Jump to content

Wikipedia talk:Manual of Style

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia

Welcome to the MOS pit[Humor]


    Style discussions elsewhere

    [edit]

    Add a link to new discussions at top of list and indicate what kind of discussion it is (move request, RfC, open discussion, deletion discussion, etc.). Follow the links to participate, if interested. Move to Concluded when decided, and summarize conclusion. Please keep this section at the top of the page.

    Current

    [edit]

    (newest on top)

    Pretty stale but not "concluded":

    Capitalization-specific:

    This section is an excerpt from Wikipedia talk:Manual of Style/Capital letters § Current.[edit]


    Move requests:

    Other discussions:

    Concluded

    [edit]

    Disclose contradictory readings of COMMONALITY

    [edit]

    In a recent TfD, the following readings of MOS:COMMONALITY were both deemed acceptable readings by quite a few editors:

    (1) It follows from COMMONALITY that some English dialects are not suitable for Wikipedia.
    main thrust: COMMONALITY advises that uncommon terms or phrases be avoided, and nothing else.[1]
    (2) It does not follow from COMMONALITY that some English dialects are not suitable for Wikipedia.
    main thrust: COMMONALITY advises on the use of uncommon terms or phrases (avoid or gloss [note the or here]), and nothing else.[2]

    Notably:

    • (1) and (2) seem to and do contradict each other.
    • (1) seems to contradict MOS:ENGVAR: "The English Wikipedia prefers no national variety over others."
    • (1) seems to contradict at least some other related policies or guidelines (eg WP:TITLEVAR).
    • (1) seems to contradict at least some TfD consensuses (eg this EngvarB one).
    • Possibly, (2) seems to contradict at least some other related policies or guidelines.[3]
    • (2) seems to contradict at least some TfD consensuses (eg this UCamE one).

    I personally (and strongly) believe COMMONALITY should not be ambiguous between (1) and (2), and that rather one or the other should be excluded as an acceptable reading (eg with explicit text to that effect in COMMONALITY). I feel the current ambiguity (and/or its non-disclosure in COMMONALITY and related policies or guidelines) might be doing more harm than good.[4]

    As I really doubt we might reach consensus on one of (1) or (2) here,[5] I'll be adding explicit language to disclose the current lack of consensus on (1) or (2). I'll add this text to either COMMONALITY only (Proposal 1) or to COMMONALITY and all related policies or guidelines (eg ENGVAR, TITLEVAR, etc; Proposal 2). Unless someone objects? Or has a preference for Proposal 1 or Proposal 2?

    – Asdfjrjjj (talk) 21:51, 14 August 2025 (UTC) Asdfjrjjj (talk) 21:51, 14 August 2025 (UTC)[reply]

    Again, as with the other large blocks of text posted by the above editor, I don't really understand the logical reasoning here. I don't think MOS:COMMONALITY is confusing. It says to write in a way that would make sense to any educated reader of English, and not to use dialectical constructions that would be seen as ungrammatical or uninterpretable by most readers of English. An example that I gave in a discussion about Euro English and Ugandan English was the sentence, apparently acceptable in the Ugandan dialect, "They told me to come and you give me the package." MOS:COMMONALITY says not to use phrasing like that in English Wikipedia articles, except when glossing or explaining them. – Jonesey95 (talk) 22:11, 14 August 2025 (UTC)[reply]
    To editor Jonesey95: Ohh geeez, I'll try to rephrase but honestly might need another editor to put this into better terms if this doesn't work :/
    • You and others claim: (A) "There are some national varieties of English which are unsuitable for Wikipedia."
    • I and others claim: (not A) "There are no national varieties of English which are unsuitable for Wikipedia."
    These stances cannot both be true at the same time, and so they are contradictory stances. Claim (A) is often made with reference to COMMONALITY in particular (by saying something like "this dialect fails COMMONALITY" or just "this dialect is unsuitable per COMMONALITY"). And claim (not A) is often upheld even with reference to COMMONALITY. So, the contradictory stances seem to be due to distinct readings of COMMONALITY, in which case these readings themselves are contradicting one another, if that makes sense? (Or Pppery seemed to recognise the distinction so they might be able to help here, if possible? But in case this was a good explanation, would you oppose Proposal 1 or Proposal 2, or have a preference? :) - Asdfjrjjj (talk) 23:07, 14 August 2025 (UTC)[reply]
    Putting words in my mouth without providing links is not a great practice, especially on a MOS talk page, where arguments should be more precise than in other places on Wikipedia. I do not think that either (1) or (2) follow logically from MOS:COMMONALITY, although I find the double negative in (2) a bit cloudy. Please link to where I have claimed that "There are some national varieties of English which are unsuitable for Wikipedia." I may have said that a given article about an English dialect provided no guidance about differences, or supplied only differences that would be viewed as ungrammatical or misspelled by most educated English speakers, but I do not think that (A) or either of the propositions at the top logically follows from such a statement. – Jonesey95 (talk) 00:18, 15 August 2025 (UTC)[reply]
    To editor Jonesey95: I'm really sorry if I misread, but both Pppery and I took your 29 Jul nomination ("These recently created templates recommend the use of a dialect of English that is not suitable for Wikipedia, per MOS:COMMONALITY.") in the linked TfD to be based on COMMONALITY.[6] If that was not the intention then, again, I'm really sorry!
    For the record though, at least one of (1) or (2) must be an acceptable reading of COMMONALITY, just logically I think.[7]
    Also for the record, I'm not seeking consensus on (1) nor (2) here! And I'm not saying stance (A) or (not A) is good/bad/whatever! It's just the lack of consensus on (1) or (2) (or (A) or (not A)) that I feel oughtta be disclosed per Proposal 1 or Proposal 2.
    Finally, for the avoidance of any doubt, here are the comments I could find in support of stance (A). At least some of these seem to explicitly or implicitly (and partially or fully) be based on COMMONALITY imo.[8]
    • 29 Jul 2025: "These recently created templates recommend the use of a dialect of English that is not suitable for Wikipedia, per MOS:COMMONALITY." [links preserved] (in this TfD)
    • 13 Aug 2025: "I continue to object to the creation of both templates whose talk pages direct to this page, per MOS:COMMONALITY." [link preserved] (in this UCarE talk)
    • 3 Aug 2025: "I disagree with the change to remove mention of unusable dialects; it is needed to avoid creation of useless templates like the recent {{Use Cameroonian English}}." [link preserved] (in this UBE talk)
    • 20 Nov 2024: "Indian English#Spelling says that British spelling is used, and any Indian-specific vocabulary would not be usable because MOS:COMMONALITY says not to use regionalisms: Use a commonly understood word or phrase in preference to one that has a different meaning because of national differences." [links, colour preserved] (in this EngvarB talk)
    • 23 Nov 2024: "Neither Indian English#Spelling nor Pakistani English makes any substantiated claims that the spelling of those variants of English are different from British English, and we would not use dialect-specific vocabulary here at Wikipedia, per MOS." [links preserved] (in this EngvarB talk)
    • 22 Nov 2024: "The rest of the article explains the differences in pronunciation (not relevant since Wikipedia is a written medium) and regionalisms (also not relevant because MOS:COMMONALITY says not to use regionalisms: Use a commonly understood word or phrase in preference to one that has a different meaning because of national differences)." [link, colour preserved] (in this TfD)
    • 22 Nov 2024: "Ugandan English explains that this dialect makes use of phrasing that would not be acceptable here on Wikipedia per MOS:COMMONALITY, and uses misspelled standard English words. We would never accept those misspellings here, so these templates should probably go away." [links preserved] (in this TfD)
    • 26 Nov 2024: "How is Ugandan English different from British or American English in a way that is applicable here on Wikipedia? We can't accept misspelled words or non-standard phrasing that would not be understandable by the majority of English speakers, so this template is not usable as a guide for writing articles." (in this TfD)
    • 5 Dec 2024: "The Ugandan English article provides examples of English usage that would be considered incorrect here at Wikipedia." (in this TfD)
    • 29 Jul 2025: "Most usage of this variant on WP will be subsumed per WP:COMMONALITY." [link preserved] (in this TfD)
    • 29 Jul 2025: "the relevant phrases are inappropriate for use in articles in any case" (in this TfD)
    • 13 Aug 2025: "Propose deletion of this template which encourages the use of terms which are not suitable for use on Wikipedia in line with MOS:COMMONALITY." [link preserved] (in this TfD)
    • 13 Aug 2025: "It is not practical to expect every known English dialect to be used when editing a corresponding article. Nor would it be helpful since articles should be written in commonly used English." (in this TfD)
    • 8 Aug 2025: "Without disparagement on my part of the numerous varieties of English that exist in the world, I don't see it as practical to designate an article to be written in one of the few varieties that, together make up the bulk of English used internationally, and that typically are even the varieties used formally in places with their own variety. Should an article restrict itself to Cameroonian English or Philippine English or Roatan Island English, it would effectively place a sharp limit on who can contribute to it or copyedit it, excluding anyone who has no idea what that variety entails." (in this TfD)
    • 14 Aug 2025: "This template is an attempt to establish a new 'variety' of English in the MOS through an inappropriate route (i.e., it is bypassing a MOS discussion). First gain recognition at MOS for this as an ENGVAR variant that may be used." (in this TfD)
    - Asdfjrjjj (talk) 02:40, 15 August 2025 (UTC) Asdfjrjjj (talk) 02:40, 15 August 2025 (UTC)[reply]
    Thanks for that exhaustive work. I think my view continues to be summarized by one of my statements above: "Ugandan English explains that this dialect makes use of phrasing that would not be acceptable here on Wikipedia per MOS:COMMONALITY, and uses misspelled standard English words." I see that as distinct from "(1) It follows from COMMONALITY that some English dialects are not suitable for Wikipedia." Phrasing and spelling differences that are far enough from standard English to be seen as errors by most native English speakers should not be used on the English Wikipedia. That's not really part of MOS:COMMONALITY; it's more like MOS:COMMONSENSE. If an article about a dialect provides only examples of differences between that dialect and American or British English that are likely to be seen as errors, then that is a dialect for which a Use X English template should not be created, since we would be giving guidance to editors that is contrary to MOS. I think I need to step away from these Use X English conversations, since they tend to be a mud pit where everyone wrestles and nobody ends up making any progress. – Jonesey95 (talk) 03:54, 15 August 2025 (UTC)[reply]
    Additional comment which seems to be in support of stance (A):
    • 15 Aug 2025: "please stop creating language templates without consensus. ... We have to discuss each in turn because some have much less merit than others." [shortened] (in this TfD)
    Notably, this editor recommends establishing consensus before creating new {{Use X English}} templates. The 14 Aug comment from the same TfD[9] seems to agree, further adding that consensus ought to be established here, in MOS talk. I feel other editors might share this view, so this stance oughtta be recorded here imo. - Asdfjrjjj (talk) 07:11, 16 August 2025 (UTC) Asdfjrjjj (talk) 07:11, 16 August 2025 (UTC)[reply]
    One thing I find confusing here is that "dialect" largely refers to the spoken language, while Wikipedia is a written work. So any "dialect" that's chiefly spoken, but hardly ever written, is automatically unsuited for Wikipedia, simply due to its character as a written work. Can we agree on that? Gawaon (talk) 07:23, 15 August 2025 (UTC)[reply]
    I don't think it's true that "dialect" largely refers to the spoken language... In my experience, most varieties of standard English can be found in written form somewhere (or at least, analysis/description of the written variety can be found in the literature). It might be less easy to find formal written texts in English-based creoles, which are often more vernacular to a particular place, but this is irrelevant here as creoles and pidgins shouldn't be used on the English Wikipedia as they aren't necessarily mutually intelligible with any variety of standard English (they're their own languages). Pineapple Storage (talk) 07:41, 15 August 2025 (UTC)[reply]
    Right. But per ENGVAR, it seems clear enough that any variety of written English that is used as the standard written form in any English-speaking country is eligible for Wikipedia, and I can't remember anyone having said anything else. So I'm frankly not convinced that the disagreement sketched by Asdfjrjjj does actually exist. Actual conflicts are rather about the question of when a standard written form used in some place is sufficiently distinct from other standard forms to get its own name. To give a made-up example, one can doubt that the English written in California differs sufficiently from that in Oregon to treat Californian English and Oregonian English as different varieties. So in practice we don't do that, instead just using the label "American English" for both. But there is no conflict about the meaning of COMMONALITY or ENGVAR (nor is the use of "Oregonian English" forbidden), it's simply a question of when a variety of written English is sufficiently distinct to get its own name and template. Gawaon (talk) 08:11, 15 August 2025 (UTC)[reply]
    Did you read this TfD? Because the argument being made by the nominator was exactly that the standard variety of English in Cameroon, where English is an official language and the national variety (ie. dialect, per my comment below) is Cameroon English, should not be used on Wikipedia. Pineapple Storage (talk) 08:16, 15 August 2025 (UTC)[reply]
    I hadn't read it before. I have skimmed it now, but would rather interpret Jonesey95's argument as meaning that Cameroonian English is a variety that's spoken or maybe used in informal writing – not but a clearly distinct written form used in Cameroonian newspapers or other formal sources. Did I miss something? Gawaon (talk) 08:32, 15 August 2025 (UTC)[reply]
    I see where you're coming from, and maybe that was was the nominator originally thought, but during that discussion we established that Cameroon English does have its own spelling rules/patterns/standards when written down (including in formalised written sources)—I just couldn't list all of them with beyond-reasonable-doubt confidence because the only exhaustive source is a dictionary that isn't available online. Pineapple Storage (talk) Pineapple Storage (talk) 08:39, 15 August 2025 (UTC)[reply]
    Well, that was your conclusion, but did everybody else in that TfD agree with you? I have some doubts. A single dictionary is also not particularly convincing proof, especially since it might focus on the spoken and informal language rather than, or in addition to, the formal written register, and it probably won't reveal how many of these forms are distinct from those used in neighbouring countries. Gawaon (talk) 08:48, 15 August 2025 (UTC)[reply]
    I interpreted this comment as conceding that CamE was a distinct variety but just arguing that there shouldn't be a template for it, and this comment pretty much underlined the fact that the delete !voters hadn't given any evidence that CamE is identical to any other variety.
    I agree that one dictionary shouldn't be the only source in consideration; I did provide other (reliable, formally published) sources that discussed examples, but these were seen as not exhaustive enough. Anyway, even if the dictionary only documents spoken usage, the spellings that are used in it (because a dictionary has to be written down) will shed some light on the spellings in use in written CamE. But this isn't a discussion about CamE specifically, so I'd rather not get back into that debate now! :) Pineapple Storage (talk) 09:09, 15 August 2025 (UTC)[reply]
    The question "does MOS:COMMONALITY ban some English dialects?" is a bit vague, and as I mentioned can be answered "clearly yes" and "clearly no" depending on the definition of "dialect".
    If we sharpen the question and ask "do recent TfD results indicate that the standard English of some English-speaking countries is unacceptable because of MOS:COMMONALITY?", I would object to writing "there was no consensus about that" into this MOS page because I think those discussions were about non-standard varieties and what to tag articles for the convenience of editors than about the acceptability of standard national varieties. Having a list of TfD discussions on this topic is useful for future discussions on this family of templates; perhaps there is a better place to document that than the MOS, like an internal category or navigation template?
    There's an underlying question about whether the standard English of a given English-speaking country is indistinguishable from say, British English, either before or after MOS:COMMONALITY is applied. As the above discussion highlights, this is an empirical question to which the answer is "maybe it is, maybe it isn't, more research is needed to show that this is in fact true for at least one country in the world". Whether or not given country or collection of countries (like the Caribbean and South Asia) even have a distinct local standard variety in the first place is also an empirical question that has been answered for some but not all.
    Then there's the philosophical question of whether to call the resulting words "Cameroonian English" or "British English" if writing in formal standard Cameroonian English preferring commonalities with British English produces the same words as writing in formal standard British English preferring commonalities with Cameroonian English. As a practical resolution to avoid offending nationalist sensibilities and align intuitively with MOS:TIES, we may simply allow people to tag articles with either of those labels to and program our spell checkers to use the same dictionary for both tags. Or we may agree on a single neutral tag that better describes the desired style of writing. What I think would not be helpful is to let a debate over tags become a whiff of "some editors are trying to ban the standard English of certain countries" into the MOS when that's definitely going to offend some people and is kind of an oversimplification or arguably not really what's happening. -- Beland (talk) 09:27, 15 August 2025 (UTC)[reply]
    I agree. Gawaon (talk) 10:18, 15 August 2025 (UTC)[reply]
    I think part of the difficulty with relying on TfD discussions (exclusively) to identify ENGVAR consensus is that there are two different questions:
    1. Is X variety of English suitable for use on Wikipedia?
    2. Does there need to be a {{Use X English}} template?
    TfD discussions are meant to be answering question 2; some respondents try to do this without addressing question 1 (eg. by saying "X variety is basically the same as Y variety, which we already have a template for so we don't need this one"), while others do it by addressing question 1 (eg. "this template shouldn't exist because X variety shouldn't be used on Wikipedia"). Really, to properly get a sense of consensus about question 1, there needs to be a discussion independent of question 2, so that we're all addressing the same question; this is why I'm glad this discussion has been started.
    Also, just a minor point about terminology: it feels unfair to label people who object to the deprecation of certain varieties as being "offended"—this word carries connotations (nowadays especially). I'm sure some people would take offence, especially if their native variety is specifically excluded (for instance, if they have to learn an unfamiliar set of spelling rules to avoid their contributions being "corrected" as misspellings) but this is far from the only argument against deprecating certain varieties.[10] Personally, I've spent hours and hours researching the issue and arguing the case for inclusion of CamE even though I don't speak (or write) CamE and have no connection whatsoever to Cameroon; I'm not offended by the idea of CamE not being allowed on Wikipedia. Rather, I have strong opinions on the issue because (1) as a linguist and a Wikipedian, I care about both Wikipedia and the English language; (2) I believe Wikipedia's use of English should represent its global focus and reach; and (3) I know that the way Wikipedia works (and progresses) is through community discussion, so I want to contribute to this process. If we approach these discussions with the assumption that the only reason why anyone would object to certain (standard, written) World Englishes being deprecated is because they take offence (ie. feel personally insulted), we run the risk of overlooking the other arguments against deprecation (eg. countering systemic bias), which don't deal with personal emotions etc. but with broad, project-wide principles of Wikipedia, and with the practicalities of managing a reference work whose authors are from all over the world. Pineapple Storage (talk) 10:21, 15 August 2025 (UTC) Pineapple Storage (talk) 10:21, 15 August 2025 (UTC)[reply]
    I don't think there's actually any serious effort to ban standard formal Cameroonian English from Wikipedia, nor to change MOS:TIES in such a way that it says articles about Cameroon should be allowed to violate the local standard English spelling and grammar rules (e.g. by using American spelling sometimes under MOS:RETAIN). That would be egregious and I can't imagine that ever getting consensus support, and I think implying that there is such an effort would needlessly upset some people. What did happen was a debate over whether to tag Standard Cameroonian English as Standard British English because there is no practical difference for Wikipedia purposes.
    I have never believed that the only reason someone would object to say, tagging Standard Cameroonian English as Standard British English is because they are personally offended. "Actually there are practical differences between these dialects" is certainly one of those reasons, and that fact violates the assumptions of the above scenario. I'm just pointing out that national identity is a sensitive topic and that some people can be offended by even something as simple as how we name things, so even if there are no practical reasons to do something, we might benefit from doing it anyway for the sake of editor harmony. -- Beland (talk) 11:10, 15 August 2025 (UTC)[reply]
    That would be egregious and I can't imagine that ever getting consensus support It's very refreshing to hear this, thank you. Unfortunately, as I see it at least 4 of the ~7 participants in the CamE TfD were arguing that CamE shouldn't be used on Wikipedia. I got this impression from some of the statements made in that discussion:
    • "the use of a dialect of English that is not suitable for Wikipedia" (here)
    • "Being inclusive might be nice but [...] There is no practical way that certain articles can be written using terms that contradict EN or US usage. Also, there would be no practical benefit from such a system." (here)
    • "how can I venture into an article expressly written in Cameroonian English?" (here)
    • "if none of this is codified, we can't really enforce a 'Cameroonian English'." (here)
    • "I don't think you could get consensus that these are anything but typos to be fixed [...] regardless of their use in some countries." (here)
    For the reasons I discussed above, I think the context of TfD pretty much took all the nuance out of the conversation re ENGVAR, which is why the wider conversation should happen somewhere else (ie. here, in an MOS discussion). I also agree with you absolutely that editor harmony is an important consideration when it comes to sensitive topics like this. Pineapple Storage (talk) 11:56, 15 August 2025 (UTC)[reply]
    Yeah, I don't think editors there were considering much beyond automated spell checking, where having a dictionary is necessary and most of the concern was about not encouraging editors to use an informal dialect.
    As for spellings like "truely", are these mandatory in Cameroonian Standard English, or are Standard British English spellings also accepted? If they are mandatory, it seems we'd have to accept them per MOS:TIES. Getting a complete list of any such exceptions would be helpful for automation purposes. -- Beland (talk) 12:40, 15 August 2025 (UTC)[reply]
    I don't think the unique spellings are mandatory, as BrE spellings are also widely used in CamE; I think "truely", "occured",[11] etc. are just valid alternative spellings. I ordered the CamE dictionary when the TfD was closed and I'll update Cameroonian English once it's arrived. Pineapple Storage (talk) 13:07, 15 August 2025 (UTC) Pineapple Storage (talk) 13:07, 15 August 2025 (UTC)[reply]
    In that case, using the same spellings as the UK does isn't forcing Cameroonians to write about their own country using spellings that are invalid Standard Cameroonian English, so I don't see the problem in using the international spellings. MOS:COMMONALITY says the most common spelling in the national variety should be used, so it's possible that technically the unusual local spelling should be preferred. Is that easy to determine?
    It might actually be worth changing this to say that where two variants are equally acceptable but only one is present in other national standard Englishes, the international one would be preferred. This could prevent confusion over editors trying to fix typos that aren't typos, and also lessen the distraction for readers to which the local variants look like errors. -- Beland (talk) 14:35, 15 August 2025 (UTC)[reply]
    Is that easy to determine? Probably not, unfortunately, beyond getting access to a corpus, as most studies don't address minor spelling differences explicitly. Preferring more international spelling variants is sensible; thankfully this is already covered by MOS:COMMONALITY, but I agree that the MOS:TIES application could be emphasised—for instance by explicitly saying something to the effect of "just because X Standard English variety has multiple acceptable patterns, most of which are shared with another variety, this doesn't mean X variety shouldn't be used or that MOS:TIES shouldn't apply to that variety". Pineapple Storage (talk) 15:58, 15 August 2025 (UTC)[reply]
    I don't see where the existing verbiage in MOS:COMMONALITY says to prefer international spellings over local ones? It says that about word choice explicitly. Are you implicitly extending that to spelling as well? I read the "most commonly used current variant" as explicitly saying not to prefer the most international spelling if it's not the most common local one. -- Beland (talk) 17:46, 15 August 2025 (UTC)[reply]
    When more than one variant spelling exists within a national variety of English, the most commonly used current variant should usually be preferred
    I read the most commonly used as meaning the one that's most common generally, ie. shared by multiple varieties, rather than most common to that particular variant. If this interpretation of it is wrong, and there is consensus for the latter reading, then I think that wording needs specifying. Otherwise, if a consensus hasn't yet been established about that specific point, then maybe that's a discussion to be had? Pineapple Storage (talk) 17:53, 15 August 2025 (UTC)[reply]
    To editor Beland: I might get us that list of TfD and ENGVAR discussions soon-ish, as I also could've really used such a list when I first waded into this part of MOS :) - Asdfjrjjj (talk) 23:47, 17 August 2025 (UTC)[reply]
    (edit conflict) Thank you for starting the conversation @Asdfjrjjj, I think it's definitely one that needs to be had at some point.
    Unsurprisingly I have quite a few thoughts on this issue! Since the CamE TfD (this one) I've been doing some research about World Englishes and have been gathering sources at User:Pineapple Storage/Bibliography/World Englishes (still very much an incomplete work in progress; I also can't claim to have read all the sources there, it's basically a reading list) so I feel like I'm saturated in WP:ENGVAR thoughts at the moment, and I will likely add comments to this discussion about multiple different aspects of the issue as they occur to me—apologies in advance for this.
    One thing that springs to mind following @Gawaon's comment above is that ENGVAR maybe shouldn't even refer to varieties at all, but should instead refer to dialects. Variety (linguistics) says:

