DZ
DefZone.Net

Kien thuc cong nghe & giai tri

  1. Home
  2. Wiki
See also: Wikipedia talk:Relevance

What is "This page in a nutshell"

[edit]

Idk ~2025-34132-29 (talk) 20:40, 16 November 2025 (UTC)[reply]

It's describing what this page is about in a nutshell. Or, in other words, it's a tl;dr of the page's content. Nil🥝 21:14, 16 November 2025 (UTC)[reply]

Guidelines for commercial products or services: buses

[edit]

Recently user @Sugar Tax nominated a number of articles about local UK bus routes for nomination (examples [1], [2],[3]). Some have now been closed as no consensus (I've asked for reviews) and I'm assuming that the argument is that they have received WP:SIGCOV. The problem seems to me to be twofold: firstly, all the coverage as I see it is WP:ROTM: service is introduced, service is withdrawn, sometimes there is small scale opposition to the withdrawal. Secondly, if these were companies, they'd fail WP:CORPTRIV, and as most of the news sources are hyper-local (either newspaper websites for a town/small city or the local section of the BBC website), also WP:AUD, but because they're a service (sometimes operated by more than one company over the years) it seems these guidelines don't apply.

I can see circumstances where a bus route might be notable - I've got a book on my shelf entirely about a certain route which is also reported to be the location where George Harrison passed his audition for the Beatles, or if a route had some lasting impact on an area's fortunes, or introduced some new product or practice - but the sources in these articles simply show that these routes exist or existed. Do we need more specific guidelines or am I on the road to nowhere, pun intended? Orange sticker (talk) 21:34, 19 November 2025 (UTC)[reply]

This sounds like a very similar problem we have with airline destinations or airport connecting cities, which are mostly frequently sourced to first party sources. They should fall undrr NCORP, as well as NOT concerns. Masem (t) 21:43, 19 November 2025 (UTC)[reply]
I've just found WP:BUSROUTE but it's only an essay. I would also argue that locals submitting a petition to save a bus route is very WP:DOGBITESMAN. Orange sticker (talk) 22:06, 19 November 2025 (UTC)[reply]

Interviews

[edit]

At pages like Teahouse and the AFC help desk, I regularly see people told "Interviews don't count towards notability; they are not independent". When and where did we decide this?

It seems to me that there is a vast difference between, at one end of the spectrum, the kind of churnalistic article that is simply a string of quotes from the subject, in a Q&A format (which certainly is not independent) and, at the other extreme, an in-depth profile of the kind published by more serious journalists and publications, which is made up of an in-depth analysis in the journalist's words, supplemented by quotes from the subject explaining their beliefs or motivations. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 11:43, 26 November 2025 (UTC)[reply]

There's WP:INTERVIEW, but from prior discussions there doesn't appear to be consensus on the question of whether interviews are independent. -- LCU ActivelyDisinterested «@» °∆t° 12:43, 26 November 2025 (UTC)[reply]
Yes, there are different types of interviews and that should be context dependent. In some interviews much "third party" or "independent" information is even in the questions themselves. Also, its silly to say in a serous outlet interview, that an independent party has not taken notice, obviously they have or there would be no interview. Alanscottwalker (talk) 12:49, 26 November 2025 (UTC)[reply]
The problem is that sometimes major publications interview people who otherwise aren't notable because they did a thing that the reporter finds interesting. voorts (talk/contributions) 14:25, 26 November 2025 (UTC)[reply]
An interview alone should not be able to confer notability, but if there are other appropriate sources that provide some significant coverage, and the interview (particularly based on the weight of importance and independence of the interviewing source) helps lend more weight to that, the interview should not be discounted from notability.
The problem that I think some are concerned about is that there are business trade magazines that one can pay to be interviewed, which makes the interview there non-independent. Masem (t) 14:35, 26 November 2025 (UTC)[reply]
WP:BLP1E should weed those out. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 14:42, 26 November 2025 (UTC)[reply]
Just one source does not do much of anything, but it does not follow from that, that you should exclude that source when looking over all the sources. Alanscottwalker (talk) 17:00, 26 November 2025 (UTC)[reply]
The problem is that the dividing line is utterly subjective. For my part, I'm quite comfortable with interviews not conferring notability; if a subject is genuinely notable, there will be significant coverage in third-party sources that aren't interviews. Ravenswing 14:49, 26 November 2025 (UTC)[reply]
Many of our criteria for sources and notability are utterly subjective. We seem to manage.
Your latter point appears to be circular. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 15:06, 26 November 2025 (UTC)[reply]
No more subjective than sigcov, etc. Alanscottwalker (talk) 17:02, 26 November 2025 (UTC)[reply]
Most interviews that are worth anything start with at least an introduction, a bit about "GMG is big and strong and very handsome" before it goes into the transcript. That bit is presumably vetted for accuracy based on the reputability of the source. Also, again, based on the quality of the source, they presumably wouldn't let someone outright lie. Responsible journalism is supposed to entail keeping your subject within the realm of the real world.
This is different than low quality pieces, especially industry, where interviews are often treated as a way to produce content without much work, the subject can say whatever they want because it's just an avenue for advertising, and the bio itself may just be pulled from the subject's website. GMGtalk 17:05, 26 November 2025 (UTC)[reply]
In reality, the real answer must be it depends, as interviews lie on a spectrum (from a quote [or serious of quotes] on a topic that the subject is familiar with to a piece where the subject themself is the topic). But the real challenge is that, correctly, we do not define how many sources (or type of sources) is sufficient to create a stand-alone article. We define notability as something "worthy of notice" (and that being worthy is receiving "significant coverage in reliable sources"). Importance is not a criteria. Because we have this fuzzy concept of notability, we run into this question of whether an interview "counts" only when the subject is of borderline notability (and in that case, perhaps it is better to err towards WP:NOPAGE). --Enos733 (talk) 17:12, 26 November 2025 (UTC)[reply]
"we run into this question of whether an interview 'counts' only when ..."—But we don't. We make blanket statements such as that paraphrased in my OP, even when we don't know the intended subject. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 17:41, 26 November 2025 (UTC)[reply]
I suspect that this nutshell edit by @Bearcat ("a person does not pass GNG if interviews are the only kind of sourcing they have") has added to the confusion. The situation is more complicated than just "Nope, non-notable".
One of the structural problems is that editors (especially us old hands) are more interested in plain old print. It needs to be text, and it needs to be something I can search with ⌘F. An interview transcript might get printed in a magazine, but back in the day, that was a novelty or a single column in a publication that was filled with everything else.
Interviews are the form used by default for radio, television, podcast, and video formats. If we say "Nope, a huge number of interviews is just not notable", we're saying that the format of the source determines whether it 'counts'. That's not what we want to do. WhatamIdoing (talk) 19:13, 26 November 2025 (UTC)[reply]
I, and I think many of the people I referred to in my OP, are thinking of print or website interviews, rather than audio-visual. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 20:47, 26 November 2025 (UTC)[reply]
What I mean Madonna is not notable because of an interview that she gave. There are plenty of GNG sources about her. The only time we have concerns about whether an interview should be considered for notability is when the subject is of borderline notability. - Enos733 (talk) 21:49, 26 November 2025 (UTC)[reply]
I agree with you that disputes only arise in the more difficult cases. After all, editors are usually trying to find out "notable or not?", and when we can do that with sources that are easy to evaluate, then it'd be inefficient to spend time and energy evaluating and classifying the other sources.
However, in principle, a source should count the same regardless of whether there are other sources, and that matters for the policy writing aspect. I don't think that Madonna's most famous interview demonstrates notability (for her, or anything else), but that's because of the contents of the interview. I'd hold that view no matter what other sources did/didn't exist. WhatamIdoing (talk) 04:37, 27 November 2025 (UTC)[reply]
I would say the interview material does not count for notability, but in more reputable outlets they often give an extended 'introduction' to the interview background independent from the interview. If that is long enough, I would count it for notability, but not the interview part. PARAKANYAA (talk) 20:46, 26 November 2025 (UTC)[reply]
@PARAKANYAA, please look at the first row of the table below. Then consider:
  • If the history professor writes a one-page article in a history journal about the battle, then that's evidence that the battle is notable.
  • If the history professor writes a one-page article in a magazine about the battle, then that's evidence that the battle is notable.
  • But if the history professor says the same things on a radio show, then that's not evidence that the battle is notable?
Whether a reputable publication is written or spoken should not make any difference. WhatamIdoing (talk) 21:33, 26 November 2025 (UTC)[reply]
I feel like radio shows are a poor example because you cannot go back and fact check that like you can with writings. I wouldn't think that to be as reliable. With a documentary you get multiple takes and edits so that would be another thing. But the medium of radio does not really lend itself to fact checking, it is more off the cuff, no? PARAKANYAA (talk) 21:37, 26 November 2025 (UTC)[reply]
No, not really. You've raised three points, so I'll address them separately:
  • Most radio shows (i.e., in the present century) are recorded and archived. The recordings can be used to check the contents for WP:V purposes. If it wasn't (or if the recording was kept private), then the source would fail our WP:Published#Accessible requirement, but that's unrelated to it being an interview.
  • Not all radio shows are live broadcasts, and even for 100% live broadcasts, there are frequently semi-prepared bits. For example, if I were interviewing you live, I'd want to ask about your experience and comfort with this, and give you some idea of what to expect (for example, maybe I'd tell you: "I'll introduce you and say you're a Wikipedia editor. We'll start easy: I'll ask you about how you got started editing and whether you think Wikipedia is important to the world. After that, I'll ask about what Jimmy Wales said last week, and whether Elon Musk really hates Wikipedia, and we'll see where it goes from there. If we run out of things to talk about, I'll ask if you have a favorite story about Wikipedia, so come prepared to tell one if we need to. When you hear the background music start, we've got 30 seconds left."
  • Fact-checking (i.e., by the source, not by Wikipedia editors, who have no business doing that) is a desirable thing in (most) sources, but it's not a requirement for notability, or even for reliability. Additionally, particularly with Gotcha journalism, fact-checking does sometimes happen in live radio and television interviews. Some interviews are a long series of "Do you deny that you ever said this?" "Yes! Of course I deny saying that!" "Here's a video of you saying this. Now, do you still deny that you ever said this?" But most fact checking is both more mundane (how to spell the interviewee's name) and unnecessary for recorded interviews (whether they actually said that thing).
WhatamIdoing (talk) 21:57, 26 November 2025 (UTC)[reply]

