCOV should not be the sole determining factor for notability because there are far too many topics that are worthy of being noted or attracting notice. We also need to craft a specific policy to ward off questionable mass deletions (ArbCom is inching closer toward an RfC on this topic) and pay closer attention to the benefits of SNG because of the nuances that require critical thinking skills and common sense in lieu of a simple binary approach per SIGCOV and GNG. After all, WP is supposed to be the sum of all knowledge. Atsme 💬 📧 10:40, 28 August 2022 (UTC)

  Can you give an example of what sources you want access to? The Wikipedia Library is quite broad Wakelamp d[@-@]b (talk) 10:51, 31 August 2022 (UTC)

• Hi, Wakelamp, Sam just mentioned a few that are back in. I don't know about the access possibilities but the AVMA Journals - American Veterinary Medical Association, I presume we now have access to The Veterinary Journal via ScienceDirect.com by Elsevier? A sideline thought might be access to The Kennel Club and American Kennel Club libraries, and NYTimes, WaPo and WSJ which are behind paywalls. As often as WP cites those news sources, one would think they would be happy to give verified WP editors free access considering we generate a substantial number of the clicks to their online publications. Atsme 💬 📧 15:11, 31 August 2022 (UTC)

  I thought the NYT times was there via Newspaper? I would like the WSJ as well, plus [The TImes]] - have they been vetoed because they are Murdoch? My wishlistalso includes full access to the whole of google books through an agreement with the Author's Guild (??) :-) And scanning of more libraries in other countries . and....:-) ~~~ Wakelamp d[@-@]b (talk) 15:40, 31 August 2022 (UTC)

  Internet Archive has so much stuff it's hard to comprehend, it exceeds what Google Books does, example. They are scanning thousands of books a day. Some will say copyright, but for research on notability it doesn't matter because links are not required to cite a work. -- GreenC 02:08, 4 September 2022 (UTC)

• @Atsme: Thanks for mentioning TWL :) We're working hard to add more publishers to the library, and we've just added both SAGE and (re-added) Elsevier, which are huge collections of content, available immediately to all eligible users. In terms of NPP, since the minimum standards for being a new page patroller are 90 days and 500 edits, almost all patrollers should meet the eligibility criteria of 180 days and 500 edits for the library, meaning they can access Nature, PNAS, JSTOR, and many other collections, without any waitlists at all. Samwalton9 (WMF) (talk) 14:18, 31 August 2022 (UTC)

• Hi, Samwalton9 – it is always good to hear from you, and thank you for the wonderful news & update! Dating back years ago to when I first volunteered as a TWL coordinator, I have considered it a lifeline for WP. I can't imagine what we'd do without it. I think it was Science Direct that I needed access but now that Elsevier is added back, I'm good to go. Thank you for all you do!!! Atsme 💬 📧 15:11, 31 August 2022 (UTC)

Maybe not on topic, but I think the English Wikipedia needs to prioritize bringing unsourced or under-sourced stubs up to a minimum level of quality before worrying about expanding coverage. At the very least, let us try to raise existing articles to a minimum standard of quality faster than we add new poor-quality articles. I have an example of the problem with letting poorly sourced sub-stubs sit around for years, described at User:Donald Albury/The rescue of a sub-stub biography. This is not the first poor-quality article that I have expanded, but it was particularly egregious in how much it had wrong about the subject. That it sat around for more than ten years with so much misinformation in it is an embarassment to Wikipedia. - Donald Albury 16:52, 28 August 2022 (UTC)

There are 1,773 Category:All unreferenced BLPs, and why we're not batch-moving them to draftspace, I don't know. There are another >100,000 Category:All articles lacking sources and >430,000 Category:All articles needing additional references. I remember a recent discussion about this and there was no consensus to batch move these out of mainspace, as I recall. Levivich 19:27, 28 August 2022 (UTC)

Who is "the English Wikipedia" that should do this, if not us? And what if I decide that my own WP:VOLUNTEER efforts are best spent doing the opposite?

I fairly often see editors who have made "only" dozens, rather than tens of thousands, of edits who are adding sources to existing content or adding new material plus a source. I wonder what we would find, if we compared the contributions of the editors in this discussion against the ideal of expanding and sourcing existing articles. My own contributions today have likely removed more sourced text than I added. Several people in this discussion haven't expanded or added a source to an article in a long while. In fact, several haven't touched the mainspace for days – or months, in at least one case. I have heard that one of the Wikipedia's has a WP:NOTHERE concept that expects 30% of all edits to be in the mainspace. It's not a hard-and-fast rule (e.g., you don't want to ban someone who solves problems with templates), but their idea is that Wikipedia needs more getting the work done and fewer people who do little except tell others what they ought to do. I suspect that many of us here would be in trouble. I suspect that, if you look at the last several years, you'd find that three-quarters of my edits were outside the main namespace. If you didn't count non-content-oriented edits (e.g., formatting edits, WP:AWB runs, etc.), then perhaps a large fraction of the Wikipedia:List of Wikipedians by number of edits editors would be consider "not here" to write an encyclopedia. WhatamIdoing (talk) 19:45, 28 August 2022 (UTC)

The juxtaposition here is a bit odd since Donald Albury has 50,000+ edits of which 60% are in mainspace, and created 300+ pages in main... Andre🚐 21:36, 1 September 2022 (UTC)

You and I, on the other hand, would not fare quite so well. Only 24% of your edits were in the mainspace last month (Donald's score was 49%). But beyond the number of edits, I feel like we (we, the core community; we, the people who hang out at the village pumps) tend to say things like "The English Wikipedia needs to do add more references to under-sourced articles", and we will all solemnly agree, but then we don't add more references. We agree that somebody ought to do that, and then we go back to the edits that we enjoy, like tagging articles written and sourced by other people so they'll know we found their work deficient. WhatamIdoing (talk) 22:54, 3 September 2022 (UTC)

I typically average only 30% in mainspace, but I do spend a lot of my edits nowadays expanding and editing references and expanding text, and recently I've cleaned up a few articles that were languishing for years. Still, I agree that there's no shame in being a gnomish editor, or spending a lot of time in discussion and meta land, or only uploading files, or mostly reverting and blocking vandals, or whatever it is volunteers want to do. I do agree though with Donald's point as well that there is plenty of work to do improving existing articles, though it's not always the most exciting or glamorous work. Andre🚐 23:54, 3 September 2022 (UTC)

This may be contributing to a pissing contest, but I think I am in a position to call for improving sourcing in existing articles, as that is a lot of what I do. (As an aside, the percentage of my edits that are in main space is fairly high because I don't post much on project pages like this.) Now, I understand and appreciate that editors contribute in different ways, and no one can be forced to do something they don't want to do, but I think we can say that some ways of editing do not improve the encyclopedia, especially if they create unnecessary work for other editors. It is policy that everything in an article must be verifiable from reliable sources. Failure to provide citations to sources creates problems for readers and other editors. I will look for and add sources when I have time and the resources, but I will also tag unsourced material when I feel that it is important to do so, and I do not have the time and/or the resources to find suitable sources. People may want to ignore it, but there is the principle that the burden of providing sourcing for material lies on those who want to keep in the encyclopedia. Donald Albury 15:12, 4 September 2022 (UTC)

• I'm not sure this is a good or productive way to view debates over our content policies - it's a bit WP:BATTLEGROUNDy. But as someone who has edited since 2004, one thing I'd point out is that prior to the broad adoption of the WP:GNG, it was not uncommon for people to argue that certain subjects or topics were inherently non-notable, regardless of the sources produced for them (or at least to demand sources far beyond what the GNG now requires) - "not notable" was a sort of vague, often hard-to-answer WP:AFD argument that people would make for all sorts of reasons. The GNG set a clear threshold, which made that less common (and made it easier to answer an assertion that something isn't notable.) --Aquillion (talk) 05:33, 4 September 2022 (UTC)

Letter-like unicode characters in quotes and cites

Sometimes websites (and particularly people posting things on Instagram, Twitter, Facebook etc.) use unicode characters that resemble plain Latin characters, substituting these for their "plain" equivalents purely for typographic effect. Examples include:

• Mathematical Alphanumeric Symbols - e.g. 𝑼𝒏𝒊𝒄𝒐𝒅𝒆
• Letterlike Symbols - ℂℹℙℍ℮ℛ
• Enclosed Alphanumerics - e.g. ⒽⒺⓁⓁⓞ
• Latin Extended-B's "turned" characters to create upside down text (or mixing those with codes from other, unrelated, pages, to simulate that effect) - e.g. ɐᴉpǝdᴉʞᴉM
• Using Combining Diacritical Marks purely for decoration - e.g. ẓͭa͚͍̓̊͗ͅl̠̺̥̃̒̐ğ̲͖̱̍͊o͍̫̜̒͑͛

There are websites which offer cut-and-paste translation into all of these, so it's easy for someone to do (without any arcane unicode knowledge, and with no ill-intention).

It strikes me that these are readability and accessibility problems when that text is copied into Wikipedia articles (either as quotations, or as the titles of references). I'd love to know what a screen reader makes of all of the above. I chanced upon Mateusz Malina today, which references a Facebook page than uses Mathematical Alphanumeric Symbols purely for bolding. I'd be inclined to simply rewrite the text in the plain Latin characters, but I'd like to cite the specific part of the MOS that covers that. Which is a problem, as it seems WP:ACCESSIBILITY and MOS:CONFORM are silent on the topic.

Is there anywhere else in MOS that covers this?

If there isn't, I'd suggest there should be. Something to the effect of "if the original text uses <the above wacky stuff> to substitute for Latin characters solely for typographical effect, and the intended Latin text is unambiguous, convert the text into the equivalent Latin characters".

Thoughts? -- Finlay McWalter··–·Talk 13:48, 12 August 2022 (UTC)

Oh, I forgot Halfwidth and Fullwidth Forms - e.g. Ｔｒａｎｓｌａｔｅ - in addition to deliberate typographical effect, you sometimes see this in text from Japanese or Korean sources, likely where the person creating the text wasn't a confident English speaker, and wasn't aware that the text they were creating would be so odd. -- Finlay McWalter··–·Talk 13:57, 12 August 2022 (UTC)

@Finlay McWalter well, for page titles: Wikipedia:Naming conventions (use English). Also, as this is the English Wikipedia, articles should be written in English, spelling words with those symbols instead of letters isn't English. — xaosflux ^Talk 14:04, 12 August 2022 (UTC)

Yes, I'd certainly that's the case for titles and the text of the article. But we do have quotes in foreign languages many places. I'd think the matter of translating these from pseudo-english to actual-english would be uncontentious. -- Finlay McWalter··–·Talk 14:10, 12 August 2022 (UTC)

If you are looking for a Manual of Style entry that says "Articles in the English Wikipedia must be written in English", I don't think I can help you there (See also: Wikipedia:You don't need to cite that the sky is blue.). The "cover" of our encyclopedia (Main Page) does say: This Wikipedia is written in English. I also can't point you to a MOS that says "Words used in articles must be spelled correctly", see again WP:BLUE. Spelling words using such symbols is not considered correct spelling, so in general they should not be included in articles. If the symbols themselves are especially noteworthy of the subject, they could quoted in article prose (which should explain why they are noteworthy, and be reliably referenced).

Regarding actual foreign language quotes, see Wikipedia:Manual_of_Style#Foreign-language_quotations — xaosflux ^Talk 14:38, 12 August 2022 (UTC)

• This seems like a solution in search of a problem. Unless and until you can produce BOTH of the following two conditions 1) Places in wikipedia where this is being done and most importantly 2) Where you have been prevented from fixing the problem yourself, then there is not any reason to do anything about this. Can you give examples of these sorts of problems, and can you demonstrate, with diffs, how you are not able to fix them yourself? --Jayron32 15:28, 12 August 2022 (UTC)
• Is MOS:CONFORM what you are looking for? A quotation is not a facsimile and, in most cases, it is not a requirement that the original formatting be preserved. Formatting and other purely typographical elements of quoted text should be adapted to English Wikipedia's conventions without comment provided that doing so will not change or obscure meaning or intent of the text. Caeciliusinhorto (talk) 07:38, 13 August 2022 (UTC)
• In the absence of any real and significant problems changes to existing guidelines are premature, see WP:BLOAT. 74.73.224.126 (talk) 15:51, 13 August 2022 (UTC)
• Way to spill the WP:BEANS. I already got to ruin Help:Link once by finding nightmare anchors in template help archives – I can't wait to see how well wiki code handles Zalgo! SamuelRiv (talk) 22:52, 27 August 2022 (UTC)

North Korea: reliability of sources

A long while back, I made an RfC on the source Daily NK. In the discussion, we discussed how sources are generally unreliable when it comes to the topic of North Korea due to the country being so closed off, so much so that even reliable sources like New York Times often reports false information on the topic. Here is an excert of the closing admin's remarks:

The Daily NK has been shown to be wrong multiple times; however, it has noticeable WP:USEBYOTHERS by some of our most reliable sources, not to mention the South Korean government, which you'd think would be the most interested ... For the area of North Korea we should treat all sources with a great deal of skepticism ... Certainly we don't use Daily NK alone for information in Wikipedia's voice, but does that mean "don't use" or "use with attribution"? It has to depend on the specific instance. That's not going to be an easy rule to follow, but a) we don't really have the option to write nothing about a nation of 25 million people that regularly makes front page international news, yet b) to be absolutely safe we'd basically have to do that, because there are no good sources. DailyNK seems as good as any, which isn't very. (There are plenty of worse ones!) We are here to present the world's knowledge, and the world talks a lot about North Korea, but what it says is often wrong.

In the discussion, it was brought up that perhaps there should be an addition to policies or guidelines, such that it addresses these specific issues. In particular, if it is the case that there are no reliable sources on something that needs to be covered (in this instance, modern internal affairs of North Korea, but would generally apply to things that are closed off or treated in a secretive manner), generally unreliable sources may be used without attribution, albeit with great caution.

