Wikipedia talk:Stand-alone lists

Creating minimum inclusion criteria for lists involving subjective categorization

For lists involving subjective categorization - such as List of video games considered the best and List of massacres in France - there is the risk that we place WP:UNDUE emphasis on a minority view, as well as the risk that we engage in WP:OR by including as comparable two topics that no reliable source would say are comparable.

Some articles, like List of video games considered the best, address this by requiring that multiple reliable sources have considered the game among the "best/greatest of all time", but others do not. Given the core policy issues that failure to do this can and does cause, I suggest we modify this guideline to include something like the following under "Selection criteria":

Selection criteria for lists involving subjective categorization

To comply with core policies on neutrality and original research topics should only be included unqualified in a list involving subjective categorization, such as List of video games considered the best or List of massacres in France, if the view that the categorization applies is the view of the majority in published, reliable sources. If the view that the categorization applies is held by significant minority then the topic can be included alongside appropriate qualification that makes it clear that its inclusion is not the majority view.

This is particularly important when the category is covered by MOS:PUFFERY or MOS:LABEL.

Much of the wording is taken from WP:UNDUE. This follows on from a discussion at WT:NOR. BilledMammal (talk) 02:39, 21 March 2023 (UTC)

I don't think it's a good idea to duplicate this kind of material across policy and guideline pages. WhatamIdoing (talk) 02:43, 21 March 2023 (UTC)

The wording taken from WP:UNDUE is the aspects on majority/significant minority view; the rest, as far as I know, does not duplicate content elsewhere. However, I don't mind what PAG this goes in, or how it is worded - I just believe we need it, due to how common it is for lists involving subjective categorization to include topics whose inclusions aren't WP:DUE. BilledMammal (talk) 02:48, 21 March 2023 (UTC)

I think the current wording at WP:LISTCRITERIA is sufficient to cover these cases, and I'm wary of being so specific in our prescriptions. I'd rather give editors more discretion to choose the most appropriate list criteria on a case by case basis. For example, a while back I created List of gay novels prior to the Stonewall riots. Whether a particular work is a "gay novel" is a subjective judgement. I suggested that for the purposes of this list, a work being described as a gay novel in any reliable source was sufficient for inclusion. Using the criteria you propose instead seems like it would be difficult to apply in practice. How do I determine whether The Picture of Dorian Gray being a gay novel is "the view of the majority in published, reliable sources"? Is it a question of whether a majority of sources about gay novels mention Dorian Gray? Or whether a majority of sources on the topic of Dorian Gray describe it as a gay novel? Does the fact that a source fails to describe it as a gay novel imply that the author has judged that it's not a gay novel? Colin M (talk) 17:19, 22 March 2023 (UTC)

That difficulty is defined by WP:NPOV; it exists already for everything we write, this proposal would just make it clear that lists do not have an NPOV exception.

With your Dorian Gray example, if it would be appropriate to describe the novel as a gay novel in the article then it would be appropriate to include it in an NPOV-compliant list - I don’t know enough about the topic to say whether it would be appropriate to do so. BilledMammal (talk) 21:59, 22 March 2023 (UTC)

I think it's more complicated than that. You need to consider the facts as represented by the sources, of course, but you also need to consider other factors. If the introduction to the list says something like "This is a list of foo that comply with the 2016 version of the ISO standard 1234-5", then you need to be pretty tight about what you put into the list.

But if the lead instead says something like "This is a list of all known foo. If the entry has already been formally accepted by the Governing body or by the Professional association, their registration numbers are listed. Sometimes, there is a disagreement over whether some foo are really foo enough to count. All foo are labeled with the earliest known year; newer foo are tentative and may be disputed later. Entries that have been verifiably disputed by experts are included by marked with 🚨. Proposals whose foo-ness has been widely rejected by the field are in a separate table below and marked with ❌." – then in that case, you can list almost anything.

The point isn't to make the list big or small; it's to be clear to the reader about what inclusion means.