I've wondered occasionally whether some examples like this might help editors reach a shared understanding:

Examples of some interviews
Source Interviewee Topic of discussion Notability?
10-minute segment on a radio show A history professor from Little University 100th anniversary of a battle in the broadcast area checkY Yes for the battle – equal to a newspaper article saying the same thing in prose

☒N Not for the radio show

☒N Not for the professor

A Current Affair (Australian TV program) John Hewson Whether his tax proposal would make a birthday cake cost more or less checkY Yes for the tax proposal – equal to a newspaper article saying the same thing in prose

☒N Not for the interview itself (the post-broadcast media coverage of the interview made the interview notable)

Late Show with David Letterman Madonna Madonna herself and her books ☒N Not for Madonna – all primary

☒N Not for the interview itself

60 Minutes Betty Ford Ford herself checkY Yes for Ford – even though the interview was carefully selected to be sympathetic to her and Ford's goal was to promote herself

☒N Not for the interview itself

The Rolling Stone Interview series Various authors/creative people The interviewees, their works, their views, etc. checkY Yes for the interviewees – Long interviews demonstrating a significant amount of attention to the subject and preparation by the interviewer
Man on the street interview Random person Usually current events or personal opinions ☒N Not for anything (no significant coverage)
Corporate newsletter CEO whatever the CEO wants to say about the company ☒N Not for anything (self-published, non-independent)
Business magazine CEO whatever the CEO wants to say about the company ☒N Not for anything (all primary)
interviewer leads CEO through a comparison of that company's marketing strategy vs other companies' marketing checkY Yes for the company

Question? Maybe for the other companies (if SIGCOV) ☒N Not for the CEO

interviewer leads CEO through analysis of how the CEO's life experiences and management style shaped their career checkY Yes for the CEO – even if it's all positive or "softball"

☒N Not for the company

Do you think that would help? (It could go in Wikipedia:Interviews.) Would you change anything? WhatamIdoing (talk) 20:25, 26 November 2025 (UTC)[reply]

I do not think the source column makes sense - as the topic of discussion is the most important aspect to determine what is notable. Certainly, there is a difference between an interview on 60 Minutes versus a segment on a radio show or quotes in a Wall Street Journal article, but for clarity to editors, the question needs to be why the person was interviewed. Was the person interviewed to talk about how they use AI, or about their company's marketing strategy, or was the person interviewed to talk about their life experiences? And secondly, what is the depth of the published interview (is the subject the feature of the published piece, an example, or an anecdote). - Enos733 (talk) 20:39, 26 November 2025 (UTC)[reply]
The WP:SOURCE (which includes both the publication/interviewer and the interviewee – the first two columns here) determines whether the interview could be a reliable source at all. "Teenager interviews friend for school project" is never going to demonstrate notability, nor is a reporter doing random man-on-the-street interviews.
I agree with you that the topic of discussion is the key point, because you could interview Alice Expert about Alice herself (notability points to the BLP on Alice) or you could interview Alice about her area of expertise (notability points to the area of expertise, but not to Alice herself). As a real-world example of that, tech analyst Rob Enderle was famous for getting quoted in tech news, largely because he was very quick to return calls from reporters who were working on short deadlines. The resulting articles demonstrated notability for whatever he was talking about, not for himself. (The reason we're stuck with an article on him is because the The Mercury News wrote a whole article about him, not because it'd be possible to fill pages with "On <date> he was quoted in the <publication> as saying that Apple might fail".)
I think the content itself is also relevant. For example, both Madonna and Betsy Ford were interviewed on television, but I think only the Ford interview could contribute towards notability. The Madonna interview didn't have any useful substance. WhatamIdoing (talk) 22:14, 26 November 2025 (UTC)[reply]
My point is largely that if we do add a table to WP:INTERVIEWS, we should try to hold the source column consistent, as editors are generally better about evaluating the general appropriateness of a source. But, if we have source "60 Minutes" and an topic of discussion a in-depth interview, I don't want a table to leave the impression that the quality of the source is the defining feature ("if it was in 60 Minutes it must contribute to notability"). Some of the segments in 60 Minutes that include a clip of an interview are more akin to interviewing a history professor (subject matter expert) and would not contribute to notability of the interviewee. - Enos733 (talk) 22:35, 26 November 2025 (UTC)[reply]

Third opinion invited

[edit]

As WP:NJOURNAL has been dismissed in toto as an essay, comments would be welcome at Wikipedia:Requested moves/Technical requests § Quantitative Biology Paradoctor (talk) 13:06, 15 December 2025 (UTC)[reply]

Because

[edit]

The description after the GNG says, in part: ""Sources" should be secondary sources, as those provide the most objective evidence of notability".

