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, 177, 178, 179, 180, 181, 182, 183, 184, 185, 186, 187, 188, 189, 190, 191, 192, 193, 194, 195, 196, 197

Unnecessary delay in publishing articles translated for $$ by an NGO

So, I just stumbled upon Wikipedia:WikiProject Intertranswiki/OKA. TL;DR, there is an NGO sponsoring translating high quality articles between Wikipedias. But on EN due to our COI/PAID policies they are required to use AfC, which means that their articles, which usually are very good, are delayed through AfC backlog, to which they also contribute. I think this is an excellent initative that however needlessly clutters AfC due to our current rules, and I'd like to suggest we consider giving it exception from the COI requirement to use AfC. It makes sense to direct paid-for spammers to AfC, as their articles are often problematic (notability, etc.) but what we have here is very different (translations of good quality articles from other wikis - ex. current drafts include Draft:Renaissance in Ferrara, Draft:Spa Conference (2-3 July 1918), Draft:Formal procedure law in Switzerland, etc.), yet this stuff is caught in the same "COI" net. (See project page linked above for links of articles already published, links to drafts waiting for review, and their instructions to translators) Thoughts? (Courstesy ping project founder @7804j). PS. A question to 7804j - how are articles chosen for translation? How is the system designed not to be abused by spammers? Perhaps if an exception is granted on en wiki, it should not apply to articles about companies, products or living persons? _{Piotr Konieczny aka Prokonsul Piotrus| reply here} 02:27, 22 June 2024 (UTC)

I would dispute that "this is an excellent initative" or "that their articles, which usually are very good". They have caused a lot of work; mostly these are machine translations by people whose English is rather poor. The titles chosen are often completely ungrammatical (Greek Classicism Sculpture was a typical one) or inappropriate, & in the past they have chosen often subjects we already have. The texts are just whatever the language taken - usually Portuguese, Spanish, French or Italian, has on their wiki, & the quality of the original is often poor, & errors introduced by machine translation go uncorrrected. There have been numerous complaints. They have got slightly better, but I think still don't publish a full list of articles they have paid for, whicgh they should. The Open Knowledge Association isn't really "an NGO" - as far as I can see it's a single Swiss guy with a bit of money to spend, who you have rashly decided to endorse. Johnbod (talk) 02:51, 22 June 2024 (UTC)

I think that the principle is sound: high-quality articles can and should be translated into languages where they're missing. Doc James ran a similar program for certain medical articles a few years ago (e.g., during the Ebola and Zika outbreaks), to public acclaim. However, he was working with pre-screened professional translators, and OKA seems to have struggled with quality control. WhatamIdoing (talk) 04:19, 22 June 2024 (UTC)

Unfortunately the ODA model makes absolutely no attempt at quality control. As will be clear to anyone who reads one of them, they are just machine translations dumped onto en:wp with no aftercare. Many that were forks were just turned into redirects, which the ODA doesn't appear to have noticed. The ones that are left take a lot of cleaning up, when some regular editor can be bothered. Johnbod (talk) 01:52, 23 June 2024 (UTC)

I am afraid that your anecdotal analysis above is different from mine. The articles from OKA I've seen seem pretty decent, at start+ class, and would survive AfD if nominated. Can you recall which articles were redirected - and prove that they are a rule, and not an exception? _{Piotr Konieczny aka Prokonsul Piotrus| reply here} 02:50, 23 June 2024 (UTC)

Whether they would survive Afd is almost all about the notability of the subject, and that is not usually an issue - the quality is. In fact the worst issues arise when they tackle very prominent subjects. I never claimed that redirected ones were the "rule" - I make no attempt to search out OKA efforts, but then clearly neither do you. Draft:Crow-stepped gable is a recent creation, objected to, for which we have a redirect already in place. Not much of it will survive, I'd imagine. If they kept proper lists of their articles on wiki I would be able to find some, I imagine. Johnbod (talk) 03:12, 24 June 2024 (UTC)

Johnbod List here; may not be everything. Mathglot (talk) 03:34, 24 June 2024 (UTC)

Thanks, but I don't think that is at all complete. The template was only set up in October 22 (by 7804j), well into OKA's project. Stuff may have been added later. You used to able to access an off-wiki spreadsheet 7804j maintained, but I can't see that you can now. User:7804j? For example, the earlier efforts of User:Racnela21, one of the most prolific OKA editors, are not templated - see the 48k bytes of Brazilian Romantic painting (typically, initially called Brazilian Romanticism Painting). Johnbod (talk) 13:08, 24 June 2024 (UTC)

This list contains all articles created by OKA after the template was created. Oka was created relatively shortly before the template was created, therefore there are not many articles without it (probably 90+% have the template). The off wiki tracker is still at oka.wiki/tracker 7804j (talk) 13:24, 24 June 2024 (UTC)

Also, I'd like to highlight that quality is not really the topic of this discussion, since this is about whether COI should require all paid editors to go through AfC and, as you pointed out yourself, AfC's goals are not primarily to check quality. I'd suggest moving the OKA discussions somewhere else such as our talkpage in the intertranswiki project 7804j (talk) 13:27, 24 June 2024 (UTC)

Piotr brings up "would survive AfD" because that's the standard AfC uses. If OKA articles typically have quality issues that wouldn't be enough for deletion, then there's no point insisting they go through AfC – assuming reviewers are doing their job properly, they'll just send them right through. – Joe (talk) 11:00, 24 June 2024 (UTC)

Things that would make them fail Afd include repeating articles we already have under a different title, a perennial problem with OKA, which reviewers don't always pick up, but sometimes do - as currently at Draft:Crow-stepped gable. Besides, some reviewers (perhaps not "doing their job properly" - how shocking) insist on minimal standards of coherent English, etc. Johnbod (talk) 14:31, 24 June 2024 (UTC)

Health translation efforts from English to other languages are still running. https://mdwiki.toolforge.org/Translation_Dashboard/leaderboard.php Our translators are mostly volunteers with a mix of Wikipedians and professional translators. Doc James (talk · contribs · email) 12:40, 23 June 2024 (UTC)

Hi Piotrus,

Thanks for initiating that discussion! I am fully supportive of such an exemption, as I see this AfC requirement as additional red tape that consumes a lot of time for OKA translators and AfC reviewers.

Our core principle is that our translators are free to work on anything that interests them. We provide them with a monthly stipend, some training on how Wikipedia works, but we then see them as volunteer contributors on whom we impose some process to ensure they do not abuse the grant and provide overall value (eg, quality checks, quantity checks). To help them find articles to translate, we curate an optional backlog (at oka.wiki/tracker). Articles of this tracker primarily consist of "Featured" and "Good" quality articles from other Wikis, as well as red links from these articles. We also complement this with articles that we find important, eg, about geographical features such as lakes, mountains, etc. The broader principles for articles prioritization are described at oka.wiki/overview

Note that there was a similar discussion in the Interwiki talkpage, which can provide useful additional context.

Regarding Johnbod's response, I would like to bring 3 points of context:

1) While overall quality is good, it may vary. Because we have many different translators, with difference levels of experience, the quality will not be uniform. We are providing them with training, and we have observed their quality improved over time. We stop providing grants to translators wjth recurring quality issues. Overall, I do not agree with Johnbod's characterizarion of a high degree of quality issues. Often, the issues raised with OKA's work were not due to the quality of the translation, but because of the source article itself. We have published several thousand of articles, most of which are still live with very minimal change vs their original published version.

2) This discussion is not about assessing the quality of the work, but whether the COI requirement to go through AfC should apply to OKA. The only reason why our translators go through AfC today is because of the COI policy, which was not created primarily to check quality of paid translations but to eliminate bias. Therefore, I don't think such arguments are appropriate in the current discussion.

3) Our funding comes from many different private individuals, but it is true that currently I am the main donor. That being said, this should not make any difference as to whether we can be called an "NGO". Would the Gates Foundation not be called an NGO just because most of its funding comes from Bill Gates? We have over 15 full time translators who agree to do this work with a very small stipend, much smaller than what they could earn in a regular job, so the work of OKA is much more than that of a single person 7804j (talk) 08:46, 22 June 2024 (UTC)

Personally, I don't care how high quality the articles end up being, if you have a financial tie to a subject you should go through AfC. Lee Vilenski ^{(talk • contribs)} 08:59, 22 June 2024 (UTC)

Getting paid to translate an article about Brazilian Romantic painting (popular in the late 1800s) is not exactly the same as having a financial tie to the subject. WhatamIdoing (talk) 04:32, 28 June 2024 (UTC)

I would prefer not to couch any action in terms of "an exception" for a named user or group. Rather, I would prefer to see an adjustment to WP:PAID to make a modification to allow "philanthropic paid editing" where the articles in question and the content added are chosen by the paid editors and there is no oversight by the payer. At that point, individual articles and editors would be subject to the same kind of oversight as any other. It seems to me that philanthropic paid editing to expand the encyclopedia is within the scope of WP:HERE, and this should not be formulated as an "exception" as if something were wrong with it in the general case. Mathglot (talk) 09:14, 22 June 2024 (UTC)

I agree with [[U|Lee Vilenski}} if you have a financial tie to a subject you should go through AfC, The given example Draft:Renaissance in Ferrara is very poorly translated. Theroadislong (talk) 09:19, 22 June 2024 (UTC)

Courtesy ping: Lee Vilenski. Mathglot (talk) 09:22, 22 June 2024 (UTC)

But that's the thing, OKA editors don't have a financial tie to the subject. They're paid by an organisation to edit Wikipedia, but the selection of topics is independent. It's basically paid editing without a COI, which is a bit of blind spot in our current policies. – Joe (talk) 09:30, 22 June 2024 (UTC)

Indeed. What "tie to the subject" is there in "Renaissance in Ferrara"? We might as well call COI and PAID for Wikipedia:School and university projects or most of WP:GLAM stuff, and various edit-a-thons, since there is $ involved in it as well. Do we require AfC from Wikipedians in Residences? _{Piotr Konieczny aka Prokonsul Piotrus| reply here} 10:48, 22 June 2024 (UTC)

Actually I would be interested to understand what are the requirements for projects such as the ones you mentioned to *not* qualify as paid editing. As you pointed out, Wikipedians in Residence do not need to go through AfC -- what are the formal criteria/policy allowing them to be compensated without being considered paid editors? 7804j (talk) 20:17, 23 June 2024 (UTC)

As per foundation:Policy:Terms of Use/Frequently asked questions on paid contributions without disclosure#How does this provision affect teachers, professors, and employees of galleries, libraries, archives, and museums ("GLAM")?, Wikipedians-in-residence are still considered paid editors for contributions for which they are being paid. isaacl (talk) 22:30, 23 June 2024 (UTC)

@Isaacl:, yes, but as I read it, they are free to make edits of their choice without even disclosing their paid status, as long as they are not making specific edits about the payer institution. The way I read it, is that GLAM employees do not need to disclose because: "Disclosure is only necessary where compensation has been promised or received in exchange for a particular contribution". That section recommends a simple disclosure for W-in-residence, but only in the case where they are "specifically compensated to edit the article about the archive at which they are employed". Paid status need not be disclosed for general edits unrelated to that. Do you see it differently? Mathglot (talk) 02:20, 24 June 2024 (UTC)

Yes, I do, and so has previous discussion at Wikipedia talk:Paid-contribution disclosure. If they are being compensated for a particular contribution, as per the section you quoted, then they fit the definition of a paid editor. :foundation:Policy:Terms of Use#Paid Contributions Without Disclosure does not distinguish reasons for the paid contributions. isaacl (talk) 06:14, 24 June 2024 (UTC)

Yes they do fit it if compensated for a *particular contribution*, and the Paid FAQ linked by the foundation Policy you cited above specifically calls out the circumstances when paid editors do *not* need to disclose their contributions. Those circumstances match those of paid OKA volunteers, who, had they been a Wikipedia-in-residence or a GLAM-paid instead of OKA-paid, would not have had to disclose their status, according to the wmf policy FAQ itself. Mathglot (talk) 06:39, 24 June 2024 (UTC)

On the English wikipedia we do require that disclosure "If you receive, or expect to receive, compensation for your contributions to Wikipedia, you must disclose who is paying you to edit (your "employer"), who the client is, and any other relevant role or relationship." Even if the foundation FAQ says that per the foundation they don't per English wikipedia they do. Horse Eye's Back (talk) 06:44, 24 June 2024 (UTC)

The FAQ is giving specific examples, and is non-exhaustive. As explained in the first paragraph of the section, you are only required to comply with the disclosure provision when you are compensated by your employer or by a client specifically for edits and uploads to a Wikimedia project. This is in accordance with the actual Terms of Use: if you are being specifically compensated for contributions, you are a paid editor, but this does not extend to your contributions that are not within the scope of your compensation. If you are being paid to edit about your employer, that's within the scope of your compensation, and so the relationship has to be disclosed (and the example is about this specific situation). isaacl (talk) 13:24, 24 June 2024 (UTC)

So in the same line of thought, this means that all articles created by Wikipedians in Residence in the context of the organization that pays them need to go through AfC (as @Horse Eye's Back suggests in the comment below), is that also your understanding? 7804j (talk) 16:21, 24 June 2024 (UTC)

Note that "Wikipedian-in-residence" is just a self-described title, without any oversight from anyone involved with the WMF or Wikipedia, so the scope of their role is entirely decided by their employer and them. Some of those who have participated at Wikipedia talk:Paid-contribution disclosure have said that they do not edit Wikipedia as part of their role; they provide education and support to the institution's staff. isaacl (talk) 16:56, 24 June 2024 (UTC)

"Do we require AfC from Wikipedians in Residences?" The outcome of the recent case involving the BYU library's Wikipedians in Residence clarified that the community does in fact expect Wikipedians in Residence to use AfC. Horse Eye's Back (talk) 03:54, 24 June 2024 (UTC)

@Mathglot "philanthropic paid editing". I like the term - hope it makes it into our updated policies. _{Piotr Konieczny aka Prokonsul Piotrus| reply here} 10:51, 22 June 2024 (UTC)

This is one reason I prefer the term financial conflict of interest. "Paid editing" focuses on a transaction—being paid to edit—but the real issue is the tendency to bias created by some financial relationships. Wikipedians in Residence are the paradigmatic example of people who are literally paid to edit but don't have a conflict of interest; it seems like OKA translators are another. If we shifted the guideline to talk about FCOIs instead of paid editing, the need for an exception for philanthropy would disappear. – Joe (talk) 11:24, 22 June 2024 (UTC)

Hear, hear. There is nothing inherently wrong with folks making $$ out of volunteering. _{Piotr Konieczny aka Prokonsul Piotrus| reply here} 14:17, 22 June 2024 (UTC)

By definition you can't make money out of volunteering, if they're making money they're working not volunteering. Horse Eye's Back (talk) 05:39, 24 June 2024 (UTC)

If you can make xxx$ out of a full tims job and only half of that when editing Wikipedia, it becomes more a hybrid role than pure full time job. Our translators usually give up much better paid opportunities for being able to work on Wikipedia. 7804j (talk) 06:10, 2 July 2024 (UTC)

7804j, I would not pursue this line; it's a distraction, and a loser. Volunteering/working is binary, there is no hybrid, in-between, or threshold of payment so low that it is not "working". Mathglot (talk) 06:18, 2 July 2024 (UTC)

In the context of Wikipedia I agree with you that there should be no distinction in the policy. I just wanted to call out that many of these paid editors do so not because they are interested financially but because they care about Wikipedia and just need some money to pay rent and food (thus why we call it a grant/stipend). Sometimes people are being overly harsh on them, so I think it's important to highlight they also do some personal sacrifices to do that job. 7804j (talk) 06:22, 2 July 2024 (UTC)

Indeed. And WiRs get paid stipends and such, and we still consider them volunteers, no? _{Piotr Konieczny aka Prokonsul Piotrus| reply here} 08:09, 2 July 2024 (UTC)

We consider WiR and such to be paid editors if they are paid (there are volunteer WiR who don't get any compensation). Horse Eye's Back (talk) 15:32, 4 July 2024 (UTC)

You seem to be making the distinction between working full time and working part time, not between working and volunteering. Horse Eye's Back (talk) 15:32, 4 July 2024 (UTC)

No, I am making the distinction between working full time in a for-profit translation company that pays well, and working full-time through stipends from a non-profit organization like OKA that pays a lot less. OKA editors accept a much lower grant than what they could earn elsewhere because they know it's an important cause. 7804j (talk) 15:41, 4 July 2024 (UTC)

And what is the distinction? Neither of those is a hybrid situation or volunteering... Taking a lower salary to work in a job you want to work in vs one which pays more but you don't want to do is not volunteering, almost all of us do that. Horse Eye's Back (talk) 15:54, 4 July 2024 (UTC)

Wikipedians in Residence all have signficant conflicts of interest, primarily in relation to their employer. Horse Eye's Back (talk) 05:43, 24 June 2024 (UTC)

Everyone has significant conflicts of interest, primarily in relation to their employers. The issue is whether they make edits in those areas or not. If a WiR at the Museum of Nowheresville was editing Museum of Nowheresville, there'd be a problem. If an OKA translator was editing Open Knowledge Association, there'd be a problem. But that's not what we're talking about here. – Joe (talk) 10:49, 24 June 2024 (UTC)

Not able to square "Wikipedians in Residence are the paradigmatic example of people who are literally paid to edit but don't have a conflict of interest" with "Everyone has significant conflicts of interest, primarily in relation to their employers" Horse Eye's Back (talk) 07:05, 27 June 2024 (UTC)

So OKA has been on my radar for some years now due to off-wiki reports sent to the paid editing queue. I was extremely suspicious of it at first and (along with others active in UPE patrolling) worried it would be a sort of front for the usual abusive paid editing. However, I have to hold my hands up and say that it's been c. five years and nothing like that has come up. From what I've seen, the selection of topics is genuinely made based on what's missing on enwiki, and the quality of the translation are at least no worse than average. @7804j: You perhaps made an initial strategic error in structuring/talking about this as "freelancers" doing "paid editing", because this puts you in a category of people that the volunteer community, for good reason, have come to be very sceptical of. Essentially identical activities that are framed as grant-making or residency do not raise the same eyebrows, especially if you can get some sort of buy-in from the WMF (which is not hard).

