Jump to content

Wikipedia:Village pump (idea lab): Difference between revisions

From Wikipedia, the free encyclopedia
Browse history interactively
← Previous editNext edit →
Content deleted Content added
VisualWikitext
Line 967: Line 967:
My suggestion is for an icon (Chain or Fetters) on these types of articles:
My suggestion is for an icon (Chain or Fetters) on these types of articles:


• biographies: slavers, slave owners, slave traders.
• [[modern era|modern]] biographies: slavers, slave owners, slave traders.


• Monuments (statues) that are of one of the above
• [[modern era|modern]] Monuments (statues) that are of one of the above


• Monuments (buildings) whose construction was financed in part or wholly by slavery
• [[modern era|modern]] Monuments (buildings) whose construction was financed in part or wholly by slavery


• Institutions who were funded in part of wholly by money originating from slavery, slavers, slave owners, slave traders
• [[modern era|modern]] Institutions who were funded in part of wholly by money originating from slavery, slavers, slave owners, slave traders


That the icon be placed at the top of the article.
That the icon be placed at the top of the article.

Revision as of 23:14, 27 October 2023

 Policy Technical Proposals Idea lab WMF Miscellaneous 
Shortcuts
The idea lab section of the village pump is a place where new ideas or suggestions on general Wikipedia issues can be incubated, for later submission for consensus discussion at Village pump (proposals). Try to be creative and positive when commenting on ideas.
Before creating a new section, note:

Before commenting, note:

  • This page is not for consensus polling. Stalwart "Oppose" and "Support" comments generally have no place here. Instead, discuss ideas and suggest variations on them.
  • Wondering whether someone already had this idea? Search the archives below, and look through Wikipedia:Perennial proposals.

Discussions are automatically archived after remaining inactive for two weeks.

« Archives, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61

Unreferenced articles

Since I don't like writing long walls of text, and nobody reads those anyways, I'll state my idea in point form:

  • Unreferenced articles (i.e. those with zero sources) are a huge problem, and a violation of one of Wikipedia's core content policies: WP:V.
  • There are currently 119,000 articles tagged as unreferenced; and new ones get added every month.
  • With articles needing more sources, one can invoke WP:BURDEN and trim off the unsourced claims. One can't do that with totally unreferenced articles.
  • So here's my idea: "unreferenced" tags act as pseudo-prods. If the tag isn't removed within two weeks (and the problem fixed), the article is deleted. Users can draftify or improve the article during that period. Tags should not be removed unless references are added.
    • Don't panic! This idea will not apply retroactively; i.e.; articles already tagged with {{unreferenced}} would not be subject to this change.
  • This would greatly discourage creation of new, unsourced articles, of which there are too many.
  • This would stop the addition of new articles to the unreferenced backlog, allowing that to be picked away at.
  • This would maintain a higher average quality of verifiability and quality on the encyclopedia.
  • This would incentivise users to add sources to their unreferenced contributions.

I have a feeling this is idea will be drastically unpopular, but I believe something should be done, so... thoughts on the proposal? Since this is the idea lab, we can modify the idea to whatever directions necessary. Edward-Woodrow :) [talk] 18:17, 17 September 2023 (UTC)[reply]

