Wikipedia talk:Notability

Currently, 2026 Maldivian local elections is under AFD. Usually, national-level summaries of local elections are never sent to AFD and say "it fails WP:NEVENT" (can someone try this at 2026 United Kingdom local elections?), but here we are. Thankfully, it looks like the article will be kept. Is it a good idea to have an WP:SNG on this, similar to WP:NPOL?

• Direct elections to the following:
  • Head of state of a country listed at List of sovereign states
  • A chamber of a legislature of a country listed at List of sovereign states, where all seats, or a specific class of seats, are contested
  • A by-election (special election) to a chamber of legislature listed at List of legislatures by country#Legislatures of sovereign states (Member and observer states of the United Nations)
• National summaries of local elections of a country listed at List of sovereign states
• Elections to the European Parliament
• United Nations Security Council elections

Howard the Duck (talk) 18:32, 6 April 2026 (UTC)

I would consider this to be part of NEVENT, as to avoid creating a whole new SNG.

We'd also want to be clear that per NOT we generally do not cover local / regional elections in depth, with some exceptions (like for mayors of major metropolises), general if those recieve detailed national coverage or better. Masem (t) 19:00, 6 April 2026 (UTC)

Specific local elections in a specific place should not be construed as being a part of this proposal, just national-level summaries, like the aforemention article on AFD. Howard the Duck (talk) 19:11, 6 April 2026 (UTC)

I'm just saying if you considering an SNG or section for elections, i would make sure it covers what we generally allow and we generally decline for all such elections, and not just the examples we want to keep. Masem (t) 19:15, 6 April 2026 (UTC)

Hmmmm I forgot on "what we generally decline". I've been a participant of some elections AFD because newspapers.con isn't a thing outside the US and you can't find anything from the 1990s or earlier.... but that's not usually what we mean when you say "what we generally decline", as we allow articles on specific elections on specific positions on specific places. Howard the Duck (talk) 19:21, 6 April 2026 (UTC)

If we are talking SNG space, you are basically saying that all national and provincial/state elections and presumed notable, any lower election would not be covered by that SNG but instead must meet the GNG with particular consideration for breadth of sources (not local ones only). So like this proposed SNG wouldn't cover the recent NYC election, but that had clear GNG coverage. Masem (t) 19:43, 6 April 2026 (UTC)

As written, this only encapsulates "National summaries of local elections of a country", not "all... provincial/state elections"; i.e. 2026 United Kingdom local elections, not 2024 London mayoral election. I assume that should be clear. Howard the Duck (talk) 19:56, 6 April 2026 (UTC)

You have it right that elections not explicitly named would need to meet the GNG and normally that wouldn't have to be said within an SNG, but here, we don't want editors to end up using a lot of local only sources to justify the local election as GNG notable, hence we should spell that out and add caution about source breadth. Masem (t) 20:03, 6 April 2026 (UTC)

There is no chance of having an SNG for elections. However, I think it is realistic to add them as an example in WP:EVENTOUTCOMES. Kelob2678 (talk) 20:10, 6 April 2026 (UTC)

I'd be in favor of this; it's just shocking people actually nominated that article to AFD. You won't see that happen on a set specific of countries, save for WP:CRYSTAL concerns. That may be the first AFD of its kind. Howard the Duck (talk) 22:05, 6 April 2026 (UTC)

The AFD has been closed already.

I think that what would help is specific examples to help people mentally anchor the description, like:

• Yes to 2025 New York City mayoral election, but no to 2025 Smallville mayoral election
• Yes to a comprehensive summary/list of dozens or hundreds of local elections in a large area, but no to a separate article on just one of them

WhatamIdoing (talk) 21:22, 6 April 2026 (UTC)

FWIW, I'm indifferent on specific local elections on specific places. I think these should not be covered by WP:SNG or even WP:EVENTOUTCOMES. New York vs. Smallville (Clark Kent would win all the time on both places) is sorta like rich country vs. Maldives... I don't think that's what WP:SNG endeavors to keep (or exclude). Howard the Duck (talk) 22:03, 6 April 2026 (UTC)

That's the point I was trying to make. If elections are put into an SNG in some form, we clearly want to have the SNG (the presumption of notability) apply to the major elections due to their scope, but that presumption doesn't apply to more local elections. But local elections still can be presumed notable if the GNG (significant coverage in multiple independent sources), with the cavaet that these should show coverage beyond just the local region.

If this all goes to OUTCOMES, I would still include that local elections can possibly met the GNG with the demonstration of significant coverage. Masem (t) 23:44, 6 April 2026 (UTC)

I don't think the intent of SNGs was "this class of articles are automatically notable, the others are not", just the first part. Like GEOLAND does not mean random villages are not notable because towns are, or NPOL does not mean congressmen are notable but mayors aren't. Howard the Duck (talk) 08:33, 7 April 2026 (UTC)

Its probably more significant if it ends up at OUTCOMES, but most SNGs do start that says, effectively, the GNG can be used for any article of that type, and then goes into the SNG criteria for the limited cases where we make presumed notability based on topic importance/merit (as with national elections). So yes, the SNGs should be clear the GNG is always a route, even if we don't spell out the cases that the SNG doesn't cover.

But that said, unique to local and regional elections, we need to caution that the GNG there must be met demonstrating non-local sources. This is similar to what NCORP does in explaining the stronger requirements for independent sourcing there. The main WP:N page does not go into whether sourcing must be beyond limited geographical regions (independence, yes, but that's not the same thing), so whereever this about elections ends up, it should state that the GNG is always a route as long as non-local coverage exists. Masem (t) 11:37, 7 April 2026 (UTC)

I have always interpreted the SNGs as:

These topics are presumed notable BECAUSE they are highly likely to have significant coverage (and thus pass GNG). However… that presumption could be wrong (ie coverage doesn’t in fact exist). Remember that there are exceptions to every rule, including GNG and SNGs.

So… Don’t nominate a topic that falls within an SNG for deletion UNLESS you can demonstrate that they are an exception (ie they don’t actually have significant coverage). Blueboar (talk) 12:17, 7 April 2026 (UTC)

To me most SNG (excluding NPROF) are saying that if this criteria is met, there is likely going to be sufficient coverage out there or in the future (as with elections) that a standalone article can reasonably be created and improved wiki fashion. If in the future that it can be demonstrated that presumption was wrong (with a good BEFORE check), then we can talk about deletion, but we want to encourage article creation for articles that we know from the past are nearly always going to have significant coverage. Masem (t) 12:49, 7 April 2026 (UTC)

IMO, in practice, larger scale elections (in US, statewide or US Senator or a the top spot in multi-million person city, and the equivalent in other countries) are of high impact / importance/ visibility and tend to get a finger on the scale towards inclusion because of that. Since there no official criteria for these qualities, one can say "suitable coverage probably exists" to reconcile that consideration with Wikipedia rules. Once you move into lower-than-that elections, only a small fraction have true GNG-coverage of the topic of the article. There's lots of coverage that really isn't that. Much of it is just stats, and then the Wiki editor turns stats into sentences and it has the appearance of GNG type content. The other common type is more in depth coverage of the candidates. This is coverage of the candidate not of the election. The latter is typically not used to generate a lot of prose in the election article, but is often considered to fulfill the GNG requirement. As a result we have a lot of election articles which are really "stats-only" where the prose is just the Wiki editor turning stats into prose. IMO somewhat tough wp:notability provisions and practices for those lower level election articles is fine. Sincerely, North8000 (talk) 18:43, 7 April 2026 (UTC)

I've seen articles on specific local elections on specific places with references made up of just database entries with absolutely no prose. That's not good enough, and that's not what I am proposing here. Now, there may be articles like that about national-level elections, or local elections summarized nationally; those may be problematic. I've had issues with not just elections, but having articles deleted because the internet did not archive newspapers from outside the Western world; I've said elsewhere that "Wikipedia is not the place for that" until someone scans those newspapers from a dusty library pit, but elections (and sports seasons/tournaments) are another matter since "if we can create an article about a 2026 election/season, then there is reaosn to believe there was WP:GNG coverage about the equivalent 1926 election/season about this but we just don't know where to find it. Howard the Duck (talk) 21:06, 7 April 2026 (UTC)

Taking the article in your OP as an example, I think that the Wikipedia norm would be "keep". And of the four main reasons, #1 official, #2 isn't but a good idea, #3 isn't but can be used to strengthen a #2 argument, #4 has no formal basis. #1 is the national level proposed change......suitable coverage almost certainly exists. #2 is that much of it is a summary type article .....basically an "up merge" of hundreds of elections that shouldn't have an individual article. #3 is that the individual elections have lots of coverage. This is not valid to satisfy GNG, (since it isn't about the topic of the article) but it sort of contributes to #2. #4 is that it is a well-developed encyclopedic article, and not a "just stats turned into prose" article. Again, there is no official criteria for #4. Sincerely, North8000 (talk) 17:16, 11 April 2026 (UTC)

• An issue I would also link to bring to the table is that in the UK, we have pages for local elections listing results, which you could say are forks of the local authorities, but then editors are then creating ward pages and repeating the same lists! Chalkwell (Southend-on-Sea ward) and Westborough (ward) are perfect examples, which we already have pages for each election via Southend-on-Sea City Council elections. Chalkwell already exists as a settlement article on here, while Westborough is just a ward within Westcliff-on-Sea. Its madness. An SNG stating what we can have is probably a good idea, especially if its as tight as NPROF.Davidstewartharvey (talk) 11:10, 12 April 2026 (UTC)

Hi all. There is a debate open at User talk:~2026-19304-32#April 2026 regarding verifiability of publications and there usage on Wikipedia on the article for any Pakistani film. Anyone is welcome to participate, thank you! M. Billoo 03:27, 11 April 2026 (UTC)

I am interested in opinions about the newish page Melanie Walker which I ran into as a routine part of new page patrol; it was created over a redirect a month ago. While the page has been tagged for academic notability, I think she is far from a pass of WP:NPROF and that is not the issue I am asking about. To me her notability depends upon the section Relationship with Jeffrey Epstein. I am not sure how best to handle that, and I dont want to jump to an AfD to judge notability in this case,

Courtesy ping of Hispalois and Mariamnei. Ldm1954 (talk) 10:01, 11 April 2026 (UTC)

I nominated the page for deletion. I couldn't find any evidence she is notable Katzrockso (talk) 11:16, 11 April 2026 (UTC)

Isn't notability one of the most important factors when determining whether we should have an article on a topic? Why is this only a guideline? FaviFake (talk) 19:17, 17 April 2026 (UTC)

Why isn't Wikipedia:Five pillars a policy? Or even a guideline? WhatamIdoing (talk) 19:26, 17 April 2026 (UTC)

That's answered in the FAQ:

Is this page a policy or guideline, or the source for all policies and guidelines?

It is none of those. It is a non-binding description of some of the fundamental principles, begun by User:Neutrality in 2005 as a simple introduction for new users. For comparison, WP:NPOV, WP:NOT and WP:IAR were first written down on Wikipedia in 2001, and WP:NOR and WP:V were written in 2003.