I wonder whether the "as those provide the most objective evidence of notability" part is needed. I doubt that it means literally what it says ('analysis, evaluation, interpretation, or synthesis'-oriented sources 'provide the most objective evidence' of 'a test used by editors to decide whether a given topic warrants its own article'? – that seems unlikely). I suspect that omitting these words would do no harm to the actual meaning. WhatamIdoing (talk) 00:24, 21 December 2025 (UTC)[reply]

This whole description is circular reasoning. Secondary sources provide evidence of notability because we have defined notability to mean the existence of secondary sources. —David Eppstein (talk) 00:43, 21 December 2025 (UTC)[reply]
It is circular, but I think it sufficiently communicates that we view secondary sources as having more weight for the lasting encyclopedic relevance of a subject than primary sources. We could replace "evidence of notability" with "indication of lasting encyclopedic relevance" or "evidence of importance" or something, but I'm not sure if that would be any more understandable. Vermont (🐿️—🏳️‍🌈) 03:51, 21 December 2025 (UTC)[reply]
I'm just not sure that it adds anything. Why not "Sources should be secondary", without the non-actionable explanation of why they should be secondary? WhatamIdoing (talk) 21:08, 21 December 2025 (UTC)[reply]
It could say “there must be secondary sources”, except that there are some tolerated topics, eg London bus routes, that are not bound to WP:N. SmokeyJoe (talk) 01:06, 24 December 2025 (UTC)[reply]
The “action” is AfD. “Why” is an explanation, not an action. SmokeyJoe (talk) 01:07, 24 December 2025 (UTC)[reply]
That "notability" is more likely meant to be read as real-world notability than WP-notability. So not so much circular, but mixing the definitions. Masem (t) 04:14, 21 December 2025 (UTC)[reply]
We should avoid mixing the definitions, but either way, I'm thinking that it's enough to say "Sources" should be secondary sources, and then take Churchill's advice about when to stop talking ("Say what you have to say and the first time you come to a sentence with a grammatical ending – sit down"). WhatamIdoing (talk) 04:40, 21 December 2025 (UTC)[reply]
I'd remove all reference to "secondary" completely. What Notability really wants is significant coverage in reliable sources that are independent of the subject. "Secondary" does not mean any of those things; at best, when we say "secondary", we mean "probably has the kind of analysis and interpretation we need to write a good article without violating WP:OR", and at worst we just mean "good". Anomie⚔ 13:17, 21 December 2025 (UTC)[reply]
We need secondary here. We don't want simply primary sources (which can include reliable, third-party sources but just repeating elements of the primary source w/o comment). WP's definition of secondary is based on the transformative information like analysis and opinion that we need to show that reliable sources consider the topic of some type of significance to give more coverage than just rote repetition of the primary source. Masem (t) 15:40, 21 December 2025 (UTC)[reply]
Wikipedia's definition of secondary sources, according to who? Why would we use a definition of secondary sources different from the entire world's? If we are going to be meaning an idea different entirely from what the reader will assume when they hear secondary, why not use a different word? PARAKANYAA (talk) 17:28, 21 December 2025 (UTC)[reply]
Our definition of secondary is based on the more common academic version of how sources are catalogued. And WP:PSTS describes this pretty well for a layperson that might not have heard this before. Masem (t) 19:36, 21 December 2025 (UTC)[reply]
In practice, our definition of secondary is subject to the opinion of participants of an AfD on whether they think the subject is important. If so, the secondary nature of the sources is taken for granted. If not, sources are put down as primary or local or routine. It's a way of pretending we're being objective when we're not. —David Eppstein (talk) 19:40, 21 December 2025 (UTC)[reply]
AFD is its own mess, where it feels most of the time its vote counting rather than the admin reviewing on the basis of policy and guidelines. Basically, AFD nowadays for under the radar articles is basically who shouts the loudest wins. I wouldn't use that to say that's practice, because that's a problem that needs fixing itself. Masem (t) 19:51, 21 December 2025 (UTC)[reply]
But what is notability for, beyond "can this survive an AfD"? —David Eppstein (talk) 20:14, 21 December 2025 (UTC)[reply]
Wikipedia-notability is an excellent predictor, not perfect, of whether the topic will pass AfD. SmokeyJoe (talk) 01:14, 24 December 2025 (UTC)[reply]
David wrote “our definition of secondary is subject to the opinion of participants of an AfD”. No, AfD participants don’t try to redefine terms, as far as I’ve ever seen, but instead decide whether to ignore WP:N. SmokeyJoe (talk) 01:14, 24 December 2025 (UTC)[reply]
So... you may have noticed that the GNG criterion itself ("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 contain the word secondary. That word appears only in the bullet points:
  • ...Availability of secondary sources covering the subject is a good test for notability.
  • "Sources" should be secondary sources...
I've wondered occasionally whether replacing that with a description/definition would be both more informative (probably yes) and still have support (dunno about that).
Specifically, I wonder whether, if editors knew that "Sources" should be secondary sources... means the same thing as "Sources" should contain analysis, evaluation, interpretation, or synthesis of the facts, evidence, concepts, and ideas taken from primary sources... if we might discover that editors didn't actually want secondary sources in the first place.
I think what would actually happen is that editors would discover just how low the bar can go for a secondary source:
  • "Big Business, Inc.'s earnings were up compared to the previous quarter" is a secondary source, because it provides "analysis" (a simple comparison).
  • "Our testing showed that the blue-green widget is appropriate for low-temperature environments" is a secondary source, because it provides "evaluation".
  • "When Trump wrote covfefe, he meant coverage" is a secondary source, because it provides interpretation.
  • "Companies in the widget industry reported higher earnings last quarter but warned of weakening demand, with our analysis suggesting a sector-wide decline of about 10% next quarter" is synthesis, because it combines facts from multiple sources.
but I could be wrong, and they might decide that they weren't actually looking for analytical, evaluative, etc. content at all. WhatamIdoing (talk) 20:57, 21 December 2025 (UTC)[reply]
IMO, what they're really looking for is "did other people whose opinion we care about, who are not involved with it, think it was important enough to take notice of?". Or they're just looking for some way to support their preconceived notion of whether or not it should have been taken notice of. Anomie⚔ 21:12, 21 December 2025 (UTC)[reply]
Wikipedia's definition of secondary sources ...but not the definition that was in place when the word secondary was added to the GNG.
The GNG began life as a single sentence – "They [several guidelines] generally assert that a minimum standard for any given topic is that it has been the subject of multiple non-trivial published works, where the source is independent of the topic itself" – in September 2006. The word secondary does not appear anywhere on the page. The page was tagged as a guideline 13 days later,[4] and whether it was actually a guideline was fought over for the next couple of months (example).
@JzG introduced the "secondary" language in October 2006, saying first that we needed "independent, neutral sources"[5], and then immediately changing it to say "multiple reliable secondary sources independent of the subject"[6]. It links to WP:RS, which at that time provided two relevant sections: #Some definitions and #What is an independent secondary source?. The latter, in particular, is worth looking at, because the definition is non-self-published (what's that got to do with independence or secondary-ness?), the author not "collaborating" in the subject/event, and that they only "may have" relied on primary sources (and so only "may" be secondary sources). In other words, the definition of "independent secondary source" at the time looks a lot like "is WP:PRIMARYNEWS".
