Wikipedia talk:Stand-alone lists

Proposed opening for the "Selection criteria" section

Here are a couple opening paragraphs to help make the Selection criteria section more accessible to new editors...

The selection criteria of a list are the requirements and characteristics needed for items to be included in the list. Some titles make this clear on their own. List of bones of the human skeleton and List of AMD processors tell you exactly what belongs.

But many titles leave room for different interpretations. For example, List of hairstyles. Is that limited to styles of hair on one's head? Or does it include hair in other areas? For lists with titles that are subjective, ambiguous, or not entirely clear, additional selection criteria are needed. To continue the example, that list specifies in its lead that facial hairstyles are included in the list.

Hope these help. Sincerely,    — The Transhumanist   09:46, 3 March 2026 (UTC)

I like it. --Guy Macon (talk) 11:54, 3 March 2026 (UTC)

Is the idea to add this to the current text, or to replace (part of) the current text with this? TompaDompa (talk) 20:40, 4 March 2026 (UTC)

I think it's usually helpful to spell out the list selection criteria, but I think that it's sometimes weird to do this in the lead of the list (e.g., if the list selection criteria are best explained with various wikijargon, like "non-redirect bluelinks"). Sometimes this should be on the talk page instead of in the mainspace. WhatamIdoing (talk) 20:49, 4 March 2026 (UTC)

Those are intended to be the first 2 paragraphs of the section. The next section could clarify where additional selection criteria (beyond what's in the title) should be placed. For some lists, it's in the lead, for some, it's in the body, some have it in their hidden comments, and then there are those that have it on the talk page. Some guidance on this may be very useful, especially to new editors, to help them understand what is going on with lists. Would you like to take a crack at the third paragraph? To lead into the third paragraph, the second paragraph could end with ", but the lead section isn't always the best place to put the inclusion criteria...".    — The Transhumanist   01:53, 5 March 2026 (UTC)

I've been wondering if we need a Template:List selection criteria (for use on Talk: pages). WhatamIdoing (talk) 03:49, 5 March 2026 (UTC)

Maybe that could help. Though, we should probably finish the section first, so we would have a better idea of what to put in the template.    — The Transhumanist   06:37, 5 March 2026 (UTC)

There's Template:List_criteria, or are you envisioning something different? FactOrOpinion (talk) 17:12, 5 March 2026 (UTC)

That's exactly what I wanted, thank you! WhatamIdoing (talk) 21:03, 5 March 2026 (UTC)

@TompaDompa: The paragraphs we are discussing here are for prepending to the existing section, which might need to be adjusted a little, but not necessarily replaced.    — The Transhumanist   02:01, 5 March 2026 (UTC)

Alright. I'm not opposed to adding something along those lines. TompaDompa (talk) 13:58, 6 March 2026 (UTC)

The selection criteria of a list are the requirements and characteristics needed for items to be included in the list.

"Requirements" would be enough: in our case, a criterion is a standard by which something may be judged or decided. Or say "the characteristics required" and drop the "needed". Take the time to pick the best word: "and" often indicates indecision. "Required" is more precise than "needed". "Characteristics" is better than "properties", which leads to difficulties as old as Aristotle.

Some titles make this clear on their own. List of bones of the human skeleton and List of AMD processors tell you exactly what belongs.

Most lists with selections also tell you what belongs: "people living in X" is exactly what is said on the tin. If it doesn't, that suggests the title needs to change. That is tangential to why we need to mention our selection criteria: it is the selection that needs to be made as explicit as possible. A list of selection criteria is there to tell readers the criteria by which selection has been made. Readers already know they won't get a list of every single being living in X; they perhaps expect that they're the most prominent ones. We still need to say it, preferably without using the word "notability", which has a technical meaning here and may confuse editors. They're already confused enough about it. Dispositional concepts are complex. Selbstporträt (talk) 14:15, 5 March 2026 (UTC)

I think we should avoid the concept of "requirements" or "needed".

Wikipedia:Article titles are frequently vaguer than the contents. Using the "People living in ____" example, the list should not include everyone who lives there. It should only include people verifiably living there, or verifiably living there and their residence seems significant, or verifiability living there and they qualify for a separate, stand-alone article on Wikipedia, or verifiably living there and an editor has already written an article about them. But we should not have an article title like "People verifiably living in ____ with articles written about them". It should just be "People living in ____" (except that we don't do "People living in" lists at all, even though that makes a nice example). WhatamIdoing (talk) 21:23, 5 March 2026 (UTC)

Your point is about identification, not selection. Here we're talking about filtering: a page of entities E with property P should not have entities without the property P. Identification alone often suffices for two list types: comprehensive lists, and partial lists. Both seek to be exhaustive. No need to spell out how we'll pick stations in the London Tube when we want every single one.

A list based on selection criteria is of a different type, call it a selective list. The selection (often called a "view") can be thought as some query Q; Q can be composed of conditions, e.g. "give me all the scientists who published papers after 2015 in molecular biology, especially in" etc. Adding conditions (i.e. with an "and" connector) restricts selection; combining conditions (i.e. with an "or" connector) expands selection. Anyone who made an advanced search online did what I just said.

To state criteria is to state conditions: they *require* something, both of the reader and of the topic under consideration. That's true even in fuzzy logic, where set membership is fuzzy. Vagueness is of no concern here for it's quite possible to give vague recipes.

Calling what we are selecting for a characteristic only makes that requirement implicit. For matters where procedures add clarity, we should not retreat from telling readers what to do. A cookbook written in denotational semantics would contain no preparation steps.

MOS:AT does not shy away from imperatives. Wikipedia:Article titles works by characterization: it states goals, not rules. Which style one prefers I could care less. So let's revise my offer:

(1) The selection criteria are requirements to include members in the list.

(2) The selection criteria are characteristics required for entries to make the list.

(3) The selection criteria are what allows editors to select entries for our list.

They should all work fine. At this point, Elves and Fairies would usually take the text out of my hands. The basic point is the word "and": "characteristics and requirements" present the same thing in two different ways. But it can be read as an alternative. And it can be read as a conjunction!

Better pick one.

This exercice takes a minute with edits. Talk pages are just an inefficient way to edit. Easier with incremental changes. Too many justifications paralyze. Selbstporträt (talk) 02:15, 6 March 2026 (UTC)

Revised proposed opening for the "Selection criteria" section

Here's a revised draft of the opening for the Selection criteria section, to make it more accessible to new editors...

The selection criteria of a list are the characteristics of items belonging to the focus of the list. Some titles make this clear on their own. List of Game of Thrones episodes and List of AMD processors tell you exactly what belongs.

But many titles leave room for different interpretations. For example, List of hairstyles. Is that limited to styles of hair on one's head? Or does it include hair in other areas? For lists with titles that are subjective, ambiguous, or not entirely clear, additional selection criteria are needed. To continue the example, that list specifies in its lead that facial hairstyles are included in the list.

When the title isn't clear enough, selection criteria may be placed in the lead, in the list's body, or on the list's talk page. Sometimes, list articles include non-list sections, with the main list in a section; for those, the selection criteria may be best placed in the section with the list. For list articles with sublists, selection criteria typically goes in their sections. Selection criteria that includes self-references (mentions Wikipedia or its website features) should be presented on the talk page.

Is this complete enough for an opening? Sincerely,    — The Transhumanist   06:34, 6 March 2026 (UTC)

Clear, understandable, and helpful. My only problem is that it doesn't explicitly say that some sort of selection criteria is required.

I have dealt with multiple editors who say that no selection criteria is required, and that they can include or exclude anyone from the list of people named in the Epstein files at will as long as one source mentions them in any context. I have seen them defend including someone because he was invited to an event that Epstein was invited to (we don't know whether they or Epstein actually attended). One person was included because an unnamed "friend of Jeffrey Epstein" (assumed from the fact that Epstein was copied on the email) invited them to a dinner in the UK while Epstein was under house arrest in Florida. And they added someone who, along with his wife, was mentioned in a draft sent from Epstein's email account (we know that some of his staff used that account) back to the same account. The only reason they were in the sources cited was because they made public denials after people on social media searched for their names and made a bunch of accusations. And only the husband was added because reasons.

I really want this page to say that some sort of selection criteria is required. I feel strongly enough about it that I am ready to create an RfC, but I am hoping that we can agree that letting any editor add or remove anyone from that list because they feel like it should be expressly forbidden using language I can quote. --Guy Macon (talk) 11:53, 6 March 2026 (UTC)

I really want this page to say that some sort of selection criteria is required. I agree. I'll follow this advice and add something to that effect. TompaDompa (talk) 15:09, 6 March 2026 (UTC)

@Guy Macon: We now have The criteria may be outlined in the article lead, the article body, or on the talk page; when they are unambiguously clear from the article title, mentioning the precise definition of the criteria may be omitted altogether. to quote. We'll see if it gets tweaked further by others. TompaDompa (talk) 15:14, 6 March 2026 (UTC)

That looks perfect. Thanks! --Guy Macon (talk) 15:21, 6 March 2026 (UTC)

All lists already have selection criteria. They start out with it, in the title. That the main store of selection criteria is in the title is ignored, and is tacked on as an inferential afterthought. The title pre-exists any other form of selection criteria, and is the first indicator of content a reader is exposed to. That the title is selection criteria should be made explicitly clear. And when the title is not enough, then additional selection criteria is required. Also, "outlined" is ambiguous. "Outline" means something very specific on Wikipedia, especially relating to content. An instruction that requires something be "unambiguously clear" should itself be unambiguously clear, as to not be hypocritical.    — The Transhumanist   21:21, 6 March 2026 (UTC)

I know you're fond of Wikipedia:Outlines, but realistically, I suspect that nobody except you is going to look at "outlined" in that sentence and think about Wikipedia:Outlines. I wouldn't bother changing that. WhatamIdoing (talk) 22:34, 6 March 2026 (UTC)

@Guy Macon, they told you what their list-selection criterion was. It was "as long as one source mentions them in any context". That's a list-selection criterion. You and I might say that it's a completely inappropriate criterion, but it's the criterion they were using and wanting to use. WhatamIdoing (talk) 22:33, 6 March 2026 (UTC)

I can't say that you are wrong. That's technically a list-selection criterion, just as "being a person", "having a name", and "being in the Epstein files" (all three criteria are contained in the article name "list of people named in the Epstein files") are. I probably should emphasize that having a well-chosen set of criteria is a necessary-but-not-sufficient means to an end (preventing it from being an indiscriminate list or a list containing BLP violations) rather than being an end in itself.

How do you comfort a pedant? There, their, they're. :) --Guy Macon (talk) 02:00, 7 March 2026 (UTC)

"you and I might say that it's a completely inappropriate criterion"

Saying it is one thing. Just like dismissing work with "it's not an improvement". Arguing it is quite another. I'm sure there's a smiley I could use here. Also, "It was "as long as one source mentions them in any context"" is plainly false. Selbstporträt (talk) 02:19, 7 March 2026 (UTC)

One way or another, editors will have to reach a consensus about what to put on that page. WhatamIdoing (talk) 05:06, 7 March 2026 (UTC)

There has been more sparks today here than since I appeared over there. Let's hope Guy's frens can cooperate ;-) Selbstporträt (talk) 07:24, 7 March 2026 (UTC)

Original research tangent

"We can't tell editors: find yourselves a catalogue of all the Norwegian garage bands, and borrow its inclusion criteria!"

To clarify the importance of that remark: I have seen editors (professors, in fact) arguing that unless we have a source to a kind of catalogue like that, any selection is original research. This is not what OR means. This implies we might to need to revise the OR page, and clarify that we have no such requirement for our criteria. Selbstporträt (talk) 14:29, 5 March 2026 (UTC)

More generally, that problem could be summarized as "deciding the scope of a page is not OR". That includes setting the list-selection criteria as well as deciding whether to merge or split a page. WhatamIdoing (talk) 17:52, 5 March 2026 (UTC)

Indeed, this returns us to the topics we're allowed to choose, and the style guide for titles. So many hours wasted on name changes and on "but is it truly neutral" frolics.

A more pressing point would be to remind people that the criteria should be somewhat cumulative, otherwise they may end up not filtering anything or worse degenerate.