FaviFake (talk) 19:30, 17 April 2026 (UTC)

And if we can get along with none of our principles being officially designated as a {{policy}}, then I suspect we can get along with this page being "just" a guideline. WhatamIdoing (talk) 19:37, 17 April 2026 (UTC)

This is a guideline for deciding whether a policy-compliant article can be written about a topic (following WP:V, WP:NPOV and WP:NOR). If one has already been written then there is no need for it. Phil Bridger (talk) 20:03, 17 April 2026 (UTC)

Here's an old discussion from 2017: Wikipedia talk:Notability/Archive 60#Promote to policy? I'll try to find more recent discussion than this if needed. Also, here's the article about itself "notability in the English Wikipedia". George Ho (talk) 20:18, 17 April 2026 (UTC)

Because people can't agree on what's notable. ~WikiOriginal-9~ (talk) 20:21, 17 April 2026 (UTC)

Or that being notable is necessary for having a separate page. WhatamIdoing (talk) 20:41, 17 April 2026 (UTC)

This reminds me of the question of why WP:Consensus is not Policy, eg Wikipedia talk:Consensus/Archive 3#Policy, in that it is a good question asked by serious editors.

Notability and Consensus are both interesting cases of where the English word as defined by dictionaries is different to how it is used by encultured Wikipedians.

WP:Notability is pseudo-policy, by virtue of being explicitly cited by WP:Deletion policy, by WP:DEL#REASON #8 (it used to be #6).

A commonly accepted answer is that Wikipedia-notability tolerates more exception, more consistent with

the guideline caveat

“Editors should generally follow it, though exceptions may apply”

than the policy caveat

“a widely accepted standard that all editors should normally follow”.

- SmokeyJoe (talk) 21:54, 17 April 2026 (UTC)

On taggery, I think that

• Wikipedia:Notability#General notability guideline should be a {{policy}} section,
• Wikipedia:Notability#Why we have these requirements is {{essay}}
• the rest is {{guideline}}

- SmokeyJoe (talk) 22:08, 17 April 2026 (UTC)

GNG as a policy? What about WP:SUSTAINED saying that GNG might not suffice? George Ho (talk) 23:44, 17 April 2026 (UTC)

Not every standalone page is required to meet WP:N, and/or WP:N is not the only decider when a standalone page is reasonable. eg NLIST is a prime reason that we don't use WP:N for this purpose. Add in that when notability is met is very subjective, and it makes for a poor policy page. Guideline, yes. Masem (t) 22:52, 17 April 2026 (UTC)

Wikipedia:Ignore all rules is an entirely subjective standard, and yet it is a policy page. WhatamIdoing (talk) 23:20, 17 April 2026 (UTC)

We have WP:GUIDES policy saying that exceptions can be made by using common sense, especially in certain circumstances. IAR is just... IAR, ya know, right? George Ho (talk) 23:47, 17 April 2026 (UTC)

Please see Wikipedia talk:Manual of Style/Visual arts § RfC: Various sections of MOS:VA. voorts (talk/contributions) 17:19, 18 April 2026 (UTC)

Why does an article primarily need secondary sources and not primary sources? is a primary source not more reliable than a secondary? REAL Mr. CRT (talk) 20:40, 20 April 2026 (UTC)

Because Wikipedia is an encyclopedia, not a publisher of original research. voorts (talk/contributions) 20:43, 20 April 2026 (UTC)

That's an argument for sources. That's not an argument in favor of sources that provide analysis and critical thinking. WhatamIdoing (talk) 20:45, 20 April 2026 (UTC)

This does not help, please specify reasoning. REAL Mr. CRT (talk) 20:46, 20 April 2026 (UTC)

Did you read the notability guideline? It answers your questions. voorts (talk/contributions) 20:49, 20 April 2026 (UTC)

Wikipedia is for things that other people (ideally experts) deem important enough to write about, not what people say about themselves or their own work. HJ Mitchell | Penny for your thoughts? 20:59, 20 April 2026 (UTC)

Yes, but Wikipedia:Secondary does not mean independent, and that's an argument for an independent source (=written by other people), rather than a secondary one (=analysis and critical thinking, even if it's personally written by the subject). WhatamIdoing (talk) 21:01, 20 April 2026 (UTC)

Neither primary nor secondary sources are more or less reliable than the other. You should, however, use them differently. A secondary source is good for "Wonderpam is effective at curing scaryitis". A primary source is good only for "This one study found that _____".

The "real" answer is that 'way back in the day, when Wikipedia was new, editors didn't distinguish between WP:SECONDARY and WP:INDEPENDENT sources. For a long while, "secondary" was defined in policy as "second-hand". In this model, if you were involved in a car wreck, and you tell the reporter what you think happened, and the reporter quotes you word-for-word, then the newspaper the next morning was called a "secondary" source, because there was the primary source (you) and the reporter is the "second", and that logically makes the reporter "secondary", right? Only it doesn't.

You may have noticed that the text of the GNG ("A topic is presumed to be suitable for a stand-alone article or list when it has received significant coverage in reliable sources that are independent of the subject") doesn't actually mention secondary sources. That's because when the wording was settled on, we treated secondary and independent as equivalent.

Consequently, the "real" answer is: A historical accident.

But: The reason that we keep it that way is because we find significant value in having a source that provides analysis, evaluation, interpretation, or synthesis of the facts, evidence, concepts, and ideas. It's okay to say that "_____ is a television show", or even "_____ is a television show, and here's a list of episodes and actors' names". All of that could be done with just a primary source. However, it's much better to say that _____ is the first television show to do something, the best-known television show for something, the television show that changed the public's view of something, the television show that was a critical moment in the early career of the now-famous director, etc. All of that kind of content is secondary source material. It helps us go beyond the bare facts towards placing the subject in context and explaining its significance. WhatamIdoing (talk) 20:55, 20 April 2026 (UTC)

Sure, reliable sources of every type, like primary and secondary, have existed for millennia, if not centuries. However, as I have realized, primarily relying on secondary sources has been highly recommended (if not required) not to just limit what the project can handle but to also convince others, readers and editors alike, into determining which topics belong in this project and which don't. Too bad certain people would like to challenge the status quo of "notability", like proposing the rename of "notability" multiple times.

Also, use of primary sources have not been without caution or care anticipated. If reliant on primary sources excessively, an article would be seen as some sort of self-promotional piece ever made or some sort of mere summary of primary sources, like a TV episode or a docuseries.

For further note, there's the policy counterpart WP:Verifiability#Notability, which tells you to seek an independent reliable source verifying a topic's notability, but then it implicitly lets you create an article about a subject whose notability has been already predetermined by (presumed/assumed existence of) such source. Sorry if I make my reply a lot longer, but I couldn't find a concise way to briefly explain all this. George Ho (talk) 22:52, 20 April 2026 (UTC)

The OP's question is about secondary (e.g., analytical sources), not about Wikipedia:Independent sources (=what WP:V mentions as a key point in a simple summary of this guideline). WhatamIdoing (talk) 23:31, 20 April 2026 (UTC)

Didn't mean to go off-topic. I just wanted to make context.

Oh... now I re-read what secondary sources are (article link). Okay, admittedly, secondary sources rely on primary sources for their material, making analytic or evaluative claims about them. Now reading Wikipedia:Party and person essay.... George Ho (talk) 23:58, 20 April 2026 (UTC)

Because "secondary" is a somewhat poor heuristic for "more likely to be reliable, independent of the subject, and to contain the sort of analysis we need to avoid WP:OR while covering the topic in an encyclopedic manner". Over the years, editors found it was a whole lot easier to just say "secondary good, primary bad" than to actually discuss whether a source is reliable, independent, and so on. So they shoved the word "secondary" into more and more policies and guidelines, and developed their own idiosyncrasies to deal with the ambiguities in the standard academic definition. Anomie⚔ 23:41, 20 April 2026 (UTC)

Secondary sources for WP's purpose provide analysis and other transformation of other information. We use the factor that if there is significant coverage that comes from such analysis from independent sources, that is showing the topic has been worthy of note in reliable sources, and thus reasonable to have a standalone article.

We do not disallow primary sources, but primary sources alone only show that the topic exists, not that it has been worthy of note; primary sources can be used in moderation with secondary sources. Similarly, we want independent coverage to make sure we're not getting excessive promotional material; we don't disallow dependent coverage but notability needs to be shown from independent sources. Masem (t) 00:02, 21 April 2026 (UTC)

• Another answer is, primary sources may be fine for a bunch of facts but the purpose of secondary sources which is also our purpose is to put facts together in a coherent, useful way, and if we put them together the way secondary and tertiary sources do, we avoid being the new arbiter of topics. Alanscottwalker (talk) 12:29, 21 April 2026 (UTC)

  +1. One of the problems with using primary sources is that we tend to try to analyze their meaning ourselves. Oh, Artemisia Gentileschi was raped as a teenager by an older respected man and an accomplice, also an older respected man...that explains why she was so focused on Susanna and the Elders. If we'd used the primary source (court records) and the fact she painted Susanna at least four times to make that connection: original research from a primary source. Fortunately many academics have made that connection, providing us with secondary sources here. Valereee (talk) 12:58, 21 April 2026 (UTC)

  +1. SmokeyJoe (talk) 02:00, 27 April 2026 (UTC)

Intro: WP:CHURNALISM

There was a bit of an unfocused discussion on "churnalism" last year at Wikipedia talk:Notability/Archive 84#On notability and scientific churnalism which resulted in the following being boldly added by Ldm1954 to WP:SUSTAINED after only brief discussion between two editors actually focused on that exact guideline:

Similarly, reproductions or close paraphrasing of press releases from, for instance, companies or other organizations (aka churnalism) does not count as sustained coverage. This is particularly the case because normally these are not secondary or independent sources.

Which was promptly shortened by WhatamIdoing to:

Similarly, reproductions or close paraphrasing of press releases (aka churnalism) does not count as sustained coverage.

with WhatamIdoing explaining in the disucssion:

I've shortened the addition, because half of what you wrote wasn't true, especially since press releases can be Wikipedia:Independent sources, and their problem isn't the lack of independence or being primary; the main problem with a press release is that it's self-published – a problem that completely goes away when The Daily News independently decides to write an article about the subject of the press release, no matter how badly the article is written (including, but not limited to, copying large chunks out of a press release).