P.S.: Please ping me on reply. Cheers, Edward-Woodrow :) [talk] 18:24, 17 September 2023 (UTC)[reply]
  • As long as this is not applied retroactively, as described in the proposed idea, this doesn't seem like too problematic of an idea, although it's not within current policy. WP:MINREF states: Technically, if an article contains none of these four types of material, then it is not required by any policy to name any sources at all, either as inline citations or as general references. That's an information page, but WP:V has always stated that All content must be verifiable. I think it was last year User:Levivich opened an RFC at VPP or somewhere proposing this language be changed to All content must be verified; the RFC did not pass. I think the present idea might be forcing through an unsupported policy change. If there is sufficient support for this, I would prefer an action less drastic than deletion, but as evidenced by other discussions currently active at VPR and VPI, something like draftification is a whole can of worms inside a barrel of worms inside an intermodal shipping container of worms. Folly Mox (talk) 19:18, 17 September 2023 (UTC)[reply]
I did consider proposing draftification, but abandoned it for pretty much that reason. I do frequently draftify articles myself, but I think institutionalized, automatic draftification would lead no where good, especially as long as G13 is around and kicking. Edward-Woodrow :) [talk] 19:21, 17 September 2023 (UTC)[reply]
What if, instead of leaping all the way to deletion, articles thus tagged, after 14 days, were __NOINDEX__ed? That's just a tiny baby step towards a fully referenced mainspace, but mitigates propagation of potential misinformation, and retains the problem articles for potential improvement. Folly Mox (talk) 15:53, 18 September 2023 (UTC)[reply]
The page would still be accessible, though, through links or URL-hopping. It is possible this idea could go hand-in-hand with the above-proposed softdeleting/archiving/whatever we're calling it this week. Edward-Woodrow :) [talk] 19:42, 18 September 2023 (UTC)[reply]
The RFC I was thinking of was indeed last year. Folly Mox (talk) 20:18, 17 September 2023 (UTC)[reply]
Although MINREF says that, WP:V which is policy boy an information page says Any material lacking an inline citation to a reliable source that directly supports the material may be removed and should not be restored without an inline citation to a reliable source. So an editor can redirect an article without any referencing, or cut it down to a stub. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 21:42, 18 September 2023 (UTC)[reply]
Agreed. Stubbifying, redirecting, sourcing, tagging, and the null action are all permitted over unreferenced articles under current policy. Folly Mox (talk) 05:26, 19 September 2023 (UTC)[reply]
  • Support. It's not 2004 anymore. Everyone in the English-speaking world with an internet connection knows what Wikipedia is and what its basic requirements are, to the extent that {{Citation needed}} is a pretty universally recognized phrase even outside the context of Wikipedia. Sourcing things with the internet is easier now than it ever has been. So we should expect much more of people when it comes to citing something at article creation than we did in 2004. I would also guess we're somewhere approaching 99% coverage of the most globally-notable non-CE topics; thus, unlike in 2004, there is virtually 0% any old as-yet-untagged unreferenced article is going to be on a subject so integral to society that we can be certain expansion and sourcing will ever happen. JoelleJay (talk) 01:32, 18 September 2023 (UTC)[reply]
I think the idea that we're approaching 99% "completion" isn't accurate; see this recent discussion at VPM, for example. Curbon7 (talk) 03:09, 18 September 2023 (UTC)[reply]
I did not say 99% "completion". I said 99% coverage of the most globally-notable non-CE topics. This means subjects like apple and knee and Muhammad and psychology and division (mathematics) and beauty and Russia. Things that are universally recognized and would always be expected to appear in any broad general-use encyclopedia. JoelleJay (talk) 03:31, 18 September 2023 (UTC)[reply]
For clarification, when you say coverage are you talking about articles created or articles fleshed out? Curbon7 (talk) 04:24, 18 September 2023 (UTC)[reply]
Articles created. JoelleJay (talk) 22:01, 18 September 2023 (UTC)[reply]
Everyone in the English-speaking world with an internet connection knows what Wikipedia is and what its basic requirements are: on this point, it's just wrong to presume that everyone understands Wikipedia. They really don't. It might be a frequently used tool but it's extremely to rare to find people who have scraped near its innards, or to have more than the vaguest idea of en.wp's reliability beyond "I know it's iffy but it seems mostly ok". {{Citation needed}} is a niche reference in the first place, but especially outside cities and outside the age group 20–35. Even for highly educated people, have they ever looked at a Wikipedia reference list? would not infrequently elicit a 'they have those?'. Notability and verifiability are pretty much Wikipedia's basic requirements, and it's totally mad to expect anywhere upwards of 5% to recognise either one. Everyone in the English-speaking world with an internet connection? That sentence struck me as a massive example of projecting oneself on others, and I'm sorry if it's tangential but just wanted to put it out there. J947 † edits 09:46, 18 September 2023 (UTC)[reply]
Agree with Everyone here being an overstatement. My experience has been that people of my generation are aware of Wikipedia's existence, and people of younger generations typically encounter us in the form of their google lady saying "according to Wikipedia," before answering a question beginning "ok google" asked aloud in their kitchen.
If nothing else, the data constantly spewing in from AfC indicate that people generally don't understand what Wikipedia is and especially don't understand its basic requirements. Folly Mox (talk) 14:39, 18 September 2023 (UTC)[reply]
By "understand Wikipedia", I mean "understand that it's an encyclopedia" and what an encyclopedia is. It's the 5th-most visited website in the world, people are using it as a resource for accurate information. Sure, "everyone" might be hyperbolic--I guess there is that cohort of old folks who haven't had to write an essay in 50 years and so might have forgotten that things should be cited--but looking things up online is universal within the demographic I mentioned so I find it highly unlikely anyone doesn't understand what Wikipedia is. We have many expectations of adults that aren't always met, that doesn't mean we shouldn't have those expectations. Likewise, we can expect (as distinct from presume) that editors provide sources when creating articles. JoelleJay (talk) 18:47, 18 September 2023 (UTC)[reply]
  • Oppose as utterly irresponsible. I don't like a subject area, I go and mass tag all the unreferenced articles at a pace that nobody can keep up with, and they are gone in 14 days. It is simple as that. Could even be done by bot. --Rschen7754 04:13, 18 September 2023 (UTC)[reply]
    I don't like a subject area, I go and mass tag all the unreferenced articles at a pace that nobody can keep up with: are you suggesting requesting sources or deleting unsourced content is a form of vandalism or harassment? Or are you suggesting that a fait accompli implies what has been done is now everyone's valuable treasure and responsibility?
    If users did not add sources it is their problem, providing sources is the responsibility of the person who added information (WP:BURDEN). And other users have to deal with this lack of sources: no user should consider themselves "bad", "frivolous", "in a deletion frenzy", or "irresponsible", because they are simply doing what's right by removing what is unsourced (deleting articles included). Again, it is the person who adds the information who should be sourcing them. There were thousands of people who did not provide a source on multiple articles throughout WP? Anyone can remove what they have added, any of those unsourced articles should be deleted. Veverve (talk) 05:38, 18 September 2023 (UTC)[reply]
    I don't like a subject area, I go and mass tag all the unreferenced articles at a pace that nobody can keep up with: Not only would this happen, we would probably see a few people who "accidentally" remove "unreliable" (according to them) sources just before tagging them, or even tagging articles while the sources are still present. That's not IMO a sound reason not to set this long, slow train in motion, but I wouldn't want anyone to be surprised when it does happen. WhatamIdoing (talk) 09:15, 18 September 2023 (UTC)[reply]
    we would probably see a few people who "accidentally" remove "unreliable" (according to them) sources just before tagging them, or even tagging articles while the sources are still present: the same argument can be said for anything that gets PRODed or XfDed. Veverve (talk) 10:32, 18 September 2023 (UTC)[reply]
    It already happens with PROD and XFD; that's why I am confident that it will happen here, too. I suppose the more subtle way to handle this is for one editor to remove sources from the articles they dislike, and another to come along later to tag them for deletion. Since we can reasonably predict that this will be interpreted as a black-and-white rule, we should expect all uncited articles to be tagged for deletion. If it follows a PROD-like process, there's a risk that an admin will notice that the article was previously sourced, but I would not expect that to be a major barrier to deleting most articles about any subject we choose. WhatamIdoing (talk) 05:43, 21 September 2023 (UTC)[reply]
    I read the will not apply retroactively as applying to articles created prior to the potential adoption of this idea, rather than articles tagged prior, but that could certainly be clarified further. Are there whole topic areas lacking references? There probably shouldn't be any topic areas lacking references created in the future, and if such a situation were to arise, I think I'd characterise the creation of all the unreferenced articles as more disruptive than tagging them all db-unref or whatever we'd call this process. Folly Mox (talk) 14:58, 18 September 2023 (UTC)[reply]
    My intent was articles tagged prior to potential adoption, but I'm willing to be talked around. Edward-Woodrow :) [talk] 19:44, 18 September 2023 (UTC)[reply]
  • Support: you add information, you source, you prove notability. WP:BURDEN is on you. A 2007 essay already proposed the same philosophy Edward-Woodrow does here: Wikipedia:No reliable sources, no verifiability, no article. Veverve (talk) 05:27, 18 September 2023 (UTC)[reply]
  • Deletion would be too harsh; consider that these may be created by newcomers, who would be bit and perhaps leave before they can become a good contributor. If we have to go down this road, then draftifying is a better solution, but with no deadline, this is way more than is necessary for all but BLPs and sensitive topics. SounderBruce 07:08, 18 September 2023 (UTC)[reply]
    If someone in 2023 does not understand that they need to cite whichever sources they're using to write a Wikipedia article, they should not be writing Wikipedia articles. JoelleJay (talk) 18:18, 18 September 2023 (UTC)[reply]
    No one begins editing with a perfect understanding of policies. Many editors receive proper guidance and become productive with some experience; turning them away at the first step would only contribute to worsening editor retention. There are also cultures that might not conform to our ideas of citations and reliable sources; are all those people not excluded from what should be an inclusive movement? Do we want to wall ourselves off to only people who fit a colonialist view of knowledge? SounderBruce 00:52, 20 September 2023 (UTC)[reply]
    @SounderBruce: They'll have two weeks to find references. Besides, speedy deletion would be far more bitey than this. I'm not sure I understand your statement There are also cultures that might not conform to our ideas of citations and reliable sources; are all those people not excluded from what should be an inclusive movement? Wikipedia policy requires sources. Period. And in what way would different cultures diverge on interpretation of "citations reliable sources?" I'm not sure I understand this. Edward-Woodrow • talk 01:16, 20 September 2023 (UTC)[reply]
    We could write more about this subject than anyone would probably care to read, but the following two points are probably the biggest general areas:
    • What counts as "reliable" varies by culture. My culture doesn't accept "I saw Karp in the elevator, and he said it was probably np-complete" as a reliable source, but personal information from an expert is a very highly valued source in some other cultures.
    • Cultures without a long tradition of written language do not expect the sources they rely upon to be fixed in a tangible form. My culture (and enwiki's policy) does.
    WhatamIdoing (talk) 05:49, 21 September 2023 (UTC)[reply]
    I see, thanks. Edward-Woodrow • talk 12:12, 21 September 2023 (UTC)[reply]
    You don't need a perfect understanding of the policies. You only need to meet the most basic expectation that you cite whatever source you're using to write an article so that other editors can verify it. If someone is from some hypothetical culture that doesn't have the concept of "this information came from this place" ends up on Wikipedia, gains autoconfirmed status, and then creates an unsourced article in passable English, this proposal would alert them to the requirement for sources (and what that means) and they could learn what to do. But I find it very unlikely that anyone creating unreferenced articles in 2023 is an editor we would want to retain, and even more unlikely that deleting unreferenced articles from 2006 would be discouraging to anyone. JoelleJay (talk) 01:31, 20 September 2023 (UTC)[reply]
    You are assuming that the person is using a source, that the source has been fixed in a reasonably permanent form, and that the source has been made available to the general public. We require that it be possible to cite such a source, but I also know that I could write, without consulting a single source, that there is one public park in the small town where my mother grew up. It would be accurate, and it would probably be verifiable (I assume the town has a website), but if you asked me to "cite whatever source you're using", I'd have to say "I'm not using any sources. This is just something I've known for decades." While I give an example of a single sentence, it's possible to do this for whole articles.
    I think we do want to retain editors whose first contributions don't have inline citations. Your first addition of content was unsourced. Your first article had a few bare URLs but no inline citations. (My first edits were no better: although my first mainspace edit was to add bare URL as a source, my second added a significant amount of verifiable but uncited information, and my first article had just one bare URL, to what's now a dead link but whose domain name makes me think it might have been a personal blog.) I think that the core community would agree that they are happy that we have both been retained. WhatamIdoing (talk) 06:05, 21 September 2023 (UTC)[reply]
    I wrote my first article 13 years ago as a high schooler, and despite clearly not having read Wikipedia's rules I still assumed I needed to cite some sources before I even published it to userspace. It had inline citations and a reflist a bit over an hour after going into mainspace.
    I'm talking about articles created in 2023 that don't have a single hint of a source, including external links. Someone so totally unaware of the concept of "published sources" that after two weeks they still don't understand what to do when their page gets the proposed tag should not be editing here. It should be irrelevant whether they come from some hypothetical culture that doesn't use/acknowledge Western sources even when they are writing on a modern platform. JoelleJay (talk) 01:21, 22 September 2023 (UTC)[reply]
    I agree with SounderBruce (talk · contribs), draftiying is a better way to tackle new unreferenced articles. Deleting unreferenced articles, and assuming that they fail WP:V does not WP:AGF. SailingInABathTub ~~🛁~~ 09:08, 20 September 2023 (UTC)[reply]
  • Support in theory. Reminds me of the Commons thing where files without copyright data can be deleted with a similar system.
    • Does the tagger have a burden to search for sources and add them directly (or through Template:Refideas or similar)? We would need detailed policy on when not to tag. Add to TW eventually?
    • This would probably be primarily for the 110 000-some articles that already are unsourced, not affecting NPP, because unsourced articles wouldnt get through anyway? If this would affect NPP (flowchart), how?
    • What would 'automatically deleted after X days' mean in practice? Would an admin check for sources before deleting, or are articles really automatically gone?
      • How about a deletion review, where like 3 RS found = restore?
    • Eventually could we have nice things, like cats and bot-updated tables. NotAGenious (talk) 08:10, 18 September 2023 (UTC)[reply]
    @NotAGenious, based on how things went when we started deleting unsourced BLPs, "how it would work" is "with a good deal of finger-pointing and blame-shifting". The group of people who demand that articles be sourced tend unfortunately not to overlap significantly with the group of people who actually source articles themselves. For example, in the last ~two months, you've made about 750 edits and added about 15 refs. This doesn't mean that your contributions are unhelpful – nothing like that at all – but overall, as a purely practical matter, we'd probably have to assume that you wouldn't contribute a lot to the work necessary to make this proposal successful. I wouldn't be surprised to discover that was the case for most editors who support the idea in theory. WhatamIdoing (talk) 08:23, 18 September 2023 (UTC)[reply]
    The group of people who demand that articles be sourced tend unfortunately not to overlap significantly with the group of people who actually source articles themselves: and why should it be otherwise? Burden is not on those who demand sources, but on those who add information.
    As a sidenote, I have noticed that those who do not add sources to their claims are very much overlapping with those who do not want those kind of methods to exist. One just has to look at the fact you yourself consider unsourced contributions as valuable. Veverve (talk) 10:27, 18 September 2023 (UTC)[reply]
    When you demand that other people do work that you do not choose to do yourself, they tend to feel like you are uncollegial, uncollaborative, and a nuisance. And when the person making those demands adds unsourced content himself (example from earlier this year), we can add hypocritical to the list of words people might apply to someone who says everything must be sourced, but doesn't actually add sources himself.
    But I wasn't actually concerned about you in particular; I'm just saying that if we're going to identify a problem that needs to be fixed on a certain timeline, we need to consider whether that is realistic with the time and energy that actual editors are willing to devote to it. Otherwise, the work won't actually get done.
    I do not consider unsourced contributions to be valuable per se, even though I opposed having your badly written essay in the project space. I consider some unsourced contributions to be valuable, and others to be garbage, just like I consider some cited contributions to be valuable and others to be garbage. WhatamIdoing (talk) 06:20, 21 September 2023 (UTC)[reply]
    Thank you, WhatAmIDoing, for expressing your concerns. While I have read a lot of info and past discussions on referencing and participated in a few discussions on reliable sources, I agree that I haven't focused much on adding sources (well, nor content creation overall - but when I write something, I always source it). But, from what I've learned, I feel competent enough to participate here, and if the idea is succesful, be a part on implementing the system. NotAGenious (talk) 10:39, 18 September 2023 (UTC)[reply]
    I'm glad to hear it. The next question is: How many others like you are willing to pitch in?
    If we have 100,000 articles, and everyone in the team cites one article per week, we'll need 2,000 editors to accomplish this goal in one year. If they can do one a day, we'll need 275 for a year. Realistically speaking, and keeping in mind that all of the other work of the wiki still has to happen, including some major challenges during the next year (e.g., the introduction of mw:Help:Temporary accounts in 2024, which may ultimately be an improvement, but which will disrupt some people's workflows), how many do you think that an editor like you could do? WhatamIdoing (talk) 06:25, 21 September 2023 (UTC)[reply]
    A rotating crew of eleven, each sourcing an average of five articles a day, could complete the task by 2030. One year sounds like a pretty ambitious timeframe. Folly Mox (talk) 17:16, 21 September 2023 (UTC)[reply]
    One article per week seems like a good goal. One per day seems like a little too much. Edward-Woodrow • talk 19:54, 21 September 2023 (UTC)[reply]
    All right: At the rate of one article per week, per editor, between now and the start of 2030, we need 300 editors signed up. Doing it over the course of six years will require continually recruiting folks to replace those who fall away, but that's the scale that we're looking at.
    Now: Is that reasonable? I'm thinking that it's maybe not entirely reasonable. For comparison, consider the one-month-long Wikipedia:WikiCup and Wikipedia:WikiProject Good articles/GAN Backlog Drives, which seem to draw about 100 participants (some of which don't do much; others of which do a lot more). We're looking at needing three times as many people, for one or two orders of magnitude longer (depending on whether you think of this as "six annual events, each of which is 12 months long" or "72 months in a row"). Either way, it's more people for a much longer period of time.
    What might make it feasible, or at least not obviously impossible, is that the request is smaller. I added two sources to five medicine-related articles just now. It took me about 45 minutes to do five articles, with some of them being faster than others. (Generally, the more specific/narrow the subject, the easier it is to find a relevant source.) So if the goal is to get just one source into an article, rather than to fully source the content, that request is just 10 minutes a week, and the challenge would be to find enough people (you'd probably have to recruit more than a thousand, in the end, because people won't do a full year), and to find ways to keep them on task (because even people who genuinely want to help will forget). WhatamIdoing (talk) 02:13, 22 September 2023 (UTC)[reply]
    It would be helpful if a couple of other editors did a similar self-experiment. I pulled five non-organizational, non-BLP articles out of https://bambots.brucemyers.com/cwb/bycat/Medicine2.html#Cites%20no%20sources There are similar lists for most of the large WikiProjects. WhatamIdoing (talk) 02:17, 22 September 2023 (UTC)[reply]
    I would say the tagger has a burden to do a cursory search, to make sure they aren't about to delete a future FA. It would be good for an admin to do the same quick searches, just to be safe. Edward-Woodrow :) [talk] 11:39, 18 September 2023 (UTC)[reply]
@Edward-Woodrow, The very first bullet point is wrong. Unreferenced articles are not "a violation of one of Wikipedia's core content policies: WP:V." WP:V only requires inline citations under certain circumstances. It is possible to write a (very short) stub that is not required to contain any inline citations. For example:
  • Lung cancer is a type of cancer. Smoking raises the risk of lung cancer.
  • Christmas candy is a type of candy associated with Christmas. Candy canes are one type of Christmas candy.
  • French Renaissance sculpture is the type of sculpture made in France during the Renaissance.
None of those sub-stubs contain direct quotations; none of them contain contentious matter about living people; none of them contain material that is WP:LIKELY to be challenged; none of them contain material that has already been challenged. If you wish to expand your view a little further, then all three of them contain solely material for which a source could be found quite easily, so it's not a NOR violation, either.
It is true that none of them prove that they are notable subjects, but I suggest to you that it is equally true that no editor will genuinely have any doubts about the notability of these subjects, and for better or worse, there is not one sentence in any of the core policies that says obviously notable subjects must be proven to be notable through the addition of an inline citation, even if every editor already knows that these are notable subjects. Past efforts to create such a requirement have failed. Maybe if you want to be able to delete articles because somebody else didn't do the thing that you haven't done yourself, then you should first try to get the undesirable behavior officially banned first. That's how we ended up with WP:BLPPROD some years back: first, you have to get the policy amended to disallow unsourced articles; only after that can you start deleting them for violating the rule. WhatamIdoing (talk) 08:13, 18 September 2023 (UTC)[reply]
I'm not saying that everything needs an inline citation, I'm saying that all material in Wikipedia mainspace, including everything in articles, lists, and captions, must be verifiable. It's not verifiable if there isn't a source anywhere in the article, is it? Edward-Woodrow :) [talk] 11:33, 18 September 2023 (UTC)[reply]
Typical cases where things are verifiable but unsourced are when articles summarise other articles, and the references can be found only in articles linked from the unsourced article in question. —Kusma (talk) 11:43, 18 September 2023 (UTC)[reply]
User:Edward-Woodrow, It's not verifiable if there isn't a source anywhere in the article, is it? To me this reads as a misconception. Information is verifiable if it has been published somewhere. It makes everything way way easier if some indication of that publication is in the article, but "takes extra work to verify" is not equivalent to "unverifiable". Folly Mox (talk) 14:47, 18 September 2023 (UTC)[reply]
That is definitely not how people evaluate WP:V in practice. An unsourced article tagged with {{sources exist}} because of some sigcov in obscure scholarship is rightfully said to “have unverifiable statements.” All information not proven false could theoretically be verified. Mach61 (talk) 23:40, 18 September 2023 (UTC)[reply]
Then 'people' are wrong. – Joe (talk) 12:45, 20 September 2023 (UTC)[reply]
@Joe Roe, you're right, but @Mach61 has only been editing for a year, and I would not be the least bit surprised to hear that some high-volume editors "stretch" or "simplify" the rules for newer editors, in ways that happen to favor the convenience of the high-volume editors. We have an endless supply of written rules, but WP:Nobody reads the directions, and most people learn the alleged rules through a sort of telephone game: one editor tells me a slightly distorted version of the rules, and I misinterpret that when I tell you what you should do, and you're stuck with bad information. WhatamIdoing (talk) 06:33, 21 September 2023 (UTC)[reply]
I've been thinking the same thing a lot lately. It seems like new editors are increasingly learning about our policies from, well, somewhere other than the policies themselves (training programmes? Condensed summaries? Word of mouth on Discord? I really don't know), resulting in baffling examples of people misunderstanding what I always considered to be very straightforward directions, mixing up concepts, applying guidance on one thing to something completely different, throwing around shortcuts without knowing what the linked section actually says (never mind the next section), etc. And I don't think I'm just being a cranky old man here; it really does seem like something has changed in the last couple of years. I wish I could put my finger on what it is. – Joe (talk) 06:44, 21 September 2023 (UTC)[reply]
I think the change predates the pandemic, and I wonder sometimes if we could trace it back to as early as 2016. That's the year when both "extended confirmed" and "new page reviewer" became addition user rights, and their main purpose is to give experienced editors control. So whatever zeitgeist at that time prompted this desire for even more control might be the start, not of "nobody reads the directions", which has been a thing on the internet since before most people had internet access, but of the idea that everything should be tailored to my convenience, and that the highest goal was reverting someone else's less-than-perfect first contribution, instead of creating content yourself. WhatamIdoing (talk) 14:40, 21 September 2023 (UTC)[reply]
I think the change has crept up on us gradually. There were people like that when I started editing in 2007, but they seem to have got more and more vocal. I am of the generation to whom you could say RTFM and I would go away sheepishly to do so, but more and more people seem to think that everything should be done their way, which usually involves someone else doing the work. Phil Bridger (talk) 15:29, 21 September 2023 (UTC)[reply]
I agree that it has been a gradual shift. WhatamIdoing (talk) 02:27, 22 September 2023 (UTC)[reply]
@Edward-Woodrow, if there isn't a source anywhere in the article, it's only uncited. It's still verifiable if you (i.e., anyone) are able to find a reliable source that matches the contents. The rule is that content must be verifiABLE, and we assume you are "able" to use a search engine. See Wikipedia:Glossary#verifiable for a short definition of the related terms. WhatamIdoing (talk) 06:29, 21 September 2023 (UTC)[reply]
Yes but as I said above WP:V says Any material lacking an inline citation to a reliable source that directly supports the material may be removed and should not be restored without an inline citation to a reliable source. So there should be no expectation that unreferenced content can standard if it is challenged. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 21:51, 18 September 2023 (UTC)[reply]
This is the idea lab, so I don't know what all the bold "support" votes above are here for. Anyway, the vast majority of the "unreferenced articles" issue is in old articles; most if not all new articles have some references (so I don't think your proposal would help discourage new unreferenced articles any more than they already are). Deleting articles that are unreferenced-but-nobody-has-tagged-them-yet (which focuses on articles that have been unreferenced for short times, not long times) seems less useful than other methods of choosing which unreferenced articles to deal with. You could randomly AfD five unreferenced articles per day. —Kusma (talk) 08:20, 18 September 2023 (UTC)[reply]
I think they're almost all tagged. We sent a bot through to tag unref'd articles some years back. I rarely encounter an untagged one (and sometimes still encounter an incorrectly tagged one). WhatamIdoing (talk) 08:25, 18 September 2023 (UTC)[reply]
Disagree. I checked Petscan. There are 307 articles in the category for this month, but only 15 of those articles were created after July 1 of this year. Many were created years ago. 276 out of 307 were created before 2018. So the assumption that because this only applies to newly tagged articles, it only applies to newly created articles, is untrue; this proposal as stated will have retroactive effects. DFlhb (talk) 09:21, 18 September 2023 (UTC)[reply]
most if not all new articles have some references (so I don't think your proposal would help discourage new unreferenced articles any more than they already are). Besides DFlhb's reply above.... I suggest you check the New Pages Feed. This month, I have draftified 104 articles because the have no sources. Edward-Woodrow :) [talk] 11:42, 18 September 2023 (UTC)[reply]
My question is how many new unsourced articles make it through New Page patrol. —Kusma (talk) 11:44, 18 September 2023 (UTC)[reply]
Well, based on my limited experience, it depends, mostly on the reviewer. Some people like to tag and leave a friendly message on the talk page, I prefer to draftify, with the goal there to get it out of mainspace, and I assume there are other methods, too. Edward-Woodrow :) [talk] 11:49, 18 September 2023 (UTC)[reply]
I've done I'd guess a few thousand NPP's. If I ever saw an article with zero references but where suitable sources clearly exist (and the article didn't have other disqualifying attributes and isn't a good situation for conversion of a stub to a redirect) I'd pass it, but that has never happened. North8000 (talk) 14:33, 18 September 2023 (UTC)[reply]
I've seen a few with zero references but where suitable sources clearly exist. When I find ones like that, I add the sources. Yes, that makes NPP take longer and is arguably not NPP's job, but I'm of the mind that it should be part of the job. ~ ONUnicorn(Talk|Contribs)problem solving 14:54, 18 September 2023 (UTC)[reply]
Personally, I'd rather have NPP focus on CSD-worthy problems (=its original remit) and leave the rest for the rest of the community, but the scope creep has extended so far that I don't know that there is any chance of reducing the load on the NPPers at this point. WhatamIdoing (talk) 06:36, 21 September 2023 (UTC)[reply]
