Wikipedia talk:Notability#GNG plural

Perhaps this guidance could have a clarification on what it means to be "direct". Since it isn't explicitly defined, some editors argue that nonspecific coverage of a group, e.g. "[6-person sports team] flew to [host country] and successfully defended its title in [tournament], with all players scoring points despite struggling in the unexpected heat", is also direct coverage of each individual member (or at least of any members mentioned elsewhere in the source). JoelleJay (talk) 14:57, 17 April 2025 (UTC)

:That view is contrary to any reasonable definitions of the word "directly" and the phrase "in detail". I don't think we need to start defining ordinary words because some people make dumb arguments. voorts (talk/contributions) 16:22, 17 April 2025 (UTC)

::@Voorts, I agree, but I think the argument is that if all the details in the coverage of a group apply equally to each member, then it wouldn't be OR to state those details just in the context of one member in an article on that member... ∴ the clause {{tq|so that no original research is needed to extract the content}} is satisfied... ∴ the "SIGCOV" part of GNG is satisfied... Obviously the intent of having SIGCOV, and its application in practice, goes beyond simply the requirement that a given source not need OR to use it, but I have encountered arguments in multiple AfDs, including by admins, where parts of a source that are talking about a group are claimed to be SIGCOV of one of its members. JoelleJay (talk) 18:34, 17 April 2025 (UTC)

:::I agree that it wouldn't be OR.

:::I'm not sure that there is a single, objective, universal answer. On the one hand, if something is said about two people, it's nitpicky to say this isn't directly about the two people. On very extreme the other end of the spectrum, we can source "All men are mortal", but we're not going to add "Sooner or later, he's going to die" to every BLP, nor would we consider that large group reference to show anything about the notability of any individual. WhatamIdoing (talk) 03:47, 18 April 2025 (UTC)

:It's going to be subjective but we pose this as being far from just a passing mention. The entire reference doesn't have to be focused on the topic in question but it should at least one paragraph devoted to directly taking that topic. Masem (t) 16:36, 17 April 2025 (UTC)

::Sure, but what does it mean for a paragraph to be "devoted to directly talking about that topic"? JoelleJay (talk) 18:23, 17 April 2025 (UTC)

:::I suggest that a paragraph is "devoted to directly talking about that topic" if it provides you with material that is incontestably suitable for a Wikipedia article about that topic. For example, if you have a passage about Bob's Big Business, Inc. that provides you useful encyclopedic information for an article on {{fake link|Bob's Big Business, Inc.}}, then that's "directly talking about that topic"; if you instead have a passage that maybe namechecks the business but is actually talking about the widget industry in general, or about something tangentially related (Bob likes baseball; Bob's sister is a possibly notable lawyer), then that's not incontestably suitable for a Wikipedia article about the business, as editors could argue that Bob's love of baseball is about him, and his sister's career ought to be described in a separate article, {{fake link|Lee Lawyer}}. WhatamIdoing (talk) 03:53, 18 April 2025 (UTC)

::::That's how I view this too. See my essay WP:SPECTRUM. voorts (talk/contributions) 14:22, 18 April 2025 (UTC)

:::::This is a good essay. I might not agree with 100% of it, but it explains why bright line tests aren't going to work. It's going to come down to what lets us build a unique non-stub article. Anything less is either ripe for deletion, or at best a merge into another non-stub article. Shooterwalker (talk) 18:38, 29 April 2025 (UTC)

IMO the main intent is we need the type of coverage of the topic of the article to build a real enclyclopedia article on the topic of the article from. "Directly" means the content is about the topic of the article. I think that that provides guidance for people who are seeking it. But maybe we should add " "Directly" means the content is about the topic of the article " to handle situations where folks prefer to not follow the intended meaning of "directly". Sincerely, North8000 (talk) 18:48, 17 April 2025 (UTC)

Pinging @Masem and @WikiOriginal-9.

I don't like exactly the wording of "assumption" but it appears that "presumption" is not favored, so taking to talk page as part of WP:BRD cycle.