Anyway. Enough chit chat. Time to write up that essay. Selbstporträt (talk) 18:15, 5 March 2026 (UTC)

@Selbstporträt, I'd suggest adding it to Wikipedia:These are not original research, if you want an essay. But it's possible that it should be directly in the policy itself. WhatamIdoing (talk) 18:29, 5 March 2026 (UTC)

We certainly should not be saying "deciding the scope of a page is not OR", because it can be—producing novel content is WP:Original research, whether that is done by adding novel information to an article or creating a novel definition/scope of a concept/topic. Similarly, deciding the scope of a page can be a WP:NPOV violation, a WP:BLP violation, a WP:FRINGE violation, or other kinds of violations depending on what scope is decided upon. Deciding the scope of a page is not a "get out of jail free"-card for abiding by Wikipedia's policies and guidelines. TompaDompa (talk) 14:06, 6 March 2026 (UTC)

The relevant bit is: the entirety of the list does not need to be documented in sources for notability, only that the grouping or set in general has been. Selbstporträt (talk) 23:01, 6 March 2026 (UTC)

What it means to have "the grouping our set" be "documented" is frequently a source of confusion. It would probably help if we provided a clear example, approximately along these lines:

• ____ is a notable subject
• List of ____ is therefore a notable subject for a list because sources talk about [something]. It is not necessary to have either any reliable sources that use a list format, or to have any reliable sources that includes words like "When considering _____ as a group or set..."

Also, naming some of the common list subjects (e.g., lists of works, lists of geographical features, lists of people holding a given office) would probably also help people avoid doomed AFDs. WhatamIdoing (talk) 05:03, 7 March 2026 (UTC)

Examples will come next, with "Common selection criteria" subsection. There should be two subsections at least, or none at all. Dividing by 1 is incoherent.

One subsection could start with the Norwegian bands, or Transhumanist's para above, which I rather like.

Titles are covered elsewhere. But good point: following WP:AT is for the best. It's a great page. Selbstporträt (talk) 07:20, 7 March 2026 (UTC)

None of this makes deciding the scope immune to OR concerns, which is why we should not be saying "deciding the scope of a page is not OR". TompaDompa (talk) 13:22, 7 March 2026 (UTC)

"None of this" lacks clarity. "The scope" misses its own scope: the scope of what? Selbstporträt (talk) 16:01, 7 March 2026 (UTC)

Nothing anybody has said in this conversation (and nothing that hasn't been said either, for that mattar) makes deciding the scope of a page immune to OR concerns, which is why we should not be saying "deciding the scope of a page is not OR". TompaDompa (talk) 16:11, 7 March 2026 (UTC)

Here is a quote I already offered:

the entirety of the list does not need to be documented in sources for notability, only that the grouping or set in general has been.

I believe that "the entirety of the list" contradicts the "nothing" you just said. This quote comes from the section Stand-alone lists on the Notability page. It is a common mistake. Here is the best place to remind it. Selbstporträt (talk) 16:41, 7 March 2026 (UTC)

You are conflating WP:Notability concerns and WP:Original research concerns. TompaDompa (talk) 17:10, 7 March 2026 (UTC)

You are failing to recognize a direct implication: "documented in sources" means satisfying reliability, which parries accusations of OR.

Since only the grouping or set in general needs to be sourced, then "OR concerns" should not be raised regarding the entirety of lists.

It is possible to see implications without making any conflation. Selbstporträt (talk) 17:33, 7 March 2026 (UTC)

Consider the following: there is a grouping or set that is documented in sources, thus demonstrating notability. An editor then makes a list article about that grouping or set, and imposes a number of criteria for inclusion. These criteria do not reflect the way the sources handle edge cases. The list is then about a notable topic which is covered in sources, yet the contents do not reflect the sources accurately. Then we have novel content on an otherwise appropriate topic, or in other words WP:Original research problems without WP:Notability problems. It is not too difficult to imagine this happening with something like List of sovereign states. TompaDompa (talk) 17:45, 7 March 2026 (UTC)

Here's how they introduce their inclusion criteria, with emphasis:

Compiling a list such as this can be complicated and controversial, as there is no definition that is binding on all the members of the community of nations concerning the criteria for statehood. For more information on the criteria used to determine the contents of this list, please see the criteria for inclusion section below. The list is intended to include entities that have been recognised as having de facto status as sovereign states, and inclusion should not be seen as an endorsement of any specific claim to statehood in legal terms.

So much the worse for the idea that a selection criteria should be objective, then! And this example supports the need for a claim you edited: If the chosen criteria rests on established norms or authorities, they should cite relevant sources. Which is exactly what has been done here.

Being sourced does not imply notability, BTW. Selbstporträt (talk) 17:56, 7 March 2026 (UTC)

It seems to me that you have an idiosyncratic interpretation of "objective". Or at least, I fail to see how this is "So much the worse for the idea". Do you think criteria should not be objective?

The current version says "If the choice of selection criteria for a list is subjective or likely to be disputed, then the criteria should be supported with citations.", which is a stronger requirement than "If the chosen criteria rests on established norms or authorities, they should cite relevant sources.", so I fail to see how the latter is needed? I'll add "to relevant sources" to the current version, if that helps. TompaDompa (talk) 18:14, 7 March 2026 (UTC)

The current version did not solve the very first thing it was meant to solve, which is that you require criteria to be objective and right after that you say "but when they're subjective".

I have no idea how you can still be under the impression that I'm the one who struggles with concepts when I keep showing you examples that break whatever you are throwing at me. Selbstporträt (talk) 18:20, 7 March 2026 (UTC)

I think the apparent contradiction has been resolved, which is one of the reasons I think we have different ideas of what "objective" and "subjective" mean in this context. Here's text from the "Common selection critera" section that was here before either of us:

Lists are commonly written to satisfy one of the following sets of objective criteria:

What would you say "objective" means in that sentence? TompaDompa (talk) 18:43, 7 March 2026 (UTC)

Look. You keep reiterating your position without arguing for it, while talking past the arguments I propose and commenting on my competence. This is not the time to play Questions. But I'll oblige. Recall what you dismissed without any justification:

If the chosen criteria rests on established norms or authorities, they should cite relevant sources.

Ring any bell? Selbstporträt (talk) 18:58, 7 March 2026 (UTC)

Let's take this thread down to the bottom of the discussion, okay? I'll start a new subsection for it. WhatamIdoing (talk) 20:21, 7 March 2026 (UTC)

If I have commented on your competence, I apologize; I have not done so intentionally.

Yes, I recognize "If the chosen criteria rests on established norms or authorities, they should cite relevant sources." My objection to it was that it replaced "If the choice of selection criteria for a list is subjective or likely to be disputed, then the criteria should be supported with citations.", and that the latter is a stronger requirement (inasmuch as it applies in a broader set of circumstances). I have stated this objection more than once. For instance, here I said:

My version's "If the choice of selection criteria for a list is subjective or likely to be disputed, then the criteria should be supported with citations." is a much stronger requirement than your version's "If the criteria chosen rests on established norms or authorities, they should cite the relevant reliable sources.", because "subjective or likely to be disputed" is a much lower threshold than "rests on established norms or authorities" (which I also think is rather unclear in its meaning).

Why would you say that this is something I dismissed without any justification? Did you miss what I said about it? TompaDompa (talk) 22:07, 7 March 2026 (UTC)

I think "If the chosen criteria rests on established norms or authorities, they should cite relevant sources" should be removed. In particular, sometimes it's silly (there are 'norms' for how many US states there are, but they do not need to be cited in an article), and if the criteria are listed on the talk page, it's usually unnecessary. WhatamIdoing (talk) 20:16, 7 March 2026 (UTC)

Norms are not established norms, tho.

Also, the statement does not say if you have norms, use them. It says if you pick a norm, cite it. Selbstporträt (talk) 21:11, 7 March 2026 (UTC)

Why should we WP:CITE a norm for List of US states? Most 10-year-old school children in the US can recite the list in alphabetical order. WhatamIdoing (talk) 23:13, 7 March 2026 (UTC)

Perhaps "if the chosen criteria" should be clarified, for what I want to say definitely does not imply you must choose one. I'm fine with pages such as "no criteria has been reached, come see us in the talk page so we can discuss". Or even nothing at all. It's the first time anyone bugged me with this kind of quest. Relatedly, thank you for this essay:WP:FETCH - it offered the perspective I needed. Selbstporträt (talk) 22:46, 8 March 2026 (UTC)

I think we're talking about different things.

OR is a potentially serious concern if someone wants to make up a novel topic for a list, like "List of Tuesdays in February with full moons". The dates of the full moons are verifiable and (therefore) not OR (because nothing that's 100% fully verifiable can be OR, by definition), but I don't think there are any reliable sources talking about full moons on Tuesdays in February. Therefore, the topic of the list would be OR, even if the individual entries aren't.

What's not OR is identifying the scope of an article (including stand-alone lists). You will never find a reliable source saying "The correct scope for this Wikipedia article is to include all the X but to exclude Y and Z". Therefore a statement like "In the List of books by Alice Expert, we're going to include all the books written by Alice alone or with a co-author, and that reference work she was the senior editor for, but we're going to exclude all the books containing just a chapter she wrote" is not something you can find in a reliable source. However, making that decision – setting the list's scope and selection criteria – is not original research. Some kinds of list topics are so common that they are never OR, e.g., lists of albums by musicians, lists of books by authors, lists of geographical entities at a given level (rivers in a continent, mountains in a country, etc.), but editors still have to decide the scope of the list-article and the selection criteria (every album the musician contributed to, or only the solo ones? What standard is being used to differentiate a tall hill from a short mountain?).

It's that second bit that we've been talking about. WhatamIdoing (talk) 16:41, 7 March 2026 (UTC)

It remains the case that we should not be saying "deciding the scope of a page is not OR", because it is possible to commit OR when deciding the scope. TompaDompa (talk) 17:14, 7 March 2026 (UTC)

Can you give me a realistic example of how editors would commit OR when identifying the scope of a list for a non-OR LISTN-compliant subject? WhatamIdoing (talk) 17:29, 7 March 2026 (UTC)

I think List of sovereign states would be a good example of a list where there are plenty of ways to include and exclude entries. It would likely not be too difficult to come up with a set of criteria that would produce a novel list. TompaDompa (talk) 17:36, 7 March 2026 (UTC)

The point is that the list itself needs to be supported by a source, not its extension. For countries, we'd have a list of all those who fit that criteria. For Norwegian garage bands? Not a chance. Selbstporträt (talk) 19:04, 7 March 2026 (UTC)

Because such lists have been made so many times, with all sorts of criteria, I'm not sure that Wikipedia editors would manage to come up with a completely novel set of criteria. But let's say that they do. The typical criteria for inclusion in such as list are:

1. The entity must have a permanent population.
2. The entity must have a defined territory, although it may be disputed.
3. The entity must not be subject to any other sovereign state.
4. The entity be recognized by at least one other sovereign state.

Now the Wikipedia editors add their own additional requirements:

1. There must be a Wikipedia article about the sovereign state ("blue link" requirement).
2. The permanent population must be at least 500 citizens residing year-round within the defined territory.

Let's stipulate for the purpose of this discussion that there are no reliable sources saying that only entities with a separate Wikipedia article are sovereign states and that there are no reliable sources setting a minimum population of 500 year-round residents.

Editors have imposed these criteria for practical reasons (e.g., ending efforts to add various Micronations), but they cannot be cited to reliable sources.

Have editors committed the sin of original research by enforcing their chosen rules? WhatamIdoing (talk) 19:13, 7 March 2026 (UTC)

"Criteria chosen for practical reasons" would be better than "subjective". Selbstporträt (talk) 21:45, 7 March 2026 (UTC)

Reasonable people can disagree about whether the requirement for a stand-alone Wikipedia article constitutes original research. WP:CSC specifically mentions it as common practice. A novel threshold for the permanent population is a much more dubious proposition. If 500 is okay to keep out entries that editors think don't belong, why should not 500,000 be okay? I think most people would agree that imposing the latter threshold (which would exclude e.g. Andorra and Iceland) would not be okay. It would be a novel definition, resulting in a novel list. It would be, yes, original research. TompaDompa (talk) 22:23, 7 March 2026 (UTC)

Originality applies to things like claims, to technical concepts, and inferences. We can't claim anything without being able to support it. We can't invent concepts out of thin air. (Substance is not far from OR.) We can infer a conclusion unless someone made it.