    [Varieties] may include languages, dialects, registers, styles, or other forms of language, as well as a standard variety.

    I would argue, the fact that ENGVAR only mentions varieties rather than dialects is probably a result of the prejudice/misconception described in the lead of Variety (linguistics):

    The use of the word variety to refer to the different forms avoids the use of the term language, which many people associate only with the standard language, and the term dialect, which is often associated with non-standard language forms thought of as less prestigious or "proper" than the standard.

    This perception (of "dialect" meaning a non-standard variety) is clearly a misconception; it's linguistically inaccurate, and can be easily dispelled by reading the first two sentences of the article Dialect.
    When we discuss ENGVAR, what we're actually discussing is dialect, because register and style are already dictated by WP:ENCSTYLE; this is an encyclop(a)edia, so the style is encyclop(a)edic and the register is formal (but not excessively/performatively formal, because it needs to be accessible to a very wide audience).
    I think ENGVAR should be altered to replace variety with dialect (and maybe to add an explicit mention that the formal register of any dialect should be used in articles, to clear up any ambiguity). What do people think about this? Pineapple Storage (talk) 08:12, 15 August 2025 (UTC)[reply]
    I think that "variety" should stay for just that reason: if a dialect, or variety, is chiefly spoken, but not written, that it's not eligible for Wikipedia – not out of prejudice, but since Wikipedia is itself a written work. "Variety" seems better suited to get that across. Gawaon (talk) 08:16, 15 August 2025 (UTC)[reply]
    Or, in words cited above: any standard variety of English is suitable for Wikipedia, but not any dialect, register, or style. Gawaon (talk) 08:18, 15 August 2025 (UTC)[reply]
    I think it's worth pointing out here that the word "dialect" can be used in at least three different ways. As our article Dialect points out, sometimes it exclusively means non-standard varieties. In that sense, there are no dialects which are suitable for Wikipedia. Perhaps what Gawaon is getting at is that non-standard varieties (which are more commonly spoken than written) are unsuitable for a written encyclopedia - and yes, I think that it generally agreed upon. I should also say that simply because a dialect is commonly written doesn't make it suitable for Wikipedia. People in Yorkshire may commonly use Northern colloquialisms when writing letters, emails, text messages, business signage, or when transcribing spoken language. But we're not interested in just any writing, we're interested in writing in what they consider their standard written dialect, which would be Standard British English - the sort that appears in newspapers and generally intelligible formal documents not written in legalese or other technical dialect.
    I tend to use "dialect" in the primary sense defined by the article Dialect, which includes both standard and non-standard varieties. In this sense, Standard American English is the dialect I happen to speak, it is the one spoken on national news broadcasts and written in national newspapers, and it is the (only) appropriate dialect for writing articles with strong ties to the United States. In this sense, some dialects are appropriate for Wikipedia.
    Asdfjrjjj seems to be using "dialect" to mean "national variety". In this sense, all dialects are appropriate for Wikipedia, because as ENVAR says: "The English Wikipedia prefers no national variety over others."
    Using the second sense, I'd say each English-speaking country can have multiple native dialects, and only the standard one is suitable for Wikipedia. -- Beland (talk) 08:22, 15 August 2025 (UTC)[reply]
    I agree, and when I talk about "written variety" etc, I mean the formal version of the variety as written in eg. newspapers, books, and other formalised publications, rather than informal contexts like personal communications (or advertising, blogs, etc) as these are more likely to reflect vernacular (informal spoken) usage, per Written language#Relationship with spoken and signed language. Pineapple Storage (talk) 08:29, 15 August 2025 (UTC)[reply]
    Still though, I think ENGVAR in its current state leaves an ambiguity in the minds of some editors, who might then feel it necessary to proscribe certain dialects as being "too informal" when actually the standard written register of that dialect is perfectly acceptable. Pineapple Storage (talk) 08:32, 15 August 2025 (UTC)[reply]
    I disagree with changing "variety" to "dialect". The whole reason linguists use "variety" is to avoid controversies in the definition of dialect. That includes the question of whether "dialect" has a connotation of informality, and also more commonly the question of whether two or more varieties are different dialects or different languages. I do agree ENVAR could be improved with some clarification; I'll come up with some verbiage. -- Beland (talk) 08:36, 15 August 2025 (UTC)[reply]
    If "variety" is still going to be used, then ENGVAR should make it very clear that any country's standard written variety can be used (unless of course consensus turns out to be that only some standard written varieties are acceptable, which I would find disappointingly predictable). Pineapple Storage (talk) 08:45, 15 August 2025 (UTC)[reply]
    There should probably also be a discussion about how we're defining "standard"... Does a "standard written variety" need to have been formally codified in order to be allowed on en-wp, or is attestation/description in other (non-dictionary) reliable sources sufficient? If a consensus can be reached on this, then ENGVAR should probably make it clear. And how are we defining "codified"? Does a single published dictionary count? If so, does the dictionary being unavailable online (as discussed in the CamE TfD) make the variety ineligible for use? And if a single dictionary doesn't count, how many dictionaries must exist? This would all ideally be made clear—if not in WP:ENGVAR, then somewhere else. Pineapple Storage (talk) 09:22, 15 August 2025 (UTC)[reply]
    MOS:TIES requires that the English of a specific nation (which I think means "country" in this context) be used for any article with strong ties to an "English-speaking nation". Our article English-speaking world defines that as "the 88 countries and territories in which English is an official, administrative, or cultural language". So I would infer we care about matching the expectations of English speakers in Cameroon, for example, but not France.
    I think the requirement here is to produce articles that are 100% compatible with the article-local standard English, and intelligible to everyone else (with glosses if necessary) even if there are spellings and other quirks that are not allowed in the reader's local standard English. It is not a goal to produce articles that demonstrate unique aspects of the article-local standard English. The question of whether unique constructions exist in standard contexts and would plausibly be used on Wikipedia is relevant to tagging and spell-checking, but if the answer is "no" I think it is not right to say that a given country's standard dialect is "not allowed" on Wikipedia. It just means that articles happen to be 100% compatible with the standard English of multiple countries.
    It's worth pointing out that MOS:COMMONALITY doesn't "ban" all localisms and by implication the national standard dialects they come from. It just requires that those that can't be avoided should be glossed for intelligibility. For example, even though for articles about Nigeria we prefer constructions that it has in common with British English, sometimes it is impossible not to use localisms that refer to elements of local culture, like danfo (which is a specific type of shared taxi and has no equivalent phrase in British English with the same precise meaning). I think we would also prefer local Nigerian expressions such as senior wife to refer to the primary polygynous spouse, whereas in American English we would say "first wife" which does come up for example in Mormon families.
    Reading Codification (linguistics) carefully, it appears that process is a process of establishing social consensus for a set of spelling and grammar and vocabulary preferences, which by definition has already happened in standard varieties. It sounds like by "formal codification" you mean someone has written down these social rules.
    Thinking about how we resolve questions of Standard American English in US articles, we often simply rely on native speakers of that variety to know what is and isn't correct grammar and spelling. That would imply we could rely on local editors to know the correct forms for a standard English that does not have a dictionary or written rules of grammar. I think that's fine; by definition there will be lots of documents in a standard written variety they can point to for attestations if there's a dispute. No version of English has a single official authority that sets the rules, so while dictionaries and books from grammarians can be very helpful in settling these disputes, they don't always agree with each other or the actual current social consensus. It's up to editors to decide if disputed practices are OK, like ending a sentence with a preposition, or how to deal with something that only one editor or maybe less than 1% of American English speakers consider an error is worth changing, presumably depending on the available alternatives. Even in the US and the UK we have turned to attestations to argue some disputes, like whether it is OK to refer to a ship as "she".
    We might think not having written dictionaries and grammar rules for a national standard English would make it difficult to write articles about that country in cases where there aren't a lot of local English speakers active on Wikipedia, but in practice targeting Standard British English seems to result in compatible outcomes. (That's the advice I've gotten for grammar-checking Indian English.)
    I'm not sure we need to put specific guidance about this in the MOS. I think editors know what it means to say, "use standard formal Cameroonian English", and how to research what the rules are for that variety, disputed and undisputed. -- Beland (talk) 10:47, 15 August 2025 (UTC)[reply]
    Again, it's very refreshing to hear this take. Pineapple Storage (talk) 12:06, 15 August 2025 (UTC)[reply]
    Interestingly, WP:ENCSTYLE and WP:TONE are in Wikipedia:Writing better articles, which is an essay, not a guideline. But the main Manual of Style page points there, and it is certainly explaining a real consensus. -- Beland (talk) 08:32, 15 August 2025 (UTC)[reply]
    To editor Pineapple Storage: Oh wait I'd missed User:Pineapple Storage/Bibliography/World Englishes, looking pretty good! I'd also written WP:CARIB/E for Caribbean English (more as a descriptivist overview than a bibliography). Maybe we might sort of consolidate these similar to how the big/well-known Englishes are consolidated in the various Comparison of forms of English articles (though these prolly oughtta stay in project namespace to focus exclusively on description of only formal, written English, or to focus exclusively on bibliography). Might add sources from Caribbean English to your page soon-ish if you don't mind :) - Asdfjrjjj (talk) 22:39, 17 August 2025 (UTC)[reply]
    @Asdfjrjjj For sure, this sounds like a good plan! The World Englishes bibliography I've been putting together is very much a work in progress, but I plan to keep going basically until I've got sources about all the places (and their national varieties—or not, in cases where English is an official language but sources say a distinct variety hasn't yet emerged, such as in Sudan, South Sudan, Tuvalu and Vanuatu, among others) currently covered by Template:Ties/sandbox.
    Absolutely, feel free to add any sources (and new sections etc. as needed) to User:Pineapple Storage/Bibliography/World Englishes, and also feel free to incorporate any sources I add there into WP:CARIB/E if you find them useful!! There are a bunch of sources about Caribbean varieties that I've bookmarked but haven't yet done full citations for, so I'll hopefully add those soon. Pineapple Storage (talk) 10:49, 18 August 2025 (UTC)[reply]
    I don't understand how the EngvarB closure could be read as saying anything about which dialects are or are not appropriate for Wikipedia. It's mostly a decision that "EngvarB" is not a good name for a template, and the choice of whether to always tag language as specific country varieties or to use tags that posit a country-spanning variety of some sort was left open. -- Beland (talk) 08:40, 15 August 2025 (UTC)[reply]
    A dialect is not defined as a set of words/phrasing that are different or unique when compared to whatever default is implicitly implied in that definition. If a word or formulation is common to that dialect as well as to the implied default, then it is a part of that dialect as it is the implied default. CMD (talk) 11:17, 15 August 2025 (UTC)[reply]
    Well put. This is why "avoid the unique features of this dialect" is not the same as "this dialect is banned". -- Beland (talk) 11:37, 15 August 2025 (UTC)[reply]
    Yes. The original question didn't give examples, but I am happy to. We would avoid terms such as "lakh" or "krore" because they wouldn't be readily understood by most Wikipedia users. This doesn't mean we are banning Indian English, we are just going for common (COMMONALITY) features. cagliost (talk) 20:48, 15 August 2025 (UTC)[reply]
    Based on the discussion so far, here's my proposed clarification:
    • Change "The English Wikipedia prefers no national variety over others." to "All national varieties of Standard English are equally acceptable on English Wikipedia, which uses a formal, encyclopedic register or tone."
    • Change "conventions of a particular variety of English" to "conventions of a particular variety of Standard English"
    • Change "within a national variety of English" to "within a national standard variety of English"
    • Change "(formal, not colloquial) English" to "standard (formal, not colloquial) English" for consistent terminology.
    -- Beland (talk) 11:36, 15 August 2025 (UTC)[reply]
    I'd support these changes. For MOS:TIES, maybe adding links might be helpful, eg. "standard (formal, not colloquial) English"? Or maybe "standard (formal, not colloquial or vernacular) English"?
    Also, after the list of example articles and varieties, currently TIES says:

    For topics with strong ties to the Commonwealth of Nations, or multiple Commonwealth countries or other former British territories, use British spelling.

    The way this is currently worded, it's pretty much a direct contradiction of the principle of TIES. What about something more like:

    For topics with strong ties to one or multiple Commonwealth countries or other former British territories, if there are no relevant national Standard English varieties, use British English spelling.

    (Note the altered link targets.) Any thoughts on this idea? Pineapple Storage (talk) 12:46, 15 August 2025 (UTC)[reply]
    I have no objection to the first suggestion of adding links.
    The second suggestion doesn't make sense to me, as there are standard English varieties in most if not all of those places.
    Can you gave an example or two of where it would contradict MOS:TIES? I think the language about the Commonwealth of Nations ended up this way because {{Use Commonwealth English}} was deleted and replaced with {{Use British English}} in articles where there weren't clearly ties to a specific country. It had previously said that Commonwealth and British orthography were nearly indistinguishable for encyclopedic writing. -- Beland (talk) 14:16, 15 August 2025 (UTC)[reply]
    From my reading of the current wording of TIES, it basically says "If a country is in the Commonwealth or used to be a British territory, use British spelling" but doesn't make any exceptions for Commonwealth nations or former British territories that have their own national variety of Standard English. So if we were to follow only that sentence (and not the rest of MOS:TIES), articles with ties to every country and territory listed at List of countries that have gained independence from the United Kingdom would have to be written in British English, regardless of whether a local national Standard English exists. This clashes with the first sentence of TIES, which says "An article on a topic that has strong ties to a particular English-speaking nation should use the (formal, not colloquial) English of that nation." The change I'm suggesting is basically to alter the sentence so that it allows local varieties to be used for ex–British-territories that have their own national variety. For example, currently, articles about Malaysia use British English, but with the new wording they would use Malaysian Standard English, in line with the first sentence of TIES. Pineapple Storage (talk) 14:56, 15 August 2025 (UTC)[reply]
    Also, just from the list of Member states of the Commonwealth of Nations there are four that don't have English as an official, national or recognised language (Cyprus, Gabon, Mozambique and Togo) let alone any non–English-speaking countries or territories at List of countries that have gained independence from the United Kingdom.[12] Pineapple Storage (talk) 15:01, 15 August 2025 (UTC)[reply]
    Perhaps it's not clear from the wording, but if an article is only about one country, this clause isn't supposed to apply. It's only when an article has ties to multiple countries or the actual Commonwealth of Nations organization. For example, Australia–Malaysia relations or Canada–United Kingdom relations.
    Both to clean up which countries we're referring to and to cover multi-country cases which don't all use UK-derived spelling, would it be better to say something like:
    • For articles with strong ties to the Commonwealth of Nations organization, use British spelling.
    • For articles with strong ties to multiple English-speaking countries, either:
      • Pick one of the related national standard varieties. (For example, Australia–United States relations uses Australian English.)
      • Use a variety common to at least some if not the majority of countries. (For example, Five Eyes uses British spelling.)
    -- Beland (talk) 15:26, 15 August 2025 (UTC)[reply]
    Okay that's where the ambiguity was. These suggestions seem sensible! What about cases like Gabon and Togo, which are Commonwealth countries without national varieties whose articles don't currently have a variety template? Should this apply to them, or should they be governed by MOS:RETAIN? Also, the wording you propose would remove any mention of former British territories. Should these be mentioned explicitly as being treated the same way as Commonwealth nations? Or should they just be assumed to fall under the main MOS:TIES clause? Pineapple Storage (talk) 15:46, 15 August 2025 (UTC)[reply]
    I added "former British territories" because not all of the English-speaking countries that have standard written dialects close to British English are still members of the Commonwealth of Nations, like Ireland and Zimbabwe. But that's also a bit odd because the United States is also a former British colony that does not use British spelling. I think we were just looking to say "it's OK to just say we're using British spelling for topics with ties to multiple countries that have almost identical spelling to the UK" and not bother trying to make an exception for Canada when it's in a group of British-spelling countries. So maybe the intended group is more like "English-speaking countries that don't use American spelling".
    Yes, MOS:TIES clearly already applies to articles about individual English-speaking countries regardless of their Commonwealth or former British colonial status. The above clarifying language only applies when we're talking about more than one of country, and if the article is about two former British colonies that still have British spelling in common, then British spelling is a valid choice, as is either national standard variety.
    According to English-speaking world, Gabon and Togo are not English-speaking countries, so I think they should be governed by MOS:RETAIN. They are members of the Commonwealth of Nations, but an article has to have strong ties to the organization, not one of its member countries, for the Commonwealth clause to apply.
    I do also intend "country" to include UK dependencies that are not part of the UK, like Bermuda and Pitcairn. -- Beland (talk) 17:21, 15 August 2025 (UTC)[reply]
    Okay, all sounds good to me.
    I'm glad you mentioned UK territories actually, as most do have their own varieties named and described in the literature.
    Discussion of English varieties in dependent territories
    For instance:
    This accidentally turned into a pretty extensive research exercise so apologies for the hefty paragraph and the ~9hr delayed response! I've collapsed it to save space and boredom for non–language-nerds.
    Still though, it proves that there are a lot of potentially eligible British Overseas Territories varieties, as well as varieties from other dependent territories. By nature of most of these territories being remote and/or island nations, the speaker communities are always going to be very small compared with other varieties, and I sense that the idea of the varieties listed above being "allowed" on Wikipedia would really raise the hackles of some of the participants in this discussion and the recent TfDs.
    So, what should we do about these varieties? Is there really any harm in "allowing" them, as some would argue? Or can they just remain as lesser-used options among the many varieties available? Pineapple Storage (talk) 01:33, 16 August 2025 (UTC)[reply]
    It is already allowed, and in fact required by MOS:TIES, to write articles on these jurisdictions in a way that is compatible with the standard varieties in all these countries. Whether or not the need for formality and clarity for an international audience means any of the unique localisms actually get used in articles, is an empirical question. Presumably that would depend on someone who actually knows something about the local English varieties attempting to use them. -- Beland (talk) 03:44, 16 August 2025 (UTC)[reply]
    I totally agree! Glad I'm not the only one. Pineapple Storage (talk) 03:47, 16 August 2025 (UTC)[reply]
    Perhaps it would be better if there were one template to cover a large number of countries with derived spelling patterns and a small number of unique elements that editors would expect to need in articles. It could say something like "This article has been identified as having strong ties to [country name]. It should use the standard formal written English of this country, which [is similar to British English, is similar to American English, is a mix of American and British conventions] in spelling and grammar. Follow MOS:COMMONALITY when using local expressions." That helps editors know what spellings to use even if they aren't familiar with the requested dialect. -- Beland (talk) 03:52, 16 August 2025 (UTC)[reply]
    This definitely sounds like a good idea. I guess editnotices would be the best option for this? Pineapple Storage (talk) 04:09, 16 August 2025 (UTC)[reply]
    Template:Ties is available for this, and I'm looking to expand my template-writing experience at the moment so I'd be happy to start the ball rolling on this. Pineapple Storage (talk) 04:13, 16 August 2025 (UTC)[reply]
    This would go very far in harmonising all the {{Use X English}} templates we have, would be veeery much in favour of this. Buuut at least some editors would like at least some varieties excluded, so would be nice to get their thoughts/go-ahead first. To editors Quondum and Dgp4004: this is another very relevant proposal to your stances re these templates in the UABE TfD! - Asdfjrjjj (talk) 07:22, 16 August 2025 (UTC)[reply]
    It is, in any case, much more practical than having more than a 100 different "Use ... English" templates! Gawaon (talk) 07:58, 16 August 2025 (UTC)[reply]
    Unsurprisingly, I remain opposed to pretty much all of these suggestions which arise from a misreading of MOS:TIES.
    Starting with the British Overseas Territories, we're now talking subnational. These are not nations. And it opens the door to even more silly subnational templates like 'Use Antiguan English' and 'Use Barbudan English', because I bet somebody can dig out a word that's used on one island and not the other. So no, I absolutely do not accept that there is any policy on Wikipedia which mandates the creation (or backdoor creation via another template) of a 'Use Falklands English' template.
    Further, this idea of creating a new mega template including dictionaries of local vocabularies is just going to be a vehicle for even more of these pedantic and poorly sourced claims. So I remain opposed. Dgp4004 (talk) 09:57, 16 August 2025 (UTC)[reply]
    Is your objection to the use of local words, or to the bureaucracy of template tracking? -- Beland (talk) 10:03, 16 August 2025 (UTC)[reply]
    I never mentioned 'the bureaucracy of template tracking'. I'll stick with my own words; they don't need paraphrasing. Dgp4004 (talk) 10:10, 16 August 2025 (UTC)[reply]
    Sounds like I'm not understanding what you mean, then. What is it then about these templates that you find objectionable? -- Beland (talk) 10:25, 16 August 2025 (UTC)[reply]
    They do not improve Wikipedia, they are like spam. It seems that editors create them, and then only when challenged do they try to find out exactly what X English means, and suddenly we have 'I've ordered a dictionary' and 'this newspaper uses -ze so it must be a local variety of English'. They are not created from need. None of this arises from an organic discussion, dispute or confusion on the relevant pages. Their purpose appears to be to promote or even create a local formal English that cannot be adequately sourced. These words that are used to justify the language template's existence are not even words that often appear in an encyclopedia, let alone on those few pages that will use the language template. I have sometimes cited the example of bread roll in British English. There are a great many words like cob, bap, barm, bun etc. which are only used in particular regions. That does not create a justification or a need for a 'Use Lancastrian English' template. Dgp4004 (talk) 11:02, 16 August 2025 (UTC)[reply]
    It seems everyone is agreed we don't find non-standard dialects acceptable, including regional ones. It seems necessary to tag articles with something so spell checkers know which dictionary to apply.
    What would be your preferred rule to determine if a separate template is needed? That an article actually use a local word or spelling? Wouldn't that potentially mean a lot of re-tagging of all a country's articles if we later discover it's actually necessary to distinguish it from British English for practical purposes? Is it better if all these templates are merged into one and the country is only indicated by a parameter? -- Beland (talk) 11:28, 16 August 2025 (UTC)[reply]
    I would also be interested in the answer to this. I write in Belizean English, and will continue doing so until ENGVAR guidance changes. I've been tagging everything {{UBE}} b/c it was the closest I could find to "not AmE" (and {{EngvarB}} confused me tbh). Sure, I can continue tagging everything I write as {{UBE}}, but that's not actually accurate, and imposes more of a burden on me whenever I need to flag "Hey guys, this is actually correctly written in Belizean English, despite what the tag says!" And ofc precludes the use of bots/scripts to correct Belizean English usage (by preventing tracking categories). And discourages Belizean editors from contributing to Wikipedia (they might be forgiven for thinking they must write in BrE/AmE/one of the English dialects with a {{Use X English}} template). - Asdfjrjjj (talk) 11:41, 16 August 2025 (UTC)[reply]
    Just for curiosity: how does (written) Belizean English actually differ from British English? Gawaon (talk) 12:26, 16 August 2025 (UTC)[reply]
    To editor Gawaon: imo? by being wayyy more tolerant of Americanisms and American-style spelling, plus some distinctive terms/senses, a few unique spellings (caye not cay, etc; cf WP:BZE/E)! My -ise spellings are actually wayyy old school in Belizean English: Oxford-style spelling is the historical, formal norm afaik. - Asdfjrjjj (talk) 12:56, 16 August 2025 (UTC)[reply]
    I don't want to go down yet another handbrake turn onto another topic as this is interminable. If you would like to discuss whether Use Belizean English should continue to be a thing, a specific TfD would be the best place to go into the depths of that. Whatever you're writing in, all I can tell you is that it is indistinguishable from British English.
    As to what would be my preferred solution, I would personally prefer guidance which asks editors to use the established templates and to seek consensus for each additional template they wish to create. Rather than the present, 'I'm doing this regardless of need and I'm going to tag as many pages as possible and then maybe later I'll think about what Use X English actually means.'
    Once a need has been established for one of these templates by consensus, I have no problem whatever with that being used. But this is the only way to stop the proliferation of these poorly sourced, unnecessary and meaningless templates which spread like wildfire. That doesn't touch on any other policies about language. It only concerns the proliferation of language templates. Dgp4004 (talk) 12:41, 16 August 2025 (UTC)[reply]
    To editor Dgp4004: Eh I was just using Belizean English as an example, just read X English in its stead! Your preferred solution unfortunately keeps us from using tracking categories, discourages some potential editors from contributing to Wikipedia, and imposes a burden on some editors. Beland's unified {{Use X English}}, on the other hand, fully addresses your goal ("to stop the proliferation of these poorly sourced, unnecessary and meaningless templates which spread like wildfire"), without any of the drawbacks of your preferred solution. If that goal is the only reason you have for opposing Beland's solution and preferring that any new {{Use X English}} templates go through a pre-approval-by-consensus process, then I'm really sorry but I don't think your stance makes much sense to me.[14] - Asdfjrjjj (talk) 13:21, 16 August 2025 (UTC)[reply]
    I completely understand why you would oppose seeking consensus on the creation of some of these templates because they can't stand up to scrutiny, just as Use Cameroonian English didn't. Hence why we're here seeking to move the goalposts. Dgp4004 (talk) 16:00, 16 August 2025 (UTC)[reply]
    Certainly it could be a waste of a lot of effort to create a template and apply it to a bunch of articles without some assurance it won't be deleted because it's widely opposed. There isn't a "Draft template" namespace; I don't have a problem with people being bold and temporarily creating templates that aren't used in articles for the purpose of prototyping or discussing them, especially if they aren't simply variations on the theme of templates that have been deleted.
    Personally, I also mildly dislike the idea of a large number of templates, one for each of the 88 English-speaking countries, because that creates a lot of overhead maintaining their documentation, doing updates if formatting changes, etc. The idea of creating one mega template addresses the objection on the grounds of template proliferation.
    If the idea is to get consensus before using such a template, great, this conversation is part of that process.
    You objected earlier: "a new mega template including dictionaries of local vocabularies is just going to be a vehicle for even more of these pedantic and poorly sourced claims."
    Yes, though it's a very interesting topic for linguists and readers of articles on the dialects of English, it is much more efficient when thinking about templates to entirely avoid the question of whether or not standard formal English in a given country is different from British English at all. (Not to mention the distraction of non-standard national varieties, which is most of what our articles cover.) This implies we want to avoid templates that make claims like "this article is written in Cameroonian English". I see Template:Ties/sandbox is doing that, which is not what I was proposing and not what I would favor. I also don't like that it allows specifying a given English variety using anywhere from one to four different codes. A, the codes are less intelligible to editors than country names, and B, having more than one per choice creates a headache for people like me who write bots. It's better to have only one, intuitive way to specify a given variety, and spit out an error for invalid choices.
    I have drafted an alternative at Template:English ties. This is in the form of an editnotice, but at this point I mostly care about getting consensus on the wording and not making it pretty or where it will be used (we could replace the talk page notices with this, for example, or use a non-displaying version embedded in article wikitext that would be more efficient for bots).
    Instead of saying "this article is written in Cameroonian English", it says "This article has strong ties to Cameroon. Per MOS:TIES it uses standard formal written English as used in this country, which for encyclopedia purposes is very similar to British English."
    For countries where there is an undisputed distinct national standard variety, it will simply link to that article, like: "This article has strong ties to the United States. Per MOS:TIES it uses standard formal American English." It also handles Oxford and IUPAC spelling cases.
    This template is not the place to put exhaustive lists of spelling differences for specific dialects. I think that joyfully pedantic work is best directed at Wiktionary, where the results can be used by everyday readers, linguistics researchers, editors trying to follow MOS:COMMONALITY, and authors of automated spell-checkers like me. (I already use Wiktionary as my list of valid words, and it also tags common misspellings which is very helpful for correction purposes.)
    You also referred to a "misreading of MOS:TIES". I agree that policy doesn't require the creation of any templates; that's a different practical question. If there's something else you're getting at, it might be good to make that explicit and clarify the MOS if needed.
    Or in general, if people feel we should be treating small English-speaking countries and their dialects differently than large ones, maybe we should have a discussion about that. My proposed template can only be used for articles that have strong ties to a country (or to note Oxford or IUPAC spelling subvariants). That means people can't go around tagging random no-ties articles like Electron and say they have to use Cameroonian English. (I'm wondering if worry that someone might do that is one of the reasons some people are resistant to creating templates for more national dialects?) Should there be an explicit list or usage cutoff for dialects that can be used for no-ties articles? In practice articles without ties to a specific country almost always use American, British, or Oxford, and because these are highly used that makes things easy for readers to understand, minimizes distracting or confusing localisms (or the need to commonalize them) and gives editors an easy target. Or we could say no-ties articles should use a dialect with at least 20 million speakers, which is 0.25% of the world population. (That would mean American, Indian, Nigerian, Pakistani, Indonesian, Phillipine, British, Canadian, and Australian, according to List of countries by English-speaking population.) -- Beland (talk) 20:46, 16 August 2025 (UTC)[reply]
    Thank you for the feedback on Template:Ties/sandbox; I've updated it so that only one (ISO 3166-1 alpha-3) code can be entered per country/territory, so hopefully that would make it easier for bots etc.? I also removed the mentions of specific varieties; I'll go through and add comparative explanation for the various lesser-known varieties (based on the sources I've been able to find that provide that kind of description), but that will take a while obviously. Pineapple Storage (talk) 22:02, 16 August 2025 (UTC)[reply]
    @Beland I just re-read your comment and realised you specifically said the codes were a problem! Facepalm Apologies for that.
    You said above that you mostly care about getting consensus on the wording for Template:English ties, so I feel like it's worth mentioning this: currently the template is worded in a way that assumes that any variety (other than IUPAC, Oxford, AmE, CanE and PhlE) is functionally the same as BrE... Is that what you meant by entirely avoid the question of whether or not standard formal English in a given country is different from British English at all, in your comment above? If so, does this not run the risk of attracting EngvarB-type objections? Many varieties, especially Caribbean and Australasian Englishes, incorporate enough AmE spelling patterns etc. to differ noticeably from standard BrE, even in formal/encyclop(a)edic writing. Pineapple Storage (talk) 23:31, 16 August 2025 (UTC)[reply]
    Sort of. That list is not complete; I only bothered putting in a few countries for demonstration purposes. If we decide we want to use the template, I will add all the other English-speaking countries that shouldn't get the "which for encyclopedia purposes is very similar to British English" message. That's only the default so I didn't have to repeat that text over and over again. I could change the template to not have a default and reject any string that's not explicitly listed as a country name or other valid choice? -- Beland (talk) 01:14, 17 August 2025 (UTC)[reply]
    To editor Beland: re {{English ties}}, looks excellent! I personally would avoid explicit MOS mention, and rather just summarise the relevant MOS guidance and link to it.[15] Could we have this edit notice track all the Category:Use X English tracking categories though? I feel this might be needed for bots/scripts, and to fully supplant all current {{English variant templates}} (maintenance and edit notice ones).[16] And on a side note, the IUPAC switch prolly oughtta not even be indexed to BrE/AmE, as IUPAC's independent of English variety as far as I can tell?