"Rebuttable presumption" doesn't sound too deletion leaning, it sounds less clunky than "assumption, not a guarantee". At least that is my thought. What do you think? Aasim (話す（はなす）) 03:48, 13 May 2025 (UTC)

:Notability is a rebuttable presumption in practice, and language elsewhere in the guideline uses this term. "Assumption" is a bit too much of a simplification because it belies the fact that passing the letter of the text of a test like the GNG is not assurance that notability has actually been demonstrated, that's the "rebuttable" part coming in. Masem (t) 11:57, 13 May 2025 (UTC)

:For reference, I assume you are discussing [https://en.wikipedia.org/w/index.php?title=Wikipedia:Notability&diff=prev&oldid=1290134574 this edit]: Although there can be times where it is warranted, I think part of motivation for using a term other than "presumption" is to avoid using a variant of the word being described ("presumed"). I think adding "rebuttable" is redundant with the clause that the established assumption is not a guarantee. isaacl (talk) 15:36, 13 May 2025 (UTC)

::"rebuttable presumption" is a legal term, see presumption, that has similar apicability to how notability is practiced particularly with SNGs, hence why we have used it here. Masem (t) 16:17, 13 May 2025 (UTC)

:::The guideline used to link to Rebuttable presumption, which used to be its own, law-specific article [https://en.wikipedia.org/w/index.php?title=Rebuttable_presumption&diff=prev&oldid=1152666670 until about 18 months ago]. Conclusive presumption was also a separate article [https://en.wikipedia.org/w/index.php?title=Conclusive_presumption&diff=prev&oldid=1152666740 back then]. These were blank-and-redirect moves after 24.5 hours' "discussion" at Talk:Presumption#Merge proposal, in which the now-retired editor says that he wants to merge them, and then says that he redirected them and would later add the contents (spoiler alert: he never did, though Dl2000 merged in the contents of Conclusive presumption).

:::A few months after that, @Espresso Addict [https://en.wikipedia.org/w/index.php?title=Wikipedia:Notability&diff=prev&oldid=1179870511 removed] the link to Rebuttable presumption from this guideline, on the grounds that the link was a redirect. This was discussed in Wikipedia talk:Notability/Archive 79#Do we need "rebuttably"?

:::The point behind "rebuttable" language is not to create or invoke a legal system, but to clarify that this presumption is not a conclusive/mandatory/irrebuttable presumption. For those unfamiliar with this language, a rebuttable presumption is something like "I saw Karp in the elevator, and he said it was np-complete." You could rebut that with evidence showing the person didn't see Karp, it wasn't in the elevator, he didn't say that, and/or that it's not actually np-complete. A non-irrebuttable presumption is "Dogs are prohibited on the premises. However, any dog trained to assist a blind member is hereby deemed to be a cat for the purposes of enforcing the ban on dogs". You can't introduce evidence that contradicts this definition, because no factual evidence can overcome the definition. The blind member's dog is not banned, because the ban itself says that the blind member's dog isn't a dog. WhatamIdoing (talk) 17:55, 13 May 2025 (UTC)

:::I'm not sure what you mean by "we have used it here". Are you signing on as a joint author for the proposed change? (The [https://en.wikipedia.org/w/index.php?title=Wikipedia:Notability&diff=prev&oldid=1290146382 longstanding version to which you reverted] doesn't use the word "rebuttable".) I agree with WhatamIdoing that I don't think it's necessary to draw a parallel with a legal term of art. In my view, saying something is not a guarantee is clear without using jargon. isaacl (talk) 21:31, 13 May 2025 (UTC)

::::If you scroll down you can see. Sometimes diffs can be confusing. Aasim (話す（はなす）) 21:46, 13 May 2025 (UTC)

:::::It would help if you would be a bit more specific about what you are proposing. I don't know where you are scrolling (the page history?). Can you write out your proposed change on this talk page? (And any specific explanation for "we have used it here"?) isaacl (talk) 04:10, 14 May 2025 (UTC)

{{textdiff|* "Presumed" means that significant coverage in reliable sources creates an assumption, not a guarantee, that a subject merits its own article. A more in-depth discussion might conclude that the topic actually should not have a stand-alone article—perhaps because it violates what Wikipedia is not, particularly the rule that Wikipedia is not an indiscriminate collection of information.Moreover, not all coverage in reliable sources constitutes evidence of notability for the purposes of article creation; for example, directories and databases, advertisements, announcements columns, and minor news stories are all examples of coverage that may not actually support notability when examined, despite their existence as reliable sources.