I am not very knowledgeable of the scientific side of "churnalism" discussed there, but there is an ongoing discussion touching on entertainment "churnalism" at Wikipedia talk:Notability (films)#Unreleased Films and Notability (again). As I stated there, there is no official policy on suspected "churnalism" in perfectly reliable sources, and the term itself can be defined in a way that it can be stretched to apply to almost all news reporting, making it unfalsifiable. Press releases are already discussed as being non-independent of the subject at the WP:GNG. So articles built upon obvious copies of press releases should fail the GNG. But that should be identifiable to the extent that it is demonstrable, i.e. logically proved. And in almost all cases I have seen (in Western media), I don't believe we have the evidence required to make that kind of judgement. In general, I am not satisfied with this "churnalism" neologism and its application. Secondary sources are built from primary sources, so if a journalist honestly believes that what they have been told was the truth, and his or her belief was reasonable, he or she is entitled to use it. "I believe what a press release said" is an independent opinion, whether you like it or not. If you are going to argue that coverage fails independence, then that claim should rest on demonstrable analysis of the editorial process in that specific instance. But it is not Wikipedia's place to speculate to that end. Remember: The goal is to reflect established views of a topic as far as we can determine them. And a topic is "notable" in Wikipedia terms only if the outside world has already "taken notice of it". As Erik wrote in the aforementioned discussion: We should not shun reliable sources' reporting just because there is some apparent thinking that such sources are being manipulated by publicists. Ultimately, whether we like it or not, Wikipedia is simply a reflection of what reliable sources tell us. A compiler of information that is presented in proportion to the coverage found in reliable sources, and not a publisher of original thought. It is rather arbitrary to selectively disqualify information that mainstream, trustworthy sources have made the editorial decision to publish as their own material. So I'm not very satisfied with this "churnalism" addition to WP:SUSTAINED. And I believe WhatamIdoing wasn't either, as they concluded their posting in the 2025 discussion here with the following statement:

On the more general question, I am doubtful that we need this in the guideline, and I predict that including it will cause problems. Specifically, just like we occasionally have editors exceed the intent of various rules – a frustrated editor declaring that they're hereby WP:CHALLENGING every single uncited sentence in every single article, or POV pushers declaring that all sources supporting the other POV are unreliable, or deletionists declaring that no _____ (fill in the blank: current events, influences, children, websites, small businesses, etc.) is ever notable – I expect we will see an increase in the number of AFD noms and delete !votes that baselessly assert that every source is mere churnalism and therefore banned by WP:N. (How does the nom know it's churnalism, you might ask? Well, first of all, no true source would ever bother writing about that kind of unimportant subject, second, everyone knows that true sources only write criticisms instead of the positive facts related in that source, and third, it's obviously UNCIVIL of anyone to question my omniscience. I just know, okay? And you will never, ever convince me that I'm wrong.)

What do we think? Οἶδα (talk) 05:53, 21 April 2026 (UTC)

I think the discussion you refer to was somewhat broad, and there was a general concensus in favor of it wrt WP:SUSTAINED. While WhatamIdoing makes a good point, the discussion was not (in my mind) on the general issue of press releases. It relates to whether press releases by companies, universities or academics about their recently published work which are then reproduced with minimal edits are a basis for a claim of sustained coverage. (If you want I can give examples in science of churnalism versus genuine press articles with independent investigation.) The discussion you reference Wikipedia talk:Notability (films)#Unreleased Films and Notability (again) is not about sustained, so I do not understand the relevance. What are you actually suggesting? Ldm1954 (talk) 12:19, 21 April 2026 (UTC)

Giving an example, with apologies as it relates to my work. I view this as churnalism, whereas this is not, but the similar article in Scientific American is because it is a reproduction, even though Scientific American is a respectable source.. Ldm1954 (talk) 12:37, 21 April 2026 (UTC)

A reprint is not what Wikipedia editors traditionally call churnalism. That's covered under Wikipedia:Notability#cite note-4, in which the same document being printed in multiple places, even if it has minor changes, is still just one source and therefore not "multiple".

The thing that Wikipedia editors usually call churnalism in this context is:

• Big Business, Inc. issues a press release about itself.
• Jo Journalist copies a large chunk of it, changes a couple of words, and puts "by Jo Journalist" at the top of the article.

In this case, we question whether the news outlet has done what we need from them, which is to exercise editorial judgment about whether the subject is worth giving attention to.

The other thing that a few editors (most memorably one now banned by ArbCom for his anti-COI work) call churnalism in the context of notability is:

• Low-Cost Wire Service puts together some information on non-time-sensitive subjects, which they sell to ordinary news sources in multiple formats. The same bit of reporting gets produced as a 200-word version of an article, a 300-word version, and a 600-word version; a 30-second news-style video, a 60-second video, and a 90-second video saying the same things; and a photo essay.

This isn't really churnalism per se. It's more like a Least publishable unit model, except that it provides all the publications at the same time. When you can determine that all of these "different" sources have the same origin (i.e., the same journalists, all working for Low-Cost Wire Service, Inc.), they should be counted as one source. WhatamIdoing (talk) 17:30, 21 April 2026 (UTC)

We are not disagreeing about churnalism in the context of science publications, they and reproductions do not have to be mutually exclusive. Ldm1954 (talk) 17:43, 21 April 2026 (UTC)

I do not agree with your statement that "the similar article in Scientific American is [churnalism] because it is a reproduction". I believe that the article in Scientific American is a reproduction and not churnalism. WhatamIdoing (talk) 19:33, 21 April 2026 (UTC)

The discussion I referenced pertains to the notability guideline for films and is revisiting an earlier discussion on the same talk page here, wherein this "churnalism" addition to WP:SUSATINED was referenced. Among multiple points raised there, the topic of "churnalism" was brought up many times to dismiss film-related announcement articles from perfectly reliable publications such as Variety, Deadline, etc, which some have argued "draw pretty heavily from a press release and promotional blurbs". You will find no disagreement from me on that. Though, as one user wrote:

The reliable sources, manipulated or not, are churning out content. That is their job. They take announcements, rewrite them somewhat for clicks, then post them without any independent coverage. So, it is not that there is coverage in reliable sources...it is that the coverage is basically from the studios. Not all articles published in a reliable source are reliable. I mean, if you read some of the sources I cited in the deletion discussions linked above, you will see there dozens of articles in reliable sources, but if you look at those sources you will see that the majority are published at the same time based on churnalism. That is not okay.

As I was saying above, where I take issue with this is: "without any independent coverage". I am doubtful the extent to which we can make such unfalsifiable "failing independence" accusations against reliable publications for the purposes of selectively disqualifying their coverage from being counted as WP:SIGCOV from secondary sources. My question is whether or not the GNG doesn't already effectively communicate about press releases failing independence without needing this mention/definition of the problematic "churnalism" at SUSTAINED, at the risk of its miscitation. Οἶδα (talk) 05:09, 23 April 2026 (UTC)

I think that it's helpful to understand why WP:N (generally) and the GNG (specifically) want "independent sources". The goal, for notability purposes, is not to prove that the publication was uninfluenced by anyone or anything. The goal is for the publication to function as a filter. To explain what I mean, let's pretend for a moment that every single news or magazine article ever had its origin in a press release or similar document. We would still want to know whether The Daily News or Monthly Magazine picked Subject A over other subjects. Even if we can prove that every single fact came straight off a page on the corporate website called "Press releases", the fact that the 100% independent editorial process at that 100% independent publication 100% independently chose to copy some facts from Subject A's press release instead of doing the same for Subjects B, C, and D is what Wikipedia's notability process actually cares about. Notability is not about the purity of the information. Notability is not about fact-checking. Notability is about whether there was voluntary attention from the world at large, and not just from people being paid to pay attention. WhatamIdoing (talk) 00:10, 24 April 2026 (UTC)

the fact that [...] chose to copy some facts from Subject A's press release instead of doing the same for Subjects B, C, and D

I would like to know that too. I understand that independent publications act as a filter. But I think we're overstating what editors can actually infer. Because unless obvious enough to the point of being demonstrable, there is no actual way for us to determine that a publication chose to do something for Subject A over B, C, and D without guesswork. Presumably notability should not be grounded in assumptions about voluntary attention, correct? I could just as equally claim that a seemingly ideal article reflects involuntary attention. My source? Why, it's obvious and it would be to you as well had you possessed my omniscience. I just know, okay? Let me take the reins here. The article must be deleted. It is the only true conclusion. And you will never, ever prove otherwise. Οἶδα (talk) 08:10, 25 April 2026 (UTC)

I think we can and should make some common-sense assumptions. We can, for example, assume that when there are thousands of restaurants in a big city, and that the daily newspaper for that city runs very few articles about local restaurants per day, then the newspaper is picking and choosing which restaurants it gives attention to.

Similarly, I think we can safely assume that when there are just two restaurants in a small town, and the small-town weekly newspaper runs an article every time one of those has a change of menu, decor, ownership, etc., that the newspaper is not picking and choosing which restaurants it gives attention to. It is instead indiscriminately trying to cover everything that happens, because not much happens in that town.

The goal of WP:AUD is to 'level the playing field' between the small-town restaurant, which gets attention from their local newspaper merely for existing, and the big city restaurant, which will be largely ignored by their local newspaper.

I think that many editors are unfamiliar with the ecosystem of small-town newspapers, but even into the early years of the current century, there were small-town newspapers writing about whose grandchildren came to visit last week. Every traveling preacher or small-time entertainer making the round of rural schools and libraries would be "notable" if we treated local newspapers as equivalent to the major dailies. WhatamIdoing (talk) 18:44, 25 April 2026 (UTC)

@Οἶδα, I agree with you that unfalsifiable accusations are a significant problem. WhatamIdoing (talk) 17:31, 21 April 2026 (UTC)

If opinions of a few editors exemplify the consensus, then perhaps I can stand corrected about wide consensus being necessary. Indeed, "churnalism" has been frowned upon, so at least something was already done about it. Right? George Ho (talk) 05:49, 23 April 2026 (UTC)

This is a hard problem to tackle. But I sympathize with efforts to address it at its root, and not on a source-by-source basis. There are many problems with churnalism. For one, a lot of churnalism is just repeating press releases or things they saw on social media, and that makes it questionably reliable. Shooterwalker (talk) 19:47, 23 April 2026 (UTC)

The current sentence in SUSTAINED says: "Similarly, reproductions or close paraphrasing of press releases (aka churnalism) does not count as sustained coverage."

Maybe we should split this up:

• The sentence "Brief bursts of news coverage may not sufficiently demonstrate notability" in SUSTAINED could get a copy of footnote Wikipedia:Notability#cite note-4, which says "Lack of multiple sources suggests that the topic may be more suitable for inclusion in an article on a broader topic. It is common for multiple newspapers or journals to publish the same story, sometimes with minor alterations or different headlines, but one story does not constitute multiple works. Several journals simultaneously publishing different articles does not always constitute multiple works, especially when the authors are relying on the same sources, and merely restating the same information. Similarly, a series of publications by the same author or in the same periodical is normally counted as one source."
• Wikipedia:Notability#cite note-5, which says "Works produced by the subject, or those with a strong connection to them, are not evidence of notability. See also: Wikipedia:Verifiability § Questionable sources for handling of such situations." could have the words about "reproductions or close paraphrasing of press releases (aka churnalism)" added.