I feel like I read somewhere that there are two actions that can be taken on new pages: patrol and review. I'm not sure whether this is a distinction without a difference or two actually separate things, but it would make sense to me if there were a first order "is this CSDable" check and a second "are there any other major issues" check.
There was also a discussion somewhere about smearing out this boolean NPP tagging, by introducing other actions that could be undertaken by people with different skillsets, such that one person could mark an article as "not copyvio", another could mark it "not spam" etc until the bar is crossed where it becomes "fully patrolled", without pushing any one editor into making determinations they don't feel sufficiently practised at. To address myself in the present comment, this tangent would probably be better suited to a different venue. Folly Mox (talk) 17:30, 21 September 2023 (UTC)[reply]
@Folly Mox: introducing other actions that could be undertaken by people with different skillsets, such that one person could mark an article as "not copyvio", another could mark it "not spam" etc until the bar is crossed where it becomes "fully patrolled" I understand the rationale behind that, but I feel like it would just make everything take orders of magnitude longer in NPP; the backlog is big enough (although it looks like the redirect backlog is levelling off? Maybe?). Edward-Woodrow • talk 19:51, 21 September 2023 (UTC)[reply]
It wasn't my idea, and it's not for this conversation, but yeah definitely if people have a single assigned station and you're always waiting on User:Example to finish the G5 check or whatever, NPP would certainly take longer. I think the original idea was to allow partial patrols to be recorded somehow, like if no one is comfortable assessing the reliability of the sourcing because it's in Armenian or Mongolian or something, but Earwig came up clean, there could at least be an easy way to record that to avoid duplication of effort. I think the thought was that ideally all NPPers would still be competent in all the areas to patrol an article fully from zero to indexed by search engines, but people who are weak in certain areas or can only dedicate time in smol chunks could still put in partial labour. Hopefully someone else remembers the conversation if anyone considers my hazy recollections to have value as a starting point for a new NPP experiment. Folly Mox (talk) 21:25, 21 September 2023 (UTC)[reply]
@Folly Mox, you are thinking of User talk:Iridescent/Archive 50#c-WhatamIdoing-20221011190200-Novem Linguae-20221010223100. It was partly inspired by the way MILHIST does B-class assessments.
A new article often gets ~50 page views on the first day. Particularly in the first hour, people are most often looking for the "quick fail". Those who have particular areas of interest (e.g., attack pages and copyvios) could save some time and reduce duplicated effort by being able to filter out articles that have already been determined not to be a hoax, attack page, etc. WhatamIdoing (talk) 02:44, 22 September 2023 (UTC)[reply]
I feel like Alice and Bob might have been there, but Special:Search has not been aligning with what my brain tells me it once read. I'll take your word for it. That sounds more reliable than my leaky memory. Sometimes it all feels like a dream. Folly Mox (talk) 07:54, 22 September 2023 (UTC)[reply]
I'm pretty sure that I was in (at least) two discussions on that topic around that same time. Another at WT:AFC or a similar page, perhaps? It's possible that an insource: search for the buttons would find it, if we assume that I copied them from one discussion to the other.
P.S. Alice and Bob are everywhere. ;-) WhatamIdoing (talk) 17:20, 25 September 2023 (UTC)[reply]
Indeed, it was the same conversation. I just had to scroll up from the posted link to the top of the subheading. That will (not) teach me to read the larger context I'm always harping on about. Folly Mox (talk) 18:16, 25 September 2023 (UTC)[reply]
Duplicated effort is not necessarily a bad thing. One person may spot what another has missed. Phil Bridger (talk) 08:20, 22 September 2023 (UTC)[reply]
But the tenth person is unlikely to find an attack page that was somehow missed by the previous nine. WhatamIdoing (talk) 17:21, 25 September 2023 (UTC)[reply]
@Folly Mox: Sorry to continue this tangent on NPP, but the patrol vs review discrepancy is only because of a complicated technical issue explained Wikipedia:New pages patrol#Other issues under the Technical details subsection. VickKiang (talk) 08:15, 23 September 2023 (UTC)[reply]
My goodness! So it's not so much a two tiered system as it is a technical glitch. I'll forgo any further commentary on what it could or should be. Tangents are beautiful. Folly Mox (talk) 08:32, 23 September 2023 (UTC)[reply]
  • I'm generally against anything categorical like this, but having absolutely zero references is a very very very very very very low bar and I'd support it. It would need some safety mechanisms to avoid unintended consequences. It would also put us a step towards the mentality that finding reference(es) is the main step 1 of building an article which would solve many of our problems. Without that somebody has done nothing of value / nothing worth preserving. It's like I say that I'm giving somebody a car and I just give them a floor mat and call that an "unfinished car" and tell them to keep a space for it in their garage while they "develop" the car. North8000 (talk) 14:22, 18 September 2023 (UTC)[reply]
  • I would support this, as long as currently existing articles are grandfathered-in and ineligible. Curious how that could be technically implemented. Could we have a bot revert this template when added to older articles? DFlhb (talk) 14:44, 18 September 2023 (UTC)[reply]
  • Question - wouldn’t it be simpler to have a bot scan any potential new article - and simply not accept the “save” into Mainspace unless there was at least one citation? It could generate an error message so the creator knows why the text isn’t being saved. Blueboar (talk) 14:58, 18 September 2023 (UTC)[reply]
    Not exactly a bot task, but presumably the Submit button on drafts could be coded to throw an error if no ref tags are present in the article rather than doing whatever it does to submit the draft to the AfC heap. Articles created directly in mainspace, expanded from redirects, or moved manually across namespaces would require at least an edit filter and possibly a software change. Folly Mox (talk) 15:05, 18 September 2023 (UTC)[reply]
    (edit conflict) It would be difficult. Many (if not a majority) of new articles go into mainspace via "move" and so that restriction would also need to get applied to "move". Many others are created by conversion of redirects and it would also need to apply to those. Finally, people creating articles in mainspace would need to know that the absolute first edit would need to contain a reference which would practically mean that only clever insiders could build an article in mainspace. North8000 (talk) 15:09, 18 September 2023 (UTC)[reply]
    There's also the issue that any warning describing how to solve the problem (e.g. "Error: All drafts must contain at least one reference before they can be saved to mainspace. Refs are added using <ref></ref> tags containing by the reference's name, ...") will just have users copy-and-paste the "magic words" so their article can be saved, without them knowing or caring what those tags mean to begin with. 2603:8001:4542:28FB:49F5:C6E1:BDC2:515F (talk) 15:24, 18 September 2023 (UTC)[reply]
    It would be non-trivial, but those can be addressed by the WMF with thought-through interface changes. It would also narrow the "feedback gap" (proactive enforcement vs reactive, so newbs are told ahead of time if they're doing something they shouldn't, which would be more newb-friendly). Given the vast amounts of work spent on maintenance and gnoming, we could do with more 'automation' of this kind. DFlhb (talk) 16:16, 18 September 2023 (UTC)[reply]
    I agree that pro-active feedback is potentially the way to go, via improvements to the AfC wizard, edit filter warnings, etc. Suriname0 (talk) 20:47, 19 September 2023 (UTC)[reply]
    See mw:Edit check, which will be nudging editors to add sources (for any whole paragraph, at the moment, but that can be changed.) WhatamIdoing (talk) 06:39, 21 September 2023 (UTC)[reply]
    This looks like a great initiative. Eager to see the results once this gets deployed. Suriname0 (talk) 03:14, 22 September 2023 (UTC)[reply]
    This should be a non-issue for drafts, as an unsourced draft is easily dispositioned as such by AfC. --Ahecht (TALK
    PAGE    ) 18:30, 18 September 2023 (UTC)[reply]
    Yes, this template would only be applicable to mainspace. Edward-Woodrow :) [talk] 19:36, 18 September 2023 (UTC)[reply]
  • Good idea. Anything that reduces the burden of AfD and cleans up the project with only high quality sources is a good thing. I would use it to amend BLPPROD. Aasim - Herrscher of Wikis ❄️ 15:45, 18 September 2023 (UTC)[reply]
  • I'm in general agreement with something along these lines - a requirement that articles cite at least one source (doesn't have to be in-line; doesn't have to be the best quality source) to remain in mainspace. I also think sending all such articles to draft is a bad idea. This seems to be similar to BLPPROD. That said, the two week time frame seems arbitrary. BLPPROD gives people 7 days to add sources. Regular PROD provides 7 days for someone to object. If instead of this psuedo prod we sent it to draft, they would have 6 months (although the drafts are harder to find than PRODs). I think this needs to be in line with some of our other processes, and either be 7 days or 6 months. ~ ONUnicorn(Talk|Contribs)problem solving 16:40, 18 September 2023 (UTC)[reply]
  • Good idea as is Blueboar & North8000's idea above to have the Submit button throw an error code if someone tries to make an article without sources. I wouldn't even mind this being retroactive for the oldest articles- say, any articles from the <current oldest year tagged> get tagged with this. I'm neutral on whether this should last a week or a fortnight.
Additionally, I would point out that requiring sources isn't just a Wikipedia thing. Every essay requiring research since high school that I've written has required me to cite my sources. It doesn't take a genius to cite sources- if the average high school idiot can manage, there's no reason not to require it of would-be editors. This would just be a change to enforce rules that keep Wikipedia better than a failing high school essay. Happy editing, SilverTiger12 (talk) 17:46, 18 September 2023 (UTC)[reply]
@ONUnicorn and SilverTiger12: About the time frame... I chose fourteen days because I thought seven days would be seen as a bit too stringent, especially since I suspect there will be more unreferenced-prods then other PRODs. Then again, seven days would be more consistent, and thus easier to remember, etc. Edward-Woodrow :) [talk] 19:39, 18 September 2023 (UTC)[reply]
  • Good goal in spirit. I don't agree with this suggested implementation for roughly 4 reasons.
  1. We should strive for clarity for our editors, experienced or new. We should improve the accuracy of template names, rather than making them mean something other than their name. Thus, repurposing {{unreferenced}} is not the right idea. I'm not fundamentally opposed to a new tag, for example {{proposed deletion unreferenced}} or even a fancy custom message for something like {{proposed deletion|unreferenced}}.
  2. {{unreferenced}} is currently often misused.
    1. For example already published books etc are sometimes marked unreferenced, but an article about an already published book has implicitly verifiable information from the published item itself.
    2. Sometimes it's also used when {{no footnotes}} would be more appropriate.
  3. Changing the meaning of {{unreferenced}} isn't worth it for the gain. Articles without references at AfD are not a major problem, they are (1) rare in comparison to other types of articles at AfD and (2) frequnetly HEY-able. The biggest threats to Wikipedia at AfD these days are WP:PROMO/WP:NOTCV things, and these articles tend to be WP:REFBOMBED with weak references.
  4. After some time, this would begin to encourage editors to add weak references. An article with only weak references is arguably worse than one with none at all. An article without any references is easy to spot and therefore improve. It's also easier to spot for readers. An article with weak references is harder to spot and could sit unimproved for much longer. For readers who only check for lists of references (as many of us do on occasion), it might fool them into thinking its better referenced than it really is.
—siroχo 19:14, 18 September 2023 (UTC)[reply]
@Edward-Woodrow Here's some feedback on the specific use of {{unreferenced}}, forgot you had asked for a ping on reply so here it is now. —siroχo 19:16, 18 September 2023 (UTC)[reply]
Excellent points, especially number 4. We want people to add proper references, not the crap that sometimes gets added in referencing contests. An unsourced article is better than one with inline citations to a Wikipedia mirror. —Kusma (talk) 11:40, 20 September 2023 (UTC)[reply]
I think in that case, an editor could re-add the tag. <ref/> tags aren't the magic word, it's what's in them. Edward-Woodrow • talk 11:58, 20 September 2023 (UTC)[reply]
Ah, so your proposal isn't just about unsourced articles, but also about the (far more numerous) articles using bad sources? —Kusma (talk) 13:34, 20 September 2023 (UTC)[reply]
No, I didn't say that. But it isn't a stretch to say that the following example would be disallowed:
  • Editor A creates an article with no sources.
  • Editor B tags it with {{unreferenced PROD}}
  • Editor A adds an unrelated citation that does not support the content, and they remove the tag, saying "look, it's referenced now".
In that case, the tag should clearly be re-added, otherwise the entire purpose is nulled. Edward-Woodrow • talk 19:42, 20 September 2023 (UTC)[reply]
Realistically, it'd be better to follow the removed-PROD process, and ship it off to AFD. Otherwise, you're just going to end up with edit wars over whether that source is reliable, supports the content, etc. WhatamIdoing (talk) 02:51, 22 September 2023 (UTC)[reply]
True. However, in egregious cases, Editor B can probably safely re-add the tag. Edward-Woodrow • talk 22:18, 23 September 2023 (UTC)[reply]
  • Support For the simple reason that this is the standard already applied to most IP and new editors, realistically only long-term editors can get away with creating such cruft. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 21:46, 18 September 2023 (UTC)[reply]
    I like with siroχo ideas of a new proposed deletion template would be a better idea than repurposing a pre-existing template. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 21:54, 18 September 2023 (UTC)[reply]
  • Support I think it is high time that we work at insuring that information in Wikipedia is in fact verifiable by requiring citation of source that varify the content. It is no longer enough to assume that something is verifiable without bothering to find and cite reliable sources. - Donald Albury 23:34, 18 September 2023 (UTC)[reply]
  • If I'm going to be 100% honest, I kind of thought something like this was already happening. You just don't see many of these unreference articles being successfully pushed out, at least not in my experience going from random page to random page. That being said, if at any point this becomes a bigger issue, such a process would be vital to have. I'm very much on board with this, but I think the month based system proposed would be better off as a set amount of time rather than the end of the month (should that be what you're actually proposing). Somewhere between 21-30 days? - Mebigrouxboy (talk) 04:52, 19 September 2023 (UTC)[reply]
Since this idea seems to be enjoying broad support, I will start a WP:VPPR discussion in early October, bearing mind the suggestions here. I support Siroxo's suggestion of creating a new template ({{unreferenced}} would be marked as deprecated). There remains the question of how long the waiting period should be. One week? Two weeks? One month?
We should also make sure that users don't:
  • Add nonsense references so they can remove the tag.
  • Add failed verification references so they can remove the tag.
  • etc.: listing the myriad methods here would be a WP:BEANS violation.
Edward-Woodrow :) [talk] 14:04, 19 September 2023 (UTC)[reply]
  • Support. This seems like a good path to take. There really is no reason in allowing new articles to be created with zero sources. Regarding the above mentioned gaming the system, a fail verification or a nonsense source should be removed and the article tagged as unsourced. Personally, I'd not grandfather any article, but that's me. Gonnym (talk) 19:56, 19 September 2023 (UTC)[reply]
  • Good idea. Given the risk of vandalism removing references, I'd prefer to have admins make all decisions about deletion, and not leave anything to bots. And there should be advice along the lines of WP:BEFORE for the use of the template. We might want to make repeatedly tagging pages that shouldn't be tagged something that is considered disruptive conduct. I would eventually want this procedure to apply to all unreferenced articles retroactively, although I realize that there would be a backlog. (I suspect, but don't know for sure, that a proposal like this could get consensus, even without grandfathering old articles.) I'm ambivalent about the length of the waiting period. On the one hand, I feel like editors should not be too rushed to find sources. On the other hand, with a likely backlog, there will be a delay anyway. I guess two weeks might be reasonable. --Tryptofish (talk) 20:31, 19 September 2023 (UTC)[reply]
    Whether two weeks is reasonable depends on how many of them get tagged in a given time span. WhatamIdoing (talk) 06:50, 21 September 2023 (UTC)[reply]
    Following up, I've been thinking about the concerns some editors have raised, that this would go against what Wikipedia has stood for. It seems to me that, nonetheless, times change, and this might be something where it's reasonable for us to do something that differs from historical practice. In its early days, the project needed to create new pages. Today, we have a much larger problem with low-quality content, than with missing content. So I believe we can move in the direction discussed here, without turning into Citizendium. --Tryptofish (talk) 22:32, 27 September 2023 (UTC)[reply]
  • In its present form, this proposal is not compatible with WP:ATD, WP:NPOSSIBLE and WP:BEFORE. Unreferenced tags are very often erroneously placed on articles that are referenced. Many erroneous tags still remain on those articles. There are editors trying to clear the backlog of unreferenced articles by adding sources, and an artificial two week deadline would seriously disrupt their efforts. If an editor is trying to work through the unreferenced tags on, to pick a random example, articles about the chronology of Ireland (which has many obviously notable articles to which tags were inappropriately added instead of sources), it is not helpful to waste his time by forcing him to look at yet another prodlist every day. What is really needed is more people working on the backlog. James500 (talk) 06:18, 20 September 2023 (UTC)[reply]
    The proposal would only apply to new articles, not the backlog Mach61 (talk) 12:01, 20 September 2023 (UTC)[reply]
    Actually I'm certain this would apply to the backlog, at least eventually. NPP and AFC already prevent new articles from not having any sources. If this debate is about new articles only, then people are passionately debating a moot point. Dave (talk) 22:27, 22 September 2023 (UTC)[reply]
  • Good idea. I'm not sure if I would ultimately support or oppose the proposal, but this has clearly been thought through and I would encourage the OP to take it to RfC when they feel it is appropriate. Curbon7 (talk) 09:18, 20 September 2023 (UTC)[reply]
    Upon reading more of the comments here, I am not sure of the feasibility of this. I think Kusma raises a good point below that this may simply lead to a rise of poor sourcing, and it feels like the proposal is getting more complicated as we're now talking about deprecating templates and de-"Prodding" procedure rather than a simple page tag. Curbon7 (talk) 08:37, 22 September 2023 (UTC)[reply]
A reminder that this is the idea lab, so please remember not to make support/oppose !votes, just suggestions for how we can improve the proposal. Thanks, Edward-Woodrow • talk 12:09, 20 September 2023 (UTC)[reply]
  • What people always seem to miss when trying to (mis)apply WP:BURDEN to deletion is that, while creating an unsourced article is analagous to adding an unsourced statement to an existing article, deleting an article is decidely not analagous to removing text from an existing article. You can't undo an article deletion without a whole bureaucratic song-and-dance. You can't even see what was deleted without the help of an admin. Removing an unsourced statement is just the first step in WP:BRD; deleting an article is a drastic and, for vast majority of editors, irreversable rejection of that content. That's why our deletion policy has always been based on the principle that we don't use deletion for surmountable problems, and unless you have reason to believe the whole article is unverifiable, lack of citations is definitely falls in that category. – Joe (talk) 12:41, 20 September 2023 (UTC)[reply]
    I think you have misunderstood how editors apply WP:BURDEN/WP:ONUS to article creations in practice; they don't delete articles, they redirect or draftify them. Such an action is easily revertable, if a consensus emerges to support the articles creation. BilledMammal (talk) 00:38, 27 October 2023 (UTC)[reply]
  • The idea behind this is valid, we definitely should not be solving it by enshrining grandfather clauses into Wikipedia's rules. * Pppery * it has begun... 02:28, 21 September 2023 (UTC)[reply]
    If you do propose this, you need to be clear about what you are proposing - is it "unreferenced articles created after <DATE> are subject to this process", or "all unreferenced articles are subject to this process, but some human has to edit the page to start the 14-day count". The former I would oppose as both too infrequent to bother with (most unref tags are old unreferenced articles that somebody decided to add a tag to today) and creating an obvious double standard, and the latter I would probably support. * Pppery * it has begun... 02:47, 21 September 2023 (UTC)[reply]
    @Pppery: The latter. Specifically, in the theoretical case this was implemented, here's what (I imagine) would happen from the technical perspective:
    • {{unreferenced}} is marked deprecated.
    • {{unreferenced PROD}} is created.
    • The {{unreferenced}} backlog stops growing at all.
    • {{unreferenced PROD}} replaces {{unreferenced}} except in the places where the old tag is already transcluded. Twinkle etc. remove the tag from their tagging options and add the new version to their "PROD" section.
    • Now, if you see an unreferenced article, instead of tagging it with the old template, you tag it with the PROD template.
    Edward-Woodrow • talk 12:20, 21 September 2023 (UTC)[reply]
    I'm not sold that "deletion within 14 days" is a good way to deal with unsourced articles (mostly because I am afraid we will see too much poor sourcing as an unintended consequence), but I agree that grandfathering is not the way to go. Instead of concentrating on articles newly tagged as unsourced, any new proposal should deal with the unsourced backlog. If we had a method to deal with two months' worth of backlog per month, we should be able to kill the backlog within ten years; if we can do four months per months, in five years. Anything faster than that is probably too fast for our limited number of volunteers. —Kusma (talk) 08:13, 21 September 2023 (UTC)[reply]
  • The current system is (or at least is supposed to be) that an editor concerned about a lack of sourcing looks for sources first, and then, if unsuccessful, nominates for deletion if deletion policy applies. What is wrong with that? Remember that the creator doesn't own the article. Phil Bridger (talk) 15:42, 21 September 2023 (UTC)[reply]
  • It's always been policy that every article should be good enough to pass a GAC review if nominated. However, until recently, that was seen as a goal; an imprefect article was better than no article. It is my honest opinion this fairly recent trend of people who by and large do not contribute content to the encyclopedia, but go around and tag every flaw they find in articles with "Fix it now or I send this sucker off to AFD" will be the eventual demise of Wikipedia. I truely believe this. What made wikipedia great and allowed it to defeat Encarta, Citizendium and a hundred other encyclopedias in the market was that Wikipedia was a "one stop shop" with at least some coverage of the core topics you'd find in any respectable encyclopedia and those "$10 bar bet" articles that were beneath coverage in a "academically credentialed encyclopedia". The fact that many of the articles were flawed didn't really matter; what mattered is Wikipedia had coverage of a topic and "they" didn't. In my opinion, the people who now want Wikipedia to switch to the very model that it defeated a hundred times over should be laughed at, not listened to. If you find a flaw in an article, the right thing to do, IMHO, is fix the flaw; not tag, not whine, not delete. I see tagging as a means of requesting help when you can't fix the flaw yourself. Deletion should be reserved for when the flaws in an article are foundational and cannot be fixed by editing.Dave (talk) 18:45, 21 September 2023 (UTC)[reply]
    Indeed, the approach to building the encyclopedia that you describe is detailed in the editing policy, specifically at WP:IMPERFECT and WP:PRESERVE. —siroχo 18:54, 21 September 2023 (UTC)[reply]
    You're saying I should be laughed at? I'm fine with that. Go ahead. I still think we should maintain quality over quantity. Edward-Woodrow • talk 19:49, 21 September 2023 (UTC)[reply]
    My interpretation of Dave's comment was not literally that you should be laughed at, but that the model of encyclopedia creation that traditional encyclopedias, or UGC encyclopedias like Citizendium have used is less effective at creating a high quality encyclopedia.
    It's a worry that workable suggestions like yours could, over time, add up to a model like that. The vast majority of articles on Wikipedia started as low quality in one language or another, it's rare that the first edit creates a true good article.
    There is value in low quality articles. So when/if we implement a policy like this, we do need to take care not to lose that value. So, personally, the question I'm trying to help answer here how do we effectively reduce the number of unverifiable articles on Wikipedia without losing the value that low quality articles continue to bring to Wikipedia. —siroχo 02:54, 22 September 2023 (UTC)[reply]
    Edward, I don't think it's "quality over quantity". I think it's "narrowness over breadth".
    I think this old comment from Iridescent might be worth thinking about. Maybe the world needs our little articles on ultra-niche subjects, even if they aren't beautiful, more than the world needs Wikipedia:Vital articles from us. You can get information about popular and generally important subjects from many websites. But if you want to know more about the Ceilings of the Natural History Museum, London, Wikipedia may be one of your few free options. WhatamIdoing (talk) 05:33, 22 September 2023 (UTC)[reply]
    I think this comment is a little inappropriate, in that it suggests editors support getting rid of articles like Ceilings of the Natural History Museum, London; that is very much false. BilledMammal (talk) 06:46, 22 September 2023 (UTC)[reply]
    Which is an FA? I'm not sure I see your point. Edward-Woodrow • talk 12:28, 22 September 2023 (UTC)[reply]
    Sure, it's a beautiful article, as one would expect from both the subject matter and the editor who created it. But also: "the ceilings in a particular museum" is not a subject whose notability is obvious from the start. What often matters is that we have some information about the subject, not that we have perfect information.
    The COVID-19 pandemic is an example of this. The highest traffic to the articles was in the gap between the disease obviously becoming a problem and "proper" sources getting their things published. For a couple of weeks, we were one of the best places on the internet for reasonably up to date, reasonably accurate information. When better (free) sources became available, our page views dropped. People might prefer beautiful articles from authoritative sources, but what they need is something useful.
    We've seen the same pattern for natural disasters, celebrity deaths, and other time-bound subjects. For those readers, what matters is that we have content. They do not need or expect us to be perfect. Material can be useful without being trustworthy, authoritative, or credible. We all know this from the real-world, too: If Uncle Bob says ____ about politics, then the opposite is almost certainly true. Uncle Bob isn't trustworthy, authoritative, or credible, but he's still useful. WhatamIdoing (talk) 17:39, 25 September 2023 (UTC)[reply]
    Are you a long-lost cousin or sister that I didn't know I had? We seem to have the same Uncle Bob. Phil Bridger (talk) 17:53, 25 September 2023 (UTC)[reply]
    So you're saying that sometimes people rely on us for information, even though sometimes it's incorrect? I don't see the advantage of misinformation. If Uncle Bob is wrong, then Uncle Bob is not useful (unless he is consistently, 100% wrong, but that's a different matter). Edward-Woodrow • talk 19:41, 25 September 2023 (UTC)[reply]
    Edward, while I would largely agree with Dave's comments, I don't find your actions or proposals offensive or laughable. However, there are two broad philosophical points I would earnestly urge you to consider. The first is that, in their time, Nupedia and Citizendium could have justified their superiority on the exact same grounds that you've given: they favored quality over quantity. Quality is a virtue, but pursuing it in certain ways endangers the entire enterprise of a crowdsourced encyclopedia. The second is to consider our overall goal as a work of reference: we are here to present true (well, verifiable) and relevant information about the generalized and specialized topics that people look up here (IMO, anyway). Things like sourcing and policy compliance are metrics we use to estimate how well we're meeting that goal, because it can't easily be evaluated at scale. However, as anyone who's been part of a large, modern organization can testify, confusing the metrics with the goal itself can have major detrimental consequences. Choess (talk) 13:10, 22 September 2023 (UTC)[reply]
    Exactly. Not only could they, Citizendium did. They bragged they had a better model of quality over quantity. Veropedia was another one that did. The only Wikipedia competitor that is still around with a quality over quantity model that I can name off the top of my head is Encyclopedia Brittanica. Even then Wikipedia has made a severe dent in their market share, and they have had to change their business model to adapt to Wikiedia's dominance. I think the heart of this difference in philosophies is the assumption "but we can't have flawed articles out there, we could mislead the public." Twenty years of experience suggests that's not the case. The public knows Wikipedia's model of "the encyclopedia that anyone can edit" will result in potential flaws in it's articles. This is well known and discussed ad-nauseum all over the internet. Yet Wikipedia has thrived. The vastness of coverage in Wikipedia's articles makes up for the fact that a number of them have flaws; the flaws are known and tolerated. I would view it more like the market leader in many products, from cars to computer operating systems has many critics who point out the numerous flaws and longtime issues that have never been fixed. Yet they remain the market leader because in the big picture they still deliver what there competitors can't. That doesn't mean we ignore flaws in articles, we should fix them as we find them. I am saying Toyota shouldn't throw away it's market dominance to change business models because Volvo is perceived as having higher quality. Dave (talk) 15:50, 22 September 2023 (UTC)[reply]
    This is edging towards the age-old inclusionist–deletionist debate, which I'd prefer to avoid in this discussion (and my leanings are probably fairly obvious, anyways), but I will say this.
    Wikipedia needs stubs and we need coverage in obscure areas; even if the articles aren't perfect. And no article is "perfect".
    However, an unreferenced article is completely different from an unnotable article. Unreferenced articles are a poison that we could do without. We shouldn't force out anything that isn't a C-class article and devolve into Encyclopedia Britannica v2.0, but we also shouldn't be willingly letting in unreferenced content without a fight. This proposal isn't a speedy deletion criterion or anything so drastic; it gives plenty of time for the creator and other interested editors to improve the article by one very small step. Really, it's BLPPROD extended to all articles.
    Perhaps deletion is too drastic, per WP:PRESERVE. Perhaps this idea could work with BilledMammal's idea of "soft-deleting" (or "archiving").
    Fundamentally, I see the core idea of this proposal (minus any specifics) – and my intent in starting this discussion – as an extension of WP:BURDEN:

    Any material lacking an inline citation to a reliable source that directly supports [...] the material may be removed and should not be restored without an inline citation to a reliable source. [...] In some cases, editors may object if you remove material without giving them time to provide references. Consider adding a citation needed tag as an interim step.

    Happy editing, Edward-Woodrow • talk 22:30, 23 September 2023 (UTC)[reply]
  • Oppose as unnecessary for cleanup and irresponsible to our preservation goals. Certainly not appropriate for deletion; maybe in some cases for converting to an archive [if we had a non-draft archive space that didn't auto-delete after 6 months]. The equivalent mass deletion on Commons of all images without modern copyright tags was a mistake and an avoidable loss. – SJ + 02:14, 22 September 2023 (UTC)[reply]
    You bring up a good point, WP:PRESERVE is a policy.-- GreenC 17:00, 23 September 2023 (UTC)[reply]
  • The proposal sounds sensible in spirit, but it has bite potential, and I feel that retaining and training new editors is a more valuable pursuit than aggressively removing unreferenced articles. Barnards.tar.gz (talk) 16:16, 22 September 2023 (UTC)[reply]
  • Support. I think this is a really good idea, but I think there should be a delay before the article is proposed for deletion. — Preceding unsigned comment added by Professor Penguino (talk • contribs) 09:58, 22 September 2023 (UTC)[reply]
  • Oppose per Rschen7754 and against the spirit of WP:BEFORE. -- King of ♥ ♦ ♣ ♠ 01:47, 23 September 2023 (UTC)[reply]
    A reminder that This page is not for consensus polling. Stalwart "Oppose" and "Support" comments generally have no place here. Instead, discuss ideas and suggest variations on them. This is not the RfC on this proposed change. Curbon7 (talk) 02:02, 23 September 2023 (UTC)[reply]
    I'm intrigued by against the spirit of WP:BEFORE. Would you care to elaborate? I was assuming the nominator would do a quick WP:BEFORE check to see if the problem can be easily fixed, like one would for PROD or AfD. Edward-Woodrow • talk 12:18, 23 September 2023 (UTC)[reply]
    "I was assuming the nominator would do a quick WP:BEFORE" That is indeed the way it's supposed to work, however, a quick scan at some of the mass deletions, including mass deletions of every article that was ever created by an editor we've now deemed isn't worthy to be on this site, will reveal that is no longer how our deletion process works.Dave (talk) 16:42, 23 September 2023 (UTC)[reply]
    But this isn't a mass-deletion question. Edward-Woodrow • talk 16:51, 23 September 2023 (UTC)[reply]
    Based on my reading of the proposal, it allows deletion of articles simply for being unreferenced, not for being unreferenceable. If the nominator is supposed to conduct WP:BEFORE (and refrain from nominating articles where sources exist), then how is it not redundant to our existing PROD and AfD processes? -- King of ♥ ♦ ♣ ♠ 02:33, 25 September 2023 (UTC)[reply]
  • Strong oppose we already have a {{prod}} tag. This proposal is to hijack a different, useful maintenance tag to become a copy of that tag. And even the proposer agrees that deleting all 100k articles tagged as "unsourced" (some of which are lists that don't need sourcing) is infeasible.
    As far as the proposals to (via edit filters, etc.) require new articles to have at least one source at the time of creation: possibly, but I'll believe it's possible when I see it. 217.180.228.138 (talk) 04:02, 24 September 2023 (UTC)[reply]
    This is a good point, but there is one major difference between current WP:PROD and this proposal. PROD must be subject to WP:DEL-REASON; unreferenced articles are not explicitly included in that policy, and thus this would exist as a sort of exception, propped up by WP:BURDEN, in the same way that BLPPROD is an exception, propped up by WP:BLP. The closest DEL-REASON comes to unreferenced articles is a little more stringent:
    • Articles that cannot possibly be attributed to reliable sources, including neologisms, original theories and conclusions, and hoaxes
    • Articles for which thorough attempts to find reliable sources to verify them have failed.
    Cheers, Edward-Woodrow • talk 13:02, 24 September 2023 (UTC)[reply]
Story
It is 2023.
A: The earth is not a cube.
B: Add references please.
A: The earth is not a cube. You know it and I know it. It’s verifiable. Our children may not know it and we’re telling them now. Why do we need that?
B: Please provide references. It’s our policy.
A: Why don’t YOU add that if you think it’s needed?
B: The BURDEN is on you.
A: Oh the atmosphere here is toxic. The people here are . . . OKOK, the earth is a cube. It’s a cube. Okay? Please don’t chase after me. I won’t edit any more. You can do whatever you want.
B: . . .
--Dustfreeworld (talk) 06:40, 24 September 2023 (UTC)[reply]
...seems like a WP:SKYBLUE scenario rather than a relevant analogy, honestly. Edward-Woodrow • talk 12:57, 24 September 2023 (UTC)[reply]
But when we set down a "rule", we can expect it to get implemented mindlessly. Consider, e.g., BURDEN, which says you may (=are permitted but not required to) remove uncited content, while noting that if you're stupid, mindless, or POINTY about it, people will be angry with you, but which is twisted by someone basically every day of the week to say that uncited content must be removed, that it is an abomination to have even exclusively SKYBLUE information on a page if there isn't a source on the page, etc.
And, of course, if you think that a particular unsourced sentence is okay (Something vulgaris is a type of insect in the Something family"), and therefore an unsourced article containing only that would be okay, then there wouldn't be any point to this proposal. You'd check an individual article, do a quick WP:BEFORE search, and then either add a source or {{subst:PROD}} it through the existing process. WhatamIdoing (talk) 19:55, 25 September 2023 (UTC)[reply]
  • Oppose Category:Articles lacking sources I am surprised to see articles created this year on the list. Just don't let new articles be created that don't have references. I thought we had that already. Didn't we have that at one point in time? A bot should be created to post on the talk pages of all those who created these articles, with a link to where to find references and instructions. Specific instructions can be given such as if they are in the catagory for artists, mention that any musuem that has their artwork will have an official website that mentions them and being featured in a permanent musuem collection confirms the notability of an artist. Dream Focus 15:50, 25 September 2023 (UTC)[reply]
    @Dream Focus: Ah, so it's NPP's fault for letting in bad articles? Edward-Woodrow • talk 19:43, 25 September 2023 (UTC)[reply]
    During the time this discussion has been open multiple new/IP editors have been blocked as disruptive for creating continuously unreferenced articles, it's only established accounts that can get away with doing this. But they are a special case and shouldn't be subjected to community policies apparently. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 19:50, 25 September 2023 (UTC)[reply]
    Wait... are there articles from this year? The category just contains articles according to when they were tagged. Birel, for example, is listed as September 2023, but it was created in 2004. — Rhododendrites talk \\ 21:20, 25 September 2023 (UTC)[reply]
    Yes, currently 123 of them. —Cryptic 23:05, 25 September 2023 (UTC)[reply]
    I created one of those. List of fire departments in United States. Navigational list don't need references. Dream Focus 15:03, 27 September 2023 (UTC)[reply]
  • What about pages like weak operator topology? There's a reason that page is still unsourced after all these years: most of us are utterly unqualified to judge if any reference even supports any non-trivial claim in the article. The responsible thing is to leave that page to someone who actually knows what they are doing, however long it takes, instead of accidentally TURGIDAXing it in an attempt to protect it from deletion.
    And it's not just math and physics. What about subjects where all the likely sources are in a language I do not speak? Again, if there's a deadline to deletion, and no speaker of the language seems to be coming along, should I just use Google Translate and hope for the best?
    Instead of this broad proposal, what I might favor is expanding BLPPROD (or a similar process) to cover at least some subjects other than BLPs. For example, anything that might realistically cause real-world harm, e.g. an unsourced page about a medical treatment. And maybe some "spam-magnet" subjects, e.g. online content, commercial products, etc. Suffusion of Yellow (talk) 23:14, 26 September 2023 (UTC)[reply]
To anyone still subscribed to this discussion and interested: I have started an RfC at VPPR. Edward-Woodrow • talk 19:13, 6 October 2023 (UTC)[reply]
  • Support, but can they be draftified instead and then deleted after six months if inactive. This is fairer than straight-up speedy deleting them. Nominating them for deletion is also a good alternative. JacobTheRox (talk) 06:40, 18 October 2023 (UTC)[reply]
  • Oppose as unworkably disruptive. Also, we should have 100s of !voters at an RfC before signing off on new deletion system, not 1-2 dozen. Great idea in theory, though. We do have a big problem.
What we need is to recruit more editors to work this backlog down. Folks talk about an admin drought -- I think the big problem is not enough rank and file to research refs, which can be time-consuming. That's all the more reason to be welcoming and supportive of newcomers.
--A. B. (talk • contribs • global count) 02:53, 22 October 2023 (UTC)[reply]

Alternative Idea

How about instead we have it so that they have to be fixed within 30 days before they can be nominated for deletion? After they are nominated for proposed deletion sources must be added then within 14 days. There can also be a tag for "An editor has identified that the contents of this article is verifiable, but the article does not cite any sources. Please help improve this article. More information can be discussed on the talk page." The PROD tag could then only be removed as soon as a single reliable source is added. Sources in WP:RSPS determined to be "deprecated" can be discounted when assessing whether any sources are cited. Aasim - Herrscher of Wikis ❄️ 18:58, 20 September 2023 (UTC)[reply]

Comment - I doubt anyone is reading this far down but could maybe someone send a bot through to add == References == {{reflist}} to the 118,000 that don't have it? It would just make the repair work slightly easier for people going through and trying to get things up to par. Who knows it might even prompt someone to add a ref just because it irritates them to see an empty unfilled field for them? jengod (talk) 20:19, 29 September 2023 (UTC)[reply]

I suppose it couldn't hurt. Also, I like the idea of empty references sections glaring at the readers. Edward-Woodrow • talk 20:27, 29 September 2023 (UTC)[reply]
Yes, for some reason that imagery also appeals to me a lot. JoelleJay (talk) 05:17, 30 September 2023 (UTC)[reply]
I like this idea for a bot. It could check the last edit time to make sure to wait at least a few hours from the last edit. —siroχo 23:32, 29 September 2023 (UTC)[reply]
I think this idea for a bot to create references sections is a great idea. --Tryptofish (talk) 20:24, 30 September 2023 (UTC)[reply]
So do I. BilledMammal (talk) 20:29, 30 September 2023 (UTC)[reply]

Bot to create References sections

Continuing directly from the subsection just above, I want to pull out an idea that I think is promising, and I don't want it to be overlooked amid a long discussion. Repeating what was said just above, the idea would be to have a bot go through the 118,000 pages in the unreferenced category, and add == References == {{reflist}} to each page. This would do nothing in the way of deleting any content. It would help editors a little bit in getting sources onto the page. And it would be a constructive way to present editors with a prompt that references need to be added. The bot could also set a minimum amount of time since the last edit to the page, before adding the section, so as not to step on editors' toes when new pages are being created. Personally, I think this is a splendid idea. --Tryptofish (talk) 21:38, 30 September 2023 (UTC)[reply]

Some care will need to be taken in implementation, since some of the tags are placed incorrectly. Special:Permalink/1173684615 had a general reference present in a subheading named "Literature", for example. Also this bot will codify the project's preference for naming this subheading "References", which I'm not against but should be recognised as an outcome. Folly Mox (talk) 22:53, 1 October 2023 (UTC)[reply]
This should be done where the article has ref tags but no references section. It could also fix up where there is a header but no template or tag below to insert the references. Graeme Bartlett (talk) 03:00, 2 October 2023 (UTC)[reply]
I agree that there should be a reasonable amount of care regarding incorrectly tagged/categorized pages (although those pages are already incorrectly tagged and categorized). There's nothing to stop editors from subsequently changing the section header. But I disagree with limiting this only to pages that already have ref tags (which aren't unreferenced pages to begin with), although I'm fine with having the bot do clean up for improper formatting. Part of the idea (see comments in subsection above) is that there is actually a benefit to having a References section with no references in it. --Tryptofish (talk) 18:19, 3 October 2023 (UTC)[reply]
As one of those "cleaning up the mess", Adding the Ref. section from a bot insures it is correctly placed, i.e. After "See also" and Before "External links". I'm clueless how a bot can handle complexity of articles with/without those sections. Regards, JoeNMLC (talk) 19:26, 3 October 2023 (UTC)[reply]
  • If the reflist being added would be empty, the bot could also add a conspicuous call-to-action to help inform the author about what they need to do next. We could make the message hidden if references are added. Maybe also a date parameter, to say that the article will need to have references added within n days or whatever? I put together a mock-up to kind of demonstrate what I mean:
  • References
    It's time to add references to this article
    This article, {{PAGENAME}}, doesn't cite any sources yet. All Wikipedia articles must be supported by references to reliable, published sources.

    Learn how to add references   Edit this article

    Find sources: Google (books · news · scholar · free images · WP refs) · FENS · JSTOR · TWL

  • Cleanup/prod tags kind of do this already, but this could be less scary looking/bitey. It could also be a supplement to those tags. Well, it's just an idea! I hope this isn't too weird. 3df (talk) 06:53, 21 October 2023 (UTC)[reply]
We were talking about doing this to old articles, so the likelihood is that the creator retired years ago. The editors dealing with the problem are generally experienced and don't need instructions on how to add sources. Espresso Addict (talk) 08:06, 21 October 2023 (UTC)[reply]
@3df and @Espresso Addict, I do like the idea of adding that box to unref. articles. Below here is a version that I "tweaked" to include the article's age.
References
It's time to add references to this article
This article, {{PAGENAME}}, was created February 2003 (21 years ago) (2003-02), and doesn't cite any sources yet. All Wikipedia articles must be supported by references to reliable, published sources.

Showing that article age can motivate update/PROD/AfD action. Regards, JoeNMLC (talk) 01:29, 25 October 2023 (UTC)[reply]

@3df and JoeNMLC: I like this idea a lot. Perhaps we should move it to a new section? @Espresso Addict: I think it's readers and (potential) casual editors that this would target, not experienced editors. We unfortunately we don't have enough of those to fix 118,000 unreferenced articles any time soon. – Joe (talk) 06:29, 27 October 2023 (UTC)[reply]
Do readers/casual editors add (valid) references to articles? That's not common in my experience. Would this encourage them? I really don't know. @WhatamIdoing: for an opinion. In the past, I've tried to get academics to help out with various problems with articles, but have invariably found that the culture here puts them off. I've tried to get wikiprojects involved as well, but that has not been all that successful.
As to the backlog of completely unsourced articles (now 117.5k), it's been trending down consistently of late, which I believe has been down to the diligent efforts of a small number of experienced editors (and a lot of prods), presumably coupled with NPP not allowing new completely unsourced articles to enter mainspace in the first place. Espresso Addict (talk) 18:12, 27 October 2023 (UTC)[reply]
@Espresso Addict, AFAIK there has never been even one study on what it takes to get a banner off a page. @Steven Walling will remember whether any research was done on this point 10–15 years ago, and @Trizek (WMF) can tell us whether the Growth team has studied this, but AFAICT if the question is "Does spamming an {{unref}} tag on an article result in references being added by a newcomer?", the answer is "nobody knows" (and my guess is "no"). WhatamIdoing (talk) 20:58, 27 October 2023 (UTC)[reply]
I suppose the useful question to ask in context of the present discussion is whether the proposed banner is better, worse or the same as the standard orange-edged banner. But, more generally, my experience with asking new good-faith editors to add sources to stave off deletion is that at best they add blogs, unreliable genealogy-type websites and the like. Espresso Addict (talk) 21:04, 27 October 2023 (UTC)[reply]
I guess a follow on there, is once they add blogs and such, how likely is it that editors will add better sources? A blog is much more valuable than nothing to an editor looking to improve sourcing. I know I do try to improve sourcing some of the time, and having something to start from is easier than doing raw searches across multiple indices. Personally, I'm a huge fan of the collaborative, evolutionary model of Wikipedia, but I have noticed that there is a bit less of that in 2023 than there was back in the aughts. I don't necessarily mean that as an indictment, as I know we also need folks focused on FAs and GAs, but I also think the encyclopedia would benefit from more gnoming and such. /Ramble over. —siroχo 22:52, 27 October 2023 (UTC)[reply]
If the bot adds a ==References== section to a page that has ref tags, then that action could inadvertently mask section-blanking vandalism.
Also, just as a probably unwelcome but important point of fact, we don't technically have a rule that says all non-BLP articles must cite sources. An article that contains only the most basic information, so that none of it falls into WP:MINREF territory, is not technically required to have any citations at all. See User:WhatamIdoing/Christmas candy for an example of what's 'legal'. Sources must be WP:Glossary#verifiable, but your suggested text claims that they must be WP:Glossary#cited, which is not the same thing. WhatamIdoing (talk) 06:30, 27 October 2023 (UTC)[reply]
@WhatamIdoing: If I'm not mistaken the proposal is to this only to pages in Category:Articles lacking sources, so it should always have been checked by a human who has verified that the lack of sources is real and really a problem. – Joe (talk) 06:37, 27 October 2023 (UTC)[reply]
I checked the first 10 in Category:Articles lacking sources. Two had references, and needed the WikiProject banner to be updated. (A bot could do that.) One already had a ref section.
I checked the first item in the first 10 subsections of the randomly selected Category:Articles lacking sources from April 2020 (i.e., first starting with A, first starting with B, etc.). Four contained ref sections, one of which had no <ref> tags but did contain a bare URL to an article in The New York Times. One of the ref sections was named ==References and notes==.
This suggests that the bot would have to be moderately smart about detecting existing sections. WhatamIdoing (talk) 16:14, 27 October 2023 (UTC)[reply]

The concept of "preserve" is based on there being some non-trivial work product to preserve. Without references, these is no work product. It's like I throw a floormat into a garage and say that it is a yet-to-be finished automobile and saying to leave a space in their garage waiting for somebody to finish it. North8000 (talk) 00:24, 27 October 2023 (UTC)[reply]

Yeah... or it's just perfectly good, accurate encyclopaedic text without citations. One of the two. – Joe (talk) 06:23, 27 October 2023 (UTC)[reply]

Use of Wikidata values in infoboxes

Is there any possibility that infoboxes for US places could start using data from Wikidata? This could be used for population and area figures, for a start. This information generally comes from the US Census. If a bot could be developed on Wikidata to import population, land area, and water area figures from US Census sources to all states, county-equivalents, municipalities, and census-designated places, would this be a good idea? It could also be used to automatically update pages in other languages as well. Kk.urban (talk) 18:05, 23 September 2023 (UTC)[reply]

  • We have had a LOT of issues with the Verifiability (and thus the reliability) of Wikidata. We have had numerous discussions on how and when to use it, almost always deciding to limit its use. So… if it isn’t allowed now, my call would be to continue to not allow it. Blueboar (talk) 19:55, 23 September 2023 (UTC)[reply]
    Indeed, any such proposal would need to demonstrate that it would significantly improve data integrity and verifiability compared to what we have now. Failing that, it would be unlikely to overcome the fairly entrenched opposition to Wikidata use here on enwiki. —Kusma (talk) 20:07, 23 September 2023 (UTC)[reply]
    Of course, it could be vandalized, which is less likely to be caught than here. But this information could be imported by a bot, and an edit filter could be created to prevent editing of these fields except by admins or other approved users. If it was bot-generated, it would be correct and would not need to be modified. It's easier to control such things on Wikidata because of how Wikidata is structured (each piece of data is specifically attached to one property). Kk.urban (talk) 22:43, 23 September 2023 (UTC)[reply]
Pulling in data from Wikidata is one of those perennial proposals that has always been rejected. Except for one, Template:Infobox software. It would be super nice to have that kind of structured data from Wikidata be pulled into all Wikipedias' infoboxes on a subject, so that one change can go to all languages. SWinxy (talk) 22:24, 23 September 2023 (UTC)[reply]
The problem is that data gets pulled in whether it's referenced or not, a minor issue on somethings a potentially major issue for BLP details. Would be great if it was possible to edit here and have the changes backflushed to wikidata, but my understanding is that that project isn't going anywhere. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 22:31, 23 September 2023 (UTC)[reply]
I was thinking of this use only where the information all comes from one source. That's why I suggested census data. Information such as a person's birthdate or spouse can come from many different sources, thus importing it from Wikidata would be less trustworthy. Kk.urban (talk) 22:47, 23 September 2023 (UTC)[reply]
Would a bot be a better answer? A bot could be created to do the repetitive task of changes the data and refs, without any need to involve wikidata. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 22:57, 23 September 2023 (UTC)[reply]
Yes, that would be good. It would make it much faster. (Of course, it can always be vandalized, and this would not translate to other Wikipedias, but this would solve the main issue.) Kk.urban (talk) 23:01, 23 September 2023 (UTC)[reply]
Well there's Module:PopulationFromWikidata for Australian places. Despite being from Australia and its use on many articles I watch, I don't personally know much about it beyond what's on the module page though. Graham87 (talk) 08:16, 24 September 2023 (UTC)[reply]
Template:PH wikidata is used to populate Philippines-related infoboxes. (There have been some on-wiki developed similar solutions to this question, like Template:UN population, but these lack significant benefits compared to wikidata templates while having the same issues as wikidata templates.) CMD (talk) 10:31, 24 September 2023 (UTC)[reply]
I would agree on this wikidata. They are used to present information, also fill-in missing information that could not able to be inserted in descriptions, events, and photos. Therefore, there should be value in this data. 205.155.225.253 (talk) 15:52, 3 October 2023 (UTC)[reply]

Technically straightforward, including only showing referenced information (and even showing those references here, e.g., see South Pole Telescope). Community consensus wise, not there yet (as indeed this conversation shows). I'd encourage you to have a look at Commons or other language Wikipedias to see what Wikidata-powered infoboxes can do nowadays. Thanks. Mike Peel (talk) 19:17, 24 September 2023 (UTC)[reply]

I suggest looking at w:es:Jimmy Wales. That large infobox has only six parameters (one of which is his name) in the article. The rest comes from Wikidata. WhatamIdoing (talk) 05:37, 25 September 2023 (UTC)[reply]
I suggest looking at ones that get less scrutiny and more issues. Take e.g. Roger Vangheluwe, a Catholic bishop who had to resign after he admitted sexually abusing his 5-year old nephew. According to Wikidata, and thus also according to the infobox at w:ca:Roger_Vangheluwe, he was convicted, but this isn't true though, the crimes had happened too far in the past. And yes, even such filth comes under our BLP policies, and we shouldn't make claims which aren't true. Of course it is easier to fill an infobox from Wikidata and leave it in their hands. The claims were unsourced on Wikidata for years, and have been "sourced" since August 2022 to a newspaper article which makes no claim of a conviction[1]. Please leave Wikidata out of our infoboxes. Fram (talk) 08:58, 25 September 2023 (UTC)[reply]
But that kind of thing isn't what I was suggesting. The material I am suggesting would all come from one database/source. Kk.urban (talk) 09:01, 25 September 2023 (UTC)[reply]
I think that if the Wikidata is actively curated, and sourced, then it could be used. It is used to a limited extent on the Chembox template. But there is enough incompatible data that not everything is just accepted from Wikidata. There are problems with scope, source and accuracy (which could be wikidata or wikipedia's problem). Even if everything comes from one database, that in turn may not be 100% accurate, and scope may vary, eg which area is included in a population count. So I think we could use Wikidata more, but on a case by case basis. By the way vandalism is not frequent on Wikidata. Graeme Bartlett (talk) 22:33, 25 September 2023 (UTC)[reply]
Something I'd like to see us commit to as a prerequisite to any deeper Wikidata integration is significantly more useful error messaging from and documentation for the modules that pull from Wikidata. As it stands they're largely inscrutable black boxes that usually work, but always require expert assistance if they don't.
This broken map display from August, which I attempted to help out on, required updating something on OpenStreetMap (not documented, and the talkpage of the module responsible was a redlink). This referencing error from June, which I filed (improperly including a link in the header), required some troubleshooting due to inadequate documentation, although the fix was made on Wikidata rather than a third-party site.
I don't expect upstream errors to be as easy to address as in-house problems, but people shouldn't be expected to read and understand the codebase in order to fix them. Some sort of debugging checklist for error types, or even a "check value in property P at Wikidata item Q" should be the expectation. Folly Mox (talk) 23:24, 25 September 2023 (UTC)[reply]