Quality is a separate issue and something that pretty much always causes friction when people who aren't very familiar with Wikipedia are incentivised to contribute to it en masse. There is no easy to solution to this. Specifically, making them go through AfC isn't going to help – AfC reviewers don't have the time to do a close reading of drafts to look for translation issues. They'll take a look through for major problems (which OKA drafts don't seem to have) and for notability (virtually guaranteed because these are substantial articles on other Wikipedias) and then pass it through. So we'll end up with the same outcome as if they were created in mainspace directly, just with some extra volunteer time wasted within an already backlogged process.

As to whether OKA creations need to go through AfC, I am usually the last person to point this out, but technically this is a request not a requirement. AfC is broken by design because generally we don't want to encourage paid editors by giving them an efficient route to publication, or encourage volunteers to do work that someone else will get paid for. As Mathglot says, Neither our COI policy or the AfC process was designed with 'philanthropic paid editing' in mind. I think it's fine for OKA editors to bypass this and create directly in mainspace. This isn't an exception our a change to the rules, it's just applying WP:IAR and recognising that forcing good faith creations into a broken process because their creator got a stipend while writing them, or because they might have some translation issues, is not in the spirit of WP:FCOI. – Joe (talk) 09:25, 22 June 2024 (UTC)

@Joe Roe "extra volunteer time wasted" - exactly, this is the problem I am trying to address. _{Piotr Konieczny aka Prokonsul Piotrus| reply here} 11:12, 22 June 2024 (UTC)

Thanks @Joe Roe!

Initially, I also thought that the AfC requirement for paid editors was a request and not a requirement. However, @Seraphimblade raised in my talk page that any OKA editor creating an article in the mainspace without going through AfC would be blocked. Hence why we started requiring all our translators to go through AfC since early May.

I agree with you that it was a mistake from my end to have initially used the term "freelancer". Our translators are volunteers receiving a grant to cover basic costs of living (~400 usd per month for the ones working full time). Going forward, I will make sure to always use the more accurate terms of "Grant/stipend recipients". I did not want to use the term of "Wikipedians in Residence" as it seemed to me that this requires that the work be related to the institution itself. I wasn't aware that there are options to get buy-in from the Wikimedia foundation, but I will explore this avenue as it will indeed help with acceptance of OKA among the community.

In general, I strongly with the idea of introducing a broader exemption to the AfC requirement of the COI policy to either philanthropic institutions that do not target specific topics and give high degree of freedom to grant recipients, or to payments that are too low to represent full wages (e.g., <xxx$ per month/ per hour).

7804j (talk) 11:54, 22 June 2024 (UTC)

Specifically you might want to look into meta:Wikimedia thematic organizations or one of the other categories of meta:Wikimedia movement affiliates. – Joe (talk) 12:06, 22 June 2024 (UTC)

Whatever avenues you explore, I would not get into proposals related to trying to find a threshold where a payment is "too low" to make a difference, and thus presumably not trigger a PAID concern. Experience with paid crowd-sourcing platforms such as MTurk shows that micropayments may attract volunteers for certain tasks, even sometimes for a larger than average task such as a translation. Mathglot (talk) 18:04, 22 June 2024 (UTC)

This might be a dumb question, but I'm tired and can't find it: where in the policies do we require paid editors to use AFC? (please do not ping on reply) Primefac (talk) 22:05, 22 June 2024 (UTC)

WP:COIEDIT states that paid editors "should put new articles through the Articles for Creation (AfC) process instead of creating them directly". _{Piotr Konieczny aka Prokonsul Piotrus| reply here} 02:52, 23 June 2024 (UTC)

I see. Primefac (talk) 12:38, 23 June 2024 (UTC)

Ah, so here's this month's OKA thread, I thought I'd miss it!

If an organization of this sentiment really wanted to help the English Wikipedia, they would be working exclusively on poorly developed vital articles. Then there would be no AFC necessary. The English WP is far past the point where creating new articles is an effective way to make meaningful improvements. Unless, of course, this creation targets areas of systemic bias where there is a genuine dearth in coverage.

To me this appears much like the organizers have gone so far in one direction that whether or not their effort is actually worthwhile is no longer a consideration. Even with their current infrastructure, it would be considerably more effective to take EN FAs and translate them into other languages. Aza24 (talk) 07:29, 23 June 2024 (UTC)

You've created 68 articles, the last one two weeks ago. Are we to understand that that was the last one we needed? – Joe (talk) 11:22, 23 June 2024 (UTC)

Halleluyah, we are done! _{Piotr Konieczny aka Prokonsul Piotrus| reply here} 13:27, 23 June 2024 (UTC)

The English Wikipedia does not need new articles nearly as much as it needs improvements on existing ones. As I said, the only exception is to fill systemic bias gaps, which yes, includes a woman poet! Comparing a single editor with an entire organization does not track.

Unfortunately, the OKA is fundamentally flawed in this regard, but it doesn’t seem like an object of concern for them. Aza24 (talk) 17:09, 23 June 2024 (UTC)

I should add that if I'm being overly critical, it's because this organization should be held to a high standard. Sine it is under the guise of effective altruism, the former "effective" qualifier needs to take more prominence. I can't see anywhere that it's even been considered how to most effectively help Wikipedia. Otherwise, the OKA would have approached the community before founding, to identify what is actually needed. Since they didn't, now we find ourselves in these same threads, time and time again. Aza24 (talk) 17:35, 23 June 2024 (UTC)

Your argument appears to be about your opinion on how work on Wikipedia ought to be prioritized, and is a red herring. One of the central features of a volunteer organization, is that volunteers work on articles of their choice, not articles of your choice, or some committee's choice. Thank goodness I didn't have to listen to you, or I never would have had the opportunity to translate that article about a medieval Catalan peasant uprising, when there were no doubt many hundreds of thousands of tasks more urgent than that one at the time. The OKA volunteers who translate articles of their choice in their own manner should be held to the same standard I was, namely, Wikipedia policies and guidelines, and nothing else. Mathglot (talk) 19:14, 23 June 2024 (UTC)

Thank goodness I don't have to listen to you either! Aza24 (talk) 19:45, 23 June 2024 (UTC)

@Aza24 I do not think this is the right place to discuss this. This thread is about whether to make changes to the AfC requirement of COI, not about how OKA prioritizes articles. So I would suggest moving that discussion for example to the OKA taskforce talkpage.

That being said, we (OKA) already operate along the lines of what you seem to recommend. Many if the articles our translators work are are about neglected topics in EN wiki, for example, articles about geographical features of non-English speaking countries (eg, Spain, Latin America) or non-English speaking historical figures. I would actually argue that improving coverage on these topics is much more important than extending already extensive articles on important topics. But most importantly, it takes different skill sets to translate vs expand articles. The editors who receive our grants would not necessarily be sufficiently familiar with these topics to be able to expand them starting from scratch.

Regarding your recommendation to translate from English to other languages: we do that already. We published thousands of articles in the Spanish and Portuguese Wikipedia, with a strong focus on under represented topics in these Wikipedia such as mathematics, computer science, etc. There's been a lot of off Wiki analysis of opportunities to maximize impact on donation that went on before we decided to set up OKA the way it is, and I'm happy to share more detail about the rationale if there is interest 7804j (talk) 19:41, 23 June 2024 (UTC)

I'm going to retract my comments. Given your response, I don't think I'm nearly as informed as I should be on the organization to be casting such aspirations/critiscms. Also, my comments seemed needly inflammatory; my apologies. – Aza24 (talk) 20:54, 23 June 2024 (UTC)

@Aza24 I just wanted to say that it is quite rare to see folks backtrack and even apologize in Internet discussions (and that includes on Wikipedia). Regardless of the issue at hand, I would like to say I very much respect and appreciate you for what you have just said above. Cheers, _{Piotr Konieczny aka Prokonsul Piotrus| reply here} 02:47, 24 June 2024 (UTC)

How did feline hyperthyroidism come up then? Traumnovelle (talk) 09:23, 5 July 2024 (UTC)

I see a nescessary delay, there is no rush and that absolutely needs to be treated the same way as other paid edits. Horse Eye's Back (talk) 16:49, 23 June 2024 (UTC)

I agree that paid editing is fishy due to the presence of inherently non-encyclopedic motivation, which may ultimately lead to poor quality translations of selection of poorly referenced source articles. As I see, OKA is fairly new and it is probably not flooded with quick buck seekers, but things may quickly change when rumors spread on how to earn some extra easy cash off google translator. I took a quick look at OKA articles submitted in AfC and all my random picks seem to have good quality. So here is my suggestion: How about vetting decent contributors to bypass AfC? - Altenmann >talk 19:19, 23 June 2024 (UTC)

I could see creating some sort of “fast-track” for reviewing these articles, but some sort of review is still necessary. If for no other reason than preventing duplication of topic with existing articles. Blueboar (talk) 19:51, 23 June 2024 (UTC)

I could get behind a separate lane so to speak, I just really dislike the idea of creating a loophole. Horse Eye's Back (talk) 19:58, 23 June 2024 (UTC)

HEB, Can you expand on what you mean by the idea of "a separate lane"? I wouldn't favor a change that referred to OKA by name (except at best in an explanatory note as an illustration of a general point in line that requires an example). Plenty of generalized guidelines have logical carve-outs that need to be explicit, for example, the guidance that strongly discourages external links in the body of an article specifically states that it doesn't apply to inline citations. We could follow that approach.

But there may be even a better way to deal with this. Currently, the first line of WP:FCOI says this:

Being paid to contribute to Wikipedia is one form of financial COI; it places the paid editor in a conflict between their employer's goals and Wikipedia's goals.

In my view, this is the crux of the problem, because it *assumes* that an employer's goals are in conflict with Wikipedia's goals. But what if that is a false assumption? I believe the general problem we are addressing could be handled without any specific carve-out, by altering it as follows:

Being paid to contribute to Wikipedia is one form of financial COI; it places the paid editor in a conflict when their employer's goals and Wikipedia's goals differ.

If the goals of an organization do not differ from Wikipedia's goals, then no separate lane or carve-out is required elsewhwere. This somewhat leaves open the question of what we would define as Wikipedia's goals, but Wikipedia:Purpose (info page) says this:

Wikipedia's purpose is to benefit readers by acting as a widely accessible and free encyclopedia; a comprehensive written compendium that contains information on all branches of knowledge. ...

The goal of a Wikipedia article is to present a neutrally written summary of existing mainstream knowledge in a fair and accurate manner with a straightforward, "just-the-facts style".

If a philanthropic organization's goals are the same as Wikipedia's, and there is no organizational oversight of payees' output, then it seems to me no special lane is required. (edit conflict) Mathglot (talk) 20:35, 23 June 2024 (UTC)

The practical question is who's going to decide which edits do or do not need independent review? If in practice this can only be done on an article-by-article basis, then I don't think much is gained by setting up a new decision branch that comes before using the articles for creation process. isaacl (talk) 22:39, 23 June 2024 (UTC)

The lane or whatever isn't me idea so I don't want to speculate on it, in general I think what we have now works. In terms of the hypothetical unless they themselves are wikipedia how can their goals be the same as Wikipedia's? Generally organizations have self promotion as a goal and that is forbidden per WP:PROMOTION. Horse Eye's Back (talk) 05:52, 24 June 2024 (UTC)

The organisation's goals may be the same but the individual's goal may be to try and make as much as money as quickly as they can which can lead to machine translations + quality issues, which I've notice in the one OKA article I came across. Traumnovelle (talk) 09:25, 5 July 2024 (UTC)

That could be a problem if the payment model is Piece work, but it's unlikely to be a problem with a set monthly stipend. WhatamIdoing (talk) 19:20, 5 July 2024 (UTC)

The New Page Patrol process should already cover most of the review requirements, no? 7804j (talk) 20:11, 23 June 2024 (UTC)

Question: do we actually have some specific consensus that these uniformly awful translations should in fact be submitted through AfC? That would be such a good thing! Every one of them I've seen so far (mostly relating to horses) has been created directly in mainspace, and requires an amount of clean-up that seems to be far beyond the editor resources we have – with the result that overall this project is making the encyclopaedia worse, not better. I've asked myself several times why these pages were not being submitted as drafts, but not until now seen any discussion of them; if there's an standing consensus that they should go through AfC, I'll be draftifying several of them in the near future. Sorry, but oppose any kind of AfC exemption for the moment. Justlettersandnumbers (talk) 20:42, 23 June 2024 (UTC)

Justlettersandnumbers, First: imho, you should draftify them regardless, if they are not ready for mainspace, not because there is or isn't some guideline stating that they should all go through Afc. Secondly, do you draw a distinction between awful translations produced by paid translators and awful translations produced by unpaid translators that go straignt into mainspace, and if so, what criteria should be used for each? Granted, the former are easier to find due to categorization. Mathglot (talk) 20:52, 23 June 2024 (UTC)

• I think enough concerns have been raised about poor translations here that the argument to skip the AFC process is quite weak. I will also add that unedited machine translations are an extreme drain on experienced editor time, resulting in diffs like this one from 2021. If unedited machine translations are occurring here, this could turn into a big problem and big cleanup effort, and once sufficient evidence is gathered, we should attempt to communicate these concerns to the event organizers. –Novem Linguae (talk) 02:19, 24 June 2024 (UTC)

  I've seen no evidence that OKA translators are creating unedited machine translations. – Joe (talk) 10:55, 24 June 2024 (UTC)

  Sounds like @Johnbod (mostly these are machine translations by people whose English is rather poor) and @Theroadislong (Has this been machine translated? There seems to be a lot of mangled content here? in Draft:Renaissance in Ferrara) might disagree. –Novem Linguae (talk) 16:13, 24 June 2024 (UTC)

  Indeed - 7804j has never denied that these are machine translations, and they normally appear on en:wp in a single edit, & are not edited further except for a couple of tidies. There is no evidence that they are edited machine translations when OKA bow out, and they should be treated as "unedited machine translations" - what other evidence of absence would there actually be? Other volunteers are left to do things like categories and links, which they normally lack. Very rarely does anyone do the complete rewrite that ones like Draft:Renaissance in Ferrara need just to be comprehensible to an average English reader. To anyone who think OKA texts are "generally good" or "decent translations" I would say: just try actually reading that one - which btw will probably get far more views than most OKA efforts, as there is a real topic there. It covers our existing School of Ferrara but that is so crap I don't object on WP:FORK grounds, though it is typical that OKA haven't addressed this. Johnbod (talk) 15:11, 25 June 2024 (UTC)

  I think you're applying a really high standard here. For example, the original title of Brazilian Romanticism Painting, and yes of course that's not perfect English, but does it impair the reader's ability to understand that the article is about Romanticism in Brazilian art? No. I see the same kind of thing reading through the rest of the article and other OKA articles: uneven English, yes, but perfectly comprehensible and, more importantly, sourced encyclopaedic content. The rest will be ironed out with time, like how you corrected the title of Brazilian Romantic painting a couple of weeks after it was created.