WhatamIdoing (talk) 00:16, 24 April 2026 (UTC)

A source that is taking a press release or equivalent material, and simply rewording or restating it without any significant additional transformation is a primary source, and thus should immediately be excluded from consideration for notability. That itself covers the basics of churnalism. This also couples when you get multiple sources doing that reiteration over a short period (the brief burst of coverage). Masem (t) 03:52, 25 April 2026 (UTC)

As in WP:PRIMARYNEWS? So an article is restating a press release and we know that because...? And the publications are doing no independent research into the factual basis of the material they are publishing and we know that because...? Either way, I agree that articles containing announcements of future events and providing no in-depth aka non-trival interpretive/filtering/original material are routine and do not establish notability warranting a standalone article. But what if, as I mentioned at NFILM, a publication publishes an exclusive story and 10 other newspapers pick it up, citing their exclusive. And it isn't obvious that the original article is restating a press release, even though it most likely is. Is that churnalism? I understand that it may not establish notability for a future event, even if such coverage were to happen 10 times over the span of a year. But is it actually "churnalism" in any provable way? It would moreso appear the media made the editorial decision to cover something, and we're dismissing it on nothing more than a series of unfalsifiable assumptions. What if CBS News posts a bombshell story and every line is actually just information from their confidential source, information the publication merely says it has "learned"? When do the sources that the media use as the basis of their material become equivalent to a press release? Everything the media decides to publish can be argued to be promotional ultimately. And likewise unfiltered, or not filtered enough. Οἶδα (talk) 06:47, 26 April 2026 (UTC)

So an article is restating a press release and we know that because... in the ideal case, we also have a copy of the press release in hand, and we can compare the text of the "news article" with the text of the press release. When that's not possible, there's a significant chance that Wikipedians are guessing based on their biases.

I have seen comments from editors about publications are doing no independent research into the factual basis that seem unreasonable to me. For example, at the List of megachurches in the United States, a few editors occasionally say that they want independent audits of claimed attendance. These Wikipedians think it's not enough for a professional journalist to call the church, ask for the number, decide the answer they received is trustworthy, and publish that number; they want someone in the parking lot, counting the number of people arriving, and preferably then running an article complaining that their count was lower than the church's reported annual average. But a newspaper reporter isn't a private investigator, so staking out a parking lot to check attendance numbers is not how they work. WhatamIdoing (talk) 23:52, 26 April 2026 (UTC)

Seemingly independent coverage can be covertly reproduced supplied promotional material. It is not “bias” to look for and to recognise the subtle signals of insider influence, marketing spin, or undisclosed conflicts of interest.

In an idealised scenario, where the press briefing sheet has entered the public domain, something has gone wrong.

In the normal case, where the press briefing sheet is never to be seen, but is 90% word for word in the publication, magical powers are not needed to spot the duck.

”New cafe has the warmth of your grandparents Christmas dinner”

”New cafe is the best new place to be seen”

”New cafe serves only the best rib eye, aged for 12 weeks, and is to die for”

“New cafe has been the place to be for the last five weeks”

”New cafe has been booked out every weekend for the last five weeks”.

The first may reflect a personal or highly subjective perspective. The second is unverifiable. The third and fifth suggest access to information not typically available without internal knowledge. The fourth may be plausible, but warrants further context.

Unsourced facts, facts unverifiable without going to the source, are an indicator of possible undisclosed non-independence. Do the facts sound like the trim pickings from a hypothetical audit, or are they more likely the work of the church spokesperson’s office assistant?

Newspapers don’t anymore usually pay reporting staff enough to do serious investigation. One can make a good relationship with local newspaper staff by supplying them with publishable words. It doesn’t take a thief to know a thief, the tells can be taught. It is reasonable to approach commercially beneficial promotional content with caution.

- SmokeyJoe (talk) 02:40, 27 April 2026 (UTC)

I can agree that It is not “bias” to look for and to recognise the subtle signals.

However, it is bias to look for and recognize subtle signals only in certain subjects. For example: Everything in a luxury goods magazine is going to be "positive". And if a Wikipedia editor says "This is positive, so it's obviously written by the company and probably illegal paid advertising", that's the Wikipedia editor's own personal bias speaking, and not anything grounded in facts. WhatamIdoing (talk) 02:47, 27 April 2026 (UTC)

A luxury goods magazine is the sort of thing I mean to talk about. It is the same as a trade magazine and the fine dining section of a newspaper. If you want coverage, you make it easy on the “journalist”. You write them briefing material.

This “bias” I call “skepticism”. It is a good bias for Wikipedia. Promoters want their name in Wikipedia.

It’s important to analyse facts. “This material is all positive” is a subjective fact. “Obviously” is not a fact but bluster. “Probably” in “probably illegal paid advertising” might be unjustified or untrue. A lot of covert non-independent promotion is not illegal. SmokeyJoe (talk) 02:54, 27 April 2026 (UTC)

If you want coverage in a disreputable source, you make it easy on the journalist. However, if you want coverage in a reliable source, then you cooperate with them to the extent that you can, and then realize you can't control them. Whole categories of publications should not be condemned en masse just because you feel "skeptical" about (aka "biased against") subjects such as "luxury goods" and "fine dining". WhatamIdoing (talk) 03:00, 27 April 2026 (UTC)

We live in an ocean of disreputable sources.

I do condemn categories of publication, according to the WP:RSPSOURCES table. I don’t even read the red ones. For others, it’s case by case on the content of the article. I do tend to trust the green ones. I hope that whoever maintains that table does a good job. SmokeyJoe (talk) 03:07, 27 April 2026 (UTC)

"The red ones" are individual, named publications. They are not "categories of publications", like "luxury goods magazines". WhatamIdoing (talk) 03:14, 27 April 2026 (UTC)

Certain subjects? This is why we have SNGs. Covert non-independence is to be more expected for topics connected to promotion. For commercial promotion, expect staff assistants. For non-commercial promotion, expect a different style. Yes, this will lead to bias. Name the bias and try to identify it. SmokeyJoe (talk) 03:01, 27 April 2026 (UTC)

Allegations of "covert non-independence" is not something that editors should just make up out of their own personal "skepticism". WhatamIdoing (talk) 03:03, 27 April 2026 (UTC)

Agreed. They should name what they consider the objective evidence, such as inclusions of facts that could only have come from inside the organisation. Others may then disagree objectively.

Personal skepticism bias? I am more skeptical of claims of significant on CORP topics than I am for ancient history or the natural sciences. Is that a problem? SmokeyJoe (talk) 03:10, 27 April 2026 (UTC)

Why do you think that "facts that could only have come from inside the organisation" is evidence of "covert non-independence" and not evidence of "the journalist interviewing knowledgeable sources"?

If a source writes that Microsoft has more than 200,000 employees, do you think that is evidence that the journalist was secretly manipulated by Microsoft's publicity department into publishing their pre-written content, or do you think that is evidence that the journalist asked a question during an ordinary interview and received an ordinary answer? WhatamIdoing (talk) 03:20, 27 April 2026 (UTC)

I don’t immediately think that Microsoft’s employee count is intended as promotion to be included in press coverage. That fact would not draw my skepticism.

If a source wrote “Copilot does [these many impressive things]”, I would be skeptical of the true author of the information. For Wikipedia content, I would expect each thing to be verifiable to an independent source. If the content of the source needs to be stripped as content in Wikipedia, I would not count it towards any WP:SIGCOV measure of that source, but I think we would appreciate a better label for what makes it not a good source. The best I think we have now is “non-independent”, for something that we think is Microsoft-supplied puffery statements. But first, I would check the journalist, because if they were already recognised as an independent software expert, it would lean to giving it a pass. (I’m imaging here that we are talking about the dubious two best GNG sources in an AfD, which would not be applicable for Microsoft, but maybe a new company). SmokeyJoe (talk) 03:54, 27 April 2026 (UTC)

I think you're letting personal biases affect your judgment of sources. If you're looking at an ordinary daily newspaper, you say that a headline like "Microsoft now region's largest employer" doesn't strike you as promotional, so you wouldn't expect that to be secretly pre-written by the company's PR department, but if the next article on the same page has a headline of "Copilot saves man's life by recognizing serious symptoms", then you suspect that of being "Microsoft-supplied puffery statements".

I suggest that the reality is probably the other way around: The "largest employer" is an annual set piece written entirely from pre-packaged information supplied to the paper (usually by a business consortium or a government report, rather than each of the individual organizations). None of the facts are discovered or otherwise elicited by the publication: you just read the report, assume everything is correct as supplied, copy the key figures, and nobody cares how little effort you put into it (except the editor, who would like you to do it quickly and move on to something else). The "software saved someone's life" pieces usually arise organically on social media (example), and a basic amount of fact-checking (e.g., reading the social media thread) leads to more information (e.g., someone saying that the same thing happened to their cousin, or that they were there when it happened).

I think we need to evaluate the source for itself, and not based on whether the content of an individual article "feels promotional" to individual Wikipedia editors. WhatamIdoing (talk) 18:34, 28 April 2026 (UTC)

It’s hard to let go of personal biases. Having seen how briefing papers to journalists work, I’m suspicious of anything that looks like it.

”largest employer” is not puffery because it is verifiable. “Great employer” is puffery.

"Copilot saves man's life by recognizing serious symptoms" sounds impressive, but I need to see verification, and to not discover after deep digging to find that the story is about a hypothetical. Anything verifiable from outside Microsoft would probably satisfy me that it’s a good story for meeting the GNG.

I agree with evaluating the source for itself, but I think I am more than you interested to read into the text for subtle signs, and to suspect an undisclosed press briefing. SmokeyJoe (talk) 23:00, 28 April 2026 (UTC)

"Great employer" is verifiable if a reliable source publishes it, and "Best employer" is a staple of surveys in multiple independent, reliable sources (e.g., "JPMorganChase ranked as Singapore’s best employer"). What I think you mean is that "Largest employer" is an objective fact, rather than an opinion. People might reasonably disagree about the details (e.g., is 100 full-time employees more or less than 110 half-time employees? How do you count seasonal employees?), but they will agree on the basic mathematical aspect of it. People would probably disagree on what makes an employer be "great" or "the best" (e.g., whether "best" is different from "highest paying").

I think that being suspicious of secret communications based on "subtle signs" results in an impermissible (potentially even NPOV-violating) level of bias. As editors, we have to take reliable sources at face value. We shouldn't be rejecting sources because of our personal guesses. We definitely don't want to end up in a situation in which editors are saying things like "This source is so anti-Microsoft that I just magically know that it's the result of sloppy work by a lazy reporter who secretly copy/pasted a bunch of stuff from Apple's press briefing materials". Or even "Everything in Category:Lifestyle magazines is unreliable because luxury goods magazines are too puffy to be real reporting". WhatamIdoing (talk) 00:02, 29 April 2026 (UTC)

“Great employer”, no qualification, is an unverifiable claim, district from a fact that someone made the claim.

Employee count, or Full-time equivalent, year-sum, whatever.

I think Wikipedia ignoring the fact that corporations pay to get coverage is foolish. Not everyone needs to worry about it, as long as some do, it will probably work out. Crowdsourcing Wikipedia content, and deletion decisons, works.

We don’t have to take anything at face value. Everything can be questioned. For efficiency purposes, I take WP:GREL sources at face value, I reject WP:RSPSS deprecated and blacklisted source without looking. All others, included sources not listed in the table, where the source is offered or challenged as a GNG source, and the topic is commercial, I adopt a skeptical mindset with regard to independence. Others might not. You don’t have to. But I strongly reject the notion that I am not allowed to. I do not reject because of personal guesses.