Perhaps surprisingly, there are not just accuracy/data integrity, but also technical problems with using Wikidata, especially for using more than one Wikidata item in an article. See e.g. the recent discussion at Wikipedia:Village_pump_(technical)#Lua_freaking_out_at_Comparison_of_web_browsers. —Kusma (talk) 11:13, 26 September 2023 (UTC)[reply]

In answer to the OP's question, Module:WikidataIB exists for this purpose. All data pulled into infoboxes must be sourced to a non-Wikipedia source, per this 2018 RFC. – Jonesey95 (talk) 04:45, 16 October 2023 (UTC)[reply]

Wikidata is not merely inadequately curated. On the occasions when it is curated, it is sometimes robotically curated to require inaccurate claims to remain as claims, in order to keep its data consistent with other databases that make the same claims. It's ok for claims like "X person has id Y in database Z" but inadequately reliable for anything beyond that. —David Eppstein (talk) 15:34, 24 October 2023 (UTC)[reply]

Sports cleanup contest rather than mass removal of sports articles

tl;dr Should we have a contest to improve poor-quality sports articles?
I've recently been discouraged by seeing proposals such as WP:LUGSTUBS and WP:LUGSTUBS2 which are going to result in the ultimate deletion of many articles which are notable, and even more upsetting is seeing ideas to extend it to 40,000+ others and potentially even more after that, which would be mass destruction of so many notable articles, something I am deeply concerned at the thought of. As an attempt at compromise, I like the idea of having a backlog drive / contest (similar to the WikiCup or GA backlog drives, which each receive large participation) to improve the sports articles which are notable and nominate for deletion the non-notable ones - I've set out a rough draft of what this would look like at User:BeanieFan11/Global Sports Cleanup Contest format - there would be plenty of different awards (I've thought of many of the barnstars etc. that we have currently and am thinking about designing others for this contest) and, if widely advertised, I suspect this would receive a large participation and would have improved many more articles than the current mass draftification RFCs would (also see Wikipedia:WikiProject National Football League/Football biography cleanup, which has improved several times the number as has the RFCs - even though it doesn't have awards or any of that). Is there support for such an idea? BeanieFan11 (talk) 12:47, 29 September 2023 (UTC)[reply]

  • Is there support for such an idea? I don't oppose it, and I'm willing to support it in whatever way is needed, but I'm not convinced it will work - the only realistic way, in my opinion, to address disruptive mass action is through mass action.
Regardless, to assist I have created the following lists, consisting of all Lugstubs potentially eligible for the LUGSTUBS process[a]:
  1. WikiProject Athletics/LUGSTUBS
  2. WikiProject Australian Sports/LUGSTUBS
  3. WikiProject Badminton/LUGSTUBS
  4. WikiProject Basketball/LUGSTUBS
  5. WikiProject Biathlon/LUGSTUBS
  6. WikiProject Boxing/LUGSTUBS
  7. WikiProject Canadian sport/LUGSTUBS
  8. WikiProject Cricket/LUGSTUBS
  9. WikiProject Cycling/LUGSTUBS
  10. WikiProject Fencing/LUGSTUBS
  11. WikiProject Figure skating/LUGSTUBS
  12. WikiProject Football/LUGSTUBS
  13. WikiProject Golf/LUGSTUBS
  14. WikiProject Gymnastics/LUGSTUBS
  15. WikiProject Handball/LUGSTUBS
  16. WikiProject Ice Hockey/LUGSTUBS
  17. WikiProject Martial arts/LUGSTUBS
  18. WikiProject Olympics/LUGSTUBS
  19. WikiProject Pakistan Super League/LUGSTUBS
  20. WikiProject Professional wrestling/LUGSTUBS
  21. WikiProject Referees/LUGSTUBS
  22. WikiProject Rowing/LUGSTUBS
  23. WikiProject Rugby union/LUGSTUBS
  24. WikiProject Running/LUGSTUBS
  25. WikiProject Sailing/LUGSTUBS
  26. WikiProject Skiing and Snowboarding/LUGSTUBS
  27. WikiProject Softball/LUGSTUBS
  28. WikiProject Speed skating/LUGSTUBS
  29. WikiProject Sport/LUGSTUBS
  30. WikiProject Swimming/LUGSTUBS
  31. WikiProject Tennis/LUGSTUBS
  32. WikiProject Triathlon/LUGSTUBS
  33. WikiProject Volleyball/LUGSTUBS
  34. WikiProject Water sports/LUGSTUBS
  35. WikiProject Women's sport/LUGSTUBS
I have also created Wikipedia:WikiProject Olympics/Biographical substubs; this isn't limited to Lugnuts creations but covers all Olympic biographies that are less than 2500 bytes, whose sources are believed to not contribute to WP:SPORTCRIT #5[b], and which have had no significant contributors other than the creator. Note that while most of these are mass created, many are not; the ones that are not mass created are more likely to be notable than those that are and while they will never be subject to the LUGSTUBS process they may be worth focusing on as the ones with the greatest potential. BilledMammal (talk) 14:18, 29 September 2023 (UTC)[reply]
@BilledMammal: Thank you for those lists - do you think you could also create lists / tables for non-Lugnuts created "substubs" in each sport (or just lists including all creations would work)? Thanks! BeanieFan11 (talk) 14:35, 29 September 2023 (UTC)[reply]
The unformatted data can be found here; I don't have time to compile it into tidy tables at the moment, but I'll try to do so in the next few days unless someone beats me to it. For those who are interested, a list without duplication and without the associated sports listed can be found here. There is a caveat with this list, and that is that my confidence in the list of non-contributory sources is low, as I didn't have much time to spend reviewing them. If another editor wants to go over my list and try to identify any sources that should not be there they are welcome to do so; a complete list of sources found in short sports biographies can be found here, and again any editor wishing to review the list further than I did - I only reviewed the first page - is welcome to do so.
I find it interesting to note that there are only 15,404 sports biographies created by editors other than Lugnuts that are under 2,500 bytes, have no significant edits by editors other than the creator, and are believed to have no sources suitable to meet WP:SPORTSCRIT #5. In contrast, there are 43,357 articles by Lugnuts that meet that criteria; I think this really emphasis just how low quality Lugnuts creations were, and why we have to do something about them.
lists / tables for non-Lugnuts created "substubs" in each sport (or just lists including all creations would work) If you want lists/tables that include both Lugnuts and non-Lugnuts creations, that is easy to do; let me know. BilledMammal (talk) 15:47, 29 September 2023 (UTC)[reply]
@BilledMammal: I think tables for just non-Lugnuts creations by sport would be best, but ones with both Lugnuts and non-Lugnuts creations would be fine as well if you want - I must say I am very surprised that there's only 15,000 that meet the Lugstubs criteria compared to 40,000 by him; that really doesn't sound right, considering there's alone 200,000 stub association football articles, 130,000 in the Olympics, 30,000 in cricket, close to 50,000 in college football (although a fair chunk of them are seasons) and 16,000 in the NFL (the latter two of which couldn't have been created by Lugnuts) - are you sure that quarry query is working properly? BeanieFan11 (talk) 16:09, 30 September 2023 (UTC)[reply]
I think so; I just manually check a couple dozen cricketers, and all the ones who I expected to be on the list were. BilledMammal (talk) 12:04, 1 October 2023 (UTC)[reply]
Though Lugnuts wasn't a troll per se, I think WP:DENY has some relevance here. If we're doing individual improvement rather than mass cleaning up of one editor's mess, then we shouldn't make it about that one editor. Just lump them in with the rest. I'll propose that instead, BLP articles should be separated as higher importance. I've tried to do un referenced BLP clean up this year, and sports stubs (Lugnuts and otherwise) have made it way more difficult than it should be. And I'll reiterate, I think a detailed "List of [athletes that meet X criteria]" is a far better solution than hundreds of individual one-sentence or one-paragraph articles. Thebiguglyalien (talk) 16:57, 30 September 2023 (UTC)[reply]
Any possibility of adding some kind of nationality into the data? I have better-than-usual access to historical Finnish newspapers and (if the amount of Finns in the list is even vaguely reasonable) could take a stab at those. Ljleppan (talk) 11:47, 1 October 2023 (UTC)[reply]
@User:BilledMammal: Would it be possible to divide Wikipedia:WikiProject Cricket/LUGSTUBS by nationality? In particular, would it be possible to have a list of the Australian cricketers? James500 (talk) 18:20, 22 October 2023 (UTC)[reply]
  • Note the case of Viggo Jørgensen, which I found randomly clicking in one of BilledMammal's Olympians list; created by Lugnuts, but recently radically improved by Mdm.Bla. How do we encourage more of that kind of useful work?
And in a separate thought, it would be a good idea to highlight sportswomen. The Women in Red folk might be willing to have a specific drive to improve these, though generally they work on creation. Espresso Addict (talk) 22:33, 10 October 2023 (UTC)[reply]
@Espresso Addict: Working through the tens of thousands of articles just like Viggo Jørgensen's unedited state (and that of his teammate Aage Jørgensen [no relation], whose page I also improved to a lesser extent) is likely being seen as somewhere between impossible and possible, but a waste of time, hence the push for mass draftification. Maybe something like adding a link to these articles in WP:JOBS#Writing?
The good news is that non-database sources are findable. I relied heavily on WP:LOONA#Denmark for sourcing these two articles, and I suspect that there are hundreds of Lugstubs that could meet the WP:SPORTSBASIC criteria using LOONA/another newspaper archive. I am also a fan of Ljleppan's idea of adding nationality data, as I believe that organizing these articles as much as possible will make their improvement much more likely. mdm.bla 02:01, 11 October 2023 (UTC)[reply]

Notes

  1. ^ Initial assessment only; these lists are likely to contain many false positives that would be removed before draftification would be proposed. However, even the false positives are in need of significant work, and so I don't see listing them on a cleanup project as an issue.
  2. ^ Initial assessment only; at some point a more detailed assessment will be made of these sources

New Vector 2022 RfC

I have been talking with editors on Wikipedia talk:Requests for comment/Rollback of Vector 2022 about a new Vector 2022 RfC, and it seems as if there is support for a new RfC. Pinging most editors involved in that discussion: @InfiniteNexus, @Certes, @Tvx1, @Æo, @Hemiauchenia, @Nikkimaria, @Casualdejekyll, @Qwerfjkl, @Tenryuu, @Aaron Liu, @Toa Nidhiki05, @Yngvadottir, @TheMissingMuse, @Cessaune, @BilledMammal, @Radlrb, @Randy Kryn. Please @ anyone else that I might have missed.

Let's use this discussion to build the next Vector 2022 RfC. I know there will be pleasure and displeasure about the skin, but we should save those arguments for the RfC. Whatever that RfC might cover. We also shouldn't be beating a dead horse, rather we should think about questions we can ask. Personally, I am more inclined to ask open ended questions rather than simple yes or no questions, like Wikipedia:Talk pages consultation 2019/Phase 1, but I am fine with yes or no questions. BTW the reason I did not move that discussion is because WP:INVOLVED but I do think a higher traffic page meant for ideation is a better place to discuss this idea rather than an obscure talk page. Aasim - Herrscher of Wikis ❄️ 20:14, 3 October 2023 (UTC)[reply]

BTW I started building something in User:Awesome Aasim/Vector 2022 Feedback Survey as this is the idea I have. Do you think my questions would get specifics? We can always alter the suggested survey questions. I also put a section for !voting on ideas. Aasim - Herrscher of Wikis ❄️ 21:09, 3 October 2023 (UTC)[reply]
Good initiative! The idea of a RfC with multiple questions is perfect, and the proposed questions are very intelligent. I would still like to see another yes/no question about reverting to V10 added to the list. Let's see what are the opinions of others. Æo (talk) 21:54, 3 October 2023 (UTC)[reply]
I made that Idea 5, which was originally about changing the default skin to a community based one, but now it is about what should be the default skin. But now it can be a little more open and we can even have other ideas brought as well, including other mw:Skins. Aasim - Herrscher of Wikis ❄️ 00:39, 4 October 2023 (UTC)[reply]
By my count, this would be the eleventh major RfC or request for community feedback on the new skin, after requests for feedback on the first (March 2020), second (July 2021), third (November 2021), fourth (March 2022), and fifth (June 2022) prototypes by the web team, the pre-deployment VPP discussion (July 2022), pre-deployment RfC (September 2022), post-deployment RfC (January 2023), close review of the post-deployment RfC (March 2023), and follow up discussion on the fixed width issue. This horse isn't just dead, it's been sent to the glue factory and turned into Ikea furniture. – Joe (talk) 07:38, 4 October 2023 (UTC)[reply]
Let's say that the horse has been shotten dead multiple times instead of having died by natural causes. Æo (talk) 09:34, 4 October 2023 (UTC)[reply]
Let's focus on the two main discussions where people actually discussed Vector 2022. There is the pre and post deployment RfCs, neither of which had any consensus to deploy or to rollback. That bridge was burned six months ago. We have extensively discussed whether we should revert the skin or not. What we can discuss is the next steps for the skin now that it is deployed. A quality review would be the best bet for what to do next. Aasim - Herrscher of Wikis ❄️ 14:15, 4 October 2023 (UTC)[reply]
There was a consensus to deploy, as assessed by two experienced closers ProcrastinatingReader and ScottishFinnishRadish: if all the concerns outlined above are satisfactorily addressed then we see community support to roll out the change. There was then no consensus to un-deploy it, implying that the previous consensus still stands. – Joe (talk) 06:45, 5 October 2023 (UTC)[reply]
I don't see how that can be read as consensus to deploy. "if all the concerns outlined above..." refers to "fixed-width...clearly visible and available to both logged-out and logged-in users", "non-intuitive icons in the sticky header", and "the behaviour of the language selector". The fixed-width toggle was only sorted out after deployment, while the sticky header icons and the language selector were (and remain) completely unchanged. CMD (talk) 07:01, 5 October 2023 (UTC)[reply]
They were addressed here. Changes were made to the fixed width behaviour and the language toggler; I'm not sure about the icons, but it's also not clear from the closing statement what exactly was wrong with them. On a consensus-based project, "address" does not mean "do exactly as we say", it means discuss further and reach a satisfactory compromise. If the community as a whole was unsatisfied with the compromise that was reached, it could have expressed that in the post-deployment RfC, but it didn't. – Joe (talk) 07:19, 5 October 2023 (UTC)[reply]
I'm not seeing how that discussion shows consensus the concerns were addressed. There was in fact much "expressed" in the discsussion you link. If the issues were addressed, then there would not have been what appeared to be a very rapid rush to address the issues with the fixed-width behaviour after Vector 2022 was already deployed. CMD (talk) 07:41, 5 October 2023 (UTC)[reply]
I don't think it really matters when exactly they were addressed? The point is that the WMF made a good faith effort to address them by changing some things and justifying why they would not change other things, and the community, given the chance to reject that response as unsatisfactory, did not. – Joe (talk) 08:03, 5 October 2023 (UTC)[reply]
I think one of these closers commented somewhere that, in their opinion, the concerns were not satisfactorily addressed. I'm sure someone can find the diff. –Novem Linguae (talk) 07:41, 5 October 2023 (UTC)[reply]
That was a conversation on ProcrastinatingReader's talkpage where both closers noted that reading their close as consensus to deploy was inaccurate, and that the concerns raised were not satisfactorally addressed. CMD (talk) 07:54, 5 October 2023 (UTC)[reply]
  • Specific questions or requests for improvement can be made without a broad RFC that will just rehash old arguments. Users who dislike the skin are free to use a different one, just as they were with the last change and will be with the next change. 331dot (talk) 09:55, 4 October 2023 (UTC)[reply]
    This is not about the personal preference of single users, but for voicing concerns of the community as a whole (or at least that 65% majority of users who were against V22 in the January 2023 RfC), and for the good of the whole Wikipedia project. For example, I already use V10 as set in my personal options but I still think that V22 is detrimental for the project as a whole, and that even anonymous editors and readers should be given the possibility to switch back to V10. Æo (talk) 10:21, 4 October 2023 (UTC)[reply]
    People who dislike change are always more vocal than those who like a change. This has been reviewed to death, the RFC was reviewed, there was a review of the RFC, a review of the review of the RFC, ad nauseum. There needs to be some finality over this general issue. Specific concerns should certainly be discussed, like availability for unregistered users(it's my understanding there are technical reasons for that, but I may be wrong), but in general, can we move past this? At this rate we will still be discussing it when Vector 2032 comes out and then the cycle will start overy. There is no way to satisfy everyone on this planet. 331dot (talk) 17:29, 4 October 2023 (UTC)[reply]
    I personally like change but V22 was a change in a completely wrong direction. By "review" you probably mean that the first and second RfCs were "closed". Æo (talk) 18:04, 4 October 2023 (UTC)[reply]
    You are certainly entitled to your opinion- and I truly hope that you expressed it in the numerous prior discussions. 331dot (talk) 19:02, 4 October 2023 (UTC)[reply]
    I assume that you are aware nothing on Wikipedia is a straight vote. 331dot (talk) 17:31, 4 October 2023 (UTC)[reply]
    Yet there was a clear majority against V22. Æo (talk) 18:05, 4 October 2023 (UTC)[reply]
    A good amount of which was WP:IDONTLIKEIT and ignored the research and study groups previously performed. Arguments are weighed in a discussion. Hey, if you don't like it, fine with me- but the horse is dead, buried, encased in concrete, yada yada. At some point we need to move on(especially since there are alternatives like switching to the old one), before we know it, we will be doing this all over again with Vector 2032. 331dot (talk) 18:57, 4 October 2023 (UTC)[reply]
    Yes I recognise that. That is why I structured the questions to be less focused on proposals and more on general questions. Sure I do have some ideas for next steps. Specifically phab:T106463. But I am also sure that some of the other ideas I put there would get moderate support. Aasim - Herrscher of Wikis ❄️ 14:26, 4 October 2023 (UTC)[reply]
We also shouldn't be beating a dead horse - then might I suggest not doing this? firefly ( t · c ) 16:37, 4 October 2023 (UTC)[reply]
That is in reference to the deployment and rollback RfCs. Neither of those RfCs were able to get consensus for deployment or rollback. However, the closers believed that if the concerns about deployment were addressed then there might have been consensus for deployment. I do want to propose the enabling of Vector 2022's "responsive mode" which so far has not been done and is the reason I am using Timeless right now instead of that skin (so far checking "responsive mode" in Preferences has no effect on Vector 2022). When that comes I certainly will use Vector 2022. Aasim - Herrscher of Wikis ❄️ 18:43, 4 October 2023 (UTC)[reply]
  • I do not think having a 10 question RFC on this issue (or most issues) is a good idea. That seems to lack focus and may not be a great use of community time. –Novem Linguae (talk) 18:43, 4 October 2023 (UTC)[reply]
    Then... should I just focus on individual steps? So individual changes to the Vector 2022 skin? Because as I said above rolling back the skin is off the table. We have essentially talked that to death. The main thing I really think should happen right now is enabling Responsive Vector 2022 (see phab:T106463 and phab:T291656 and phab:T319305) (or at least the checkbox in preferences), but I did put some other questions up about the ToC, etc. There absolutely should be more civility and less hostility towards the WMF as an organisation as, even if they make a mistake, they are likely doing it out of negligence not out of actual malice. Aasim - Herrscher of Wikis ❄️ 19:43, 4 October 2023 (UTC)[reply]
    Unlimited width as default achieved consensus during the rollback RFC but was never actioned, right? With that in mind I do not have high hopes that further RFCs would lead to software changes, and therefore I do not recommend them.
    Speaking more generally, to craft a successful RFC of any kind, I would recommend a laser focus. Just one or two questions, separated into different sections, encouraging a binary answer to make things easier for the closer, and very specific ("Should we X?" or "Should we ask the WMF to X?") –Novem Linguae (talk) 21:02, 4 October 2023 (UTC)[reply]
    Yeah, I can see how that was completely pointless. WMF's solution was to have width be persistent. That is a good compromise.
    However, as seen on Phabricator, there is a bigger push to roll out responsive mode. The ticket with the highest number is the one created by User:AHollender (WMF) (before they no longer worked for WMF). Aasim - Herrscher of Wikis ❄️ 23:09, 4 October 2023 (UTC)[reply]
The way the WMF handled the Vector rollout was irresponsible and short-sighted. But I really doubt that there's any will at this point to undo it, which would likely be as painful of a process. Our focus is better spent on bringing WMF actions that affect the English Wikipedia under increased scrutiny, which is what's being done right now with WP:2023 WMF RfC Hub. Thebiguglyalien (talk) 19:33, 4 October 2023 (UTC)[reply]
I am genuinely curious as to what they should have done instead. They spent years developing it, more time testing it, asked focus groups, and solicited community input. The WMF can't ask every Wikipedia user on this planet what they think, nor can they design a skin acceptable to every user. The grievance from opponents seems to be that the WMF didn't ask them personally. 331dot (talk) 19:49, 4 October 2023 (UTC)[reply]
Ditto. The rollout of V22 was exemplary: as I've listed above, the web team organised seven rounds of community feedback over the course of nearly three years before deployment, made numerous changes in response to that feedback, did not launch until they'd addressed all the issues that the final pre-deployment RfC found to be community 'blockers'. If anything, I think they were too solicitous to the community, considering that frontend design is more within the WMF/MediaWiki developer remit than enwiki's. – Joe (talk) 06:51, 5 October 2023 (UTC)[reply]
I strongly dislike Vector 2022. I dislike the way it is forced on readers – I think all readers should choose what skin they want to read Wikipedia in.
On the other hand... give that poor horse a break. Edward-Woodrow • talk 19:48, 4 October 2023 (UTC)[reply]
They can choose, they can create accounts. There is no entitlement for IPs to use a particular skin any more than guests to my home can demand that I arrange my furniture to their preference. 331dot (talk) 19:51, 4 October 2023 (UTC)[reply]
So essentially we're saying register an account just so you can read Wikipedia in a usable (i.e., non-Vector2022) way? No thanks. Their own choice doesn't affect us; they're not demanding we re-arrange their furniture, they're choosing to sit in the chair they choose. We shouldn't grab them by the shoulders and force them to pick a chair they might not like. Edward-Woodrow • talk 19:54, 4 October 2023 (UTC)[reply]
"Useable" is a matter of opinion, of course. No one's forcing anyone to do anything. "Welcome to my home. Please sit here for the moment. If this chair isn't to your liking, I have others available". They aren't entitled to pick the chair before they arrive. 331dot (talk) 20:08, 4 October 2023 (UTC)[reply]
Except we don't have others available. At least, not to – *sniff* – non-editors. Edward-Woodrow • talk 20:09, 4 October 2023 (UTC)[reply]
Non editors can create accounts to save their preferences. It's my understanding that there are technical issues with allowing IPs to choose a skin. It would involve cookies, for one(I think). 331dot (talk) 20:12, 4 October 2023 (UTC)[reply]
That's unnecessarily complicated. Not everyone wants to create an account just to read Wikipedia in a skin of their choice. Would you want to log in every time you just want to read an article? Edward-Woodrow • talk 20:14, 4 October 2023 (UTC)[reply]
I accept that you and others may see it as complicated but it takes seconds and people can save their login information. I can't design Amazon as I like it, or my bank's website. We are fortunate to have choices at all. 331dot (talk) 20:21, 4 October 2023 (UTC)[reply]
Technical issues right now, yes. But it doesn't have to be, but it would require a fundamental rethink of how MediaWiki handles caching. At least from what I heard in past discussions. Aasim - Herrscher of Wikis ❄️ 21:51, 4 October 2023 (UTC)[reply]
If you're logged out but browsing on mobile, you still get Minerva. Most of our logged out readers connect via mobile. I won't dissuade anyone against talking something to death – our favoured community pastime – but I don't foresee another Vector 2022 conversation bearing fruit. Folly Mox (talk) 22:25, 4 October 2023 (UTC)[reply]
I agree that further Vector2022 discussion will probably be pointless. Edward-Woodrow • talk 22:26, 4 October 2023 (UTC)[reply]
@Folly Mox: I have never edited via mobile apart from making tiny corrections. It is simply impossible to do serious editing by mobile. I don't get your point. Æo (talk) 23:46, 4 October 2023 (UTC)[reply]
You're not alone! My point was in reference to the argument that logged out readers don't get to choose their default skin: since most don't see Vector 2022, any argument that Vector 2022 should be improved for their convenience has to take into account the actual audience. But like, if yall want to hold a ten piece RfC about Vector 2022 again, have at! People love RfCs. Folly Mox (talk) 00:31, 5 October 2023 (UTC)[reply]
I'll just raise the next logical step: enabling responsive Vector 2022 so that mobile users can, if not get the perfect editing experience, at least get one that is functionally identical to desktop while not having wonky problems with zoom and fat finger. Aasim - Herrscher of Wikis ❄️ 03:14, 5 October 2023 (UTC)[reply]
Could you give an example of another large website that gives non-logged in users the ability to choose a skin? – Joe (talk) 06:52, 5 October 2023 (UTC)[reply]
Reddit has maintained access to its old skin. Can't find any overall numbers, but looking at a few thread discussions the old one is used by ~5% of traffic, with the new skin used by ~25% (much like Wikipedia both are dwarfed by mobile/mobile app, although unlike Wikipedia it's much easier to participate in reddit through mobile). CMD (talk) 07:11, 5 October 2023 (UTC)[reply]
That's a good example, thanks. As far as I know the only way to access Reddit with the old skin is with an altered URL: old.reddit.com. You can actually alter Wikipedia's skin by altering the URL too (e.g. en.wikipedia.org/wiki/Main_Page?useskin=vector) but because it's a parameter not a different domain it doesn't persist between pages. I wonder if anyone has ever asked the WMF about the feasability of e.g. en-vector.wikipedia.org? – Joe (talk) 07:25, 5 October 2023 (UTC)[reply]
The option has been raised; iirc it was not looked on favourably as it would require caching multiple versions of each page. (There were more detailed explanations but unfortunately I don't remember from who or where.) CMD (talk) 08:05, 5 October 2023 (UTC)[reply]
Yes, the specific issue raised is that it would require a rethink of the caching system. Currently only complete MediaWiki pages, skin and all, are cached. But to have IP skin preferences will require the skin preference being stored somewhere, probably in local storage, and will result in a flash of unstyled content on almost every page load. It is not like that is a bad thing; many web apps have this flash of unstyled content. But it is something to be aware of. Aasim - Herrscher of Wikis ❄️ 13:16, 5 October 2023 (UTC)[reply]
Apparently there was a MediaWiki extension to persist ?useskin= (though not maintained since 2008, mw:Extension:PersistUseskin), and of course it's possible to do via the browser as well. Same issues apply as well of course. Alpha3031 (t • c) 13:45, 6 October 2023 (UTC)[reply]
The last question seems off topic to me, and more appropriate in the context of the WMF-enwiki RfC. Im my opinion, the other questions seem a bit biased.
Here are the questions I would ask.
  • What do you like about Vector 2022?
  • What do you dislike about Vector 2022?
  • What features, if any, would you like to be added to/removed from Vector 2022?
  • On the whole, how do you feel about Vector 2022? This could simply be a generic scale (love it, like it, neutral, dislike it, hate it) with an accompanying short paragraph. Cessaune [talk] 14:58, 12 October 2023 (UTC)[reply]

Drafting an RfC for whether news coverage counts toward GNG

WP:GNG and WP:NEVENTS say that for an event to be notable, it requires sustained significant coverage from secondary sources. As the guidelines stand, just being reported in the news does not establish notability, because that fails sustained and secondary. But for some reason, there are thousands of stub articles that just describe a news story without any meaningful encyclopedic coverage. They're sourced exclusively to news articles, and more are created just about every day. Traffic incidents, fires, mass murders, and explosions are common offenders, among others. These events simply don't have the coverage that warrant a separate article, but they get enough "it was in the news once so it's notable" that it brings an AfD to no consensus. That happened at this AfD, where the closer recommended an RfC. So here we are.