  It's actually quite easy to verify whether a machine translation has been edited or not: just run the original through the same translator. For example, here's how DeepL handles the first paragraph of the first section:

  The Este court in Ferrara was one of the most vital in northern Italy from the end of the 14th century, when Niccolò d'Este started the university and initiated the construction of the castle[1]. The courtly connotations were pronounced, as evidenced by the interest in the world of fairy tales of medieval heritage, as evidenced by the numerous novels of chivalry that enriched the famous library, in astrology and esotericism[2]. On an artistic level, Pisanello, who produced various medals for Lionello d'Este, was highly appreciated, as was the illuminated production, both of an international nature, in which Belbello da Pavia (author of the Bible of Niccolò d'Este) stood out, and updated to humanism, such as that of Taddeo Crivelli (Bible of Borso d'Este)[2].

  Compare that to the draft:

  The court of the Este in Ferrara was one of the most vital in northern Italy since the late 14th century, when Niccolò d'Este funded the University of Ferrara and started the construction of the Castello Estense. His courtly features were prominent, as evidenced by his interests in the fable world of medieval heritage, astrology and esotericism. On the artistic level, Pisanello, who produced several medals for Lionello d'Este, was highly regarded, as was the illuminated production of both international in which Belbello da Pavia (author of the Bible of Niccolò d'Este) stood out, as well as update to humanism, such as that of Taddeo Crivelli (author of the Bible of Borso d'Este).

  Again, it's not perfect, but it's not somebody just acting as a conduit for automated translations, which is what the practice of draftifying these is supposed to filter out. OKA editors are using a machine translation as a base and then proofreading it, which in my experience is what practically everyone that works in more than one language does these days. – Joe (talk) 15:41, 25 June 2024 (UTC)

  Not sure what you think this demonstrates. It could be that they used a different translator. If you are suggesting they used the same one, then manually touched it up, the effect of their changes has on the whole made things worse, no? To someone who doesn't know the area, both versions of the passage are basicly gibberish in the details. To bring either up to even mediocre WP standards, a total rewording is needed. This is typical (ok, this example, which Piotrus selected, is worse than most of theirs these days). Johnbod (talk) 15:02, 28 June 2024 (UTC)

  Including estabilished and experienced editors like myself. (I machine translate and proofread my own articles between en and pl, for example). Nothing wrong with using MT as long as one knows how to proofread stuff (and if the original article of course is of decent starting quality to begin with). _{Piotr Konieczny aka Prokonsul Piotrus| reply here} 02:53, 26 June 2024 (UTC)

  See [1]

  Deepl translate of the German lead gives me: Feline hyperthyroidism is a disorder of the endocrine system in domestic cats (feline, adjective from the Latin felis "cat"), which is characterised by hyperthyroidism. It is the most common hormonal disorder (endocrinopathy) in cats over ten years of age, whereas hyperthyroidism is much less common in other pets. The disease is often characterised by weight loss despite increased food intake, is usually detected by blood tests and is easily treatable.

  I believe the whole article is probably just a straight up machine translation. Traumnovelle (talk) 09:34, 5 July 2024 (UTC)

  Novem Linguae, one of the points of this discussion, I believe, is that there is a difference between poor translations in general on the one hand, and translations by paid OKA editors on the other. Can you confirm that the translations in your 2021 link above as added to Cemetery of San Fernando were from OKA editors? Because if they weren't, everyone, I think, is in agreement that there are very many poor translations by new editors. The question at issue here is whether that applies to OKA editors as well, to such a degree that Afc is necessary for their contributions. Mathglot (talk) 11:21, 24 June 2024 (UTC)

  Can you confirm that the translations in your 2021 link above as added to Cemetery of San Fernando were from OKA editors? They were not OKA editors. That link is just a generic example of how much work machine translations are to clean up. –Novem Linguae (talk) 16:13, 24 June 2024 (UTC)

@Justlettersandnumbers: I'm not sure if you're asking about this specific case or translations in general. If it's the specific case of OKA, it sounds like you've found a bad run of horse-related translations, but myself and others have seen a lot of decent translations from them too. The reason some are asking OKA translations to go through AfC is because they're paid for them, not because they're translations.

If you're asking whether there is community consensus for draftifying poor translations in general, I'd say the answer is no. Unedited machine translations are fair game (a legacy of the WMF's failed experiment with auto-translation, I believe), but if it just needs copyediting then draftspace will not help. AfC reviewers don't routinely do anything about translation issues, as long as it's a viable article. Instead there's the {{Cleanup translation}} family of templates and an active patrol that deals with them in mainspace. – Joe (talk) 11:22, 24 June 2024 (UTC)

The WMF has never attempted to do anything with auto-translation. They accidentally (and briefly) enabled exactly the sort of "machine translation as a base, but then proofread it and clean it up" system that many good editors use themselves, from Spanish to English (only that language pair) here, and then turned it back off when the error was pointed out to them.

In the meantime, one (1) editor dumped a bunch of unedited Spanish mis-translations in the mainspace, and we panicked and created Security through obscurity restrictions on all editors ever since. Which is to say: I can, and have, used machine translation to English in the Wikipedia:Content translation tool, but most editors, including those with far better translation skills than me, won't be able to figure out how to do that on their own. In the meantime, most editors are pasting the contents into machine translation in another tab, and thereby screwing up links, templates, categories, and formatting. Anyone who's been paying attention will know that this is typical of our community. WhatamIdoing (talk) 05:05, 28 June 2024 (UTC)

@WhatamIdoing Indeed. My students do translations for class assignments, and I often tell them not to bother with the official Wiki translation tool because it doesn't work due to the reasons you discuss (i.e. their work can't be easily published). Then, of course, they struggle with code etc. eating our class time, so instead of having let's say a discussion about free culture or such I have to spend time doing activities about how to add hyperlinks or templates or such. On the bright side, they eventually learn the code, at least some of it. But it is still embarassing that I have to tell them "don't use the official tool, it is not friendly enough". _{Piotr Konieczny aka Prokonsul Piotrus| reply here} 09:50, 29 June 2024 (UTC)

Yes, that's what I was referring to. A promising tool that was killed by a botched deployment – typical of the WMF in that era! – Joe (talk) 06:30, 1 July 2024 (UTC)

If you are referring to Wikipedia:Administrators' noticeboard/CXT, then your summary is extremely misleading, as it was about the extremely poor translations from many editors, with that Spanish editor as the most visible example. But upon rereading that discussion, I see that you were trying to muddle the waters and defend the indefensible by providing wrong numbers there already, so I guess hoping that you will change now is rather useless. Fram (talk) 08:58, 1 July 2024 (UTC)

I moved Draft:History of Caraquet back to draft space yesterday. It would be nice if such articles didn't start with presenting speculation by one local amateur historian and genealogist as if it was accepted truth, even though it disagrees with nearly all actual historians and the available evidence. The remainder of the article isn't much better. Fram (talk) 08:58, 1 July 2024 (UTC)

@Fram What policy allows you to draftify such an article without consulting the community? I believe AfD is the only acceptable option (or perhaps PROD/CSD if not contested). _{Piotr Konieczny aka Prokonsul Piotrus| reply here} 09:40, 1 July 2024 (UTC)

WP:ATD, why? The topic is probably salvageable, the article is largely rubbish, so the paid editor can make sure they write a decent article which at least follows accepted science, instead of blindly copying what another Wiki has produced. Fram (talk) 10:07, 1 July 2024 (UTC)

I do not see draftification listsed as an acceptable ATD. Sure, the article needs various fixes, but I don't see why they cannot be done in the mainspace. If you think it should not be in the mainspace, we need a community consensus (i.e. through AfD) on whether it should be de-mainspaced. Single editors do not have the power to delete (hide) articles - this is a task we relegate to the community (outside CSD-level garbage) and this is hardly at that level. See also WP:DRAFTNO. _{Piotr Konieczny aka Prokonsul Piotrus| reply here} 08:13, 2 July 2024 (UTC)

I see it listed under incubation: "Recently created articles that have potential, but do not yet meet Wikipedia's quality standards, may be moved to the draft namespace ("draftified") for improvement, with the aim of eventually moving them back to the main namespace, optionally via the articles for creation (AfC) process..." (the whole incubation subsection is actually about draftification, incubation and draftification appear to by synonyms... Maybe we should just use one term as it seems to be causing confusion?) Horse Eye's Back (talk) 17:45, 5 July 2024 (UTC)

Before the Draft: space was created (late 2013), that section of the deletion policy was talking about the Wikipedia:Article Incubator. Before the Wikipedia:Article Incubator was created (in 2009), we moved such articles to the creator's userspace. WhatamIdoing (talk) 19:27, 5 July 2024 (UTC)

The problem is the ambiguous "Wikipedia's quality standards". Some AfC reviewers seem to decline anything that's not GA-level ready. _{Piotr Konieczny aka Prokonsul Piotrus| reply here} 10:30, 6 July 2024 (UTC)

"Wikipedia's quality standards" does not mean GA and I don't think you will find a single editor who will publicly say that. If an AfC reviewer is doing that on the DL then bring a case against them and get their privilages stripped, someone being an abusive jerk isn't the wording's fault its the absusive jerk's fault. Horse Eye's Back (talk) 16:45, 6 July 2024 (UTC)

We do have a mismatch between the mainspace's actual standards and what it takes to get an article out of AFC. For example, we had a chat last week about why "too short" was listed in Wikipedia:WikiProject Articles for creation/Reviewing instructions as a reason to decline an article. We agreed to change it.

Looking at 10 recently accepted articles, the Page Size gadget shows a median new article accepted by AFC is around 400 words. A quick visit to Special:Random indicates that the median Wikipedia article (most of which are not new, and some of which are very well developed) is less than 200 words. I don't think that AFC should be expecting the typical new article to be twice the length of established articles. WhatamIdoing (talk) 18:18, 6 July 2024 (UTC)

@Horse Eye's Back I've seen some articles declined, by various reviewers, where I am sure those articles would not be draftified or deleted at AfD. Declining articles because not all content is referenced, for example, is I think pretty common (but I don't have a solid sample to say if this is a systemic problem, or I just stumbled upon some expections). Now, it's great to prod new editors and tell them to make sure everything is referenced - but if they don't do this, should their content be declined and even deleted, even through the same article, if published in the mainspace, would at best get some {{fact}}s? For example, recently Draft:Battle of Pinsk was declined due to no inline citations (it only had general links). The creator, fortunately, addressed this and now the article languishes in draft queue, even through it's obviously good enough for mainspace. But even without inline citations, it would've been fine as a stub/start-class; having inline citations is not necessary (not that I am not saying we should not push for their addition, I am just saying - the rules don't say lack of inline citations is sufficient reasons for not publishing content). _{Piotr Konieczny aka Prokonsul Piotrus| reply here} 14:04, 9 July 2024 (UTC)

PS. Even AfC reviewing rules linked just above state that declining article due to lack of inline cites is an error: "Avoid declining an article because it correctly uses general references to support some or all of the material. The content and sourcing policies require inline citations for only four specific types of material, most commonly direct quotations and contentious material about living persons." _{Piotr Konieczny aka Prokonsul Piotrus| reply here} 14:05, 9 July 2024 (UTC)

I have accepted this draft, though the lead section needs re-writing per WP:LEAD. Theroadislong (talk) 17:27, 9 July 2024 (UTC)

Thank you, @Theroadislong. I know some reviewers worry about what will happen if they accept WP:IMPERFECT articles. If the reviewer doesn't accept such an article, they break the nominal rules and often lose content (because the original editor will give up), but if they do, then someone who doesn't know the actual rules or who disagrees with them might come yell at them. If they do it more than rarely, the reviewer can end up at risk of a serious attack. One 'mistake' in 100 adds up to a lot of mistakes if you review thousands of articles, but nobody at ANI says "1% possible error rate"; they say "Obvious WP:COMPETENCE problem; here are a dozen he screwed up on!" WhatamIdoing (talk) 04:14, 15 July 2024 (UTC)

Here's what I think about the issue:

• AFC is essentially broken by design (See Wikipedia:Broken by design)
  • It takes an enormous amount of time
  • Reviewers do thankless work and don't want to exhaustively review a translation
  • Reviewers are technically speaking supposed to allow things that are notable & not promotive through regardless of translation qualify.
• The articles are accurate & unbiased but badly translated
  • Looked at a couple, consider Brazilian Romantic painting. The subject is notable and the article is probably going to be helpful to somebody, but the sentence "This pictorial production was part of the local evolution of the Romantic movement" seems typical. "Pictorial" is a word I assume is more common in portguese, but when used for no reason makes things jarring and hard to understand in English.
  • That one can probably be improved (though it would be a lot of work, given the length). But if you consider Draft:Renaissance in Ferrara, even the lead is genuinely difficult to understanding the meaning behind. Content can't be fixed if it can't be understood in the first place.
• Normal Wikipedia articles are also terrible
  • The current policy on badly translated articles is to leave them up, even if they're just -- in French.
  • There is Wikipedia:Pages_needing_translation_into_English

So, probably let them be created without AFC, but maybe the NGO should have someone who has native-level proficiency in English review them if they're not by professional translators? Because some of the text in Renaissance in Ferrara is bad, and Brazilian Romantic painting is obviously very oddly worded as of the first sentence. Maybe it's ok if some of the articles on Wikipedia are badly worded. Especially because AFC is not designed for this. Mrfoogles (talk) 23:00, 14 July 2024 (UTC)

The one thing that AFC *is* designed for is finding articles that duplicate material already present on Wikipedia, as in Draft:Wooden_house (see reviewer comment). That is important. Mrfoogles (talk) 23:03, 14 July 2024 (UTC)

I don't see how this is a COI. The editors are writing on random topics they choose. Presumably they have no financial or other interest in the topics they are writing on. If individual editors have a COI, they should of course declare it and go through AfC. If an article is really terribly awful and not notable at all, it will be deleted, merged, BLARed, etc.; if an article is really terribly awful and of marginal notability, someone at NPP will probably draftify it. voorts (talk/contributions) 21:20, 18 July 2024 (UTC)

arbitrary break (translated for $$)

There have been a lot of assertions unsubstantiated opinions about the quality of OKA-generated content that range roughly from it sucks to very good, with little to back it up. As of yesterday, articles which have been assessed for quality and which carry the {{OKA}} template on the Talk page now appear in the standard, quality-assessment categories; the parent category is OKA articles by quality. (A flat, quality-agnostic view is available here.) I am not knowledgeable about how these ratings are assigned, but afaik it has something to do with the Afc process. It might be interesting to compare the quality distribution here with that of all translated articles. In any case, at least we have some data to look at, instead of just raw opinion. Mathglot (talk) 20:03, 2 July 2024 (UTC)

Anything B and below is pretty much meaningless in terms of measuring quality as anyone can assign these ratings and they are not given much oversight/critical evaluation. I don't doubt some quality translated articles can exist but offering money for a task that can be very easily automated is a terrible idea as proven by the multiple examples of terrible articles.

@7804j I don't know how your payment model works but if you're paying per article that's a bad idea. Why not pay for good/featured articles instead? It would be much harder to game such a system and would result in better quality if editors were required to work on an article beyond creation. Traumnovelle (talk) 09:47, 5 July 2024 (UTC)

We're not paying per quantity, but per hour of work and instructing that people should focus on quality. Our translators are also paid when they work on improvements of existing articles. 7804j (talk) 09:49, 5 July 2024 (UTC)

How do you measure hours worked given people are working remotely, is it just a trust based model? I can still see someone abusing that through using a machine translation then claiming they did it manually to inflate hours worked. Time clock fraud. Also what put feline hyperthyroidism on the radar, if I may ask? Traumnovelle (talk) 10:08, 5 July 2024 (UTC)

I would say that while hours worked could be an interesting question and relevant for OKA's bookkeeping and financial health, the question of whether OKA is being defrauded by its users is irrelevant as far as Wikipedia article quality is concerned, so can we drop this line of inquiry, or move it to the OKA external website, and stick to the question of how this relates to Wikipedia?