• {{shortcut|WP:SIGCOV}} "Significant coverage" addresses the topic directly and in detail, so that no original research is needed to extract the content. Significant coverage is more than a trivial mention, but it does not need to be the main topic of the source material.

|What this means:

• "Presumed" means that significant coverage in reliable sources creates a rebuttable presumption, not a guarantee, that a subject merits its own article. A more in-depth discussion might conclude that the topic actually should not have a stand-alone article—perhaps because it violates what Wikipedia is not, particularly the rule that Wikipedia is not an indiscriminate collection of information.Moreover, not all coverage in reliable sources constitutes evidence of notability for the purposes of article creation; for example, directories and databases, advertisements, announcements columns, and minor news stories are all examples of coverage that may not actually support notability when examined, despite their existence as reliable sources.
• {{shortcut|WP:SIGCOV}} "Significant coverage" addresses the topic directly and in detail, so that no original research is needed to extract the content. Significant coverage is more than just a trivial mention, but it does not need to be the main topic of the source material.

}}

::::::Essentially this is what was supposed to be shown by the diff. MediaWiki failed at indicating exactly what changed. Aasim (話す（はなす）) 16:53, 14 May 2025 (UTC)

:::::::Yes, that's exactly what I understood was being changed in the diff. isaacl (talk) 08:59, 15 May 2025 (UTC)

::::IMO the actual longstanding version is the one that contained a link to Rebuttable presumption for 13 years, representing 70% of this guideline's existence. WhatamIdoing (talk) 21:46, 13 May 2025 (UTC)

:::::I was just quoting the edit summary. The initial reply said "language elsewhere in the guideline uses this term", thus stating that the term is in current version. I understand the guideline used to contain related language before. isaacl (talk) 04:09, 14 May 2025 (UTC)

::::::The link (not always the word rebuttable, but always the link itself) was in the lead for 13 years. I haven't checked other parts of the guideline. WhatamIdoing (talk) 17:58, 14 May 2025 (UTC)

I am raising this issue here for comments since I see it as an issue that is coming up indirectly at AfD discussions and related questions of notability. It relates to what is called churnalism, which is the proliferation of reproduction of press releases from universities and companies. One example is the 58 articles listed [https://acs.altmetric.com/details/167943289/news here], mainly IMO because the Internet likes cats. (I am using that example as I don't want to denegrate the work of others.) I do not consider those 58 articles as WP:SIGCOV, it is WP:TOOSOON.

When it comes to BLP academic notability, churnalism articles are generally discounted or ignored, the key test is peer recognition. While I have often seen them ignored for proposed products or new science papers, I have also seen them invoked. I think opinions plus some general edits to the various notability pages might be useful. (Apologies if there is already material and/or an essay or three that I have missed.) Ldm1954 (talk) 11:49, 4 June 2025 (UTC)

:N.B., there is an essay WP:CHURNALISM which is relevant, but not exactly on this. It could be expanded. Ldm1954 (talk) 12:27, 4 June 2025 (UTC)

::Think there is one existing aspect of notability that can already be used, in that a burst of coverage is not considered a factor for notability, but enduring coverage. Churnalism is not enduring.

::The other factor that could be argued is that when there is clearly one primary source that dozens of other sources have reported on (this is not limited to scientific fields), those additional sources are not independent nor secondary coverage generally if they are just repeating the points of the primary source. (I have seen good secondary sources discuss the impact that the primary source describes in a larger manner, transforming that information, which is fine). We don't use primary sources for determining notability. Masem (t) 12:35, 4 June 2025 (UTC)

:Thanks for raising this issue. WP:CHURNALISM is an issue in many topic areas and I think it's inevitable that we'll need to confront it across the board. For now, I've only seen reactions on a source-by-source basis. The problem with that it fails to address the problem across the board. It also fails to consider that some sources do put out low-quality fact-checked reporting mixed in with low-quality clickbait for engagement. And that's going to become more and more the norm, as the journalism industry suffers economic hardship. Shooterwalker (talk) 23:37, 5 June 2025 (UTC)

::It's already in the guideline: Wikipedia:Notability#Notable topics have attracted attention over a sufficiently significant period of time. {{xt|Brief bursts of news coverage may not sufficiently demonstrate notability.}} It doesn't matter whether that's technically churnalism (e.g., a quickly copyedited press release) or just the thing that everyone independently decided to write about last Thursday (e.g., "Look at what she wore to the gala") and never thought about again afterwards. If it's only in the news for one week, it's not notable/doesn't qualify for a separate article. (Something that is only in the media briefly could still be appropriate to include as a paragraph in a larger article.)