Microsoft is a really bad hypothetical example. This sort of analysis, studying the source text for writing style and likelihood of the author independently knowing, or being able to find out, what they are reporting, is required for borderline notable CORP topics. Winston Weinberg is a very interesting case that I am studying. Many interviews, all of highly dubious independence. SmokeyJoe (talk) 10:34, 29 April 2026 (UTC)

And once again, you've resorted to accusing "corporations" and reliable sources of engaging in illegal activities: corporations pay to get coverage. Paying for undisclosed "coverage" is illegal in the developed world.

Remember when you went through a restaurant review in a luxury travel magazine, and declared the whole thing to be written by the restaurant, because it was too puffy for your taste, except that it turned out this was the house style, and everything that author wrote had the same style? We agreed at the time that writing positive content was not a reason to suspect non-independence. Can I remind you of this exchange?

I don't think that we should be judging the independence of sources according to the POV the source expresses. WhatamIdoing (talk) 15:17, 25 May 2022 (UTC)

:Definitely agree. I’ve been talking about bias and POV as confounding attempts to infer whether the source is independent. I had been straying into thinking that pro-bias or pro-POV is an indicator of non-independence, but I am now convinced that this is a poor method, both for it being subjective, and because there can be many other reasons for a pro-bias or pro-POV. Bias and POV are not reasons to exclude a source from meeting the GNG.

:I am now inclined, in close call cases, to look at whether the source contextualises (with which you agree), and whether the source is distant-perspective (with which I think you have not agreed, and for which I am not aware of any written guidelines or essays). SmokeyJoe (talk) 03:34, 27 May 2022 (UTC)

And yet you're here again, making the same claim that your subjective impression of the biases in a review is supposed to be evidence that the source is not independent. WhatamIdoing (talk) 19:44, 29 April 2026 (UTC)

I am not making the accusation, instead, you infer it.

I’ve been thinking on this ever since you first did this. I think that you are trying to lead me down a dark path of irrelevance. I think you appear to be assuming that illegal things don’t happen. I think that what I imply is not illegal. A person or company is allowed to have staff paid to support public relations, or communications, or education. These paid staff are allowed to write briefing notes. These briefing notes are allowed to be in the style of newspaper articles. The journalist is allowed to read the briefing notes, and write a story in their name that bears strong resemblance to the briefing notes. This may be described with criticism, but I do not believe it is illegal. Maybe there is copyright practice infringement, claiming something is yours when another could prove it is theirs? Copyright infringement happens frequently.

I have read that in India, a developed nation, with nuclear power and a space program, it is commonplace for positive newspaper stories to be paid, and I have never heard of a court conviction for doing it.

For Nigeria, newspaper sources are treated harshly at AfC and AfD, despite their being no stories of publications by bribe.

I do not make legal calls on newspaper practices, and I think such calls are worthless to Wikipedia.

I do read text for tone, perspective, and the authors ability to know things that an independent distant perspective author would not know. I do not think that questions of legality are relevant.

I continue to believe that contextualisation, for for-profit companies, their products, CEOs and founders, is at the top of a list that should be considered, at AfC or AfD, for the sources proffered as the minimum GNG-meeting sources.

”subjective impression of the biases in a review is supposed to be evidence that the source is not independent” is a gross misreading. Tone, perspective, and insider facts can all be assessed objectively.

“Evidence that a source is not independent” is not the test. For a WP:CORP topic, I expect objective evidence (or conclusion from objective analysis) that the source is independent.

Independence of the source, it’s author, and the information it publishes, is not for being presumed in the absence of evidence either way. WP:PAID editing is rife. Companies advertise it. Some make a show of asserting that they comply with Wikipedia policy, but it would be foolish to assume this reflect universal PAID editor practice. These editors have incentive to push the boundaries to get their clients’ articles into mainspace. If editors don’t resist that, then Wikipedia will be polluted with promotion-biased information. AfC and NPR-ers are indeed suspicious of new CORP topics. I am working to describing objective methods that achieve what they are already doing.

If you think that you can infer that I write things that imply that illegal things happen in practice, why should you or I or anyone care, in respect to Wikipedia policy discussions? SmokeyJoe (talk) 13:38, 3 May 2026 (UTC)

In re the authors ability to know things that an independent distant perspective author would not know: Would an "independent distant perspective author" know enough to write an news article?

Would you describe Bob Woodward and Carl Bernstein as having that "independent distant perspective" when they broke the Watergate scandal? Was Daniel Ellsberg an "independent distant perspective author" with respect to the Pentagon Papers? How exactly does an "independent distant perspective author" find out new information, if not getting that information from someone – possibly an "inside source" – who has that information? WhatamIdoing (talk) 23:39, 3 May 2026 (UTC)

You want to explore my ideas as they apply to obviously notable topics? No. Wikipedia works fine for obviously notable topics, and for deletion of obviously non-notable topics. I’m trying to make progress where progress would be useful. The trouble is at the boundaries.

Improved methods for evaluating sources for meeting the GNG, are only needed for borderline notable topics. The {{Source assess table}} linked from WP:SIRS is one such attempt that seems to work for experienced Wikipedians, but not for newcomers. Can we explain better, even if it tends to being lies-to-children?

I have suggested Winston Weinberg, where if we can ignore the complicating sockpuppety, experienced Wikipedians at AfD rejected sources for reasons that I find hard to explain to a newcomer.

Your style seems to me to be accepting of the sources offered. Do you approve of the !votes in that AfD, their conclusion, their style? How would you assess the sources, whether in the article in the history behind the redirect, or those offered in the AfD?

- SmokeyJoe (talk) 02:02, 4 May 2026 (UTC)

I want to explore why:

• when an employee quietly gives documents to a newspaper reporter,
  • causing the reporter to know something that the average "independent distant perspective author" did not know, and
    • the reporter turns that into a newspaper article saying something negative about a politician,
      • Wikipedia editors would hail the newspaper as the epitome of an independent source, but
• when another employee quietly gives documents to a newspaper reporter,
  • causing the reporter to know something that the average "independent distant perspective author" did not know, and
    • the reporter turns that into a newspaper article saying something positive about a business,
    • you want me to believe that the newspaper is not an independent source.

Even when it's the same newspaper. WhatamIdoing (talk) 03:25, 4 May 2026 (UTC)

Please forget my description of a hypothetical.

Here is a real example. What is wrong with the sources? SmokeyJoe (talk) 06:24, 4 May 2026 (UTC)

User:WhatamIdoing, what is wrong with the sources? Is it what the editors at the AfD wrote? SmokeyJoe (talk) 23:19, 13 May 2026 (UTC)

"New cafe serves only the best rib eye, aged for 12 weeks" does not suggest access to information not typically available without internal knowledge. It suggests that the reporter read the menu.

But let's pretend that it suggests internal knowledge. Reporters are supposed to be getting that "internal knowledge" through a process known as interviewing relevant people, such as the chef, manager, and/or owners. So you complain that they're not doing their job, and your proof is that they learned things that they would learn if they'd done their job. WhatamIdoing (talk) 02:53, 27 April 2026 (UTC)

"New cafe serves only the best rib eye, aged for 12 weeks" does not suggest access to information not typically available without internal knowledge. It suggests that the reporter read the menu.

On that line of thought, it is plagiarism, failure to explicitly cite the menu. SmokeyJoe (talk) 02:55, 27 April 2026 (UTC)

No, it isn't. The standard for what needs to be explicitly cited depends on the genre.

Also: What do you expect them to do? Are they supposed to break into the refrigerator, check the packaging labels, intercept the invoices, and go investigate the meat packaging plant, to make sure that it really was "aged for 12 weeks", and not just 11.5? WhatamIdoing (talk) 03:02, 27 April 2026 (UTC)

Yes, it is. The menu in a commercial restaurant cannot be consider such an obviously reliable source that it doesn’t need explicit citing as the sole source of information.

What do I expect quality sources of Wikipedia to do? I expect them to take a distant perspective, such as noting other people’s apparent liking of the food.

If the aging of the beef is really a point of focus, I want the journalist to be talking to an independent chef who ate the meal or visited the kitchen. SmokeyJoe (talk) 03:14, 27 April 2026 (UTC)

I think you're applying a completely inappropriate standard. The community's standard is what ordinary reliable sources actually do. The community's standard is not rejecting ordinary journalism and only accepting sources if they follow an individual editor's paranoid standards, like sending in secret evaluators to determine whether the restaurant was lying. It's not possible, BTW, for "an independent chef" to determine whether a piece of meat was aged 10 vs 12 vs 14 weeks just by eating the meal. WhatamIdoing (talk) 03:22, 27 April 2026 (UTC)

I hold a different opinion then. For content that is announcing something commercial and new, I expect stronger evidence that others have noticed. This means, to some extent, more distance of perspective. I think that Wikipedia should not be publishing diners experiences in new restaurants, but can cover reliable sources discussing diners experiences. It is not sufficient that the source is reliable.

The point about the beef aging is not the number, my point was aimed at a source that is reproducing information supplied by the source. I had in mind an unpublished press briefing, which you reveal to be similar to content that could have been on a restaurant menu. I agree. A restaurant menu is part of the promotional material of a restaurant, although without any attempt to hide it. SmokeyJoe (talk) 03:41, 27 April 2026 (UTC)

Yes, I agree. I certainly would never claim that "churnalism" is not a thing. Journos converting press releases into articles is a thing. But for the purposes of assessing sources, the leap from "This article looks derivative of a press release" to "This article has had no original reporting or fact-checking" is perfectly unreasonable. Unless a publication has a history of doing just that (i.e. WP:NEWSORGINDIA), which would likely render it a non-RS in the first place, I don't believe we can selectively dismiss its publishing on such baseless assumptions that call for personal involvement in the editorial process and even, at times, for the operation of the author's mind. There exists as much proof for that latter claim that there does in the opposite direction. So much of journalism is ultimately just "reproducing information supplied by the source". So if that is what disqualifies articles suspected of copying press releases then so much of what is reported can be as equally and arbitrarily thrown into the bin. I understand that articles built on such "churnalism" do not state that the material therein has not been fact-checked by the publication, but publications mostly don't state that when they have done fact-checking either. Οἶδα (talk) 22:52, 7 May 2026 (UTC)

Besides what has been said above, if you work long enough in an area, it becomes to spot regurgitations of press releases even if you dont have the press release in front of you. I know for video games, when a new game is freshly announced, all the major sites will discuss the announcement without any real details, that's churnalism. At the same time, there are times that the publisher has treated the media ahead to sneak peaks, so that when the game is announced, these sites discuss their impression of the sneak peak. That fact, that the information is based on a sneak peak, is nearly always present, as to make those articles reasonably secondary and not regurgitation of a press release, so not churnalism. The film project has similar metrics, that for most films, there's lots of pre-production press (announcement of actors and main crew), but per NFILM, editors shouldn't create articles until production starts, as to avoid that pre-production churnalism. Masem (t) 03:20, 27 April 2026 (UTC)

I agree with this and I think it offers a (partial) solution to churnalism. Repeating press releases or social media reactions should not be the WP:SIGCOV basis for an article. (They might be used to fill in non-controversial gaps in coverage, if used judiciously and/or with clear attribution.) Shooterwalker (talk) 23:33, 26 April 2026 (UTC)

If it repeats a ton of information, it's SIGCOV. SIGCOV is about an amount of information (e.g., that could realistically be used to fill in gaps in coverage). SIGCOV is not about the quality of the information. What it might not be is independent, secondary, or even reliable. WhatamIdoing (talk) 23:54, 26 April 2026 (UTC)

Repetition of information, without another layer of commentary, without even acknowledging the repetition, is lazy and cheap. It may be part of a comprehensive collation and valid and valuable, but it is a bad quality source for Wikipedia. (Don’t use a phone book to determine the population of a town). For repeated information, go to the source. SmokeyJoe (talk) 02:47, 27 April 2026 (UTC)

A source may have all sorts of problems, but if it contains a ton of information, SIGCOV is not one of the problems it has. WhatamIdoing (talk) 02:54, 27 April 2026 (UTC)

It seems common for Wikipedians to misidentify/mislabel the point of failure of a bad source. Primary, non-independent, SIGCOV, reliable. Maybe we need a good picture? SmokeyJoe (talk) 02:57, 27 April 2026 (UTC)

I agree that it is common for Wikipedians to misidentify/mislabel the point of failure, including whether the problem is a bad source (see, e.g., all the people crying "MEDRS!" when they ought to be talking about NPOV). Accurately identifying the problem is not usually important to editors: they want to 'win' with as little effort on their part as possible, and if WP:THIS wins quicker than WP:THAT, then they'll say WP:THIS every time. However, when talking about changing a guideline or policy, we need to be precise and accurate, or we'll end up writing nonsense. WhatamIdoing (talk) 03:06, 27 April 2026 (UTC)

Yes.