    re the proposal to restrict ENGVAR (the 20 mil cutoff for non-TIES articles), I think that's a great compromise, as I do feel this might be a fear amongst stance (A) editors (but I've misread their stance/reasoning before, it seems, so dunno).
    Should maybe any of these proposals be RfCs? I dunno how we establish consensus on them, and I don't really see a lot of stance (A) editors participating here so I fear this might be an echo chamber with only stance (not A) editors :(..? Asdfjrjjj (talk) 22:10, 16 August 2025 (UTC)[reply]
    I must apologise to you Beland, and to Asdfjrjjj and others, as my tone has been belligerent. I am far calmer now, and against all the odds, Beland's template Template:English ties has won me over!
    It elegantly marries the freedom to use any national variety of English with the commonalities they share with British English, and allows for flexibility for those varieties that don't. And it will be far easier to discuss and agree edits to one template than trying to keep track of many and putting them each through TfD. Great work!
    However, just to muddle things completely and swap sides... whilst I'm certainly not against your proposal to restrict neutral articles to the major English varieties, it might be a little unfair! I wonder if it mightn't be better to expand the scope to also replace - or at least have the potential to replace - all the Use X English templates, even the large ones. Dgp4004 (talk) 22:33, 16 August 2025 (UTC)[reply]
    And something else that just occurred to me while reading the template's examples: I know it sounds silly, but 'English ties' could be perhaps provocative. It could be misread as ties to England. 'English ties|Scotland' or 'Northern Ireland' or 'Falklands'. And it also sets an example that it can be broken down into tiny sub-national areas, each varying from another. Perhaps it would be better to restrict it to nation states. English ties|United Kingdom, Ghana etc. Dgp4004 (talk) 22:57, 16 August 2025 (UTC)[reply]
    The name "English ties" was chosen hastily just to get something drafted. Any suggestions for a better name that avoids confusion with the country of England? {{strong ties}} is available. We could also clobber {{ties}} with whatever final design is preferred. I don't have strong feelings, other than that it should be intuitive and hopefully short.
    I threw in Scotland as an example because {{Use Scottish English}} and {{Scottish English}} exist and are used on hundreds of articles. A 2023 TfD found no consensus to merge {{Scottish English}} to {{British English}}, on the grounds that words like "church" and "barrister" should not be used over the Scottish alternatives when talking about Scottish entities. Though it is not listed at English-speaking nation, Scotland is considered one of the four "nations" of the United Kingdom, and so arguably is one. This implies we actually do need to say to use standard formal Scottish English, with deference to commonality with British English. I will update my draft.
    I think "Falklands" needs to be a valid choice because even though it is under British sovereignty, it is not part of the United Kingdom, so using {{English ties|United Kingdom}} would produce factually incorrect text. It's possible that the standard written English of the Falklands, Bermuda, and England are all the same for encyclopedic purposes, but given how geographically separate they are and how Bermuda is influenced by Jamaica and the United States and the Falklands is influenced by Argentina and Spanish, I wouldn't be surprised if one day we discover we need to treat them differently. Since part of the point of this template is to avoid having to re-tag, I think we are better future-proofed by allowing the names of dependent territories. ISO 3166 can be somewhat helpful figuring out what entities exist and which are internal territories vs. external. I think I know enough about some places to say flat out to use the sovereign power's national variety, like "strong ties to the United States Virgin Islands, use American English". But if we discover later that there actually is a difference that requires different spellings or more gentle wording, that's easy to change in the central template without re-tagging hundreds or thousands of articles. -- Beland (talk) 02:01, 17 August 2025 (UTC)[reply]
    Hmm, if we want this to replace all the "Use X English" templates, then presumably a name that has "ties" might be too narrow? -- Beland (talk) 03:10, 17 August 2025 (UTC)[reply]
    Or actually, we could use {{ties}} for choices based on strong national ties, and {{English variety}} or something else for choices based on MOS:RETAIN. One could be a redirect to the other so they could share overlapping code and mappings and whatnot. -- Beland (talk) 03:45, 17 August 2025 (UTC)[reply]
    Yes, Ties and English Variety might be a good idea. Just for the record though, I must disagree with your assertion that the overseas territories are apart from the UK. You're absolutely right that they're not in it, as such. But they are annexed to the Crown of the United Kingdom. That 'of the United Kingdom' part is often overlooked. Dgp4004 (talk) 18:14, 17 August 2025 (UTC)[reply]
    Well, the "as such" part actually matters. We don't want to put text in our template that directly contradicts our articles: United Kingdom says the UK is only England, Scotland, Wales, and Northern Ireland. It explicitly says that the Crown Dependencies are not part of the UK. British Overseas Territories explicitly says they are not part of the UK. That's why I say they are under British sovereignty but not part of the UK. -- Beland (talk) 03:56, 18 August 2025 (UTC)[reply]
    I would second this proposal, seems an elegant and reasonable solution to me, and allows for ENGVAR restriction for non-TIES articles (if that comes about :) - Asdfjrjjj (talk) 18:56, 17 August 2025 (UTC)[reply]
    I also think this sounds like a good option. Template:Retain is available, so could either be the main version of the non-TIES template, or a redirect for simplicity. Pineapple Storage (talk) 13:03, 18 August 2025 (UTC)[reply]
    Also interesting about the Scottish English template still being a thing. Use Scottish English already redirects to Use British English. Just one of the many Hydra's heads I suppose! Dgp4004 (talk) 18:17, 17 August 2025 (UTC)[reply]
    To editor Ohconfucius: hi I know you mentioned preferring to not be pinged re ENGVAR/{{Use X English}} discussions in the CamE TfD, but Beland's proposals above (the {{English ties}} template, the ENGVAR restriction) seem super relevant and possibly an acceptable way to address everyone's concerns?[17] - Asdfjrjjj (talk) 23:33, 16 August 2025 (UTC)[reply]
    The original {{Use British English}} and {{Use American English}} templates were devised by me with help from User:Rich Farmbrough (if my recollection serves me correctly) as a technical solution not only to identify which spelling variants are used in any given article (in accordance with WP:ENGVAR) but also to ensure that there are means to ensure that an article's spellings remain consistent going forwards. As such, each template is dated for when it was last audited. Its functionality and facility for maintenance are predicated on its placement within the article as a hidden category. Templates marking the 4 canonical varieties have to be retained to achieve those twin objectives.
    As to use of the {{English ties}} editnotice, it would be a big plus if by adopting this system we managed to merge down the number of {{Use X English}} templates, and curtail their multiplication beyond the 4 canonical varieties.  Ohc revolution of our times 19:30, 21 August 2025 (UTC)[reply]
    But I see that this mega template has already been created, not even in draft space, so there's little point in taking part in this discussion. It's like hydra's head. You object to one template, a dozen more spring up. Dgp4004 (talk) 10:06, 16 August 2025 (UTC)[reply]
    Template proliferation was always going to be the result of the engvarB close, this discussion isn't the cause. CMD (talk) 10:39, 16 August 2025 (UTC)[reply]
    It's perfectly possible to replace EngvarB with a single template that just has a better name. -- Beland (talk) 10:49, 16 August 2025 (UTC)[reply]
    It’s British English. Where a variety has sufficient differences from British English to merit its own template, like Australian English, use that where appropriate. Where it doesn’t, then those places are effectively using British English, maybe with a bit of added local vocab. MapReader (talk) 12:35, 16 August 2025 (UTC)[reply]
    It was and perhaps is possible, but the tfd decided against that. CMD (talk) 13:16, 16 August 2025 (UTC)[reply]
    If you're talking about the EngvarB TfD, I closed that discussion, and that wasn't the outcome. I specifically wrote that "editors will need to decide if they want to revive or create templates" for spelling styles that span countries, and gave a few examples of different approaches to doing that. -- Beland (talk) 20:57, 16 August 2025 (UTC)[reply]
    The outcome was that the template was deprecated, not set for renaming. Many of its instances have been replaced by other templates, and existing instances will continue to be replaced as time goes on. That can't be unwound even if a new template is created, so whatever the intention proliferation was baked into the outcome. CMD (talk) 02:10, 17 August 2025 (UTC)[reply]
    Certainly many articles were more appropriately tagged with country-specific templates since they did not have ties with multiple countries. But those templates already existed, so I'm not sure I would consider that proliferation. If {{English ties}} replaces those country-specific templates, we may in fact see the reduction in the overall number of language variety templates. -- Beland (talk) 03:02, 17 August 2025 (UTC)[reply]
    That's not certain at all, unless there's an analysis of engvar b template replacement I've missed. CMD (talk) 03:13, 17 August 2025 (UTC)[reply]
    @Kusma, MPF, EEng, MapReader, Jc3s5h, Oknazevad, NebY, Trovatore, Ham II, Vestrian24Bio, HouseBlaster, Tony1, Amakuru, The C of E, and Keith D: I feel Beland's proposal here (for a unified {{Use X English}} template) might be of interest to anyone who participated in the previous EngvarB discussions here on MOS talk and TfD? (But apologies if anyone did not wish to be pinged.) - Asdfjrjjj (talk) 11:29, 16 August 2025 (UTC)[reply]
    "Use X English" is fine for me. The C of E God Save the King! (talk) 11:32, 16 August 2025 (UTC)[reply]
    I largely agree with User:Beland, but I try not to overthink the situation.
    Commonality has to be synonymous with accessibility for our wide readership. Local vernacular, before it acquires the status of loan word, is something that I would generally find acceptable, provided that its use within any given article is rendered comprehensible through glossing or otherwise defining. Classic examples that immediately spring to mind of words that must be defined/glossed are the Indian measures "lakh" and "crore", which, through being exceedingly common in Indian sources for dealing with metrics cause confusion to readers elsewhere in the world – I would note in passing that we seem to have fewer problems eliminating numerical separators that are not common in the English-speaking world (viz: the full stop, the space, the comma). All that remains are the more common spelling variants denoted by nationalistic phonetic simplification ("color", "gray", "liter", "edema"), or a particular morphology ("traveled", "paneling"), or the Oxford ~ize, which are the object of the EngvarB script.
    These permutations are the foundational variants that define/form the 4 "core varieties". As my good friend User:Tony1 has remarked, barring semantic variations, there really are precious few words in Aus and NZ English that are differently spelt compared to British English that would warrant using their own language tag – except of course nationalistic considerations, and for which I still favour {{EngvarB}} over {{Use British English}}, but I digress.
    Following the principles outlined in our discussion about the {{Use Cameroonian English}} template and in the absence of further canonical variants, I would argue that all non-core {{Use X English}} templates could be replaced by or redirected to one of the 4 core varieties.  Ohc revolution of our times 09:02, 21 August 2025 (UTC)[reply]
    Thank you for the ping @Pineapple Storage. For myself, I oppose your suggested changes the the MOS above. It won't surprise you that I believe the MOS needs tightening up rather than loosening. I think that putting in terms like 'Standard English' is just going to lead down a rabbit hole of whether a variety is indeed 'Standard English'. There will no doubt be some editors of the belief that a single obscure study or dictionary about some national variety is sufficient to make it 'Standard English'. In my view, these changes just creates more wriggle room to crowbar in dozens and dozens of obscure 'Use X English' templates. Dgp4004 (talk) 15:52, 15 August 2025 (UTC)[reply]
    If the following are true:
    • All articles are legible and intelligible to readers who can read English
    • All {{Use X English}} templates are about varieties that have their own Wikipedia article, which either currently contains or could in the future contain comprehensive documentation and guidance on the spelling and grammar rules of that variety
    then what exactly is the problem with dozens and dozens of obscure 'Use X English' templates?
    Aside from the fact that obscure is relative (and the suitability of an English variety shouldn't be decided by someone who's never heard of it), there are dozens and dozens of countries and territories that do have a Standard English variety. And there are dozens and dozens (likely hundreds and hundreds) of linguists around the world producing documentation of (and other literature about) these varieties. As with everything on Wikipedia, what matters is whether the reliable sources exist, and as long as they do, there's no real reason (other than WP:IDONTLIKEIT) why a certain variety shouldn't be used. Also, we're not working on paper, so there are no practical limits that stop us from making as many templates as there are (attested) Standard English varieties. Pineapple Storage (talk) 16:10, 15 August 2025 (UTC)[reply]
    @Dgp4004 I would also direct you to @Beland's comment above, if you haven't already read it. Pineapple Storage (talk) 16:13, 15 August 2025 (UTC)[reply]
    What Standard English looks like for any given country is pretty easy to figure out these days, given that by definition it's used in newspapers and government documents. There should be plenty of samples of that online for all but the smallest of countries. A variety is not considered standard because an academic paper is published about it, it's because of its status and use in society. Having a dictionary published for a dialect does not make that dialect standard. Consulting a dictionary is a good way to know if a word is colloquial or regional or standard. We also have Wiktionary which already knows this about most words, and which can be updated as we learn about new country-specific English words.
    I'm not sure why adding "standard" would create new disputes over actual usage. Editors should already be complaining about and removing non-standard, informal, or colloquial language and avoidable regionalisms.
    MOS:TIES apparently already requires us to write compatibly with 88 different national standard dialects. This language doesn't change that, it just makes explicit what should already be obvious, that there are not dozens of dialects beyond that to choose from. Whether or not we need to have 88 or 90 different templates to indicate the dialect choice of all articles is a different question, and I'm skeptical the template tail should be wagging the "how should I spell words" dog.
    -- Beland (talk) 17:41, 15 August 2025 (UTC)[reply]
    Beland's suggested changes read good to me. I also think that they adequately match what is already the meaning of ENGVAR – they are a clarification rather than an effective change of the guideline. Gawaon (talk) 17:38, 15 August 2025 (UTC)[reply]
    These seem very sensible to me :), agree with Gawaon above re how they're not changing MOS meaning. Buuuut I and Beland read ENGVAR the same way I think, so would be nice to hear from editors who hold the opposing view first before making any changes, I feel?[18] - Asdfjrjjj (talk) 19:02, 15 August 2025 (UTC) Asdfjrjjj (talk) 19:02, 15 August 2025 (UTC)[reply]
    @CROIX, Johnuniq, Retro, PharyngealImplosive7, Quondum, Rich Farmbrough, Cagliost, Arnav Bhate, Fylindfotberserk, Amakuru, Charcoal feather, PBS, SUM1, and LlywelynII: just in case anyone had any thoughts on Beland's proposals here? or the broader/vaguer Proposal 1 or Proposal 2 in general? And sorry if anyone did not wish to be pinged for this, or if I missed anyone! - Asdfjrjjj (talk) 20:35, 15 August 2025 (UTC)[reply]
    Not sure how I got included but thanks.
    Nothing much to add. This all seems to fall under WP:BLUE to me, going back to the original post. The reading that all forms of English are acceptable down to a single person's idiolect and (e.g.) William Faulkner's article should be written from the POV of a mentally disabled southerner, ee cummings's article should have no punctuation or capitalization, and China should be written in Chinglish with Chinese grammatical rules... it's just patent nonsense.
    One thing: The Ugandan example above seems to point to occasions when a standard national form of English might accept constructions not grammatically included in most others. If this is really a large issue people can't be rational about at this late date, it might be necessary to also strengthen the wording that we should default to phrasing acceptable across multiple dialects. We really should, honestly, but some Brits would probably get their dander up about being told to just use "while" on all occasions and it's not like we want to discourage anyone's work on the project. — LlywelynII 03:25, 16 August 2025 (UTC)[reply]
    How would people feel about adding a sentence to the "When more than one variant spelling exists within a national variety of English" item so it reads:
    When more than one variant spelling exists within a national variety of English, the most commonly used current variant should usually be preferred, except when:
    • The less common spelling has a specific usage in a specialized context, e.g., connexion in Methodist connexionalism
    • Two spellings are accepted in a national variety and they are equally common or usage frequency is difficult to determine, the spelling which is more broadly used internationally is preferred
    -- Beland (talk) 17:54, 15 August 2025 (UTC)[reply]
    Pineapple Storage points out above that "most commonly used" may have been intended to mean "most commonly used across all standard English varieties" rather than "most commonly used in the national standard English variety. If that's so, we could take care of this with a much smaller change, like saying "most commonly used current variant (internationally)" or "(across all national varieties of Standard English)". -- Beland (talk) 17:57, 15 August 2025 (UTC)[reply]
    Just a quick note… remember that we are discussing written varieties of English, and not spoken varieties. Formal written English (ie what one would use in writing an encyclopedia article) has a lot less variation than colloquial spoken English. Blueboar (talk) 18:13, 15 August 2025 (UTC)[reply]
    My impression so that we here we come into overly specialized territory where it's just not practical to have a rule. In practice, such issues rarely arise, and if they do, the statistics needed to apply the rule likely won't exist, or their reliability won't be clear. Not every theoretically possible detail needs to be covered in the MOS (avoid CREEP). Gawaon (talk) 21:25, 15 August 2025 (UTC)[reply]
    • Comment - I'm not sure what the latest proposal is, the discussion seems to have become fragmented... But if there is a consensus forming around the proposed {{English ties}} template, then I think that sounds a good proposal. My main points of concern in the past have been that we should not remove opportunities to specify that articles have ties to their own locations, e.g. by deleting "Use Ugandan English" and forcing articles from that country to be written in "British English", while also retaining the principle of EngvarB, which gives editors unfamiliar with local dialects guidance as to the broad category that it falls under. As long as we allow all the ties that might be necessary, including those for Cameroon and Tanzania etc, even where a formal specification of that country's English can't be found, then {{English ties}} looks great. A nice corollary if this is that it looks like it avoids the need to be too prescriptive about terminology from on high, and allows decisions to be made at article level on whether to use local terms or default to commonality. Thus the danfo and senior wife examples mentioned above could be accepted in an article, while other obscure local terms might be avoided, with local consensus and normal editing determining this. Does this sound like the proposed plan? Cheers  — Amakuru (talk) 08:06, 17 August 2025 (UTC)[reply]
      Exactly! -- Beland (talk) 15:55, 17 August 2025 (UTC)[reply]
      Yes! sorry I think I'm meant to summarise this discussion, will figure out how to/guidance and do shortly :) - Asdfjrjjj (talk) 20:10, 17 August 2025 (UTC)[reply]
    {{English ties}} probably shouldn't be using the more American "encyclopedia" in templates specifying similarity to British English. It's also worth noting that there is variatoin within countries, including the US and UK, and it is unclear whether this drive for directly linking formal English standards to individual countries is meant to assert a single standard for each country. CMD (talk) 09:36, 21 August 2025 (UTC)[reply]
    Hehe, that's a tricky problem! Personally I think it should rather be "encyclopedic/encyclopaedic purposes", since an adjective is called for in that position, but I don't really know now to resolve the ENGVAR issue. Though maybe it's possible to tweak the output to generate the correct form depending on the ENGVAR actually specified by individual invocations of the template? Gawaon (talk) 10:23, 21 August 2025 (UTC)[reply]
    Ah, good point. We could also just say "Wikipedia" instead of "encyclopedia". -- Beland (talk) 16:13, 21 August 2025 (UTC)[reply]
    To avoid the e/ae debate, could this be an opportunity to introduce the adjective "Wikipedic" to the lexicon?[Humor] Pineapple Storage (talk) 18:19, 21 August 2025 (UTC)[reply]
    Oxford dictionaries list "encyclopedia" before "encyclopaedia", so the former is an point of WP:COMMONALITY between British and American English. I agree with Gawaon about "encyclopedic purposes". Ham II (talk) 17:54, 21 August 2025 (UTC)[reply]
    Heh, that implies all instances of "encyclopaedia" should be changed to "encyclopedia" if the latter is more common in British English. -- Beland (talk) 18:21, 21 August 2025 (UTC)[reply]
    Is that so? I don't have access to the full content, but the OED lists encyclopaedia first. Gawaon (talk) 18:45, 21 August 2025 (UTC)[reply]
    My Concise Oxford English Dictionary lists "encyclopedia" first, and my memory is that Lexico did, but I stand corrected about the OED! Ngrams have "encyclopedia" outnumbering "encyclopaedia" in British English after 1988. Ham II (talk) 18:57, 21 August 2025 (UTC)[reply]