@Uncle G merged in User:Uncle G/On notability on 20 November 2006. This marks the beginning of the guideline that we're familiar with, complete with three bullet points to define words in the GNG criterion. It also re-labels the proto-GNG criterion as ==The Primary Notability Criterion==: "a subject must have been the subject of multiple non-trivial published works, whose sources are independent of the subject itself." There are some bold edits to the lead (examples: [7], [8]), and an insistence that the definition – "A topic is '''notable''' if it has been the subject of multiple, non-trivial published works whose sources are independent of the subject itself" – be the first sentence,[9] As of the end of the year, the word secondary still does not appear anywhere on the page.
What is now the GNG was renamed "The common notability criterion" in February 2007,[10] was switched back to "primary notability criterion", and finally landed on "general notability" on 26 March 2007.[11] By that point, the GNG criterion was "A topic is notable if it has been the subject of at least one substantial or multiple non-trivial published works that are reliable and independent of the subject", and there were six bullet points providing definitions, but the word secondary appears only in a (since removed) section called ==Notability is not popularity==, and nowhere in the GNG.
WP:PSTS on that day provides this example of a secondary source: "A journalist's analysis or commentary of a traffic accident based on eye-witness reports is a secondary source". We'd call that WP:PRIMARYNEWS these days.
Secondary was introduced a month later,[12] the day after this discussion about primary vs secondary sources started. In the discussion, you can see editors struggling with the idea that Wikipedia:Secondary does not mean independent (e.g., "Second, changing "published works" to "secondary sources" makes the part about the sources being "independent of the subject" redundant. Secondary sources are, by definition, independent of the subject", "You and I know offhand that "secondary" includes by definition "independent", but some people don't"). Specifically, in this discussion, secondary is proposed as a replacement in the GNG for published. There was also a simultaneous discussion noting that WP:NOR explicitly authorized the creation of articles using only primary sources (and had done so for at least a year, if memory serves). There were also comments such as "Why must these be secondary sources? Some primary sources are legitimate" and "As long as independent is there, that's the important part - independent primary sources can establish notability". In short, I think that there is no good reason to believe that editors, when they put the word secondary into the GNG, actually meant the same thing that we mean now.
It may be useful to keep in mind that the GNG wasn't some widely accepted touchstone back then. There were still editors arguing that it was okay to have articles about anything that seemed important, even if there were no sources. WhatamIdoing (talk) 20:25, 21 December 2025 (UTC)[reply]
Thank you for the in depth analysis. 🙂 IMO this further goes to show that throwing "secondary" around into various policies and guidelines these days is more about giving people a legitimate-seeming way to say "WP:PRIMARYBAD" than anything necessary to define policy. Anomie⚔ 20:52, 21 December 2025 (UTC)[reply]
Strongly disagree, secondary has been tied to notability for nearly two decades, and it has precise meaning. We simply don't want articles composed of only primary sources which do nothing to transform the information. This for example is why we have a NOTNEWS problem, because editors rely too much on primary sourcing in these areas and do not wait to see what the longer term significance of the event may be. Or to take another example, NSPORTS used to say that any player that played one pro game was considered notable, but the bulk of such areas were just primary-sourced biographies which was harming the encyclopedia. We need that secondary aspect to help establish why the topic is important just beyond the data that one can write about it.
Now, I will agree that how much something is secondary can have degrees. 99% of an article could be rote primary and then end with a secondary sentence. Or that some articles can be primary for one topic and secondary for another. But key to having significant coverage that is independent of the topic itself is nearly all the time going to be from secondary sources. Masem (t) 21:51, 21 December 2025 (UTC)[reply]
Masem, what/who are you strongly disagreeing with?
Yes, the word secondary was added to the GNG about 18.6 years ago. It has had multiple meanings during that time, and at any given time, editors use a variety of meanings (some of which could fairly be described as "precise", though these precise definitions are not always fairly described as "accurate").
I think everyone agrees on that? WhatamIdoing (talk) 23:20, 21 December 2025 (UTC)[reply]
“Secondary source” is an excellent fit. In early versions, it was not written perfectly. SmokeyJoe (talk) 01:17, 24 December 2025 (UTC)[reply]
Sure, you made such a proposal back before it was called the GNG. It didn't get adopted, but I admire your consistency. WhatamIdoing (talk) 06:00, 24 December 2025 (UTC)[reply]
Interesting to note here that in this discussion, the idea that news coverage can be a secondary source is taken for granted (It also needs to say "and independent of each other", to weed out, for example, 20 papers republishing the same Associated Press column-that's one source (the original AP piece), not 20). Katzrockso (talk) 08:14, 24 December 2025 (UTC)[reply]
What we have today is close to what I think it should be. I think the GNG would be improved if we could flesh out what is meant by “significant”, and if we could recognise that for secondary sources, it is reputability not reliability that matters first. SmokeyJoe (talk) 10:36, 25 December 2025 (UTC)[reply]
I've been hoping to get a definition of SIGCOV in this guideline for years. I'm willing to help with another attempt if folks think it's got a chance. WhatamIdoing (talk) 18:23, 25 December 2025 (UTC)[reply]
SIGCOV.
You have mentioned elsewhere, 500 words. I’m not sure the 500 words all needed to be directly about the topic? You’ve also mentioned the concept of an ability to write something, based in the sources.
100 words, WP:100W, is the shortest serious proposed threshold for SIGCOV that I remember.
Sometimes, I’ve fallen back to “two running sentences that contain adjectives” as a pretty extreme low end for SIGCOV.
I maintain that the SIGCOV measure must be secondary source material. If the best coverage is data or statistics, these topics are reliably deleted at AfD.
I wonder, does anyone have data from AfD where the decision hinged on SIGCOV.
The least SIGCOV I remember was for the maternal grandmother of a Roman politician. It had names, but no further information, and none of it was secondary, and it was kept. SmokeyJoe (talk) 21:23, 25 December 2025 (UTC)[reply]
How much coverage: If the goal is to write a non-stub Wikipedia article, then you need to be able to write about 300 words. (This is also approximately the length of the median Wikipedia article.)
Given that encyclopedic style favors concision instead of verbosity, it would be difficult to take 100 words of prose (i.e., counting only those parts of a publication that are relevant/usable in an article) and turn it into 300 words of encyclopedic content. I think therefore that a single WP:100W source would usually prove inadequate to the goal. (100 facts in a non-prose format might be feasible, as the two words "Birthplace: London" turns into the five-word sentence "He was born in London".)
A couple of 100W sources might work. I think that if you get 500 words (added together, though not just repeating the same things as the others), that you'll probably be able to write a decent article.