Originality does not apply to editorial choices like creating articles, presentation structure, or material limitation. Other factors are involved, with consensus as backstop.

We already have a whole section on the content policies for stand-alone lists. They're the same as always. All we need to do is to say that adding criteria must not break these policies. You're right that OR extends beyond what is said over there.

Your insistance on OR makes me wonder: perhaps we should take the time to write a whole paragraph on the OR question for criteria, with examples? We already have a subsection. We should have another one. Selbstporträt (talk) 22:42, 7 March 2026 (UTC)

I've started a section below. WhatamIdoing (talk) 00:19, 8 March 2026 (UTC)

Back to main discussion

I've done a lot of thinking on this. The overall list (particularly those with compound criteria) is itself inherently subjective unless some additional standard applies. The more common problems there are where it is unlikely to ever be sought out (e.g. List of red-haired football players who are over 6' tall) or POV (list of convicted rapists who are supporters of politician John Smith) . IMO a good criteria there is how likely a Wikipedia user would seek out that particular list for enclyclopedic type information. Next, if the scope of the list is so broad that it will never be even 10% covered, even with somewhat objective criteria ("list of in important world events in 1914") what's in the list will also be subjective because the 1% of those events that fulfill the criteria that made it in there are whatever the editor picked. Sincerely, North8000 (talk) 22:28, 4 March 2026 (UTC)

Totally agreed. Here would be one way I would conceptualize the ask:

1. Like every other Wikipedia article, stand-alone lists should introduce their topics.

When that introduction suffices to characterize all of the entries, you're good to go. This often happens with exhaustive lists, or with partial lists that seek to be exhaustive. This seldom if ever suffices for what we should call selective lists.

(We can link to these types of lists, of course.)

In cases of selective lists, editors need to introduce their selection criteria. They can do so right in the lead and make direct statements about the criteria by which members of the list were selected. They can write a section dedicated to it, called Selection criteria. They can introduce the criteria in the Talk page, using a template. They can do more than one of the above, as long as the criteria are the same at each place.

See for instance [...]

(I will never get bored of insisting that we give examples! That's what makes the MOS so great and so uncontroversial.)

2. The selection criteria ought to be designed for consistency: they should be simple enough for any new editor to get what needs to be satisfied to add another entry. For edge cases (there always are!), editors could be invited to discuss them in the Talk pages.

See for instance [...]

(I will never get bored of insisting that we give examples! That's what makes the MOS so great and so uncontroversial.)

3. When there is no selection criteria or when place is limited, editors may be asked to make their suggestions to the Talk page first. In that cases, the Talk page should include a reminder of the conditions that the new entries must meet.

See for instance [...]

(I will never get bored of insisting that we give examples! That's what makes the MOS so great and so uncontroversial.)

4. There is no fourth step.

That's it! We're done! No need to impose any more conditions for BLP pages. Everything is covered. Clean and direct.

Would that work? Selbstporträt (talk) 01:50, 5 March 2026 (UTC)

I see some parallels between this and the discussion Wikipedia talk:Stand-alone lists#"Ask yourself if any of the following are true" above, which is about the OP removing this content from the guideline. Specifically, the general theme seems to be an approach to the guideline that favors mechanical, rigid thinking over editors using common sense and their best judgement.

Unfortunately, the world doesn't fall neatly into discrete "objective and unambiguous" categories. We cannot require that the subject of all stand-alone lists have neatly defined boundaries such that editors will always and universally agree that X or Y definitely does, or definitely doesn't, belong on that list. ("Universality" is usually what people mean when they say "objective"; that is, everyone looking at a cat will agree that it's a cat and not a dog or a bird.) However, Wikipedia has to cover subjective information, and even more difficult, it has to cover information that is vague and fuzzy. For example: What is a restaurant? Is an ice cream shop a restaurant? What if it's just a kiosk or booth? How small can it be, and still be a restaurant? There is no universally accepted rule that says restaurants sell at least four kinds of food in a fixed location with a minimum size of 6 square meters and at least one chair, bench, or other seat – no matter what quality (except, probably, "food") you pick, there will be some restaurant that does all of these things except that one, and someone will say "But they don't charge money, so it's not a restaurant" or "Customers can't come inside, so it's not a restaurant" or "It doesn't have a fixed location, so it's not a restaurant". And yet we manage to write the article Restaurant, and we manage to make Lists of restaurants anyway. This is successful because most editors can and do use common sense and their best judgement to determine what belongs in each list, even if the "restaurant" turns out to be a food truck that gives away food for free. WhatamIdoing (talk) 03:15, 5 March 2026 (UTC)

Exactly. I suppose you know this story about how the burrito became a sandwich. Just in case you don't:

When you slap some meat inside two slices of bread, you have a sandwich, at least according to the U.S. Department of Agriculture, which enforces the safety and labeling of meat and poultry. "We're talking about a traditional closed-face sandwich," says Mark Wheeler, who works in food safety at the USDA. "A sandwich is a meat or poultry filling between two slices of bread, a bun or a biscuit." That excludes items like burritos, wraps or hot dogs. For this definition, Wheeler consulted the agency's 202-page Food Standards and Labeling Policy Book.

https://www.npr.org/sections/alltechconsidered/2014/06/26/325803580/what-burritos-and-sandwiches-can-teach-us-about-innovation

Not unlike what happens when you tell New Yorkers to only bring in subways dogs that fit in a bag... We don't need to burden ourselves with any of that: we just need a simple guideline to help newcomers, not a rulebook to galvanize enforcers!

Speaking of which, I note from the exchange the claim that no discussion means there's no consensus. I don't think that's true. Silence, to be the weakest consensus, needs to be consensus! This applies even to BLP matters, the only place where the burden of proof favors stonewalling. The idea of reverting an edit because there's no consensus is so counterproductive. Imagine: I make an edit, you revert it; I ask why, you say there's no consensus. I say OK, let's discuss it: why do you disagree? You say: oh no, I agree with you, it's just that there's no consensus! What kind of sweet hell is that?

Sometimes, like now, it makes more sense to go directly to the talk page. One can expect some pushback if someone would like to change a page like this one. So it saves time. Or does it?

I doubt the wiki would have gone as far as it actually went if we always had to have palavers each time we want to change a few lines from each page! Selbstporträt (talk) 05:20, 5 March 2026 (UTC)

I love the burrito-sandwich story. That might make a useful example for this guideline, or for an essay explaining this problem.

Silence can indicate consensus; however, as soon as someone objects, it's no longer silence. I agree that people shouldn't revert edits that they personally agree with. WhatamIdoing (talk) 19:12, 5 March 2026 (UTC)

"I revert because there's no consensus" presumes the question has already been discussed, resulting in no consensus. Otherwise it means "I revert because I object" without stating the objection. It's no better than to say "I revert because I revert, but don't know why yet". Editors should not fear change. Nor should they fear being reverted, and then working out a solution to preserve the work being done. So here's a second draft of what I just said:

Like every other Wikipedia article, stand-alone lists ought to introduce their topics. That often suffices to characterize all the entries of comprehensive lists, or of partial lists, both seeking to be exhaustive. This seldom if ever suffices for selective lists. A list of selection criteria allows editors to select entries for selective lists. They should be stated in the lead, in a section on the page, or in the Talk page. Criterias ought to be designed for consistency: simple enough to help any editor add relevant entries, and clear enough to reduce the number of edge cases. When no selection criteria have been established, editors may be asked to suggest entries to the Talk page first; the Talk page should then include a template reminding editors of the conditions the new entries must meet. See for instance [...]

Then we can follow up with the kinds of selection criteria worked best so far, and provide examples. Then, and only then, should discussion of fine points about Norwegian garage bands be made.

Again, it's just easier to simply try something (anything!) on the page itself and see how it goes. Small increments work best with more iterations, not with one suggestion per two days. We're not building a sacred text. It's more like a handout. Selbstporträt (talk) 02:53, 6 March 2026 (UTC)

We see people revert with explanations like "I support this change, but somebody else will object, so I'm reverting it". They should not do that.

I don't want to introduce a new term of the art like selective lists. All lists are "selective", in the sense that no lists contain everything or a random assortment of words. We need something much simpler/easier, and with more focus on "here's what's helpful" than on formalities that editors "should" do. WhatamIdoing (talk) 05:16, 6 March 2026 (UTC)

Either one lists all the elements of a set or one can pick the most salient ones. Can't do both. Were we to list all the defunct and relocated NHL teams, we don't need any "list" of inclusion criteria. We want *all* the NHL teams that satisfies the predicate. Readers and editors don't need to be told anything else. Were we in "edit mode", you wouldn't be able to revert all I just wrote based on equivocation. You could propose to call such list a selection, a collection, or else. You could describe it instead of calling it anything, say with "list with explicit inclusion criteria". (After two days, we should realize that to call the criteria a "list" creates confusion.) Since you need to preserve, your edit would naturally mark the legitimate scope of your disagreement.

Only if you insist would we need to come here. Then I would tell you that titles acts like a predicate, and that we need to state the scope of the page when that predicate isn't enough. We always provide our inclusion criteria for lists where titles may confuse editors. Once this is clear, we should see that there is absolutely no need to provide BLP persons any kind of special treatment.

We would go over examples of exhaustive lists, and of lists for which the selection criteria needs to be shown, for instance the list of prime numbers: "The first 1,000 primes are listed below, followed by lists of notable types of prime numbers in alphabetical order, giving their respective first terms." Go ahead: try to argue that the concept of prime number is somehow fuzzy. At some point I'm sure you'd get what I'm trying to say. Which is never clear when we just "talk" but without any non-verbal cues.

We could even come up with the conclusion that sometimes a page just needs a different title. Which is what the guy who "warned" me and other editors over a week now should have suggested instead of playing gotcha in various meta pages without doing much on an article page he must have edited by now. That changing titles may help needs to be added to the section we're currently trying to edit: it's a good way to solve recalcitrant inclusion criteria.

I'm not here to talk. I'm here to make sure I won't get bogged down by purported taskmasters ever again. I'm here to rock. Editing mode wins. If metapedians can't edit the rulebook over which they lord all day, those who spend their time here writing and editing surely can. Selbstporträt (talk) 13:18, 6 March 2026 (UTC)

About this: sometimes a page just needs a different title.

One of the things that I think many editors struggle with is matching the title to the scope. The process is:

1. Figure out what the article/page is about.
2. Choose a good title for that subject.

The dispute usually sounds like "Well, it says List of people who... at the top of the page and not List of people who have Wikipedia articles and who also...", so that means that 100% of every person ever needs to be in the list, and I get to add all of these non-notable/out-of-scope people, and if you want it to be restricted to notable people, then you have to rename it to List of notable people who... [which would violate Wikipedia:Stand-alone lists#Titles, BTW], because otherwise the article title controls the page content and the article title says the article is about everyone in the whole world".

See also Wikipedia:Scope first, then article title. You cannot choose a title until you already know what the article is about. While sometimes we do need a different title, often we need editors to write an article about the subject instead of writing one about the current title. WhatamIdoing (talk) 22:24, 6 March 2026 (UTC)

Another version of that paragraph has has been made below, in the #Gatherings section. I think it does everything we need. Selbstporträt (talk) 12:40, 12 March 2026 (UTC)

I wouldn't say the general theme [is] an approach to the guideline that favors mechanical, rigid thinking over editors using common sense and their best judgement. I would say it is an approach that favours relying on sources. I don't want us to tell editors to use common sense and their best judgment, because all too often that gets taken as a license to engage in WP:Original research, non-neutrality, and other policy violations.

Anyway, I think this discussion has got more than a little side-tracked. What do we want to do about the following paragraph?

Selection criteria (also known as inclusion criteria or membership criteria) should be unambiguous, objective, and supported by reliable sources. Avoid original or arbitrary criteria that would synthesize a list that is not plainly verifiable in reliable sources. In cases where the membership criteria are subjective or likely to be disputed, it is especially important that inclusion be based on reliable sources given with inline citations for each item.