::I assume this relates to the dispute at Superwood, which the OP has sent to Wikipedia:Articles for deletion/Superwood, and the complaint the OP filed at Wikipedia:Administrators' noticeboard/Incidents#Aggressive editing by CresiaBilli. WhatamIdoing (talk) 01:18, 6 June 2025 (UTC)

:::While that is one example, I can think of many other instances in AfC/AfD/PROD from science press releases; as in the example I mentioned earlier, it is not hard to get 50 or so hits. As the other comments above indicate, I am not alone in my concerns. While WP:Sustained as you invoke is relevant, people have responded by claiming WP:Sigcov. I think the policy you point to could be usefully reinforced by being more specific, including mentioning churnalism. Ldm1954 (talk) 01:45, 6 June 2025 (UTC)

::::A bunch of articles reiterating a press release is not significant coverage, because those aren't secondary or independent sources. Masem (t) 03:12, 6 June 2025 (UTC)

:::::SIGCOV is about how much (relevant) content the source contains. SIGCOV is not about whether the source is otherwise useful or any good. A press release, like any other type of source, can be full of useless fluff and have no SIGCOV. A press release, also like any other type of source, can also be full of relevant and encyclopedically appropriate facts, in which case it's SIGCOV (but still worthless for proving notability due to its non-SIGCOV failings). WhatamIdoing (talk) 06:32, 7 June 2025 (UTC)

::::::To expand: Press releases can be independent of the source. We tend to think of them as "buy our product" or "look at what I did", but that's not the only use.

::::::To give a real-world example, the most recent press release from Consumer Reports (which prefers to call them "news releases") is https://www.consumerreports.org/media-room/press-releases/2025/05/consumer-reports-investigation-uncovers-krogers-widespread-data-collection-of-loyalty-program-members-to-create-secret-shopper-profiles/ The subject is a US retailer named Kroger and their privacy practices. Consumer Reports and their press release is WP:INDY of Kroger. It's still not proof of notability, but it's an independent source if you want to expand Kroger#Controversies.

::::::What the OP is saying is that if Consumer Reports investigates a problem, and proactively provides information about the problem to the news media, then any resulting/related news articles should be ignored as worthless churnalism. I think you'll find that the role of press releases and publicity in the news is more complicated than that simplistic judgement. WhatamIdoing (talk) 06:41, 7 June 2025 (UTC)

::Specific suggestion (wordsmithing possible). After the paragraphs in Wikipedia:Notability#Notable topics have attracted attention over a sufficiently significant period of time add:

::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 in most cases these are not secondary or independent sources.

::My intent is to include some specific wording that can be used in WP:NPP, perhaps eventually be added to the tagging scripts. At the moment I argue that there is a gap.Ldm1954 (talk) 03:38, 6 June 2025 (UTC)

:::So... the answer kinda depends on what job you think the sources are supposed to be doing for us.

:::If your idea is "Wikipedia shouldn't have an article on ____ unless real-world, for-profit businesses thought that spending at least n hours paying someone to research the subject, because spending money on reporters' time is how the newspaper proves that it's really worthwhile and important content", then of course you're going to object to "shortcuts" like reading information supplied by the subject.

:::If your idea is "Wikipedia shouldn't have an article on _____ unless professional editors decided that ____ to give them space in their publication", then the method by which the article gets written becomes much less important. What matters in that model is that the editor/publisher chose to have something in their newspaper/magazine/whatever about _____. WhatamIdoing (talk) 06:30, 7 June 2025 (UTC)

I've got no objection to the suggested wording, but feel that this is fixing the smaller problem (because scientific churnalism isn't really different to any other churnalism, on which we have policies already). The huge, huge elephant in the room of science and technology is the vast increase in barely-selective journals with dubious peer review, merging into the substratum of predatory publishing and paper mills. I'll give a specific borderline example, Kirtiraj Gaikwad, recently kept at AfD based on WP:NPROF#1. He produced 45 publications in 2024, and has produced 13 already this year. Two thoughts: (1) honestly, do you believe that a researcher can produce a meaningful, novel paper in little over a week? (2) if someone produces 45 papers per year and cites themselves in each publication, is it surprising that their publications are highly cited? This sort of thing runs a risk of drastically undermining the spirit of NPROF, with the risk that the non-academic wikipedia-world will eventually rebel, and argue that peer-reviewed output isn't evidence of notability because (1) there's good documented evidence that in many cases peer review is compromised or simply not happening, and (2) citations are so contaminated by non-independent citations (self-citation, and you-scratch-my-back-I-scratch-yours citation rings with friends) that #C1 is fatally flawed. And I'd have to agree. Elemimele (talk) 12:05, 6 June 2025 (UTC)