What kind of coverage: I generally agree that secondary source material should be required. NPROF seems to get along without it, as does NSPECIES, but the absence of secondary source coverage restricts what you can write in an article (e.g., without secondary sources, you can write "This pinky-russet beetle is 2.2 cm long", but you can't write "At 2.2 cm, this pinky-russet beetle is shorter than Scarabaeus ille, and it is a more mauvy shade of pinky-russet than Scarabaeus alius").
However, SIGCOV ("a lot of coverage") and SECONDARY ("analytical or evaluative") are separate considerations. The GNG requires both, and neither having a tiny bit of secondary source material nor having a huge amount of primary source material is sufficient. The tiny secondary source fails because it's not SIGCOV, and the huge primary source fails because it's not SECONDARY. WhatamIdoing (talk) 22:54, 25 December 2025 (UTC)[reply]
How much coverage? Why require non-stub? Stubs are welcome. Even permastubs. Aren’t they? As long as they aren’t CORP topics. Ancient named people, and species are very welcome stubs.
Should we document that SIGCOV is field-dependent? SmokeyJoe (talk) 23:22, 25 December 2025 (UTC)[reply]
WP:NOPAGE says "Sometimes, when a subject is notable, but it is unlikely that there ever will be a lot to write about it, editors should weigh the advantages and disadvantages of creating a permanent stub". Also, WP:WHYN: "We require "significant coverage" in reliable sources so that we can actually write a whole article, rather than half a paragraph or a definition of that topic. If only a few sentences could be written and supported by sources about the subject, that subject does not qualify for a separate page". (Of course, since I wrote most of WHYN once, I see no inherent reason why it couldn't be re-written again, assuming the consensus about that point has changed since it was originally discussed and accepted.)
Mostly, though, I'd say that we should gently discourage (but not ban) doomed permastubs because permastubs irritate editors. Think about how much of the NSPORTS and NOLY discussions amounted to "These are too short". The opponents of WP:NSPECIES frequently complained about articles being too short for their tastes. WhatamIdoing (talk) 02:33, 26 December 2025 (UTC)[reply]
I don’t think permastubs are a problem, when separated from the problem of mass article creation. SmokeyJoe (talk) 03:34, 26 December 2025 (UTC)[reply]
I know that permastubs don't bother you. They don't usually bother me, either. But they do bother some other editors, and those editors respond with complaints, mass deletion attempts, time-consuming proposals, claims that the mere existence harms Wikipedia's reputation, etc. Just the sour attitude emanating from some of the more immediatist of them (How dare nobody WP:VOLUNTEER to fix this problem that I refuse to fix myself?!) is a problem for social cohesion. We need editors to be unified in fighting against ignorance, rather than divided by whether a two-sentence stub that almost nobody reads[1] is a problem.
[1] I did some massviews work a while ago. The median Wikipedia article gets one page view per week. It's not physically possible for these ultra-low-traffic articles to have a statistically significant effect on readers' views of Wikipedia. WhatamIdoing (talk) 06:47, 26 December 2025 (UTC)[reply]
Do we share disapproval, even disdain, for editors that expect others to actively support their views. I am most ambivalent about the MOS-aficionados. Yes, it’s great that Wikipedia has a style guide, but is it going to far to impose a style guide more firmly than elsewhere in the English speaking world?
As I keep mentioning, I’m very permissive of history and the natural sciences, but these would fail the permastubs test, due to further sources being always possible to appear or be created.
Sports bio-stubs are complicated to me. They include many BLPs, taking a very narrow snapshot of a young person’s life. But they only repeat reliable sources for what can easily be found anyway, so is it a problem?
I have taken on what I thing is the dominant aversion against use of Wikipedia for commercial promotion, which is reflected in the harsh wording of WP:CORP, and the corresponding firm treatment that WP:CORP articles receive at AfD.
Talking through this, I’m not seeing any good reason for SIGCOV to be particularly deep. For topics that we don’t like, promotion and native advertising, it’s the independence test, of the two GNG secondary sources, that serves to exclude unliked topics. Two running sentences containing qualitative adjectives, commenting on the subject, feels good enough, as long as the authors are completely independent of the subject and each other. SmokeyJoe (talk) 08:17, 26 December 2025 (UTC)[reply]
WP:VIEWSSTATS disagrees with the data from here, according to which the average median page gets roughly 700 views per year. I think it is due to the Quarry list containing redirects. For example, among the first 10 links, 8 are redirects, and they skew the views down. Kelob2678 (talk) 00:16, 27 December 2025 (UTC)[reply]
@Kelob2678, I refer you to How to Lie with Statistics, chapter 2, "The Well-Chosen Average". The mean average is very different from the median average and the mode average. WhatamIdoing (talk) 00:21, 27 December 2025 (UTC)[reply]
I meant the median page, not the average. English Wikipedia had 133B views over the last year across all pages, assuming that at least half of those views are to the 7M pages in the mainspace, each page got 9,500 views on average, bots included. Kelob2678 (talk) 00:39, 27 December 2025 (UTC)[reply]
I don't think we want to include bots when looking at traffic, especially with the devs reporting unusually high levels of bot activity this year. WhatamIdoing (talk) 01:08, 27 December 2025 (UTC)[reply]
Roughly one-third of all views were from bots. The point is that the average number of views per page is an order of magnitude larger than the views for the median page, which is itself an order of magnitude larger than the number listed at WP:VIEWSSTATS. Kelob2678 (talk) 08:49, 27 December 2025 (UTC)[reply]
Here are the page views for a list of 10 articles. The mean is 100/week, but the median is less than 10/week. All of these are non-redirect articles.
Would you like to re-run the VIEWSSTATS numbers with another random set? I'd suggest doing that some time after the New Year, so we can have 2025 numbers. Articles created/deleted this year should be filtered out, since it's not fair to calculate page-views-per-year on an article that might have only existed for a few years. Massviews was unhappy with the dataset size, so I had to run it in batches. Other than that, it's an easy thing to do. WhatamIdoing (talk) 18:42, 27 December 2025 (UTC)[reply]
The median is 446.5 annually, which is significantly larger than 1 per week. I don't think there is a need to recalculate, as the dataset I linked above contains annual views in 2024 for 6,961,382 pages, which is more complete than the Quarry method. The page ranked 3,480,691 is List of teams and cyclists in the 2006 Tour de France, which received 722 views in 2024. They also allow downloading it, so newly created pages can be filtered out, but I don't think this would change the number by much. Kelob2678 (talk) 19:07, 27 December 2025 (UTC)[reply]
I wonder if we could get these numbers across all namespaces and in the mainspace separately for redirects and maybe Wikipedia:Disambiguation pages. WhatamIdoing (talk) 20:31, 27 December 2025 (UTC)[reply]
Can we define "secondary source material" more specifically, instead of with the vague and often argued over term "secondary source"? What's the important part: That it has analysis, evaluation, interpretation, or synthesis rather than only bare facts? That it explicitly digests primary sources? That it's "one step removed from the event", whatever that means? Personally, I'd focus on the first. But just saying "secondary source material" opens it up to people arguing that a source full of analysis, evaluation, interpretation, and synthesis doesn't "count" because it's not far enough removed from the event or the like. Anomie⚔ 14:05, 26 December 2025 (UTC)[reply]