As far as feline hyperthyroidism, I don't understand what you are asking; afaict, you were the first to mention this article. If you meant, "How did this topic get picked up by an OKA editor?" then I would say that my understanding is that OKA editors get to work on any topic of their choice. Is that what you were asking? Mathglot (talk) 10:35, 5 July 2024 (UTC)

Thanks for starting the category Category:OKA articles by quality, Mathglot, as a useful list of OKA articles. Unfortunately our assessments are normally almost entirely based on length, regardless of quality - and many OKA articles are all too long. I see there are ZERO A/FA/FL/GA class articles, & the great majority are B or C. Taking Brazilian Abolitionist Confederation, a 37 kbyte B-class slab from the dreaded User:Racnela21, this is in principle the kind of article I'd support, as being something we are unlikely to cover otherwise, even if it has only had about 40 views a month, a quick skim finds "The document also reports on other aspects of the history of the advances and setbacks made in the Empire's path towards abolition, which is described as a fatality that "caused slavery to become a fact and, what is more, to obtain universal tolerance". Huh? Johnbod (talk) 16:34, 9 July 2024 (UTC)

Zero sounds like the right number. The total number of A/FA/FL/GA class articles in the encyclopedia is 59,491^[source] and I would expect a sample of 60,000 articles drawn randomly from 8M articles in Wikipedia to have about 0.007 articles rated A/FA/FL/GA class. By that reckoning, we should see the first high quality OKA articles appear when the total number of OKA translations reaches somewhere between 200 and 1,000 times the current number. Mathglot (talk) 19:12, 9 July 2024 (UTC)

You're putting this paid editing on par with mass stub creation by now-banned users and all the terrible articles that wouldn't (or shouldn't) survive AfD. A lot of these articles are machine translated without any work in fixing them put into them by the 'creator'. Traumnovelle (talk) 19:19, 9 July 2024 (UTC)

My calculation is bonkers and I was going to redo it, but frankly trying to play a numbers game and somehow measure that against an unknown number of articles that shouldn't survive Afd is a distraction. In any case this is only a sliver of a much bigger issue, already being discussed in other forums, namely that of machine translation and AI output being added to the encyclopedia. This sliver is getting more attention because of the paid aspect, but it goes far beyond that. How are we to deal with that? Somebody said, "If you can't measure it, you can't improve it," and quality ratings seem like the first step. If they are strictly connected with length, do they have any value at all then? Mathglot (talk) 19:54, 9 July 2024 (UTC)

Let's be clear to those looking on; the calculation is not just wrong, it's wrong by many orders of magnitude and should be ignored. if about 60,000 out of 8M articles are A/FA/FL/GA, then that is a rate of 0.007 per article. If you then sample 60,000 articles, you would expect (60,000*.007) articles to me that class... not zero, but 420. If the OKA articles were as good as typical for reaching those classifications, we would expect to see them. (A reference above suggests there are 7000-and-some Oka articles, so we'd expect around 50 meeting that class.) This is not to say that there aren't other considerations, such as the age of the article; how many articles as new to engWiki as the Oka ones are of that class? -- Nat Gertler (talk) 21:59, 9 July 2024 (UTC)

It also takes time and motivation to achieve GA or FA. So don't expect new articles to easily reach that standard. For B class someone should have checked that it was well written. So hopefully B class OKA assessment is not just based on size and pictures. Graeme Bartlett (talk) 00:10, 10 July 2024 (UTC)

Well, all too clearly they haven't - see various complaints above and elsewhere. Many of these are translations of FA/GA articles on other wikis. Even allowing for different standards, if they were in comprehensible English, many ought to at least pass GA. But neither OKA or anyone else is interested in nominating them, if only because they would often need a total rewrite. Johnbod (talk) 01:26, 10 July 2024 (UTC)

The featured article on de.wiki I saw translated by OKA had multiple prose lines that were unreferenced. Whether that is due to article deterioration or if it passed like that I am unsure of. But machine translation introduces many issues that would result in an article failing GA class. Traumnovelle (talk) 01:30, 10 July 2024 (UTC)

No, that's just how dewiki does things. They use very few inline citations, even in Featured Articles. Today's Featured Article there is w:de:Paul Maas (Altphilologe), which has ~1,300 words and 10 refs. About half the paragraphs have no citations at all. Their wiki, their rules. WhatamIdoing (talk) 04:22, 15 July 2024 (UTC)

Policy against demands of proof of non-existence

Answered to my satisfaction - Altenmann >talk 18:17, 23 June 2024 (UTC)

Now and then someone tells me something like "What proof do you have that J. Random was not a Christian?" I know this is a logical blunder, but I cannot remember any rule against this in our WP:V rules. Neither I remember the name of the fallacy. Can someone remind me? - Altenmann >talk 17:20, 23 June 2024 (UTC) Proving a negative? Similar to but not the same as Argument from ignorance? Idk if it is in WP policies, but I would want proof (sourcing) that he was. Selfstudier (talk) 17:30, 23 June 2024 (UTC) but I would want proof (sourcing) that he was -- My question is about demanding a proof that 'he was not. - Altenmann >talk Proving a negative is philosophically too broad. But Evidence of absence seems to suit Wikipedia's approach to WP:TRUTH: our WP:V requires evidence. - Altenmann >talk 17:54, 23 June 2024 (UTC) I assume you're talking about this for statements within an article context, in which case I would need to see an example statement in which it's a problem. If the article on Judy Random states that she was a Christian, I would expect that to be sourced, as well as any statement that she was not a Christian (which is a sourcable thing.) If you're talking about in discussion, that seems quite allowable thing to ask, depending on what was being discussed. -- Nat Gertler (talk) 17:55, 23 June 2024 (UTC) It does not matter. Talk pages are not an idle chat: they are about article content. Of course you can say in talk page anything you want, but if the implications are to change article content, then the arguments must be based on reliable sources. Of course, there are discussions where opinions of editors do matter, such as article titles (heck, take AfDs), but still, they must involve arguments, not opinions, and arguments boil down to shat is said in "real world"- Altenmann >talk 18:01, 23 June 2024 (UTC) OK, the point is, if an article wants to claim that Random was not a Christian, you do actually need a source that says Random was not a Christian. I don't see what's hard about this. WP:V requires verifiability for all claims, including negative ones. --Trovatore (talk) 18:04, 23 June 2024 (UTC) Absolutely it does matter. Your initial post seemed to be seeking a rule against it, and you're on a page for discussing policy. The verifiability policies already cover this for article content, and there's no particular need for a rule against it elsewhere. The example is weak, as it seems quite possible to source a statement that Judy Random was not a Christian or to specify that she held some other religious belief. But if someone is asking that on the talk page, it seems quite a reasonable response to a talk page statement that she was not a Christian. It should not be disallowed to ask that as a response for a claim. -- Nat Gertler (talk) 18:08, 23 June 2024 (UTC) Sorry, I stated my question incorrectly. Let me set it closer to the issue: Someone added Category:Buddhists to a bio. I removed it and I was reverted because I didnt provide an evidence that a person was not a Buddhist. What would be my proper counter-argument. WP:CATV didnt enlighten me. Sorry for my fussy brains. - Altenmann >talk 18:13, 23 June 2024 (UTC) The WP:ONUS is on the person doing the adding to justify the addition. Usually, one could expect WP:BRD but that's not compulsory. So discussion on talk to resolve. Selfstudier (talk) 18:17, 23 June 2024 (UTC) Got it. WP:ONUS is what I needed. - Altenmann >talk 18:20, 23 June 2024 (UTC)

The above collapsed discussion does raise a point that sometimes troubles me. Category links don't have footnotes. In theory they're supposed to be justified by sourced material in the article, but you can't necessarily tell which cite justifies the category.
Of course in most cases this is not that much of a problem, but it can become one when someone adds a category that makes a potentially contentious claim. I remember this specifically over someone wanting to add category:Whitewashing in film to The Last Temptation of Christ (film), which struck me as an uncited criticism of the casting. --Trovatore (talk) 21:46, 23 June 2024 (UTC)

Perhaps one way to resolve this for categories without an clear justification in the prose (or which might do if prose is removed from the article for any reason or perhaps even just reworded) would be to put a hidden comment next to the category link with a source or explicit link to the relevant section of the article (e.g. "see criticism from XYZ Group", "source: P.D. Michaels, 2024", "Ref name=BBCNewsApril29"). Thryduulf (talk) 23:02, 23 June 2024 (UTC)

Hmm, it's better than nothing, but it seems more aimed at editors than at readers. --Trovatore (talk) 23:20, 23 June 2024 (UTC)

(A distinct but related concern is that categories can appear to make assertions in Wikivoice, which we have to be careful about.) --Trovatore (talk) 23:23, 23 June 2024 (UTC)

Categories are supposed to be for defining characteristics. If it's a defining characteristic, it really should be in the prose (although with the way we create categories like "Left-handed Inuit arcwelders from Texas", it may be a combination of different sections of prose.) Per WP:CATV, "It should be clear from verifiable information in the article why it was placed in each of its categories."-- Nat Gertler (talk) 02:01, 24 June 2024 (UTC)

Something else related to the collapsed part of this discussion, but not mentioned there, is that sometimes justification for a category can be implicit. For example if a person is verifiably Swedish and verifiably a member of an organisation that requires members to be Buddhists, you don't need an explicit citation to add Category:Swedish Buddhists to the article unless there is evidence they are/were not Buddhist (perhaps they renounced that religion later in life). Thryduulf (talk) 23:10, 23 June 2024 (UTC)

I think someone adding a category which casts the subject in a negative light, most especially if a BLP, ought to be prepared to defend the addition if challenged. Wehwalt (talk) 01:26, 24 June 2024 (UTC)

Everybody who adds anything needs to be prepared to defend it if challenged. In the example above the defence would be exactly as I've laid out - they are/were Swedish, are/were a member of an organisation that requires members to be Buddhists and there is no evidence the person adding it has seen to the contrary. Thryduulf (talk) 08:37, 24 June 2024 (UTC)

Hmm, so out of curiosity I took a look at that category, which has only two individual bios at the top level, one of which is Malin Ackerman. Ackerman's bio categorizes her as both a "Swedish Buddhist" as an "American Buddhist". However, the body asserts that she was raised Buddhist, and mentions her "Buddhist upbringing", but does not assert that she is currently Buddhist.

Not sure there's a broad policy conclusion here, but I think it's worth noticing that articles are not always entirely careful about these things. Thryduulf, this is arguably similar to the case you mention. She was raised Buddhist, with sources (I haven't checked them, but that seems not on-point in this discussion), and we have no active assertion that she decided she wasn't a Buddhist anymore. Is that enough to put her in the cats? My intuition is no, not when the article uses language that seems noncommittal on her current status. --Trovatore (talk) 17:38, 24 June 2024 (UTC)

I had a similar issue with people adding categories like Jewish Conservatives to Benjamin Disraeli, who was certainly not both Jewish and Conservative at the same time ... Wehwalt (talk) 17:50, 24 June 2024 (UTC)

Disraeli is not even an edgecase - the lead of the article makes it very clear that that category is incorrect and so should not be on the article. Thryduulf (talk) 17:56, 24 June 2024 (UTC)

@Trovatore: I picked the category out of thin air, so it's interesting you found an edgecase! Reading Ackerman's bio (but not the sources), I'd say that if the standard is "on the balance of probabilities" then the category is correct but if the standard is "beyond reasonable doubt" then it isn't (not because it's necessarily incorrect, but because there is reasonable doubt).
When it comes to BLP anything contentious or potentially defamatory absolutely needs to have the higher standard of proof, something innocuous is usually fine at the lesser standard (although obviously better is always preferred if possible). A person's religious beliefs are something that can be contentious and some people would regard some mischaracterisations as defamatory, but not everybody and not always. Given the content in the article I am completely confident that describing Ackerman as Buddhist would not be defamatory even if correct, and I'm not seeing anything to suggest it is contentious. My gut feeling is that they are probably nominally or casually Buddhist - someone who doesn't actively practice the faith but would tick that box on a form. Thryduulf (talk) 17:54, 24 June 2024 (UTC)

So the analogy with legal burdens of proof could get a bit strained, but I'd kind of suggest that the (underused) clear and convincing evidence might be a better way of thinking of it. "Eh, it's probably true" doesn't strike me as good enough to add a cat, particularly to a BLP, even if we think the subject probably doesn't mind being called a Buddhist. --Trovatore (talk) 22:33, 24 June 2024 (UTC)

I don't think categories are (or should be) limited to current status. Babe Ruth is not currently a baseball player, but he's probably properly in those categories. -- Nat Gertler (talk) 17:56, 24 June 2024 (UTC)

Wow, new one on me. I did not know that Babe Ruth was a Swedish Buddhist.

Anyway I think that's a bit of a different issue. Ruth's profession was ballplayer, until he retired. That's what he was known for. Ackerman is not particularly known for being a Buddhist, as far as I'm aware.

It does raise some interesting questions. Eldridge Cleaver became a conservative Republican, but is most known for what you could call "far left" activism, to the limited IMHO extent that that terminology makes sense. Does he belong in e.g. "socialist" categories? I really don't know. --Trovatore (talk) 22:51, 24 June 2024 (UTC)

Maybe we need "Lapsed ..." categories. Donald Albury 23:03, 24 June 2024 (UTC)

I think the idea that categories should be a single-moment snapshot rather than reflecting the wide range that has been noted is wrong. We have a list of American politicians who switched parties in office -- which party's categories should they be under? Both! There may be some categorization that only applies to non-notable periods of their life -- Jane was baptized but declared herself an atheist when she was 12, long before she became a professional cat juggler, so she certainly doesn't belong in Christian cat jugglers and perhaps not even in Christians at all, but if she switched from atheism to agnosticism mid-career, then she does belong in both atheist cat jugglers and agnostic cat jugglers. -- Nat Gertler (talk) 23:14, 24 June 2024 (UTC)

Categories are fundamentally for navigational purposes. If someone is looking for articles about ____, then they should find the articles related to _____, even if occasionally that article says "Well, you might have thought he was a ____, but the truth is rather more complex and interesting than that". WhatamIdoing (talk) 05:21, 28 June 2024 (UTC)

I'm aware of that viewpoint but I don't really agree. The problem is that an article's presence in a category often appears to be an assertion (in Wikivoice no less) that the subject of the article satisfies the category's defining criterion. If there were a way to make it clear to readers, including casual ones, that that is not the case ... but there isn't. --Trovatore (talk) 17:21, 28 June 2024 (UTC)

That's the biggest problem imo, with cats. I stopped paying attention to them for that reason, as long as people are not using cats to enforce or contradict content in actual articles, fine with me. Selfstudier (talk) 17:30, 28 June 2024 (UTC)

In some cases it can be mitigated by renaming the cat to make the criterion more objective. For example I happened to see that the category I called out, category:Whitewashing in film, is actually at CfD. I think a lot of the problem would go away if we renamed it to something along the lines of category:Controversies over whitewashing in film. It's reasonably objective whether there was a controversy; you can support that with one reliable cite. Whether the film is actually an example of whitewashing is much more fraught. --Trovatore (talk) 01:20, 29 June 2024 (UTC)

Trovatore, the second sentence of the Wikipedia:Categorization guideline literally says "The central goal of the category system is to provide navigational links to pages in Wikipedia within a hierarchy of categories".

I conclude from this that categories are therefore fundamentally for navigational purposes, equivalent to something in a navbox. WhatamIdoing (talk) 23:21, 30 June 2024 (UTC)

That might be the goal, but it doesn't trump V or NPOV. I sharply disagree with the idea of providing categories that might appear to make contentious claims just because they might help someone find something. --Trovatore (talk) 19:34, 1 July 2024 (UTC)

We follow the same basic rules for all forms of navigational content. That means that if it doesn't have to be cited in a navbox, it doesn't have to be cited in a category, or a ==See also== list, or a disambiguation page. None of them should be unfair ("non-neutral"), but the primary point of all navigation is to help people find things, not to hide appropriate content away because someone might jump to conclusions. WhatamIdoing (talk) 21:02, 2 July 2024 (UTC)

Putting an article in category:Foos, on its face, makes the claim that the subject of the article is a foo. That ineluctably implicates V, and the claim that it is a foo must be cited (if contentious). --Trovatore (talk) 20:45, 3 July 2024 (UTC)

See also why {{unreferenced category}} exists.