Do we want to keep it as-is? Do we want to tweak it somehow? Do we want to do away with it entirely? My suggestion is changing it to the following:

Selection criteria (also known as inclusion criteria or membership criteria) should be unambiguous, objective, and supported by reliable sources. Avoid original or arbitrary criteria that would synthesize a list that is not plainly verifiable in reliable sources. If the choice of selection criteria for a list is subjective or likely to be disputed, then the criteria should be supported with citations. Likewise, if inclusion of an item in the list is subjective or likely to be disputed, a reliable source in the form of an inline citation must be provided for that list item. If the topic of the list is itself subjective or likely to be disputed, the criteria must be possible to implement consistently by applying an objective standard. The criteria may be outlined in the article lead, the article body, or on the talk page; when they are unambiguously clear from the article title, mentioning the precise definition of the criteria may be omitted altogether.

I think that would be an improvement compared to the current version. I am certainly open to revising it further, but it seems to me that it might be better to implement a change we can agree would be an improvement, even if we find it to be flawed, instead of holding out for an ideal version before we make any change at all. TompaDompa (talk) 14:17, 6 March 2026 (UTC)

I already said why the first three sentences don't work. The fourth sentence could stay, although it's only a reminder that applies for every disputed claim made in wiki-voice. The fifth sentence not only reiterates the contradiction that motivated all this, but it doubles down on taking "we'll list notable people" as somehow objective.

So I duly submit we should just make edits we see fit, and negotiate here if we can't make progress in edit mode. Selbstporträt (talk) 14:56, 6 March 2026 (UTC)

I have to admit that I find your position a bit difficult to glean from the large amounts of text and various digressions above, but at any rate the first three sentences don't work would amount to doing away with the paragraph entirely, then. Still, I'll take your advice and implement my suggestion. TompaDompa (talk) 15:12, 6 March 2026 (UTC)

Go right ahead. At the very least, we'll have other versions to talk about, and we'll have edit comments to orient ourselves. Nay not worry about my impatience: I'll cooperate, and will preserve as much as I can. I won't tell you "we don't have consensus". Promised. I'm sure there's a way to distinguish objective and subjective stuff if that's what you really want! Selbstporträt (talk) 15:16, 6 March 2026 (UTC)

I object to insisting that list-selection criteria be "unambiguous" and "objective". Here are my main reasons:

• Reality is not unambiguous; therefore, demanding that list-selection criteria be unambiguous is tantamount to demanding that Wikipedia's lists not reflect reality.
• Editors have different ideas of what "objective" means. The concept of objectivity itself is subjective: One editor will say that this criteria is objective, and another will say that it's not.

Instead, I think we should focus on encouraging criteria like "possible to implement consistently"; I would even like them to be "easy for editors to apply" and to "provide practical information".

Overall, thinking back to many years of discussions about this, what people generally want is to not have complexity and especially not to have discussions in which reasonable people have come to opposite conclusions. I imagine, e.g., that there are some discussions about List of people named in the Epstein files that many editors would prefer to avoid, and which some editors hope could be avoided by insisting upon by applying an objective standard. Even saying that the criteria should be plainly verifiable is going to produce disputes between editors who say that ____ is verifiable, and editors who say that it might technically be verifiable, but it's not plainly verifiable. WhatamIdoing (talk) 22:12, 6 March 2026 (UTC)

I have revised the first paragraph. Each edit has been justified. I believe the specification is correct, both informally and quasi-formally. Editors should be able to understand it, at least after stylistic revisions.

This took me around 45 minutes. For comparison, I estimate that writing the comments on this talk page took me between 4-6 hours. And I'm quite sure my editing comments are easier to follow. While talking here informs my editing, to proceed in the fashion of a rewriting system definitely helped me.

I will wait before getting to the second paragraph. There are other affairs to which to attend anyway. Until later. Selbstporträt (talk) 01:51, 7 March 2026 (UTC)

I looked into it, and stating that criteria should be "unambiguous, objective, and supported by reliable sources" goes back to this edit back in May 2010. I don't agree with changing it to "recognizable, natural, concise, and precise", and I don't agree with the rationale that these are the usual criteria we use for titles; titles of lists and list criteria are different, and different considerations apply.

I think this edit that adds "express the characteristics required for entries to be included in an article" is fine.

I disagree with changing "Avoid original or arbitrary criteria that would synthesize a list that is not plainly verifiable in reliable sources." to "The selection criteria should preserve the notability of the grouping; entries need not be notable." and the rationale replaced the reconstruction of WP:NLIST with a paraphrase of WP:NLIST; the formulation is more direct and cites the relevant guideline; this is a conflation of issues of WP:Notability and issues of WP:Original research.

I disagree with replacing "If the choice of selection criteria for a list is subjective or likely to be disputed, then the criteria should be supported with citations." with "If the criteria chosen rests on established norms or authorities, they should cite the relevant reliable sources." and the rationale replaced the subjective with appeals to established norms or authorities; these two versions are stating very different things and the latter is a much weaker requirement.

I disagree with removing "Likewise, if inclusion of an item in the list is subjective or likely to be disputed, a reliable source in the form of an inline citation must be provided for that list item. If the topic of the list is itself subjective or likely to be disputed, the criteria must be possible to implement consistently by applying an objective standard." and the rationale deleted two sentences: the first repeated WP:RS, and the second has already been introduced by WP:CRITERIA. Removing the first sentence removes the requirement about inline citations for entries. The rationale for removing the second sentence runs into the same issue as before with treating list criteria as though they were list titles.

I disagree with replacing "when they are unambiguously clear from the article title, mentioning the precise definition of the criteria may be omitted altogether." with "For comprehensive lists or partial lists, the description of the article's title should suffice: no need then to state criteria." as borderline incomprehensible—a straightforward reading of "comprehensive lists or partial lists" would be "lists that are exhaustive, and lists that aren't".

I think replacing "outlined" with "presented" (which was done by a different editor) was a good change.

In summary, I think nearly all of your changes were for the worse. I have undone the ones I disagree with. TompaDompa (talk) 13:51, 7 March 2026 (UTC)

Telling me that you disagree ain't enough. You got to give reasons too, preferably more reasons than statements expressing your disagreement. Disagreeing five times and offering one argument makes for weak support.

Also, I've received other feedback on these edits. They contrast to yours. Thus it's quite possible that incomprehension rests on the lack of shared background knowledge between us. By "Partial list", I was referring to lists that seek to be complete, but are not. Dynamic lists are of that kind. With the template, they look like it:

This is a dynamic list and may never be able to satisfy particular standards for completeness. You can help by editing the page to add missing items, with references to reliable sources.

We shouldn't call them comprehensive, for they're not. Yet they too seek completeness. The fact that there are lists that seek completeness implies that the paragraph that follows needs to be clarified too. For completeness doesn't imply lack of discrimination.

There are lists that don't seek completeness. Quite the opposite, in fact. It is especially for them that we have inclusion criteria. So the distinction I have in mind is as critical as it should be obvious.

I'll take a look at your edits, and expect to read more than "I disagree" in the comments that accompany them. Selbstporträt (talk) 15:29, 7 March 2026 (UTC)

I read your edit. The only comment provided is "see talk page". Presumably, this refers to the comment to which I just responded. To preempt further stonewalling, I made a small edit, with an argument I made here once, and twice in the comment. I will demonstrate my point one more time, using a real example of inclusion criteria:

The first 1,000 primes are listed below, followed by lists of notable types of prime numbers in alphabetical order, giving their respective first terms.

This is not a claim about the world, but about editorial choices. The very idea that this kind of claim needs to be supported by reliable sources is absurd.

So the claim that inclusion criteria should supported by reliable sources is incorrect. By the same token, the claim that selection criteria should be objective is also incorrect.

Also, we should bear in mind that inclusion criteria inherit the goals titles have. This inheritance isn't marked by insisting on using "unambiguous" and delinking to WP:PRECISE, which already addresses that issue. Compare and contrast:

(1) Usually, titles should unambiguously define the topical scope of the article, but should be no more precise than that.

(2) Usually, inclusion criteria should unambiguously define the topical scope of the article, but should be no more precise than that.

We can keep "unambiguous" if you prefer, but we need to say that it's the same logic that applies, and refer to it. Selbstporträt (talk) 15:52, 7 March 2026 (UTC)

In the old version, we said "Ideally, the selection criteria will be unambiguous, objective, and supported by reliable sources". This has become "They should be unambiguous". The guideline has gone from a statement of aspiration (this is the ideal state; get as close as you can under the sometimes messy circumstances) to a rule (the list is wrong and bad if it doesn't do what it ought to do).

This change is worse because it is farther from reality, which is not always unambiguous. Also, what's unambiguous to some people is confusing to others. WhatamIdoing (talk) 16:13, 7 March 2026 (UTC)

In fairness, this change revealed the category mistake behind the prescription. Perhaps a better example is a list of criteria stated in the Talk page: do they need to be supported by reliable sources too?Selbstporträt (talk) 16:16, 7 March 2026 (UTC)

(edit conflict) I don't think the claim that inclusion criteria should supported by reliable sources is incorrect simply because editors come up with size-limiting threshold criteria such as "the first 1,000 primes" or "entries need to be notable", and if you do it would be better to add a carve-out (as already exists at WP:CSC, mind you) than to remove the notion that criteria, like all content on Wikipedia, should be supported by reliable sources. I've added such a carve-out. By the same token, the claim that selection criteria should be objective is also incorrect. similarly falls flat when you consider that WP:CSC says Lists are commonly written to satisfy one of the following sets of objective criteria.

Maybe your idea about complete versus partial lists is a good one. The text you added was not. As you say, it's quite possible that incomprehension rests on the lack of shared background knowledge—instructions should be readily understandable. The jargon of "comprehensive lists" and "partial lists" is not one that is commonly understood in the sense that you mean them in the context you presented them. Moreover, I don't think it is even necessarily true that the description of the article's title should suffice for those lists—see List of sovereign states for an example of a list that seeks completeness within its scope but where additional clarifying criteria are described (and probably necessary). That the title is unambiguously clear is what makes explicitly described criteria unnecessary in some cases, not that the list is what you term a "comprehensive list or partial list". That's why my version is better.

My version's "Avoid original or arbitrary criteria that would synthesize a list that is not plainly verifiable in reliable sources." is about something different than your version's "The selection criteria should preserve the notability of the grouping; entries need not be notable." My version is about WP:Original research, whereas yours is about WP:Notability. These concepts are not interchangeable here. If you think it is important to mention notability here we can do so in addition to the stuff about original research (but notability is already mentioned in the "Common selection criteria" section), but we cannot simply replace one with the other as though they are saying the same thing, because they aren't. That we ensure that we do not engage in original research when covering notable topics is more important than ensuring that the topics we cover are notable, because OR violations are more serious than notability violations (policy violations are typically more serious than guideline violations, though not necessarily always). That's why my version is better.

My version's "If the choice of selection criteria for a list is subjective or likely to be disputed, then the criteria should be supported with citations." is a much stronger requirement than your version's "If the criteria chosen rests on established norms or authorities, they should cite the relevant reliable sources.", because "subjective or likely to be disputed" is a much lower threshold than "rests on established norms or authorities" (which I also think is rather unclear in its meaning). That's why my version is better.

My version's "Likewise, if inclusion of an item in the list is subjective or likely to be disputed, a reliable source in the form of an inline citation must be provided for that list item." is not redundant. It explicitly states that some entries need citations to show that they meet the criteria. That's why my version is better.

My version's "If the topic of the list is itself subjective or likely to be disputed, the criteria must be possible to implement consistently by applying an objective standard." is not redundant. It covers enforceability, and explains what to do about subjective topics. That's why my version is better.

we should bear in mind that inclusion criteria inherit the goals titles have We certainly shouldn't be treating inclusion criteria as titles, which is what you are proposing. Consider e.g. Dinosaurs in Jurassic Park—the title prioritizes concision and naturalness over precision and accuracy. That list includes entries that are not dinosaurs (e.g. Mosasaurus), and excludes avian dinosaurs (i.e. birds). The same considerations do not apply to titles and list criteria. That's why we shouldn't say that it's the same logic that applies, and refer to it—because that's not correct. TompaDompa (talk) 17:05, 7 March 2026 (UTC)

"I don't think "the claim that inclusion criteria should supported by reliable sources is incorrect" simply because editors come up with size-limiting threshold criteria such as "the first 1,000 primes" or "entries need to be notable""

I should start to count the number of times you argue by assertion.