Should there be an addition to policies or guidelines in regards to this, whether using the logic I have conveyed or using a different path of logic? Or perhaps a different set of actions should be taken, or this concern is already adequately addressed by policies and guidelines? TheGEICOgecko (talk) 18:09, 15 August 2022 (UTC)

I can't remember where I saw this, but I recall editors asserting not too long ago in a source discussion about NK (unfortunately, I think it was located on a talk page about an NK-related article and thus can't find it) that despite the popular perception of NK as a hermetically-sealed state totally inaccessible to outside reporting or research, that hasn't really been true since the 90s, and high quality research is available in scholarly publications. I'm not a subject matter expert on NK, but a quick glance at Google Scholar results for articles about NK published in the last decade appears to confirm this. There may still be cause to doubt normally-reliable English-language journalistic RS on NK, or to use Daily NK for uncontroversial subject matter, but my sense is that NK is not significantly less accessible to RS than other countries with minimal press freedoms such as Uzbekistan or Russia, and may actually be more accessible than other countries such as Eritrea or the self-proclaimed DPR/LPR. _signed, Rosguill ^talk 18:33, 15 August 2022 (UTC)

I think this whole idea of "generally unreliable" needs to be re-worked from the ground up. Being from a "closed off" society does not make a source "generally unreliable". If that were the case, then all historical sources would be generally unreliable, because there's no society on Earth so "closed off" as one that requires time travel to visit it.

Think about that source. There are many things that we would not want to use it for ("The Glorious Leader announces that poverty has been eradicated"). There are also many things that it's useful for ("On Monday, Lee Kim was appointed the new Minister of Foo").

Consider the equivalent in any struggling or repressive country. Maybe you will think of Cuba or Venezuela. Maybe you will think of Iran or Syria. Maybe you will think of the difficulties of reporting associated with Ukraine–Russia crisis. It doesn't matter what the specific situation is. As a Wikipedia editor, you take the publication (newspaper, government press release – it doesn't really matter for this purpose), you look at its context and reputation (hmm, you get jail time for "insulting the monarch" in Thailand; might want to take that into account, if the subject is criticism of the Thai government), and you figure out whether this source is strong enough to support the thing you want to write about. If it is, it's "reliable". If it's not, find another.

Something isn't reliable if it could be used for anything or everything. It's reliable if a good editor would accept that for the specific claim. WhatamIdoing (talk) 20:43, 15 August 2022 (UTC)

I get what you're saying, but "generally unreliable" is only a general description of the reliability of a source. Of course a source should be used as if it were reliable, regardless of how reliable the source generally is, if it can be determined to be reliable in that specific instance. And no, being closed-off isn't a particular requirement to make a source generally unreliable, me mentioning the closed-offness only describes the problem rather than implies that it inherently causes a problem, closed-offness does not inherently make something unreliable. Historical sources are only considered reliable if experts come to a consensus that it is reliable (or something of the like that externally gives the source credibility). The fact that it is historical means it probably should be treated with a great deal of skepticism, without any external/secondary credibility. With the topic of North Korea, the case is the same, any source talking about the topic should be treated with automatic skepticism, rather than assumed general reliability, as one might be able to do with New York Times. TheGEICOgecko (talk) 06:24, 16 August 2022 (UTC)

I just used the Daily NK thing as context, but what I'm proposing isn't specifically about North Korea. If it is the case that North Korea is not nearly as inaccessible as people generally believe, I assume there are other topics, less broad than that of a whole country, that would also fall under this category of there being no reliable sources, despite it being a topic that needs to be mentioned? I'm not particularly familiar with topics that have sparse public knowledge, but perhaps things that are classified? Or what a user mentioned in the Daily NK RfC, "What are conditions inside of China's Uighur camps, for example? Or how many Russians died in WW2? Or in the Gulags?" TheGEICOgecko (talk) 06:30, 16 August 2022 (UTC)

In theory, that problem doesn't exist. If a point is not mentioned in any reliable source, then that point is automatically undue and mentioning it would be a violation of NPOV. And now that I've told you that you're completely wrong and that the problem doesn't exist, let me give you about a thousand examples of this problem on wiki:

• MEDRS declares all primary scholarly sources and all mass media to be basically unreliable; and yet we still have articles on Category:Experimental drugs, some substantial fraction of which cannot be sourced to MEDRS's ideal.
• See also all COVID-19 articles, especially when considered from the start of the lockdowns in March 2020: no proper peer-reviewed secondary sources in academic journals for 99% of the content, barely any truly independent secondary sources (especially about vaccines and the efficacy of each government's actions), and yet we decided that we needed a huge number of articles on those subjects.

In other cases, there are things we feel "needs to be mentioned" but which cannot be sourced to a "generally reliable" source. Consider, e.g., birthdates (or at least years) for most BLPs. Self-published posts to social media are not generally reliable sources. If you are writing about a typical notable-but-not-exactly-famous person, like a business owner or a typical professor, the only source for the birthdate is: whatever the subject posts on the internet. NPOV would tell you not to bother putting that in the article (because it wasn't important enough to independent sources to mention it, so we shouldn't), but some editors still feel like a biography needs to include that information. Is a statement that "Today's my birthday" on Twitter a reliable source for objective facts? Not "generally", but it is "actually" reliable for this purpose, because editors actually do accept that as a strong enough source for that statement. WhatamIdoing (talk) 20:01, 16 August 2022 (UTC)

I guess what I was trying to say is for the instances when there is no reliable source, even when generally reliable sources cover it. But I think you sort of implicitly answered that too: policies and guidelines on the matter uses the term "reliable sources", not "generally reliable sources" (since Perennial sources is only an essay). So if I'm interpreting NPOV, RS, etc. correctly, if there were to ever be a time a topic is extensively covered by all the news organizations, and if it can be determined that all of the sources are reporting in an unreliable manner, then the topic should not be covered, or it needs to be covered but only with attribution or something of the sort. TheGEICOgecko (talk) 14:57, 17 August 2022 (UTC)

I have been thinking this year that we need an actual definition, of the sort that would be approved by a professional dictionary editor, for the term reliable sources. So far, I'm thinking that the definitions are:

Reliable source

A source that experienced Wikipedia editors accept as being appropriate and adequate support for the specific statement in question. For example:

Y A specific tweet from Donald Trump's Twitter account, to support a statement that he tweeted something.

N A chemistry textbook, to support a statement about Donald Trump's use of Twitter.

Generally reliable source

A source that experienced Wikipedia editors would probably find useful for writing an article, or at least a substantial paragraph, about the main subject of the source. For example:

Y An organic chemistry textbook.

N Anything on Twitter.

So far, we've avoided having definitions. We tell people how to identify a reliable source, but not what it is. I think we need this partly because of the rise of RSP, which re-uses the "reliable" language to it means something different. WhatamIdoing (talk) 20:00, 28 August 2022 (UTC)

Your definition is by community norms, elucidated through example. That is ('by definition') a definition defined solely by praxis, as in how to identify a reliable source.

And I'll accept that there are a sizable number of people in the world who find dictionary-style definitions illustrative in themselves. But only if they'll accept that there are a sizable number who find exactly the opposite. There's a reason it's a comedy meme (TvTropes). SamuelRiv (talk) 20:16, 28 August 2022 (UTC)

@SamuelRiv, I think that "by community norms" actually is our definition, and I'm concerned that not owning up to this (in the form of writing the real definition in the official documentation) leads people to assume that there is, somewhere, a different, consistent, objective, "better" definition – maybe even one that could be turned into an algorithm, so that Friend Computer could tell you whether the source you cited is reliable. If we write down that "reliable source = whatever sources we accept for a given statement", then (after the usual two-year lag for people to notice) people might stop saying "But it can't be reliable, because it's biased!" (or self-published, or non-independent, or in a newspaper, or whatever). WhatamIdoing (talk) 23:02, 3 September 2022 (UTC)

And this might stray a bit off topic, but if I accurately described Wikipedia policy, should WP:DAILYNK be changed? Or is there a reason it is still sound reasoning that is in line with NPOV? TheGEICOgecko (talk) 15:10, 17 August 2022 (UTC)

Yeah it should probably be changed. I'm not seeing it as a good source if it's the only source for a fact. If there are other sources, and we are reasonably confident that they have independently checked the fact rather than credulously taking NK Daily at its word, that's different. But then you can use them instead. So, some other observations:

Yes, it us correct that most all sources are less reliable than you probably think. But we have to use something.

Persons are generally poor sources for information about themselves, as they are not a neutral observers and are subject to wishful-thinking memories (or sometimes flat spin or cherry-pick facts). Regarding birthdays specifically, Stan Kenton, Gene Tierney, Jackie DeShannon, and Mariah Carey are four people I know of who either thought their birthday was different than it really is, or else played cute with their birthday for professional or personal reasons. In all four cases forensic digging was required to get the right date. I'm sure there are many more. Birth certificates reliable for our standards, but can (rarely) be wrong also.

WP:RS correctly notes that there are only two sources that are assumed (not proven) to be probably reliable for a given fact: sources with robust independent fact-checking operations (such as Der Spiegel for instance) or that are known to have been independently fact-checked, and peer-reviewed academic journals. Everything else is up for discussion.

The "attributed, but not given in our voice" can be a bullshit workaround. "According to [source], John Smith is a mountebank" has, basically the same impact on many readers as "John Smith is a mountebank". Kind of like "I'm not saying it, I'm just saying that a lot of people are saying it". Sometimes attribution is OK, but a lot of times not.

Some of the things I would like or need to know about Daily NK before considering it reliable are

• How many fact-checkers do they have? How good are they? How much time are they given to fact-check articles, and how do they do it? Are they genuinely independent -- that is, can or cannot an editor overrule them? (Granted, this is hard to find out for most any publication. Sometimes you can make an educated guess.)
• What's their business model? What does their target audience want from them (for instance, the target audience of the The Economist expects them to get their facts right or else they'll stop subscribing; is NK Daily using that sort of business model)?
• Do they have any agenda? Do they have a motive to spin or cherry-pick facts?

And more, but those are some key questions. Are the answers to these questions satisfactory, and do we know these answers with a sufficient level of confidence?

If there is any fact that the is reported inn NK Daily and not elsewhere, we should leave it out. Yes I know sometimes we really want to report some fact that we consider important, that is important, and we consider NK Daily to very likely be correct on this fact, but "very likely" is below our bar for reporting facts. Herostratus (talk) 01:18, 19 August 2022 (UTC)

And for problems with self-sourceing for birthplaces, see Talk:Burt Reynolds from 2007. Donald Albury 14:36, 19 August 2022 (UTC)

@Herostratus wrote Birth certificates reliable for our standards, and I would finish that sentence "...but their use is banned by WP:BLPPRIMARY.

We don't actually require fact-checking. We especially don't require "independent" fact-checking (which almost doesn't exist, for any reasonable definition of "independent"). I think editors would do better to focus on the line in WP:RS about whether the source is "an appropriate source for that content". State-controlled media is a very appropriate source for some content (e.g., names and titles of various state officials). It is a very inappropriate source for other content (e.g., the motives of states they're in conflict with). WhatamIdoing (talk) 23:03, 26 August 2022 (UTC)

[Re: "...but their use is banned by WP:BLPPRIMARY"]: It's a wiki, nothing is "banned" that improves the project, see the core and founding rule WP:IAR. Not using birth certificates is obviously silly, and I ignore anything that tries to make me do silly things, and I recommend it.

[Re: everything else]: Alright. Yes, the requirement for fact checking is not absolute. (There is independent fact-checking, in publications where, by policy and culture, the fact-checkers don't report directly to the editors and have the final say on what goes in, or a least the power to make their case fairly; and the peers in "peer-reviewed" are also independent and able to make their rulings stick, or are supposed to be.)

Yeah state-controlled media is pretty good. If the main paper in Dictatorland says that Pinckey Pruddle has been made Minister of Defense, you can be pretty sure that that has been stone cold verified (cos if not, and it's wrong, somebody is going to get shot). As is often true, we don't know what their internal fact-checking operation is actually like, but we can deduce that it must be sufficiently rigorous for our purposes.

But the Daily NK isn't state media. It's a private organization. Its article says "Its sources inside North Korea communicate with the main office using Chinese cell phones"... well I mean if they get a call saying that Kim Jong-un was spotted on a train heading to China, Daily NK can use that if they want. Can we? No, not even close. Maybe the caller didn't see what she thought she saw, maybe she misunderstood where the train was headed, maybe she didn't see it but was told it by a person whom she considers reliable but got it wrong, or whatever. If it was the New York Times reporting on Justin Trudeau flying to Mexico, we can be confident that they have checked this with the Canadian government and/or otherwise fact-checked it. The New York Times is not going to publish "Trudeau Making Mexico Visit" based on what one person thinks they saw at the airport.

Not only that, but come on: agenda, much? Daily NK looks to be a fine publication. But I mean they're anti-North Korea. If they're not publishing or withholding or polishing or spinning or accepting items to fit their agenda, that violates what they're about. They might not do it anyway, but I'm not sufficiently confident of that, knowing people.

Again, we are talking about if NK Daily is the only source for our fact. Sorry, but I can't as a general rule have sufficient confidence that the fact is almost surely true. Heck, even the New York Times and everybody else gets things wrong often enough. Of course there are exceptions, but really the only exception I can think of is that the fact has also been verified somewhere else. Herostratus (talk) 02:49, 27 August 2022 (UTC)

I disagree; IMO not using birth certificates is very sensible. There is no way to know be certain that the "Robert Smith" in the birth certificate is the same Robert Smith the article is about.

I feel, overall, like you might not be very familiar with the practical side of fact-checking processes. I think you might be interested in reading articles like these: [1] [2] WhatamIdoing (talk) 21:39, 28 August 2022 (UTC)

Well, we can't always trust our feelings. I've read those articles and many others, and I rationally believe that I know what I'm talking about. — Preceding unsigned comment added by Herostratus (talk • contribs)

First, RSN is not the determinant of WP policy, not even when they "deprecate" a source. It is Essay-equivalent. Second, the MOS and guides to citations don't yet detail how to do secondary citations, which is a pity. In cases where something like the NYT cites NKD, then you do a secondary citation: "NKD, cited in NYT" or if you prefer, "NYT, citing NKD". RSN recently affirmed this for another deprecated source. Whatever you think of NKD or any other source with similar dubiousness, if a mainstream high-quality news outlet is using them as their sole primary source in their reprinting (as in, "according to NKD reporting" is on every single line), then you do a secondary citation.