But I wonder whether we have the same idea about what's being "contended" in that small minority of cats that are actually contentious. I think that what matters is whether the article would be of interest to someone looking for articles about _____. This would include articles that are not, sensu strictu, actually about _____. For example, if you look in Category:Planets, you will find 13 pages and one redirect that are not planets. If you take the POV that putting an article in Category:Planets means you are defining that article's subject as "being a planet", then you will be unhappy to discover pages like Definition of planet and Equatorial bulge in that cat, because those subjects are related to planets but not actually planets themselves. OTOH if you take my POV, which is that putting an article in that cat means that someone looking to learn more about planets might be interested in those articles, then you won't have any concerns about the contents of that cat at all. WhatamIdoing (talk) 00:25, 4 July 2024 (UTC)

As I said, if we could make your POV clear to all readers, including casual ones ... but we can't. --Trovatore (talk) 02:00, 4 July 2024 (UTC)

We could. If we wanted to, every content cat page could have an explanation at the top that explains what a category is, what it means for a page to be listed there, and how to use the page. We haven't chosen to do this yet, but there's nothing stopping us from doing so, if we thought it was really important to explain to readers why Category:Planets does not exclusively contain articles about planets.

Actually, we already do, to a very limited extent; it looks like there's a link to Help:Category at the top of every cat page. That's more of a how-to/editor-help page, but we could change that to a reader-help page, if we wanted to. WhatamIdoing (talk) 03:21, 8 July 2024 (UTC)

But that's only if they go to the category page. The casual reader is only seeing the bottom of the page where it lists that this person is in Category:deadbeat dads and Category:pervs and skeezes with nothing covering the fact that he's in those categories as a critic of those people. (Obviously, I'm feeling far too lazy to look up a real example this morning.) -- Nat Gertler (talk) 13:30, 8 July 2024 (UTC)

If you're in the article, and it's important enough for the person to be in the cat, then there should normally be some content in the article itself that makes the relevance reasonably clear. That doesn't require the article to say "Ro Righteous is a deadbeat dad", but it should say something related to the subject (perhaps "Ro Righteous has repeatedly introduced legislation about child support payments"). WhatamIdoing (talk) 04:29, 15 July 2024 (UTC)

Because if there isn't any relevant content in the article, then there's no navigational point in putting the article in [Cat:Deadbeat dads]. WhatamIdoing (talk) 04:33, 15 July 2024 (UTC)

Categories are definitely a navigational tool...... when Wikipedia began we thought it'd be a good way of collecting data and analyzing relationships between articles. However categories are so unstable that data can never be reproduced for any real academic analysis. This is also a problem with our vital articles Moxy🍁 00:54, 4 July 2024 (UTC)

I think they're still a pretty good way of collecting data and analyzing relationships/various other things. They're certainly useful for building graphs and graphs can be very useful. There can be a lot of instability and confusing weirdness, but for the one fairly large topic area I've looked at, there are plenty of relatively stable structures in the networks too (that can be easier to see when edges are bundled e.g. here). Sean.hoyland (talk) 02:11, 4 July 2024 (UTC)

Idea is cat= List= navbox sort of thing, I get that but they are susceptible to manipulation and lots of editors can't be bothered to check, including me, although I used to. I might correct if I happen to notice something weird or outlandish, but all this diffuse, parent/child blah, nah. Selfstudier (talk) 11:12, 4 July 2024 (UTC)

Unmaintainable lists

Although we have generally avoided creating insane lists like List of notable men, List of albums (a redirect), List of roads in California, List of bird species, etc., there doesn't seem to be any clear policy against creating such lists, other than the vague guidance at WP:NOTDIRECTORY, WP:SIZE, and WP:LISTCRIT, which collectively discourage such lists, but don't seem to prohibit them. Because of this vague guidance, we sometimes end up with lists that are basically unmaintainable (especially with our shrinking editor base). Lists like:

• List of gay, lesbian or bisexual people
• List of people who died in traffic collisions
• List of women writers
• List of kidnappings (may lock-up your browser)
• List of monarchs of fictional countries
• etc.

Many of these lists are either arbitrarily populated (and misleadingly short) or they were heroically filled out by one editor and then left to rot. My questions are:

1. Is this a problem?
2. If so, how can we better prevent such lists from being created?

Nosferattus (talk) 01:13, 9 July 2024 (UTC)

FWIW, some lists display Template:Dynamic list, stating that the list might never be complete.—Bagumba (talk) 06:35, 9 July 2024 (UTC)

We used to have lists of people by name. Many of the broader lists, some of which pre-date the introduction of categories in May 2004, have been deleted/rationalised over the years. Graham87 (talk) 08:00, 9 July 2024 (UTC)

In some cases the scope of the lists should be clarified and/or long lists split into more focused sublists. e.g the List of kidnappings would benefit from tighter inclusion criteria and splitting (probably by date or location). Each of the lists needs to be evaluated individually though and I don't think there is anything we can (or should) do to prevent lists being created. Thryduulf (talk) 08:35, 9 July 2024 (UTC)

I'm not sure these lists are a bad idea in principle, or that we should have a policy against them, but I agree that the implementation is problematic. It feels like the ideal solution would be improving the category UI (current category views don't support easy full text search across all items, for example), or some sort of "Autolist" feature that derives equivalent list pages from category data (optionally pulling in the lead paragraph of each article, or wikidata items, to form a table). Barnards.tar.gz (talk) 10:32, 9 July 2024 (UTC)

We already have what is essentially a much improved version of both categories and lists in the form of Wikidata. You can generate such an 'autolist' from Wikidata using {{Wikidata list}}. Unfortunately, a vocal group of editors managed to block us actually using Wikidata on enwiki early on, based on a severe case of not invented here syndrome, so we can't have nice things like fr:Archives_de_l'État_de_Neuchâtel#Type_de_fonds_conservés_et_consultation. – Joe (talk) 10:56, 9 July 2024 (UTC)

Interesting... do you have a link to where that consensus was established? Was it about Wikidata generally or about the Wikidata list template specifically? I can see some limitations of the template, but the approach seems promising. Barnards.tar.gz (talk) 11:17, 9 July 2024 (UTC)

There's a nice summary of the major discussions at Wikipedia:Wikidata#Appropriate usage in articles. There's never, as far as I know, been a clear consensus against using Wikidata-generated content in mainspace (be that Wikidata lists or Wikidata infoboxes like {{Infobox person/Wikidata}}), but a few editors opposed to it were very effective at simply edit-warring out all instances of them, citing the lack of affirmative consensus to use them. – Joe (talk) 12:03, 9 July 2024 (UTC)

Current consensus is based on this RfC from 2013 -- a few months after Wikidata went online -- which resulted in It is, on the other hand, not appropriate to use Wikidata in article text on English Wikipedia at this time. There were many reasons. Especially at that time nobody really knew how Wikidata even worked. The biggest remaining objection, expressed in another RfC last year (which found no consensus to change the status quo), is that it requires going to Wikidata to edit such articles, plus some fears that Wikidata is subject to less scrutiny as those changes won't show up on watchlists.
The thing is, enacting a full ban like we have basically kills the idea of those improvements ever happening. WMF builds things they know will have some kind of measurable impact. Despite "Wikidata editing from Wikipedia" being a perennial request and even sometimes goal for the WMF for the past decade, it's not realistic to expect them to devote real resources to actually getting it done when there's no guarantee we'd even allow it. But it would also make no sense to invest resources into drawing from our category system, which is both localized to enwp and a clumsy, ancient system compared to what Wikidata can do.
Back to the point of this thread, yes, an article like "list of women writers" is a crazy, unmaintainable list (btw, Nosferattus, I think the most relevant guideline is WP:SALAT), and would make a lot of sense to be updated via Wikidata. But since we can't have nice things, and because we do allow index articles, the best we can do is split it up by e.g. nationality or era. — Rhododendrites ^talk \\ 15:33, 9 July 2024 (UTC)

@Thryduulf and Barnards.tar.gz: What would be your opinion of a list like List of notable women or List of albums? Clearly more granular lists would be preferable in those cases, but why? What is the threshold for a list being unmaintainable? It's interesting to me that even Wikipedia:Miscellany for deletion/List of people by name was a contentious discussion (and originally closed as no consensus). Surely, our resources are better off not devoted to such Sisyphean tasks. Nosferattus (talk) 15:35, 9 July 2024 (UTC)

I think such lists are beset with practical problems. In practice, there aren't enough heroes available to watch and maintain them. Many aren't just lists of articles, but also contain content, which needs sourcing and verifying and copyediting and all the other things that make an article good. Now, that's a problem that many niche-subject articles have, but at least niche subjects can have a small enough scope that an article can be plausibly materially complete, without degrading over time. In contrast, these list articles go out of date immediately and need constant effort to maintain. A list article might be considered maintainable if it was popular enough to attract the requisite army of editors, but that might then result in another practical problem: List of notable women might be hundreds of thousands of entries long, likely hitting technical limits.

I took a look at {{Wikidata list}}, and although I like the result that has been achieved on frwiki, I could understand reticence to start using it to replace big list articles. For a start, it's much less accessible to editors, requiring proficiency in SPARQL as well as Wiki templating. It seems to depend on a bot rather than being updated in real time? That's a departure from the normal editing workflow. I'm also not sure how closely Wikidata's notability and sourcing practices match up to Wikipedia's. Might we be auto-importing junk? Nevertheless, exploring the option of using Wikidata or Wikipedia's existing category data, or some other system-based solution, seems to me a more promising line of enquiry than expending tons of manual effort maintaining unusably large articles. Barnards.tar.gz (talk) 16:01, 9 July 2024 (UTC)

List of notable women and List of albums should be (redirects to) indexes of smaller, more focused lists. What is the threshold for a list being unmaintainable? There is no single answer to this question as it depends entirely on the context of the list - for example how many entries there are, how much information there is about each entry, how complete is the list (can it ever be complete? c.f. Wikidata's concept of expected completeness)? How objective is the definition (compare e.g. List of US States and List of shades of green)? How objective are the inclusion criteria? How granular is the list? What sublists do or could exist? In some cases the answers to these questions vary over time, e.g. prior to circa the 2010s a single list of exoplanets would have been regarded as a manageable finite list, since then the number known has increased exponentially. Thryduulf (talk) 16:05, 9 July 2024 (UTC)

Just to echo Thryduulf, it is a very common practice to split very long lists into a group of smaller more refined lists, and leave the original title as an index of lists. This could very easily be done with all of the current lists named by time period or geographic scope. BD2412 T 17:09, 14 July 2024 (UTC)

There are "only" about 6+ million pages on Enwiki. The universe of overlap among those 6 million pages, and traffic accidents, or kidnappings, is maintainable. The advantage of a list vs. a category is the list allows short descriptions, sortable columns for data points, etc.. This is not a bad thing. There are encyclopedias devoted to woman writers (often by country or time period). There is also an encyclopedia of kidnappings. We are not creating topics out of thin air, in most cases. Something like List of monarchs of fictional countries is a little strange but maybe it's a topic covered by RS? The statement "left to rot" is another way of saying "left to be further improved by someone else", which is how Wikipedia is designed. I've done this myself, and seen others do it. If it takes 1 months of 10 years, what's the rush? Wikipedia is not a finished product. -- GreenC 17:49, 9 July 2024 (UTC)

I've thought about this a lot because I see them in NPP and there is not much guidance on this and because it spans multiple guidelines (lists, notability, wp:not, and verifiabiltiy) none of which really cover it. ("Verifiabiltiy" because inclusion on a list is implicitly a statement that they are/meet the criteria of the list) IMO one good criteria would be that there is a reasonable likelyhood that some readers would look for it and find it useful. Which would include these criteria:

1. Some selectivity So "list of humans" would not fulfill this criteria
2. Not overly specialized in some abstract way (usually on "compound criteria" lists) So no "List of US senators who are 5' 7" tall"
3. Somewhat objective (vs. subjective) criteria. So no "list of good US Senators". But "list of notable violin players" is somewhat objective and OK.

Sincerely, North8000 (talk) 18:33, 9 July 2024 (UTC)

Not to get off the original topic, but as far as the secondary topic of how to populate lists and the discussion above of things like using Wikidata, it may or may not interest anyone that a Wiktionary user developed a javascript for populating ~lists by scraping categories' contents, discussed here, which is another possible idea (it pulls and displays definitions, but it's possible to imagine a Wikipedia implementation pulling e.g. shortdesc or infobox contents or lead sentences instead), though the benefit of "this is the contents of category X, just as a list" is admittedly debatable. (Or perhaps Wikipedia already has its own such gadget...) -sche (talk) 17:01, 14 July 2024 (UTC)

If you don't mind me going even further off topic, I've proposed a split at Talk:List of common misconceptions#Split proposal and I'd value any and all opinions. WhatamIdoing (talk) 00:35, 15 July 2024 (UTC)

Talk pages of contentious topics

Per WP:CTOP, many pages don't allow certain users to edit the page, or post in the talk page, but allow for all editors to request an edit on the talk page. Is the requesting editor allowed to respond to a reply on their request, in order to further explain the request/its reasoning, or to question a denial of the request? JoeJShmo^💌 01:16, 9 July 2024 (UTC)

IMO, no. WP:ARBECR intends that non EC editor contributions be limited to (straightforward a la WP:EDITXY) edit requests only. If an EC editor were to specifically request clarification, then that could be responded to, but otherwise once the edit request has been made, then it is a matter for EC editors to decide whether to implement it or not. Selfstudier (talk) 08:25, 9 July 2024 (UTC)

In my opinion they absolutely should be able to do the first, because it is very clearly within the spirit of the exception and is a clear benefit to the project. The second is a much greyer area, I think it should be allowed as long as it doesn't stray into discussion - a single request for explanation (especially of jargon or references to past discussions) if the denial is unclear or a "I think you've misunderstood my request because..." as long as it's done in good faith. Do note though that some editors prioritise strict adherence to the letter of the rules much higher than I do (some accept even edit requests only begrudgingly). Thryduulf (talk) 08:30, 9 July 2024 (UTC)

My rough metric is "could this response reasonably have just been a stand-alone edit request". If an IP requests and edit adding the language "She was born in 1887", and it's declined due to a need for reliable sources, a response with links to such sources is fine. It's just bureaucracy to require them to submit a second edit request.

If the decline is at all an indication that the change itself is opposed, further replies from the IP could not reasonably be an edit request, since edit requests are for uncontroversial changes. Firefangledfeathers (talk / contribs) 12:03, 9 July 2024 (UTC)

Non-extendedconfirmed users are excluded from the consensus forming process. So, unless the responding party has clearly misunderstood the request, they are limited to responding to a request for clarification from an extendedconfirmed user. That's it. And I would say (based on experience in the PIA topic area), if they show any sign of a sense of entitlement by not complying with the rules they should just be blocked. Sean.hoyland (talk) 05:28, 14 July 2024 (UTC)

@Sean.hoyland @Firefangledfeathers@Thryduulf@Selfstudier There is now a case about this editor at AE Doug Weller talk 09:43, 14 July 2024 (UTC)

Thank you Doug, and thank you everyone for your responses. It seems a certain user (unfortunately with a previous vendetta against me) has jumped the gun and opened a case relevant to this discussion. I don't want anyone to waste their time, so I'd recommend not worrying about it too much. Take care. JoeJShmo^💌 10:34, 14 July 2024 (UTC)

WP:V relating to photographs

How to photographs align with WP:V? Consider an example "Jane Jones drove to work in a blue Chevrolet" where sources describe Jane Jones driving to work, have a picture of her blue Chevrolet, but do not explicitly describe the vehicle as blue. Would we be able to describe the vehicle in our articles based on the photograph publish in a WP:RS or only based on prose descriptions published in WP:RS? I took a glance at WP:V and couldn't find anything. --24.125.98.89 (talk) 13:18, 14 July 2024 (UTC)

If a photograph is published in a RS, we generally trust that the photograph is what the RS says it is. Synthesis can still creep in though. In your example, if the RS just says “this is a picture of Jane Jones’s car”, then it’s possible that she drove to work in a different car which might not have been blue. If the RS says clearly “this is the car in which Jane Jones drove to work”, then I think it would be verifiable to say she drove to work in a blue car. If this turns out to be an important or controversial fact (e.g. for Blue–green distinction in language reasons), we might want to ask for the assessment of “blue” to be made by the RS rather than by the editor.