I agree with you about the "analysis, evaluation, interpretation, or synthesis" being the key part.
One of the problems with this is that editors don't always realize how small that analysis, evaluation, etc. can be. "X is bigger than Y" is analysis. "The theme of this art is the struggle of man vs nature" is interpretation. WhatamIdoing (talk) 18:38, 26 December 2025 (UTC)[reply]
Can we define, and use, “secondary source material”. Absolutely, I recommend it. The topic here is quite niche. It is only about source typing for WP:GNG, WP:N or WP:PSTS purposes. A narrow application, but quite serious, because article deletions decisions hang off it.
I like “secondary source material” because it is more important what material is being used, and less important the source typic of the source as a whole.
Several of the academic definitions are wrong, where they assert or lean to asserting that a source is a primary source or a secondary source. Very often, it changes from paragraph to paragraph. For example, in an interview, it very often begins with secondary source material in the introduction, which is followed by primary source material in the interview-proper. These are often very easily distinguished, by tone, perspective, formality of the writing, even tense sometimes.
But please do not abbreviate “secondary source material” to SSM. Not without it being the (singular) shortcut of an essay.
“Secondary source material”: “That it has analysis, evaluation, interpretation, or synthesis”? Yes. I like to include “comment” and “contextualisation”. Anything transformative of the facts.
“rather than only bare facts”? Yes. But a few bare facts doesn’t prevent the whole from being a secondary source. A source can be both primary and secondary, depending on how it is being used.
“That it explicitly digests primary sources”? Yes, but I’d drop the “explicitly”. We are getting away from encyclopedic purposes, but a poem will be definitely a secondary source, digesting primary source material with clever subtlety. “Encyclopedic” generally means “in formal boring writing style”, although it doesn’t have to.
"one step removed from the event" is straying from historiography towards journalism, and second hand sources, so be careful, but it is generally true. A common illustrative example of an eyewitness account of an event, eg “Tom reported seeing Jerry violently and callously damage Alex’s property”. Violently and callously are subjective judgmental terms, which makes the statement secondary source content about the damage done. However, with the passage of time, better sources will judge the style of damaging, and the source will become mostly only usable for a primary source material on the opinion in the moment of the eyewitness called Tom.
"secondary source material" opens it up to people … “. I definitely agree.
”… arguing that a source full of analysis, evaluation, interpretation, and synthesis doesn't "count" because it's not far enough removed from the event or the like”. This is complicated. I’d want to process the example directly. People arguing “count” is a warning flag for someone oversimplifying.
- SmokeyJoe (talk) 23:37, 26 December 2025 (UTC)[reply]
Would you accept changing the current text:
  • "Sources" should be secondary sources, as those provide the most objective evidence of notability.
to this?
  • "Sources" should contain some secondary source material, such as analysis, interpretation, evaluation, contextualization, or comparison.
WhatamIdoing (talk) 23:49, 26 December 2025 (UTC)[reply]
How about Sources should contain material that analyzes, interprets, evaluates, contextualizes, or compares the topic.? Leave out the invitation for people to argue over historiography versus journalism and so on. Anomie⚔ 01:12, 27 December 2025 (UTC)[reply]
Material doesn’t analyse, authors analyse. SmokeyJoe (talk) 01:30, 27 December 2025 (UTC)[reply]
Instead of nitpicking every word choice, why not suggest improvements? 🙄 Anomie⚔ 01:45, 27 December 2025 (UTC)[reply]
Already done.
”Sources” should be secondary sources.. WAID’s #2 below.
Leave it to the article, and it’s sources, to explain what it means to be a secondary source.
Wikipedia should not go down the path of rewording real world definitions. SmokeyJoe (talk) 02:09, 27 December 2025 (UTC)[reply]
The need for independent sources comes from WP:V, not WP:GNG. Guy (help! - typo?) 13:14, 25 December 2025 (UTC)[reply]
I think it's the other way around. WP:V is about how to verify a particular sentence/claim, and that can be done with non-independent sources. The WP:GNG has always been focused on independent sources. WhatamIdoing (talk) 18:13, 25 December 2025 (UTC)[reply]
Core principle of WP:V has been for a long time "If no third-party sources exists, WP should not have a topic about it." Masem (t) 18:29, 25 December 2025 (UTC)[reply]
...in a section that exists only to summarize WP:N. WhatamIdoing (talk) 19:01, 25 December 2025 (UTC)[reply]
If a reliable independent source repeats a "primary" source, why exactly does that coverage not count? Why is the information somehow tainted by being previously published elsewhere, assuming the reliable source's publication really is independent? Anomie⚔ 20:52, 21 December 2025 (UTC)[reply]
You've crossed streams here, so is your question really:
  • Reliable independent source repeating a primary source (which could also be independent), or
  • Reliable independent source repeating a non-independent source?
Also, how much "repeating" is involved in this scenario? Most editors don't accept a press release reprinted word-for-word in an ordinary newspaper. Most editors do accept a bit of copying (there are only so many ways to say "won the Notable Prize") and getting information from non-independent sources (how else do you say how many employees the company has, or why the artist made that artwork?). WhatamIdoing (talk) 21:02, 21 December 2025 (UTC)[reply]
I was asking "why not?" to Masem's assertion that We don't want simply primary sources (which can include reliable, third-party sources but just repeating elements of the primary source w/o comment). Anomie⚔ 21:14, 21 December 2025 (UTC)[reply]
Part of why we want secondary sources is because WP is not simply a directory or a who's who or many other WP:NOT issues. Just because a lot of primary sources exist about a topic does not mean that topic is appropriate for WP, because that's generally not where significant coverage is going to come from; this may be most critical in business and corporations to avoid self-promotion, but applies universally across all topics. We want that significant coverage, which is going to come nearly all the time from secondary (and tertiary) sources which are going to go in depth on a topic with the work's own take of establishing the topic (using interpretation and analysis and other transformative aspects).
Lets put it the other way - I cannot imagine a case of significant coverage coming only from third-party, primary sources, that would not fail one or more of the NOT tenets. Masem (t) 21:45, 21 December 2025 (UTC)[reply]
If you can't imagine "significant coverage" coming from primary sources, then you don't need the word "secondary" in there because "significant coverage" is already part of when it has received significant coverage in reliable sources that are independent of the subject. Anomie⚔ 22:14, 21 December 2025 (UTC)[reply]
Significant coverage is a significant amount of (media) coverage. It is not coverage that says something is significant/important. Look at the original wording: "the subject of multiple non-trivial published works" and "the subject of at least one substantial or multiple non-trivial published works". This isn't about saying that the subject is important/significant; it's about the sources containing more than a passing mention.
Also, WP:N has disclaimed the "notable subjects = important subjects" thing for a very long time. WhatamIdoing (talk) 23:23, 21 December 2025 (UTC)[reply]