:That might be something that is covered by what are considered reliable sources under WP:SCIRS and WP:MEDRS , and works that are not reliable sources (including non-peer reviewed journals) do not qualify to demonstrating notability. However, this also feels like more a problem specific to NPROF in handling academics that publish in this fashion, since the criteria there are a bit different from GNG due to NPROF's predating of GNG. Masem (t) 12:17, 6 June 2025 (UTC)

::Oh, that's really interesting. I didn't know NPROF came before GNG, I just sort-of-assumed GNG was the original gold standard and academics got a special dispensation. I think you're right that it's specific to NPROF and I'm grateful that Ldm1954 started this discussion (I don't want to derail the topic though). Academics are unique because we're assessed so heavily on the sheer bulk of our publication output, and because of the sheer volume of outlets to be assessed; I'm not sure how we can do it without overwhelming the reliable sources noticeboard. It's really hard to distinguish a lowish-impact niche journal respected in its field and genuinely peer-reviewed from a lowish-impact predatory journal that claims to be peer-reviewed and isn't (let alone from lowish-impact non-predatory journals with stressed editors who can't find any peer reviewers and therefore fudge the issue). Elemimele (talk) 13:05, 6 June 2025 (UTC)

:::Yes, NPROF is handled uniquely both due to predating GNG and that in general academics are difficult to write about from a biographical side but their research is what gets the focus. But all that said, chugging out papers in non-peer-reviewed or predatory journals should not be applicable for that demonstration. But that's all under the focus of NPROF there. Masem (t) 14:30, 6 June 2025 (UTC)

This thread is interesting but confusing because it seems to be about very different topics. Some parts of it are about academics and that SNG. But even there it seems to imply that (mere) publishing of papers by the article's subject is a way to meet criteria #1 of the SNG but I don't see where that is in the SNG or on the noted example article/AFD. But then one of the discussed examples is about a product which is very different. North8000 (talk) 14:13, 6 June 2025 (UTC)

:I may have side-tracked things from Ldm1954's original concerns about academic churnalism. But yes, {{u|North8000}}, mere publishing of papers by the article's subject is indeed deemed a way to meet criterion #1 because (1) the peer review process is take as indicating that the work is independently viewed as valuable by someone other than the author, and (2) high-citation is taken as evidence that it's made significant impact in the scholarly discipline. My argument is that paper-mills and citation-rings and even just splitting your work into hundreds of micro-papers in low-end journals make a nonsense of both aspects, and that this is very much akin to churnalism in that it's the creation of a very large number of near-meaningless sources. Elemimele (talk) 17:51, 6 June 2025 (UTC)

::The Least publishable unit is rewarded by some systems, so it'd be surprising if we didn't see it. I thought I'd heard that (some?) citation metrics discount self-citations, though. (Or maybe that was just a proposal for how to improve metrics?) WhatamIdoing (talk) 06:26, 7 June 2025 (UTC)

:::Citation rings, as mentioned by Elemimele, are about getting others to cite you (and you cite them) so that the high citations are not self-citations. It can be difficult to distinguish these from legitimate but specialized subfields where everyone cites everyone else because that's all there is to cite in the subfield. But in the occasional case when dubiously inflated citations are suspected in an academic AfD, my experience is that it is typical to look for other indicators of academic impact and other WP:PROF criteria instead. A recent example (ending in delete): Wikipedia:Articles for deletion/Roshdi Khalil. —David Eppstein (talk) 07:41, 7 June 2025 (UTC)

:Churnalism: can we please return to the issue of churnalism, and leave citation metrics to WT:NPROF. I think I saw in the comments a concensus developing for adding something like the text I gave. Yes? Ldm1954 (talk) 07:32, 7 June 2025 (UTC)

::Not hearing any disagreements, I have been bold and added two sentences (slightly ce) to the text as I indicated above. Ldm1954 (talk) 13:58, 12 June 2025 (UTC)

:::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 also wonder whether churnalism ought to be handled as a problem of "multiple sources". The guideline says "Similarly, a series of publications by the same author or in the same periodical is normally counted as one source." This is the main problem with churnalism: it's basically the same source.

:::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.) WhatamIdoing (talk) 03:07, 13 June 2025 (UTC)