    Where to put English variety notices?

    [edit]

    Sounds like we want to create separate {{ties}} and {{retain}} templates for articles that have strong ties to an English-speaking country vs. use an arbitrary national variety following MOS:RETAIN. I can make drafts. Before I put work into that, I was wondering if we need all three possible placements - talk page, editnotice, and non-displaying bot tag? Also, do people have any preference for making separate templates for each placement (e.g. {{ties talk page}} and {{ties editnotice}}) vs. making one template for all placements using more complicated logic? The latter would at least prevent anyone from putting one in the wrong place. It seems to me that having both talk page and editnotices is a bit redundant, but it looks like editing of editnotices is restricted to admins and template editors and page movers. -- Beland (talk) 17:08, 21 August 2025 (UTC)[reply]

    I've been intending to come in on this so I'd better do so now before the discussion moves on. What would we gain from having separate {{ties}} and {{retain}} templates instead of a single {{English variety}} template? I imagine it would be a lot of work to divide up the existing instances of {{Use American English}}, etc. – to what end?
    I think we should take this opportunity to do away with the clutter of ENGVAR templates on talk pages, and (if it's feasible) have them in editnotices where they can be seen by users editing articles. Ham II (talk) 17:38, 21 August 2025 (UTC)[reply]
    The main reason for having two different templates is so that there can be two different wordings. Articles that have strong ties to a place have to use Standard English as written in that place, but we want to avoid telling editors that articles about, say, the Falkland Islands, have to use British English because it's disputed as to whether there is a separate Falklands Standard English. There's also the dispute over whether it's worth creating lots of rarely-used templates even when there is a demonstrable difference, because it may not actually make a difference to the spelling for encyclopedia purposes. The English variety for these articles also cannot be changed by local consensus, or at least that would create a conflict with the general MOS rule.
    For articles with no strong ties, we can't use the wording that says the spelling and grammar are controlled by that of a particular place; we're just declaring which variety has arbitrarily been chosen. This wording can also say the variety can be changed by local consensus, and point to the method for finding the earliest substantial edit in a particular variety. Presumably the list of available dialect choices will also be a lot smaller for this template. There seems to be broad support for using American or British English for no-ties articles, but tagging a no-ties article as Ghanaian English may or may not have consensus (especially if it's practically indistinguishable from British English as used in that article). We could debate that and enforce an outcome separately from the need to write articles about Ghana in Ghanaian English.
    For articles currently tagged for American English, this may improve the accuracy of spell checking for topics tied to places that use mostly but not entirely American-style spelling. I think it's probably more of a pressing issue for small countries; if the bulk of AE-tagged articles stay the way they are now for a few years, that may be unsatisfying for those that want to tidy up template space, but not really creating disputes or confusion over spelling choices. -- Beland (talk) 18:07, 21 August 2025 (UTC)[reply]
    Agree with User:Ham II]. I don't see how having additional templates for {{Ties}} and {{Retain}} would truly add value for the editor, at the cost of contributing to screen clutter.
    OTOH, as to the direction of travel, my preference would be to keep only a slimmed-down collection of {{Use X English}} templates. We could even resurrect that {{EngvarB}} (instead of British) template and {{EngvarA}} (instead of American) as palliatives to "[anti-]imperialist" sentiment. Templates that do not represent canonical variants (while ignoring dialectic vernacular) could be consolidated through a merger into a template of one of the 4 canonical varieties.  Ohc revolution of our times 18:17, 21 August 2025 (UTC)[reply]
    That doesn't seem to fit with what was agree to above. -- Beland (talk) 18:44, 21 August 2025 (UTC)[reply]
    @Ohconfucius There simply are not 4 canonical varieties though. You say ignoring dialectic vernacular as if the only formal written varieties of English are AmE, BrE, CanE or Oxford, which is clearly not the case. Also, I'm sorry to have to ask again, but canonical according to whom? I've looked, and I can't find any evidence that consensus is to ignore WP:ENGVAR when it says "All national varieties of Standard English are equally acceptable on English Wikipedia"—or, prior to recent changes, its previous wording "The English Wikipedia prefers no national variety over others"—and limit the encyclop(a)edia to four varieties (well, three varieties and one spelling system). Pineapple Storage (talk) 19:18, 21 August 2025 (UTC)[reply]
    If only admins can edit editnotices, that would be a serious hindrance, so I'd consider talk page + and non-displaying bot tag the way to go. (Just as we do it with {{Use Oxford spelling}} and friends, which all have their talk page cousin.) Gawaon (talk) 18:50, 21 August 2025 (UTC)[reply]

    Summary

    [edit]

    The following is a quick summary of the above discussion as of the datestamp here:

    • Consensus against adopting either Proposal 1 or Proposal 2.
    • Consensus for creating {{English ties}}.