I will also add: Trying to exclude "subjective" information unless you can produce six sources for it can cause pretty serious NPOV problems. A standard like that may sometimes be necessary, but in other cases, it can bias the list towards entries that got a lot of publicity, instead of providing a more comprehensive list. The list selection criteria that make sense for a list of video games will not necessarily make sense for a list of artists. Whether someone is "really" an artist is at least as subjective as whether a video game is best in its class, and if you require six approved sources that say someone is an artist, you could end up with a List of Ruritanian artists who can afford to hire a publicist instead of a List of Ruritanian artists. WhatamIdoing (talk) 04:33, 23 March 2023 (UTC)

Trying to exclude "subjective" information unless you can produce six sources for it can cause pretty serious NPOV problems. I agree, which is why I haven't proposed that - but my question for the rest is if we can't say that x is y in the article on x because it would violate WP:UNDUE, why can we say that x is y on a list? BilledMammal (talk) 04:42, 23 March 2023 (UTC)

What's appropriate depends on the subject of the article. There are things you can say in Rolls-Royce Silver Ghost that would be undue in Rolls-Royce Limited; there are things you could say in that article that would be undue in Luxury car; there are things you could say in that article that would be undue in Car; there are things you could say there that would be undue in Vehicle. Being undue in article X does not mean that it's undue in every single article. WhatamIdoing (talk) 04:17, 31 March 2023 (UTC)

This issue needs to be addressed with care. The AFD for Wikipedia:Articles for deletion/Horror icon (2nd nomination) (no WP:CANVASSing) is an example of something that is heavily discouraged on Wikipedia. This kind of WP:SYNTH cascades into other issues.

First is that you transform the subjective opinions of several authors into the objective voice of an encyclopedia, elevating their opinions with false appearance of reliability and neutrality. This is different from including a critic's opinion in the reception area of an article, where the attribution is clear, and it is accurately noted as an opinion. There's a major difference between summarizing two or three critics who all call something "cute", versus an authoritative List of cute media. Presenting it as an encyclopedic topic is misleading and does a disservice to our readers.

The second issue is we can create endless arbitrary articles based on any subjective descriptor. For example, List of disgusting foods, or List of purple paintings, or List of most violent movies. It doesn't matter if you rename these articles List of foods regarded as disgusting by a food critic, or List of paintings that have been noted as purple by at least three art historians, or An aggregated list of movies ranked as most violent. The rename just makes the editor's chosen construction more clearly arbitrary.

I don't know what the solution is. But this might deserve one or two RFCs: one to see if there are any consistent opinions about how to handle these lists, and a second to see if we can turn that consistency into a guideline. Shooterwalker (talk) 03:18, 29 March 2023 (UTC)

I think the solution to your second problem can be found in WP:NLIST --- like other articles, we should only create sets that are notable, whose notability is supported by multiple reliable sources. Specifically, the topic of a list should be discussed as a group or a set in independent reliable sources. — hike395 (talk) 09:37, 29 March 2023 (UTC)

• So, with regards to "view of the majority in published, reliable sources." that seems like an unweildy standard to hold something to. How are we to assess "the majority" of "published, reliable sources" There may be hundreds or thousands of possible sources for such list articles. Do we really expect people to manually check each one and then add the total to assure 50%+1 of all available sources, for example, call an event "a massacre" specifically? What about describing an event using terms that make it clear that it meets the dictionary definition of "massacre", but what if the source doesn't use word "massacre"? What if the source uses clear and unambiguous perfect synonyms of the word "massacre"? What if the source is in a language other than English, which lacks a really good single word that is equivalent to "massacre". Determining a "majority in published reliable sources" is an absolutely unworkable standard. Even at best, it is open to gaming and wikilawyering in ways that are a hinderance to a positive work environment. "We have 100 sources that say "massacre" and 99 that done mention "massacre". I found 2 more that don't mention the word massacre, so now we must remove it by the rules!" Seriously, that's what this is going to lead to. Just, no. Not more of that kind of bullshit. I know it's a problem, but this is absolutely NOT the solution.--Jayron32 15:30, 30 March 2023 (UTC)