I don't know exactly what the wording for an RfC should be, which is why I'm posting it at the idea lab. But essentially the idea would be to determine whether being sourced exclusively to news sources is enough for an event article or if that's a valid reason for deleting/merging/BLARing. Other problems that might be addressed include: what WP:SUSTAINED means, whether local coverage is significant coverage, and how to respond to an article about an event that happened in the last few days.

Pinging everyone who participated in the last discussion on this: EvergreenFir, Silver seren, SMcCandlish, Orange Suede Sofa, Masem, BilledMammal, Moabdave, Newimpartial, Davidstewartharvey. Whatever result we come up with in this discussion, I'm expecting to post it as an RfC at WP:VPP before this goes stale like the last one did. Thebiguglyalien (talk) 05:13, 5 October 2023 (UTC)[reply]

Wikipedia talk:Notability, imo. Andre🚐 05:26, 5 October 2023 (UTC)[reply]
A key point that is regurgitated in both GNG and NEVENTS is that notability is a rebuttable presumption. An event article may be created the day it happens, with news coverage giving to the GNG in the short term, but well after the event has completed and there's no signs of any enduring coverage, then the notability can be fairly challenges, and the claim that "but there were newspapers article that meet the GNG!" should no longer hold true due to the lack of enduring coverage. We need to be clear that passing the GNG once doesn't mean it can be challenged later.
Also, what came up in that AFD is the aspect that much of coverage from news can be seen as routine, which is specifically called out in the GNG. While a crime like that shooting may not itself be routine, the coverage of it was routine (a burst of coverage at the start, but very little after the fact). Contrast to racially-motivated shootings or school shootings that have very long tails of coverage of how to stop those types of crimes. That should be a major consideration but requires editors to get their heads out of trying to capture every news detail and instead thing about topics that will be of interest 10-25 years from now. Masem (t) 12:36, 5 October 2023 (UTC)[reply]
Contrast to racially-motivated shootings or school shootings that have very long tails of coverage of how to stop those types of crimes. Even for many of these (but not all, of course) the "long tail" referred to seems likely to be about the politics, with mention of the event itself just an excuse to bring the politics up. Anomie⚔ 21:53, 6 October 2023 (UTC)[reply]
Certainly, if there's such an event that has a normal short tail of coverage, but some local politicians try to use the event to push their own works, that's not the same as a long tail of coverage. We're expecting, for example, secondary analysis on the event or further significant events that results from that one, not just name dropping particularly for political plays. Masem (t) 00:09, 9 October 2023 (UTC)[reply]
As you mention Alien, it's kind of already written into the relevant PAG that primary and routine coverage are not suitable for establishing notability. If this no longer reflects AFD practice, then either AFD or the PAG no longer reflect the community consensus. I'm not sure anyone would even read more detail guidance, so the most an RFC could do is to clarify which one is the case. Alpha3031 (t • c) 09:02, 7 October 2023 (UTC)[reply]
Coming back to this, I think we could use a better definition of ROUTINE though. Alpha3031 (t • c) 03:36, 12 October 2023 (UTC)[reply]
Actually, yeah, I like that idea. Andre🚐 03:50, 12 October 2023 (UTC)[reply]
There a couple of distinct questions here. "what WP:SUSTAINED means" is a different question to "how to respond to an article about an event that happened in the last few days". An event that happens in the last few days is by definition going to lack sustained coverage, yet we create them anyway (with I suspect broad community acceptance). There is an intersection in that the question of SUSTAINED eventually comes up for news events, and if that is the target question it would be best addressed with a narrowly tailored RfC that doesn't go into related wider questions. CMD (talk) 10:31, 7 October 2023 (UTC)[reply]
RFCs about changing a (single) policy or guideline almost always happen at on the talk page of the page that will be changed.
Perhaps a sensible question to ask is how long "a sufficiently significant period of time" is. For example, if the event (e.g., sports game, natural disaster, election) happened at the start of April 2010, is it sufficient if news/other sources write about it:
  • the month in which the event happened only?
  • the year in which the event happened only?
  • occasionally during the next (two? – five? – ten?) years?
  • according to a scheme that is not primarily related to time (e.g., the next time either of those teams reach the championships, which could be next year, or it could be 20 years from now)?
Were editors to say, e.g., that sustained coverage requires five years, then we would presumably permit the article for (at least) five years, and then consider it at AFD during year #6. This is probably a point that you'd want to make clear from the start, because there is no appetite in the community (or from our readers) to exclude content about recent events, and if people think you are proposing that the 2024 United States presidential election or the 2024 Summer Olympics or the Death of Queen Elizabeth II can't be written about until five years after they happen, they'll stop reading and vote against your proposal without a second thought. WhatamIdoing (talk) 19:41, 8 October 2023 (UTC)[reply]
Yup. Andre🚐 20:47, 8 October 2023 (UTC)[reply]
This would affect both WP:N and WP:NEVENTS (and possibly WP:OR and WP:RS if the critical primary/secondary issue is addressed). I think "length of time" is a red herring. To me, a logical reading of WP:SUSTAINED/WP:PERSISTENCE is that the source needs to be retrospective in nature based on previous reporting (i.e. a secondary source). "Let's look back at the earthquake that happened three months ago" would be sustained coverage while "new development about building code violations in the earthquake six months ago" would not be sustained coverage because it's still reporting on the event. Thebiguglyalien (talk) 21:02, 8 October 2023 (UTC)[reply]
Disagree. Firstly, the line between a primary source and a secondary source is blurry. A primary source is generally unambiguously primary if it's a direct source, but when it comes to at-the-time news reporting, then it starts taking on more secondariness as time goes on, and I would argue it's not wise or practicable to make that secondary source line a bright one. Nowadays, since reliable outlets are releasing new content every day, there is going to be primary and secondary material about recent and past events every day. E.g. 1 article that's mostly just breaking news facts and interviews about the earthquake, which is essentially a completely primary source. 1 article that is a retrospective aggregation analysis about earthquake coverage. That could definitely land on the same day, week, or month easily. Andre🚐 21:08, 8 October 2023 (UTC)[reply]
If you have a secondary source, then you have a secondary source. Get one or two more of those, and it meets GNG. Even if there are still primary sources being written. If I understand WP:OR and WP:RS correctly, the whole point of requiring secondary sources is because that's what's needed to write a proper article. All that should really matter with notability is the existence of sources which could reasonably be considered secondary. The problem I wish to solve is that this primary/secondary distinction has been lost in favor of an "any coverage at all" standard that results in newspaper-like articles for events that will never see serious retrospective coverage. Thebiguglyalien (talk) 21:26, 8 October 2023 (UTC)[reply]
I agree it's a potential problem and there may be a potential solution to it. I'm not engaging to just poke holes in your idea. But I think the problem can be solved either through notability, which seems to be where the action is in wiki policy as WAID says, or thinking about reforming AFD. On the latter point. Isn't the real problem that Wikipedians came to a "wrong or bad non-consensus"? Andre🚐 21:29, 8 October 2023 (UTC)[reply]
AfD reform is another thing that I thought of, but I wasn't sure if that would be effective. The problem I found is that a lot of editors in AfD discussions about events like crimes or disasters will either say "X people died so it's notable" or "it was in the news so it meets GNG", and they won't hear any challenge as to whether these are relevant to notability. Closing admins then accept these !votes at face value, even though they probably shouldn't. It's often the same editors, so it seems that AfD has developed its own culture independent from the actual expectations of the community. My hope was that an RfC would affirm these expectations. Thebiguglyalien (talk) 21:38, 8 October 2023 (UTC)[reply]
My suggestion as a first start would be to write a >=~3 paragraph essay articulating the above point. Why admins should discard bad rationales at AFD. Andre🚐 21:42, 8 October 2023 (UTC)[reply]
The problem of a general "rule" saying X and the specific consensus of editors saying not-X is not really one that we are likely to solve. According to WP:NOT and WP:PG (both policies), when the written rules diverge consistently from actual practice, it's the written rules that are wrong, not the actual practice. WhatamIdoing (talk) 22:42, 8 October 2023 (UTC)[reply]
The question is whether that consensus exists among the community or just among a small handful of AfD regulars. In theory, an RfC would answer that. Thebiguglyalien (talk) 23:17, 8 October 2023 (UTC)[reply]
It's axiomatic that any group disagreeing with "me" is only a tiny fraction of the community, and if that's hard to say with a straight face, then it's obvious that they were confused or did not fully understand the situation. Anyone who understands as much as "me" is always going to agree with "me", right?
So having said that, I think that you will find that it's difficult to claim that there are more people participating in discussions about any given policy or guideline than are participating at AFD. To put some numbers on that, Wikipedia:Articles for deletion/Log/2023 October 2 got 240 comments on 2 October 2023 (NB: not the whole week; just the one day). Wikipedia talk:Notability had 19 comments that day (it usually averages around five). Wikipedia talk:What Wikipedia is not had zero. WT:V, zero. WT:NPOV, zero. WT:CONSENSUS, three. This village pump, four.
When you look at those numbers, I think it is pretty clear where the community's attention is. The chattering classes might congregate on the policy/guideline/village pump pages, but we're a very small minority. WhatamIdoing (talk) 04:14, 9 October 2023 (UTC)[reply]
Well, you've !voted in just 10 AfDs since December 2020 and yet you participate in pretty much every single PAG discussion that comes up, so isn't that pretty strong evidence that there are editors who are heavily involved in community-wide PAG RfCs who nevertheless decline to help enforce the resulting consensuses? So, no, the editors who !vote regularly at AfD on any particular topic definitely comprise a very small sliver of the community broadly interested in that topic who would participate in PAG discussions if properly notified. We saw this with the 120+ !votes at the NSPORT RfC, despite NSPORT AfDs having maybe 20 regular !voters in a given month. Local PAG-contradicting consensuses in the form of AfD outcomes can accumulate rapidly if even a couple regulars from one "side" are absent for a bit. We saw this in a set of cricketer AfDs in the months after NSPORTS2022 was settled where a group of cricket project members continued !voting "per NCRIC" despite it having been resoundingly deprecated, and these closed as keep despite the topics objectively failing a criterion that had received global consensus as a requirement for retention.(See, e.g., [2]; @Dlthewave might have more background). Those editors then cited those closes as precedent at more AfDs. It was only when a handful more people from the larger community stepped in that these articles were deleted.[3] JoelleJay (talk) 01:32, 10 October 2023 (UTC)[reply]
That's right: I participate in relatively few AFDs, and when it comes to AFD, I advocate for paying more attention to the people who do the work than to people who talk about it. You could argue that I am advocating for less influence from me, and people like me, and you'd be right. My goal is to get the guidelines to accurately reflect what happens at AFD. My goal is not to change what happens at AFD; my goal is to give content creators a fair understanding of what reality looks like. One way to do this is to provide definitions of our jargon, so that they know, e.g., that SIGCOV is necessary, and that SIGCOV has a meaning that goes beyond whatever Humpty Dumpty wanted it to mean this time, all in the service of determining who is to be master.
If the community refuses to enforce the supposed consensus, then we don't actually have a consensus. It's against our core principles to elevate the words from those of us in the talking shops over the actions of people doing the work. WhatamIdoing (talk) 15:45, 10 October 2023 (UTC)[reply]
Well, that is a good point. But it still wouldn't necessarily preclude an essay encouraging admins to discard bad rationales at AFD. Which then if admins were to follow, would become the practice. Making it a rule, as you say, is fraught since we tend to ignore all of those when needed. Andre🚐 23:57, 8 October 2023 (UTC)[reply]
I believe that we already have an essay for that at Wikipedia:Arguments to avoid in deletion discussions. WhatamIdoing (talk) 03:55, 9 October 2023 (UTC)[reply]
I think WP:ROUGHCONSENSUS, which is incorporated as a section in the deletion guidelines for administrators, is technically more applicable as it actually says that those !votes are discounted. Alpha3031 (t • c) 01:51, 12 October 2023 (UTC)[reply]
Both are good answers; yet we're still faced with the quandary. Is it education? Various areas of the project have a lot of structure now that they didn't have 10 years ago. Andre🚐 01:54, 12 October 2023 (UTC)[reply]
Well, the admins are in a bind. If the tiny minority of editors in the talking shops say that generally speaking, ____ should be deleted, but the tiny minority of editors who actually evaluated the specific subject say that, in this particular case, this particular article should be kept, the admin has a choice between Wikipedia:Ignore all rules or a brisk trip to Wikipedia:Deletion review.
A couple of weeks ago, I asked another editor (sorry, I don't have the link handy, and I don't remember who it was) what they'd advise an admin in such a case, and if memory serves, I got no direct answer. Especially for some of our newer editors, whom we educated to believe in The One True™ Set of Rules, the idea that AFD might exist to make the right decision in individual cases rather than to implement The True™ Rules, seems to be a foreign concept. WhatamIdoing (talk) 02:38, 12 October 2023 (UTC)[reply]
I think a use of IAR should generally explain how the situation is different from the general case. Otherwise, either the policy should change or the discussion is defective. (or, for that matter, both)
I think, though , maybe those cases should be taken to DRV whichever way it's closed... or, perhaps there are ways we can make DRV less intimidating to admins. Of course, there might be a positive effect of promoting deeper consideration of the appropriate close, but I think it's less than ideal for the likelihood of DRV itself to influence which direction the discussion closes. Alpha3031 (t • c) 03:56, 12 October 2023 (UTC)[reply]
What should end up happening is these articles on sensational breaking news topics (if initially kept at AfD) are deleted after a year or so when no further coverage occurs. But then we have editors who insist any background (i.e. secondary) material in a news piece automatically constitutes sustained coverage, because it is coverage of a "sustained" period of the subject's existence... See this AfD and its related discussion at WT:N. So that would end up resulting in a keep even if 10 years from now there has not been a single additional piece of SIGCOV of the subject since the initial news burst. JoelleJay (talk) 23:30, 9 October 2023 (UTC)[reply]
So far based on this discussion, the wording for an RfC would be a simple one sentence question like Does news coverage satisfy WP:GNG for articles about events? or For articles about events, does news coverage about that event satisfy notability requirements?. Is this what would work? It still leaves unanswered what constitutes news coverage (such as whether coverage of a trial or other aftermath would still be news coverage of the same event). And what would be the appropriate place to hold it? WP:VPP, Wikipedia talk:Notability, and Wikipedia talk:Notability (events) all seem reasonable. Thebiguglyalien (talk) 03:37, 11 October 2023 (UTC)[reply]
There may be a difference between a news article published the day after an event and one published much later. I think the concern here is about recent events, so that difference won't arise, but I wouldn't want to see use of retrospective news coverage to help establish notability for historical events inadvertently being discouraged by the wording of this proposal. Perhaps adding "recent" would deal with that, although there is the potential for much squabbling over what "recent" means. Donald Albury 12:31, 11 October 2023 (UTC)[reply]
Another problem… not all events are equal. there are some events that are obviously notable the instant they occur - the election of a new Pope comes to mind. We know that there will be significant (sustained) coverage of the new Pope… even if at the time of writing all we have are breaking news reports announcing the election. I see no reason why this initial news coverage should be deemed non GNG compliant. Blueboar (talk) 12:52, 11 October 2023 (UTC)[reply]
Do we have articles about the elections of popes? I'm not sure if those would be notable, which sort of proves the point here. The election itself would need sustained coverage if it were to have its own article. Thebiguglyalien (talk) 15:10, 11 October 2023 (UTC)[reply]
Of course we do. See Category:Papal conclaves. These are major events with sustained coverage over centuries. —Kusma (talk) 15:26, 11 October 2023 (UTC)[reply]
All right, it seemed like something that would be covered in the given pope's article. In this case WP:EFFECT would apply. But even then, it would be difficult to write an article that wasn't just a bare sequence of events until it was analyzed. Thebiguglyalien (talk) 15:32, 11 October 2023 (UTC)[reply]
Are you referring to updates or retrospectives? Taking the 2020 Beirut explosion for example. You can have breaking news about an update, which is routine coverage, but you can also have a retrospective, which is a great source for an event article. Both of these are 2023 articles, but only one provides coverage that suggests this is an event that's being studied and is of encyclopedic interest. Thebiguglyalien (talk) 15:09, 11 October 2023 (UTC)[reply]
I did write "retrospective". Donald Albury 15:30, 11 October 2023 (UTC)[reply]
As it happens, I cited a retrospective published in a newspaper just yesterday. The source was a "what happened in the last decade" kind of piece. Many US newspapers also do a similar thing at the end of December, to review the year's news.
Nobody would doubt the notability of the biggest public corruption scandal in the US, and it might be interesting to imagine Wikipedia in spring of 1981, with the FBI arresting dozens of elected officials on charges of bribery. It doesn't take a crystal ball to imagine that an event of this magnitude will produce sustained coverage. What would you recommend to editors who create the article right away? WhatamIdoing (talk) 16:02, 11 October 2023 (UTC)[reply]
@Thebiguglyalien, I would only recommend that you ask an RFC question like "Does news coverage satisfy?" if your goal is to get a resounding "Yes, of course!" on record. WhatamIdoing (talk) 16:03, 11 October 2023 (UTC)[reply]
Yeah, that's something I'm worried about. If that's how most of the community feels, then so be it, but obviously I don't want to influence such an answer with the wording of the question. Thebiguglyalien (talk) 16:15, 11 October 2023 (UTC)[reply]
I wonder whether you might be able to get a more nuanced answer if you ask something more specific, like "If an event (e.g., a fire) happened more than a couple of years ago, but a good-faith search finds no sources beyond the year of the event, should those early sources be considered to constitute 'sustained coverage' of the event?"
In general, I'm not sure that this will produce the results that you seem to want, because the real policy is what experienced editors do, rather than exact words we put into a page with a fancy tag at the top (see the fifth of the Wikipedia:Five pillars), so they might still vote to keep them at AFD, and they wouldn't be wrong. But it's possible that this would give you a written ruling that would let you argue at AFD that The Community™, or at least the tiny fraction that responds to RFCs, says your way is better. Or, you know, you could just PROD or AFD a couple of those every day until someone complains. WhatamIdoing (talk) 17:15, 11 October 2023 (UTC)[reply]
I think maybe we should discuss it more, have more facilitated problem solving brainstorming. E.g. what is the problem, what are the possible opportunities, a strengths/weaknesses/actions/opportunities analysis for each, make a tree and start pruning the tree until you arrive at somewhat of a rigorous proposal. Currently, I agree with WAID that having an RFC for a result that would simply reaffirm the value of news coverage without any narrowing or clarification of the bright lines and tests we want to approximate, is not valuable for anyone and wouldn't have a meaningful change. I do think you've touched on a few troubling or interesting ends to tug on. 1) why, when closing discussions, don't admins or non-admin closers clarify the consensus by asking questions or by being more transparent about discarding invalid logic, according to a consensus and precedent based understanding of policy, leading to some outcomes that never get a DRV or a challenge, and then we are left holding a bag, this seems like a generalized weakness of how some closers close discussions (consensus is not a vote/not bean counting) 2) why do some commenters at AFD refuse to engage or change their mind and stick to an invalid view. without naming names, I have AFD'd or !voted to delete an article and found that some people left comments that I thought didn't make sense, then the article was kept for no consensus. is it education? is it the process? Andre🚐 17:24, 11 October 2023 (UTC)[reply]
I have AFD'd or !voted to delete an article and found that some people left comments that I thought didn't make sense, then the article was kept for no consensus. is it education? is it the process? Yeah...I have a whole bookmarks folder of such AfDs. And another folder for the ones that were only closed correctly after lengthy needless debate over something that should be obvious. Like where editors insisted school newspaper coverage of a student was independent despite historical precedent, the PAGs themselves, and a concurrent question at RS that came to a unanimous consensus before the AfD ended resoundingly confirming they are not independent of the student for the purposes of notability. Or the one where editors insisted the results of an anonymous, one-off google docs poll hosted by a football fan club's Twitter account was of sufficient significance to pass ANYBIO, and the 40 non-quote words of coverage of this "honor" was SIGCOV even after it was shown to be a word-for-word copy of a press release on the football club's facebook...that one was originally closed as "no consensus" even with all that info available! JoelleJay (talk) 20:53, 13 October 2023 (UTC)[reply]
  • I think it's going to be difficult to have a hard-and-fast rule for this. The main issue that immediately comes to mind is that some events obviously have long-term notability. If a US president is assassinated, you'd better believe we're making an article for that the instant the news breaks, and anyone complaining that we're WP:NOTNEWS or demanding we wait for WP:SUSTAINED coverage isn't going to be treated seriously. But according to policy, we're not really supposed to rely on our "gut feeling" about events like that (see WP:CRYSTALBALL), so... what do we use to determine that a breaking event is so significant that it definitely requires an article? I would say that the tone and level of coverage both matter as well; if an event gets only cursory coverage in the local news, whose tone is "here's a thing that happened today", it may not need an article until we have either non-news coverage or WP:SUSTAINED coverage; but if it gets wall-to-wall coverage in national / international news and the tone of that coverage is "oh my god, this changes everything forever / a day that will live in infamy" etc etc etc, then it probably can support its own article. Another thing that occurs to me that might support a hypothetical refined GNG for news is what you might call "secondary news" coverage, news coverage that isn't about the event itself but its long-term effects - if you look at the recent attacks in Israel, say, or think about a hypothetical presidential assassination, there will be massive amounts of articles that aren't just covering the event directly but which devote an entire article to eg. "here is what this means in a specific context; here is how it might affect the price of X, Y, and Z; here's the impact it will have on the stock market" etc etc etc. Basically I don't think that we can use WP:SUSTAINED as the sole criteria, we need criteria that helps us determine what news stories really do require their own articles immediately. (And while the examples I gave might seem obvious, I suspect that the main borderline articles are going to be political grist ones - lots of stories get a ton of coverage as "the most important thing EVER, politician XYZ is TOTALLY DESTROYED" in the partisan press but little sustained coverage elsewhere. OTOH sometimes they represent genuinely important scandals that do deserve an immediate article. So what to do with those articles and what sort of sources would demonstrate we can write a neutral article about them is probably the main thing to keep in mind when writing guidelines.) --Aquillion (talk) 17:51, 11 October 2023 (UTC)[reply]
    I'd like to point out that CRYSTAL only applies to article content. It does not apply to discussions on how much sourcing we might expect to find in the future, or in XYZ location that we haven't looked at yet. In fact, most of NOT applies primarily or exclusively to content in or intended for mainspace. Same with OR or V. Editors may or may not find such arguments less persuasive, but they are not proscribed by policy. Alpha3031 (t • c) 16:32, 13 October 2023 (UTC)[reply]
  • General comment. I used to work a lot at ITN, and my impression was that lots of editors, often not well-known ones here on en-wiki, were extremely interested and active in starting and developing articles on "disasters" especially where people died. There are some topics (disasters, sports events, television/films...) where there is abundant online sourcing that is easy to understand and free to access, and so inexperienced or non-native English speaker editors feel comfortable working to build articles in those areas. There's also considerable reader interest in such topics, far more so than any of the kinds of more obviously encyclopedia topics I tend to edit. For example my article from last year George Checkley gets only a handful of daily views on a good day, whereas 2022 Taitung earthquakes, from a similar timeframe, gets around 14 daily views; also September 2022 Afghanistan earthquake got a recent spike of hits, after the recent earthquakes. So there is low-level sustained reader interest beyond the year mark. Not sure where I'm going with this exactly, but I think (1) starting to prune this content (which I don't disagree with personally) is likely to be very difficult and contentious; and (2) any guidelines developed will need to be very clear, because many of the editors working in these areas are not experienced, don't have fluent English, care passionately about the topic, and often don't work much in other areas of en-wiki. Espresso Addict (talk) 01:35, 12 October 2023 (UTC)[reply]
    Current events are definitely popular with readers.
    Perhaps one of the points should be to merge such articles into a "List of earthquakes in..." rather than a full deletion. Even if you don't think that a given event is worth having a separate article on it, it might still be worth mentioning it somewhere. WhatamIdoing (talk) 02:32, 12 October 2023 (UTC)[reply]
    Maybe we could just create a space somewhere on this project but somewhat separated from mainspace and just get rid of NOTNEWS and IINFO in that space. Somewhere adhering to WP:COPO but without necessarily being as attached to what an encyclopaedia is or is not, or what precisely counts as encyclopedic content. Alpha3031 (t • c) 16:42, 13 October 2023 (UTC)[reply]
    The first of the Wikipedia:Five pillars indicates that Wikipedia is not exclusively an encyclopedia, as it also includes the sort of information one would expect to find in almanacs and gazetteers. WhatamIdoing (talk) 17:16, 13 October 2023 (UTC)[reply]
    Yes, but while IINFO and NEWS are both excluded (in 5P1 as well as NOT) there's no reason why we can't create a space for it. Alpha3031 (t • c) 08:10, 14 October 2023 (UTC)[reply]
    WP:NEWS says nothing relevant; I think you've got the wrong WP:UPPERCASE.
    WP:NOTNEWS says that Wikipedia articles aren't supposed to be news articles, which is different from being about whatever is in the news. That policy says Editors are encouraged to include current and up-to-date information within its coverage, and to develop stand-alone articles on significant current events. Whatever subjects are in the news are not excluded from Wikipedia; indeed, there are officially encouraged according to long-standing policy. WhatamIdoing (talk) 22:58, 14 October 2023 (UTC)[reply]
    Yes, I was referencing them as sections within NOT, which also says that most newsworthy events do not qualify for inclusion. Alpha3031 (t • c) 05:55, 15 October 2023 (UTC)[reply]
    TL;DR: Most is measured against what's in the newspaper, not what editors start articles about.
    I agree that most newsworthy events do not qualify for inclusion, and most newsworthy events are not even attempted to be included in Wikipedia. Here's a list of the "bigger" articles in a daily newspaper for one day last week: Ten articles about sports. Three articles about crimes. One article each about a Nobel prize, city meetings, statewide demographic changes, travel, a celebrity–charity kerfuffle, a local business, a routine report about an airport, a local hospital, a car wreck, update on a previous wildfire, pollution worries, an invasive species, and a photo spread about seasonal events. Two articles each about new state laws, music, and the Israel–Hamas mess. There were also half a dozen nationally syndicated columns and another half-dozen pieces about homes that sold recently.
    The number of these articles that I expect to be cited on Wikipedia is: zero. However, a couple of them (e.g., 2023 Israel–Hamas war) are about notable subjects, and a couple might be usable to expand or update existing content (e.g., Demographics of California).
    For bigger newspapers, here are some 2016 numbers on content:
    That's a whole bunch of newsworthy events just on the one day. Most of them do not qualify for inclusion. But even if just 1% of the newsworthy events qualify for inclusion, that could be thousands of current events per year. We don't want articles about most newsworthy events, but we probably do want articles about most of the newsworthy events that editors actually start articles about. WhatamIdoing (talk) 21:23, 15 October 2023 (UTC)[reply]
    Strong agree on this analysis. Andre🚐 21:30, 15 October 2023 (UTC)[reply]
    I like WAID's idea to officially advise a merge. Current events articles already exist. It might be a way to deal with portals, too? Andre🚐 17:21, 13 October 2023 (UTC)[reply]
    As far as I'm concerned, merging all of the non-notable events into "list of" articles is the optimal solution. The trick is making it happen, because right now we basically have to RM or AfD each article one at a time and convince the "news coverage = new article" editors that merging is a better way to organize these things. I could easily come up with a list of hundreds of event articles that should almost certainly be merged, and that's to say nothing of the borderline cases. But just getting one of them addressed is a struggle. Thebiguglyalien (talk) 05:10, 14 October 2023 (UTC)[reply]
    Maybe you could propose a speedy mergification similar to how some articles are draftified or prodded. Andre🚐 05:38, 14 October 2023 (UTC)[reply]
    The discussion Wikipedia:Village pump (idea lab)/Archive 49#Addressing non-notable event articles: Traffic incidents might be of interest. I raised the issue of addressing lots of articles at the same time, and it was completely unproductive. I expect any other idea that involves dealing with large amounts of articles would be met with similar nitpicking. Thebiguglyalien (talk) 13:59, 14 October 2023 (UTC)[reply]
    @Thebiguglyalien, how many merges have you proposed, using the directions at Wikipedia:Merging#Proposing a merge? What were the outcomes? WhatamIdoing (talk) 23:03, 14 October 2023 (UTC)[reply]
    None. I don't want to start merging if it's not clear what type of merging we're talking about here. For example I can say that 2021 Cairo clothing factory fire probably doesn't need an article. But the most likely target, List of building or structure fires, already mentions it. It's not clear what these merges would look like. Thebiguglyalien (talk) 23:40, 14 October 2023 (UTC)[reply]
    Looking at the English-language sources about that one fire, it appears that a List of fatal fires in Egypt would be a valid subject. Several of the longer sources included a list of fatal fires in Egypt during the previous months. The same (large) city had another fatal factory fire in 2015. WhatamIdoing (talk) 22:02, 15 October 2023 (UTC)[reply]
    An article on major fires in Egypt, and what factors contribute to them happening, would be much more interesting than just a list. Espresso Addict (talk) 22:43, 15 October 2023 (UTC)[reply]
    I think a list along the lines of Floods in California would work. A plain bullet point list says only something like like:
    • Fire at Business in City on 32 Octember 2021
    would be boring. (The overall cause appears to be a systematic failure to enforce safety regulations, and once the fire has started, the fire departments do their best but don't have the amazing resources that some others do. It is not a wealthy country, after all.) WhatamIdoing (talk) 04:44, 16 October 2023 (UTC)[reply]
    Lists like this would be a perfect solution. The easiest way to cover everything would be to split the current List of building or structure fires by decade. Then bullet points can be expanded into a paragraph or two by merging the relevant details from the main article where applicable. Thebiguglyalien (talk) 01:49, 17 October 2023 (UTC)[reply]
    Late to the party , don't know how I missed the ping, but I agree that the example above, Floods in California is a good example of what should be recommended. Earlier this year I was involved in an AFD where a road accident had been nominated, but there was coverage years later as it was one of the worst UK road accidents so it was kept. But as I pointed out then, on the same day on the same road a former rugby player had died in a separate accident which didn't have an article. I did put forward that we could create page about the serious road accidents on that occured on that actual road. Davidstewartharvey (talk) 14:20, 17 October 2023 (UTC)[reply]
    That one is quite nice. I think lists are a neat solution to this problem. DFlhb (talk) 15:31, 17 October 2023 (UTC)[reply]
    I've up-merged several stub articles into List of tanker explosions as a proof of concept, and it looks much cleaner and more useful than a bunch of disparate stubs. There are still a few blank sections where I want a second opinion about whether merging or WP:SUMMARY is appropriate. Thebiguglyalien (talk) 17:38, 17 October 2023 (UTC)[reply]
  • This entire discussion is crazy to me. I have written hundreds of articles on 19th-century state supreme court justices, relying almost entirely on news coverage of their careers and their deaths. Many of these are sourced to local newspapers of the time period. BD2412 T 21:37, 15 October 2023 (UTC)[reply]
    Yes, absolutely. Historical news coverage is critical to write articles properly. Andre🚐 22:31, 15 October 2023 (UTC)[reply]
    Are you writing "inauguration of [justice]" or "tenure of [justice]" articles? That's the only way I could see this being relevant to that. Thebiguglyalien (talk) 22:36, 15 October 2023 (UTC)[reply]
    I believe I raised this topic earlier and it was clarified that the current proposal only relates to events, not people or places. Espresso Addict (talk) 22:40, 15 October 2023 (UTC)[reply]
    @Thebiguglyalien: Obituaries often provide fairly comprehensive biographical coverage, although occasionally newspapers will provide biographical summaries for all current members of their state's high court. This was more common in the 1800s, of course, when there tended not to be other sources for this information. @Espresso Addict: I am concerned about creep. This sounds like the beginning of saying that news coverage is not appropriate as an encyclopedic source. BD2412 T 22:56, 15 October 2023 (UTC)[reply]
    I too am worried about creep, both intentional and unintentional (eg discussion about a minor change to minor changes spiralling into a call to nuke 'em from orbit). Espresso Addict (talk) 23:03, 15 October 2023 (UTC)[reply]
    BD2412, my problem is with articles about events that have no lasting effect and no retrospective coverage. When a deadly car crash happens and makes the news, can I say it meets GNG? To me the answer is obviously no for a number of reasons, but there are enough people at AfD who say yes that it needs to be settled one way or the other. Thebiguglyalien (talk) 23:18, 15 October 2023 (UTC)[reply]
    I hope this won't be settled "one way or the other", but by intelligent merging of less notable event articles, something that doesn't require AfD or noticeboard discussions. —Kusma (talk) 15:00, 17 October 2023 (UTC)[reply]
    I agree. As tempting as it is to create the One True™ Rule for this, it's better just to do the work. There is enough information in the sources in 2021 Cairo clothing factory fire to create (e.g.,) a List of fatal fires in Egypt. Creating an official rule will have basically no effect, but merging up related articles solves the actual problem directly. WhatamIdoing (talk) 15:21, 17 October 2023 (UTC)[reply]
    I'm optimistic that this can happen based on the thread of comments above this one. It's just a matter of figuring out what any potential target lists would look like and how they're organized. Thebiguglyalien (talk) 15:23, 17 October 2023 (UTC)[reply]
    As others have pointed out, the existing policy on merging already is fine with what you're doing, so go ahead. Andre🚐 21:46, 17 October 2023 (UTC)[reply]
List of tanker explosions looks how I'd expect most lists of events would look after merging up. Shorter instances are collected on the same page, while more in-depth coverage is provided for a few with Template:Main. If there's general agreement that this works, then maybe it can be applied to some of the other events-related topics that have a lot of stubs. For larger topics, they can be divided into multiple lists divided by location ("List of X in country" or "List of X in continent") or by time ("List of X in year" or "List of X in decade"). Thebiguglyalien (talk) 15:01, 18 October 2023 (UTC)[reply]
So how do you decide which articles to retain? The answer seems in this case to be practically none. I remember the Bahawalpur, Pakistan (2017) incident distinctly, which received coverage in depth in the UK and had a well-developed article classed as C-class and mid-importance for some projects. I would not have thought that kind of article would be reduced to a redirect by such proposals without any individual discussion. Espresso Addict (talk) 05:44, 19 October 2023 (UTC)[reply]
There are a few relevant guidelines regarding merges of inadequate articles. The most immediate ones are WP:PAGEDECIDE and WP:AVOIDSPLIT. The former recommends keeping it within a larger topic if there isn't adequate sourcing, and the latter encourages such merges if sourcing can't be found to demonstrate notability. Regarding notability, relevant guidelines obviously include WP:N, which requires significant secondary coverage sustained over time, as well as WP:NEVENTS, which has similar requirements. Now if someone was willing to do the work to replace most of the primary sources with secondary ones and write up a few paragraphs about analysis that's taken place in the past few years, then I would encourage them to revert the merge and do so. If it's not possible to do that, then it was inappropriate to ever give it its own article in the first place. Thebiguglyalien (talk) 14:22, 19 October 2023 (UTC)[reply]
For the without any individual discussion worry, the Wikipedia:Merging process recommends discussions of at least a week, so I don't think we should assume that there would be no opportunity for discussion. WhatamIdoing (talk) 16:29, 19 October 2023 (UTC)[reply]
WhatamIdoing/Espresso Addict/anyone else interested: What do you suggest as far as getting consensus for merges if a bold merge is challenged? For List of tanker explosions, I merged the articles that were short enough to be stub or start class articles. Ideally any article that's all primary sources should be rewritten, merged, or deleted, but that's not going to stop some editors from bringing together some primary sources and asserting notability. Case in point, I've laid the groundwork for List of mass stabbing incidents (before 2010) which is still a work in progress, but one of the redirects was reverted, restoring an "article" that's just regurgitating primary sources and reads like a news story. Thebiguglyalien (talk) 23:05, 19 October 2023 (UTC)[reply]
WP:PROPMERGE. Curbon7 (talk) 00:10, 20 October 2023 (UTC)[reply]
I think another issue that needs to be borne in mind is the appearance of racism, when all non-recent articles on a topic are merged, except one lone American one. Espresso Addict (talk) 05:02, 20 October 2023 (UTC)[reply]
I would appreciate if you discussed the merge with me beforehand and not just done it immediately without any consensus. Lettlerhello • contribs 13:42, 20 October 2023 (UTC)[reply]
Also, I do agree that most of the articles on the List of mass stabbings are not really articles and do not follow the lasting coverage guidelines at all. E.g. Nanital wedding massacre, it might not even be a real story if I'm being honest. However, you did merge some articles that definitely cover notable topics and have lasting coverage through retrospectives, etc. like the Rackham's stabbings. Again, I would appreciate if you and other editors had a discussion prior to this instead of you going on a one-person crusade based on your own opinions. Lettlerhello • contribs 13:47, 20 October 2023 (UTC)[reply]
  • I will point out there is a closely-related discussion at WT:NOT#WP:NOTNEWS/unfolding news stories related to how to write/clean up event articles well after the event is over. That is tightly related to establishing that most news cover of an event in the short term is primary, and what we're looking for is more secondary (analysis or reflection, or describing what impacts an event had). --Masem (t) 17:30, 20 October 2023 (UTC)[reply]
I don't know if the answer is a policy change, but I like the idea of a clean-up. It does seem people rush to crate articles that don't really hold-up over time. And I certainly agree with some of the commentary here about some of the fluff stuff added to articles to try and establish their notability. To me, the problem is the phrase "encyclopedic" and how its used here. It's really easy for people to justify any information they put in an article by saying that it is such, and really is quite difficult to refute. I think trying to define what content is relevant and not in articles is a good start in filing down the number of articles in general. DarkSide830 (talk) 16:58, 21 October 2023 (UTC)[reply]
I agree that merging some subjects that don't hold up over time into a single, bigger article has a lot of potential for improving their encyclopedic value.
I'm not sure that we need to reduce the number of articles in general. Some estimates suggest that we're still short by at least several million (e.g., politicians from previous centuries that were mostly covered in non-English sources – look at the red links in List of mayors of Berlin, all of which are merely waiting for an interested German-speaking editor to create the articles). Consequently, I don't see this as a way of having "less"; I see this approach as a way of having "better" (and possibly even "more"). WhatamIdoing (talk) 17:36, 21 October 2023 (UTC)[reply]
I agree. We used to say WP:NOT paper. Andre🚐 17:50, 21 October 2023 (UTC)[reply]
I'm sure we're missing relevant older articles, but that's not what I'm talking about. I'm talking about the breaking newsy stuff that get articles before we can establish their long-term significance, nor am I talking about people unless we're talking about 1E bios. DarkSide830 (talk) 19:23, 24 October 2023 (UTC)[reply]
  • Here from a ping by WhatamIdoing, thank you. We tried to lance this boil with WP:NEVENT back in 2009, to reconcile WP:GNG, WP:NOTNEWS, and WP:NTEMP. It seems we failed. Please read the discussion starting at Wikipedia talk:Notability (events)/Archive 1 to see how the current guidance was arrived at and whether there were any useful ideas left on the cutting room floor. Fences&Windows 10:02, 22 October 2023 (UTC)[reply]
    The thing is, NEVENT does cover nearly all cases here. The problem is that some editors just don't care, so they keep creating weak articles and then !voting keep as a group at AfD by saying that appearing in a newspaper is enough to meet GNG. Instead of proposing a change to NEVENT (which would almost certainly be shot down), they just blatantly ignore sitewide consensus, and no one does anything about it. The small handful of admins who close AfD discussions seem to be unwilling or unable to weigh arguments against notability guidelines like they're supposed to. So now we're stuck with thousands of articles that read like news and provide no meaningful analysis. Thebiguglyalien (talk) 15:10, 22 October 2023 (UTC)[reply]
    Wikipedia:Nobody reads the directions. The written rules exist for our convenience, and as something we can wave at people when we're fixing things up, and not because we expect others to seek them out and carefully follow them. The expected path is:
    If you are expecting this path:
    • A person interested enough to contribute information to Wikipedia is able to find all the directions.
    • That person puts their excitement on hold long enough to carefully read and understand all the directions.
    • The excited person then dispassionately and correctly evaluates the subject that they're excited about, and decides that Wikipedia isn't interested in hearing about a potentially big-deal event,
    then you have not been paying attention. WhatamIdoing (talk) 19:20, 22 October 2023 (UTC)[reply]
    My frustration stems from the fact that waving this tends not to do anything. I can list too many AfDs where I pointed to these things, was ignored or flat out told that our guidelines were wrong, and then it was closed as keep. This, this, and this are some examples. On the latter, I had to withdraw after an ANI discussion was opened. This is what we're dealing with at AfD. I'm tempted to start taking it to deletion review any time a keep is decided based on news coverage. Thebiguglyalien (talk) 19:48, 22 October 2023 (UTC)[reply]
    Deletion review is definitely one option that we have under the present system. Andre🚐 19:56, 22 October 2023 (UTC)[reply]
    The wider problem is that there is a disconnect between the editors who create much of the content, and the editors who frequent discussions that generate policy/guidelines. Content editors are more likely to comment in AfD than other backroom areas. Espresso Addict (talk) 21:21, 22 October 2023 (UTC)[reply]
    That's a very good point. Maybe the problem we should focus on is how to open up policy backroom to more content editors who don't have experience in such discussion. Andre🚐 22:55, 22 October 2023 (UTC)[reply]
    That's the whole point about setting up an RFC at a Village Pump and advertising it at CENT and other places. These are the equivalent to the town square, and there is no better way to mass advertise key PAG discussions. Editors that do not engage in these areas and then complain about PAG override what they think is consensus are not working within the proper context of what we expect. Masem (t) 22:59, 22 October 2023 (UTC)[reply]
    I think we should put on our product thinking hat. If people don't do things you have to ask why they chose another path, and try to guide them better. With better tools, or better UI, or better design, or better ideas, or a new workflow, etc. It's problem solving. This is the idea lab. Andre🚐 23:23, 22 October 2023 (UTC)[reply]
    Someday, we should probably pull the numbers for VP participation, so that we can get over this idea that the talking shops are where editors hang out. For this Village pump (alone), so far, less than 3,300 editors have ever posted here. About 13.9 million registered editors have made at least one edit to the English Wikipedia. That's about one in 4,000 (four thousand) editors, or one fiftieth of one percent of registered editors. The other 99.98% of editors have never posted here.
    If you think that it's "unfair" to include editors who haven't made many edits, then let's look at the numbers a different way:
    • 99% of editors who have made 100+ edits have never posted to this village pump.
    • 95% of editors who have made 1,000+ edits have never posted to this village pump.
    • 75% of editors who have made 10,000+ edits have never posted to this village pump.
    My conclusion: This village pump is not the way to reach the whole community. WhatamIdoing (talk) 02:07, 23 October 2023 (UTC)[reply]
    @Andrevan, I think you will be interested in the work that the Editing team did on uploading images to Commons from inside the visual editor a few years ago. The goal is to get people to upload relevant images (e.g., they're writing about a type of food and upload a photo of their dinner) while they're editing.
    • What's wanted: Here's my own photo, taken by me personally, that I'm donating to Commons.
    • What happened: I copied this logo on the internet, but when I truthfully said that it wasn't my own work, it wouldn't let me upload it. So I lied, because the only way to make the stupid thing work was to say that it was my own work, promise to give my first-born son to Rumpelstiltskin, and click through all these pointless warnings about copyright law. What a lot of nonsense. Everyone knows that it's okay to have a corporate logo in a Wikipedia article!
    Education in the UI only goes so far. When people believe that they are doing the right thing, they will do or say whatever is necessary to convince the computer to let them do it. WhatamIdoing (talk) 02:12, 23 October 2023 (UTC)[reply]
    I agree, but that's why you need to do before-the-fact user testing and optimization. Andre🚐 02:27, 23 October 2023 (UTC)[reply]
    I'm telling you the results of the user testing and optimization. There was no amount of education that actually stopped people from doing what they believed was the right thing to do. Eventually Commons put up an AbuseFilter to stop nearly all of the in-editor uploads (except from highly experienced Commons editors, who rarely use the feature anyway). WhatamIdoing (talk) 02:50, 23 October 2023 (UTC)[reply]
    Fair enough. I believe you. And that's an interesting result. And it's definitely one kind of result. But I wouldn't extrapolate that to mean no UI change could make people participate on the village pump more. I bet, if we did an AB test of a giant red callout to the village pump versus not, it would increase traffic. Andre🚐 02:54, 23 October 2023 (UTC)[reply]
    It is unfortunately not a unique result. It's a truism in computer security that if you give people a choice between following the policies (e.g., "No uploading confidential documents to external websites like Dropbox") and getting their job done (e.g., "He needs this doc, and it's too big to send in e-mail!"), they will break the policy every time. The average person, if promised a video of dancing bears, will click on anything necessary to watch that video. A person might not click the buttons labeled "Download malware", but they will click buttons that say "Enter your password to download the dancing bear video. Note: You may have to override your antiviral software to see this".
    It is highly likely that we could to drive some traffic (=page views) here. It is possible that we could drive some participation here (=edits). It is much harder to retain people, and much harder to convince them that their preferred contribution (e.g., updating the stats from the big game) is more important or more fun than chatting with people about abstract ideas. (Also, since Wikipedia:Policy writing is hard, they might not be any good at it.) WhatamIdoing (talk) 18:35, 23 October 2023 (UTC)[reply]
    Policy writing is hard, but people often don't even try to get good at it. And we don't offer them many venues to break into it or to polish their skills. Part of this is how bad threaded discussion is on Wikipedia. The reply tool is very helpful, but it's still a long way from solving the problem. I know they've been working on that for a while. Anyway, one thing I think is part of the issue is that people like to go straight to the hairiest, meatiest topic, like NPOV for controversial current events. There are not a lot of "training ground" low stakes policy/meta discussions. Andre🚐 18:44, 23 October 2023 (UTC)[reply]
    I have a more detailed set of numbers about who posts at the Village pumps. See Wikipedia talk:Village pump#Who posts at the Village pumps? if you're interested in statistics. WhatamIdoing (talk) 18:38, 23 October 2023 (UTC)[reply]
    If people want their voices to be heard on Wikipedia, then the onus is on them to speak up. We can and should make the venue as available as possible, but they ultimately have to come here on their own. If they fail to do that, then they're endorsing whatever conclusion we come to here by default, and that's their decision. Thebiguglyalien (talk) 02:17, 23 October 2023 (UTC)[reply]
    Your opinion does not seem to be consistent with either the Wikipedia:Ignore all rules policy or the Wikipedia:What Wikipedia is not#Wikipedia is not compulsory policy. WhatamIdoing (talk) 02:20, 23 October 2023 (UTC)[reply]
    WP is not compulsory, but also we expect editors to be editing from the start within PAG (that's why the Welcome message standard for new editors goes to the top level set of these pages). We absolutely do expect editors to understand what PAG are and how they are developed, and that if they don't like how some things are handled on WP, they should be speaking up at the appropriate forums. Outside of rare cases, we take silence on such matters as implicit agreement with PAG. Which does lead to the difference between the small number of editors that spend a lot of time at PAG pages (and typically the same group with the largest contribution numbers on WP), and those editors that come by once a week or month to add something and get annoyed when they get reverted, but make no effort to engage beyond that level. Masem (t) 03:11, 23 October 2023 (UTC)[reply]
    I don't think any of that's true.
    Is anyone here actually surprised if a newbie screws up their first edits? I'm not. You're not. We don't actually expect them to know what the rules are, or to follow them even if they do know the rules. We actually expect them to get it wrong, and we're pleasantly surprised when a newbie's edit summary says "typo" and the diff shows them actually fixing a typo. We enforce the rules – Ignorantia juris non excusat – but we don't expect them to know or follow our rules.
    Even most experienced editors, who generally know some distorted telephone game version of some of the rules, have no idea how policies and guidelines are developed. I'd bet that if you asked the next three RFA candidates to tell you how they're developed, you'd get three different answers – and that I wouldn't agree with any of them, as the candidates would be too focused on writing a diplomatic reply to tell the truth, which is that if you want a change (e.g.,) to the FAC rules, it matters more to get a short list of specific individuals on board than to get any other random editors to vote your way.
    Wikipedia:Nobody reads the directions. This is true on Wikipedia, and it is true everywhere on the internet. And in the real world, too. Think about how few people actually read the directions before trying to use the device (or build the flat pack furniture) they just bought. WhatamIdoing (talk) 18:22, 23 October 2023 (UTC)[reply]
    I agree with this. Wikipolitics is the thing. Andre🚐 18:27, 23 October 2023 (UTC)[reply]
    New editors, we do not expect them to know off the bat, which is why we welcome them with PAG links.
    And I agree that no experienced editor knows the PAG word for word, and likely have their own internal concept of them that differs from the actual language. However, the point is that we do expect editors to be aware PAG exists, that they are derived from consensus, and that their are ways to request changes to that. An editor that consistently edits against a PAG even after being told of the correct PAG, and that complains bitterly about the PAG, but never participates in discussions about changing PAG is likely going to be seen as disruptive.
    So knowing in general the PAG framework and editing within it aren't compulsory. You aren't required to participate in discussions on changing PAG, but you have little recourse if you constantly complain the PAG are bad and hampers your editing. Which goes back to making sure that if we want to change any OAG related to news coverage, it needs to be advertised as widely as possible, as I suspect many editors that do work on current event articles have never really participated in PAG discussions. Masem (t) 23:29, 23 October 2023 (UTC)[reply]
    What's that bumper sticker? "If you didn't vote, you don't have the right to complain", or something like that? And yet people complain anyway.
    I agree with you that many editors that do work on current event articles have never really participated in discussions about policies and guidelines, or even AFDs. We would all benefit from having more of our discussions include people who are writing the articles, instead of just those of us who like to hang out in the talking shops. WhatamIdoing (talk) 01:07, 24 October 2023 (UTC)[reply]
    @Thebiguglyalien, "delete the whole thing" is not the same as "salvage what you can". The Wikipedia:Editing policy says Wikipedia summarizes accepted knowledge. As a rule, the more accepted knowledge it can encapsulate, the better it is. Outright deleting information is not a path towards having "more accepted knowledge". It is a path towards making Wikipedia useless to readers, because the thing they want to read about isn't here.
    Looking at the first one, let's leave aside the absurdity of an editor looking into his WP:CRYSTALBALL just 11 hours after the article was created, and barely a day after the event, and declaring that he already knows that this event has no lasting significance. There might be a policy basis for not having a separate article for any given mass shooting, but there's no policy basis for not having this information somewhere in the English Wikipedia. The goal at AFD is to make sure that information is excluded completely – that it does not survive as even a brief little paragraph in a larger subject (e.g., the city, the police department), or as a redirect to the item in the List of killings by law enforcement officers in the United States, May 2023. If this wasn't the goal, then the editor wouldn't have taken the subject to AFD. I ask: Why would an editor who agrees with our policy ("the more accepted knowledge, the better") want to do have less knowledge? If more = better, then less = worse. WhatamIdoing (talk) 01:54, 23 October 2023 (UTC)[reply]
    If your interpretation of the editing policy is correct, then the logical conclusion as I see it is covering everything ever mentioned in any source. I believe that there should be a limit on what knowledge we cover, and that limit should be what's analyzed in secondary sources. I disagree with the premise that events should be assumed notable simply because they happened recently. If lasting coverage doesn't exist yet, then it doesn't exist. We don't get to say that it will some day exist. That's what would constitute a crystal ball. I'm happy with the metric set by WP:EVENTCRIT for when a recent event should be assumed notable. Thebiguglyalien (talk) 02:14, 23 October 2023 (UTC)[reply]
    There are certainly editors who support a "everything in any [reliable] source" standard, but most of us want something a little stronger – most usually a filter for what's encyclopedic (e.g., yes to mass shootings, no to what color suit Queen Elizabeth wore to an event). WhatamIdoing (talk) 02:18, 23 October 2023 (UTC)[reply]
I've set up a slightly larger test case with mass stabbing incidents:
I merged many of the articles where there was clearly no case for a WP:SIZESPLIT. More could reasonably be merged, if anyone wishes to do so, and many of the remaining articles are likely non-notable as well. The problem of bad sourcing also still exists and needs to be addressed among this category of events. Thebiguglyalien (talk) 15:04, 22 October 2023 (UTC)[reply]

The RFC in the OP would create a mess in many ways. Rather than getting into that I'll just note that the discussion has evolved since then. I think that the gist of the OP is that in principle the SNG/GNG look OK but that they really arent getting followed. When it's suits them, people generally ignore the vaguer generally ignore general guidance in policies and guidlines and only feel obliged to follow the more explicit stuff. For some articles (for example, those on the "In the news" on the main page), the events are or very high prominence /impact/importance, have lots of sourcing and lots of content I think that people generally say "wp:notable enough". IMO giving (only) some consideration to degree of prominence /impact/importance is how the fuzzy notability ecosystem actuall works, even though it's not explicitly in any guideline. I think that the sooner that we acknowlesdge that the sooner we can solve these otherwise unsolvable quandaries. North8000 (talk) 20:18, 24 October 2023 (UTC)[reply]