    The following is a more comprehensive summary of the above discussion as of the datestamp here:

    • The original poster clarified (here) that the controversy they'd seen was on whether or not some national varieties of English were unsuitable for Wikipedia. An editor noted (here) that the original poster may have misinterpreted their stance or the basis for their stance. Another editor noted (here) that the original poster may have misinterpreted the controversy, noting it might rather be on whether or not some national varieties of English were distinct enough to need a separate label or template. A third editor agreed (here) that the original poster may have misinterpreted the controversy, noting it might rather be on whether or not some varieties of English were non-standard, or on whether or not this or that English variety tag would most help editors to maintain a consistent style. A fourth editor noted (here) that at least some editors on TfDs for English variety templates seemed to take a stance on whether or not some national varieties of English were unsuitable for Wikipedia.
    • An editor noted (here) that it would be useful to have a list of TfD discussions on English variant templates.
    • An editor noted (here) that ENGVAR maybe should not make reference to varieties of English, but rather to dialects of English, to avoid ambiguity.
    • An editor proposed (here) four specific copy edits to ENGVAR. Three of four editors seemed to agree; one of four editors seemed to disagree.
    • An editor proposed (here) creating {{English ties}} to preclude the need for discrete English variant templates for TIES articles. Five of five editors seemed to agree. The proponent later proposed (here) splitting {{English ties}} into {{ties}} for TIES articles, and {{English variety}} for RETAIN articles. Two of two editors seemed to agree. An editor proposed (here) one specific copy edit to the {{English ties}} notice. The proponent later asked for guidance (here) on the placement of the {{ties}} and {{retain}} (ie {{English ties}} and {{English variety}}) notices.
    • An editor proposed (here) possibly restricting ENGVAR for non-TIES articles.
    • An editor proposed (here) one specific copy edit to COMMONALITY. One of one editors seemed to disagree.

    Ends. Asdfjrjjj (talk) 23:40, 17 August 2025 (UTC); updated 01:27, 5 September 2025 (UTC)[reply]

    Table

    [edit]

    The following is a list of previous ENGVAR discussions and TfDs for English variant templates.[19]

    Date Link Result Description
    14 Aug 2025 MOS talk null editor proposed a non-specific copy edit to COMMONALITY to flag its contradictory readings
    13 Aug 2025 TfD null on {{Use Antiguan and Barbudan English}} nominated because template encouraged use of terms not suitable for Wikipedia per COMMONALITY
    4 Aug 2025 MOS talk null editor sought clarification on consensus reading of COMMONALITY
    29 Jul 2025 TfD delete {{Use Cameroonian English}} and {{Cameroonian English}} nominated because template recommended use of non-standard terms not suitable for Wikipedia per COMMONALITY
    26 Jul 2025 TfD delete {{Use Fooian English}} nominated because template was unused or not needed
    3 Jul 2025 EngvarB talk null editor noted {{EngvarB}} would be redirected to {{Use British English}}
    24 Jun 2025 MOS talk redirect {{EngvarB}} to {{Use British English}} editor asked for guidance after TfD; consensus seemed to be: "update {{Use British English}} doc to explain and emphasise that {{Use British English}} is not exclusively for articles with TIES to UK, but rather any article with certain words and spellings"
    12 Jun 2025 MOS talk null editor asked for help applying ENGVAR and SIC to a particular quote (loth vs loath)
    27 May 2025 TfD deprecate {{EngvarB}} nominated after vars EngvarB talk discussions and because dialect-specific tags existed and were better; consensus seemed to be: "TIES articles should use dialect-specific rather than non-specific tags, and EngvarB is not a good name, and bots should respect dialect-specific tags or not be used, and articles with TIES conflicts should follow RETAIN or create and use a spelling-specific (rather than dialect-specific) tag like {{Use Oxford spelling}}"
    1 Apr 2025 TfD delete {{Use Tanzanian English}} nominated because template was unused, and template nor main article provided guidance on how to write in Tanzanian English
    10 Mar 2025 MOS talk null editor proposed to soften the wording in COMMONALITY re the preferred spelling variant when more than one coexisted within a national variety of English (towards vs toward in AmE and BrE)
    28 Feb 2025 TfD delete {{Bangladeshi English}} nominated because it provided invalid guidance to editors
    20 Feb 2025 TfD delete {{Use Bangladeshi English}} nominated because template advised use in articles with "Bangladeshi English spelling" despite main article not mentioning any distinctive spelling, and because template provided guidance contrary to COMMONALITY
    27 Jan 2025 EngvarB talk null editor noted there was consensus on to deprecate {{EngvarB}}
    27 Jan 2025 TfD delete {{Antarctic English}} nominated because template was unused, and there was no Antarctic English dialect
    21 Jan 2025 MOS talk null following a Bicolor cat talk discussion, editor sought clarity on whether an article's having a title in one particular English dialect should determine whether that dialect was likewise used in the body per ENGVAR
    14 Dec 2024 MOS talk null editor sought to extend TIES beyond English-speaking nations
    23 Nov 2024 EngvarB talk null editor noted {{EngvarB}} may not be well-named
    23 Nov 2024 TfD keep {{Use Ugandan English}} and {{Ugandan English}} nominated because templates were created without prior discussion, and because main article noted phrasing that would not be acceptable per COMMONALITY, and noted misspelt standard English words that would not be acceptable; consensus seemed to be: "UgE may not be standardised but is nonetheless its own Engl variant"
    22 Nov 2024 TfD keep {{Use Hiberno-English}} and {{Hiberno English}} nominated because main article noted IrE standards aligned with BrE, and regionalisms were to not be used per COMMONALITY, so templates were redundant to their BrE counterparts; consensus seemed to be: "IrE might look like BrE but is distinguishable from it, eg by use of Irish loanwords"
    19 Nov 2024 EngvarB talk null editor noted the X English main articles of vars {{Use X English}} templates did not note substantive spelling differences from BrE, and so vars {{Use X English}} templates might be redundant to {{Use British English}}
    15 May 2024 MOS talk null editor sought to copy edit TIES to make it clear it did not apply to titles (Bangalore vs Bengaluru in Indian English)
    1 Mar 2024 MOS talk null editor sought clarification on whether § International organizations was a TIES-like principle for articles with strong ties to international organisations
    23 Feb 2024 MOS talk null editor proposed TIES
    17 Aug 2023 MOS talk null editor sought ENGVAR guidance on percent vs per cent in {{Use British English}} articles
    10 Aug 2023 EngvarB talk support deprecating {{EngvarB}} nominated because its name or purpose or function were unclear; consensus seemed to agree
    26 Jul 2023 MOS talk null editor sought to copy-edit RETAIN to clarify that there was nothing special about the originally-established English variant nor editor who originally established it, in the face of consensus to change the established variant
    26 Jul 2023 MOS talk null editor sought advice on whether to change the established English variant and date format of Tartan Day after no one had objected nor consented to the change in the talk discussion
    24 Apr 2023 MOS talk null editor sought COMMONALITY clarification for theatre vs theater
    22 Apr 2023 TfD null on {{Scottish English}} nominated because Scottish English was not a national variety of English for ENGVAR
    20 Mar 2023 MOS talk null editor sought COMMONALITY clarification for TIES articles (lakh vs hundred thousand)
    1 Apr 2022 MOS talk null editor sought ENGVAR clarification on whether the main spelling variants of a dialect (gray, theater, draft in AmE) were required over minor spelling variants (grey, theatre, draught in AmE)
    27 Feb 2022 TfD keep {{Antarctic English}} nominated because it was used only in one page where {{Use British English}} would do
    30 Nov 2021 TfD delete {{Use Talossan English}} nominated because it was unused and unlikely to be used
    10 Nov 2021 MOS talk null editor sought ENGVAR clarification for articles that contained different sections for different countries (eg driver's license, wireless microphone licensing)
    21 Oct 2021 TfD delete {{Commonwealth English editnotice}} null
    22 Jul 2021 MOS talk null following a suggestion to use Indian English in biographies of Indian people, a new page patroller sought ENGVAR clarification on whether they were supposed to be adding {{Use X English}} templates to articles they patrolled, and whether there was a tool to convert articles they patrolled to the correct English dialect
    14 May 2021 TfD delete {{Use Commonwealth English}} and {{Commonwealth English}} nominated because there was no single written standard for Commonwealth English or there was no Commonwealth English dialect, and because templates prescribed spellings not backed by sources; consensus seemed to agree
    7 May 2021 TfD delete {{Israeli English}} nominated because Israel is not an English-speaking country for ENGVAR
    6 May 2021 TfD redirect {{Bermudian English}} to {{Commonwealth English}} nominated because there was no evidence that a distinct written standard of Bermudian English existed, nor that any article was written in such a dialect
    6 May 2021 TfD redirect {{Sark English}} to {{British English}} nominated because there was no evidence that a distinct written standard of Sark English existed, nor that any article was written in such a dialect
    29 Apr 2021 TfD relist {{Israeli English}} null
    29 Apr 2021 TfD delete {{Nepali English}} nominated because there was no evidence that a distinct written standard of Nepali English existed (and main article noted no one in Nepal could agree on a standard), nor that any article was written in such a dialect
    29 Apr 2021 TfD delete {{Pitcairn Islands English}} nominated because no identifiable written standard variety of Pitcairn Islands English existed, and no evidence that any article was written in such a dialect
    22 Apr 2021 MOS talk null editor sought to overhaul TIES as they identified only two standard written varieties of English (BrE, AmE) since COMMONALITY precluded use of distinctive dialectal expressions, and current wording gave rise to numerous useless templates, led to nationalist or territorial tagging of articles, and TIES rather ought to focus on spelling differences than on varieties of English
    22 Apr 2021 TfD relist {{Israeli English}} null
    22 Apr 2021 TfD relist {{Nepali English}} null
    22 Apr 2021 TfD relist {{Pitcairn Islands English}} null
    21 Apr 2021 MOS talk null editor proposed deprecating all Varieties of English templates, noting they served no good purpose but rather usually served as a form of nationalist territorial marking, whereas an IP editor was mass-tagging articles with {{Ghanaian English}} despite there being no written standard of Ghanaian English that differed from standards of written American or Commonwealth English (w/ or w/o Oxford spelling), and despite COMMONALITY precluding use of distinguishing dialectal expressions
    6 Mar 2021 TfD delete 23 X English talk page templates nominated because all but two templates were unused, and because nearly all corresponding English variants had either no main article or had just a redirect as one, and because the corresponding English variant of some (eg {{Jersey English}}) did not seem sufficiently different from BrE to warrant their own templates; the two used templates ({{Malawian English}} and {{Botswana English}}) were nominated because they otherwise had the same problems as the rest
    3 Mar 2021 TfD delete {{Montserratian English}} nominated because template was unused, had no Montserratian English main article, and corresponded to an English variety that was too niche to need a separate template
    28 Feb 2021 TfD delete {{Use English English}} nominated because there was no difference between English English and British English
    10 Feb 2021 TfD delete {{Herm English}} nominated because template was unused and unlikely to be used
    10 Feb 2021 TfD delete {{English English}} and {{English English editnotice}} nominated because templates were unused and redundant to their BrE counterparts
    2 Dec 2020 TfD redirect {{Scottish English editnotice}} to {{Scottish English}} nominated because templates were exact duplicates in displayed text and function
    17 Nov 2020 MOS talk null editor sought guidance on English variant for EU-related articles
    9 Oct 2020 TfD keep {{EngvarB}} nominated because template was redundant to {{Use British English}}
    13 Sep 2020 MOS talk null editor sought guidance on whether English English was a recognised variety of English, as it seemed covered by British or Commonwealth English, whereas they had noticed an attempt to tag articles with templates flagging English English
    13 Sep 2020 TfD delete {{South Georgia and the South Sandwich Islands English}} nominated because template was unused and not likely to be used, and might be redundant to {{British English}}
    31 Aug 2020 TfD delete {{EngvarB spelling}} nominated because template was rarely used and seemed to duplicate {{British English}}
    31 Aug 2020 TfD delete {{EngvarO spelling}} nominated because template was rarely used and seemed to duplicate {{British English Oxford spelling}}
    22 Aug 2020 TfD delete {{Brecqhou English}} nominated because template was unused and not likely to be used
    24 Jul 2020 TfD keep {{British English editnotice}} nominated because it served a similar purpose to {{Use British English}}
    12 Jul 2020 TfD delete {{EngvarC spelling}} nominated because it was used only once where {{Canadian English}} would do
    9 Jul 2020 TfD delete {{Jethou English}} nominated because template was unused, and nominator could not confirm existence of a Jethou English dialect
    20 May 2020 MOS talk null editor sought to revisit the 2013 deletion of {{Malaysian English}} as articles with TIES to Malaysia had to be flagged as written in Malaysian English per ENGVAR
    1 May 2020 MOS talk null editor sought TIES clarification for EU-related articles
    20 Mar 2020 MOS talk null editor sought clarification on why a copy-edit to TIES had removed mention of Maltese English and added mention of British English for the Institutions of the European Union article
    28 Nov 2019 TfD delete {{Hiberno English editnotice}} nominated because Hiberno-English referred to spoken, vernacular dialects, not a standard, written English dialect, such that no Wikipedia article ought to be written in Hiberno-English
    22 Nov 2019 MOS talk null editor sought TIES clarification on whether features are on (BrE) or in (Shetland Scots) vars Shetland islands
    20 Nov 2019 TfD relisted {{Hiberno English editnotice}} null
    11 Nov 2019 TfD relisted {{Hiberno English editnotice}} null
    28 Oct 2019 MOS talk null editor sought TIES clarification on what constituted strong ties
    6 Apr 2019 TfD keep 17 {{Use X English}} templates nominated because they didn't display edit notices (like their X English talk page template counterparts); consensus seemed to be: "these are non-displaying maintenance templates for bots or scripts or categorisation"
    27 Mar 2019 MOS talk null editor sought clarification on whether dmy dates established BrE for RETAIN
    5 Mar 2019 TfD delete {{Zambian English}} nominated because template was unused
    6 Jan 2019 TfD null on {{Batswana English}} nominated because template was unused and corresponding dialect had no main article
    29 Dec 2018 TfD relist {{Batswana English}} null
    14 Dec 2018 MOS talk null editor sought TIES clarification on what constituted strong ties
    22 Feb 2018 TfD keep {{Use Jamaican English}} nominated because Jamaican Standard English used British spelling and grammar, so template was redundant to {{Use British English}}
    11 Feb 2018 TfD relist {{Use Jamaican English}} null
    24 Jan 2018 TfD merge {{IUPAC spelling US}} to {{American English}} and {{IUPAC spelling}} and {{British English Oxford spelling}} to {{British English}} nominated because IUPAC spelling did not only apply to AmE and BrE, and Oxford spelling did not only apply to BrE, and so these spelling systems might be better tracked by parameters in each {{Use X English}} template or their talk page counterparts
    24 Oct 2017 TfD delete {{Zimbabwean English}} nominated because template was unused and had no main article
    8 Jul 2017 TfD delete {{EngvarMP}} nominated because template was unused
    31 May 2017 TfD delete {{Upper Class English}} nominated because there was no defined Upper Class English, and upper class English usage might differ across varieties of English, and an upper class English dialect ought not to be used in articles
    28 May 2017 TfD keep {{Engvar}} nominated because consensus in this RfC was against template use; consensus seemed to be "nominator misunderstood template's use"
    15 Mar 2017 TfD redirect {{EngvarO}} to {{Use British (Oxford) English}} nominated because template's name was unclear, and seemed redundant to older {{Use British (Oxford) English}}
    6 Mar 2017 TfD relist {{EngvarO}} null
    26 Feb 2017 TfD relist {{EngvarO}} null
    23 Dec 2016 TfD delete {{Hawaiian English}} nominated because template was unused and there was no Hawaiian variety of standard written English
    23 Dec 2016 TfD redirect {{EngvarA spelling}} to {{American English}} nominated because template was redundant to {{American English}}
    20 Oct 2015 TfD null on {{Scottish English}} nominated because template was used only twice, there was no difference between Scottish English and BrE, and Scotland was part of the UK such that the template seemed to be a POVFORK
    9 Aug 2015 TfD null on {{British English Oxford spelling}} nominated because template had only minor stylistic differences from {{British English Oxford spelling editnotice}}
    30 Dec 2013 TfD delete {{Malaysian English}} nominated because Malaysian English was not a native variety of English
    30 Dec 2013 TfD delete {{Brunei English}} nominated because Brunei English was not a native variety of English
    5 Oct 2013 TfD null on {{EngvarB}} nominated because template was redundant to {{Use British English}} and {{British English}}, there was no evidence that script maintainers were ensuring pages were being properly tagged as written in BrE per ENGVAR, no evidence that pages written in AmE had a similar script, and template produced unsightly white space in at least one page
    27 Sep 2013 TfD relist {{EngvarB}} null
    7 Sep 2013 TfD keep {{English variant notice}} and all templates in Category:Varieties of English templates nominated because these talk page or edit notices did not guide editors on how to match established style, rather serving only to scare away those that might make silly orthographic changes, whereas a show of nationalism was not needed for this effect
    21 May 2011 TfD null on {{Philippine English}} nominated because template was used only once, and per main article the English variety was substantially similar to AmE so edit notice might be redundant to {{American English}}
    2 May 2011 TfD withdraw nomination of {{Use British English}} nominated because template's flag topicon looked disgusting and served little purpose, whereas newbies were unlikely to see or care about dialect consistency requests, while experienced editors could just tell the established dialect via spelling, and talk page notices were available, and edit notices might be created
    28 Apr 2010 TfD delete {{Quebec English}} nominated because template was redundant to {{Canadian English}} and we did not have separate templates for regions of large English-speaking countries eg US
    9 Mar 2010 TfD keep {{Scottish English}} nominated because Scottish English was substantially similar to BrE per main article so template might be redundant to {{British English}}, and because Scottish English might also refer to Scots, which would be a separate Wikipedia and so confusing
    9 Mar 2010 TfD keep {{Hiberno-English}} nominated because main article noted Hiberno-English was substantially similar to BrE so template might be redundant to {{British English}}
    9 Mar 2010 TfD keep {{New Zealand English}} nominated because main article noted NZE was very similar to AusE so template might be redundant to {{Australian English}}
    9 Mar 2010 TfD keep {{Pakistani English}} nominated because main article noted PakE was virtually identical to Indian English so template might be redundant to {{Indian English}}
    15 Apr 2009 TfD delete {{Singapore-English}} nominated because template was unused, and Singaporean English was obscure, and its differences from BrE were marginal such that very few article would demand SingE particularities
    26 Oct 2008 TfD keep {{Singapore-English}} nominated because template was unused and did not have a main article
    15 Aug 2008 TfD delete {{European-English}} nominated because template was used only once, and European English referred either to BrE or English as used by European organisations or Euro-English (a stub noting "euro-English has no central norm"), and template did not note differential spelling from AmE or BrE so might be redundant to {{British-English}}
    21 May 2008 TfD delete {{AmEng notice}} and {{BrEng notice}} nominated because templates were non-temporary article message boxes, whereas talk page notices ({{American-English}} and {{British-English}}) already existed
    2 Jun 2005 TfD delete {{AmE}} and {{BrE}} nominated because templates were not used and both English dialects were accepted by MOS so corresponding notices ("This page is in <American/British> English") served no purpose