Here's the inclusion criteria of Dinosaurs example:

The following list includes on-screen appearances. Some animals listed here have also made prior appearances in the novels.

Do you have a reliable source to support that claim?

Not the appearances themselves. The claim that the following list includes on-screen appearances.

You're conflating the selection function with its domain. That's a type error. Selbstporträt (talk) 17:43, 7 March 2026 (UTC)

Hold on, do you think "inclusion criteria should be supported by reliable sources" means "sources must state the criteria"? TompaDompa (talk) 17:48, 7 March 2026 (UTC)

I changed "supported by reliable sources" to "reflect reliable sources". Does that resolve your concerns? TompaDompa (talk) 17:52, 7 March 2026 (UTC)

That's not what I'm saying at all!

Read the first sentence again:

Selection criteria (also known as inclusion criteria or membership criteria) express the characteristics required for entries to be included in an article

I'm saying that we should not conflate what the criteria express and the characteristics themselves! I'm trying to stick to the expression of the criteria.

To repeat, it is the expression of the criteria that needs to be precise, concise, etc. Just like titles need to. It would make no sense to say that the number 7 (one of the entries in the list of prime numbers) is precise, concise, etc. Just like it would make no sense to say that criteria should be objective, always sourced, etc.

You keep focusing on characteristics when it's obvious we can't subsume them under any clear umbrella: they span from established norms to properties that belong to the inner working of Wikipedia itself, e.g. N! Selbstporträt (talk) 18:15, 7 March 2026 (UTC)

NB. I don't consider WP:N an established norm, like say a dinosaur taxonomy. We can't cite Wikipedia policies to support claims made in articles anyway. Even if we could, they rest on dynamic pages: that'd be a nightmare. Selbstporträt (talk) 18:37, 7 March 2026 (UTC)

I apologize for misunderstanding you. I can't say I find the expression of the criteria to be a particularly interesting or important subject, leastways not compared to their substance. The substance of the criteria is where providing guidance to editors, in my opinion, is much more important. Saying that the criteria should be "unambiguous" gets us most of the way there when it comes to their expression. Anyway, WP:CRITERIA says that titles should have the following five characteristics: (1) recognizability, (2) naturalness, (3) precision, (4) concision, and (5) consistency. Of these, only precision is important for list criteria (and largely covered by "unambiguous"). Concision is important for titles because of the limited space available, which is a much lesser concern for list criteria. The other three are important for titles to help readers find the articles they want and recognize that they have arrived where they think they have, and those are basically non-issues for list criteria. As I've said, titles of lists and list criteria are different, and different considerations apply. TompaDompa (talk) 19:00, 7 March 2026 (UTC)

"I can't say I find the expression of the criteria to be a particularly interesting or important subject"

Nobody forces you to edit this page. Titling is the first way we delimit the domain of discourse of an article. When that scope does not represent its entries, then we need to tell readers and editors about our selection criteria.

Once we have them on the main page, we minimize WP:SURPRISE. When we have them on the talk page, we guide future editors who would like to add entries to a stand-alone list.

Once we take care of how we express our criteria, then we turn to what we consider valid domain restrictions. There are three main cases:

Case 1: we have established norms or authorities: we cite them and we're done.

Case 2: we're on our own: we cite internal policies, like we will need to do for just about every single BLP page.

Case 3. we have comprehensive lists and dynamic lists: criteria are may be facultative.

There is no need to say what we mean by objective and subjective when we can directly specify the relevant cases where the various criteria apply. These cases should cover the range of everything we need.

Unless you can find examples of criteria that would break this specification. If you don't, then that's what we should write, for you'd be left with no argument.

PS: I tried to evoke various fields for my analogies to get through. It's hard for me to tell where you come from. Selbstporträt (talk) 19:24, 7 March 2026 (UTC)

Of course nobody forces me to edit this page. But I am interested primarily in instructions that relate to the substance of the criteria, not the formulation of the criteria. That's why I started this section in the first place.

Let's say, for the sake of argument, that the only valid domain restrictions are external norms/authorities on the one hand, and internal policies on the other. Writing then in our instructions here that The complete set of valid domain restrictions consists of external norms and authorities and internal policies. would not be helpful. It would be hopelessly opaque to the uninitiated. I think it is fair to say that you have a tendency to write in a comparatively opaque manner, using a fair amount of jargon and otherwise technical language. When it comes to writing policies and guidelines on Wikipedia, this is unhelpful (in other contexts such as lawmaking and academic writing, it can on the contrary be a necessity). One of the functions of Wikipedia's policies and guidelines is to provide instructions to editors who may not yet have become familiar with Wikipedia-specific terminology, in order to help them contribute to Wikipedia productively. Another function is to have something to quote when disagreements arise. In both of these cases, it is helpful both to use relatively plain language, and to illustrate how more general principles apply in more specific situations (so there is no need to "derive it from first principles", so to speak). TompaDompa (talk) 23:31, 7 March 2026 (UTC)

Perhaps you should not conflate what I write here and what I write on the main page. Also, you shouldn't presume to write plainly. If we compare the two paragraphs we wrote side to side yesterday, there are fair chances you'd win neither on style or clarity. So please rest assured that I definitely don't think "we should cite sources for our criteria, except for most pages for which this section will be useful" is a masterclass of elegance.

Besides, you definitely are not the most flexible editor either, and your ways of trying to get what you want are not forthright. I keep telling you to stick to arguments, and that's just not something you seem willing to do.

If the only role that suits you is to lead write, I don't mind going through your output. As long as it leads us to a better section, think whatever you want.

So no more patronizing crap, pretty please with sugar on it. Selbstporträt (talk) 23:56, 7 March 2026 (UTC)

I was thinking of both what you write here and what you write on the main page. For instance, For comprehensive lists or partial lists, the description of the article's title should suffice: no need then to restrict the list with inclusion criteria. and Criteria also ought to facilitate entry selection were written on the main page. I'm not saying that I'm necessarily better at writing this stuff, I'm trying to explain what my goals are. Ideally, we would work together to improve the text incrementally and eventually arrive at a version that resolves all of our concerns. That may not be possible, of course—we have might have some aims that are mutually exclusive. Likewise, I do not mean to be patronizing; I apologise if that's how I come across. I think you are at times patronizing, both in the things you say and the ways you choose to say them. I would genuinely prefer collaborating in good faith and in good spirits.

I think I shall have to ask you to elaborate on your ways of trying to get what you want are not forthright. I keep telling you to stick to arguments, and that's just not something you seem willing to do. TompaDompa (talk) 01:04, 8 March 2026 (UTC)

Reread my first comment today. You should see that I emphasize five "I think" and only one argument. That's the opposite of being helpful. What you think is secondary. It's why you think so that matters. Then reread my second comment today. I emphasized that your comment to your edit was "See talk page". Which means you have bulldozed a series of edits and dismissed their justification with the sheer power of your opinions. This is worse than unhelpful. Repeat that a few times and it becomes disruptive.

"For comprehensive lists or partial lists, the description of the article's title should suffice: no need then to restrict the list with inclusion criteria" may certainly be improved, but it parses well enough. The concept of comprehensive list has the advantage of being easy to understand (all-inclusive, complete, exhaustive) and formally sound. It just means we enumerate every single element from the list. If you prefer a synonym, suit yourself. A partial list is the same, except that we have yet to enumerate all the elements. There many such lists, including dynamic lists, which seemed to surprise you.

All in all, that just means that requiring an explicit inclusion criteria for an all-inclusive list would be silly. Which is something we seem to be agreeing on: "when they are unambiguously clear from the article title, mentioning the precise definition of the criteria may be omitted altogether." How is that sentence any clearer: it's verbose, "unambigously clear" makes little sense, you introduce "definition" for no reason, precision is not enough, and "altogether"? It's only clearer to you because you wrote it.

Which is why it matters that we co-edit. A piece of text can always be improved. Granted, it can also be worsened. On the aggregate, more people is better to fight kudzu.

That being said, what I wrote didn't deserve your morning comment. I have a thick skin and it was to be expected. Just don't push it. I definitely can respond in kind. Selbstporträt (talk) 01:43, 8 March 2026 (UTC)

The clause "no need then to restrict the list with inclusion criteria" is a sentence fragment. It does not parse well enough.

(Gentle reminder: What one person believes is the opposite of being helpful is not necessarily the opposite of being helpful to everyone else, and we will catch more flies with honey than with vinegar.) WhatamIdoing (talk) 05:21, 8 March 2026 (UTC)

I did see Disagreeing five times and offering one argument makes for weak support. I'm not sure we have a common understanding of "argument" in this context. You see one argument, wheras I see more than one—and it is not immediately obvious to me which one is the only one you see.

All in all, that just means that requiring an explicit inclusion criteria for an all-inclusive list would be silly. I don't think that's the case. As the "Lead" section says, "Even when the selection criteria might seem obvious to some, an explicit standard is often helpful to both readers, to understand the scope, and other editors, to reduce the tendency to include trivial or off-topic entries."

Which is something we seem to be agreeing on: "when they are unambiguously clear from the article title, mentioning the precise definition of the criteria may be omitted altogether." How is that sentence any clearer: it's verbose, "unambigously clear" makes little sense, you introduce "definition" for no reason, precision is not enough, and "altogether"? It's only clearer to you because you wrote it. I agree that "precise definition" is not a terribly good phrasing and that "altogether" is not strictly necessary. Those are things relatively minor rephrasings are good at addressing. As for "unambiguously clear", the reason that exact phrasing was used is that the "Lead" section says A stand-alone list should begin with a lead section that [...] makes direct statements about the criteria by which members of the list were selected, unless inclusion criteria are unambiguously clear from the article title. I think my version is clearer because it does not require the reader to know the specific meanings of "comprehensive lists" and "partial lists" that you intend (it may be the case that The concept of comprehensive list has the advantage of being easy to understand (all-inclusive, complete, exhaustive) and formally sound. It just means we enumerate every single element from the list. If you prefer a synonym, suit yourself. A partial list is the same, except that we have yet to enumerate all the elements., but it is not terminology that is widely used here), and it matches the "Lead" section.

I also think "may be omitted" is better than "should suffice: no need then to restrict the list with inclusion criteria" because it errs on the side of presenting the criteria explicitly. This is more in line with what the "Lead" section says (and I think the "Lead" section has it right).

I don't know why you think I am surprised at the existence of dynamic lists.

As a sidenote, it's a bit difficult to know for sure which comments you mean when you speak of "today" and times of day. I don't know which time zone you are in.

I'm not trying to be difficult here. I genuinely find it hard to communicate effectively with you—whether that's due to a problem on my end, yours, or both. I try to make sure I understand you, and you me. Thus far, with limited success. TompaDompa (talk) 12:33, 8 March 2026 (UTC)

No hard feelings. Sorry for being terse or prolix at the wrong times. Will try to be more considerate. Selbstporträt (talk) 15:48, 8 March 2026 (UTC)

Should be objective, but they are subjective

The wording under discussion is this bit:

Selection criteria...should be unambiguous, objective, and reflect reliable sources. If the choice of selection criteria for a list is subjective...

As you can see, we have a paragraph stating:

• that the criteria "should" be objective, and
• that the criteria sometimes aren't objective.

Are we all agreed that One of These Things (Is Not Like the Others)? Does anyone want to say that these will never be confusing or seem contradictory to editors?