  Wrt "the majority of published reliable sources" – which is language you will never find in WP:UNDUE, which cares about which viewpoints are major and minor, but never recommends simply counting up the number of sources – I also worry about low-quality sources being put forward as a reason to "outvote" the high-quality sources. Your view only has a few scholarly sources, but my view has hundreds of social media posts, and as WP:V makes clear, even social media posts can be reliable for saying that someone posted something, so they're "published reliable sources"! WhatamIdoing (talk) 04:25, 31 March 2023 (UTC)

  @WhatamIdoing and Jayron32: What I am trying to do here is make it clear to editors that WP:DUE applies to inclusion in lists. I don't think either of you are disagreeing with that goal, just the wording that I am using for it.

  In regards to view of the majority in published, reliable sources, the intent of that sentence is to determine whether a view is the majority view, using the same definition of "majority view" as WP:UNDUE does. However, if you feel my wording deviates from the meaning used at UNDUE then it would need to be changed. BilledMammal (talk) 14:23, 23 April 2023 (UTC)

  Yes, that wording deviates from UNDUE, and it would need to be changed.

  I'm not sure that trying a straight-up transposition of UNDUE to inclusion/exclusion for a list is a good idea. For one thing, the list-selection criteria for List of video games considered the best, which you seem to support, would probably be invalid. (Why? Because with UNDUE, you would either restrict yourself to comparative sources, such as low-quality listicles about "The best video games ever", or you would compare only sources that refer to this particular game, so if 100% of the only two reviews for "WhatamIdoing's Game" say it's the best ever, then that's the majority viewpoint for that video game, and it would be included. UNDUE doesn't have an escape clause like "Ignore the majority viewpoint for your exact subject if there aren't very many sources about your subject".)

  Second, on your broader goal, the community has a problem with the idea of subjectivity. We don't agree on what's subjective and what's not. We are relatively weak on deciding whether something is an opinion or a fact. For example: Is an event in which four people die a "mass killing"? It depends on your definition of mass killing. Do Heartbeat bills ban abortions at six weeks of pregnancy, or at four, or at three? It depends on your definition of pregnancy. If a tree falls in a forest, and no one is around, does it make a sound? It depends on your definition (in that case, whether "a sound" is the soundwave, which does not require an observer to exist, or the perception of that soundwave, which does). These aren't really "subjective" or "opinions", though. Anyone could look at the definitions and say "Under this one, yes; under that one, no", and everyone looking at the definitions would come to the same conclusions. (And if they are much over the age of 10 and genuinely can't do this, then (a) Wikipedia:Competence is required for Wikipedians, and (b) it's probably time for neuropsych testing, because the inability to apply different definitions or see other points of view is a symptom of several neurological conditions, including autism.) For lists, we have traditionally used the approach of providing the definition that we're following, which we call the Wikipedia:List selection criteria, and letting editors sort it out from there.

  Thirdly, sometimes what's UNDUE in an article is not inappropriate for a list. I don't know if you've read Richard Feynman about visiting the Sistine Chapel, but he writes that his guidebook said this: '"Below the paintings by Michelangelo there are fourteen panels by Botticelli, Perugino" – all these great artists – "and two by So-and-so, which are of no significance".' A List of paintings in the Sistine Chapel should include all of them, because even the paintings of no artistic or historical significance are still paintings in the Sistine Chapel (and including them would not make the list too long). The article on the Sistine Chapel, on the other hand, should probably not mention them, as including material of no significance is UNDUE.

  Finally, I think that if we want to make progress on the underlying problem, the next major step is to help editors understand the difference between subjective and uses a definition that is less familiar to me. WhatamIdoing (talk) 20:08, 23 April 2023 (UTC)

  One thing I want to clarify is that I didn't use the wording the majority of published reliable sources; I used the wording view of the majority in published, reliable sources. The difference as I see it is that it isn't talking about counting sources, it is about determining what the majority view is. However, I wonder if it would be clearer if reworded to view of the majority, substantiated with references to commonly accepted reference texts, which is very close to the current wording at WP:DUE.