As for other cases, such as for example an article about some random off-street in Pyongyang, can you use NKD as a source to note that there is a pothole? Other editors vehemently disagree, but I think this is the exactly thing where WP:ContextMatters should apply. A more important example is for quotations from government officials and agencies -- surely state and propaganda media are fine sources for what some general said to them was their public position, even if he had a position of significant power over them. Is an official statement printed in a popular newspaper somehow not the public position of a government official or agency? It is performative. SamuelRiv (talk) 04:01, 27 August 2022 (UTC)

I think you're conflating RSP and RSN, which is understandable. But I do want to quibble that, when a consensus of editors in an RFC determine that sources are reliable or unreliable, that does have the binding power of editors' consensus, and is not equivalent to an essay. The board itself simply summarizes a binding consensus, and is not itself a policy. Andre🚐 23:18, 27 August 2022 (UTC)

The RfC said that NK News was not reliable. That does not mean that when their reporting is cited in reliable sources it cannot be mentioned. If for example, a NYT article says, "according to NK News, Kim Jong Il cuts his own hair," we can say, "according to NK News, Kim Jong Il cuts his own hair." We just cannot report it as fact. There is no reason why Wikipedia articles should assign greater certainty to claims in NK News than American news media does. TFD (talk) 04:11, 27 August 2022 (UTC)

I agree. Generally reliable sources can be used for truth. Generally unreliable sources can't necessarily be trusted, but that doesn't mean they can never be mentioned and attributed in the appropriate context. They just can't be used to cite the truth of a statement. Andre🚐 04:19, 27 August 2022 (UTC)

Yes, I suppose. However, caution here. As a practical matter, the difference between "Smith has declared bankruptcy over 20 times" and "According to the Daily Unreliable, Smith has declared bankruptcy over 20 times" [and with the ref being "Daily Unreliable, cited in National Reliable") is probably lost on most readers: the reader comes away thinking "Wow, Smith has had a lot of bankruptcies". Sometimes this is used to put something before the reader for sketchy reasons, with the deniability of "well, we're not saying it in our voice".

If the New York Timesreports a fact from NK Daily, have they made phone calls to Korea, or had a local reporter or stringer go and double-check that it's true. I dunno, and "who knows?" is not a very good standard for considering reliability. The New York Times has a different business model than we do, and for all I know their policy is "enh, probably true, print it, if it's wrong we can issue a correction". I don't know. But they do have to usually publish facts pretty quickly, I guess, and their overall top goal is to sell papers. It'd different for their own reporting, they do have some sort of fact-checking operation for that I am pretty sure. Herostratus (talk) 21:59, 27 August 2022 (UTC)

I agree to a point, but I think for a source like the NYT that we consider generally reliable, if they will report "According to NK Daily, blah blah," I think we can report that according to the genrel NYT, NK Daily blah blah. That seems unlikely to burn us. Andre🚐 22:01, 27 August 2022 (UTC)

There's an audience who views CNN as reliable and Fox as not, an audience who views the reverse, and a comparably large audience who only has an opinion on Dainik Jagran versus Dainik Bhaskar. The point of inline in-prose attribution to a notable outlet or author is not because we presume the reader to have familiarity with their reputation (if that reputation is relevant they are wikilinked or their work is described further in text) -- it is primarily to detach a potentially opinionated, nebulous, controversial, unconfirmed, etc. claim to the reputation of a notable person or institution. Whether the reader wants to explore that reputation further is up to them, but that attachment to something notable both strengthens the justification for including the controversial claim, in that someone's reputation and/or career prospects are on the line, and also provides the claim with implicit caveats in wikivoice per the tone of prose the encyclopedia has already established. All this, again, is independent of whatever prior knowledge the reader may or may not have about the source, or whether or not the reader chooses to look into the source further. SamuelRiv (talk) 23:13, 27 August 2022 (UTC)

If the NYT publishes attributed statements, then the assumption is they have some credibility unless the NYT article says they don't. While Wikipedia editors cannot distinguish between true, credible or false claims in unreliable sources and cannot engage in fact-checking, professional journalists can.

Incidentally, a lot of information in news reporting comes from unpublished unreliable (per Wikipedia) sources. For example, an article might say "according to a State Department official," or "sources have told the NYT." Notice the wording an Al Jazeera article published yesterday, "Risk of leak at Ukraine’s Zaporizhzhia nuclear plant: Operator":

Ukraine’s state energy operator said

Energoatom said

The agency said

Russia’s defence ministry said

The ministry said

It said

Reuters could not verify the battlefield report.

Recent satellite images from Planet Labs showed

Regional authorities also said

said Valentyn Reznichenko

Energoatom said

Ukrainian President Volodymyr Zelenskyy said

Zelenskyy warned

Officials said

Ukraine has claimed

Moscow, for its part, accuses

Even if we wait until historians attempt to determine the facts, there will still be gaps. But we couldn't write a balanced article without mentioning unconfirmed information.

TFD (talk) 13:25, 28 August 2022 (UTC)

Authors' works lists

I came across some edits that used citation templates for scholarly footnotes in author bibliographies, gave ISBN numbers and other information where they were not important or could be disruptive as pointed out in the Wikipedia:Manual of Style/Lists of works, resulted in clearly substandard layouts (to my eyes at least) like

• —— (1999). Slapboxing with Jesus: Stories. Vintage. ISBN 978-0375705908.

from my third link below, and generally didn't conform to standard in most articles about authors (see for example Paul Auster, Honoré de Balzac, Günter Grass (albeit with ISBNs steering towards certain editions), Henrik Ibsen, Toni Morrison (same as Grass), Joyce Carol Oates).

I tried to give some first aid to these lists and explain why, but got them all reverted (almost) without comment. So i wonder what others think about this. Does the MOS/Lists of works need to be revised, for example, including the difference between a works list item and a scholarly citation made clearer?

Pertinent articles: [3] [4] [5] [6] [7]

Thank you. 151.177.58.208 (talk) 23:41, 16 August 2022 (UTC)

MOS:WORKS does say {{Cite book}} may be used to format bibliography entries Schazjmd (talk) 23:58, 16 August 2022 (UTC)

Template:Cite book#Display options discusses using author-mask, etc. MOS:WORKS#ISBNs suggests using them. Adakiko (talk) 00:05, 17 August 2022 (UTC)

In Victor LaValle, the editor has chosen to use author-mask=2 which results in the long line. If author-mask=0 is used, there is no line. Perhaps suggest that option on the article's talk page? (Personally I think it looks much better without the line.) Schazjmd (talk) 00:07, 17 August 2022 (UTC)

Both as Schazjmd says and year shouldn't be first. A literary work isn't a piece of research that may be inaccurate in a year. Plus all the things pertaining to an exact edition of which there may be or be going to be several and which aren't important to finding the work anyway (unless, exceptionally, the work is revised, and then a "revised edition [year]" or similar will usually suffice). That too is quite another thing with research. 151.177.58.208 (talk) 00:14, 17 August 2022 (UTC)

When author-mask=0, the year will no longer be listed before the title. Schazjmd (talk) 00:19, 17 August 2022 (UTC)

Ok, thanks. The technical side isn't my thing. 151.177.58.208 (talk) 00:20, 17 August 2022 (UTC)

The year of first publication is important, because someone might want to know which book came first, or to know something about the general time period. It really does matter whether a literary work was published in 1938 or 1998, and there are authors whose careers span more than 50 years.

In general, both the year of publication and the ISBN would ideally be for the first edition. WhatamIdoing (talk) 05:13, 17 August 2022 (UTC)

=== 1 ===

I usually try to use the 1st ed, particularly hardcover version if it is released first; otherwise, I use the first paperback. If it is released only in ebook format on the day of publication, I use that ISBN. I also often write out that it's 1st editi like this:

• —— (2007). Water Logic (paperback 1st ed.). Small Beer Press. ISBN 978-1931520232.

Some people might think it looks "weird", though it is as systemetized as possible. It leads to less anomalistic structures or punctuation mistakes and standardizes the order and what comes first. it's always author, year, edit, imprint, isbn of the ed listed. It gives the optimal amount of info about the novel, etc. in the shortest compact amount of space.

Please let me know if any of this does not make sense or is incorrect usage of the parameters in cite book. I would like to avoid misusing parameters.

=== 2 ===

When it comes to using mask=2, I prefer it since it makes the year and names of the work align with each other in the listing. Having the author name in every single line is redundant, in honest opinion, and I only have authors visible if a co-author is involved, like this:

• ——; Bear, Elizabeth (2011). The Tempering of Men (hardcover ed.). Tor Books. ISBN 978-0765324702.

In this case, the article author is Sarah Monette, with co-author Bear.

I see that some find the line looking bad or unattractive. Is that a consensus among most people? Does anyone know? I think it looks fine and result in less space wasted by author name listed again.

=== 3 ===

Another question is ISBNs. obviously, there's multiple ways to display ISBN, and I've always liked 13-digit over the shorter one, and I type / display only 1 dash. Which is preferable?

• ——; Bear, Elizabeth (2011). The Tempering of Men (hardcover ed.). Tor Books. ISBN 978-0765324702.
• ——; Bear, Elizabeth (2011). The Tempering of Men (hardcover ed.). Tor Books. ISBN 9780765324702.
• ——; Bear, Elizabeth (2011). The Tempering of Men (hardcover ed.). Tor Books. ISBN 978-0-76-532470-2.
• ——; Bear, Elizabeth (2011). The Tempering of Men (hardcover ed.). Tor Books. ISBN 0765324709.

The 10 digit version became essentially obsolete after 2007 as far as I've heard. So for works before 2007, should we be using 10 or 13 digit?

Adding all the dashes to a 13 makes it cluttered, except I know that it's useful to some people since it indicates registrant and location of publication. Infobox standards suggest use the dashes like here https://en.wikipedia.org/wiki/Template:Infobox_book. So practically and aesthetically, overall, how important are dashes?

=== 4 ===

Additionally, I'm interested in page numbers. Normally, citations discussing a particular part of say a scholarly journal would say the page numbers that are relevant. I find it might be a useful way of saying how long the work is like this:

1. —— (1990). The Moonbane Mage (paperback ed.). DAW. pp. 1–254. ISBN 978-0886774158.

however, I realize that this might not be accepted, since a) that is use beyond the original intent of pages; and b) print size and number of words per page can vary. Any recommend on the use of this parameter would be helpful. Create a template (talk) 09:11, 18 August 2022 (UTC)

and sorry it should be 978-0-7653-2470-2 Create a template (talk) 09:23, 18 August 2022 (UTC)

@WhatamIdoing: please note I did not say the year of (first) publication could be omitted. I said it shouldn't come first. 151.177.58.208 (talk) 14:53, 18 August 2022 (UTC)

@Create a template: in the articles I linked to (or for Oates a link from that), and many others, as per a moment ago, you can see how it's usually done.

That includes being done without ugly and unnecessary extra lines after the item dots (let alone semicolons, which don't even belong in a listing. Perhaps colon is what you, like many others nowadays, are after, but that too is unnecessary here). What you mean by "author name listed again" I don't understand. In the works list of an article about a certain author, of course you needn't list neither the name of that author nor a placeholder for it again before every work. Therefore, in many of your own edits and examples, the author isn't what comes first, but the year. As you can see from most articles here (and similar mentions in other places), for literary and artistic works the expected "first" is the title. When they are not actually used and (foot)noted as references, that is.

As for edition, it's the first that counts. Not ever the first paperback unless there's no bound one before. Same goes for revisions if there are any.

Reasons against ISBN numbers are given both in the MOS page and by me. Have you read them?

I, at least, think it's bad netiquette undoing serious and explained edits without explaining. 151.177.58.208 (talk) 15:23, 18 August 2022 (UTC)

Contrast the above entries with the current state of co-author Elizabeth Bear#Published works eg (with header codes replaced by boldface)

Novels

The Jenny Casey trilogy

• Hammered (January 2005, Bantam Spectra)
• Scardown (July 2005, Bantam Spectra)
• Worldwired (November 2005, Bantam Spectra)

...

The Iskryne series

• A Companion to Wolves, co-written with Sarah Monette (October 2007, Tor)
• The Tempering of Men, co-written with Sarah Monette (August 2011, Tor)
• An Apprentice to Elves, co-written with Sarah Monette (June 25, 2015, Tor)

I find this far less noisy and much easier to read, and I suspect it's easier for editors too. NebY (talk) 15:57, 18 August 2022 (UTC)

I think the point is that MOS:WORKS permits multiple ways of formatting an author's list of works. Generally, when multiple ways of doing something are acceptable, an editor who wants to change one acceptable method for a different method should discuss it on that article's talk page first to make sure there are no objections. Schazjmd (talk) 16:08, 18 August 2022 (UTC)

@Schazjmd: agree. And, perhaps, the editor might look how it's done in other articles and outside Wikipedia and give a thought as to why. But let's discuss this here then? 151.177.58.208 (talk) 16:53, 18 August 2022 (UTC)

We also have the problem of stating the obvious, because what's obvious to one person isn't for another and if it's considered obvious in circles that engage in whatever the subject is (literature in this case), it can be hard to argue for just because it's obvious.

Case in point (which I mentioned above): repeating the author's name or providing a placeholder for it for each item on a works list for the article's subject. I wonder if the MOS page editor who stated that citation templates are permissible for a works list meant that such a repetition also should be – and if, then neither I nor common usage agree. Same for year-before-title. 151.177.58.208 (talk) 18:48, 18 August 2022 (UTC)

The CS1 templates have a number of capabilities/advantages over simple unformatted text, but it should be used "properly" to employ them all without detracting from others. First is metadata, which would require the explicit filling out of the author name, and then author-mask if desired, for every field, to be useful. Metadata may not be all that important for a simple list of an author's works on an author's page (since a researcher wouldn't likely want to cite/mine an author page in that way), but doing it correctly may still be helpful for those parsing it. This also precludes a simple wrapper template that omits author info, or uses a dummy -- however it would (might? currently checking on that) make viable a wrapper for a list of a single author's works which simply duplicates the author's (or any other) fields on each entry, with appropriate masking. CS1 also offers nice language features and data checking. But if none of that's really necessary, a non-CS1 template of this simplicity would have less overhead and be significantly easier for novices to maintain. SamuelRiv (talk) 17:12, 23 August 2022 (UTC)

You are quite right: researchers go elsewhere for their info and an encyclopedic article shouldn't be encumbered with technical overhead that precludes both easy reading and editing. Like other places, everything else (including an agreeable layout) can't be sacrificed for a few very technical aspects for a few users or instances. – If author's name (and similar) is that important in some technical ways, is it possible to make a special template for this kind of lists, reasonably easy to use for the non-nerd, meriting recommendation on the MoS page, containing some device that adds the name automatically only when it's needed? Takes it from the head of the page, for example? 151.177.58.208 (talk) 21:52, 23 August 2022 (UTC)

I'm sorry, that's not what I said at all. Wikipedia is used extensively by researchers, both meta and directly, as well as in LIS, AI training, and live data mining. Getting CS1 to implement low-overhead metadata was a critical development. I'm simply saying that biography pages specifically, in a section that simply lists the author's works specifically, does not necessarily have much to offer from metadata alone if they're templated. That does not mean they shouldn't be templated. CS1 still has fairly low overhead. SamuelRiv (talk) 23:22, 27 August 2022 (UTC)

RFC on naming conventions and Articles regarding districts and Councils

An RFC has been raised regarding Wikipedia talk:WikiProject Geography.