We want more than just repeation of primary information, which frequently happens with things like breaking news, press releases, sports results, and the like. We need sources providing significant coverage to estaish why the topic is more that just a collection of facts and how it fits in the world's knowledge, otherwise we are engaging in original research to justify having articles, which leads to navel gazing, trainspotting, and walled gardens of info. That type of info is only going to come from secondary sources that transform facts into something more. Masem (t) 18:09, 23 December 2025 (UTC)[reply]
Sources that esta[bl]ish why the topic is more that just a collection of facts and how it fits in the world's knowledge are WP:SECONDARY sources. That's separate from providing substantial/non-trivial/significant coverage. WhatamIdoing (talk) 23:53, 23 December 2025 (UTC)[reply]
Contextualisation. SmokeyJoe (talk) 21:26, 25 December 2025 (UTC)[reply]
Contextualization is usually provided by secondary sources.
If you define contextualization very broadly (and I might ;-) ), then a primary source can provide some sorts of context (e.g., time period or place for a person; taxonomic classification for a species), but that kind of context also doesn't require significant coverage, as it can be communicated simply and briefly (e.g., "born: 32 Octember 1901, London" or "Callicarpa dichotoma is in Lamiaceae"). WhatamIdoing (talk) 23:06, 25 December 2025 (UTC)[reply]
Yes. SmokeyJoe (talk) 03:30, 26 December 2025 (UTC)[reply]
Could secondary be removed by this logic? Yes, but only if all editors fully understood what significance coverage is. But I know from numerous AFDs that editors think even name dropping in sources is significant coverage. We need to be clear that significant coverage is most often going to come from secondary sources to make sure editors are on the same page. Masem (t) 18:12, 23 December 2025 (UTC)[reply]
So, in other words, you want "secondary" in there so you can say "primary is bad" instead of having to justify your definition of "significant" which excludes breaking news and such. I find that poor justification, but also par for the course when having "secondary" in policies comes up. Anomie⚔ 18:17, 23 December 2025 (UTC)[reply]
We could just say, simply, that "notability requires significant coverage", omitting "reliable" and "independent", for the same reason to remove "secondary", because I'd expect any rational editor to know that these are necessary aspects of significant coverage. But we know many editors do not read significant coverage that way, so all these terms and necessary to define what sources we evaluate when it comes to evluating norability. Making it clear that notability is best judged through what secondary sources say is part of this explanation.
I still challenge to find any topic on WP that would clearly meet WP:N based only on independent but primary source coverage. Masem (t) 20:04, 23 December 2025 (UTC)[reply]
It definitionally can't because Wikipedia defines notability as requiring significant coverage, so this challenge is incoherent. Katzrockso (talk) 20:14, 23 December 2025 (UTC)[reply]
Normal meanings of "significant" don't include "independent" or "reliable", so trying to roll those into "significant" would be creating more confusing wikijargon. Yes, all three of those already have wikijargon aspects, but that suggestion would make it worse. On the other hand, "secondary" as used here is already almost entirely wikijargon (it originated in academic definitions, but has since shifted greatly), and worse it's wikijargon that we don't all even agree on the meaning of. And I maintain that your disbelief that we could have a topic pass WP:N with only "primary" independent sources just further proves that we don't need "secondary" here for the guideline to function as intended. Anomie⚔ 21:08, 23 December 2025 (UTC)[reply]
Our definition of secondary sources is still 100% the academic version. The problem is that some take the "one step removed" to be a step removed in authorship (eg independence or third party) and not in thought (the transformation of ideas) Masem (t) 21:27, 23 December 2025 (UTC)[reply]
Secondary sources were created by someone who did not experience first-hand or participate in the events or conditions you’re researching. [13]. Here's the first definition I pulled off a search engine. Does this comport with "[o]ur definition of secondary sources"?
Here's another one I found [14] listing out what things can be listed as "primary" vs "secondary" sources. Here's a very revealing one [15] that distinguishes what a primary and secondary source is by discipline (though this page seems to be internally incoherent when providing a description of secondary sources). Katzrockso (talk) 08:30, 24 December 2025 (UTC)[reply]
All three seem to be consistent with how WP defines a secondary source as well as the academic definition. Harvard's mentions the writing being from someone not associated with the original work (the "one step removed" idea) but goes on further to explain that writing generally includes transformative actions, so supporting the "one step removed in thought". A simple recap of a sports game by a sports writer (the game being primary) is still primary, but an analysis of what the best plays in that game would be secondary. Masem (t) 13:28, 25 December 2025 (UTC)[reply]
User:Anomie wrote: “ "secondary" as used here is already almost entirely wikijargon (it originated in academic definitions, but has since shifted greatly)”
I think Anomie is plain wrong, partly confused, and challenge him to substantiate that statement.
To start with, “secondary” is not a standalone word, it has to be the term “secondary source”. SmokeyJoe (talk) 21:34, 25 December 2025 (UTC)[reply]
Looking again at the actual text at WP:SECONDARY, you're right, it derives fairly straightforwardly from the academic definitions in secondary source. On the other hand, it also acknowledges a fundamental ambiguity ("Whether a source is primary or secondary depends on context"), and many large-scale discussions I've seen that include WP:PSTS bog down in different sides asserting that their interpretation of the ambiguity is the right one and arguing past each other rather than actually trying to make use of the academic definitions. So, yes, most of the time I see "secondary" thrown around, I find it's being used as wikijargon. Also, I find abbreviation of "secondary source" to "secondary" valid usage when context is clear. Anomie⚔ 14:16, 26 December 2025 (UTC)[reply]
I do not I find abbreviation of "secondary source" to "secondary" valid usage when context is clear. Clear to who? It is not good abbreviation because it is a whole word with similar but different meaning. Jargon creates barriers, and hinders comprehension. SmokeyJoe (talk) 23:10, 26 December 2025 (UTC)[reply]
And yet you're the one pushing for more jargon like "secondary source material", with lots of vagueness and things that only sometimes apply. Anomie⚔ 01:02, 27 December 2025 (UTC)[reply]
That reads like a cheap shot. SmokeyJoe (talk) 01:31, 27 December 2025 (UTC)[reply]
I do sometimes fall into the trap of shooting back when someone shoots at me first. Anomie⚔ 01:52, 27 December 2025 (UTC)[reply]
Ok. I don’t agree that “secondary source material” is jargon. “Secondary source” has real world definitions, some better than others, but it is a serious academic term. Secondary source material is material from a secondary source carrying the qualities of a secondary source.
It is not everyday discourse, but it is not complicated, and it is not vague.
Secondary source typing only sometimes is needed, but those times are questions of article deletion, which are pretty important. SmokeyJoe (talk) 02:14, 27 December 2025 (UTC)[reply]
"Secondary source" is a serious academic term, yes, but it lacks a clear and unambiguous definition. Whether a source is "secondary" or not depends on usage and context and which academic field is being talked about. Better IMO to avoid that and say what we actually mean. But obviously I'm not going to convince you, and you're not going to convince me. Anomie⚔ 19:08, 27 December 2025 (UTC)[reply]