I think the way to resolve this is to stop saying that criteria "should" be objective. Maybe "For more lists, editors usually choose criteria that are objective"? WhatamIdoing (talk) 20:23, 7 March 2026 (UTC)

"For more lists"

We don't know that. We probably can't. But if there's a bet, I'll take the under and point at BLP pages. Selbstporträt (talk) 21:14, 7 March 2026 (UTC)

I think we know that a majority of lists have objective criteria, because most lists are so boring. (Take a look at some lists, and see how many you think require any description beyond the title.) WhatamIdoing (talk) 22:51, 7 March 2026 (UTC)

If the description of the concept expressed by the title suffices, then I don't see any criteria being applied. Otherwise we would not have a discussion on selection criteria here, but on WP:AT. Are there selection criteria for other types of pages? Selbstporträt (talk) 22:58, 7 March 2026 (UTC)

Yes, but we tend not to write them down, or even discuss them, and when we do, they're not called "selection criteria". For an exception to the not-writing-it-down practice, see the archives of Woman, whose scope was determined through many discussions to include women by any definition (biologically, psychologically, social role, etc.). Another fairly common example is discussions about whether "_____" needs to be renamed as "_____ in the United States", on the grounds that the current content is biased towards the US. For example, I have had a years-long struggle at Standardized test trying to keep the article from being taken over by specific US-only tests for school-age children/teens. The scope of Standardized test is meant to be global and to apply to all ages. Deciding whether to have an article about Standardized test, Standardized testing in the United States, or Standardized testing in education is an editorial decision equivalent to setting the selection criteria for a list. WhatamIdoing (talk) 23:52, 7 March 2026 (UTC)

Interesting. My own intuition makes me distinguish the two, but I've just seen that what I'd call a selection criteria can be put right in the title:

List of women's footballers with 100 or more international caps

There is also a section on FIFA criteria.

Alright. No need to be that precise.

We could mention this in the subsection that is useless for now. Selbstporträt (talk) 00:25, 8 March 2026 (UTC)

@WhatamIdoing: The intended reading is one that differentiates between "selection criteria" on the one hand, and "the choice of selection criteria" on the other. Say, for instance, that we have a list of large XYZ. For such a list, there must be an objective measure by which XYZ is considered "large". Which measure to use for size (length, area, volume, mass, et cetera) and at which threshold XYZ may be considered "large", on the other hand, can be up for debate—or subjective. Is there a different way of phrasing this that you think would more clearly convey the intended point? TompaDompa (talk) 21:56, 7 March 2026 (UTC)

See List of largest cities:

The United Nations uses three definitions for what constitutes a city, as not all cities in all jurisdictions are classified using the same criteria. Cities may be defined as the cities proper, their metropolitan regions, or the extent of their urban area. A complicating factor is that many large cities in the world have not only homeless or the unhoused, but also vast slum communities. This leads to official census data being less accurate in representing the actual number of residents in a given area.

See also List of richest Americans in history:

Comparing wealth of individuals across large spans of time is difficult, as the value of money and assets is heavily dependent on the time period. There are various methods of comparing individuals' wealth across time, including using simple inflation-adjusted totals or calculating an individual's wealth as a share of contemporary gross domestic product (GDP). For this reason, there is not one decisive ranking of the richest Americans in history.

There is no "must be". There is only a "what we can do". Selbstporträt (talk) 22:05, 7 March 2026 (UTC)

For a List of large XYZ, I don't think we need an objective measure. That's one good way to go about it (and it's the one we use at List of tallest buildings), but another way to go about it is to have the criteria be any XYZ that is called large by a reliable source. WhatamIdoing (talk) 22:59, 7 March 2026 (UTC)

Fair enough, but "called large by a reliable source" can also be considered an objective measure in some sense. For the purposes of having enforceable criteria that don't boil down to editorial judgment, it does the trick.TompaDompa (talk) 23:04, 7 March 2026 (UTC)

List of video games listed among the best uses that standard, and there are endless disputes from editors who say the list is inherently non-objective. You say it's "objective", but other people seem to have a different idea of what the word objective means. WhatamIdoing (talk) 23:54, 7 March 2026 (UTC)

List of video games listed among the best does not use "called the best by a reliable source" as the threshold for inclusion. It uses "included on at least six separate best-of lists from different publications (inclusive of all time periods, platforms and genres)". The issue there—at least as I see it—is not whether the criteria themselves are objective(ly enforceable), but whether the set of criteria used (and thus, the list that they give rise to) is novel. More to do with Avoid original criteria that would synthesize a list that is not plainly verifiable in reliable sources. (and perhaps with Selection criteria [...] should [...] reflect reliable sources) than with Selection criteria [...] should be [...] objective.

But yes, List of video games listed among the best is an example of a list that covers a topic that is itself subjective (which video games are among the best is of course subjective). I think Wikipedia should allow lists on such topics. I don't think Wikipedia should allow such lists to have criteria that boil down to editorial judgment in their implementation, nor do I think the criteria should be allowed to be chosen so freely as to effectively WP:Synthesize a list that is not plainly reflective of the sources (reasonable people can disagree about where to draw that line). This is one of the things I would like this page to provide clear guidance on. TompaDompa (talk) 00:15, 8 March 2026 (UTC)

Do you think that "included on at least six separate best-of lists from different publications" constitutes "editorial judgment in their implementation"? WhatamIdoing (talk) 00:29, 8 March 2026 (UTC)

Not in the implementation, no. Editorial judgment is not needed to determine whether that criterion is met for a particular entry. TompaDompa (talk) 00:37, 8 March 2026 (UTC)

It's possible to have editorial judgment without ever breaking originality.

I prefer "requires editorial judgment" to "subjective". I also prefer it to "could be disputed", for anything can be disputed. Classification accuracy could test for that.

When criteria are recognizable, natural, concise, and precise, editorial judgment should be fine for the most part. Selection can be sometimes easier, sometimes harder, but whatever we'll say to this effect will amount to "easier is better than harder", with is as useful as a prescription as fires are hot. Selbstporträt (talk) 00:56, 8 March 2026 (UTC)

TompaDompa, I think that figuring out whether two publications are really different, and whether they're reliable, and whether a list with a somewhat different headline constitutes a "best-of list" are all things that require "editorial judgment in their implementation". WhatamIdoing (talk) 05:03, 8 March 2026 (UTC)

@WhatamIdoing: I can see that, though it is a somewhat broader notion of "editorial judgment" than I had in mind. There would not seem to be much that wouldn't require editorial judgment by that interpretation? We always have to assess whether the sources we use are fit for purpose, for instance. TompaDompa (talk) 12:35, 8 March 2026 (UTC)

Yes, editors always have to use some amount of editorial judgment. Since we have many inexperienced editors who attempt to apply rules mindlessly (and frequently don't even know what the rules actually are), I think it is particularly important to to acknowledge here that using editorial judgment is valid and sometimes important in this context. WhatamIdoing (talk) 19:40, 8 March 2026 (UTC)

This may be where our perspectives diverge due to differing experiences. In my experience, editors who attempt to apply rules mindlessly is not much of a problem compared to editors failing to apply rules when they should. TompaDompa (talk) 18:26, 11 March 2026 (UTC)

I also prefer Y "editorial judgment" to "subjective". WhatamIdoing (talk) 05:07, 8 March 2026 (UTC)

Then we could and perhaps should add this in a the subsection on examples! One possible candidate: List of metric units. Also, check the section List of tallest buildings#Ranking criteria and alternatives.

I think @WhatamIdoing: needs us to recognize that outside policy, editors decide for themselves anyway. We're only here to offer guidance. We could encourage them to pick criteria based on the finest instrumentration or models, even then it's hard to know which one is best most of the times, and in the more models we use, the merrier.

So by objective criteria you mean: a model? Selbstporträt (talk) 23:13, 7 March 2026 (UTC)

I agree with this: editors decide for themselves anyway. Editors have to take whatever complex and contradictory reality they're working with, and come up with something that they can agree to live with. And in the end, consensus is king. WhatamIdoing (talk) 23:59, 7 March 2026 (UTC)

I thought we all agreed that it was Queen! Selbstporträt (talk) 00:03, 8 March 2026 (UTC)

I don't agree that in the end, consensus is king, and I think it's rather important that it is not. There are some things that are not subject to consensus, including matters of WP:BLP, WP:COPYRIGHT, and WP:NPOV. TompaDompa (talk) 00:21, 8 March 2026 (UTC)

The WMF board demands that we have a BLP policy, but doesn't dictate its contents beyond some vague statements. WMF Legal will enforce copyright via DMCA takedown notices if we don't. NPOV is technically something we can change if we ever (lose our minds and) decide to do so, despite the verbiage in it about the neutral point of view being non-negotiable.

The role of consensus is inherent in the design of a wiki. Consensus, in the sense of reaching an agreement, is the only thing that will stop people from trying to change the page to something else. If we don't have an agreement, and a good, shared understanding of what that agreement is, then nothing else matters. WhatamIdoing (talk) 00:27, 8 March 2026 (UTC)

I thought my "outside policy" was clear enough. My bad. Selbstporträt (talk) 01:00, 8 March 2026 (UTC)

Your sentence was ambiguous. "Outside policy" ("Außerhalb der Richtlinien"?) could mean that editors make all decisions except for writing the Wikipedia:Policies and guidelines. It could also mean "so long as they meet the requirements of policy", editors make all decisions. WhatamIdoing (talk) 05:15, 8 March 2026 (UTC)

Indeed. Guidelines are also outside policy, "beyond" was more appropriate. I find it hard to believe that the five pillars are real policies. They're more like principles. I might be biased, for I can be accused of many things, but not being a knight!

This page mustn't be written or interpreted as policy. It's a guideline, i.e. "best practices supported by consensus". Selbstporträt (talk) 06:24, 8 March 2026 (UTC)

Wikipedia:Five pillars is not a {{policy}} page. WhatamIdoing (talk) 19:41, 8 March 2026 (UTC)

@WhatamIdoing: This is perhaps getting a bit overly abstract. For the purposes of this page, what is important is that the instructions to editors do not suggest that anything goes as long as there is consensus for it. Experience has shown that such an approach to establishing criteria for a list can create quite a mess. TompaDompa (talk) 12:37, 8 March 2026 (UTC)

"anything goes as long as there is consensus for it"

That's exactly what you did yesterday with your edit. That applies to any claim that has yet to be challenged. Without that liberty, the whole project changes. Without your first move, we wouldn't be where we are. Thank you for that. Selbstporträt (talk) 15:41, 8 March 2026 (UTC)

No, that was being WP:BOLD in WP:Projectspace, not "anything goes as long as there is consensus for it" in WP:Articlespace. Policy tells us that there are some things that are not subject to consensus. Policy also tells us that claims that are likely to be challenged are to be treated the same as claims that have already been challenged (leastways in terms of WP:V). TompaDompa (talk) 15:54, 8 March 2026 (UTC)

Policy doesn't tell us anything. It's not a person. It's not even a thing. It's just a rhetorical trick you use to evade what I just said. Since you don't seem to have read anything on the page except the section you work on to strengthen, allow me to quote:

This page documents an English Wikipedia content guideline. Editors should generally follow it, though exceptions may apply. Substantive edits to this page should reflect consensus. This page in a nutshell: Stand-alone lists, like other articles, are subject to Wikipedia's Core content policies. They should be used for appropriate topics, and have clear selection criteria. A well-written lead section is important, especially if there is little or no other non-list content.

I believe I made my disagreement quite clear regarding your policy changes. Strictly speaking, you should not have done so. Let's say it's an exceptional measure to bypass the blockade that is happening at least since the exchange with the gentle proposal from @Transhumanist: ut supra. We should add it in the Examples section, or in the Relevance subsection.

The stone wall is behind us, now comes the slow grind. You still cling on "objective", you still use "unambiguously clear", and you carved two exceptions to make sure your "should" remains on the page. You forgot a third exception: when the list of criteria is in the talk page. We don't need any damn reliable sources for talk pages. They're talk pages!

Yet, I commend you for how it's going. You conceded that we should introduce the concept of list of inclusion criteria. We now have Transhumanist's characterization. It fits well. We also have the description I borrowed from WP:AT. It took a day to distinguish meta and object language (my fault, really) but it's still there. Before that there was the week spent on distinguishing the various ways criteria could be announced on the page. You wrapped up these concessions and put them first.

Nevertheless, you still are trying to follow up with implementing an obligation that has never existed. That, my friend, is original research. This seems to be prompted by the need to impose on BLP page creators the requirement to introduce their inclusion criteria. That's not here or there in the original text. So I'm not sure where you're trying to go with that.