Davidstewartharvey (talk) 07:16, 23 August 2022 (UTC)

Consistency

According to https://en.wikipedia.org/w/index.php?title=Special:Search&limit=500&offset=0&ns0=1&search=%22later+George+IV%22 the phrase "later George IV" appears in over 100 articles. Sometimes it appears bracketed, sometimes preceded and followed by commas, at others without punctuation marks. Should it appear in the same way in all articles? Is there any policy generally on consistency? Mcljlm (talk) 12:19, 24 August 2022 (UTC)

No, there's no general policy on consistency and we don't seek to impose it at that level. Indeed, there's a family of articles I'm working on now that are far more inconsistent, and there's nothing wrong with that. I can't imagine the variation you describe is causing the readers any problem and I would hope that Manual of Style guidance isn't required to resolve any current disputes over it between the editors of any of those 100+ articles (so long as no-one makes changes solely for the sake of consistency across articles, of course). NebY (talk) 13:03, 24 August 2022 (UTC)

The WP:OTHERCONTENT essay-bit is a little related. Gråbergs Gråa Sång (talk) 19:42, 24 August 2022 (UTC)

If you are looking to get some sort of informal guideline regarding (UK, or especially European) nobility formatting, or even just a handful of specific royal conventions (I've been inkling to get some guidance on the various countries' stylings of Charles V for a while), I would take your link and some specific suggestions to one of the relevant WikiProjects: Britan or Europe (that's an exclusive or, and don't you forget it! -- unless you're in NI), and/or Royalty and Nobility (I recommend posting in one, and cross-posting a link to the discussion in the other). SamuelRiv (talk) 04:27, 27 August 2022 (UTC)

Notability (sports) Basketball guideline proposal

Hello. I am seeking wider community input for a discussion taking place at the Notability (Sports) talkpage. There is a proposal to add a new SNG to the Basketball guideline of Notability (Sports). The discussion is an RFC. Here is the link to that discussion [8].---Steve Quinn (talk) 19:52, 26 August 2022 (UTC)

Articles for deletion

Why does a "no consensus" vote at an article for deletion (AfD) discussion result in keeping the article. It would seem that if editors cannot agree an article should exist, then it shouldn't.

Since most AfDs attract little attention, the outcome is already weighted in favor of keep, since the creator and other contributors are likely to vote to keep.

I have seen cases where it took several tries before an AfD was successful.

TFD (talk) 13:02, 28 August 2022 (UTC)

I agree, it runs counter to the principles of our policies like WP:ONUS and WP:BURDEN, all of which require affirmative consensus for inclusion. I would support a "no consensus" outcome being a default "draftify". Levivich 15:42, 28 August 2022 (UTC)

I would disagree, draftify should be presented as an option, or the closer can make a judgment call to take a poorly attended AFD no consensus as a draft. What we do not want is a high traffic AFD that is no consensus to be suddenly drafted. Masem (t) 15:49, 28 August 2022 (UTC)

There may well have been legitimate grounds for a 'no consensus results in keep' policy in Wikipedia's early days, when expanding the encyclopaedia took priority over adequate sourcing. That seems no longer to be the general consensus amongst most regular contributors, who quite rightly expect new articles to demonstrate notability (through proper sourcing etc) from the start. So yes, per WP:BURDEN, draftification for no-consensus content would seem a very good idea. As it stands, we are including content of debatable merit to our readers (and to search engines), with no indication whatsoever that it may be problematic. That cannot inspire confidence... AndyTheGrump (talk) 16:00, 28 August 2022 (UTC)

Plenty of AfD discussions net only a couple unsubstantial comments. Outside commentators might just "vote" and leave. A closer has nothing to work with. Does that justify deletion? SamuelRiv (talk) 16:09, 28 August 2022 (UTC)

• Chances are, when there are no sources at all, an AFD will not result in a “no consensus”. That usually occurs when sourcing is “iffy”… or when it seems likely that reliable sources should exist, but simply have not YET been added to the article.

The idea behind “no consensus = keep” is to give editors time to WP:FIXTHEPROBLEM. Remember that there is no rush… should it turn out that the problem can’t be fixed (because we assumed wrong, and reliable sources don’t actually exist), we can always hold a second (follow up) AFD, noting that we tried and failed to find sources. Blueboar (talk) 16:20, 28 August 2022 (UTC)

No-consensus draftification still gives time to fix problems - without displaying questionable material to readers in the meantime. If there is no rush, why the urge to display it? AndyTheGrump (talk) 16:25, 28 August 2022 (UTC)

The idea behind no consensus keep is to give editors time to FIXTHEPROBLEM. What is that based on? This happens on articles that have been around for years and years with poor (often primary) sourcing and questionable notability. The keep !votes are often from fans of a particular niche type article. MB 16:34, 28 August 2022 (UTC)

So what if there are fans of a niche article? The world is like that, a long tail distribution of interest in topics. I often see people deleting because they consider something far down the tail curve as inherently non-notable. Like, how could this community fire station in podunk town be notable?! It conflates popularity with notability. -- GreenC 18:17, 28 August 2022 (UTC)

Notability is based on the existence of significant coverage in independent sources. Fans show up and say keep because they want to see articles on all community fire stations regardless of the coverage. MB 19:01, 28 August 2022 (UTC)

Articles without coverage don't usually pass Keep at AfD. The problem is some see community fire stations and presume Delete first, then figure out how to discount sources second. -- GreenC 00:56, 29 August 2022 (UTC)

This was about No Consensus. Articles without significant coverage can end as Keep or No Consensus if there is little participation except for a few editors who have a much lower standard for what constitutes SIGCOV and a very idiosyncratic take on what is "independent" and "primary". Those are not my words, but a quote from a related discussion. MB 01:25, 29 August 2022 (UTC)

It might be true that "draftification still gives time to fix problems", but research indicates that articles get fixed faster if they're left in the mainspace. If you want an individual article to get edited, then you need to leave it out there where someone will feel like it's worthwhile to fix it. If you want an individual article to stay broken, then put it out of sight, and out of mind in the draftspace. WhatamIdoing (talk) 22:03, 28 August 2022 (UTC)

It seems like this is something that depends on the nature of the article and AFD in question--for recently created articles, TFD's criticism applies. For longstanding articles being brought to AFD due to forking, OR or WP:PAGEDECIDE concerns, keep makes more sense as a status quo outcome in the event of no consensus. _signed, Rosguill ^talk 16:25, 28 August 2022 (UTC)

Other way around. New articles have a higher chance to be actually improved than old articles - the problem(usually that there is no consensus upon notability) has evidently not been fixed in a long time if an article has no consensus, and if it is between "keep" and "Redirect" the option "redirect" should always win(because it preserves the content and allows people to work with the old content if necessary and still applies WP:ONUS and WP:BURDEN).Lurking shadow (talk) 16:56, 28 August 2022 (UTC)

Lurking shadow, I think that perspective presupposes that "the problem" with the hypothetical article in question is real in the case of a no consensus outcome and that we should move towards the most likely long-term solution (that a new article can be fixed and that an old article cannot), whereas my view would be that a no consensus outcome means that there is no consensus and that we default to whatever the prior status quo was. _signed, Rosguill ^talk 17:10, 28 August 2022 (UTC)

Except that there isn't always good reason for giving the status quo extra weight. Article age isn't one of them! Not all articles have been extensively edited(other than automated copyedits).Lurking shadow (talk) 18:56, 28 August 2022 (UTC)

Sure, I can respect that as a perspective, but note that it would retrench rather than resolve the disagreement between AfD processes and our general "status quo wins when in doubt" rule that appears to motivate TFD opening this discussion. _signed, Rosguill ^talk 18:59, 28 August 2022 (UTC)

Expanding on "presuppos[ing] that "the problem" with the hypothetical article in question is real", here are the most four recent AFDs I could find with an outcome of no consensus:

• Wikipedia:Articles for deletion/Roger Kirk (presenter) – No individual claims disputed; the only "problem" is whether we should inflict an article on this BLP. New sources added, which some editors believe meet the usual rules and others don't.
• Wikipedia:Articles for deletion/Jim Cliffe – Dispute about whether WP:GNG requires multiple non-local sources. (Spoiler: It doesn't; nothing does. Even WP:CORP only requires a single non-local source.)
• Wikipedia:Articles for deletion/The Mother of All Talk Shows – Dispute centers on whether the talk show should have an article that is separate from the talk show's regular host.
• Wikipedia:Articles for deletion/Schön Properties (2nd nomination) – Third attempt to get rid of a WP:CORP article about a 50-year-old real estate business operating only in Arabic-speaking countries, which means we have WP:NEVERHEARDOFIT and also can't read/search the national news easily.

None of these sound like seriously problematic articles. The owners of Schön might prefer that their dirty laundry wasn't aired out for all to see, but there's no obvious harm to having the articles vs not having them. Also, I had to check three days' worth of AFDs last week to find just four AFDs that closed this way, so it's not a common outcome. WhatamIdoing (talk) 22:29, 28 August 2022 (UTC)

Only four NCs in three days? Something like 10% of all AfDs I participate close in NC, I would expect that number to be a lot higher... JoelleJay (talk) 01:37, 30 August 2022 (UTC)

At the "mass creation/AfDs ArbCom RfC" workshop-workshop, specifically in the context of NSPORT, I suggested a watchlistable pseudo-draftspace with a longer or indefinite incubation time before auto-deletion eligibility, as well as restrictions on how many drafts could be nominated at AfD or moved into mainspace per week. I wonder if something like that, if feasible at all, could work for NC closes. Users could watchlist the categories they're interested in to see what's added and moved out of purgatory, and the lists could be transcluded in relevant wikiprojects. JoelleJay (talk) 02:11, 30 August 2022 (UTC)

I think there has been good input here. Now the TFD has brought this up I think it is worth looking at. If "no consensus" allows poorly sourced long term articles to remain then I think this should be changed to "dratify" or "redirect." This allows for the option of improving the article without having it listed on search engines (outside Wikipedia). This improves the quality of Wikipedia overall and, as has been mentioned, readers don't run away due to poor quality.

I realize this retrenches the status quo and doesn't resolve the disagreement mentioned by Rosguill, but it is better than the current status quo. Also, as Masem says, for high traffic AfDs "no consensus" should be optional draftify. Optional dratftify allows for a decision that would cause the least disruption, i.e., editors angrily going to DRV. And such high traffic AfDs can always be re-nominated. Concerning a "no consensus" new article, I'm not sure the best way to deal with that - let consensus about that rule the day. ---Steve Quinn (talk) 19:54, 28 August 2022 (UTC)

• I think this starts with a serious rule, like a policy, that in order to exist in mainspace, an article must meet certain minimum criteria (WP:V, WP:N, WP:BLP?). If a mainspace article's eligibility is questioned (like at AFD), there must be affirmative consensus that it meets the minimum criteria, or else some WP:ATD must be applied (e.g. merge, redirect, draftify), unless there is affirmative consensus to delete. Levivich 21:11, 28 August 2022 (UTC)

  So when editors don't agree that there's something wrong with having a separate article about this subject, then you'd like us to assume that there's definitely something wrong with having a separate article? WhatamIdoing (talk) 22:30, 28 August 2022 (UTC)

  Yes exactly. If editors don't agree whether or not an article should be in mainspace, it should not be in mainspace. Only that which we agree should be there, should be there, whether it's parts of pages or entire pages. Affirmative consensus ftw. Levivich 23:01, 28 August 2022 (UTC)

  Let's start by agreeing to delete WP:QUO. Wikipedia:Nobody reads the directions anyway, so as long as we point the redirect somewhere, probably nobody would be any the wiser anyway. ;-) WhatamIdoing (talk) 01:05, 29 August 2022 (UTC)

Our general stance on everything is that if something is going to be controversial, we want to see an affirmative action to do it, and no consensus defaults to no action being taken. I don't think it has anything to do with AfD in particular. Just how the project works from a governance perspective. TonyBallioni (talk) 21:17, 28 August 2022 (UTC)

@The Four Deuces: The main issue here is that you're interpreting this as no consensus to keep, but the alternative no consensus to delete is equally valid.

So we err on the side of inclusion because you can easily renominate the same article for deletion later, and WP:NODEADLINE/WP:NOTPAPER also apply. This also gives the option to find an alternatives to deletion, like a bold merge to some other topic. Headbomb {t · c · p · b} 22:17, 28 August 2022 (UTC)

"No consensus" means no consensus either way. Perhaps the confusion appears because we sometimes say "no consensus" because we don't always want to say "really bad idea, dude". WhatamIdoing (talk) 22:32, 28 August 2022 (UTC)

A deleted article can always be userfied so that the Keep side can keep working on it. But I disagree that articles have no consensus to keep because they are poorly sourced. Usually, it is because the delete editors have found there are too few if any reliable sources available to write an informative and balanced article. Since the article therefore lacks weight, it could actually misinform readers.

In my experience, articles that have no consensus to delete never get developed into reasonable articles. Can you provide any examples where they have?