One of our oldest policies says that when it suits the situation, you're supposed to ignore all rules, vague and explicit alike. It's really hard to say that it's bad when people don't follow the rules when one of the rules is that you shouldn't follow rules that (initially just in your own opinion; later, iff your opinion is challenged, in the group's view) make Wikipedia worse.
One way to see this issue is:
  • The written rules neither encourage nor forbid a certain action (e.g., creating separate articles for subjects whose lasting significance is possible but unclear).
  • An editor believes that taking this action will improve Wikipedia, and so takes this action.
  • A second editor worries that doing what improves Wikipedia, instead of following The Rules™ is bad, because everything which is not explicitly allowed is probably forbidden.
I wonder whether we are best served by promoting rule-following as a primary value. WhatamIdoing (talk) 01:55, 25 October 2023 (UTC)[reply]
Conversely, the action, under eventualism and mergism, did improve the encyclopedia. We just need to reduce friction to that state. Andre🚐 02:54, 25 October 2023 (UTC)[reply]
@WhatamIdoing: Many good points there. My belief is that, even though seldom invoked explicitly, WP:IAR is immensely used as an influence on other decisions. For example, giving some consideration to prominence/importance/impact to let it tip a wp:notability decision towards inclusion. North8000 (talk) 12:27, 25 October 2023 (UTC)[reply]
Or to tip it towards exclusion (e.g., towards merging). I'd still rather that we consider putting WP:N's note about editorial discretion into the GNG itself, since that would deal with the "prominence/importance/impact" issue directly, and maybe then we'd quit seeing people claim that Wikipedia:Significant coverage means "[media] coverage claiming WP:ITSIMPORTANT". I think a lot of discussions would be smoother if we all agreed that SIGCOV means "an amount of attention that is significant" (the dominant view, as you know, but when an editor understanding it as "an amount" and an editor understanding it as "evidence that it's an important subject" are in a dispute, they can't understand each other). WhatamIdoing (talk) 15:17, 25 October 2023 (UTC)[reply]
Alternatively, we could add some tests for "importance"/"significance" to the guideline. I've always considered GNG to be a proxy metric. Which is why there's so much wiggle room around quantifying GNG. "Notability" in wiki-jargon means significant coverage which means the amount of coverage. But notability the term is a proxy metric for the abstract quantity of being notable, which we judge based on significant coverage. Going back to the idea that the rules are principles, the point of IAR is that you're not supposed to rules-lawyer the letter of the law. You are supposed to internalize the deep values and principles, from which the practices emerge. That is why people often conflate notability, the measured quantity, with notability, the common sense common language idea that the wiki-jargon metric is attempting to capture a proxy for. Andre🚐 15:20, 25 October 2023 (UTC)[reply]

{{Article for deletion}} refinement

When an article is nominated for deletion it can create controversy outside Wikipedia, here's the latest for example: https://twitter.com/stillgray/status/1707181074001121633. This problem I think can be because of how the template is worded and designed. The use of red color can be misleading, as if it's "warning this/your article is going to be deleted" and as if deletion is a bad thing (whatever the outcome, deleted or not, is actually a good thing as long as it's according to the policy); probably the color should be changed, or no color is needed at all. Only one user is needed to nominate an article for deletion, but the template doesn't really indicate that. Also the wording could be not neutral, as if the article is leaning to deletion. The bolded text could be changed to something like: "An user has nominated this article for deletion. An uninvolved administrator will decide whether this article, after discussion, is kept or deleted according to Wikipedia's deletion policy." Hddty (talk) 08:30, 5 October 2023 (UTC)[reply]

I think it's fairly normal to use red to represent something been cancelled or deleted. A deletion nomination is more serious and urgent than most tags that appear at the top of articles, and I think it's a good idea to draw attention to that. Few authors of articles would agree that having their work removed is a desirable outcome or a "good thing", even if it accords with policy.
But I do agree that there's a common misunderstanding that a deletion nomination means that "Wikipedia" (which most people seem to assume is some sort of tech company that controls articles, even if the know that in theory we're the encyclopaedia that "anyone can edit") has decided that the article should be erased. We could easily avoid that by using more active wording, e.g. An editor has nominated this article for deletion [...]. We should probably avoid giving the impression that a single administrator decides anything, and perhaps make more explicit reference to the concept of consensus (since "please share your thoughts" often leads to drive-by voting). – Joe (talk) 10:07, 5 October 2023 (UTC)[reply]
WRT red, it's not just that it can signify deletion (though you're right it does), it also ties into some more general visual shorthand. There is an escalating colour scheme for reader-facing tags - blue for the most neutral things like {{current event}}, yellow and orange for ones which indicate varying levels of content problem like {{very long}} or {{POV}}, and red for the most dramatic ones like {{copypaste}} or the various deletion templates. I think it's quite helpful to have that little visual flag. Andrew Gray (talk) 12:32, 5 October 2023 (UTC)[reply]
Personally I feel a lot of the template messages come of rather cold. "There is a current discussion to decide if this article should be deleted" might come across better than "This article is being considered for deletion in accordance with Wikipedia's deletion policy".
Although I wouldn't worry touch about cranks on twitter using Wikipedia discussions for part of the culture war, whatever is done there will be some (of either side) who try to use it to create outrage. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 19:13, 5 October 2023 (UTC)[reply]
I've made some changes to the wording prompted by this discussion. Please take a look. – Joe (talk) 04:12, 6 October 2023 (UTC)[reply]
Convenience diff. Folly Mox (talk) 04:24, 6 October 2023 (UTC)[reply]
Thanks. Text at WP:PROD and WP:CSD templates should probably be changed too. Hddty (talk) 07:54, 6 October 2023 (UTC)[reply]
For PROD, how would "An editor has proposed that this article be deleted" sound? Going off of the changes to the AfD template. Deauthorized. (talk) 18:55, 6 October 2023 (UTC)[reply]
I don't really know, better ask native English speakers. Hddty (talk) 03:11, 7 October 2023 (UTC)[reply]
@Joe Roe and Sdkb: On my mobile screen there's a space missing before "Feel free", so it reads "retain it.Feel free" – but when I view it in landscape mode, "Feel free" drops down to a separate paragraph so I don't know what's going on there.
This seems like an unexpected place to be discussing major changes to a heavily-used template. Could we move this to VP/PR? For my part, I like the active voice but I don't think the link to the deletion policy should be hidden the way it is, so I'd like to see the wording workshopped a bit. Sojourner in the earth (talk) 13:22, 9 October 2023 (UTC)[reply]
I can't reproduce that exactly, but it behaves strangely on mobile, both before and after our changes. Maybe something to do with this: Template_talk:Article_for_deletion/Archive_8#Edit_request_(mobile_support). Might be worth raising it at WP:VPT.
As for where to discuss, this is a template message, not a policy. I don't think it's necessary to discuss changes centrally, and it does not appear to have been done for previous changes. They were just proposed on Template talk:Article for deletion (because the template is protected) or done directly. – Joe (talk) 14:13, 9 October 2023 (UTC)[reply]
I've linked to this discussion from Template talk:Article for deletion, because that's where editors wishing to discuss this will naturally go (and they might not find their way to Template:Afd/dated, where the changes were actually made). Sojourner in the earth (talk) 15:51, 9 October 2023 (UTC)[reply]
User:Sdkb, in Special:Diff/1178838002 you mention avoiding easter eggs in the edit summary, but link Wikipedia:Deletion policy to the text "whether or not", which confused even me, a person who knows what to expect this template to link to.
Might I propose a mostly-revert? Maybe something like "which will decide whether it will be retained per our deletion policy" or something similar? (I would personally prefer retained in accordance with our or retained pursuant to our over retained per our, but I know we're trying to keep it concise and non-technical.) Folly Mox (talk) 17:30, 9 October 2023 (UTC)[reply]
Or like which will decide whether to retain the article per our deletion policy. I don't know; I'm not really good with this kind of words. I just found "whether or not" to be confusing. Folly Mox (talk) 17:37, 9 October 2023 (UTC)[reply]
Done. {{u|Sdkb}} talk 18:13, 9 October 2023 (UTC)[reply]
I preferred sdkb's more concise wording. I don't think placing the link on "whether or not" was an easter egg; that's what the policy is about. By contrast, using "may" and "per" here makes the message sound colder and more legalistic ("per" here sounds straight-up ungrammatical to me, but I know it's used more loosely in American and Indian English), which is unfortunate given the goal of the sentence is to invite people to a discussion. @Folly Mox: Could you clarify, was it the placement of the link you didn't like, or the actual phrase "whether or not"? – Joe (talk) 04:05, 10 October 2023 (UTC)[reply]
User:Joe Roe, it was the linking of just those three words that had me confused (I even clicked through, so maybe it was a feature). I think "which will decide whether or not to retain it" might be both clear enough for easily confused types like me, and also concise as is preferred. Folly Mox (talk) 04:36, 10 October 2023 (UTC)[reply]
I feel like the struggle for concision is resulting in a sacrifice of clarity in one part of the message or another. Does it really matter if the message is on two lines? My suggestion would be: You are welcome to participate in the deletion discussion, which will decide whether or not the article meets our criteria for deletion. (I prefer "criteria" because it's a little ambiguous whether "meeting the policy" is a good thing or not.) Sojourner in the earth (talk) 04:41, 10 October 2023 (UTC)[reply]
I'm fine with either suggestion, but the deletion policy almost always uses the word "criteria" to refer to speedy deletion, so that could be a source of confusion. – Joe (talk) 05:28, 10 October 2023 (UTC)[reply]
Hmm. What about whether or not there is a valid reason for deletion. (With or without naming delpol after in the same sentence) Alpha3031 (t • c) 06:38, 12 October 2023 (UTC)[reply]
This looks good to me and I think it helps address the problem -- whenever there are high-traffic AfDs, the coverage (in the news and on le social media) tends to be something along the lines of "I can't believe WIKIPEDIA is TRYING TO DELETE this thing". Which is understandable given what the message is -- the old version really did sound opaquely bureaucratic. (especially frustrating when the AfD in question is an inappropriate nom and very few people want to delete it, we're still forced to tell the world that for seven days we're unable to tell whether it's good or bad) jp×g 09:01, 21 October 2023 (UTC)[reply]
You may participate in the deletion discussion, which will decide whether or not to retain it But the discussion doesn't decide whether the article will be retained, the editors in that discussion (and closing administrator) do. – Teratix ₵ 14:53, 23 October 2023 (UTC)[reply]
Yes, that sentence still looks awkward to me, and I agree with Joe above that "per" sticks out as ungrammatical. I'm also not sure about the relevance of the deletion policy to the actual AfD discussion. It makes sense to say that an editor has AfD'd an article in accordance with the policy that describes the deletion process, but we don't really use that policy to decide whether or not the article should be deleted. Could we maybe just revert back to Joe Roe's original edit, which changed the passive voice to active but left everything else untouched? That change seems to be uncontroversial, it's all the later copyedits that are causing difficulties. Sojourner in the earth (talk) 16:04, 23 October 2023 (UTC)[reply]
@Hddty, just in case it wasn't already obvious to you (or others), the big red box shown in that tweet is something that the poster drew. That's not what an AFD notice actually looks like. We use a different shade of red, and a lot less of it. WhatamIdoing (talk) 15:22, 25 October 2023 (UTC)[reply]
I'm already aware that the poster drew the big red box. Hddty (talk) 17:23, 25 October 2023 (UTC)[reply]

Name order change for Japanese (and other) names

I want to amend Wikipedia:Manual of Style/Japan-related articles#Name order to permit the use of Japanese name order in articles. Fundamentally, I think we should use a name order that is consistent with how the person wants to be referred to per WP:BLP. For example, we have some guidelines on how to handle name changes. While this may seem like a minor stylistic change, on Wikipedia talk:Manual of Style/Japan-related articles#Why_doesn't_Wikipedia_respect_Japan? there is a concern that the use that we are having is falling out of favor.

I want to develop more clear and cut guidelines for when to use Western name order and when to use CJKV/Eastern name order. Especially since there can sometimes be no consistency when one or the other is used in reliable sources. This is a balancing act between how a person wants to be addressed vs what reliable sources say. Just as we should not always be using previous names as a privacy interest, we should respect how a person wants to be addressed. Awesome Aasim 18:20, 11 October 2023 (UTC)[reply]

What text would you like to see added / amended at Wikipedia:Manual of Style/Japan-related articles#Personal names? And why was Wikipedia talk:Manual of Style/Japan-related articles not an adequate venue for this? Folly Mox (talk) 18:34, 11 October 2023 (UTC)[reply]
Isn't the point of VPIL to brainstorm ideas and proposals? Even though they might be proposed elsewhere? Or is this just an idea brainstorm page for VPPR? Awesome Aasim 19:08, 11 October 2023 (UTC)[reply]
See the long history of Hikaru Utada page moving. Even determining what a living subject prefers in English/Western media can be challenging and a moving target (and may require rather extensive research).  — SMcCandlish ☏ ¢ 😼  19:25, 11 October 2023 (UTC)[reply]
That is probably why it was added to the MoS to start :D
But yeah, I can see how this can be very confusing. Some sources will use Japanese name order, while others will use English name order. In this case it would be best to defer to what sources closest to the subject use. In the case you mentioned here, shortening the title to Utada (singer) would be a good compromise. Awesome Aasim 22:57, 11 October 2023 (UTC)[reply]
Given that Japan is telling news sources to change their name order, why couldn't we just keep following what sources say, and switch as they do? Won't the problem solve itself? (in most cases anyway) DFlhb (talk) 09:44, 12 October 2023 (UTC)[reply]
I am not seeing where Wikipedia:Manual of Style/Japan-related articles#Name order does not permit the use of Japanese name order. CMD (talk) 12:11, 12 October 2023 (UTC)[reply]
What I am seeing though is that in almost all cases, Western name order is de facto preferred over Eastern name order for Japanese names. Even as both are used interchangeably and without consistency in reliable sources. People want control in how they are addressed. If they don't mind either, then sure we can interchange: Pick one and be consistent. But if they do mind, and that is the whole point, then it crosses into BLP. Oftentimes their website is only in one language so we should look at what name order is used there. We can also use sources that are close to the subject to determine how they should be addressed. Awesome Aasim 14:03, 12 October 2023 (UTC)[reply]
How about:
−
In all cases, the spelling and name order used [...]
+
Unless the subject has stated a clear preference, the spelling and name order used [...]
"clear preference" may address SMcCandlish's point; I think BLP concerns are minimal when we're left to make educated guesses as to the subject's desire DFlhb (talk) 14:21, 12 October 2023 (UTC)[reply]
I can imagine a new template as well: "This article uses Eastern name order to refer to East Asian individuals. In accordance with the relevant style guide, this should not be changed without consensus." Awesome Aasim 17:03, 12 October 2023 (UTC)[reply]
@DFlhb I think this is an acceptable WP:BOLD change. Awesome Aasim 13:53, 13 October 2023 (UTC)[reply]
There should not be an article-wide template, it should be name by name. Politics of Singapore for example, mentions Lee Li Lian and Sylvia Lim in the same paragraph. CMD (talk) 14:25, 13 October 2023 (UTC)[reply]
Agree DFlhb (talk) 06:20, 15 October 2023 (UTC)[reply]

Rewriting WP:N to reflect community consensus

I always thought how we lay out our notability guidelines are a) convoluted b) does not reflect community consensus, most glaring example being

"It meets either the general notability guideline (GNG) below, or the criteria outlined in a subject-specific notability guideline (SNG)"

This lumps all the SNGs into one group wholly seperate from GNG when in fact they have different levels of acceptance and independence from GNG. What is your opinion on replacing the lead of WP:N with something like User:Ca/Notability restructuring? Ca talk to me! 08:45, 12 October 2023 (UTC)[reply]

Well, it does say "A topic is rebuttably presumed to merit an article if:" That seems pretty correct. Gråbergs Gråa Sång (talk) 12:07, 12 October 2023 (UTC)[reply]
At this point I would call NSPORT a supplement to GNG, not a standalone notability guideline. Apart from that, the general principle is good; making it clear that meeting SNG's is not as good as meeting GNG will hopefully help editors realize that to future proof their creations they should ensure they meet GNG. BilledMammal (talk) 13:26, 12 October 2023 (UTC)[reply]
Honestly at this point I think we would be better served by turning NSPORT into an essay, removing all language implying it's a notability guideline, and renaming it to make point clear (even deleting the NSPORT and "Notability (sports)" redirects, which implies an equality with actual notability guidelines that no longer exists.) The simple reality is that the idea of having a sports-specific notability guideline with any actual force to it has been rejected. As a result, the page not useful in its current form and isn't really a "notability guideline" because it is completely subsidiary to the GNG (IMHO all subject-specific guidelines are subsidiary to the GNG, of course; ultimately we need enough sources to write a neutral article. But NSPORT lacks even the force that the others have.) It's just an essay discussing things that might meet the GNG and possible places you might look for sources to do so. We need to put a stake in it once and for all, basically. --Aquillion (talk) 20:47, 12 October 2023 (UTC)[reply]
There are aspects of NSPORT that remain important and suggest a higher bar than GNG to meet the expectations for notability - especially how it discusses prep and college athletes. But it also provides guidance on individual sport seasons. - Enos733 (talk) 22:05, 12 October 2023 (UTC)[reply]
I would not be in favor of the change. While some SNGs are more akin to WP:OUTCOMES than guidelines, there are good reasons why the SNGs exist and why they make sense as subject-specific alternatives to GNG (such as WP:NPROF). In general, the presumption of notability in some SNGs can and do limit which articles are brought to AfD and can provide some guidance to (new) editors about which subjects are likely to merit an article. - Enos733 (talk) 15:54, 12 October 2023 (UTC)[reply]
I think NPROF is actually unique in this way, there are no other "such as". Gråbergs Gråa Sång (talk) 16:54, 12 October 2023 (UTC)[reply]
There are NAUTHOR and NGEO. Newimpartial (talk) 17:25, 12 October 2023 (UTC)[reply]
WP:GEOLAND is a little special, acknowledged, but since when is NAUTHOR "explicitly listed as an alternative to the general notability guideline."? Gråbergs Gråa Sång (talk) 18:02, 12 October 2023 (UTC)[reply]
It isn't "listed" in that sense, but it is "listed" as an exception to NOTINHERIT and offers a presumption of notability independent of the GNG, as NGEO and NPROF also do (as does ANYBIO, for that matter). Newimpartial (talk) 21:00, 12 October 2023 (UTC)[reply]
I'd also note that NAUTHOR, by being an exception to not inherited, often allows us to have fewer articles on the author/their works as several notable books/&c can often be covered in the article on the author. Espresso Addict (talk) 00:31, 13 October 2023 (UTC)[reply]
Also, the real "teeth" of NPROF - the part that matters relative to the GNG - is that it allows us to use a very narrowly defined category of sources that wouldn't usually be considered WP:INDEPENDENT for GNG purposes, on the basis that some academic sources are sufficiently high-quality to overcome that hurdle. Despite what it says, it doesn't really really replace the GNG in that regard, it just clarifies how to apply it in a particular area. --Aquillion (talk) 20:56, 12 October 2023 (UTC)[reply]
Often the sources are actually independent (ie citations in research papers by other scholars) but the coverage would not be considered sufficiently in depth. Espresso Addict (talk) 00:10, 13 October 2023 (UTC)[reply]
Can I ask you what part of the proposal contradicts what your views? I tried to adjust the "tiers" so that it aligns with popular opinion. Ca talk to me! 03:46, 13 October 2023 (UTC)[reply]
I know you weren't addressing me, but I think the biggest problem with the proposal is that it gives GNG notability the highest "level" as (perhaps) more than a rebuttable presumption for a topic to have its own article, compared with other grounds for Notability. I don't think this is true at present, I don't think this view has community consensus, and I think this interptetation of the Notability ecosystem (as a hierarchy, with GNG on top) actively works against the encyclopaedic treatment of topics. Newimpartial (talk) 16:22, 14 October 2023 (UTC)[reply]
The prose of WP:N had a recent (last 2 or 3 years) addition to spell out the complicated nuances of the SNGs to the GNG. It is more complicated than the nutshell statement but the nutshell statement captures 95% of that otherwise. There is no need to change. Masem (t) 16:59, 12 October 2023 (UTC)[reply]
The key thing is "presumed" - which means that if it meets these guidelines, it is up to editors at a deletion discussion that the page should not be included for WP:NOT reasons. Awesome Aasim 17:42, 12 October 2023 (UTC)[reply]
The thing that I think we need to make clearer regarding SNGs is that, for the most part, they are guidelines to tell us when sources probably exist and where to look for sources. The thing we need to make clearer about the GNG is that it is explicitly not about if something is worthy of an article, it's about the fact that we cannot write an article without sources of information. Something is presumed to be notable (worthy of an article) if there are sufficient sources of information that we can write about it. If there are no acceptable sources, it's not that it's not worthy of notice, it's that we do not have the necessary raw materials to build an article. We have consistently made a mistake by calling topics for which there is insufficient available sourcing "non-notable". I do not think your proposed text clarifies that, and indeed I think it makes the situation more confusing, not less. ~ ONUnicorn(Talk|Contribs)problem solving 18:18, 12 October 2023 (UTC)[reply]
I do think that this bears repeating. The reason why the GNG finally put the first round of major disputes over deletion to rest is because it recognized that the key question wasn't "is this notable" but "do we have enough sources to write a neutral, independent article?" I don't think that an article that fails to pass the GNG is ever acceptable - it's the absolute minimum baseline below which it doesn't matter how "notable" an editor feels something is, we just don't have enough to write even the most basic stub in compliance with our policies. The real problem with articles that fail the GNG isn't that they're obscure or non-notable or whatever, it's that they involve original research, putting undue weight on a single source, relying too heavily on non-WP:INDEPENDENT sources, or relying too heavily on low-quality sources like raw databases and the like, which should never be used as the sole basis for an article. --Aquillion (talk) 20:50, 12 October 2023 (UTC)[reply]
Its also important to remember that the GNG itself is also a rebuttable presumption. Just showing two or three sources may be reasonable at the start of an article's life, but the article cannot reasonably be expanded past that, it still can be put to deletion. It's part of the complex nature of the GNG and the SNGs that can't be easily expressed in one sentence, only the majority of cases as we already do in the lede. Masem (t) 00:14, 13 October 2023 (UTC)[reply]
I do feel that the SNGs might be useful if, instead of trying to replace the GNG, they became more loose guidelines for "what should guide our judgement after the bare minimum of the GNG is satisfied." It might be worth at least considering an attempt to change the GNG to that end, ie. the GNG ought to be a hard bare minimum (with perhaps a small number of footnotes for cases where sources might be more useful than they'd appear at first), with SNGs mostly serving to answer where to go after that bare minimum has been satisfied. --Aquillion (talk) 05:21, 13 October 2023 (UTC)[reply]
I don't think the SNGs should be trying to serve that purpose only. We want the GNG and the SNGs to encourage editors to create articles that they can demonstrate may likely be notable (due to either existing sources or some type of high merit), but after that, when the rebuttal presumption comes into play is if there is simply no way to readily expand the article with more sources beyond a stub shape or other problem with NOT (eg overreliance on primary sources), and then we can talk whether retention in a larger article comes around or whether deletion is better. The SNGs can serve to say "look for additional coverage via these routes..." or "It may be better to cover this type of topic in a larger article..." with advice tuned to the field the SNG covers, but I don't think we should expect the SNGs to be for this purpose. Masem (t) 16:45, 14 October 2023 (UTC)[reply]
A side effect of the levels, is that we shift from: GNG is for everything, SNGs are for some stuff, to: GNG is the strongest form of notability, SNGs are weaker forms. I agree with BilledMammal that there are benefits. But sources may fulfill GNG without being sufficient to let us to write a basic, neutral outline of the subject. GNG is already applied too rigidly (i.e. nominally meets GNG = must Keep). These levels, with GNG placed at the top, risk making that rigidity worse, even omitting the are very rarely deleted wording. DFlhb (talk) 18:55, 12 October 2023 (UTC)[reply]
I agree--I changed the wording around a little bit to reflect this. Ca talk to me! 06:22, 13 October 2023 (UTC)[reply]
  • More generally, no simplistic attempt to put GNG and SNG into tiers is ever going to reflect the complex relationships. Sometimes GNG trumps SNG, sometimes SNGs are held to overrule the GNG. The truth is the guidelines are often in conflict with each other; different editors understand that tension in different ways, and AfD exists to allow those different interpretations to be discussed as they apply to one individual case to come to a consensus on that individual case. If there was an easy codeable answer we could use just use speedy deletion. Espresso Addict (talk) 00:17, 13 October 2023 (UTC)[reply]
    I agree with your point entirely. To be realistic, notability is a mess, and no single systematic treatment would be sufficient to describe how notability is applied across all of Wikipedia. However, I believe that the adage "perfection is the enemy of good" applies here. It took me insanely long time grasp what notability is(I used a different account in the past), and some more time to understand the complex interplay between SNGs and GNG. I incorporated your suggestion and softened the wording to show that this tier system is not meant to be rigid—–exceptions may, of course, occur. Though it is not perfect, I believe this change would provide at least a general guidance to new editors. Ca talk to me! 00:55, 13 October 2023 (UTC)[reply]
What might be helpful is some sort of Notability for Newbies essay, though presumably several already exist. Indeed I believe one main point of the more-explanatory SNGs is to try to explain what is often considered notable in a particular topic area to assist newer editors. Usually when I talk to new editors I try to explain the individual case at least briefly, and link to the Teahouse for further advice. In an ideal world, very new editors should not be creating new articles nor trying to get existing ones deleted. Espresso Addict (talk) 01:48, 13 October 2023 (UTC)[reply]
I wonder if NPPSCHOOL training materials couldn't be partly integrated into our notability guidelines. Most would be out of scope, but they do a robust job of explaining the way GNG and each SNGs fit together, and which takes priority in each case. DFlhb (talk) 06:05, 13 October 2023 (UTC)[reply]
  • GNG is not "the highest form of notability". In particular, for WP:NEVENTS meeting GNG is necessary but not sufficient to demonstrate notability. Some SNGs are stricter than GNG rather than being a run-around. Fences&Windows 10:57, 22 October 2023 (UTC)[reply]
    I interpreted the "lasting coverage" requirement in NEVENT as just showing how WP:NOTNEWS is applied when considering whether to have an article or now.
    After all WP:N currently says

    A topic is rebuttably presumed to merit an article if:

    1. It meets either the general notability guideline (GNG) below, or the
    2. criteria outlined in a subject-specific notability guideline (SNG); and It is not excluded under the What Wikipedia is not policy.
    • NEVENT is just clarifying how WP:NOT applies in practice. Ca talk to me! 12:10, 22 October 2023 (UTC)[reply]

Code of conduct for outreach projects

Do we have anything like a code of conduct for the organizers of outreach projects? For example, I suspect the following statement reflects our wishes but I don't know if we have it stated formally anywhere: "Organizers of outreach projects are expected to monitor the quality of participants' contributions. They should ensure any negative effects of the project such as copyright violations are promptly removed. It is not acceptable for a funded project to leave serious problems in Wikipedia for volunteers to find and clean up." Clayoquot (talk | contribs) 22:15, 12 October 2023 (UTC)[reply]