If the source isn’t explicit in text, then it may also be a question of whether that information is due. Barnards.tar.gz (talk) 13:41, 14 July 2024 (UTC)

24.125, is this related to the tag you added to List of killings by law enforcement officers in the United States, July 2024 about whether the would-be assassin was actually a white person? I don't think anyone has any genuine doubts about that, and a couple of minutes turned up sources like these:

• "a tall, slender, white man who wore glasses and had long sandy-brown hair" [2]
• "According to the FBI, the shooter was identified as Thomas Matthew Crooks, a white male" [3]

I don't think you should assume that the information was added on the basis of a photo. WhatamIdoing (talk) 03:33, 15 July 2024 (UTC)

That series of lists does seem to have both V and DUE issues in the race column. I spot checked a few. There are cases where the person’s race is not mentioned and there is no photo, making it look like race has been inferred from name by the editor. There are cases where the person’s race is not mentioned and an image is given, and the editor may have made their own judgement about what their race looked like. Barnards.tar.gz (talk) 08:39, 15 July 2024 (UTC)

WP:Logo's stance on ® and ™ in logos

Following discussions at Wikipedia talk:WikiProject Ice Hockey#® and ™ in logos. Read it for more information.

WP:Logo doesn't say whether ® or ™ in logos are discouraged or not. If the entity itself uses its logo with these marks on its website, should the marks then be edited out, or remain? Here are some pages using logos with these marks: Amtrak, American Airlines, Delta Air Lines, Southwest Airlines, Hawaiian Airlines, United Airlines, Five guys, New York Yankees, Boston Red Sox, Chicago White Sox, Titleist, Puma (brand), Discovery, Inc., Spotify, Hammarby Fotboll and 2026 FIFA World Cup. As far as I can tell, most pages on Wikipedia uses logos without these marks. Jonteemil (talk) 15:11, 14 July 2024 (UTC)

If a mark owner is consistently presenting their mark with the registration symbol (as they should be), then it is really not our place to alter the image to something different from its normal presentation. BD2412 T 16:23, 14 July 2024 (UTC)

As was stated in the hockey Wikiproject discussion, the reason that entities themselves use such marks when presenting their logos is in the active defense of their trademark. That neither compels Wikipedia (or any other entity) to do the same, and as you were told, MOS:TMRULES already sets forth "Do not use the ™ and ® symbols, or similar, in either article text or citations, unless unavoidably necessary for context." While you responded that TMRULES does not explicitly include images, the same principle remains valid. If it ain't broke, don't fix it. Ravenswing 17:06, 14 July 2024 (UTC)

Note this is a slightly different question than you posed on the WikiProject Ice Hockey talk page. I don't know if I'd recommend editing a trademark symbol out of an image. As I stated in the other discussion thread, though, trademark symbols are not part of the trademark. Thus an accurate depiction of the trademark does not include the trademark symbol. If other qualities between a choice of two images are relatively equivalent, I feel a version without trademark symbols would be a better illustration of the actual trademark. isaacl (talk) 17:07, 14 July 2024 (UTC)

I agree with isaacl that I would not recommend editing a trademark symbol out of an image, but that's just because it's not important. This is so unimportant that if we had multiple versions of a given logo, it would be extremely low on my list of reasons to prefer one or the other. It'd be more pointful to argue over the relative merits of SVG vs PNG vs JPG than to worry about this.

I think this type of question tends to arise because someone is trying to do the right thing, but please let us reassure you: Nobody is legally required to display the ® symbol on someone else's trademark. WhatamIdoing (talk) 03:44, 15 July 2024 (UTC)

Are new rules needed for high-profile or previously contested proposals?

We have recently had a slate of discussions at Wikipedia:Move review wherein discussions had run for the requisite seven days (or had been relisted and run for fourteen days or more), and had the typical handful of participants, and were thereafter closed as moved, and where post-closure challenges were made asserting that because of the scope of the subject, more time or participation was required. An example is the still-pending discussion of the disambiguation of ABC News, and the move of the subject previously at that title to ABC News (United States). Another discussion with a comparable move review was the move of Chairperson to Chair (officer), for which it was noted that despite near-universal support for moving the article away from chairperson, a pervious discussion had rejected Chair (officer) as a title.

We currently have one set of rules for carrying out a contested or potentially contested requested move. We have no mechanism for identifying proposed moves, merges, or other changes that could arguably require some lengthier discussion time or higher standard of community participation. We do routinely notify relevant Wikiprojects, and allow anyone to relist a discussion if there is an absence of clarity on consensus. I am bringing this here to figure out if we do need some extra set of rules for discussions that deal with higher-profile topics, or will involve fixing a lot of links and templates, or deal with a subject that was previously discussed and reached a different consensus. I am not comfortable with the idea that some unwritten rule exists requiring different treatment of such cases. The rules should be written. Cheers! BD2412 T 00:45, 16 July 2024 (UTC)

WP:RMRELIST is a fine rule, and already contains guidance telling editors to publicize an RM as widely as needed. We shouldn't start carving out exceptions, especially when we can't predict them from the outset (for example, I would not have expected ABC News to be all that contentious). Additionally, "I didn't get a chance to participate" is not a good reason to overturn an otherwise correct close. If somebody thinks the consensus developed in the first discussion was horrendously wrong, they can open a new one. voorts (talk/contributions) 01:05, 16 July 2024 (UTC)

I had thought about this, having made closures on highly linked articles before when I initially started nosing around at WP:RM. Checking link count wasn't really in my workflow, and still isn't because there are not many of such discussions that I had closed.

While WP:RMRELIST is sufficient for most pages, but there are some which may bode well have better outcomes if the discussion was directed to more people. The banner on the article is placed by a bot, which means that many active registered editors may miss it if the option to exclude bot edits in the watchlist is enabled. Not many would go on WP:RMC to look for articles to be moved as well, despite the high pageview counts (I suspect that many of the regulars load the page multiple times in the day).

It would be good to have at a minimum that the link count of the pages requested for be displayed in the notice generated by the {{requested move/dated}} template. At the very least, closers may take a pause before closing the RM discussion early.

It would be nice to have a centralised notification somewhere on a highly trafficked discussion page (Open tasks at WP:AN?) to list some RM discussions based on number of links to the articles involved. Exact criteria can be fleshed out separately. – robertsky (talk) 02:00, 16 July 2024 (UTC)

I think creating bright-line prescriptive rules on this subject is likely to lead to undesirable bureaucracy rather than any actual improvements in the space. Rather than the arguments about "Topic X is prominent enough to have merited a longer RM" being resolved, I think they'd just change shape and instead manifest as "Topic X is prominent enough to have fallen under the Popular RM Ruleset". (I also share voorts' stance that these assertions are generally fairly weak. It's not practical to keep discussions open indefinitely for fear of someone missing out.) I'm also concerned that, if we create additional procedural hoops for RM closers to jump through, it would exacerbate the existing issues where controversial or complex discussions linger unclosed for weeks beyond their intended end dates.

By and large I think the status quo already works, but if there is consensus to implement a change, I think robertsky's suggested minimum options – displaying link counts on the banner, or having centralized notifications somewhere – would strike a decent balance between increasing visibility and protecting the closure process from overcomplication. ModernDayTrilobite (talk • contribs) 05:23, 16 July 2024 (UTC)

One factor that inflates link counts is that if a template containing links to articles is included on a page, then every article linked in that template shows up in "What links from here", even though the link is only in the template, not elsewhere in the article. Some templates have hundreds of links in them. In other words, link counts are not a particularly accurate way of judging how many links will need to corrected after a move. Donald Albury 12:51, 16 July 2024 (UTC)

Yes, the widespread (over)use of navbox templates has made "what links here" virtually useless for many types of articles, sadly. Johnbod (talk) 15:30, 16 July 2024 (UTC)

Then we should have tools to determine how many actual non-template generated incoming links an article has. BD2412 T 15:35, 16 July 2024 (UTC)

Actually, it would be more useful to have "What links here" ignore links in templates. Donald Albury 16:30, 16 July 2024 (UTC)

Ideally I think it should be an option for "What links here" to show or hide links present only in transclusions, and also an option to show only pages that link via transcluded templates (grouped by template) given that its possible that some links wont show as being from the template namespace if they only appear based on parameters. Thryduulf (talk) 16:39, 16 July 2024 (UTC)

@Thryduulf: Agreed. It is always best to have options. BD2412 T 16:43, 16 July 2024 (UTC)

There's a nice plugin here to find source links that are not part of the template. Check it out at User:PrimeHunter/Source links.js and add it to your Special:MyPage/common.js file. ~ 🦝 Shushugah (he/him • talk) 16:43, 16 July 2024 (UTC)

I think this falls under WP:IAR and WP:NOTBURO. If you're closing a move request where there have been half a dozen previous requests that closed the other direction, or you're closing a move request that requires editing tens of thousands of pages, it's common sense to wait for more input. Similarly, if you close a move request, and lots of editors come out of the woodwork and say they would've participated had they seen it, then it should be clear that the discussion didn't necessarily capture the wider community consensus and should be re-opened. WP:RMRELIST is a process page, not a policy or a guideline, and it shouldn't be treated as prescriptive. --Ahecht (TALK
PAGE) 17:50, 16 July 2024 (UTC)

Maybe it is "common sense" but it's not the rule, and closes that are valid under the rule should not be overturned based on some amorphous notion of common sense. Also, for a discussion that has already been relisted once, or has been moribund for days, how long should we "wait for more input"? What is the signal that it is enough? BD2412 T 17:56, 16 July 2024 (UTC)

When no more comments are being made, and or the arguments are exclusively repeating themselves, then the closer can determine the strengths of differentΩ arguments rather than popularity of said arguments. ~ 🦝 Shushugah (he/him • talk) 18:51, 16 July 2024 (UTC)

We have seen discussions where that is exactly the case, but where the discussion once closed is still collaterally attacked for having been closed with too little participation, because the change was to a high-profile topic. BD2412 T 19:01, 16 July 2024 (UTC)

@BD2412 What rule are you referring to? And even if there were a rule, a rule being overturned based on some amorphous notion of common sense is exactly why we have WP:5P5 as a pillar and WP:IAR as a policy. --Ahecht (TALK
PAGE) 21:12, 16 July 2024 (UTC)

I am referring to the operations of Wikipedia:Requested moves that are laid out on that page. There is a process for proposing a move, a set time period for discussion, and a process for resolution of what has been proposed. There is no allowance there for declaring an article to be special for purposes of carrying out this process. BD2412 T 21:25, 16 July 2024 (UTC)

There doesn't need to be an allowance because that page is descriptive, not prescriptive. It is neither a policy nor a guideline. --Ahecht (TALK
PAGE) 02:28, 19 July 2024 (UTC)

In my extremely limited experience, the problem lies less with how long RMs are open and more with whatever smorgasbord of reasons causing move review to be unpleasant to closers (and potential closers). Even in less controversial RMs, there's an unfortunately non-negligible probability of some individual going to closer's talk, failing to articulate which significant evidence or fact was overlooked or WP:RMCI clause violated (per MR's instructions), ignoring any response by the closer that isn't acquiescence, and dragging the closer to MR, where the discussion remains open despite the giant "this is not the place to relitigate RMs" sign on MR's front door. Even if I snapped my fingers and RMs, effective immediately, have to be open for at least three months (like the I/P-related RM open since April), someone's going to come after closure at 3.5 months and argue I don't agree with it and didn't get a chance to voice my opinion so clearly it wasn't open long enough and not enough of the right people were notified, etc. What we need is a culture of compliance with and better enforcement of existing rules, so that RM closers have the support of the community and its norms when closing controversial RMs. If we do so in a way that helps people who are actually following the MR instructions by allowing the community to focus their resources on closures that actually need fixing, even better. Rotideypoc41352 (talk · contribs) 05:29, 18 July 2024 (UTC)

I agree with all of this. If in fact MR closers are reopening RM discussions based on these kinds of weak/irrelevant arguments, than there's an issue with how consensus is being evaluated at MR that needs to be addressed with the regular closers there. voorts (talk/contributions) 20:49, 18 July 2024 (UTC)

We also need people to relax a bit about these things. Some high-strung vitriol comes out over what amounts to a trivial matter, and one that can always be subject to a new consensus-testing process in the future. BD2412 T 21:04, 18 July 2024 (UTC)

This is an evergreen comment. voorts (talk/contributions) 21:08, 18 July 2024 (UTC)

• The people posing the question do have the option of using RFC rather than RM, giving a (minimum) 30 day discussion period.—S Marshall T/C 07:52, 18 July 2024 (UTC)
  • @S Marshall: They have the option of using RFC, yes, but there is no requirement for this. It remains permissible to use Wikipedia:Requested moves to propose even a move that is bound to be highly controversial, for a high-visibility subject, with substantial implications for incoming links. The friction is arising where that process has been used in its normal way, and people disagree with the outcome because the article to be moved was of that higher level of controversy. BD2412 T 21:07, 18 July 2024 (UTC)

    I thought RM was the exclusive forum for requesting moves. I think there'd be several valid procedural objections for using an RfC instead of RM. voorts (talk/contributions) 21:11, 18 July 2024 (UTC)

    • I believe we have used RFCs before where it proved impossible to reach a consensus as to a move target, and a broader rule needed to be established for the handling of the entire class of cases. BD2412 T 21:16, 18 July 2024 (UTC)

      The ongoing Wikipedia:Requests for comment/Titles of European monarchs, which while not a move request in and of itself, will be used as the justification for multiples moves pretty much regardless of what the outcome is. Thryduulf (talk) 21:45, 18 July 2024 (UTC)

• I voted to relist at that MR, but for the broader procedural question, I think the status quo is fine. The healthiest use case of MR is actually precisely this - when there's a close that is totally valid by the rules, but Other Stuff (TM) came up, and whoops it's time to IAR it. There's no need to change policy - 98% of closes can continue at a fast pace, and MR can quibble on the 1% of bad closes, and the 1% of times where a close was good but New Developments means that the close should be overturned anyway. That seems like a healthy ratio. SnowFire (talk) 21:58, 18 July 2024 (UTC)
  • The express purpose of MR is to determine whether the closer acted incorrectly, and expressly not to introduce new variables that should have been introduced in the initial move request. I see too many complaints of move requests not being extended from editors who could have extended it themselves while it was ongoing, too many complaints of lack of notification from editors who could have done some notifying, and too many efforts to relitigate the underlying issue rather than evaluating whether the close was within policy. BD2412 T 22:14, 18 July 2024 (UTC)

    To analogize to the legal space, an appellate court generally won't address (or will give more deference to the trial court) issues that weren't raised in the court below. MR is designed to review whether a close is reasonable—a pretty low standard—not to force the reopening of long and contentious RMs. voorts (talk/contributions) 22:30, 18 July 2024 (UTC)

    • Er, editors can't extend it? A relist is the equivalent of a close, so someone who's !voted shouldn't relist, since if they could there could be a naysayer's veto of the losing side just spamming relists. (They can obviously make "please extend this comments!" but there's no guarantee that they'll be heeded.)
    • And yes, MR is not RM round 2, but I don't think that's the case here nor is it accurate to portray it as such. Especially in cases where the issue is simply turnout, there isn't really a great need for "finality" or some such (that's what RFCs for if an issue is truly radioactive). Take the broader question. A nominator proposes a move, and it's largely ignored. 2 editors show up out of nowhere on Day 7 and !vote support. A closer sees this and reasonably closes as non-controversial and moves it. 99.5% of the time, the story stops there, it was indeed non-controversial. 0.5% of the time, people involved in the topic come out of the woodwork and say "whoa, that's not right." There is very little harm in letting a courtesy relist happen, so that's what the nominator should do to such a good-faith request, and they should be taken to MR otherwise. Maybe nothing changes, but maybe 20 other editors in the field pop up to say that the move was a terrible mistake. Who knows. But there is 0 harm in at least exploring the possibility of the latter case. (The other option is giving leave to file a new RM, of course, but then that might annoy !voters in the previous RMs who feel like they have to copy-paste their arguments to ensure their !vote "counts.") SnowFire (talk) 22:57, 18 July 2024 (UTC)
      • @SnowFire: There is no limitation provided as to who can relist a discussion, nor is there any policy that says that "a relist is the equivalent of a close", so, yes, any editor can relist the discussion, including one who has !voted. I think it's fairly obvious that an editor who spammed relists would get called out on that, and it is not a behavior that has occurred in practice. BD2412 T 23:04, 18 July 2024 (UTC)
        • That's not the case. See WP:RELIST: "Editor qualifications to relist a discussion are the same as required to close a discussion." Someone who has !voted can't close, so therefore they can't relist either. I agree that the behavior does not happen in practice, but that's because it's prohibited, and we should not encourage involved editors to perform a relist - they are not a neutral source for if a discussion has sufficient consensus for a close. (Oddly enough, a position I feel is closer to yours, in the name of making speedier closes!) SnowFire (talk) 23:09, 18 July 2024 (UTC)
          • Well, I've learned something new, then. As far as I recall, I have relisted discussions in which I was a participant on several occasions, usually because someone made a new proposal that was worth additional time to consider. I have never been criticized for so doing. BD2412 T 03:08, 19 July 2024 (UTC)
            • I'm sure your relists were fine, but point is we also shouldn't hold it against one side that they didn't force a relist themselves, because technically they shouldn't do that. SnowFire (talk) 10:26, 19 July 2024 (UTC)

No. Phil Bridger (talk) 22:59, 18 July 2024 (UTC)

Yes. High-profile, or previously contested, or technically complicated cases, all need a Rule that the next random nominator must point to and summarise the previous RMs, and mention technical issues.