TFD (talk) 22:57, 28 August 2022 (UTC)

I think it depends on your idea of what constitutes a reasonable article. Wikipedia:Articles for deletion/Islamic fascism (the article has since be renamed to Islamofascism) closed as "no consensus", and it's currently a B-class article. H.V. Dalling looks reasonable to me. Kinetite is short, but still looks reasonable to me. Aziz Shavershian looks reasonable to me. List of largest shopping centres in Australia isn't a subject that interests me, but it looks like there is an inline citation for every entry. All of these ended with "no consensus" at AFD. WhatamIdoing (talk) 01:04, 29 August 2022 (UTC)

Setting the very high bar of "reasonable article" being the same as a Featured or Good article, there are 26 that had previous NC results:

• Global Buddhist Network
• Jason Heyward
• Chrono Break
• Josh fight
• Snitching (song)
• 2010 United States Senate election in Pennsylvania
• Steve Palacios
• Michelle Leslie
• The Liquor Store
• Palestinian enclaves
• Ashley Smith inquest
• David Bednar (general manager)
• Watford Gap services
• West Coast Shit - No consensus result after becoming a Good Article
• Ludwig Augustinsson
• Downward Dog Pose
• Ashlyn Harris
• Cultural depictions of tuberculosis
• Terra Branford
• Adolf Hitler and Stefanie Rabatsch
• Chris Hussey
• Serious (Gwen Stefani song)
• Need for Speed: ProStreet
• Truth in Numbers?
• Tomo Miličević
• Neil Harvey with the Australian cricket team in England in 1948

-- GreenC 01:39, 29 August 2022 (UTC)

Except article rating doesn't assess actual notability and is not a particularly consensus-driven process in the first place. That Neil Harvey article is a prime example of the overly-detailed, UNDUE trivia that accumulates when no one is actually discussing the subject directly, but which when well-crafted appears to satisfy article reviewers. JoelleJay (talk) 01:34, 30 August 2022 (UTC)

I have some sympathy for the idea given that the bar for deletion is quite high. There are so many ways to get to nocon, I think one cannot easily legislate for them all. Selfstudier (talk) 22:40, 28 August 2022 (UTC)

Deletion has balances of power because 1 person can nominate 10 articles in 10 minutes (or less) while to save those articles can take days of effort researching sources, improving the articles, arguing at AfD. It usually never gets done in practice for that reason. The valuable commodity is time. That's why we let it sit until someone has the time to work on it. -- GreenC 01:02, 29 August 2022 (UTC)

And one person can create 10 articles in 10 minutes, while to delete them it takes at least 7 days and multiple other editors each. If we actually valued community time we'd enforce greater restrictions on creation such that most of the time spent on any one article is spent by one editor who wants to document that subject, rather than that plus the effort of 8 other editors with no interest in the subject doing x% of the same work redundantly and in parallel over the course of a week. JoelleJay (talk) 01:48, 30 August 2022 (UTC)

• Just a random idea, maybe a bad one maybe not, but what if articles closed as no consensus were to be automatically added to a list of "AfDs closed as no consensus" & relisted for discussion a year down the line? That'd give people plenty of time to work on it; keeps the article in mainspace during that time which does attract more potential editors than draft/user-space; but still ensures there isn't a risk they'll get forgotten about for years and years until someone stumbles upon it again. AddWittyNameHere 02:21, 29 August 2022 (UTC)

  I like this idea more than NC==keep. JoelleJay (talk) 01:40, 30 August 2022 (UTC)

  And a year later, WP:NODEADLINE will still apply. Just like in 3 months, or 5 years. Automatic relisting is just busy work and bureaucracy, we don't need that. Headbomb {t · c · p · b} 10:16, 30 August 2022 (UTC)

  Sure, some of the time, that will be the exact outcome, and in some cases, the article that gets relisted will so obviously have improved it gets into speedy keep territory. I'm not unaware there are downsides to my suggestion.

  However, in other cases, consensus on notability of a subject may well have changed and the article gets deleted after all. (After all, there's more than enough cases where an article survives one or two AfD listings before eventually getting deleted.)

  As things stand right now, a good bunch of these gets relisted down the line anyway and many of the others would have gotten relisted if folks actually remembered they exist. But it puts the burden of doing so on individual editors who have to remember those articles exist and judge when it's been long enough since the last time at AfD, and in some subject areas it exposes these editors to unnecessary drama and ALLCAPS shortcut accusations like WP:IDHT and WP:POINT levied towards them for re-AfDing it.

  At least if it's an automatic process, the burden of remembering the article's existence doesn't end up on individual editors; folks aren't required to possibly open themselves up to drama and accusations in order to get such articles up at AfD again; and it rebalances the outcome of "no consensus" more closely towards it's actual meaning: there was no consensus for either keeping or deleting (so let's try again some point down the line to see if an actual consensus in either direction exists now). AddWittyNameHere 08:57, 1 September 2022 (UTC)

...it took several tries before an AfD was successful

No, no, you're not understanding what we're about, here. An AfD is not necessarily "successful" if an article is destroyed; most times yeah, but often enough, it's a cockup. The attitude shown by that statement is just silly, in my view. It's not 2010 anymore. There's a whole culture of editors backslapping each other for destroying articles, and we are destroying more OK articles than we should be. Making it easier to destroy more is the opposite of what we need. Suggestion rejected. Herostratus (talk) 03:02, 29 August 2022 (UTC)

A lot of your examples are articles created about non-notable people who subsequently achieved notability. For example, the article about the baseball player Jason Heyward was created before he had ever played as a professional. When it was nominated for deletion, the full article read:

"Jason Heyward is an outfielder and first-baseman drafted by the Atlanta Braves. He played baseball in high school for Henry County High School in McDonough, Georgia. He was selected 14th overall in the 2007 Major League Baseball Draft. He is a 6 foot 1 inch, 220 pound player."[9]

At that time [18 June 2007], the subject lacked notability per Sports personalities as there were no sources providing significant coverage.

Your argument would therefore be a form of WP:CRYSTALBALL, which is creating an article now in anticipation of the topic becoming notable in the future.

There's an upcoming movie starring Pamela Anderson, Paris Hilton, Luiz Guzman and other notable actors, but little has been released about it at this time. An editor submitted it to Articles for Creation, but it was rejected and they were told not to re-submit until the film had attracted sufficient media coverage to meet notability. But if they had created the article, it probably would have survived an AfD because of no consensus. I am sure however that it will attract attention, good or bad, based on the high profile of the actors.

AfDs BTW provide an opportunity for editors to find and add sources to articles. They don't need another four weeks, four months or whatever and then put the community through another AfD.

TFD (talk) 04:32, 29 August 2022 (UTC)

There are cases too where a notable topic exists, but the article is so poorly written that WP:BLOWITUP is the best approach. For example, Left-wing terrorism is a defined concept in terrorism studies with relative agreement on their objectives, methods and which groups it applies to. However, the original article was terrorists who happened to be left-wing, which is not the definition. There was overwhelming consensus for deletion. (See Wikipedia:Articles for deletion/Left-wing terrorism.) Four years later, I re-created the article based on reliable sources. I notice that AndyTheGrump is also a contributor. It was far easier to create a new article than to fix a bad article. And there was no public benefit to have kept a bad article for four years, waiting for someone to fix it. ([[I also recreated Right-wing terrorism which had been deleted at the same time.) TFD (talk) 14:33, 29 August 2022 (UTC)

Oh yes, absolutely, most articles sent to AfD should be deleted. They're memorials, or ephemera, or unsourceable, or COI advertisements, or not-easily-fixable BLP or NPOV violations, or resumes, and so on and so forth. For all the rest, simplify your life. Throw away all the noise, throw away all the THIS CAPITALIZED LINK and THAT CAPITALIZED LINK and the general war of capitalized links. Instead, ask a simple question:

This article has X daily readers. Overall, it would improve the experience of people searching on this term to get a 404 rather than article, because _______.

If you can't fill in the blank with something useful, go do something else. There are cogent reasons that can go in the blank. It's just that "Rule X or Rule Y or Rule Z says to delete, beep beep" isn't one of them,

So, as you say, if the article needs to get blown up, its worse than nothing. If the article says things that are false or might be false (since there's no reliable source) and we probably can't source those with reasonable effort and deleting them all would ruin the article, the article's not much use. If the article cherry-picks to spin the subject, and we can't easily fix that, the reader would be better off getting nothing. If X is at or near zero, there's not much point in having the article. If the subject is so emphemeral that we can guess that X will be at or near zero in ten years or twenty, same. And there's lot of other reasons.

Even if you can, there are some other reasons. Sometimes the article is too far beyond our remit. A how-to. A bare recipe. An essay. Many other things. We've decided not to publish stuff like that, and that's fine. Or, the article might be a net drag on the project, for some reason.

Other than that, what's the harm of having an article about some bohunk footballer from Franistan or whatever. People like to write about that, people like to read about that. You might not like it, but you can't stop them. And our remit is to be a very large and detailed encyclopedia of football ("Wikipedia ... incorporates elements of general and specialized encyclopedias"). Tell me what the harm is. If you can't, move on.

As to "AfDs BTW provide an opportunity for editors to find and add sources to articles", good grief no. I hope editors aren't of the mind "well, this article could use more sources, but I don't wanna do it, I'll send it to AfD so it'll be improved". That would be... not what AfD is for. I mean it is hard to add new sources to an article if we've deleted it. Right? Sure some few articles sent to AfD get improved and saved per WP:HEY. But a lot just slip into the grave. I mean this is an extremely risky way to build an encyclopedia, I really don't want editors to ever be thinking this. Herostratus (talk) 23:13, 29 August 2022 (UTC)

Agreed. Unless there has previously been an affirmative consensus to keep an article, no consensus should default to the article not being kept - either through it being redirected, or through it being moved to draft space. This is in line with policies such as WP:ONUS and WP:BURDEN, and would also partially address some WP:FAITACCOMPLI issues related to article creation. BilledMammal (talk) 02:32, 30 August 2022 (UTC)

• Remember we are here to build an encyclopedia. Deleting pages does not achieve that. And as stated above, no consensus means we don't change the way it is, so if the article exists, we don't delete it. If you want it to be a redirect, then argue for that, and get consensus. If the policy is not clear on what the deletion decisions are based, perhaps we need more discussion on the policy. In my opinion we have far too many biographies that do not pass the GNG, but I don't normally waste time arguing about their deletion, because there are so many special criteria that allow keeping. Graeme Bartlett (talk) 05:18, 30 August 2022 (UTC)
  • I don't see why building an encyclopedia means giving new page creators "first mover advantage", such that I can create any mainspace page and unless there's consensus to delete it, it stays. I'm here to build an accurate encyclopedia--and a curated one, WP:NOTEVERYTHING. Levivich 05:58, 30 August 2022 (UTC)

    I don't see why building an encyclopedia means giving AFD nominators a "second mover advantage", such that they can delete any mainspace page, and unless there's a consensus to keep it, it goes – especially since only one person can create a given article, but there are thousands of us who could try to get it deleted. WhatamIdoing (talk) 23:12, 3 September 2022 (UTC)

  Remember we are here to build an encyclopedia. Deleting pages does not achieve that. Only if the sole definition of "building an encyclopedia" is "increasing the total number of standalone articles". But one way Wikipedia defines itself is by what it is NOT, so removing articles that violate NOT is "building the encyclopedia" just as much as creating articles on encyclopedically-worthy subjects. And anyway, who would hire a gardener who not only doesn't weed your garden, but also actively plants weeds? JoelleJay (talk) 22:07, 30 August 2022 (UTC)

• The problem is that article creation can be done unilaterally without consensus (or even proof of notability), but deletion requires research and at least 3 editors to reach consensus if the article creator contests it. –dlthewave ☎ 06:12, 30 August 2022 (UTC)
• Oppose change (unclear status of the "proposal") - No consensus means default to inclusion because it is no consensus to change - no consensus by a new page patroller should mean return to draft. So the original burden is on the new article, unlike what is noted above. But the burden is instead on the change, which is in fact the norm on Wikipedia. That is why no consensus defaults to keep. Those suggesting it should be userfied are, in effect, proposing deletion by the backdoor. Nosebagbear (talk) 09:48, 30 August 2022 (UTC)

  I don't believe your assertion that there is a consensus for these articles due to the existence of NPP is accurate; the opinion of a single new page patroller isn't enough to form a consensus, and articles created by autopatrolled editors are not reviewed by NPP.

I also think we should avoid bolded !votes until there is a formal proposal. BilledMammal (talk) 12:09, 30 August 2022 (UTC)

@BilledMammal on the former point, it's meant to be akin to that one person could add some content with a source, and if someone reverted it, the onus would be on them to prove it should be included. But if they added such, and someone else disagreed a year later, the onus would be on the remover. On the latter, I would do so, except for the fact we're in VPP, not VPI, and so it's supposed to already be a full proposal. If we don't want bolded !votes then we can shift the convo over to VPI and I'll happily strike. Nosebagbear (talk) 21:30, 1 September 2022 (UTC)

• Oppose change per Nosebagbear. Exactly my thoughts. Pavlor (talk) 11:52, 30 August 2022 (UTC)
• If Wikipedia's mission is to sort the known universe into things that are notable and things that are not notable, defaulting AfDs to keep makes sense. If its mission is to write an encyclopaedia that keeps growing and improving, not so much. – Joe (talk) 12:01, 30 August 2022 (UTC)
• I think this is merely a discussion. The editor who opened this thread did not put forth a proposal. They asked a question and a discussion has ensued. No need to "oppose" or "support" because there is no proposal on the table. ---Steve Quinn (talk) 14:09, 30 August 2022 (UTC)
• As others have pointed out, while keeping recently-created pages as a result of "No consensus" does create some WP:Fait accompli issues with regards to the article creator, I don't think this is really the major issue. The real problem is that some "No Consensus" closes aren't a reflection of genuine and plausible disagreement over something being notable, but are instead the result of !vote counting without due regard for those !votes actually incorporating policy into their reasoning. -Indy beetle (talk) 18:42, 30 August 2022 (UTC)
• The answer to this question is simply that the burden of proof rests on the person who asserts the claim. For a deletion proposer, that's on them. --Rockstone^{Send me a message!} 08:17, 1 September 2022 (UTC)

I have to disagree with Since most AfDs attract little attention, the outcome is already weighted in favor of keep, since the creator and other contributors are likely to vote to keep. An AfD with no participation results in a soft delete, and there are just as many "other contributors" who will come along and vote delete (or WP:PERNOM). If the nominator writes an effective rationale the burden is then on the keep voters to counter that. An effective rationale countered by a keep vote with no or poor reasoning is likely to be closed as delete. I would oppose any change here. NemesisAT (talk) 09:01, 1 September 2022 (UTC)

• Oppose change. The article by definition already exists. Someone has bothered to write it. If there was consensus to delete it then this would result in deletion at AfD. We should err on the side of keeping articles that someone has bothered to work on rather than erring on the side of deleting articles that someone has decided they don't like. If you examine AfDs you will see that there are some editors who never saw an AfD they didn't want to vote delete on; it's just their dogma - if it's been nominated for deletion then it clearly should be deleted. Sometimes the same group of editors votes delete one after the other. As long as there are enough other editors who vote keep then the onus should always be on the deletors to say why it should be deleted and convince other editors of that, not the other way around. It's very easy to vote delete; it's not so easy to put work into writing an article. -- Necrothesp (talk) 15:07, 1 September 2022 (UTC)