    1. ^ This is only the primary reason for (1); other reasons for (1) might exist. This isn't my position so I might not have correctly identified the primary reason for it, or else rendered it poorly here, but maybe Jonesey95 might verify? Seems like Jonesey95's and Ohconfucius's [not sure they wanted to be pinged] and Johnuniq's and Dgp4004's and Largoplazo's position per linked TfD.
    2. ^ This is only the primary reason for (2); other reasons for (2) might exist. Copied from 5 Aug comment by Asdfjrjjj in linked TfD. Seems like Pineapple Storage's position per the same linked TfD. Seems like Beland's and Trovatore's position per this MOS discussion (ignore Asdfjrjjj's mangled comments there :).
    3. ^ Could not think of an example for this but other editors might have some (maybe Jonesey95)?
    4. ^ For instance, by leading editors to waste their time and effort, eg by creating {{Use X English}} and related templates that then just go to TfD where (1) and (2) both have the same normative weight, such that discussion just becomes a matter of how well-known the X English dialect is (ie a popularity contest), rather than a discussion on whether the template itself breaks any policy or guideline (or meets any WP:TFD#REASONS).
    5. ^ Though that'd be nice. It seems to come up quite a bit in vars TfDs and {{Use X English}} talk pages over the years, eg this UBE discussion and table. This is contrary to my naive comments (saying there was consensus) in this MOS discussion.
    6. ^ Compare to Pppery's closing comment in the same TfD: "Sure, you can interpret guidelines like MOS:COMMONALITY they way Asdfjrjjj did in favor of specific dialects, but you can also interpret them in the way the nominator did, as requiring only broad-scope dialects that can't be included in more common cases."
    7. ^ Else, we'd be saying both that X does not follow from COMMONALITY and X does not not follow from COMMONALITY (= X does follow from COMMONALITY).
    8. ^ But I could be way off, ofc. This is not an exhaustive list of stance (A) comments. The last couple in this list seem to pretty clearly not be based on COMMONALITY, but were included just for variety. Some of these comments might be referring to a dialect's informal register rather than COMMONALITY, maybe, but in that case they'd be straightforwardly mistaken and wrong in their stance (informal registers being present in every English dialect, and obviously inappropriate in a formal encyclopaedia like this one), in which case we should prolly read them charitably and interpret them as referring to COMMONALITY imo.
    9. ^ "This template is an attempt to establish a new 'variety' of English in the MOS through an inappropriate route (i.e., it is bypassing a MOS discussion). First gain recognition at MOS for this as an ENGVAR variant that may be used."
    10. ^ Again, to emphasise, I'm using the word "deprecate"/"deprecating" in the sense of Deprecation, not in any other sense of the word.
    11. ^ Incidentally, since the TfD discussion, I realised that "assured" is spelt with only one r in (AFAIK) all varieties. I know this is because of the etymology from French assurer, whereas "occurred" derives from Latin occurro, but in practical terms in Modern English it means that ured vs urred is already an inconsistent rule.
    12. ^ Afghanistan, Bahrain, Egypt, Iraq, Israel, Jordan, Libya, Myanmar, Oman, Qatar, Somalia, Yemen, etc.
    13. ^ My best guess would be "largely BrE, but with lexical influences from local creoles, and AmE influences on spelling etc. in territories that are geographically close to the US". But again, this is just an educated guess.
    14. ^ I feel like stance (A) above would be a better/more understandable reason for your position here, in which case the question would be: is not the burden on stance (A) editors to establish a list of unsuitable dialect (eg by consensus), given ENGVAR as it currently stands, so that the burden is not on incoming editors who might be forgiven for thinking their English variety is welcome on Wikipedia?
    15. ^ Eg "This article has strong ties to <country>. It uses standard formal written English as used in this country, which for encyclopedia purposes is very similar to <AmE/BrE/blend-of-AmE-and-BrE>. Remember to prefer vocabulary common to all varieties of English, and to consistently follow the conventions of this variety of English." This might be less scary to newbies I feel, and saves editors a click imo.
    16. ^ Not a template editor, just assuming tracking categories are needed due to their existence/the Engvar script.
    17. ^ Also just felt that as Engvar script creator, we'd really like to know your thoughts if possible :) but I apologise if this was still an unwanted ping, just lmk!
    18. ^ Currently I only see Dgp4004, and Jonesey95 would prefer not to be pinged it seems. I might not have pinged editors correctly, maybe? So will try again :)
    19. ^ Adapted from the 4 Aug table in this UBE talk discussion. List meant to be complete to 2019 for MOS talk, and to 2005 for TfDs, but for sure errors of ommission possible. Errors of interpretation also highly possible in Description column. An editor has suggested (here) that TfDs in this table be documented in a place more appropriate than MOS, eg an internal category or navigation template. Wasn't quite sure how to technically do that, so made this table instead. MOS talk seemed a more appropriate place for it than userspace.

    ENGVAR and gaol / jail ?

    [edit]

    Can I replace uses of 'gaol' with 'jail' per MOS:COMMONALITY because gaol is a dated spelling? This would apply to existing articles like Cork County Gaol or Dic Penderyn.

    See Talk:Beechworth. More eyes on that thread would be useful. Andy Dingley (talk) 23:52, 21 August 2025 (UTC)[reply]

    Here's an article showing the Australian view: https://www.abc.net.au/listen/radionational/archived/booksandarts/jail-or-gaol-how-should-australia-spell-it/7532694
    As a child in school in Australia in the 1970s, I would see "gaol" and be confused by it. As an adult I still have to stop and re-read it. I suspect many Brits are the same with thinking "gaol" is archaic (can you confirm?). And of course, Americans and most non-native English readers will have vast trouble with "gaol".
    Of course, if a building has an official name including "gaol" then we use that.  Stepho  talk  00:09, 22 August 2025 (UTC)[reply]
    Per Garner's Modern English (5th ed., 2022), jail has been more common in British sources since circa 1967 and the current ratio in print in "World Englishes" is 17:1 in favor of jail. The Ngram also shows that jail took off around 1967 and has been the dominant form in BrE ever since. Fowler's Dictionary of Modern English Usage (4th ed., 2015) begins the entry for gaol with: the traditional and treacherous spellings in the UK, are now under severe and probably unstoppable pressure from jail, jailer, which are dominant in most other parts of the English-speaking world. The jail entry also says this is the more common spelling in BrE and is the preferred spelling except in historical contexts in which the gaol- forms might be more appropriate. Learner's dictionaries from Cambridge and Oxford call gaol "old-fashioned". I don't have access to OED but several sites say they label 'gaol' "archaic", including this 2013 post which states all four British dictionaries consulted list jail as standard. --MYCETEAE 🍄‍🟫—talk 14:40, 24 August 2025 (UTC)[reply]
    I don't have access to the full OED content either, but one can already see in the public part that they list "gaol" as a mere "variant of jail". Gawaon (talk) 17:35, 24 August 2025 (UTC)[reply]
    The Shorter OED (6th edition, 2007) says:

    gaol (noun & verb) var. of jail

    and under jail:

    – NOTE: In Britain gaol is used in some official contexts, but otherwise is restricted to literary use, jail being the usual form. In American English jail is the usual spelling.

    Mitch Ames (talk) 11:06, 25 August 2025 (UTC)[reply]
    Just as an aside, as a kid I read 'gaol' before I knew how it was pronounced and thought is was pronounced with a hard 'g' and almost rhymed with 'cowl'. I assumed it was a different type of jail, and that there were three things: gaol, jail, and prison! I also thought Geoff was 'gee-off'. Ah, the dangers of too much reading! TreeReader (talk) 22:34, 27 August 2025 (UTC)[reply]
    I read Tale of Two Cities in high school or so, and when we watched a video (or film, maybe?) in class, I was so confused that he said "are you the jailer's daughter", instead of the line as written. :) I think we should use "jail" as appropriate, since it's also considered proper BE. "they built the gaol to replace the old jail" doesn't work for me, though. Maybe "old facility"? SarekOfVulcan (talk) 17:12, 10 September 2025 (UTC)[reply]
    Not in that article. "Gaol" is the official spelling. Hawkeye7 (discuss) 00:15, 22 August 2025 (UTC)[reply]
    • So how do we proceed here? The Beechworth thread seems to have broken down into useless WP:IDHT retrenchment and demanding proof that tourist attraction flannel doesn't stand as WP:RS. Should we really reword a whole bunch of articles about 18th century Irish prisons on the basis of modern-day usage in Australia? Do a whole bunch of recent changes need to be rolled back? Andy Dingley (talk) 09:16, 22 August 2025 (UTC)[reply]
    • MOS:COMMONALITY is a fairly gentle suggestion -- using vocabulary common to all varieties of English is preferable), not use vocabulary common to all varieties of English or similar. "Gaol" is slightly old-fashioned but perfectly correct in BrE, so while there might be good reasons to opt for "jail" (or "sulfur", "fetus" and similar), it would be inappropriate to insist on a change on the grounds of COMMONALITY in an article with WP:TIES to the Commonwealth. UndercoverClassicist T·C 09:33, 22 August 2025 (UTC)[reply]
      MOS:COMMONALITY also says "When more than one variant spelling exists within a national standard variety of English, the most commonly used current variant should usually be preferred" and there are exceptions listed when there's a different meaning with a different most-common spelling, proper names, and direct quotations. Wikipedia:Manual of Style/Spelling#English spelling comparison chart points out that even in British English, "jail" is the more common modern variant, and so should be the preferred one when it's lowercase. In fact, that chart shows "jail" is the most common modern variant in the UK, Ireland, South Africa, New Zealand, Canada, and the United States. I'm not sure any standard dialect in the world actually uses "gaol" more commonly. -- Beland (talk) 17:03, 22 August 2025 (UTC)[reply]
    • Avoid both, and use "prison" except in names of institutions with "Gaol" (and any that might have "Jail"). Ham II (talk) 09:54, 22 August 2025 (UTC)[reply]
    There is a legal technical difference between prison and { jail, gaol }. Also most of these are using 'Gaol' in the first place because that's some fundamental part of that specific article, like a historical title, e.g. Beaumaris Gaol [13]. Of course if we're writing ab initio we would choose clear, common language. Nobody is looking to write an article, 'Should electric motorbike riders be sent to gaol?' Andy Dingley (talk) 11:07, 22 August 2025 (UTC)[reply]
    How can we tell which institutions are jails and which are prisons? The article Beaumaris Gaol describes it as both a "gaol" and a "prison". -- Beland (talk) 16:55, 22 August 2025 (UTC)[reply]
    The distinction of terms, in the UK (but not in the US), is mostly down to chronology (as I found out during my recent self-revert on the Wilde poem) -- the prison system was nationalised in 1878, after which most prisons were officially called "Her Majesty's Prison Suchandsuch". Beaumaris closed in 1877, so wouldn't ever have officially held that name. Other laws earlier in the C19th used "gaol" for all of them. The Beaumaris article correctly says "gaol, or prison" -- "prison" being provided as an alternative word for "gaol" rather than an alternative function it may have served. UndercoverClassicist T·C 17:09, 22 August 2025 (UTC)[reply]
    I'm a bit confused by this. Words have meanings; if "prison" means something different than "gaol", it should not be presented as a synonym for "gaol". For example, in the US "prison" formally means long-term detention and "jail" means short-term detention. "Jail or prison" would not be a correct formal descriptor of most facilities (perhaps there are some that do both). If "gaol" and "prison" mean the same thing in the UK system, then for the sake of native English speakers who don't have "gaol" in their standard national dialect and who would have a lot of trouble figuring it out, shouldn't we use "prison" instead of "gaol" because that word is universally legible? -- Beland (talk) 17:30, 22 August 2025 (UTC)[reply]
    There's a few different things we might scope out that could also help us with spelling and word choice issues:
    • The words used in the names of institutions (in laws and other proper names in use)
    • The words used in the names of laws
    • The common nouns used to describe facilities in everyday speech
    • The legal classes of things and their names
    To start out, I took a look at the text of Gaols Act 1823. That name does not appear in the text, and that article says the same law is also called the Prisons Act 1823. It appears the names of many laws are applied retroactively and in multitudes, and many early laws may not have had names at the time they were enacted.
    Turns out it defines two classes of prison: gaols and houses of correction, with different legal requirements for each. I started documenting at Prison#UK and Ireland.
    There's a whole bunch of confinement-related laws in the UK and I don't have time to research them all now, but it would be interesting to figure out a beginning and an end time for the legal category of gaol. From what UndercoverClassicist wrote above, the end point may be Prison Act 1877. There are references on Gaols Act 1823 to laws talking about gaols or gaolers going all the way back to the 1300s, and it will take a bit of research to go past that to see if there is any notional beginning of the idea of a building of confinement, and also figure out if legal categories changed at all in those 500 years. Though that far back the statutes are in Middle English, when spelling was unstable. Wiktionary lists the spellings "gayole, gaiol, gaylle, gaille, gayle, gaile".
    I'll also note Category:Defunct prisons in the United Kingdom is a good place to find examples of names, and there is a lot of variation, many without either "Gaol" or "Prison" in the name. -- Beland (talk) 08:08, 23 August 2025 (UTC)[reply]
    In my own poking around, I found Inveraray Jail (built 1820 in Scotland), which looking at Google Books is generally referred to as a "jail", substantiating the claim on Talk:Inveraray Jail by someone who says they currently run the place that it was never known as "Inveraray Gaol". The one exception I found was in the law that created it, which of course was written in England, not Scotland. From this I infer that "jail" and "gaol" were both used historically to refer to the same class of institutions, which seems to undermine the idea that the only way to refer to UK confinement institutions from a certain era is "gaol". (Of course that law also capitalizes all nouns and uses long s and has lots of other archaic spellings we don't use, so it has forms like "Court-Houſe" instead of the modern "courthouse".) Beland (talk) 18:01, 22 August 2025 (UTC)[reply]
    Let anyone be misled: the use of long s does not represent a spelling variation; it's merely a typographical style. Long s is still ess. EEng 00:15, 23 August 2025 (UTC)[reply]