  Finally, I think that if we want to make progress on the underlying problem, the next major step is to help editors understand the difference between subjective and uses a definition that is less familiar to me. "Subjective categorization" would be any categorization that is not based on measurable and universally accepted criteria. For example, List of paintings in the Sistine Chapel is not subjective categorization because whether a painting is in the Sistine Chapel is measurable, and being in the Sistine Chapel is universally accepted as being in the Sistine Chapel. However, List of mass killings is a subjective categorization, because, as you said, there is no universally accepted criteria for what constitutes a mass killing.

  There are also three articles currently at AfD that I believe is evidence for this change needing to be made, due to the constant inclusion of non-compliant text:

  1. Wikipedia:Articles for deletion/List of right-wing terrorist attacks (2nd nomination)
  2. Wikipedia:Articles for deletion/List of left-wing terrorist attacks
  3. Wikipedia:Articles for deletion/List of Islamist terrorist attacks

  BilledMammal (talk) 01:08, 1 August 2023 (UTC)

IMHO lists are in the twilight zone in Wikiepia not really covered regarding suitability to be an article, wp:notability or criteria/procedures for inclusion. So I think that work needs to be done. Regarding the specific topic at hand, to navigate, it should be noted that inclusion on a list is essentially a statement (in the voice of Wikipedia) that the listed item IS what the title of the list is. The fundamental concept should be that the more controversial that implicit statement is, the stronger the sourcing that is required, and where it is controversial it should be noted as such and where appropriate, it should be done with attribution. Of course dealing with that implicit statement as such that doesn't fit neatly into a typical list article format, which is one of the many unresolved areas that need work regarding list articles. Sincerely North8000 (talk) 23:53, 27 December 2023 (UTC)

Listing every cast member for television shows

I've noticed a user who regularly appears to be adding every any and every actor who appears in any (mostly Australian) television shows, regardless of notability to the television program (for example, the actor may have appeared in only one episode in a unnamed background role). These actors are often added to the cast lists as "guest cast" to entries. The additions seem to be more for "completeness" purposes of listing every cast member who may have an existing page of wikipedia, similar to a directory or IMDB.

Examples: Recent edit of The Lost World (television series) from January 26, 2024

• This edit has added two actors who appear in only one episode each in the series.

Tales of the South Seas (current page as of January 26, 2024; refer to edits all done on January 26, 2024)

• User recently added 7 "guest cast" members with minor or single appearances, some actors do not have a listed role.

Are there guidelines on whether listing every actor credited in a television, regardless of notability, is permitted on wikipedia as per policies? I am assuming the editor is contributing in good faith, but the contributions appear to be creating a directory similar to what IMDB would be if someone wanted to see if complete cast lists. 50.68.30.200 (talk) 00:51, 27 January 2024 (UTC)

This is a bit out of place here, as your question doesn't relate to stand-alone lists. That said, I'd recommend starting with MOS:TV. DonIago (talk) 04:50, 27 January 2024 (UTC)

Thank you for the reply and the reference to MOS:TV - it looks like it has the guidance I was struggling to find on notability and cast lists on television series entries.

50.68.30.200 (talk) 05:27, 27 January 2024 (UTC)

Discussion at List of genocides § List ordering: Reverse or regular chronology

 You are invited to join the discussion at List of genocides § List ordering: Reverse or regular chronology. —Alalch E. 00:52, 30 August 2024 (UTC)

CSC from bullet points to numbers?

Does anyone object to changing the three common selection criteria from a bulleted list to a numbered list? I can't think of a reason why not, as those three have been stable there, in that order, for years, and similar lists (deletion reasons, speedy keep reasons, etc.) seem to be generally numbered. Jclemens (talk) 20:57, 2 November 2024 (UTC)