The bad things to be prevented are a surreptitious move in a holiday period, or exhaustion of opponents allowing win by tendentious teamwork, or a consensus based on inadequate information.

Rules should be documented, YES!

Inadequate nominations should be speedy closeable for being inadequate. SmokeyJoe (talk) 04:11, 19 July 2024 (UTC)

I wonder if the rule that would be most useful might sound more like: If a consensus forms to move the page, and the result will require updating links in other articles, then no more than ____ articles may be updated per day (perhaps 50 or 100 per day, for the first week).

If you move the page, and the redirect stays in place, and someone complains (bad close, new information – the reason doesn't matter) and those complaints result in reversing the decision, then it's no big deal. You move the page back, and you're done.

If you move the page, and the redirect doesn't stay in place (e.g., gets converted to a DAB page), and the decision gets reversed, then you may need to update all of the linked pages, which can be a huge, watchlist-flooding task. Also, it's more likely to happen, because the watchlist flooding will attract attention and irritation to the recent move request. WhatamIdoing (talk) 21:33, 19 July 2024 (UTC)

Extent of WP:CHANGELOG

@HyperAccelerated entirely removed the "Version history" section of KDE Plasma. Upon being informed that DragonFly BSD had a similar section, they also removed it from that article, citing WP:CHANGELOG in both cases.

The policy specifically states against an "exhaustive" log and additionally encourages third-party sources. In both cases, third-party sources were used in addition to first-party sources. I believe that in general, the two version history sections were useful, although perhaps they did not need to be as detailed as they were.

How should this policy be applied? Blanket removal of the two sections seems a bit of an overreaction to me. iczero (talk) 16:33, 18 July 2024 (UTC)

The section that was on KDE Plasma strikes me as a good example of the kind of thing that WP:CHANGELOG is about - I would support that removal. It was an list of every version, many with little or no information on the contents. And the article also has prose sections on the major releases separate from the table, which serves as the 'summary of development' which WP:CHANGELOG suggests should be considered instead. MrOllie (talk) 16:39, 18 July 2024 (UTC)

@Iczero, I think the main problem is that different editors have different ideas of what "exhaustive" means, and that you would probably get a more practical answer from Wikipedia talk:WikiProject Computing.

One idea that I've had (wrt to previous disputes, because this happens several times a year) is that it would be helpful to have one or two Wikipedia:Essays that illustrate it. I imagine, for example, someone finding a freely licensed changelog for a notable open-source product, pasting the whole thing into the essay, and then saying "This is what exhaustive looks like" (a section that repeats the original almost in its entirety) and "This is what an encyclopedic summary looks like" (a good summary, with third-party sources) and perhaps one or two other examples (one that's borderline and one that's far too short to useful?).

In the meantime, it's possible that the short list at Wikipedia:Featured articles#Computing would contain a good example. WhatamIdoing (talk) 21:38, 19 July 2024 (UTC)

People like to cite WP:ALLCAPS links as justification for all sorts of things, including giant slash-and-burn edits that take out half an article.

WP:CHANGELOG was deliberately written to preserve articles like Android version history. The people who wrote it explicitly said they intended the policy to not be applied to those articles, or to prohibit articles from mentioning version histories.

The text was added sua sponte by this edit in February 2011, discussed at this talk page section, and its current wording arrived at by Wikipedia_talk:What_Wikipedia_is_not/Archive_45#WP:NOTCHANGELOG (consensus to retain here). At no point was the intention ever to remove them entirely -- again, the sole focus of the discussion that wrote the policy was how to preserve Android version history and articles like it. Given this, it seems like a rather tortured by-the-letter reading to invoke it while slashing out tens of kilobytes of text. jp×g🗯️ 12:19, 23 July 2024 (UTC)

Those past discussions and the current wording also requires the use of third-party (not self-published or official) sources which all these list fail as the use only the official sources, so they at least need improvement. -- LCU ActivelyDisinterested «@» °∆t° 12:31, 23 July 2024 (UTC)

Sure -- they aren't great -- but I don't think it's a reasonable basis for wholesale removal. How many versions are there listed in the table? Here are my two-minutes-and-no-flashlight pulled out of my keister search results:

References

jp×g🗯️ 13:23, 23 July 2024 (UTC)

Agree with MrOllie that KDE Plasma is definitely a good example of "bad" changelogs. The vast majority of it isn't cited to secondary sources and it reads as an exhaustive list of minor changes rather than high-level things, violating summary style guidelines to boot. As for "NOTCHANGELOG isn't supposed to apply to these articles" suggested above, NOTCHANGELOG applies to all changelogs. There's no carveout exemptions. You can argue it was crafted to make articles like Android version history compliant, but no articles get a pass from our content rules. Der Wohltemperierte Fuchs ^talk 12:33, 23 July 2024 (UTC)

Taking the temperature on NACs in CTs

This is not a formal proposal, but may be a precursor to one. WP:NAC is generally cited as the roadmap for non-admin closures. It cautions non-admins against closing potentially contentious discussions, but does not prohibit them. It is also only an essay; but the relevant policy pages that I skimmed do not appear to make distinctions between admin and non-admin closures.

We have a good few experienced non-admin closers. Their decisions are, best as I can tell, not inferior to admin ones. However, based on the caution against contentious closures by non-admins in WP:NAC, I believe they are challenged far more frequently, and consequently their closures often end up costing, rather than saving, the community time (if this comes to a formal proposal, I will do the archival research needed to show this).

I'd like thoughts on a) whether we should prohibit non-admin closures in contentious topics, as a means of saving community time on close reviews; and b) what the best way to do this would be, given that WP:NAC does not currently carry formal weight. Regards, Vanamonde93 (talk) 21:57, 18 July 2024 (UTC)

Notified: Wikipedia talk:Non-admin closure. voorts (talk/contributions) 02:59, 20 July 2024 (UTC)

A few thoughts.

First, there are levels of contentiousness, with ARBPIA at one end and ARBBLP at the other. While non-admin closures may be more likely to be challenged for the former to the point of costing community time, I am certain that is not true of the latter.

Second, I am certain this does not apply to request moves. As a NAC, I close a lot of RM's, and as expected given the volume a number are challenged. I haven't found it any more likely that those I close within contentious topics are challenged than those outside, and given the number closed to the number contested I am certain that these closures have saved community time. BilledMammal (talk) 22:11, 18 July 2024 (UTC)

I generally think we should be offloading bureaucratic workload from admins, not piling more on. If there are specific subject areas that become magnets for poor NACs, existing processes are sufficient to curtail those without instruction creep. VQuakr (talk) 22:17, 18 July 2024 (UTC)

As a regular NAC, I'm not sure NACs are challenged more often than admin closures; even if they were, close reviews are relatively rare (for example, as of December 2023, there was an average of 2 close reviews per month as discussed here). More importantly, we shouldn't be taking editor's powers away just because some editors spuriously choose to challenge closures solely on the grounds that they were done by non-admins. As for NAC, I read that just as you do—as a word of caution, not as a command. voorts (talk/contributions) 22:35, 18 July 2024 (UTC)

Thinking out loud, but if NACs are costing the community time in contentious areas, I think it would be better to have a new "userright" (for want of a better term) called "discussion closer" that gives users who have it no extra tools but the same weight in closing discussions as an admin has. Such a right would need to be conferred in a process only slightly heavier weight than file mover - I'm thinking something like request open 2-5 days, consensus in a discussion that has at least 5 supports from admins and/or discussion closers (no consensus after 5 days = not granted). We would then recommend that discussions that are or which are likely to be contentious be closed by admins or discussion closers.

NAC would be explicitly not a permissible ground on which to challenge a closure by a discussion closer - if that's the only reason given the challenge would be speedily declined, if it was accompanied by other reasons then the portion of the statement relating to being an NAC would be struck.

Actually, even without the discussion closer status, speedily declining any review request where NAC is the only ground would be a good thing. Thryduulf (talk) 22:44, 18 July 2024 (UTC)

speedily declining any review request where NAC is the only ground would be a good thing – This, 100%. As Vanamonde noted, there's no actual policy or guideline that says we give admin closures more weight. Indeed, per WP:NOBIGDEAL and WP:ANOT, being an admin doesn't give anyone special authority over content decisions or determing consensus. To the extent that people read WP:NAC as implying that admin closes are better than NACs just because the closers have a mop, it ought to be clarified. I'm against the new "userright" because I don't like the idea that some editors determinations of consensus are weightier than others. voorts (talk/contributions) 22:55, 18 July 2024 (UTC)

@Voorts: I agree with you about the admin-non-admin distinction, but in general, there most certainly are editors who shouldn't be doing NACs. We need to allow genuinely bad closures to be reversed: we also don't want the closer's status as a non-admin to become a distraction. I'm open to other ideas on how to achieve that. Vanamonde93 (talk) 01:04, 19 July 2024 (UTC)

In such cases, the review request needs to say "this summary does not accurately/fully represent the outcome of the discussion" rather than saying "the wrong kind of person wrote this". WhatamIdoing (talk) 21:41, 19 July 2024 (UTC)

I think making it clear that only very experienced closers should close complicated or contentious discussions is sufficient. Relatedly, one of the things I look for at RFA is difficult closes the candidate has made. ScottishFinnishRadish (talk) 23:35, 18 July 2024 (UTC)

As a somewhat irregular NAC-er focussed on AfDs, I can appreciate where this is coming from - although my (purely anecdotal) observations of DRV would say this is not, relatively speaking, a problem in that sphere. I think the question is less about status (admin or not) and rather experience; I wouldn't necessarily be opposed to setting some thresholds (edit count, participation etc) for NAC on CTOPs, but I don't see a strong enough case yet for exclusion. Regards,--Goldsztajn (talk) 23:29, 18 July 2024 (UTC)

Like anything else on the wiki, the qualification to do most things is that you know how to do it, and having a mop is no promise that you do. I'm sure most of the regulars at WP:AfD, admin or not, know the details of our notability guidelines better than I do; it's absurd to suggest that I'm better qualified to close a complicated AfD just because 19 years ago, 27 people thought I'd be an OK admin. Our current NAC mindset is an anachronism and should be done away with. RoySmith (talk) 23:54, 18 July 2024 (UTC)

• We should root out and destroy every suggestion that discussion closure is an admin task. It makes sense for the admin noticeboards, and is vestigial in most other places. I like Thryduulf's user right suggestion (I know others have suggested something similar in the past) and speedy close suggestion, though I've rarely seen a situation where NAC is the only objection. We should guide newer closers to less controversial discussions, and we should explicitly indicate that experience multiple discussions is necessary for closing contentious, major discussions. We should still allow for challenges based (in part) on lack of experience, it's just that many non-admin closers are much more experienced than almost all admins. Firefangledfeathers (talk / contribs) 01:14, 19 July 2024 (UTC)

  This isn't any different from any other process. I spend a lot of time at DYK. For all the chaos that goes on there, there's an effective culture of new people being groomed to take on greater responsibility. You start out by doing your obligatory initial reviews and move on to more complicated things like building prep sets. People inevitably make mistakes, the mistakes get fixed, experience is gained, and the cycle continues. WP:GA works the same way. And WP:FA. And dozens of other nooks and crannies of the wiki where just plain editors sans mop keep everything going. As it should be. When somebody's been working in an area for a long time, they become an expert at it. The idea that some random admin who's never worked in that area could possibly do a better job is absurd. RoySmith (talk) 01:37, 19 July 2024 (UTC)

  I agree with Roy.

  @Firefangledfeathers, even a "major" discussion on an officially Contentious Topic™ can sometimes be easy and uncontentious to summarize. The ideal result of an RFC is that everyone already knows what the outcome is. A given participant might be inclined to privately summarize that outcome as "The community is a bunch of jerks who'll be the first up against the wall when the revolution comes", but even the most passionate editor on the "losing" side can often recognize when a consensus has been reached for the "wrong" result. In such cases, we don't necessarily need a highly experienced editor to state the obvious. WhatamIdoing (talk) 21:53, 19 July 2024 (UTC)

  Sure. I meant contentious in the non-trademarked sense. Firefangledfeathers (talk / contribs) 22:19, 19 July 2024 (UTC)

• Outside of deletion discussions and other outcomes that need an admin to carry out (where it absolutely does make sense), I view invoking BADNAC as essentially scope creep. You don't need to be an admin to close RfCs, at all. But I don't think it's fair to dismiss people who bring it up as baseless wikilawyers either. Usually what they're trying to allude to is that contentious discussions are hard to close and therefore that the community expects someone with experience in making successful closes to do it. This is a good and widely agreed upon principle, but it's not written down with a handy shortcut, so BADNAC gets invoked instead. If we articulate that broader principle somewhere, I think we'll see BADNAC cited less often. – Joe (talk) 07:32, 19 July 2024 (UTC)

  Outside of deletion discussions and other outcomes that need an admin to carry out (where it absolutely does make sense). There's no fundamental reason why a non-admin can't close an AfD as "delete" and then find an admin to actually push the button. In fact, it looks like {{Db-xfd}} covers exactly this use case. This is similar to how non-admin SPI clerks can determine that an account is a sock and should be blocked and then have to go find an admin to do it for them. RoySmith (talk) 14:57, 19 July 2024 (UTC)

  No fundamental reason, no, but why create the extra work and complexity when we have no shortage of admins willing to close AfDs? Any process that requires admin intervention should be left to admins unless and until it becomes obvious they need the extra help (as with SPI), as matter of efficiency if nothing else. – Joe (talk) 15:04, 19 July 2024 (UTC)

  I think that's over-broad. I think you shouldn't have to be a sysop to close an RfC that's about making a change to a fully-protected page, for example.—S Marshall T/C 18:11, 19 July 2024 (UTC)

  Pressing delete from a closed AfD feels like a situation where an admin would need to verify consensus in the first place and so now you've spent the time of two editors where one could have done. Beyond that, I am pretty staunchly opposed to admin close creep in places like RfCs. When I was a regular at AfD, I found non-admin work to be far less "right" than with RfCs; I think it attracts a different kind of non-admin. Best, Barkeep49 (talk) 04:02, 20 July 2024 (UTC)

  @Barkeep49 your point about the button-pusher needing to verify the result is valid, and indeed I've made similar arguments myself. But a good close can make that job a lot easier. A good close won't just say "Consensus to delete". It'll summarize the main points made on both sides, list the minority opinions, and talk about which arguments were rejected by other discussants (or by the closer) and for what reasons. With a good analysis like that, you can get your head around the discussion without having to read every word. And, yes, the button-pusher is ultimately responsible for their actions, and I assume all responsible button-pushers will dig as far as they feel is necessary to validate the summary. RoySmith (talk) 15:55, 21 July 2024 (UTC)

  Most AfDs don't require a long closing statement or often don't require any closing statement. This lack of need for closing statements is a way that AfD is different from RfCs. This also doesn't change my point - it's not a good use of editor time to close something which will require substantial re-verification to implement (outside of processes like DYK which are designed to have these multiple levels of checking). Best, Barkeep49 (talk) 17:15, 21 July 2024 (UTC)

  It's true that "Consensus to delete" is an adequate close for many AfDs. I would expect somebody to write the kind of detailed analysis I outlined above only for discussions that warranted it due to their complexity. RoySmith (talk) 13:35, 22 July 2024 (UTC)

  now you've spent the time of two editors where one could have done is true, but we don't stop people from wasting their own time. The admin would have to spend time to process the deletion regardless of whether it was NAC'd first or not. So in the NAC scenario, the only person who's time is arguably wasted is the NAC who volunteered their time to do this, and if that's how people want to spend their time, I don't see why policy should prohibit them from doing so. Arguably, two sets of eyes is a benefit anyway, so it's not necessarily wasted time at all. Levivich (talk) 23:51, 21 July 2024 (UTC)

• I would distinguish between a discussion close that's a content decision, and one that's a conduct or technical decision. There are philosophical and principled reasons why we absolutely must not give sysops special authority to make content decisions. Conduct decisions in CT areas, on the other hand, are best reserved for sysops even where an unelected person could make them.—S Marshall T/C 07:58, 19 July 2024 (UTC)
• There are non-admins (like S Marshall) who would be in the top 10% of admins regarding closures (even if I disagree with one) if they were one. And vice versa. It's just that the odds and optics are better when it's an admin. I think that the current guidance on this is about right. North8000 (talk) 15:46, 19 July 2024 (UTC)
• There a many inherent reasons for people not to close long, complicated, or contentious discussions, so I see the lessening of the pool of willing closers as throwing the baby out with the bathwater, so to speak. Alanscottwalker (talk) 16:06, 19 July 2024 (UTC)
• It cautions non-admins against closing potentially contentious discussions, but does not prohibit them. WP:NAC should do more than caution, but still should not prohibit. It should advise against NAC closes of contentious discussions where consensus is not abundantly clear.