  We should err on the side of keeping articles that someone has bothered to work on rather than erring on the side of deleting articles that someone has decided they don't like. would be putting the editor above the reader. The fact that someone "bothered to work on" something doesn't make it necessarily good or valuable, and it certainly doesn't mean it meets any of our policies or guidelines. We should not show the reader anything that we don't agree meets our policies and guidelines. The reader should know that everything they're reading has consensus as policy-compliant. Levivich 15:46, 1 September 2022 (UTC)

  No, it doesn't. But this about a no consensus result, not a delete result! That suggests that at least some editors think the article is worth keeping. The reader should know that everything they're reading has consensus as policy-compliant. Anyone would think that Wikipedia had fixed rules that must be obeyed! We don't and we never have had. That's why we have AfDs and not admin deletion of articles that are non-"policy-compliant" (which is clearly very often highly subjective in any case) without discussion. -- Necrothesp (talk) 16:23, 1 September 2022 (UTC)

  Articles are regularly deleted because they don't fit a strict definition of some guideline despite having many views. This deletionism is not putting the reader first. NemesisAT (talk) 14:05, 2 September 2022 (UTC)

  I bet you can't name three articles with "many views" that have been deleted. Levivich 14:50, 2 September 2022 (UTC)

  I don't know how we would get the page view counts for articles such as 2022 New Mexico parade ramming. WhatamIdoing (talk) 23:19, 3 September 2022 (UTC)

  Straight off the top of my head is List of largest towns in England without a railway station which was the top result in Google for searches on British settlements without a railway station. Tell me how deleting that is putting the reader first. NemesisAT (talk) 23:23, 3 September 2022 (UTC)

  Where to even begin with this comment... JoelleJay (talk) 04:28, 5 September 2022 (UTC)

• Oppose. There are cases where non-policy-based "Keep" votes incorrectly lead the closer to a "No consensus" result. The solution is to better educate closers on discounting such votes. But where there is a legitimate lack of consensus, we should err on the side of WP:PRESERVE. Cbl62 (talk) 16:38, 1 September 2022 (UTC)
• oppose, no consensus has long meant status quo. There are speedy deletion criteria also. Andre🚐 21:43, 1 September 2022 (UTC)

I mean, as usual, I find these discussions depressing. Before the slow Eternal September starting in IDK the mid to late oughts, there was a different attitude. Now we have many people robotically and rigidly following immutable rules and/or using rules as a club to battle for their ideology -- which is destructive often enough -- and other editors teaching new editors that that's the way to roll. It's hard to contribute by writing articles; trolling thru the project and finding articles that don't meet this rule or that rule or the other rule and trying to have them deleted is much easier -- and there are now many editors who will high-five new editors who do that. There are editors who are on a long-term class-warfare crusade to find grounds to have articles about low culture subjects deleted. And there are lot of nominated articles where the nominator, thru either misfeasance or malfeasance, hasn't done due diligence, and often enough nobody checks this. You get a few driveby "Delete per nom" votes from editors who have been brought up in this mindset, and then a busy admin who sees her job as to clear the backlog as quickly as possible, which is most easily done with a headcount..

From the days of Nupedia the rubric was (formerly) that if you had a good article, that people wanted to read, that was within our remit, then you wouldn't delete it. I had an article deleted, a good article, because we're working here with people... how to put this... maybe lack a subtle and nuanced understanding of how ref vetting works and really what we're supposed to be doing here... it is the encyclopedia that anyone can participate in, there's no threshold for subtlety of mind or commitment to the project goals. Anyway, I found this experience both alarming and alienating. Herostratus (talk) 01:45, 2 September 2022 (UTC)

I'm going to BOLDly suggest a couple compromises, for the sake of discussion. So, "Delete" results could be split into to -- something like "Delete with prejudice" where the subject is inherently no good or the article was terrible or a BLP violation or what have you, and Deletes where the article was not that bad, didn't meet the GNG or whatever (or did, but was considered of interest only to the lower classes) but is not actually harmful for people to read. For the latter, we could have a process where: 1) The article is blanked (but not deleted) and protected 2) The reader is instructed how to go into the history and find the last good version and access that That way, the editors who like to delete OK articles get satisfaction, but the reader is also able to access the article.

Or, if we don't want the readers to access deleted articles at all, for the latter we could, instead of having the actually pretty insulting suggestion that reader make it herself (which is OK for articles that have never been made, that's different), we could have a page like this:

I mean, we ought to be straight with the reader and not beat around the bush. Herostratus (talk) 01:45, 2 September 2022 (UTC)

Yes, because you can never have too many unsourced articles on people who played one cricket match in 1845. Dennis Brown - 2¢ 01:58, 2 September 2022 (UTC)

Especially ones that are basically mirrors of the sports database websites where they get their only mention. We definitely need those! -Indy beetle (talk) 08:44, 2 September 2022 (UTC)

Now you're getting it! Except for the "unsourced" part (if it means unsourceable with reasonable effort); we don't want to tell readers wrong things, or things that might be wrong. As you say, per the first sentence of the First Pillar, "Wikipedia combines many features of general and specialized encyclopedias" (emphasis added). It's important to keep this in mind, I think, because after all that is what I signed on to and so did others. People like to make these articles and people like to read them, and people who don't like it are advised to consider the Wikipedian's Meditation.

"Combines many features of" does not mean the same thing as "is". WP is not a specialist encyclopedia, almanac, or gazetteer even if it combines many features of them, but if you signed up to write one of those, you're in the wrong place. Levivich 14:57, 2 September 2022 (UTC)

And FWIW, if you really want to go after articles about very obscure subjects with sources to bare mentions in obscure databases, how about articles like Gogana conwayi? We have thousands upon thousands of articles like that. Call them "biocruft" and get them destroyed, why not. Oh wait, I forgot science is for our sort of people. Sports is for the peasantry. Phhht. Herostratus (talk) 10:20, 2 September 2022 (UTC)

actually, the more recent issue is recognition that these stubs around geolocations,sports figurrs, or special is to namw A few is that they highlight the plight around trying to create articles on underrepresented groups ( dye to systematic bias) like women and minorities, even in just Western cultures. these stubs would never have gotten through the current AFC or drafting processes to be put to mainspace, which the same issue faces those trying to create articles on women/etc.. this doesn't mean that we should delete the existing stubs, but the attitude (currently be drafted into an RFC about mass article creation and deletion) is that mass creation of these stub like article is not recommended without seeking community concurrence. But at the same time, we have to be aware that there are other ways to present the same info without creating microstubs, such as covering the arching genus of a spevies, each known species under it as list entry rather than a separate article, until GNG notability can be shown. Masem (t) 13:51, 2 September 2022 (UTC)

• We have a rule that AfD isn't for cleanup. We also have a rule that says there is no deadline to improve content. Because AfD isn't for cleanup and we don't have any other process with deadlines, any request to improve the sources for an article can be put off by other editors, and there is no limit to how often this can be done. I call this "infinite deferral".
  We also have a rule that says when we're talking about biographies of living people, editors should be very firm about requiring the highest quality sources, but because of infinite deferral, there is no venue at all where editors can be firm about sources. The "no-consensus-defaults-to-keep" facet of AfD rather exacerbates this, I think. I'm coming to the view that we need a place where we can enforce WP:BLP, WP:BURDEN and WP:ONUS, all of which are paragraphs of core policy. If AfD can't be adapted into that place, then I would tend to suggest a new process.—S Marshall T/C 20:54, 4 September 2022 (UTC)

• I sort of 'Oppose'. This is because there are cases when people can't be bothered to be involved as the afd does not interest them, don't either bother looking for any sources, or don't actually look at the sources to see if they are any good. We therefore get no concensus, when Articles should either be deleted or kept. No consensus does not mean keep. Drafty became an option at AFD because editors could see that articles could be improved but didn't have time to do the donkey work at the AFD. My other issue is closers. There are editors closing or extending AFDs making decisions not actually based on fact. I recently nominated County Borough of Southend-on-Sea, for hardly any participation, which the editor extending the AFD deadline made a snide comment about the nomination, which a further editor pointed out was wrong.Davidstewartharvey (talk) 05:33, 5 September 2022 (UTC)

pre-RfC mass-article creation discussion has begun

As part of the Conduct in deletion-related editing case, the Arbitration Committee decided to request community comments on issues related to mass nominations at Articles for Deletion in a discussion to be moderated and closed by editors appointed by the committee.

Workshopping for the first of two discussions (which focuses on mass article creation) has begun and feedback can be given at Wikipedia talk:Arbitration Committee/Requests for comment/Article creation at scale. As previously announced, Valereee and Xeno will be co-moderating these discussions.

For the Arbitration Committee, –MJL ‐Talk‐^☖ 16:33, 2 September 2022 (UTC)

Discuss this at: Wikipedia talk:Arbitration Committee/Noticeboard § pre-RfC mass-article creation discussion has begun

RfC: Updating BLOCKEVIDENCE

Should WP:BLOCKEVIDENCE be updated to explicitly allow administrators to consider off-wiki evidence when making blocks for on-wiki misconduct, as long as that evidence will be made available to all uninvolved administrators and recorded with the Arbitration Committee? 21:52, 6 September 2022 (UTC)

It is well-established that blocks for off-wiki conduct fall outside of individual administrators' blocking authority. However, the situation is less clear when off-wiki evidence contributes to a decision to block a user for on-wiki misconduct: for instance, a user who denies a COI on-wiki, while a LinkedIn profile tells a different story. Currently, WP:BLOCKEVIDENCE, WP:ADMIN § Special situations, de facto community practice, and ArbCom's recent statement on Special Circumstances blocks provide guidance to administrators in inconsistent ways, open to varying interpretations. This disagreement recently received significant attention at the arbitration noticeboard talk page. The proposers of this RfC disagree on the current meaning of BLOCKEVIDENCE, but agree that it is both ambiguous and out-of-date with respect to current practices.

This proposed change to BLOCKEVIDENCE would explicitly allow administrators to block based on off-wiki evidence as long as that evidence will be made available to any uninvolved administrator upon request. In order to ensure the retention of evidence supporting these blocks, administrators would be required to record the evidence supporting these blocks with the Arbitration Committee when making these blocks. The intent of this proposal is to allow administrators to continue to make blocks for spam and undisclosed paid editing, while establishing safeguards for evidence retention.

Proposed new text

If an administrator blocks a user based on information to which not all administrators have access, that information should be submitted to the Arbitration Committee before the block to ensure that the information is recorded in the event of any appeal. Evidence supporting these blocks must be made privately available by the blocking administrator to any uninvolved administrator upon request (for the purpose of peer review or appeal). In the event that the blocking administrator is unavailable to transmit the evidence, the Arbitration Committee will do so. These blocks typically should not be marked as "appealable only to ArbCom" and are reviewable by any uninvolved administrator.

If the blocking administrator is unwilling to share this evidence with any uninvolved administrator upon request, the administrator may not issue a block. The community has rejected the idea of individual administrators acting on evidence that cannot be peer-reviewed. Instead, the administrator should request action from the Arbitration Committee, or from the Checkuser or Oversight team, as appropriate. These editors are qualified to handle non-public evidence, and they operate under strict controls.

A separate set of requirements apply to administrators holding checkuser or oversight privileges. Those administrators may block users based on non-public information accessible only to checkusers and oversighters without emailing the Arbitration Committee. This may include information revealed through the CheckUser tool, edits that have been suppressed ("oversighted"), and information recorded in the checkuser-en-wp or paid-en-wp VRTS queues. These blocks are considered to be Checkuser or Oversight actions, as appropriate, although the technical action to issue a block is an administrative one. All such blocks are subject to direct review by the Arbitration Committee.

Unified diff

If a user needs to be blocked an administrator blocks a user based on information that will not be made available to all administrators to which not all administrators have access, that information should be sent to the Arbitration Committee or a checkuser or oversighter for action. These editors are qualified to handle non-public evidence, and they operate under strict controls. submitted to the Arbitration Committee before the block to ensure that the information is recorded in the event of any appeal. Evidence supporting these blocks must be made privately available by the blocking administrator to any uninvolved administrator upon request (for the purpose of peer review or appeal). In the event that the blocking administrator is unavailable to transmit the evidence, the Arbitration Committee will do so. These blocks typically should not be marked as "appealable only to ArbCom" and are reviewable by any uninvolved administrator. If the blocking administrator is unwilling to share this evidence with any uninvolved administrator upon request, the administrator may not issue a block. The community has rejected the idea of individual administrators acting on evidence that cannot be peer-reviewed. Instead, the administrator should request action from the Arbitration Committee, or from the Checkuser or Oversight team, as appropriate. These editors are qualified to handle non-public evidence, and they operate under strict controls. An exception is made for A separate set of requirements apply to administrators holding checkuser or oversight privileges; such . Those administrators may block users based on non-public information accessible only to Checkusers and Oversighters without emailing the Arbitration Committee. This may include information revealed through the checkuser CheckUser tool, or on edits that have been suppressed ("oversighted") and are inaccessible to administrators , and information recorded in the checkuser-en-wp or paid-en-wp VRTS queues. As such, an administrative action is generally viewed to be made in the user's capacity as an oversighter or checkuser, although the action itself is an administrative one. These blocks are considered to be checkuser or Oversight actions, as appropriate, although the technical action to issue a block is an administrative one. All such blocks are subject to direct review by the Arbitration Committee.