Your reading of "anything goes" is too broad: it only includes what is under the editors' remit. We all know that there is a layer of obligations above that. But it's hard to tell where principles end and when legislation starts. Even there just about anything goes.

The point is that we, on this page, cannot offer anything more than guidance. They're guidelines, not policies. They could be instructions, as you say but perhaps not as you intend. They *inherit* policy restrictions. Read back the quote.

So here we don't need to mention *anything* about policy except - refer to "Stand-alone lists, like other articles, are subject to Wikipedia's Core content policies". That has already been said on the page. Hence why I linked to the sections Notability and Content policies. These links have disappeared from the section for a while.

Why? Selbstporträt (talk) 18:22, 8 March 2026 (UTC)

Kudos for retracting your WP:ASPERSION unprompted. I don't know if it makes any difference to you at this point, but you have it wrong about my motives. I suspect the well has been poisoned too thoroughly already, though.

you carved two exceptions to make sure your "should" remains on the page. You forgot a third exception: when the list of criteria is in the talk page. We don't need any damn reliable sources for talk pages. They're talk pages! I'm guessing this refers to the explanatory footnote that reads Size-limiting measures, such as only including notable entries (see #Common selection criteria) or restricting ranked lists to a predetermined number of entries, need not derive from similar thresholds in reliable sources. That's not two exceptions, that's two examples of a single exception. If you think that's two exceptions (and don't see that "such as" means that it is non-exhaustive), you must think that "when the list of criteria is in the talk page" fits alongside "only including notable entries" and "restricting ranked lists to a predetermined number of entries". It doesn't. The examples I gave make the list smaller. Having the criteria on the talk page doesn't make the list smaller.

You conceded that we should introduce the concept of list of inclusion criteria. I have no idea why you think that's a concession on my part. This isn't a WP:BATTLEGROUND—it's entirely possible to simply agree or not have any strong opinions one way or the other.

Nevertheless, you still are trying t follow up with implementing an obligation that has never existed. *That*, my friend, is original research. No, it most certainly does not fall under the heading of WP:Original research. That's just you using a rhetorical trick.

This seems to be supported by some concern prompted by the need to impose on BLP page creators the requirement to introduce their inclusion criteria. But that's not here or there in the text. So I'm not sure where you're trying to go with that. I would say that I don't know why you think this is about BLP pages specifically for me, but you have kind of tipped your hand already so I think I do. Anyway, you are wrong. TompaDompa (talk) 18:28, 11 March 2026 (UTC)

Hello again,

What is retracted refers to a moment that starts here:

There must in principle be some way of determining whether an entry belongs on the list or not, yes. I wouldn't be opposed to adding something to state this outright if others feel it necessary to do so, but frankly I think it goes without saying.

It follows up there:

We now have "The criteria may be outlined in the article lead, the article body, or on the talk page; when they are unambiguously clear from the article title, mentioning the precise definition of the criteria may be omitted altogether." to quote. We'll see if it gets tweaked further by others.

The second quote omits the most objectionable bits. As you can see, motivations have nothing to do with what I was alluding to. On the other hand, anything that starts with "If you think" rests on mental attribution. They should be taken off talk pages; search for Apophasis as to why the "if" doesn't change that. When people say "I believe", it's more a way to attenuate the certainty of their assertion, it's just polite.

Reading what is on the page saves a lot of mindreading time. For instance, That's not two exceptions, that's two examples of a single exception or No, it most certainly does not fall under the heading tells me everything I need to know as to where you are directing this exchange. It won't work. Here's where it'll go instead: every time you'll make a quote fest, I will repeat all the parts you omit. For instance: everyone here ought to know by now that OR does not apply to talk pages. Note: I said "original research", not WP:OR. We should not conflate original research with WP:OR. This nitpick evades my point, which is that you are not supposed to strengthen a policy or a guideline without consensus.

As for how you count exceptions, it doesn't matter one bit: one is already too many. Our specification is too simple to have any. Why repeat MOS:SALLEAD if it's to introduce more inconsistencies? We should rather solve the one over there first: selection criteria are implicit in the first sentence of leads. Definitions by comprehension work by description, not naming. Solve that, solve most problems we have been discussing. Selbstporträt (talk) 19:17, 11 March 2026 (UTC)

You said I duly submit we should just make edits we see fit, and negotiate here if we can't make progress in edit mode. You can hardly complain about others making the edits they see fit—edits that you, or indeed any other editor, could revert at any time as not being aligned with consensus—after that.

The retracted bit said that I was sent by somebody else.

You omitted that the part before "If you think [...]" started with "I'm guessing this refers to [...]". I find it genuinely difficult to tell what you mean a lot of the time. I can make a best guess and respond to it, or I can ask you to clarify. Here, I made a best guess and responded to it. I twice noted in my response that I'm not sure whether I'm interpreting you correctly: First with "I'm guessing this refers to [...]", and then with "If you think [...]".

You used the phrase "original research", which has a specific meaning on Wikipedia outlined at WP:Original research. That's a rhetorical trick. You said that I used a rhetorical trick. I'm saying you use rhetorical tricks. TompaDompa (talk) 19:49, 11 March 2026 (UTC)

Writing "No U" might have saved you time. I disputed your edits before you made them. I offered reasons then, and I offered more after. I just did offer more reasons, in the comment you responded by "No U".

In the middle of the first quote you just provided there is and negotiate here. Is this your way to negotiate? Selbstporträt (talk) 19:59, 11 March 2026 (UTC)

Right now I believe we are discussing conduct rather than content. I don't think there is much hope of negotiating between the two of us, given that you do not appear to believe that I am acting in good faith, but I will keep trying: see e.g. this. TompaDompa (talk) 20:29, 11 March 2026 (UTC)

You misread what I said. I explained you why. And you still evade the last two points I made:

(1) you are not supposed to strengthen a policy or a guideline without consensus;

(2) I disputed your edits before you made them, offered reasons when you did, and offered more after; I still offered more in the recent comments you are using for your food fight, and in fact I still do.

Over and out. Selbstporträt (talk) 20:57, 11 March 2026 (UTC)

I took I duly submit we should just make edits we see fit, and negotiate here if we can't make progress in edit mode. as meaning "Let's edit the page as we see fit, altering/undoing the parts of each other's edits we disagree with as we go, and go to the talk page only when we reach an impasse". Perhaps it was then ill-advised to present my rationales at the talk page rather than in edit summaries. Nevertheless, it was my intention to go by the WP:EDITCONSENSUS approach just described. You can still edit the page if you think I have strengthened the guideline without consensus (which I haven't done intentionally—my intention has been to clarify it and to avoid weakening it), just as I can edit the page if I think you have strengthened or weakened the guideline without consensus. At this point it seems fairly likely that some other editor will just restore the status quo ante and declare the whole thing a complete mess, but still. TompaDompa (talk) 21:26, 11 March 2026 (UTC)

I don't think that "strengthen a policy or a guideline without consensus" is the right model. I think "impose additional restrictions on editors" is nearer the mark. WhatamIdoing (talk) 05:50, 12 March 2026 (UTC)

Why shouldn't we say that (almost) anything goes, as long as there is consensus for it?

Obviously, having a real consensus would require proper application of the ordinary policies, guidelines, and procedures, or a clear decision to Wikipedia:Ignore all rules for the betterment of Wikipedia, because there's a community-wide consensus in support of them, and "a consensus of me, myself, and I" would not overturn that.

Looking at some of your recent comments about WP:LSC at Talk:List of video games listed among the best, perhaps a clearer statement would help: I believe that this guideline should support the editors' choice of criteria that restrict the size/scope of that list. WhatamIdoing (talk) 20:01, 8 March 2026 (UTC)

Saying "editorial decision" instead of "subjective" would be a start. Then we could add that we need to balance discretion with consistency, and "objective" can drop, along "not arbitrary". Speaking of which, if we still believe that equivocation is OR, as I just read on a gaming talk page, then no wonder it is I who must have an idiosyncratic conception of what's objective. Selbstporträt (talk) 21:30, 8 March 2026 (UTC)

I think your caveat about what is needed for real consensus illustrates why we shouldn't say that (almost) anything goes, as long as there is consensus for it. If we tell editors that without including the caveat, I would expect the interpretation in practice to be that local consensus is sufficient. If we include the caveat, I would expect the reaction to be more-or-less split between "duh" and "huh?", with some finding it borderline-tautological along the lines of "you can do what you want as long as you stick to what's allowed" and some finding it impenetrable. TompaDompa (talk) 18:29, 11 March 2026 (UTC)

To know what you're saying, we'd first have to agree on which notion of local consensus you're using. Speaking as one of the editors who helped create WP:LOCALCON, I find that what that section says and how other people use that term have relatively little positive correlation. WhatamIdoing (talk) 05:53, 12 March 2026 (UTC)

If we tell editors that (almost) anything goes, as long as there is consensus for it without including the caveat that having a real consensus would require proper application of the ordinary policies, guidelines, and procedures, or a clear decision to Wikipedia:Ignore all rules for the betterment of Wikipedia, because there's a community-wide consensus in support of them, and "a consensus of me, myself, and I" would not overturn that, I would expect the interpretation in practice to be one that does not take the requirement for proper application of the ordinary policies and so on into consideration, but instead acts as if consensus at the level of the article overrides those policies (and so on). TompaDompa (talk) 16:26, 12 March 2026 (UTC)

We might be able to adapt some of the language in WP:CON to address that. Consider these:

• "Consensus...involves an effort to address editors' legitimate concerns through a process of compromise while following Wikipedia's policies and guidelines."
• "The goal of a consensus-building discussion is to resolve disputes in a way that reflects Wikipedia's goals and policies while angering as few editors as possible."
• "Consensus is ascertained by the quality of the arguments given on the various sides of an issue, as viewed through the lens of Wikipedia policy."

Consensus is king, including a consensus to IAR, but consensus takes the whole community's views into account via adherence to relevant rules and norms. WhatamIdoing (talk) 18:53, 12 March 2026 (UTC)

Honestly, I doubt it would have the desired effect. I have invoked the third point myself a number of times—suffice it to say that it doesn't always persuade people that policy considerations can render certain arguments irrelevant when assessing consensus. TompaDompa (talk) 21:50, 12 March 2026 (UTC)

What is your desired effect? WhatamIdoing (talk) 22:19, 12 March 2026 (UTC)

That editors, in the type of situation where this might be applied, understand "consensus" in the narrow Wikipedia-specific sense where policies and guidelines are taken into account (and so on – what you referred to as "real consensus") rather than a broader sense of agreement among editors, and apply this (correct) understanding in practice. I don't expect that consensus takes the whole community's views into account via adherence to relevant rules and norms will be the universal (or close enough to universal for practical purposes) understanding. Basically, if we start by saying something that at first glance seems permissive and then elaborate on the restrictions ("X is allowed! Conditions Y and Z apply..."), the expected outcome is that people (enough of them to make a difference, at least) will remember the permissive part and forget/overlook the restrictions. TompaDompa (talk) 23:19, 12 March 2026 (UTC)

I'm curious now: What kind of editing have you been doing, that you're so pessimistic about other editors?