We have a good few experienced non-admin closers. Absolutely. Adminship is not a requirement for being a good closer, but being a good closer is something tested at RfA, or at least, not being a bad closer and knowing your own strengths is tested at RfA.

Some challenges to NAC closes may be unfair, but this depends on perspective. It is a fact that many ordinary Wikipedians do not consider a non admin close of a contentious discussion to be a satisfactory close. This is not a reason to slap down ordinary wikipedians, but for non admins engaging in advances functions to do it more conservatively. A good skillful close should make a contentious-looking discussion look no longer contentious.

If a non admin's close of a discussion produces another, longer, more contentious discussion, then their close was not a net positive contribution, and they should not do such closes.

We should advise non admins to not close contentious discussions unless they are very confident that they will explain their close to the satisfaction of all the participants. Alternatively put: If an admin is confident that they can close a discussion to the satisfaction of all participants, then they should be encouraged to do so. Despite being very confident, non admins are allowed to be wrong, sometimes. Don't make a habit of it. If a challenge to their close surprises them in any way, then strongly consider reverting the close and listing it at Wikipedia:Closure requests. Then, sit back and see if someone else closes it the same way.

In any discussion, the closer should be the least important person, not the most important person.

All of the above should apply equally to XfDs, RM, and RfCs. It should apply moreso to closes at AN, DRV, MRV and XRV.

Spurious challenges should not be feared. Spurious challenges are characterised by a SNOW endorse at review.

I don't think a special user-right for closing is warranted. If credentials are wanted, I suggest a category of barnstars for good closes.

--SmokeyJoe (talk) 08:40, 21 July 2024 (UTC)

Amending BADNAC?

I want to be very clear that when I opened this it wasn't because I felt some NACs were inappropriate, but because I wanted to avoid spurious challenges, or challenges where non-admin status was muddying the waters. It's fairly clear that there is strong support for not limiting NACs; so what do folks think of an alternative approach to address the problem I mention, and making BADNAC contingent strictly on experience rather than admin status: that is, essentially striking BADNAC#2, and perhaps strengthening the reference to experience in BADNAC#3? Vanamonde93 (talk) 02:45, 20 July 2024 (UTC)

I am in favor of striking #2. I think #3 could be struck too; if somebody inexperienced is really good at evaluating consensus and has read Wikipedia's PAGs, I don't see why the close ought to be overturned on those grounds alone. But, I can live with #3 as it's currently if there's consensus that that kind of limitation should be in BADNAC. Also, I've notified WT:NAC of this discussion. voorts (talk/contributions) 03:04, 20 July 2024 (UTC)

The likelihood that someone using an account with few edits will do a passable job on anything except the most obvious cases is low. Also, it gives anyone on the "losing" side an opportunity to suggest that the newbie is a bad-hand sock, which is more drama that we would like to avoid.

BTW, "experienced editor" appears to be a label that there are different views about. @Levivich and I were chatting about this at Wikipedia talk:Wikipedians#Higher volume: Can someone who has "only" been editing for a year (the median account activity is one day), with "only" 500 edits total (more than 99.25% of accounts that have ever made a first edit), averaging "only" one edit per day during the last month (less than 10% of currently active accounts), be truly considered "an experienced editor"? If you'd like to provide a third opinion (or fourth, or fifth), please share your idea of what the minimum standard for "an experienced editor" could be over there. WhatamIdoing (talk) 05:14, 20 July 2024 (UTC)

It sounds like we need to temper our expectations regarding how we treat newcomers on Wikipedia, instead of limiting them because others might have bad faith objections.

As for a rare but pertinent counterexample, I noticed Chrhns's close of an RFC within their first 50 edits. It was well reasoned, not "exceptionally obvious" and pretty much the exact close I would make too. I did end up suggesting they edit other parts of the Wiki first, simply because I know how contentious challenges can get. But should they (or editors like them but with 500 more edits) be restricted from one part of the encyclopedia just because they read the rules before they start editing?

I think it's far more important that close challenges cite an actual policy being broken instead of just BADNAC. If a close is flawed, it will be flawed on multiple grounds. Soni (talk) 11:23, 20 July 2024 (UTC)

Agree in general, CTs excepted, non EC editors cannot close or even participate in internal project discussions. Otherwise I don't object to NACs in principle, if they are messed up, as some will be, we have the procedures to deal with that. Learning by doing is not a bad thing. Selfstudier (talk) 11:30, 20 July 2024 (UTC)

Learning by doing is how Wikipedia works. WhatamIdoing (talk) 17:02, 20 July 2024 (UTC)

Your observation that they get challenged more often could be right and a reason to advise non-admin to be careful (thus keeping it included as advice at NAC) without doing the wrong thing of making that advise a prohibition, which is what people are objecting to here. I think some data more than than the philosophical discussion above could be useful in making this kind of decision. Best, Barkeep49 (talk) 04:04, 20 July 2024 (UTC)

I will attempt to compile data in a few days. I think the problem exists regardless of frequency, however. A close challenge in which the closer's admin/non-admin status has become a factor is, I think, a priori a bad use of the community's time. Evaluations of closures need to focus on other things. Vanamonde93 (talk) 01:33, 21 July 2024 (UTC)

I oppose simply striking #2 ("The outcome is a close call (especially where there are several valid outcomes) or likely to be controversial") but suggest instead rewording it. As it reads, I don't think it is very good. I suggest, it get ideas started,

"The discussion is contentious and your close is likely to be controversial."

I think it is a good idea to put the judgement of the appropriateness of the NAC in the control of the NAC-er, and point to "the close" as the thing that will be judged. (The number of valid outcomes parenthetical is wordy verbosity)

"#3 The non-admin has little or no experience editing Wikipedia generally or has little or no previous participation in discussions." is good and important. It doesn't require touching. It is important for newbies. To make it easier for newbies to understand, I would suggest closers should have one year experience editing Wikipedia, and 500 edits in projectspace, and 100 AfDs participated before closing AfDs.

- SmokeyJoe (talk) 08:54, 21 July 2024 (UTC)

Nobody can ever know in advance if their close will be controversial. Sometimes people get upset over the silliest issues. voorts (talk/contributions) 17:39, 21 July 2024 (UTC)

A few days ago, I said this, and I think it's apt here too.—S Marshall T/C 20:57, 21 July 2024 (UTC)

Critique or criticise the close, not the closer, absolutely yes, start there. Where the same closer repeated has their closes criticised, maybe there is a pattern of evidence to suggest a change in behaviour.

NAC-ers should be advised to be cautious in closing, not prohibited in closing. NAC-ers should be advised on how the can best help. On a quick review of old-admin closes, I think you'll find they tend to be terse. A newcomer may think that a good close is a terse close. SmokeyJoe (talk) 00:50, 22 July 2024 (UTC)

nobody can ever know in advance if their close will be controversial? Nonsense. Wrong. If you can't tell that the discussion is contested, with heated participants, and that your close does not address their positions, then you should not be closing. Someone^TM getting upset over something silly is life, not controversy. SmokeyJoe (talk) 00:45, 22 July 2024 (UTC)

A topic of discussion can be controversial, but the outcome can sometimes be so obvious that the closure isn't challenged. For example, the recent RM for Gaza genocide. voorts (talk/contributions) 01:49, 22 July 2024 (UTC)

I am in favor of striking #2 and strengthening the reference to experience in #3. I think experience, not the admin bit, is the strongest predictor of good closes. Levivich (talk) 23:56, 21 July 2024 (UTC)

As I see it, BADNAC serves a couple of purposes. Preventing bad quality closes by telling editors when they might not be appropriate before they attempt to close. Providing the outcome of a close a degree of "authority" against challenges from without or within by setting some minimum standards and allowing closes procedurally to be set aside regardless content. I see the second as being less important than the quality of the close but in terms of optics, if the DAILYMAIL close was made by a 2 day account it would not have had the same weight. Therefore I do see some value in retaining a version of the current #2 somehere in BADNAC which sets a higher bar in terms of required experience for something controversial or complex than a simple snowclose. In terms of how that experience is defined, it should be related to actually closing discussions, not just general editing, and admin status should not be relevant. Scribolt (talk) 08:26, 22 July 2024 (UTC)

Stricter policy for unregistered ipv6 users vs ipv4 users

It is well established that unregistered (ip address) users are an important part of the Wikipedia community. However, a clear distinction can be made between ipv4 and ipv6, as the latter can be easily manipulated to create sock puppets and obscure other bad faith actions. The nature of ipv6 makes tracking, blocking, banning nearly impossible.

But the same is not true of ipv4 users, where the limited address space is a reasonable restraint, vs the virtually unrestrained nature of ipv6 unregistered users. Ipv6 creates a dense forest of ip addresses that bad actors can easily use to hide within.

I have personally experienced certain users (who are otherwise registered) logging out, and using a number of ipv6 addresses to harass and/or make bad faith edits. If unregistered users were restricted to ipv4 addresses, it would maintain access for unregistered users, while usefully hampering bad actors rolling through the nearly unlimited ipv6 addresses.

Of course, there should be no restriction connecting by ipv6 for registered user accounts.

Summary

Restricting only ipv6 unregistered users improves admin control, while not unduly hampering good-faith unregistered users' access to edit.  Myndex talk   09:27, 20 July 2024 (UTC)

Most people have no control over whether they use ipv4 or ipv6 addresses, and many who might have control won't know this. For most, the restriction would be completely arbitrary. There is also not a lack of harassment from ipv4s. CMD (talk) 10:01, 20 July 2024 (UTC)

Seems to unfair and unnecessarily discriminate against ipv6 users, who likely don't know they're ipv6 users, and can't control it. I don't know whether my IP is "ipv4" or "ipv6". I have noticed that some IP users are just numbers, like this: 40.7.211.792 or something, while some are more like 2f:00e::56a8:9000s. I assume that's the difference. Cremastra (talk) 10:22, 20 July 2024 (UTC)

You seem to have got the right idea. IPv4 addresses (like your first example) are usually expressed as 4 decimal numbers in the range 0-255 and are limited to about 4 billion in number, and IPv6 addresses are expressed in hexadecimal (base 16 with A-F representing the digits 10-15) with, in theory, over 10³⁸ addresses available, which means that they will "never" run out, in the same sense that IBM mainframes didn't run out of accessible memory with the increase in addressibility from 24 bits to 31 in the 1980s. Of course there's no need for the average encyclopedia writer to know or care about any of this. Phil Bridger (talk) 15:00, 20 July 2024 (UTC)

@CMD The user doesn't need control—any user's agent will use either ipv4 or ipv6, depending on the server connecting to. Wikipedia's servers could ask for ipv4 if an anonymous editor wanted to make an edit. And I am not suggesting banning nor blocking ipv6 addresses—only suggesting that ipv6 addresses be required to either log in or create an account. Perhaps temporary accounts might help.

An objective here is to prevent anonymous sock puppets. Perhaps there is a better method to require logging into some registered account from certain ip addresses. Myndex talk   10:04, 22 July 2024 (UTC)

• That's not how IPv6 users are treated, though. In effect, most IPv6 are allocated a /64 range of addresses (for example ABCD:EF12:3456:7890::/64) which is functionally equivalent to a single IPv4 address. When admins block or restrict IP users, they generally do this to the whole /64 range, not the individual IP (here's my blocking log, note the difference). Also note that we can extend that /64 to a wider range if necessary (there's a /32 in that list near the top). Therefore the differences between IPv4 and IPv6 are not a major issue. Black Kite (talk) 16:19, 20 July 2024 (UTC)

WP:LOUTSOCKing is a major pain for good faith IP editors, as it instills distrust towards them. But I don't see how tarring all IP editors using IPv6 addresses helps. Someone switching between different IPv6 /64 ranges is no different from someone switching between IPV4 addresses, and as Black Kite says there are methods for blocking ranges. -- LCU ActivelyDisinterested «@» °∆t° 16:26, 20 July 2024 (UTC)

Okay, but in practice ABCD:EF12:3456:7890::/64 does not seem narrow enough due I assume to DHCP/dynamic allocation at the ISP. YEs, this is still an issue with ipv4, but at the moment it doesn't appear that Wikipedia is grouping edits by only the first 64. And to be clear, I am not proposing that any ip address be blocked per se, only that certain IP addresses/ranges can be restricted to require login.  Myndex talk   09:46, 22 July 2024 (UTC)

In general on all but the busiest ranges you can go as wide as /48 and it will still be the same editor, so more than one editor in a /64 is unlikely. Anything wider than /48 and it's probably more than one editor. You can group edits simply by adding /64 or such to the end of the IP address in the contribution screen, it's not automatic but it only takes a second.

If IPv6 editors are required to create an account then they are blocked from editing as an IP. -- LCU ActivelyDisinterested «@» °∆t° 01:21, 23 July 2024 (UTC)

Isn't this all going to become redundant anyway when WP:Temporary accounts become a thing? Thryduulf (talk) 16:37, 20 July 2024 (UTC)

This (IPv4 v IPv6) isn't an issue IMO, but no it won't really, since any reasonably experienced editor (6 months tenure and 300 edits) will be able see the IPs behind the temporary accounts anyway. Black Kite (talk) 23:49, 20 July 2024 (UTC)

Inappropriate use of templates?

Is there anywhere on Wikipedia that details inappropriate use of templates? I have the feeling that {{rut}} and {{rus}} only exist to replace simple links, but I can't find a policy that outright forbids this. If I were to send those templates to WP:TfD, would they likely survive deletion? – PeeJay 14:08, 22 July 2024 (UTC)

{{rut}} and {{rus}} are redirects to other templates. Are you talking about these redirects (saving some typing vs their full {{Rugby union team}} and {{Rugby union stadium}} names) or those underlying actual templates? DMacks (talk) 12:58, 23 July 2024 (UTC)

I don't know where else to ask this & this might be a hornets' nest but here goes...

Recently there has been a spate of changes & reverts to the articles Martha Jefferson, John Wayles, Ulysses S. Grant National Historic Site, Sally Hemings and other articles concerning the usage of the terms "enslaved"/"enslaved persons" instead of "slaves". It is my understanding that "enslaved" & its associated terms are considered correct in modern usage as they describe a condition that could be considered reversible while the terms "slave"/"slaves" describe a person who is owned by another as that state of being and only as that state of being, plus the term being considered pejorative, etc. Do I have to open a Rfc for general usage of the terms "enslaved" to see if this is really the editorial consensus/Wikipedai accepted usage or whatever? It seems obvious to me that enslaved is preferred/accepted but there is disagreement over the usage. Not sure what to do and gawd help me RfCs can be such sinkholes of time & energy but then again I personally dislike long-simmering back&forth word-wrestling... Thanks, Shearonink (talk) 13:40, 23 July 2024 (UTC)