Just want to support Masem's advice here. We need significant coverage in reliable secondary sources that are independent of the subject. Yes, we need to be 100% clear, because people will debate whether secondary means independent or whether secondary means independent. For a guideline this important, leaving out clarity just invites semantic debates, let alone bad faith willful ignorance. Shooterwalker (talk) 17:41, 23 December 2025 (UTC)[reply]
We "need" these things only because GNG is defined as needing these things, and only for certain subjects (the ones for which GNG or GNG-like principles are the basis for notability). Our need for these things was a choice, not something handed down to us on stone tablets. —David Eppstein (talk) 18:02, 23 December 2025 (UTC)[reply]
I agree, but my question at the top isn't about whether there should be secondary sources. My question is whether this particular line needs to contain an explanation. The choice I want to discuss is whether this list of ~definitions should say:
  1. "Sources" should be secondary sources, as those provide the most objective evidence of notability – or –
  2. "Sources" should be secondary sources.
I prefer #2. it's short, direct to the point, doesn't change the rules, and doesn't try to justify itself with an explanation that some people might disagree with.
Which do you prefer? WhatamIdoing (talk) 23:59, 23 December 2025 (UTC)[reply]
I think we do want to explain why we prefer secondary over primary, but I don't think the current statement of #1 explains this well; I don't think dropping any explanation is helpful, but we should have a clearer explanation. What explanation to replace it with I don't know immediately. Masem (t) 01:16, 24 December 2025 (UTC)[reply]
Does that explanation belong in a list of definitions, or elsewhere in the guideline? WhatamIdoing (talk) 06:01, 24 December 2025 (UTC)[reply]
Given the way all other bullets under GNG are written, it should be there. That section serves as both definition and rationale for each point. Masem (t) 04:48, 25 December 2025 (UTC)[reply]
I don't think the other bullet points in Wikipedia:Notability#General notability guideline provide a rationale. Can you quote any phrase that you think is providing a justification for the requirement? WhatamIdoing (talk) 22:27, 25 December 2025 (UTC)[reply]
For the purpose of demonstrating notability, sources should be secondary sources. That statement works standing alone.
Its a nice explanation that secondary sources (with caveats) provide evidence of notability, but where “notability” is defined as a Wikipedia neologism Wikipedia-notability.
I’m finding #2 preferable. The less use of the word “notable”, the better. SmokeyJoe (talk) 21:44, 25 December 2025 (UTC)[reply]
The need is a consequence of WP:NOR. Adopting WP:NOR was a choice, but having made that choice, WP:N, as written, mainly in the GNG, is necessarily implied.
There are some additional unchallenged choices, such as requiring source mutliplicity. SmokeyJoe (talk) 21:40, 25 December 2025 (UTC)[reply]
Wikipedia’s definition of secondary source is found in the article secondary source, as it should ideally be, as Wikipedia should not needlessly create barriers, such as new technical jargon. At secondary source, read Wikipedia as an encyclopedia being in the field of historiography, not science or journalism. If you do try to force science, you get into trouble with science not having a definition for a secondary source, at all, and for journalism, you’ll find yourself latching onto the nearest match, being the second hand source, which is not the same thing. SmokeyJoe (talk) 01:11, 24 December 2025 (UTC)[reply]
Wikipedia's definition of secondary is found in the policy WP:SECONDARY. WhatamIdoing (talk) 06:01, 24 December 2025 (UTC)[reply]
which immediate links to the mainspace article. “Shortcut, WP:SECONDARY
SmokeyJoe (talk) 08:53, 24 December 2025 (UTC)[reply]
So? WP:SECONDARY is our own definition. The link to the article Secondary source is hopefully helpful to editors, but it's not our own definition. WhatamIdoing (talk) 21:39, 24 December 2025 (UTC)[reply]
The WP definition matches with what Secondary source says, particularly highlighting "Secondary sources involve generalization, analysis, interpretation, or evaluation of the original information." from that article. Masem (t) 04:49, 25 December 2025 (UTC)[reply]
WP:SECONDARY points unambiguously to the academic definitions, and accepts it on face value. SmokeyJoe (talk) 10:39, 25 December 2025 (UTC)[reply]
Maybe cut “the most” but keep the rest. Wikipedia-notability means others have previously written about it. This implies an independent secondary source. SmokeyJoe (talk) 00:59, 24 December 2025 (UTC)[reply]
Wikipedia-notability is much more specific than "others have written about it". We need those others to have written about it in large enough units rather than piecemeal over a larger number of smaller units. We need them to have published their writings through publication processes that we deem to impart reliability to the writing. We need the writing to be evaluative and to summarize other writings rather than directly reporting news events or research results. And we need both the others doing the writing and their publisher to be at some remove from the subject. All of this is much more arbitrary and subjective than a lot of GNG-advocates would pretend it to be. —David Eppstein (talk) 00:42, 25 December 2025 (UTC)[reply]
I have my doubts about rejecting the Least publishable unit. It's true for NCORP subjects (by fiat), but I don't think the question is settled for other subjects. We need enough verifiable information to be able to write a decent (~300 word) encyclopedia article, but it's not clear why we would actually need that information to be in two bigger publications instead of five smaller ones. WhatamIdoing (talk) 03:37, 25 December 2025 (UTC)[reply]
The same reason we have all those other requirements: because GNG says so. We need to have some threshold for notability (on this I agree), and GNG provides this specific threshold for notability, so we need what it says, and what it has been interpreted to say, that the sources providing notability be individually and not merely collectively in-depth. —David Eppstein (talk) 04:34, 25 December 2025 (UTC)[reply]
The GNG doesn't say that SIGCOV has to come from a single source. The GNG says:
"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."
It's reasonable to read that as requiring significant coverage across all independent reliable sources collectively, rather than individually. After all, it doesn't say "has received coverage in reliable sources that are independent of the subject and that each contain significant coverage, when considered in isolation". WhatamIdoing (talk) 05:46, 25 December 2025 (UTC)[reply]
But is the same one line report repeated 100 times "significant coverage"? Slatersteven (talk) 10:56, 25 December 2025 (UTC)[reply]
No, because a hundred copies of the same single line doesn't give us enough information to write an article.
But a hundred short sources that talk about a hundred different things might, in which case they collectively meet the goal of SIGCOV. WhatamIdoing (talk) 18:21, 25 December 2025 (UTC)[reply]
David, Wikipedia-notability is not more specific than "others have written about it", it cops out with the undefined term “significant”. I agree with everything else you wrote (as I usually do).
When depth of coverage is challenged, I sometimes point to WP:100W. SmokeyJoe (talk) 10:43, 25 December 2025 (UTC)[reply]
  • If there is some overlap in some people's minds (or in objective reality (ha, ha)) between "significant" and "secondary" , I don't think that's a problem (concepts often overlap or fade into others in different directions and people often use multiple words that may have some similarities in conceptualization, especially when trying to write down guidance (viz. a secondary source likely shows significant, sustained thought on something). The issue being addressed at Afd is an editorial judgement call, not an algorithm, and people can work on it subjectively themselves but they are still working in the same direction, under guidance: 'in my editorial judgement (our collective judgement) this thing is an article topic or it should be covered in another way, if it all.' --Alanscottwalker (talk) 13:58, 25 December 2025 (UTC)[reply]