Similarly, reproductions or close paraphrasing of press releases from, for instance, companies or other organizations (aka churnalism) does not count as sustained coverage. This is particularly the case because normally these are not secondary or independent sources.

The second sentence is untrue. SmokeyJoe (talk) 03:17, 27 April 2026 (UTC)

I'm not sure that I understand this. So:

• Bob's Big Business, Inc. posts a press release that says "Supply chain disruption slows widget production: YOY production declined 10%, causing hiring freeze", and goes on for another 300 words about this.
• The Daily News posts a lightly edited, 36-word version in its "Brief announcements" column: "Bob's Big Business, Inc. announced on Monday that supply chain disruption is slowing widget production. Widget production is down by 10% at the business compared to last year, causing the company to institute a hiring freeze".

I'd say that the newspaper is independent (e.g., not being paid by or beholden to the company), the material is still primary because there was no intellectual transformation of the press release, and the content all came from the company (not "intellectually independent", if you like that phrase). WhatamIdoing (talk) 18:50, 28 April 2026 (UTC)

I agree with what you’d say. SmokeyJoe (talk) 22:48, 28 April 2026 (UTC)

I'm late to this reply, but our policies are not separate from each other. WP:N and WP:SIGCOV requires that the information be reliable. Quantity is assessed after' we filter for reliability and independence. Shooterwalker (talk) 17:55, 19 May 2026 (UTC)

I don't care which order the sources are filtered in. If someone wants to look for SIGCOV first, and then has to throw out long sources as being unreliable or non-independent, then neither I nor any rule on Wikipedia says they're wrong. WhatamIdoing (talk) 22:37, 19 May 2026 (UTC)

I've been referred here by my colleagues at NPP. What are the notability criteria for articles with titles like "1977 in San Marino"? Do we need two reliable sources discussing 1977 in San Marino, indicating that the topic meets WP:GNG? Or is some other consideration applied? Could it even be that we know San Marino existed in 1977, and things definitely happened there, so it's always a pass?

Feedback would be really useful, as we have many such articles in our backlog.Boynamedsue (talk) 09:48, 3 May 2026 (UTC)

This is not really a Wikipedia-notability question. It looks more like a WP:SPINOUT question. Is Events in San Marino so large that it needs splitting, and at the level of year?

Reliable sources for 1977 in San Marino might be the answer if 1977 in San Marino was a spectacularly unusual year. SmokeyJoe (talk) 11:50, 3 May 2026 (UTC)

The timeline articles (not just for countries but a vast array of other topics) likely fall under WP:NLIST which means notability is not required to be considered, though obviously the major topic itself should be notable (San Marino in the example). All other content policies must be met still, and its expected that the list items to include have some selectivity to keep the list relevant and useful to the reader. Masem (t) 11:50, 3 May 2026 (UTC)

Great, that's really helpful, thanks.Boynamedsue (talk) 13:04, 3 May 2026 (UTC)

I was actually coming here for a similar reason, so I will use this thread, unless someone thinks I should start a new one. While looking at some infobox/template consolidation work, I encountered Template:Infobox year in country, which is essentially a "Year in Australia" template at the moment. After digging further and finding Template:Year nav and examples such as Template:Year in France, I realized through transclusion counts that we have thousands of such list articles (Template:Year in Europe is a very visible case, but not an isolated one). The overwhelming majority of them suffer from the same issues: minimal sourcing that never include anything actually about that year in that country (clearly failing WP:NLIST), completely arbitrary inclusion criteria regarding the contents (WP:SELCRIT), and very low page-views and editor activity (which compounds other issues). I do not actually think these issues can be fixed, because the necessary sources that would actually qualify as counting towards notability for such articles simply do not exist. However, while I am not afraid of suggesting I believe almost all of them should be deleted, I am not sure how to go about it in all honesty, so I would appreciate your input. To be clear, I am not lumping in there articles about calendar years, decades in specific places, and so on, as many such lists still clearly serve a purpose and can be adequately sourced; I just do not believe this is the case for most (and therefore thousands of) "year in country" pages. Choucas0 🐦‍⬛ 14:09, 11 May 2026 (UTC)

I have seen users arguing that natural languages are exempted from WP:GNG, and one reliable secondary source for them would be sufficient whether or not it contains WP:SIGCOV. Is this correct? Boynamedsue (talk) 10:03, 3 May 2026 (UTC)

For reference, this is the discussion, regarding Magiana language. @Kepler-1229b:@Kwamikagami:--Boynamedsue (talk) 10:06, 3 May 2026 (UTC)

Verifiable natural languages should be notable.

But. The lede is

Magíana is a poorly attested and extinct Arawakan language of Bolivia.

”poorly attested”? Attested is a weak word, weaker than verifiable. Redirect to [Arawakan languages]], note it at Talk:Arawakan languages and ask whether the content in the redirect history is worth merging. SmokeyJoe (talk) 11:41, 3 May 2026 (UTC)

If the one reference says “poorly attested” then the source is not reliable for the verification of the topic. SmokeyJoe (talk) 11:44, 3 May 2026 (UTC)

Thanks for your answers. In linguistics "poorly attested" usually means we have very few direct examples of the language, rather than of people writing about the language. For example, the Pictish language is almost as poorly attested as Magiana, as we have no brief wordlist giving examples of vocabulary and their translation, only personal and tribal names and Ogham inscriptions that might not be in the language and might not mean anything. However, there are whole books written about Pictish so it obviously passes WP:GNG.

We only know Magiana exists from two very short mentions in primary sources from the 18th century (one being a word list). No secondary sources have been written about it, except an entry on Glottolog which mentions the existence of the secondary sources and says it is Arawakan. This is a fail of WP:GNG and I don't feel there's been enough scholarship to definitively state it even existed as an independent language. I mean, I think it did, but I wouldn't bet anything I cared about on it.--Boynamedsue (talk) 12:25, 3 May 2026 (UTC)

I believe, from long experience at AfD, albeit mostly many years ago, that the application of the GNG to topics varies wildly between topic areas. Distant history and natural sciences get an easy run, currently trading for-profit companies get a tough run. Natural languages in distant history is the sort of thing I think will get an easy pass at AfD. I’m saying, the GNG doesn’t really apply meaningfully to this topic, rather than that its test fails. “Secondary source” is particularly poorly defined for fragmented coverage. “Reliable” depended heavily on what you mean by “reliable”, the topic existence, or the details, especially where there is very little secondary source detail. SIGCOV is poorly defined everywhere, and is especially poorly defined for topics that get easy passes in the GNG. SmokeyJoe (talk) 13:08, 3 May 2026 (UTC)

Cheers, interesting to hear. I might AfD it as a test case. If it gets a keep, it's pretty clear GNG doesn't apply to natural languages and it's worth trying to get that into GNG or write a subject specific guideline.--Boynamedsue (talk) 13:23, 3 May 2026 (UTC)

User:Boynamedsue, please be sure to nominate with a well articulated rationale for deletion, or redirection or merging. AfD does not work well with terse or ambivalent nominations. If you get an interesting result, as in contradicting this guideline in any way, please let us know. SmokeyJoe (talk) 13:42, 3 May 2026 (UTC)

I’m not agreeing that the “GNG doesn't apply to natural languages”. I would prefer to say that the GNG was not written to apply to distant historical topics with very little sourcing. SmokeyJoe (talk) 13:48, 3 May 2026 (UTC)

There's no exemption for natural languages.

That said, keep in mind the GNG is not black or white. If you do create an article based on one secondary source that has reasonable secondary coverage, that's a good enough presumption to at least start an article and get improvements via the Wiki-way, but you still should be expected to seek out more sourcing to improve how well it meets the GNG. Masem (t) 11:46, 3 May 2026 (UTC)

In my view the one secondary source isn't what you'd call reasonable. It doesn't really give any detail and merely notes the existence of two primary sources. It's more the principle I'm interested in though, as this is a real edge case.Boynamedsue (talk) 12:13, 3 May 2026 (UTC)

An article with one secondary source will not survive at the long term at AFD, however, it should be given the benefit of the doubt for initial creation (though pushing this to draft space over mainspace if possible) and remaining at the short term on the basis it is being worked on. But like if someone creates such an article and then does nothing with it for several months, that's a good reason to delete it. Masem (t) 15:46, 3 May 2026 (UTC)

At this point I'd contact the linguist who first brought up the existence and wordlist of Magiana at Glottolog. 🪐Kepler-1229b | talk | contribs🪐 15:49, 3 May 2026 (UTC)

Interestingly, Ramirez and Franca (2019) list Magiana in a map, but fail to mention it as an Arawakan language. 🪐Kepler-1229b | talk | contribs🪐 15:52, 3 May 2026 (UTC)

For a topic as difficult to define as languages in a field as regularly revised as linguistics sources are probably needed. That said, it also feels improbable there would be a clear natural language that doesn't meet GNG. CMD (talk) 12:36, 3 May 2026 (UTC)

Remember that GNG and all our SNGs have a banner at the top of the page noting that they outline general rules, but that there will be exceptions. So, while most natural languages may be deemed notable, a specific one may not make the cut. Blueboar (talk) 13:59, 3 May 2026 (UTC)

Endangered, if not extinct, languages are, honestly, very heartfelt or touchy topics to discuss, IMO. As I must say, failing GNG doesn't make the language any less notable than it already is, but that's my personal opinion.

Unfortunately, we don't have a separate notability guideline about languages. Rather you may have other considerations of notability, like WP:NTEMP and WP:SUSTAINED.