Side-by-side diff

− If a user needs to be blocked based on information that will not be made available to all administrators, that information should be sent to the [[WP:ARB|Arbitration Committee]] or a [[Wikipedia:Checkuser|checkuser]] or [[WP:SIGHT|oversighter]] for action. These editors are qualified to handle non-public evidence, and they operate under strict controls. The community has rejected the idea of individual administrators acting on evidence that cannot be peer-reviewed.<div class="paragraphbreak" style="margin-top:0.5em"></div> An exception is made for administrators holding [[Wikipedia:Checkuser|Checkuser]] or [[Wikipedia:Oversight|Oversight]] privileges; such administrators may block users based on non-public information revealed through the checkuser tool, or on edits that have been suppressed ("oversighted") and are inaccessible to administrators. As such, an administrative action is generally viewed to be made in the user's capacity as an oversighter or checkuser, although the action itself is an administrative one. All such blocks are subject to direct review by the [[Wikipedia:Arbitration Committee|Arbitration Committee]]. + If an administrator blocks a user based on information to which not all administrators have access, that information should be submitted to the [[WP:Arbitration Committee|Arbitration Committee]] before the block to ensure that the information is recorded in the event of any appeal. Evidence supporting these blocks must be made privately available by the blocking administrator to any uninvolved administrator upon request (for the purpose of peer review or appeal). In the event that the blocking administrator is unavailable to transmit the evidence, the Arbitration Committee will do so. These blocks [[Wikipedia:Arbitration_Committee/Noticeboard/Archive_13#Special_Circumstances_Blocks|typically should <em >not</em> be marked]] as "appealable only to ArbCom" and are reviewable by any uninvolved administrator. If the blocking administrator is unwilling to share this evidence with any uninvolved administrator upon request, the administrator may not issue a block. The community has rejected the idea of individual administrators acting on evidence that cannot be peer-reviewed. Instead, the administrator should request action from the Arbitration Committee, or from the [[WP:Checkuser|Checkuser]] or [[WP:Oversight|Oversight]] team, [[Wikipedia:Arbitration_Committee/Noticeboard/Archive_13#Special_Circumstances_Blocks|as appropriate]]. These editors are qualified to handle non-public evidence, and they operate under strict controls. A separate set of requirements apply to administrators holding checkuser or oversight privileges. Those administrators may block users based on non-public information accessible only to checkusers and oversighters without emailing the Arbitration Committee. This may include information revealed through the CheckUser tool, edits that have been suppressed ("oversighted"), and information recorded in the [[Wikipedia:Arbitration_Committee/Noticeboard/Archive_13#Special_Circumstances_Blocks|checkuser-en-wp or paid-en-wp VRTS queues]]. These blocks are considered to be Checkuser or Oversight actions, as appropriate, although the technical action to issue a block is an administrative one. All such blocks are subject to direct review by the Arbitration Committee.

References

If this proposal is successful, the change would be communicated to all administrators via MassMessage, as has been done with past changes to blocking procedure. Wikipedia:Appealing a block would also be updated to reflect this change to blocking policy. Finally, the Arbitration Committee would be recommended to establish a new unmonitored VRTS queue to receive evidence supporting these blocks (distinct from its handling of "appeal only to ArbCom" blocks), with ticket numbers that can be included in the block log.

Co-signed 21:52, 6 September 2022 (UTC):
-- Tamzin^{[cetacean needed]} (she|they|xe)
KevinL (aka L235 · t · c)

Discussion (BLOCKEVIDENCE)

• Support as proposer. I thank L235 for suggesting creating this RfC, in light of our disagreement about the policy's current meaning, one that sees multiple experienced administrators on both sides. It's clear to me that BLOCKEVIDENCE is not in keeping with current practices, and creates a dangerous dead letter of essentially unenforced wording amidst a very important policy. As a new administrator, I have already run into scenarios several times that fall into this ambiguity. To highlight a few examples:
  1. A user recreated an article on a non-notable person, which had been deleted several times in the past. Past versions appeared autobiographical, but the new account seemed to be someone different. I Googled the username and found that someone with that exact name worked in marketing for a company affiliated with the article's subject. I blocked for meatpuppetry, directing other admins to contact me for evidence.
  2. A user was reported to SPI for AfD !votestacking. At issue was whether they knew another user off-wiki, so I Googled their username and found a LinkedIn profile where someone with that name claims to be a Wikipedia editor and claims to work for a company that the user had edited about extensively. I asked the user if they were affiliated with the company. They denied it, and I blocked for UPE, directing other admins to contact me for evidence.
  3. The other day, GoodPhone2022 (talk · contribs) emailed me, (mostly-)confessing to being a sock of AlfredoEditor. In this case, given the username similarity to past sock GoodPhone2020 (talk · contribs), I was comfortable blocking, but if not for that, I would be in an area of policy ambiguity.
• In each of these cases, ArbCom's current prescription is that I would have had to forward the email to a CU-staffed queue, in two out of three cases for a routine sockpuppet. In all three cases, the reason for blocking is on-wiki misconduct, most of the evidence is on-wiki, and the off-wiki evidence was straightforward.
  We have four conflicting rulesets here: BLOCKEVIDENCE, if interpreted to forbid all blocks based on private evidence, forbids these three blocks, even the routine sockblock. If interpreted to forbid only blocks based on evidence that cannot be shared off-wiki with other admins, it allows this (since under WP:OUTING admins can discuss such evidence by email). ADMIN § Special situations, meanwhile, complicates this, in that it could be interpreted to allow these blocks but require making them "appeal only to ArbCom". ArbCom's recent statement, however, forbids that designation for the most part, and, depending on how literally one sentence is taken, forbids making the block at all. And finally, de facto current practice is that administrators do make such blocks and either explicitly or implicitly direct other admins to contact them off-wiki for evidence.
  Since ArbCom has sent no admin-wide bulletin, I suspect that the upshot of ArbCom's recent statement is that it will be ignored and business will continue as normal until someday it doesn't and we get some drama-filled desysop or admonishment that pits admins saying "But we all do this!" against arbs saying "But we said you can't!" I don't like that outcome, and I don't like the status quo of a policy that is both ambiguous and ignored. Critically, even if BLOCKEVIDENCE does allow these blocks, we have the problem that it's not a very good system. Admins resign or get for-caused or die. Admins forget why they blocked someone. LinkedIn profiles get taken down, Upwork contracts get closed. By both formalizing the permissibility of blocks like these and creating a system for admins to store evidence (mandatory when off-wiki evidence is involved, optional but encouraged for complex interpretation of on-wiki evidence), we solve that problem while clarifying the situations under which admins can and can't make blocks like these. -- Tamzin^{[cetacean needed]} (she|they|xe) 21:57, 6 September 2022 (UTC)

  @Tamzin: I would be very surprised if it is "current practice" for admins to make blocks based on nonpublic evidence without having signed the confidentiality agreement for nonpublic information. Do you have any examples of anyone other than yourself doing it? Historically as a project we are very careful about private information and, while I have no doubt that your different interpretation of WP:BLOCKEVIDENCE was reached in good faith, in the subsequent discussion on ARBN, the majority of admins, including functionaries with years of experience in actually handling these kind of blocks within the established processes, agreed with ArbCom's interpretation. – Joe (talk) 09:17, 7 September 2022 (UTC)

References

• Support as proposer. In my view, policy currently prohibits many blocks that are frequently made by administrators (e.g., low-level UPE blocks based on Upwork profiles). But those blocks seem to have become accepted by the community and the administrative corps. This proposal catches policy up to reality while adding a safeguard: the evidence will be recorded in case it’s needed in the future. I therefore support the change.
  Also, because there’s disagreement about what policy currently requires, I’d ask any folks who oppose this proposal to indicate what they think the current policy says. (Are blocks based on info with an “email me for the evidence” note permissible? Does that count as information that will [] be made available to all administrators?) With thanks to Tamzin and everyone who discussed and ideated on this, KevinL (aka L235 · t · c) 21:58, 6 September 2022 (UTC)
• Support - currently we are caught between the former status quo, with the risk of block evidence being lost putting appellants and unblocking admins in an unenviable position and the Arbcom created new rules that simply would put too many tasks on CU-only where they don't need to be. This proposal offers a solution to that, especially in the "low-hanging fruit" part of off-wiki evidence. Tamzin's reasoning is detailed and the examples are a good set of those frequently seen. Nosebagbear (talk) 22:04, 6 September 2022 (UTC)
• Support. Seems pretty straightforward to me. –MJL ‐Talk‐^☖ 22:24, 6 September 2022 (UTC)
• Support. We shouldn't prohibit good blocks just because policy doesn't allow the evidence to go public. I do have one quibble with the proposed text. It says the info will be recorded by arbcom and can be retrieved there by admins, but it also says the blocking admin MUST supply the info to other admins on request. That implies all admins using this policy must retain a permanent duplicate record of the info. That seems pointless. I'd drop the unreliable requirement for admins to supply the evidence. Alsee (talk) 22:26, 6 September 2022 (UTC)
• Support, with thanks to both proposers. The problem of UPE is a significant one, and I'm pretty sure there is community consensus that we need to allow some degree of "research" about users suspected of UPE or even just COI, regardless of the WP:HARASS prohibition of "opposition research". This proposal makes it clear how admins can be effective while still protecting private evidence, and solves problems with forgotten evidence. --Tryptofish (talk) 22:30, 6 September 2022 (UTC)

  @Tryptofish: We already allow this kind of research, by anyone, just with the caveat that only functionaries can make a block based on it. The evidence for these blocks is then logged in one of various VRT queues or private wikis (depending on the kind of block). For example, if you find evidence that a user is making undisclosed paid edits, you can email it to paid-en-wp@wikimedia.org for recording and action. – Joe (talk) 09:02, 7 September 2022 (UTC)

• I support the idea behind this proposal. However, I'm unsure about the following sentence: "In the event that the blocking administrator is unavailable to transmit the evidence, the Arbitration Committee will do so." Can we really require them to do so? This seems to convert ArbCom into a marketplace for off-wiki outing. I guess "will do so if possible" or "may do so" would be more precise. ~ ToBeFree (talk) 22:39, 6 September 2022 (UTC)

  I'm definitely OK with "if possible" or "may do so" if people feel very strongly, but my first thought is that it seems like this is an edge case that doesn't need to be spelled out – presumably we can trust ArbCom not to engage in impermissible OUTING, as they handle all sorts of private information normally. Best, KevinL (aka L235 · t · c) 22:52, 6 September 2022 (UTC)

  Well, as mentioned in Tamzin's support comment, the harassment policy does contain an outing exception specifically for emailing, so it's less of a concern than it looked to me first anyway. Formalizing the existing way for all administrators to access the evidence behind such blocks is a positive development.

  I have reviewed paid-editing blocks based on admin-only evidence (and requests for them) a few times and found it difficult to come to a clear conclusion whether the blocked user was actually lying into our faces or genuinely pointing out a case of mistaken identity. I guess those active at WP:SPI got used to this feeling. Transparency, as far as possible, increases the number of eyes that need to make the same mistake for an incorrect block to happen/stay. I can't really complain about that. ~ ToBeFree (talk) 23:14, 6 September 2022 (UTC)

• Support. This is a reasonable way out of the current conflicting-norms problem. Unlike ToBeFree, I have no issue with the fact that the en.WP community can require its own ArbCom to do something. ArbCom answers to us, not the other way around.  — SMcCandlish ☏ ¢ 😼  23:11, 6 September 2022 (UTC)

  (Now that might be oversimplified as ArbCom consists of volunteers... somehow... within the boundaries of WP:ADMINACCT and similar principles. And, although probably not applicable to the type of evidence we're discussing, there are of course even additional restrictions on what we can require them to do, described in their NDAs.) ~ ToBeFree (talk) 23:20, 6 September 2022 (UTC)

  I agree the community can compel arbcom to do something through Arbpol. This is not that. However if this passes I will absolutely be in favor of setting up an email queue for things to be sent to. Best, Barkeep49 (talk) 06:14, 7 September 2022 (UTC)

• Support This is a good solution for a difficult problem and is necessary to reduce disruption. Johnuniq (talk) 23:26, 6 September 2022 (UTC)
• Support makes sense and clears up an ambiguity/conflict in policies/procedures/best practices. I don't see any reason not to make this change. --TheSandDoctor ^Talk 23:46, 6 September 2022 (UTC)
• Absolute and unequivocal Oppose. While clearly a good-faith proposal, I think this may be the single worst idea I've seen advanced to the community in quite some time, full of ill-considered potential knock-on effects that aren't even contemplated within the proposition, let alone addressed. With due respect to Tamzin, I think they have seriously misinterpreted both the wording of relevant policy and the established community consensus on which it is based.
  For example, the first scenario that they use as evidence for why this system is needed (in their support !vote)... I'm sorry Tamzin, but that's just not the kind of investigation I want you to be undertaking under any circumstances, nor do I think it is properly within your remit as an admin: in fact, I think it is a brightline violation of the wording of WP:OUTING. The exception contemplated in that policy is that a user might utilize information relating to a generic posting by a company seeking COI editors--not the notion that users (admins included) would be tracking down potentially doxxing information regarding specific editors, just so long as they have socking concerns to justify it. That clearly goes against the spirit of the policy and longstanding community consensus.
  I suppose it's true that nothing currently prohibits any community member from acting as a non-sanctioned investigator and submitting such information to VRTS--whom I hope routinely ignore it in the (probable majority) of problematic cases and focus on on-project information and technical assets. But I am deeply concerned about how enabled similarly-thinking admins as Tamzin (again, no personal offense intended, but I feel strongly about this issue) might feel if they perceive a further institutional greenlight on such activities. And note that the outing policy would also need to be rewritten here in order to facilitate this new system, since it currently expressly forbids some of the activity that would be involved, and expresses a very different philosophy with how off-wiki information (and linking it to on-project accounts) is meant to be handled.
  Whats more, in order to facilitate this new and highly problematic role for admins, there is to now be a new log of sorts containing any amount of potentially sensitive personal information on any number of community members (and indeed, where the admin-inspector's instincts are off, personal information of people who may have nothing to do with the project whatsoever), creating one of the greatest systematic doxing risks generated by the project? All it would take is one bad actor getting access to that system, through legitimate or illegitimate means to create a world of harm. Nor should we expect any potential disruption to be limited to just a handful of overzealous admins, since this new system would encourage anyone of such a mindset, and on good terms with an admin who views their authority in this new area as broad, to seek out potentially damning information on other editors to relay it to said admins.
  I'm sorry, but our current policies with regard to the collection, dissemination, and storing of off-project personal information (which may or may not relate to community members) did not evolve in a vacuum: they are meant to place a premium on the protection of anonymity on a project that presents a massive risk of real world harm for many of its members. This proposal would be a significant erosion of that framework, which would invite all manner of potential problems. Far from being a "constitutional overreach", the rules promulgated by ArbCom (which have in any event been status quo for a long time), are, by comparison, much more in conformance with the traditional principles and concerns regarding privacy on this project, and it is (in my opinion) this proposal which would violate existing community norms and important checks and balances.
  In short, very much a case of the cure being much worse than the disease it proposes to address (and which is already effectively controlled by an existing treatment, if one that moves a little slower. I'm sorry, but we cannot, in the name of combating paid editing, vitiate some of our most important privacy policies. It just is not remotely a balanced reaction to that situation. SnowRise ^{let's rap} 01:50, 7 September 2022 (UTC)

  The outing policy, a section of the harassment policy, does not prevent administrators (or anyone for that matter) from investigating users' off-wiki activities, and explicitly notes Nothing in this policy prohibits the emailing of personal information about editors to individual administrators, functionaries, or arbitrators, or to the Wikimedia Foundation, when doing so is necessary to report violations of confidentiality-sensitive policies. If you think it ought to prohibit these things, you should propose a change to that policy. But the status quo is that such investigations and discussions are allowed; the question we're discussing here is who should block based on them, and how. -- Tamzin^{[cetacean needed]} (she|they|xe) 03:02, 7 September 2022 (UTC)

  I find that quoting of the policy incredibly selective, considering the very next two sentences read: "Only the minimum information necessary should be conveyed and the minimum number of people contacted. Editors are warned, however, that the community has rejected the idea that editors should "investigate" each other."(emphasis added). The expansion of an administrator's permitted activities that you are proposing (and at least one of the examples of conduct which you seem to already engage in) are clearly not in keeping with that principle. You are just fundamentally wrong about what the kind of behavior the community has long proscribed with that rule: no, neither admins nor rank-and-file community members are meant to be tracking eachother off-project--that is quite simply a very foolish (and for some, expressly dangerous) notion, and the outing policy is the first and perhaps most fundamental layer in a firewall that exists to protect the privacy (and in many cases even the safety) of our volunteers.