I don't think the movement has any written rules to that effect. I'm not sure that it's entirely applicable everywhere, or that this would be entirely appropriate. Consider:
  • The project has funding, but nobody's being paid for their time. Is it logical to say that the volunteers who organize an event be expected to deal with "any negative effects", so that volunteers (a group that includes the organizers themselves) don't have to do any work?
  • The project affects a small wiki, but there's no material difference between "the organizers" and "the volunteers" at that wiki. Do we need a rule that says they have to do the work that they're going to do anyway?
More broadly: Is it actually a bad thing for us to "clean up" after newbies who are trying to contribute positively? I don't know about you, but my first edits were not all perfect. I calculated recently that for the movement to get one high-volume editor like me, we have to put up with thirty thousand (30,000!) first edits. To get someone like you, we have to suffer through three thousand first edits (and 2,000 second edits, and so forth). Many of those edits are going to be imperfect. Some of them will be outright vandalism. This means, in turn, that the editors who welcomed us back in the day put up with many thousands of first edits just to find the two of us. If we want Wikipedia to survive us, we will need to extend that grace to the next generation.
And, realistically, outreach efforts might be better than the average newbie. Vandalism is much rarer with organized outreach events, and most of them provide some advice and check some of the contributions. They might not catch or fix everything, but they aren't worse than the completely unguided, unchecked contributions from equally new contributors. WhatamIdoing (talk) 17:34, 13 October 2023 (UTC)[reply]
Thanks for sharing your thoughts. This discussion is at the English Wikipedia Village Pump as I have no ambitions to tell anyone how to conduct their work on other Wikimedia projects. After reading your response I'm thinking perhaps some context on where I'm coming from might be helpful. I am not anti-outreach. I've devoted many hours to outreach over the years. I share your belief that attracting and welcoming new editors is critical for the future of our project. I'm also aware that many outreach events go fine and that these are the ones that we tend to not remember.
A subset of outreach events don't go well. Some go badly enough to overwhelm our quality control processes and cause experienced volunteers to temporarily burn out or worse. The worst events produce large volumes of crap that is harder to get rid of than most vandalism is. When that happens, we yell at the organizers to clean up the mess and we are unhappy if they don't. But the organizers 1) don't know if the people asking them to clean up actually represent Wikipedia, and 2) haven't budgeted any hours for cleanup because they thought Wikipedia stays crap-free by some sort of magic. This process is the status quo. Perhaps we can improve on the status quo by giving organizers our expectations in advance.
Another reason to share our expectations is that some donors hate the idea of sponsoring a project that ends up harming Wikipedia. If the conversation around grant-making can involve assurances that the organizers will follow Wikipedia's code of conduct for organizers, that can provide donors with a level of comfort that the risk of harm will be well-managed. Without this kind of assurance, some donors don't want to sponsor Wikipedia projects at all. Clayoquot (talk | contribs) 22:07, 14 October 2023 (UTC)[reply]
Thinking about it from the good-faith organizer's side, imagine that you are organizing an online event. You've got some experience and you've done your homework, but something goes spectacularly wrong – say, you were expecting 20 people but you got 200, and half of them apparently subscribed to the theory that if it's not a copyright violation, then it's a violation of Wikipedia:No original research.
Now what?
Would you really want us to tell you that you have to personally clean up everything by yourself, while we all sit on the sidelines and complain about what a 'bad organizer' you are? Or would you like some help? WhatamIdoing (talk) 06:13, 15 October 2023 (UTC)[reply]
I think the best would be for organisers be aware that something has gone wrong and ask for help. Unfortunately it seems all to often organisers are unaware that anything is wrong. I would hope that the community would help anyone genuinely asking for help, I know I certainly would. I think of lot of poor sentiment comes from the community discovering this issue, and being unable to get an reply from the organiser. Maybe what is needed is better avenues of communications and some standards of expectations of communication. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 20:55, 17 October 2023 (UTC)[reply]
@WhatamIdoing, regarding the spectre of editors making uncivil comments such as 'bad organizer', the possibility of incivility by people who think they are upholding behavioural standards is always a possibility. This possibility doesn't usually inhibit us from trying to articulate what our behavioural standards are.
I think both of you have made very valid points around flexibility and collaboration. "Clean everything up by yourself" seems to be too much to ask of everyone. How about starting with something along the following lines:
  1. Watch participants' user talk pages so you can be aware of issues raised by the community
  2. Spot-check participants' edits for quality
  3. Monitor your own user talk page and engage with questions and concerns that are brought to your attention there
  4. If people are raising concerns that you do not understand or know how to fix, ask at the Teahouse
  5. When planning a project, set aside some organizer time/budget to monitor and resolve issues with the quality of contributions
Clayoquot (talk | contribs) 23:32, 18 October 2023 (UTC)[reply]
That sounds like a model that is common for in-person events (e.g., student group meets at the library). I'm not sure it would scale to, say, Wikipedia:Wiki Loves Monuments. WhatamIdoing (talk) 15:26, 25 October 2023 (UTC)[reply]
Outreach projects vary considerably in their format. Organisers often don't establish a relationship with the attendees or even know their accounts. For example, I recently attended an event which was organised as a project of the Oxford Food Symposium. The format was an online presentation which lasted just an hour and so just covered the basics of creating an account. The main presenter mainly had to cope with delivering the presentation which wasn't simple as she was using an unfamiliar computer. I and other experienced attendees helped out in various ways. One did some screen sharing while I focussed on tracking down and welcoming the newbies. As an event coordinator, I also gave them confirmed editor status for ten days to minimise trouble like captcha.
Sometimes, the dashboard feature is used to help identify participation and follow-up but that usually has a limited timescale. For a longer-term relationship, there's the mentorship features but I've not used them much yet. See the FAQ for details.
Andrew🐉(talk) 09:13, 24 October 2023 (UTC)[reply]
I've been scratching my head over this for the past few days. The frustrations I have - and that I think we as a community have - are with well-funded projects where the organizers' success metrics are number of edits, number of articles created, number of participants, etc., with no attempt to measure or mitigate negative effects. Some of these projects involve series of editathons. Another type of project that I have in mind is where professional Wikipedians are paid to improve articles on sustainability issues. Sometimes a group gets a second round of funding and uses it all to do more stuff rather clean up past mistakes. A behavioural guideline would need some kind of scope about whose behaviour it applies to but 'm not sure how to articulate it yet. Clayoquot (talk | contribs) 20:03, 26 October 2023 (UTC)[reply]

Project Res-Up

Project Res-Up by Adobe uses generative AI to upscale video. However, according to The Verge, when a sample video was upscaled, "[t]he resulting footage was much sharper, with the AI removing most of the blurriness and even adding in new details like hair strands and highlights." Given the addition of new details to old film, would this be copyrightable when used on a PD film, and should we be banning it on WP because it adds new details to video? Just to be clear it's not out yet so I wanted to heads up users here. Therapyisgood (talk) 07:38, 14 October 2023 (UTC)[reply]

Have you read m:Wikilegal/Copyright Analysis of ChatGPT?
As for replacing "blurry hair" with "hair that shows some individual strands", I suggest that this isn't really a case of creating content. The original people in the old films assuredly had individual strands of hairs on their heads. WhatamIdoing (talk) 06:19, 15 October 2023 (UTC)[reply]
By extension, would you accept dialogues between historical persons invented by an author? After all, we can be fairly sure that historical persons who we know were in contact had conversations. When does the addition of invented details cross the line? (Edit) Perhaps more pertinent is the new function in Google's Pixel 8, which will switch images of faces from old photos on to new photos you take.Flawless or fake? Google's new AI now fixes smiles. Donald Albury 14:02, 15 October 2023 (UTC) Edited 14:12, 15 October 2023 (UTC)[reply]
Since the exact words sometimes matters, I'd tend to reject a made-up conversation (although there are very few encyclopedia articles that should be quoting whole conversations). Since the exact hairs on someone's head never matter (in an encyclopedia article), I wouldn't worry about that. We need to focus on what actually matters (see, e.g., Materiality (law) or Materiality (auditing)).
What we don't need is another round of an editor saying that the Portraits of presidents of the United States are all false and should be deprecated because the artist might have put a few strands of hair in the wrong place. If you'd accept a reasonable level of artistic license in a hand-painted image, you should accept the same level of artistic license in a machine-generated image.
We shouldn't misrepresent the works. Any level of image manipulation that goes beyond the routine and ordinary (e.g., cropping) needs to be declared. WhatamIdoing (talk) 20:22, 15 October 2023 (UTC)[reply]
The Pixel 8 feature has been wildly misunderstood by many. All the feature does is let you switch between faces taken of the same person in the same shot. As in, if you take a group selfie and hit the shutter button three times, you can switch Person A's face in Photo #1 with Person A's face in the Photo #2 or #3. It's not as wild as you might think — and the technology has been around for decades, just never to a wide audience. InfiniteNexus (talk) 02:43, 16 October 2023 (UTC)[reply]
That sounds like a great solution to the group photo problem: She blinked in photo #1, he blinked in photo #2, and now you can make a sort of collage that gives you one picture containing nobody blinking. WhatamIdoing (talk) 04:18, 16 October 2023 (UTC)[reply]
Yeah, that's the pitch. (Meanwhile, the biggest piece of news from Apple's event was ... USB-C?) With that being said, other features do involve generative AI "content" creation, for instance artificially generated pixels to replace deleted ones. So perhaps that is a better example than Adobe's AI to ask whether partially AI-generated images are copyrightable. InfiniteNexus (talk) 05:33, 16 October 2023 (UTC)[reply]

Idea for Entry on Banned Board Game

Hey all, hope this is the place to do this. Im not quite experienced enough with this site to do this myself, but I found a very interesting banned board game that doesn't have a Wikipedia entry. Here are a few links if anyone would like to get started on this:

https://www.youtube.com/watch?v=cUXkBsfSZ_Y

https://archive.org/details/Public_Assistance_Game/

https://boardgamegeek.com/boardgame/3393/public-assistance

https://www.nobleknight.com/Publisher/Hammerhead-Enterprises

Good luck, have fun, and good night! JikiScott (talk) 17:18, 17 October 2023 (UTC)[reply]

Read Wikipedia:Notability and Wikipedia:Reliable sources. There are innumerable things that don't have Wikipedia articles, and almost all never will.AndyTheGrump (talk) 17:26, 17 October 2023 (UTC)[reply]
And who exactly banned this? I can see mentions of stores not stocking this, as is their right, but no evidence that this was banned, or that anyone had the power to ban it. Phil Bridger (talk) 18:27, 17 October 2023 (UTC)[reply]

RFC Close Review

This is a follow-up to a topic that is being mentioned at WP:AN in the context of a close review of an RFC, and the topic is how the review of closes of Requests for Comments should be conducted. One editor said that maybe the discussion of procedures for close reviews should be done at Village pump Idea Lab, so here we are.

It had been mentioned that close reviews of RFCs are sometimes problematic because the editors either defend their original positions on the RFC, or each provide their own close of the RFC as what they think should have been done. Sometimes the first issue has been partly addressed by providing two sections in the close review, for Involved Editors, who participated in the RFC, and Uninvolved Editors. I will comment on the second issue that the most active close review forum is Deletion Review, and DRV has a rule that "DRV is not AFD round 2". The regular editors at DRV know that their role is to review whether the close was a reasonable judgment call by the closer, not whether it was the close that they would have provided. There should be a rule that "RFC Close Review is not RFC Round 2". Even after the discussion that a close review is not a new close, some of the editors at WP:AN are proposing their own closes.

It has also been proposed that a separate noticeboard may be in order for RFC close reviews. It might not be used any more often than AARV.

So I see two possible improvements to RFC close reviews. First, continue to review them at WP:AN, but with a set of procedures that include "Close Review is not RFC Round 2" and the partitioning of involved and uninvolved editors. Second, set up a separate noticeboard. My recommendation is the first, but this Idea Lab can hash out ideas for both concepts. Robert McClenon (talk) 21:32, 17 October 2023 (UTC)[reply]

Much of which is already covered at Wikipedia:Closing discussions#Challenging other closures, creating WP:RFC review as a redirect to that section might be a start. Although I'd prefer it actually start as a guide of what is expected and how to setup the review. The current process is freeform, which often results in a maul, so some kind of structure to the discussion could help. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 11:10, 18 October 2023 (UTC)[reply]
A maul??? Is that like biting the newcomers? EEng 13:03, 19 October 2023 (UTC)[reply]
Depends how much the newcomer likes to ruck. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 17:23, 19 October 2023 (UTC)[reply]
What kind of football is that? Robert McClenon (talk) 17:54, 19 October 2023 (UTC)[reply]
The one where you handle the balls, without protection. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 18:48, 22 October 2023 (UTC)[reply]
Okay. The football game that is more or less the oldest, and is not familiar to North Americans. Robert McClenon (talk) 23:47, 22 October 2023 (UTC)[reply]
Cuju? Or maybe Episkyros? Anomie⚔ 12:00, 23 October 2023 (UTC)[reply]
I agree with ActivelyDisinterested that some kind of structure will help. The idea that has been suggested more than once is that the close review be divided into sections for Involved editors, that is, those who participated in the RFC, and Uninvolved editors. I would also suggest that the close review begin by stating that the question is whether the conclusions of the closer are consistent with the results of the RFC, because a close review should not be RFC round 2. In my experience, if the close was anything other than Yes or No, each of the participants is likely to provide their own close that is a different compromise. Robert McClenon (talk) 21:09, 20 October 2023 (UTC)[reply]
What's the benefit to having separate ===Sections=== for un/involved editors, that we can't get merely by encouraging editors to self-label when they comment (as they often do already)? WhatamIdoing (talk) 05:16, 26 October 2023 (UTC)[reply]
Is Wikipedia:Discussions for discussion not the place for this? Nardog (talk) 21:08, 21 October 2023 (UTC)[reply]
That page is for potential evaluators of consensus to get opinions on a specific discussion regarding how to evaluate and close the discussion. isaacl (talk) 21:23, 21 October 2023 (UTC)[reply]
Is there a reason that Discussions for discussion could not be expanded to be the place for RFC reviews? Robert McClenon (talk) 23:47, 22 October 2023 (UTC)[reply]
I think expanding the page's scope to encompass reviewing the results of RfCs would attract a lot of participants with topic-specific interests in mind and thus diminish the focus in discussions regarding how consensus should be determined. I don't think it would be efficient to essentially re-use the page name for a new forum. isaacl (talk) 00:13, 23 October 2023 (UTC)[reply]

Discussion on changes to the edit notice of WP:Reliable sources/noticeboard

I've started a thread at WT:RSN#Suggested changes to the edit notice, to discuss a proposed change to the edit notice of the Reliable sources noticeboard. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 21:04, 21 October 2023 (UTC)[reply]

Adjusting the extended-confirmed criteria

At the moment, extended confirmed is a useful tool, but it suffers from a few deficiencies, most significantly that it is easy to game.

I would suggest we change the criteria to the following:

  1. 500 significant edits overall
  2. Including at least 250 in main space
  3. Including at least 100 in talk space
  4. Significant edits defined as "larger than 200 bytes" (the definition the community has been using at the various WP:LUGSTUBS requests)

From what I understand, this is not possible to implement on the MediaWiki side; instead we would need to create an adminbot that automatically grants the permission when the criteria is met.

This discussion is related to this ArbCom clarification request. BilledMammal (talk) 03:01, 26 October 2023 (UTC)[reply]

The criteria can be gamed, but in examples I recall seeing it was always quite obvious that it was gamed. Unless I'm mistaken, the user right can be removed in those cases. (Am I mistaken?) CMD (talk) 04:41, 26 October 2023 (UTC)[reply]
It can, but it isn't always, and that is even under circumstances where it is identified and reported. BilledMammal (talk) 04:48, 26 October 2023 (UTC)[reply]
@BilledMammal, your account is 4.5 years old now. Have you calculated how long this would keep you locked out? I'm betting it's somewhere between 2 and 2.5 years, but I haven't bothered to see how many of your early edits were "insignificant". Do you have a rationale for us wanting to keep editors like you away from difficult subjects even longer?
It might also be interesting to see how much this would disfavor some of our patrollers. We have good editors whose net contribution is a negative number of bytes. WhatamIdoing (talk) 05:12, 26 October 2023 (UTC)[reply]
Looking it up, it would have delayed me by just under six months; that might have been reasonable, but Redfiona's "still not extended confirmed" is definitely not. Clearly this proposal is faulty; I'll suggest a different one below. BilledMammal (talk) 09:35, 26 October 2023 (UTC)[reply]
I have a distinct feeling that I'd still not be extended confirmed by these criteria, and I've had an account for 17 years. (Edited to add, I had a quick look on xtools, I became extended confirmed in 2016, I think these would likely have extended it to 2018 so WhatamIdoing's "an extra 2 years" maths works out for me.) Red Fiona (talk) 07:22, 26 October 2023 (UTC)[reply]
Even without checking the "significant edits" criterion, my 500th edit (as caeciliusinhorto) was in October 2015, a little under a year after I joned; my hundredth talkpage edit was in April 2016, the same month I hit 1000 articlespace edits and got my second good article. The significant talkspace edits criterion would push me just over a month further, to 9 May 2016. In my first year and a half I didn't have any interest in ECP areas (and to be honest largely still don't) so it wouldn't have personally affected me, but tightening up the restrictions like this would certainly have a significant impact on honest editors. Caeciliusinhorto-public (talk) 09:28, 26 October 2023 (UTC)[reply]
Actually, you still wouldn't be; you've only made 49 edits considered "significant" under the above definition to the talk namespace. My thinking was that it might be beneficial for editors to demonstrate they can communicate and collaborate, but that isn't practical because clearly you and editors like you should be extended-confirmed.
Perhaps we just change it to:
  1. 500 significant edits overall
  2. Significant edits defined as "larger than 20 bytes"/"large than 10 bytes" ("Larger" includes both edits that add content and edits that remove content)
20 bytes would have delayed my ECP by two months and RedFiona's by three months; 10 bytes would have delayed mine by two weeks and RedFiona's by two months.
The most common form of gaming that I have seen is adding a single wiki-link to article, which is a four byte edit; even if we only count edits over 10 or 20 bytes we will still create a substantial barrier to that gaming while having a negligible effect on good faith editors. BilledMammal (talk) 09:35, 26 October 2023 (UTC)[reply]
That feels more reasonable to me. Even the ten byte threshold is enough to count out edits which only add or remove one or two wikilinks, change some punctuation, or fix a simple spelling error; I don't think it's unreasonable not to consider those edits as counting towards demonstrating the experience we would want to see from users in contentious ECP areas. Of course it's impossible to prevent gaming entirely without making life very difficult for good-faith editors, but if we think the current system is too easy to game this seems like a reasonable tweak to make things slightly more difficult. Caeciliusinhorto-public (talk) 09:56, 26 October 2023 (UTC)[reply]
I wonder if it'd be not too hard to use a definition of significant that would count changing the sky is pink and fully fluorine to the sky is blue and mostly nitrogen. That gets a +1 but is a significant edit. Bit of an edge case but just looking at the +/- can be misleading with how much it changes.
Another threshold, which I think is simpler but captures some of the earlier proposed talk page edit count could be something like "Made significant edits to at least 10 distinct pages, excluding their own user and user talk space." Skynxnex (talk) 17:09, 26 October 2023 (UTC)[reply]
@BilledMammal: I feel like this might just kick the can to other gaming strategies like adding shortdescs or expanding bare URLs (often +1kB). People who are really determined to game EC always will. If there's a group we can stop, it's the one's doing it on the spur of the moment—"Eh, I've got a few hours to kill, I can make 500 easy edits so I can tell the world how the real bad guy is $country". To that end, we could make it so you don't actually get EC for 4 days after qualifying. Even would leave room for admins to glance at who's on-deck to get EC, and whether any gamed it, and disable autopromotion if so. -- Tamzin[cetacean needed] (they|xe|she) 17:21, 26 October 2023 (UTC)[reply]
If that doesn't tell me to use the talk pages more, nothing will :) Suggestion 2 works for me. I don't tend to edit pages that are covered by ECP, I was trying to think of a me who happened to be in a country whose articles are more frequently covered by that sort of thing. Red Fiona (talk) 18:54, 26 October 2023 (UTC)[reply]
An alternate suggestion: exclude reverted edits (and probably reverts also) instead of counting bytes. That both takes care of the add-link/remove-link 250 times strategy, and makes it more likely that, even if they do just do something like 500 shortdescs, we get some minimal benefit out of their gaming. (Under the original proposal here, I'd have been an admin well before being extended-confirmed. That 100 200-byte talk page edit requirement's a killer.) —Cryptic 03:38, 27 October 2023 (UTC)[reply]
  • How many editors gain this permission on average every day? I'm wondering whether it would be feasible to delay granting permission for a short window, and bot-add a permission request somewhere, so that admins/experienced editors could vet the contributions and filter out obvious bad-faith editors. Espresso Addict (talk) 22:01, 27 October 2023 (UTC)[reply]

Centralized source assessment project

Would it be a good idea to create a source assessment project for subjects that are not (yet) notable or where it's unclear if they are notable or not?
Recently I created Wikipedia:SSSniperWolf sources overview and I think it's working pretty well. I moved that to project space on purpose because I didn't want to do everything by myself. All the potential sources are in one place instead of being scattered over various deletion discussions and deleted articles only admins can see. This makes it much easier to track the notability of a subject which often progresses over time. Questions like "doesn't (insert link to article) make this subject notable?" can be answered with a link to the centralized source assessment for that subject.
I'd imagine creating a project page named Wikipedia:Source assessment and creating subpages for subjects. An obvious concern would be for such assessments to turn into quasi-articles, but with some clear rules I think that's avoidable.
Thoughts? — Alexis Jazz (talk or ping me) 16:39, 26 October 2023 (UTC)[reply]


Collaborating with other encyclopedias such as Britannica

Instead of Wikipedia just dominating what's stopping you guys from collaborating with other encyclopedias to share/pool knowledge or help each other out in some other way? Americanfreedom (talk) 15:27, 27 October 2023 (UTC)[reply]

We collaborate by default by licensing our content permissively: Wikipedia:Reusing_Wikipedia_content. Other encyclopedias are generally free to use Wikipedia content within the terms of that license. If the other encyclopedias were to return the favour, we could do likewise. Barnards.tar.gz (talk) 15:42, 27 October 2023 (UTC)[reply]

Slavery Icon - near title for articles relating to the exploitation via slavery

My suggestion is for an icon (Chain or Fetters) on these types of articles:

• modern biographies: slavers, slave owners, slave traders.

• modern Monuments (statues) that are of one of the above

• modern Monuments (buildings) whose construction was financed in part or wholly by slavery

• modern Institutions who were funded in part of wholly by money originating from slavery, slavers, slave owners, slave traders

That the icon be placed at the top of the article.

That the icon can easily be placed/removed by any editor.

That the article must have an an explanation in the body of the article relating to the links on slavery.


I think there is a need for this as:

• it reflects a modern political consensus

• it reminds the reader that everything about the person/building/institution is tainted (a sort of shadow)

• it is an ironic branding for those who branded.


This is very much a draft type of suggestion, feel free to suggest alternatives, adjustments, or even rejection.

Yours ever, ~~~~ Czar Brodie (talk) 15:33, 27 October 2023 (UTC)[reply]

Why slavery specifically? There are hundreds (probably more, probably infinite) topics that articles could be icon-tagged against, and this seems a particularly controversial one. Barnards.tar.gz (talk) 15:45, 27 October 2023 (UTC)[reply]
Slavery is something that I noticed is becoming very topical in labelling (people, buildings, institutions) outside Wikipedia, and its organisational Wikipedia absence caught my eye in articles I was editing. (I was looking for a way to label articles I noticed had slavers/slave exploitation/slave profit.)
Yes I agree that this concept could be expanded , but I do not think something should not be done because it has potential to be expanded. Yours ever, Czar Brodie (talk) 16:02, 27 October 2023 (UTC)[reply]
The usual Wikipedia way of approaching this is categories, e.g. Category:Slave owners. There are mature guidelines to pay attention to that may temper the instinct to tag/categorize everything that looks a little bit slavery-adjacent. Barnards.tar.gz (talk) 16:29, 27 October 2023 (UTC)[reply]
Yes, very good point, but my suggestion is for something bold and visible other that an academic grouping tucked at the bottom of the page with a bunch of other categories. i.e. John Threlkeld. Yours ever, Czar Brodie (talk) 16:48, 27 October 2023 (UTC)[reply]
apologies I forgot to address your good point : "a particularly controversial one". Yes, agreed, but I do not think that is reason to shy away from the issue. Yours ever, Czar Brodie (talk) 16:16, 27 October 2023 (UTC)[reply]
  • No. This is political activism, and an inherently-contentious form of that. Should University of Virginia be tagged with your scarlet letter? Virginia? Genghis Khan? There are no answers, and that is why this is an idea that shouldn't be pursued. Walt Yoder (talk) 16:10, 27 October 2023 (UTC)[reply]
    University of Virginia, probably yes, with explanation in body of the article as to why; Virginia, no, states, countries in my view are outside my suggestion (needs to be focused). Genghis Khan, probably, the idea would not be limited to European/American slavery. Is it really that contentious? Those who think slavery is a good idea are now very much a minuscule minority. The contention would probably people who do not want to be reminded of this historical event. But is it Wikipedia's place to not upset them for consensus? Yours ever, Czar Brodie (talk) 16:26, 27 October 2023 (UTC)[reply]
    • It has nothing to do with anyone believing slavery is a good idea. It has to do with WP:NPOV and with the proliferation of such icons one could add (religious oppression? Gains from violence, e.g. the art in the Louvre acquired by the Napoleonic Wars? Women's oppression? Child abuse? Child labour? LGBTQ+ discrimination? ...) All it would do is clutter our pages, create countless heated discussions and edit wars, for very little actual benefit (more moral posturing than anything else). Fram (talk) 16:42, 27 October 2023 (UTC)[reply]
      I disagree that something should not be done because it has the potential to proliferate. In a sense, for me, this this goes against the philosophy of Wikipedia. I see proliferation as beneficial, so it is difficult for me to argue. My idea of an icon is kind of designed to avoid clutter. I think only certain pages would have heated discussions. i.e. the example I used above, John Threlkeld, is not controversial so I doubt there would be edit wars, but George Washington on the other hand....I agree, but there will always be edit wars on these subjects with very notable people. I get your point, but I personally do not think this is reason to not proceed. Yours ever, Czar Brodie (talk) 17:07, 27 October 2023 (UTC)[reply]
  • I think this idea would bring more heat than light to Wikipedia. Yes, I agree that those who think slavery is a good idea are now very much a minuscule minority, but in the ancient world it was very much accepted. Would this icon be applied to nearly everyone from Ancient Greece or Rome that we have articles about? If we were to apply it consistently then it would. We should just describe people or institutions that were supported by slavery as reliable sources do, and in many cases that would mean inclusion of such content, but not a yes/no icon. Phil Bridger (talk) 19:48, 27 October 2023 (UTC)[reply]
    Very strong point @Phil Bridger that forces me to reconsider and adjust my proposal. According do you think it would be viable to limit the icons for those people/buildings/institutions in the modern era? Or still problematic? Yours ever, Czar Brodie (talk) 20:01, 27 October 2023 (UTC)[reply]
    I imagine there would also be difficulty and contention in labelling, for example, the Kafala system, which some have called "modern slavery" and others have rejected this description. Curbon7 (talk) 20:12, 27 October 2023 (UTC)[reply]
    Thanks for that @Curbon7, yes Kafala system is a good example. As with all articles it is best to go with the sources. From what I understood from reading the article it is not slavery but creates a situation where slavery becomes possible, so no, probably no icon for Kafala but rather icons for those who use it for slavery. My suggestion is for an icon to be be used on people, structures, and institutions, and I think Kafala system is more a law. But I did wonder if the football stadiums mentioned in the article could come under what I am suggesting. My conclusion is that if a stadium in mentioned by reliable sources as being constructed with slave labour (not just the Kafala system) then, yes, I would not have a problem with an icon identifying such, and think it would be helpful and informative to have this highlighted at the top of the article. Czar Brodie (talk) 20:48, 27 October 2023 (UTC)[reply]
Retrieved from "https://en.wikipedia.org/w/index.php?title=Wikipedia:Village_pump_(idea_lab)&oldid=1182225883"