Perhaps the language's notability is not temporary not because it has existed and tried to survive. Rather it's been documented by existing reliable sources.

What matters more is the growth of the article about an endangered language, like the one you referred us to. If there aren't enough reliable sources that can help the article grow further, then I guess a redirect is more likely a suitable option... sadly. George Ho (talk) 14:11, 3 May 2026 (UTC); edited, 14:13, 3 May 2026 (UTC)

For me, it is improbable that an endangered language could ever fail WP:GNG. If we know it is endangered, there are enough sources. This language was only ever mentioned briefly in two written sources in the 18th century. Given where it was spoken, we can be fairly sure it is not currently spoken by a group in voluntary isolation. So it must either be extinct or a form of a language that does still exist but which was recorded so poorly that we can't tell.Boynamedsue (talk) 14:20, 3 May 2026 (UTC)

• I have nominated the page for deletion, here.Boynamedsue (talk) 06:47, 9 May 2026 (UTC)

This edit by BilledMammal changed:

However, once an article's notability has been challenged, merely asserting that unspecified sources exist is seldom persuasive, especially if time passes and actual proof does not surface.

to

However, once an article's notability has been challenged, notability must be proven to retain the article.

I've restored the earlier text for now. I agree with BilledMammal that their update reflects current practice. My worry is that, in reflecting current practice, it may change practice. I'd be concerned about people citing the punchy "notability must be proven" line in AfDs. There are non-zero exceptions and our guidelines are optimally more nuanced than this. However, I agree that the current text should be changed....it is quite verbose and imo doesn't convey its message particularly clearly. Vermont (🐿️—🏳️‍🌈) 01:38, 6 May 2026 (UTC)

It feels like there's a gap between "merely asserting that unspecified sources exist is seldom persuasive" and "notability must be proven".

The first says something about consensus. The latter makes me think – how? to whose satisfaction? Is it enough to list sources on the AFD page, or will someone say that it's not "proven" until the sources have been cited in the article itself? What if it's an offline source (like an older book)? Is it enough for me to tell you that Alice Expert's famous book, The Sun is Very Big, is definitely SIGCOV for Size of the Sun, or is that not really "proven"? WhatamIdoing (talk) 02:20, 6 May 2026 (UTC)

Maybe "notability must be demonstrated"? I think this would address WAID's objections and reflect that notability isn't always proven through reference to sources, but demonstrated by reference to SNG's. BilledMammal (talk) 02:24, 6 May 2026 (UTC)

SNGs are an additional complication, but what I really mean is: Please look at Wikipedia:Articles for deletion/White cake and tell me: At what point in time was the subject's notability "proven"? WhatamIdoing (talk) 02:48, 6 May 2026 (UTC)

Independent of SNGs, I do think "demonstrated" helps get this both closer to current practice and keeps WP:N internally consistent. If notability requires verifiable evidence (WP:NRV), it follows that determining notability in an AfD requires verifiable evidence. I'd +1 your earlier comment that "proving" has unhelpful implications; in the white cake AfD, I would consider your comment to constitute demonstrating notability, though whether it proves notability depends on the reader. Vermont (🐿️—🏳️‍🌈) 02:55, 6 May 2026 (UTC)

I think this comment did a lot to prove notability. Best, Barkeep49 (talk) 02:57, 6 May 2026 (UTC)

I do think it's important to underline, though, that all a SNG does is assert that the criteria it mentions usually imply that the bare minimum sourcing for the WP:GNG exist. (Per WP:SNG, The subject-specific notability guidelines generally include verifiable criteria about a topic which show that appropriate sourcing likely exists for that topic. Therefore, topics which pass an SNG are presumed to merit an article, though articles which pass an SNG or the GNG may still be deleted or merged into another article, especially if adequate sourcing or significant coverage cannot be found, or if the topic is not suitable for an encyclopedia.) Ultimately all articles must eventually have two independent reliable sources with significant coverage; the purpose of an SNG is that you can point to it and say "this implies that those sources probably exist." But if people are pointing to SNGs and saying it doesn't matter whether those exist, or if they're leading to us retaining articles for which those sources clearly don't exist, then that indicate that the SNG in question isn't valid and needs to be removed or revised. --Aquillion (talk) 02:45, 8 May 2026 (UTC)

I don't think that's true, and if we want it to be true, then WP:NPROF needs radical changes. WhatamIdoing (talk) 02:47, 8 May 2026 (UTC)

Just to note because several years we spent a long time on this, this applies to most SNG, but with the notable exception of NPROF, primarily because this predated WP:N and its get grandfathered. But yes, otherwise SNGs are likely indicators that with work and source additions, notability should be readily demonstrated. Masem (t) 02:47, 8 May 2026 (UTC)

Ultimately all articles must eventually have two independent reliable sources with significant coverage what you quoted from WP:SNG certainly does not say that and the existence of many articles that do not comply with your invented requirement is a strong reason to believe the community doesn't agree with you. Katzrockso (talk) 02:56, 8 May 2026 (UTC)

I think it more-or-less exactly reflects what I said, in terms of the purpose of WP:SNGs; SNGs exist to indicate that something would pass the GNG with work, not to provide alternate bottom lines. If something could never pass the GNG, we cannot write a neutral, balanced article on it. And this answers your other question - WP:EVENTUALISM and the like apply; we can have articles with less than two sources (even none) as long as editors believe the sources to pass the GNG exist somewhere. Other things can pass without notice simply because Wikipedia is huge and it takes time to find issues. The other issue is WP:FAIT; deleting articles is more difficult than creating them. But I am not aware of any article that unequivocally has no hope of passing the GNG, which has survived without controversy; there is of course some back-and-forth, but my view is that all such articles are the subject of ongoing disputes (largely driven by a reluctance to make mass-deletions even in cases, like this, where the article violates policy) that will eventually result in their deletion. Single-source articles are not acceptable, ever, anywhere, in any topic; they are inherently unencyclopedic and clearly violate multiple core policies. While some have been grandfathered in for now I do believe that the overarching consensus behind this will either force people to find additional sources eventually or accept a deletion or merge. --Aquillion (talk) 16:22, 21 May 2026 (UTC)

• The former is more convincing than the attempted change, IMO, despite current practices. Indeed, "seldom persuasive" is more specific and clearer than just "must be proven". The attempted rule doesn't specify what should happen if a topic's notability is not yet proven or has been not well verified to that point. (My further reply below...) George Ho (talk) 02:42, 6 May 2026 (UTC); expanded, 06:33, 6 May 2026 (UTC)

How about, once notability is challenged, "evidence is required"? Andre🚐 03:35, 6 May 2026 (UTC)

I'm not sure that's true. Consider:

• A: Tags with {{notability}}.
• B: Rv tagging; all ____ are notable.

In certain areas (e.g., species, modern heads of state, winners of the Oscars), there's a decent chance that this evidence-free argument will be accepted by the tagger and anyone else who happens to notice it. WhatamIdoing (talk) 04:34, 6 May 2026 (UTC)

Those arguments aren't evidence-free. The article needs to include evidence that the subject is a species, or a head of state, or an Oscar winner.

I think Andre's proposal is a good one; However, once an article's notability has been challenged, evidence is required. BilledMammal (talk) 05:05, 6 May 2026 (UTC)

Is B's assertion that the BLP is an Oscar winner "evidence"? WhatamIdoing (talk) 05:20, 6 May 2026 (UTC)

If all we have is B's assertion - if the article doesn't contain WP:V-compliant evidence to support the assertion that the subject is a species, a head of state, or an Oscar winner - then the argument won't be accepted by the tagger and anyone else who happens to notice it. BilledMammal (talk) 05:26, 6 May 2026 (UTC)

If you create a plausible-looking article that says something like "Paul Politician was the president of Ruritania from 1994 to 1998. He became president upon the death of Bob Bigman in 1994, but lost the 1998 election to Sam Slimy", people will generally believe you, even if the article is unref'd. WhatamIdoing (talk) 05:37, 6 May 2026 (UTC)

Then we definitely need to revise the current wording of WP:NEXIST to avoid encouraging that behavior. It creates both WP:V, and WP:BLP issues, and makes it easier to get hoaxes into Wikipedia.

Andre's suggestion is good, but maybe we should also consider a change to make it clear that new articles are expected to provide evidence of notability even prior to being challenged. BilledMammal (talk) 05:43, 6 May 2026 (UTC)

The next baby step towards that goal is to get a rule into WP:V that says all articles need to cite at least one source. So far, every proposal has failed (usually for overreach: they've not been content to leave it at just "any one source", but instead demand that the source must be X, Y, and Z). It's been more than a year since the last attempt, so we could try again. However, I think that Category:All articles lacking sources might be emptied in late 2027, so trying then might be logical. WhatamIdoing (talk) 17:42, 6 May 2026 (UTC)

And? People believe a lot of things without evidence, but that isn't good or consensus-oriented policy necessarily. Andre🚐 05:44, 6 May 2026 (UTC)

In practice, when people believe that, they stop challenging the article's notability. This is consensus in action, even if we think it's a suboptimal approach. WhatamIdoing (talk) 17:43, 6 May 2026 (UTC)

However, I agree that the current text should be changed....it is quite verbose and imo doesn't convey its message particularly clearly.

• Oh, I see what it's all about... methinks. My proposed wording wouldn't be that good. All I might come up is the following in a working draft quality to reflect current practices but still in the spirit of the former:

  However, once an article's notability has been challenged, in order to help you user provide a reason to keep the article, merely asserting that an existing source already verifies its notability without evaluating their reliabilities and stating a type of that source would not help support your reason to keep that article.