Most editors, given something that says "BTW, consensus" aren't rubbing their hands in glee and thinking this is their chance to ignore a bunch of pesky content policies and guidelines. They're mostly like "Dudes, I'd love to but WP:V WP:NPOV WP:NOT WP:BBQ WP:UPPERCASE says no". WhatamIdoing (talk) 23:22, 12 March 2026 (UTC)

Among other things, I have spent a fair amount of time at WP:AfD, where arguing about differing interpretations of policy is very common. Experiences there have contributed to my view that policy and guidelines should ideally be written in a way that allows excerpts to be quoted and the meaning understood without requiring the context of the surrounding text (or indeed some other text, which may be located somewhere else entirely). TompaDompa (talk) 23:36, 12 March 2026 (UTC)

TompaDompa, I'm not see a distinction between "selection criteria" and "the choice of selection criteria". Are you talking about the result vs the process? WhatamIdoing (talk) 22:57, 7 March 2026 (UTC)

Hm, let's see if I can make this clear. I'll use your example of List of tallest buildings. The height requirement is currently a height of at least 350 metres. That's one of the selection criteria for the list, and it is objective. It could instead have been e.g. 300 meters or 400 meters. Those are also objective selection criteria that could be used. The decision to use 300, 350, or 400 meters is ultimately a subjective judgment call. That decision of 300 vs 350 vs 400 is what I mean by "the choice of selection criteria", and it is subjective. TompaDompa (talk) 23:10, 7 March 2026 (UTC)

But that's not subjective: no editor decide by and for themselves. It's intersubjective: a convention based on collective decision. How about convention? Selbstporträt (talk) 23:16, 7 March 2026 (UTC)

Perhaps "editorial decision" is the accepted term for that kind of decision. For instance:

When an article is too large, consider breaking it into smaller articles, spinning part of it out into a new article, or merging part of it into another existing article. When an article is too small, consider merging with one or more other existing articles. Such editorial decisions require consensus. Guidelines on the size of articles, and detailed solutions, are provided below. The licensing policy mandates that whenever any content is copied from one article to another new or existing article, an edit summary containing the required copy attribution must be used.

I'll check for other instances. Selbstporträt (talk) 23:24, 7 March 2026 (UTC)

All roads lead to WP:CON. Selbstporträt (talk) 00:30, 8 March 2026 (UTC)

I read this that when editors are deciding what are the bounds of the list, they are free to develop rules that are based on their subjective assessment of how best to do it, but the goal is that those rules should be as objective as possible. For example List of video games notable for negative reception is a list where we decided that must not include certain classes of games (like mobile and indie games), where the game must fall below a review aggregate score, and where there is demonstrated impact and legacy related to the game being bad. This is because there are a lot of games released that just get bad reviews, but the subset that we as editors have decided here helps to limit the list to once we can source well to be on it; we didn't decide that list from sources, so the criteria are 100% subjectively chosen, but when it comes to their application, we generally stay objective with common sense exemptions. Masem (t) 00:52, 8 March 2026 (UTC)

That'd be similar to or related to the various concepts of consistency that are discussed in WP pages. We want criteria that can be applied as consistently as possible. The reason why this increases accuracy can be omitted in the guideline. We only tell editors to find themselves something that eases their job.

To dispense ourselves from having to discuss objectivity and subjectivity would improve consistency. There is no parallel concept for objectivity in the WP docs as far as I can see. Perhaps neutrality, but to suggest neutrality is easier would be farfetched. Selbstporträt (talk) 01:11, 8 March 2026 (UTC)

This is a IMO goal: Y "criteria that can be applied consistently". WhatamIdoing (talk) 05:07, 8 March 2026 (UTC)

I agree with WhatamIdoing and Selbstporträt that "criteria that can be applied consistently" is what we are looking for here. "Subjective" and "objective" get in the way, since those are very loaded words with innumerable definitions. Katzrockso (talk) 05:48, 8 March 2026 (UTC)

Not disagreeing with consistency being a good thing here, just that objective criteria - which should be easy to back with reliable sources - tend to be far easier and less debatable for consistency's purposes than subjective ones. Masem (t) 05:54, 8 March 2026 (UTC)

When different people have different ideas of what "objective criteria" and "subjective criteria" might entail (as is quite evident from this discussion here), it might be more useful to use a more specific formulation of what you mean by "objective criteria". Katzrockso (talk) 06:05, 8 March 2026 (UTC)

Objective criteria, at least to me, is something you can usually quantify, rather than qualify. "After 1950", "greater then $1 million in revenue", "article/blue link exists" and so on. Then you just need a reliable source to show that metric is passed, and generally something that any reliable source that discusses it will also demonstrate the criteria is met because its a noncontestable fact. Subjective criteria is where there is leeway, where you're not always going to get the same answer from reliable sources and so you have to start bringing in WEIGHT to make sure you're reflecting what the majority of sources say. A "list of mass-murderers" where the criteria is "must have been called a mass murderer by sources" is going to be wide open because you get a lot of edge cases where even with multiple death there's a caution of applying the word "murderer" (eg Phillip Adams (American football)) and a wide range of commentary from sources. And yes, criteria can fall in between objective and subjective too. Masem (t) 13:28, 8 March 2026 (UTC)

"article/blue link exists" isn't synonym of notability. Notability means a topic could have a page, not that it has one. The whole WP playbook rests on dispositions, e.g. a substance (wink wink) is soluble if, once in water, it dissolves. Salt is still soluble when on your kitchen counter. The same applies to reliability, verifiability, even originality, i.e. facts, allegations, and ideas—for which no reliable source for which we could *never* find a source. Think about it: judgment is required to project properties in the future like that. Operationalism, which was one of the most objective way to conceive the scientific enterprise, rests on disposition concepts. It didn't pan out. Good luck trying to dispense yourself from decision-making about what wiki entries may bring.

Also, we're talking about criteria as a whole, with an s, and which was once named a "list of inclusion criteria". (For stand-alone lists, no less: poetry is everywhere!) We don't want to characterize them one by one: we want to characterize them as a whole. As soon as criteria reduces the number of entries editorial decisions will have to be made. Numbers don't stand alone: they rely on some model. Cutoffs need to make some sense. They may be updated as time drifts. And they are supported by other editorial decisions, like proportionality. (Yes, DUE is another dispositional concept.)

The example of the list of mass-murderers illustrates why extending the concept of inclusion criteria to titles is problematic: it borrows every single problem from the page on the property of being a mass-murderer. We're not trying to decide how to characterize (not legislate over!) what are inclusion criteria. We're trying to decide how to characterize inclusion criteria for lists. Selbstporträt (talk) 14:32, 8 March 2026 (UTC)

There are two notability-related standards that are commonly used. One is that the subject is WP:Notable ("red link standard") and the other is that the article already exists ("blue link standard"). The latter is mainly appropriate when the list would otherwise be extremely large. However, the latter is also often chosen by inexperienced editors, because it's easier for them to figure out whether an article exists vs whether it could be created. For example, the List of alumni of King's College London could have more than 2,000 people on it with the bluelink standard, and under the redlink standard, I wouldn't be surprised to learn that it could be ten times that size. WhatamIdoing (talk) 20:10, 8 March 2026 (UTC)

Thanks. Good example. That might be one way to turn lists of people in X projects into something manageable and less prone to patchwriting. I never thought word cutting could stop being enjoyable. Bluelink is a good way to deal with size matters. It prompts editors to stop their drive-bys and go create the pages they would like to see in the world. That makes them discover the joy of seeing that when you're build, people come.

Our actual task would be simpler if the parent page, WP:AT, had something to say about selection criteria. WP:TITLECON supports our inclusion: "Wikipedia:Article titles states as its fifth naming criterion, after recognizability, naturalness, precision, and conciseness: consistency". This also supports replacing the actual dichotomy with consistency. Clarification on what's a naming criterion might help our task of specifying what's a selection criterion. I also like this other bit:

This essay collects general examples of applications of title consistency that have been widely accepted by the community in various discussions, as well as exceptions that have also been widely accepted by the community. It should not be read in a way that overrides any topic-specific convention, but it may be helpful where existing conventions do not address consistency.

The subsection on examples should kick off with that. Let me gather examples I liked. Selbstporträt (talk) 21:12, 8 March 2026 (UTC)

If that is the definition of "objective criteria" under consideration, then I strongly oppose any attempts to make list selection insist upon "objective criteria". Katzrockso (talk) 01:51, 9 March 2026 (UTC)

The only sense of "objective" that I would consider would be something like some kind of transparency: a criteria is objective iff we don't need to make it explicit and it has a nearly perfect success rate. We already provide something like that: "As long as [selection criteria] can be applied consistently, stating them may be omitted." When objectivity and analyticity coincide, no synthesis is allowed! Selbstporträt (talk) 02:54, 9 March 2026 (UTC)

Gatherings

I gather that the most up-to-date proposed version is this:

Selection criteria (also known as inclusion criteria or membership criteria) refer to the characteristics required for entries to be included in an article. Many articles are clear enough that no criterion need to be made explicit. For instance, the List of bones of the human skeleton tells exactly what to expect: the complete list of bones in the human body. Not one more, not one less. The lead announces the exact number in the first sentence, followed by a citation, so readers should be able to induce that the list is exhaustive and follows actual scientific convention before consulting its entries. Yet the lead also says that this number varied historically and that there are anomalies; editors could wish to "pick a bone" with the actual count and include alternative ways to identify bones. Were that the case, they would need to inform readers about their decision and tell them under what conditions they added a bone to their list. The principle of least astonishment indicates that they should, common sense too.

For comparison, the current version is this:

Selection criteria (also known as inclusion criteria or membership criteria) express the characteristics required for entries to be included in an article. As for titles themselves, their formulation ought to be recognizable, natural, concise, and precise. The criteria need to preserve the content policies of the article as well as notability. They can be omitted as long as they can be applied consistently. By consensus, the criteria may be presented in the article lead, the article body, or on the talk page.

As for the proposed new version, I have the following suggestions:

• Avoid "refer to" in the first sentence; it's sloppy writing (see WP:REFERSTO about this issue in articles).
• Make it shorter. This is overly long and intricate.
• Write it as instructions, not a set of reflections.
• There needs to be something about when explicit criteria should be used, not just stating that they may not always be needed. The purpose they serve should come before the notion that they can be omitted—that way, it is easier to follow why they might be superfluous in some instances. Saying that they are "the characteristics required for entries to be included in an article" is not sufficient.
• Rely less on examples. The guidance here should be easily applicable in a wide variety of situations, and should therefore make general points rather than specific ones.

I will probably add further suggestions later. TompaDompa (talk) 20:10, 11 March 2026 (UTC)

I'll add that I think the current version is probably a better starting point to work on improving than the proposed one. TompaDompa (talk) 20:41, 11 March 2026 (UTC)

For a better source on reference, confer to this entry Selbstporträt (talk) 03:09, 12 March 2026 (UTC)

Revising an earlier version, here would be a first paragraph:

Like every other Wikipedia article, stand-alone lists ought to first describe the characteristics of their subjects or topics. This often suffices to prepare the reader for exhaustive lists; this seldom if ever suffices for lists that only offer a selection of entries. An explicit list of selection criteria allow readers to know the scope of the list, and future editors to learn about the conditions under which new entries should be added. Such list of selection criteria can be stated in the lead if they are short, in a section on the page if they are long, or in the Talk page if they don't concern the readers. Criterias ought to be designed for consistency: simple enough to help any editor add relevant entries, and clear enough to reduce the number of edge cases. When no criteria have been established, editors may be asked to suggest entries to the Talk page first. Templates can be used to in talk pages to provide editorial guidance.

This, I believe, is all we need to say on selection criteria themselves. We don't even need to define it as a term of art. If we did, they should in principle be defined, for it is ours. The laconism that the wiki is not a dictionary applies to words of others.

Nothing (absolutely nothing) needs to be said about reliable sources or any other policy in that section. That has been taken care of elsewhere on the page. All we need to say, for that first paragraph to work, is to tell readers (in our case, editors) what they are, what they are used for, and where to put them. We could add the other paragraphs I just wrote if we want to explain things a little further, with examples.

Yes, with examples. Even more examples are coming. Selbstporträt (talk) 12:21, 12 March 2026 (UTC)

Replied below, at #Simpler introductory paragraph, to avoid having parallel discussion about the same material. TompaDompa (talk) 16:27, 12 March 2026 (UTC)

Informal Straw Poll

• Option # 1: with the understanding that sometimes the criteria is obvious, which is pretty much the opposite of having no criteria other than the restrictions that that apply to all pages. --Guy Macon (talk) 18:47, 11 March 2026 (UTC)
• Option 1 This is basically the same as saying that a list article must have a scope, as any article must. There must be some way of determining what is in scope and what is out of scope, and for a list article that's what inclusion criteria do. TompaDompa (talk) 18:59, 11 March 2026 (UTC)

Comments

• Comment: I will not responding in any way to the editor who is WP:HOUNDING me and accusing me of wrongdoing again and again whenever and wherever I post any comment. Focus on content, not editors, and do not confuse silence with agreement. --Guy Macon (talk) 18:47, 11 March 2026 (UTC)