  There is already a system in place for users (admins included) to act on off-project information suggestive of on-project disruption: WP:VRTS. That system seems to aggrieve you because you perceive it as ArbCom somehow dictating the purview of administrators, but there's clearly a lot of important policy rationale for why the system is set up like that in the first place: that information is simply not meant to become part of the record on contributors here, even in cases of disruption. Nor are our community members meant to be openly policing eachother in the manner you would have use normalize.

  What's more, you would have us log all the information thus collected in some fashion broadly available to at least editors of a certain class of permissions--and all that would need to happen in order for such doxxing data to be collected and retained for a user is that any one of our admins thought that maybe it was possible that they were socking... I'm sorry, but do you really not see all the ways that any such system would be vulnerable to exploitation or penetration, deeply undermining our traditional commitment to prioritizing the privacy and safety of our volunteers? I'm afraid that neither allowing for slightly speedier responses to a small subset of COI cases, nor giving a particularly defensive segment of the administrative corps an opportunity to thumb their nose at ArbCom are sufficiently good reasons to abrogate the principle of user anonymity so significantly. SnowRise ^{let's rap} 06:15, 7 September 2022 (UTC)

  (edit conflict)There's a clause in the outing policy that Editors are warned, however, that the community has rejected the idea that editors should "investigate" each other. My reading of the proposal is that this policy is complimentary with WP:OUTING and that both the public meaning and intent of this proposed policy maintain respect for the principle that editors should not investigate each others' private lives. @Tamzin and L235: Yes or no, is this reading of policy yours as well? — Red-tailed hawk _(nest) 06:20, 7 September 2022 (UTC)

  Kevin and I are not suggesting any change to WP:OUTING. My reading of that clause in OUTING—taken in context alongside other language that, as noted, explicitly allows reports of sensitive information—is that editors should not try to "dig up dirt" on one another, especially not speculatively or vindictively. OUTING is not a blanket ban on ever looking at anything anyone does off-wiki, and the community has repeatedly rejected attempts to make it one, something reflected in its current wording. Give that functionaries and ArbCom are bound by OUTING too, any stricter reading of that clause would mean that they commit blockable/desysoppable offenses anytime they block someone for off-wiki harassment—a block that by necessity involves some level of looking at what someone has been doing off-wiki. -- Tamzin^{[cetacean needed]} (she|they|xe) 06:43, 7 September 2022 (UTC)

  That's clearly a non-sequitor: no one (that I have seen anyway) is suggesting that an admin who in good faith comes into possession of evidence of off-wiki harassment should refuse to act to protect the harassed party. And in fact, that's the very reason we have the system that we presently have, which balances user privacy with the possibility of administrative action in the fashion it does (with appropriate non-public oversight). But that is a very different animal from permitting admins (and potentially cohorts working in close collaboration with them) to unilaterally (and on their own onus) begin digging into the off-project identities of users. That is an exception that just cannot do anything but ultimately swallow the rule. It's very clearly the exact bridge too far that inspired the very plainly worded prohibition on investigating your fellow editors in the outing policy. Acting on information brought to you about an especially harmful and chilling class of harassment is one thing; every admin having the power to self-appoint themselves an inspector-general, in any random case of any user they can say they genuinely thought might be a sock, is a very different thing, and something I pray the community will have the good sense to reject here. SnowRise ^{let's rap} 07:11, 7 September 2022 (UTC)

• Support, obviously. If you organized a gang of vandals off-wiki, you should be blocked. Not blocking due to the coms being off-wiki is just taking advantage of a technicality to me. CactiStaccingCrane (talk) 02:10, 7 September 2022 (UTC)

  As for the potential for a power-grab/privacy concerns raised by SnowRise, I agree that the proposal should be flushed out before being implemented. But you need to propose your idea first. CactiStaccingCrane (talk) 02:11, 7 September 2022 (UTC)

• Support admins need more help and more tools to prevent bad editing and problematic accounts. Andre🚐 03:56, 7 September 2022 (UTC)
• Support in principle. The particular part of the policy that states that the information has to be submitted strictly before the block seems to be a bit arbitrary and might delay an admin taking action against actively coordinated off-wiki vandalism organized on something like Twitter. Giving the admin some time after making the block (i.e. within 24 hours or something to that effect) would be superior to strictly requiring administrators to submit evidence before a block is made. — Red-tailed hawk _(nest) 06:20, 7 September 2022 (UTC)
• Oppose largely per SnowRise. I find Tamzin's examples bizarre; they seem to me to be right on the line of desysopable offences (even blockable ones), if not over it. Blocking someone because you googled their user name and found apparent connections to their editing is not okay. Sharing personal information of other editors by email (as proposed here) with any admin who asks for it is not okay. We have teams of people who deal with off-wiki evidence precisely to avoid this situation - those people have to have signed an agreement with the WMF to protect the confidentiality of data they use. Normal admins have not. GoldenRing (talk) 08:45, 7 September 2022 (UTC)

  Also, I really think WMF Legal need to be involved here, as I can see potential implications for the site ToU and privacy policy. IANAL but if admins are sharing editors' personal information at will for the purposes of maintaining the site, does this not expose the foundation to a degree of legal risk? GoldenRing (talk) 08:50, 7 September 2022 (UTC)

  I agree that this should be run past WMF Legal if implemented. Especially the part where ArbCom is supposed to act as a sort of information broker that provides personal information on editors to any admin that asks. – Joe (talk) 12:04, 7 September 2022 (UTC)

• Oppose. This is a terrible idea. as long as that evidence will be made available to any uninvolved administrator upon request – what if the blocking administrator isn't available? Or has left the project? Or lost the evidence? Or died? Just as the evidence for regular blocks are documented on-wiki, blocks for private evidence need to be documented somewhere so that no one person is a bottleneck for an appeal or unblock request, which could (and regularly does) come years after the initial block. ArbCom set up processes for doing precisely that years ago, using secure, WMF-maintained software, staffed by experienced and vetted functionaries who have signed the confidentiality agreement for handling nonpublic information, and it has been working perfectly fine for years. I don't see what problem this is supposed to solve and frankly it seems to stem entirely from one of our newest admins misunderstanding WP:BLOCKEVIDENCE. – Joe (talk) 08:51, 7 September 2022 (UTC)

  I'm also a bit worried that good number of the supporters so far seem to be supporting because the opening text of this RfC gives the impression that we currently don't block people based on non-public information (e.g. of UPE). This is not correct. Users are regularly blocked based on non-public information, but policy restricts these blocks only to CheckUsers, Oversighters, or Arbitrators who have signed the confidentiality agreement. The proposal here is to alter WP:BLOCKEVIDENCE so that all administrators are permitted to make these type of blocks. – Joe (talk) 08:59, 7 September 2022 (UTC)

• Comment The more I think about it, the less comfortable I am with this solution. I like it in principle, but I do think it needs a bit more consideration regarding the personal information. In Europe, we are governed by GDPR with regards to data protection, how long data can be kept, what purposes it can be kept for. Legally, the servers may be in the USA, if a user is accessing the data from Europe I'm not sure how that falls - and that's assuming the data is all held on the servers in the US. At present, the suggestion is that data is sent to the arbcom list, which is then immediately disseminated to Arbitrators mail boxes, which they have full control over. It's hard to argue the "US servers" point of view there, an arbitrator is clearly a data controller.
  Then there's the technical side of things. Based only on information that the user chooses to share - username they created, information they've given etc, you are investigating the individual online. That opens up a lot of risk - of mistakes, of joe jobs, of abuse. We shouldn't be encouraging this sort of behaviour, especially in our administrators.
  I still need to think more about it, so won't outright oppose, but I'm certainly uncomfortable. Worm^TT(talk) 09:01, 7 September 2022 (UTC)
• Strong oppose per GoldenRing, Joe and SnowRise. This is not something administrators should be doing at all, if you are doing it you need to stop it now. Use the existing channels to report any off-wiki coordination you stumble across. Thryduulf (talk) 10:11, 7 September 2022 (UTC)
• Strong oppose Run-on-the-mill admin blocks should be for stopping sustained atrociously poor editing that cannot be solved by attempts at communication, and should be held accountable by the community at large. This entire proposal throws out a basic principle of adminship, public accountability, out the window. In the proposal's scenarios with dishonest COI editors, I fail to see any benefits to blocking users just because of their IRL jobs, instead of concretely visible on-wiki activity like actually writing promotional articles, actually posting spam, vexatious restoration of deleted content, and the like. — Ceso femmuin mbolgaig mbung, mellohi! (投稿) 12:53, 7 September 2022 (UTC)
• In general, I think that we should only block people on the basis of on-wiki evidence. With few exceptions, whatever happens off-wiki should be irrelevant. The first exception that springs to mind if off-wiki harassment, but there are also other cases, such as UPE etc. However, those should remain just that: exceptions. After all, opposition research is strongly discouraged even when it is based on information that may be publicly available on-wiki (see WP:OUTING and Wikipedia:Harassment#How to deal with personal information). This policy change, in my opinion, normalises blocks based on off-wiki evidence too much, for my tastes, and risks encouraging opposition research. In addition, it runs counter to two fundamental principles of the project: transparency and the idea that all editors are equal regardless of their user rights. In fact, when an editor is blocked on the basis of off-wiki evidence that is only made available to administrators, we are preventing non-administrators from evaluating whether the block was appropriate, without a good reason. We should not forget that administrators are not really qualified to handle non-public information, since they have not signed the WMF confidentiality agreement and have not been vetted for those responsibilities. On the other hand functionaries and arbitrators have been vetted and have signed the confidentiality agreement and policy already recognises that they are qualified to make blocks based on non-public information. I don't think the policy should be changed, I don't find the change necessary or wise. If it seems complicated to have someone blocked on the basis of off-wiki evidence, in my opinion it's because that's how it should be. Salvio 13:34, 7 September 2022 (UTC)

  @Salvio giuliano: I think it's a valid outcome of this RfC for the community to clarify that it is unacceptable for admins to make many blocks they already make: I know administrators other than Tamzin make those blocks frequently. If the community does so, I would like it to speak clearly – as I wrote above, Also, because there’s disagreement about what policy currently requires, I’d ask any folks who oppose this proposal to indicate what they think the current policy says. (Are blocks based on info with an “email me for the evidence” note permissible? Does that count as information that will [] be made available to all administrators?) If the community says "no", we ought to update BLOCKEVIDENCE to say so. Best, KevinL (aka L235 · t · c) 15:49, 7 September 2022 (UTC)

• Wrong question. Before we figure out how to handle the record keeping for non-CU blocks involving off-wiki evidence, we should figure out if we even want that to be happening at all. If we're going to officially condone people poking around in LinkedIn, UpWork, etc, looking for evidence for blocks, then we need a wholesale rewrite of WP:OUTING. I know it goes on, so I guess it's good that this RfC got started because it brings the practice out in the open for discussion. Only after we figure out if we want admins (or anybody) doing that, then it's time to figure out how we want to handle the record keeping. -- RoySmith (talk) 13:41, 7 September 2022 (UTC)
• I am a bit surprised to see so many editors suggest that combatting professional, some of it very sophisticated, attempts to manipulate our content through paid editing is not a top flight concern. Our social policies are not a suicide pact and OUTING is a social policy. It is designed to protect good faith and bad faith editors alike. But that doesn't mean we should say to those administrators who wish to stop bad faith editors "oh sorry, you're not fit to do so until you get Checkuser and/or Oversight". L235 makes a crucial point that in actual practice these efforts seem to be accepted. What we are talking about here are efforts to ensure our readers are met with content that matches core policies and pillars like "Wikipedia is written from a neutral point of view". I'm a bit ambivalent on this proposal on the whole - hence why I'm not supporting - but I actually expected to be here with some of my concerns (some of which are capture above) rather than with what I see as the merits of this proposal. Best, Barkeep49 (talk) 14:39, 7 September 2022 (UTC)
• Support and question why anything easily google-able is being considered private information for purposes of OUTING. If any editor can enter the article subject, connected parties or editors handle into a search engine and find the connection, it's not private. UPE is only getting worse and we need every tool available to shut them down quickly. Slywriter (talk) 14:49, 7 September 2022 (UTC)
• I'm a bit confused by the proposal and several support and oppose votes above; there seems to be a pretty fractured understanding of outing vs. pursuing UPE cases vs. disruptive offwiki behavior vs. public and non-public information vs. NDAs vs. etc. above. My main question, though, is the text at Wikipedia:Administrators#Special_situations that appears to approve these sort of blocks and contradicts BLOCKEVIDENCE per my reasoning here going to be changed as a result of this? Or will this just create more policy headaches? Or does this proposal and the text and ADMIN actually align in some way I'm not noticing? Moneytrees🏝️(Talk) 14:57, 7 September 2022 (UTC)