  Doesn't look good, huh? George Ho (talk) 06:33, 6 May 2026 (UTC); edited, 06:36, 6 May 2026 (UTC)

Notability is a guideline: a best practice that should be met in usual cases. It is not a policy--something that must be adhered to regardless of circumstances. As such, no guideline, even notability, should have a must in its wording or explanatory wording. Jclemens (talk) 07:55, 6 May 2026 (UTC)

@Jclemens, please see Wikipedia:The difference between policies, guidelines, and essays, and then to go count the number of times that the word must appears in Wikipedia:Manual of Style. The MOS says must more than all of our legal policies combined, and it is not wrong to do so. WhatamIdoing (talk) 17:47, 6 May 2026 (UTC)

The MOS is absolutely and incontrovertibly wrong to use the word "must". I've had this beef with it for years. People have cited MOS pages--which deal with how we present information--in AfD discussions, which deal with whether we should have a particular page at all. Lots of people, including most of the people who contribute to or opine on Wikipedia policy, don't understand how to construct policy. Wikipedia is a poster child for how not to do things. Jclemens (talk) 19:03, 6 May 2026 (UTC)

Just reminding that every one of our guidelines start with a banner noting that occasional exceptions apply. There are a lot of strong “Shoulds” on WP… and very few “Musts”. Blueboar (talk) 20:52, 6 May 2026 (UTC)

And IMO most of those musts are basic points of grammar and punctuation, which is why the MOS ends up having more musts than the legal policies. It'd be silly to tell people that they only "should" use proper grammar and punctuation in articles. WhatamIdoing (talk) 21:41, 6 May 2026 (UTC)

Yes… but… sometimes there is disagreement about what the proper punctuation and grammar actually is. And even when there are no disagreements there are situations where using incorrect grammar or punctuation is appropriate (when quoting for example)… Which means that even in those areas there are “exceptions” to the generalized rules (and so we should avoid firm “musts”). Blueboar (talk) 18:24, 7 May 2026 (UTC)

I agree with you, which is why I think that the wording of the quoted text must be faithfully reproduced is a perfectly reasonable thing for the MOS to say. WhatamIdoing (talk) 01:27, 8 May 2026 (UTC)

Isn't this semantic though, JClemens et al. Shall, must, or coulda woulda, the question is: should the policy or guideline such as it is, Notability, a key touch point for wikipedia decisioning, advise and/or legislate that articles which do not substantiate their claim of notability with evidence of an acceptable form may be deleted? Andre🚐 02:17, 8 May 2026 (UTC)

No, because that would be contrary to WP:ARTN and WP:NEXIST. Katzrockso (talk) 02:42, 8 May 2026 (UTC)

It's not clear that it would be, but a better question is, should it? Do we want to encourage editors to create insufficiently sourced articles, or do we want to encourage them to demonstrate their creations notability from the start? BilledMammal (talk) 02:45, 8 May 2026 (UTC)

How is it not clear that even a completely unsourced article can be notable, since notability applies to topics/subjects? Do you disagree with WP:NEXIST and/or WP:ARTN? Katzrockso (talk) 02:51, 8 May 2026 (UTC)

All I'm asking is whether encouraging the creation of unsourced articles is a positive for the encyclopedia, or if we would be better of discouraging them.

Personally, I think we should discourage them, as unsourced articles have WP:V, WP:NPOV, and WP:BLP issues, and make it far easier to introduce hoaxes to Wikipedia. What do you think? BilledMammal (talk) 03:00, 8 May 2026 (UTC)

Some unsourced articles have WP:V, WP:NPOV, and WP:BLP problems – not all.

And if I were going to create a hoax article, you can bet that there would be impressive-looking sources in it. WhatamIdoing (talk) 03:07, 8 May 2026 (UTC)

Per WP:V, completely unsourced articles can only exist so long as nobody challenges them. Any material that has been challenged and which lacks an inline citation must be removed immediately until a source is provided (the responsibility for which lies on whoever wants to retain it, per WP:BURDEN.) Now, "challenge" is a thorny concept in this context and carries an implication that someone isn't just going to go around challenging everything indiscriminately (ie. it can remain as long as nobody has an issue with it and people aren't supposed to be challenging clearly uncontroversial stuff solely for the lack of a source, otherwise the implication in WP:V that things can remain until challenged would be meaningless), but once you're at the point of having an active dispute over the article's content, V is clear - everything that gets challenged and which you can't provide a source for has to go. There's also some things which fall under the category of "clearly likely to be challenged" which can't be retained in an article without a source at all. (And ofc anything BLP-sensitive and any quotes have to go instantly and can never be retained without a source, no exceptions, but I think we all understand that.) --Aquillion (talk) 03:12, 8 May 2026 (UTC)

@Aquillion, this bit: Any material that has been challenged and which lacks an inline citation must be removed immediately until a source is provided (the responsibility for which lies on whoever wants to retain it, per WP:BURDEN.) is wrong.

WP:V specifically contradicts this, particularly in the paragraph beginning "Whether or how quickly material should be removed for lacking an inline citation..."

WP:V does not impose any deadlines on adding citations when they're needed. WhatamIdoing (talk) 03:18, 8 May 2026 (UTC)

four types of information must be accompanied by an inline citation to a reliable source that directly supports the material: ... material whose verifiability has been challenged

.

We're getting off topic here, but I do not understand these objections. Do you actually want editors to add content without a source? Do you want us to have content that doesn't have a source? I don't, and I cannot understand why any editor would. BilledMammal (talk) 03:21, 8 May 2026 (UTC)

I don't want editors spreading false rumors about what's in the policies, because that leads to avoidable drama and to never being able to fix the policies/guidelines.

• Y Yes: The policy says these four types of information must have an inline citation.
• N No: The policy does not say that the required inline citation has to be supplied immediately. It also does not say that this inline citation has to be supplied before NPP or AFC accepts the article, within the 90-day default window for search engine indexing, before the end of the year, or even within my lifetime.

WhatamIdoing (talk) 03:28, 8 May 2026 (UTC)

Well, it's true that it doesn't have to be removed immediately. But it may be removed immediately, and when it is, restoring it would be a violation of WP:V (since it has now been challenged.) Again, the key point is that V allows it to remain only so long as it's not in dispute. --Aquillion (talk) 03:31, 8 May 2026 (UTC)

No. WP:V allows it to remain so long as it's not been removed as part of a WP:CHALLENGE. Something can be "in dispute" indefinitely. Category:Articles with unsourced statements from February 2007 has more than 800 articles in it. That's almost 20 years ago, and WP:V allows that. WhatamIdoing (talk) 03:50, 8 May 2026 (UTC)

If we all agreed on what constitutes an "insufficiently sourced article", it might be worth pursuing that. But we don't – we really don't, especially when you take editors out of their favorite subject areas and ask them to pass judgment on subjects they know little about – so there's no point. If someone writes an article that says:

• A periodontal cyst is a cyst in or near a tooth.

and cites https://www.ncbi.nlm.nih.gov/mesh/C04.182.089.530.690.790 (i.e., a webpage that contains two sentences), the wrong responses sound like:

• That's just {{one source}} and you need to prove notability with at least two sources.
• That source is primary, and you need to prove notability with secondary sources.
• That's too short, so it's not WP:SIGCOV, and you need to prove notability with SIGCOV sources.

An appropriate response sounds like something like this:

• Thanks for creating that article. I can't believe that it wasn't done years ago. Do you need any help finding sources so you can expand it?

But perhaps you're more of the m:immediatist POV, so you think that having nothing is better than an accurate, appropriately sourced starting point? WhatamIdoing (talk) 03:06, 8 May 2026 (UTC)

Can you explain why those are the wrong responses?

However, that example does demonstrate one of the issues with permitting unsourced and undersourced content. The correct title is Lateral Periodontal Cyst, which we already have an article on - and if the editor had spent the time to find sources, rather than creating a microstub and saying "other people can do the work", they would have realized that and not introduced issues to the encyclopedia that other editors would need to spend time cleaning up. BilledMammal (talk) 03:13, 8 May 2026 (UTC)

The correct title is not lateral periodontal cyst, because a periodontal cyst comes in a couple of flavors, only one of which is the lateral type. (Today we have proven that you're not a dentist, so to give an analogy with a more familiar subject, what you've written is like saying that we don't need an article about the general subject of lung cancer, because we've already got one about small-cell lung cancer.)

The reason those are the wrong responses is because:

1. WP:NEXIST, rather than "already cited", is the rule for notability.
2. The article fully meets the requirements of WP:V. It contains only a definition of the term, and it links to a source that is reliable for that definition.
3. If you know something about the particular cited source, then you know you're three clicks away from finding out whether there are any MEDRS-compliant sources on this subject.

Also: Who says that the editor is saying "other people can do the work"?

Also also: Who says that letting other people "do the work" is bad? We have a rule in WP:OWN prohibiting editors from refusing to allow other people "do the work", but we have no rule requiring any editor to do some greater amount. You (and I) might wish that they would Wikipedia:Beef up that first revision, but there's no rule requiring it. WhatamIdoing (talk) 03:40, 8 May 2026 (UTC)

Periodontal cysts are odontogenic cysts related to periodontal tissues. These include Lateral periodontal cyst, Botryoid odontogenic cyst, and Gingival cyst, but periodontal cyst will typically refer to the lateral periodontal cyst. However, the correct article for the content you referenced might be Odontogenic cyst. Either way, that article isn't appropriate - and would only create cleanup issues.

As for the rest of your comment, we're discussing here whether the wording of NEXIST actually reflects practice and consensus (it doesn't). Further, any editor creating a microstub that doesn't demonstrate WP:N is met is saying "other people can do the work", as they want to create the article but can't be bothered to do the work of actually demonstrating an article is suitable. One editor doing that isn't an issue - but thousands of editors doing it hundreds of thousands or millions of times creates an insurmountable backlog. BilledMammal (talk) 03:49, 8 May 2026 (UTC)

I think that article is appropriate and that it would not create clean-up problems. What actual {{clean up}} could be needed in such a short, accurate, and 100% cited article? The only "problem" is that it's very short, and that "problem" is just a personal preference.

(If you'd read the source I cited, you'd know that periodontal cysts are a subtype of odontogenic cysts.) WhatamIdoing (talk) 04:13, 8 May 2026 (UTC)

WP:IMPERFECT states Hence, perfection is not required, as Wikipedia is a work in progress. Collaborative editing means that incomplete or poorly written first drafts can evolve over time into excellent articles. Even poor articles, if they can be improved, are welcome. For instance, one person may start an article with an overview of a subject or a few random facts. Another may help standardize the article's formatting or have additional facts and figures or a graphic to add. Yet another may bring better balance to the views represented in the article and perform fact-checking and sourcing to existing content. At any point during this process, the article may become disorganized or contain substandard writing.

Do you disagree with this part of the Wikipedia:Editing policy? Katzrockso (talk) 04:14, 8 May 2026 (UTC)

I certainly have the impression, formed over years of discussion, that BilledMammal does not "welcome" any "poor articles" at all, even if they can be easily improved. WhatamIdoing (talk) 04:16, 8 May 2026 (UTC)

Rather than try to repeal existing policy in a roundabout way (by making imperfect articles unwelcome), it might be more productive for BilledMammal to directly try to change the Wikipedia:Editing policy. Katzrockso (talk) 04:19, 8 May 2026 (UTC)

It duplicates existing content, and it doesn't demonstrate notability. Poor articles are acceptable, but there becomes a point where the cost of cleanup is greater than the benefit the article provides. Articles created by LLM are clear example of this.

However, as I have said to you elsewhere, we are getting off topic, so I am going to leave this here. BilledMammal (talk) 04:21, 8 May 2026 (UTC)

(I uncollapsed this section as it was collapsed while I was in the middle of writing this comment)

I don't think it's "off-topic" to clarify the understanding of existing policy that we are attempting to discuss here.

If I say "According to WP:ABCQ, we should do X" and you say "According to WP:ABCQ, we should do Y", our contrary interpretations of WP:ABCQ are perfectly topical to whether we should do X or Y. Katzrockso (talk) 04:26, 8 May 2026 (UTC)

• It duplicates existing content – No, it doesn't.
• and it doesn't demonstrate notability – We don't have any rule requiring this yet. You're trying to create that rule because it doesn't exist yet.

WhatamIdoing (talk) 05:04, 8 May 2026 (UTC)

Indeed, as Wikipedia:Wikipedia is not compulsory, disallowing improvements to the encyclopedia on the grounds that they are incomplete goes against the spirit of WP:NOT. Katzrockso (talk) 04:22, 8 May 2026 (UTC)