    Short answer? No, you may not replace the official names of places with names you prefer. Cork County Gaol and Cork City Gaol and Kilmainham Gaol are their names. BastunĖġáḍβáś₮ŭŃ! 10:01, 22 August 2025 (UTC)[reply]

    But what about re-writing inside that article so that Cork City Gaol now states the building of a new Cork City Gaol to replace the old jail at the North Gate Bridge (the old jail, [14] ? Andy Dingley (talk) 11:07, 22 August 2025 (UTC)[reply]
    With this diff, I have just boldly changed The Ballad of Reading Gaol to replace the second use of the word gaol to make HM Prison Reading explicit. Let's see if it survives. 𝕁𝕄𝔽 (talk) 12:19, 22 August 2025 (UTC)[reply]
    • What about this revert on Dorothy Smith, Lady Pakington? It just says "...Pakington was forced to appear before the court of high commission, and was committed to gaol." The capitalized word "Gaol" does not appear anywhere in this article, and the name of the facility that this person was sent to is not mentioned in the article. @Roger 8 Roger: on the other thread, you seemed to think "jail" was better in these instances; is "prison" OK? It's the most common variant in both British English and across national standard dialects. @Ham II: It sounds like you'd prefer "prison" here, or is "jail" OK? @Bastun: Unclear what your position is? @Andy Dingley and UndercoverClassicist: Do we need to research whether the thing being referred to was named "Gaol" or "Prison" or was part of a distinct legal class of facility regardless of name to decide between "gaol" and "prison"? There seems to be some notion that context flips the more-common "jail" to the less-common "gaol" and I'm trying to figure out for those favoring "gaol" in some circumstances what the scope of that is. -- Beland (talk) 17:45, 22 August 2025 (UTC)[reply]
      Nope -- in this case, while MOS:COMMONALITY would suggest "prison", it doesn't insist, and the relevant question is whether "gaol" is a valid word for this thing in in British English (which it is). UndercoverClassicist T·C 19:31, 22 August 2025 (UTC)[reply]
      OK, so it sounds like you're ambivalent between "gaol" and "prison" in this case? Or would you object to changing it to "prison" if others object to "gaol"? -- Beland (talk) 19:49, 22 August 2025 (UTC)[reply]
    • Use "Gaol" if it's part of the official name of an institution (hence with a capital letter), "jail" everywhere else. In that way, no commonality is wasted and spelling sanity prevails as far as possible. Gawaon (talk) 19:52, 22 August 2025 (UTC)[reply]
      In the case of the Beechworth article, the article refers to the Old Melbourne Gaol, an official name. There is a semantic difference in meaning between "gaol" and "prison": the former is a facility for the temporary holding of persons, the latter is for long-term detention. Hawkeye7 (discuss) 01:02, 23 August 2025 (UTC)[reply]
      We solved the controversy in that article by not using lowercase "gaol" but the proper names "Melbourne Gaol" and "HM Prison Beechworth". Some of Andy's recent reverts could be resolved in that way, though we'd have to verify that "X Gaol" (or "HM Prison X" or something else) is in fact the correct name. But there would be other reverts where we are not using a proper name and can't because it would be too repetitive or the text doesn't specify which facility. -- Beland (talk) 06:29, 23 August 2025 (UTC)[reply]
      What's the problem with "gaol" anyway? Hawkeye7 (discuss) 08:35, 23 August 2025 (UTC)[reply]
      Mostly that it's highly irregular, so it can be nearly impossible to sound it out and figure out what word it is supposed to be unless you've seen it before and memorized it. (As Prison#cite_note-5 says, soft g does not normally occur in English before "a", and "ao" is not a normal English diphthong.) When it's not coming across as a nonsense word, it's looking like a misspelling of "goal". This is in contrast to typical spelling differences where maybe a silent "e" gets added to the end of a word or a consonant gets doubled or a "u" jammed in. That might be an invalid spelling in one's native dialect, but it doesn't change the sounding out of the word enough to make it unrecognizable. And those changes follow rules, so once you've seen a few examples, new words that follow the same pattern are easily understood.
      Americans probably have the biggest problem because "gaol" is just not a valid spelling of any word in Standard American English, and we don't have historic buildings with "Gaol" in the name or "gaol" in any laws that we still use, as far as I know. In order to have memorized this word you'd need to be reading about jails in a foreign dialect, and that just doesn't happen very often. Even for native users of Standard British English, "jail" is becoming more common so there are fewer chances to learn "gaol" and an even lower probability that documents with "gaol" will get exported.
      We expect readers to know words in order to be able to read, but only the words in the national dialect they've learned. We can bridge this irregularity with a gloss in the same way we bridge terminology differences, like "lift (elevator)" and "pants (underwear)" - "gaol (jail)". This is perfectly comprehensible but a little bumpy. To me it seems cleaner to just write "jail" in the first place, so I'm probing the scope in which we have consensus to do that. Given that "jail" is now either the highly dominant or only spelling in all standard English dialects, we never have to gloss in the reverse direction like "jail (gaol)". -- Beland (talk) 10:55, 23 August 2025 (UTC)[reply]
      @Hawkeye7: Would you happen to have a pointer to any reliable sources that say that Australia uses jail for short-term and prison for long term, either now or during the era of the Old Melbourne Gaol? This is asserted a lot for the United States, but I couldn't find anything establishing that for Australia. (What I did find I suspect was AI-generated and might be completely inaccurate.) I'm mostly trying to document this for readers, as Prison#Australia doesn't mention any such distinction at the moment.
      Looking at List of prisons in Australia, it seems none of the modern facilities have "Jail" in the name, mostly either "Prison" or "Correction Centre". Looking at Punishment in Australia, it seems that the term "remand centre" is used for short-term pre-trial detention, and many of these are located within prisons and correction centres. It would be interesting to find out when and why "Gaol" stopped being used. -- Beland (talk) 06:18, 25 August 2025 (UTC)[reply]
      "Gaol" and "prison" fell out of use when they moved to change the names to "correction centre" and "remand centre" many years ago. I presume this happened at different times in different states. I don't think these terms have currency or recognition outside Australia. Hawkeye7 (discuss) 08:31, 25 August 2025 (UTC)[reply]
      For a reference, see History of the Criminal Justice System in Victoria, Chapter 1. Hawkeye7 (discuss) 09:23, 25 August 2025 (UTC)[reply]
    • OK, this has turned into the same mess of minutiae as happened at Talk:Beechworth
    To try and reset it, can anyone please tell me which policy (not specifically about crime and punishment) support bulk changes to an article text (not title) like these [15][16] ? What's the fundamental reason for it, and benefit derived from it?
    Unless we have a broad reason for doing it at all, we don't even need to start poking at the details. Andy Dingley (talk) 10:54, 23 August 2025 (UTC)[reply]
    A proximal reason is compliance with MOS:COMMONALITY. The fundamental reason is desire to convey information seamlessly to readers in the face of the unreadability of "gaol" for speakers of certain standard dialects (which I explain in detail above in reply to Hawkeye7 at 10:55 today).
    I'll note that the existing text of Derry Gaol uses both "gaol" and "jail", and "Derry Gaol" and "Derry Jail" without explanation. This seems untidy. -- Beland (talk) 11:01, 23 August 2025 (UTC)[reply]
    What part of MOS:COMMONALITY supports arbitrary changes to existing articles? In particular, this goes against the underlying principle of WP:ENGVAR which is to avoid churn. Churn itself being seen as harmful, even outside the argument to 'be right'. A recommendation to use accessible language is one thing, but automated bulk changes to existing articles quite another, and one we've always opposed.
    ENGVAR supports the idea of 'ties'. Now I can't think of a stronger tie than an article about one specific named Gaol then using 'gaol' within its own context. If you claim that this term is too 'unfamiliar' to readers, then they're not going to get past the title.
    There is no justification here, or in any of our policies, for these bulk, blanket changes to stamp out one particular word. Andy Dingley (talk) 21:40, 23 August 2025 (UTC)[reply]
    The changes I have made are not arbitrary, they are to implement "most commonly used current variant (across all varieties of Standard English)". Because "gaol" is not the most commonly used current variant in any national variety of Standard English, it logically follows it should not be used on English Wikipedia outside of exceptions like proper names, direct quotations, citations, and when discussing the word itself. That guideline should not result in churn, because no one should be going around changing "jail" to "gaol". A burst of editing that brings a large number of articles into compliance with consensus guidelines is not undesirable churn, it's desirable cleanup. Churn does occur if there is not agreement as to which direction the guidelines are pointing or there is a desire to change them, which is why we have stopped changing these spellings and are discussing the question.
    Based on what has been written so far, it seems some editors want there to be an exception to MOS:COMMONALITY for articles on certain facilities, and others do not. That's fine; it's a normal thing to have a discussion over. It would be helpful to the discussion to articulate the scope of that exception. For example, does it only apply to jails in the UK up to 1877? MOS:TIES only discusses "strong ties to a particular English-speaking nation". Other editors have objected to adding any other entities to this rule, such as states or provinces or regions with their own dialects. If you want to expand this rule to have something beyond ties to a "nation" influence spelling, that's fine; specific language would be helpful for the purposes of discussion. But this rule currently only concerns choice of dialect, and we all agree that articles about British jails should be written in British English. If you want to introduce the notion of ties to choice of spelling within a dialect, it's possible MOS:COMMONALITY is the right place for that, since it would effectively be an exception to the "the most commonly used current variant (across all varieties of Standard English)" rule. Specific language for discussion purposes would be helpful. Is the thing you want to see happen something like "articles with strong ties to a specific historic period should adopt the spellings of that period if those spellings are still considered correct in the national variety of modern standard English, even if they are considered old-fashioned or are less clear to some readers"?
    Nothing in MOS:ENGVAR prevents edits to "existing articles". MOS:RETAIN prevents changing an existing choice of dialect, but it does not apply to articles with MOS:TIES, which are the ones we are discussing. There are plenty of reasons to change spellings in bulk across all articles with ties to the UK, for example to change "defense" to "defence" because the latter is the British spelling.
    I think it's a bit overboard to say that the fact that "Gaol" is in the title of an article means that readers who don't know what "gaol" means can't find out by reading the article. We can just tell them what it means. We already do this implicitly in Old Melbourne Gaol which starts out "The Old Melbourne Gaol is a former jail"; I think that's fine. We could make it more explicit for readers who don't pick up on the correspondence between the two words by adding '("gaol" is an old-fashioned spelling of "jail")' but I'm not sure that's necessary. You also reverted instances of "gaol" where a specific prison is never named and it's just being used as a general concept, one time in the entire article. Cases like that seem to be single instances of an unclear word that could be improved, uncomplicated by worries about consistency with the rest of the article or confusion that could result from proper nouns with "Gaol" in them. -- Beland (talk) 01:46, 24 August 2025 (UTC)[reply]
    (What I wrote is not quote right; the clarified version of the "most common" rule not only means "use jail because it is more common in British English than gaol" but also "use jail because it is the only valid spelling in American English and also exists in British English".) -- Beland (talk) 04:12, 24 August 2025 (UTC)[reply]
    I agree. Despite the title of this topic, "gaol" vs. "jail" is not really an ENGVAR issue at all, since the latter spelling is preferred in all varieties of English (for BE, according to OED). So changing "gaol" to "jail" is (outside of proper nouns) always fine and improves the article, if in a minor way. Gawaon (talk) 07:01, 24 August 2025 (UTC)[reply]
    While I think jail should generally be used in preference to gaol per MOS:COMMONALITY, I think it's unnecessary and even confusing to require it where the article title uses Gaol (because that's it's the actual name of the institution as in the examples above). I looked at the "mass" changes to the 2 articles that Andy Dingley linked to and they really seem unnecessary. Clearly someone starting to read both those articles will know what a "gaol" is from the article text itself - there's even a reference it to being "jail". There's no need to then replace gaol thereafter - it doesn't add any extra understanding for the (say) American reader. In fact, I can see readers being puzzled why there's one spelling in the title and one in the body. I think the principle would be use jail in preference to gaol except where it would lead to inconsistency with the article title. DeCausa (talk) 09:13, 25 August 2025 (UTC)[reply]
    This seems backwards to me. Oftentimes I will only be reading part of an article, because I have skipped down to the section I'm interested in, because I've followed a link that points to a specific section, or because I'm reading a snippet in search results or an AI answer. If I encounter a word I can't read in the body of a section, it's not going to help me that word is explained in the intro to the article (which I haven't read, and I don't know to go there to decode this word). As a result, I might just not know what certain sentences mean.
    Compare that to the negative consequences of having "Gaol" and "jail" in the same article. Maybe it looks a little weird? Seems a bit incongruous? But everyone knows old spellings and foreign names are weird sometimes, so is it that out of place? If that difference is explained in the intro ("gaol" is an old-fashioned spelling of "jail") that puzzlement is cleared up for everyone reading the article from the beginning. But what if someone skips the intro and goes right to a section? They might still be a little puzzled, but able to understand everything that's written. That seems really minor compared to not being able to read some words at all.
    If an incomprehensible word only shows up inside a proper noun, then e.g. American readers who miss the intro clarification don't really need to worry about decoding it to understand the meanings of sentences - it's at least clear it's the name of a thing, if not the name of a jail. Knowing that "Gaol" in a proper name means "jail" makes that weird name make a lot more sense, but for example I don't need to know that "Lacroix" means "the cross" to know that "Thierry Lacroix" is the name of a person. I don't even need to know how Western names work - that the last part is a family name - in order to understand more about this person's life. If I continue reading I will eventually notice that the same word appears in the names of their family members, but I don't have to wait for this realization to understand other facts. -- Beland (talk) 10:07, 25 August 2025 (UTC)[reply]
    Well, you certainly need to infer a heck of a lot about our readers and the way they read articles to make your argument work. I guess if we (ever) have a reader that matches your template exactly, you'll be right! DeCausa (talk) 11:30, 25 August 2025 (UTC)[reply]
    My impression is rather that you make a lot of odd assumptions about our readers if you seriously think people would be confused by encountering the spelling "jail" in any article. Gawaon (talk) 14:57, 25 August 2025 (UTC)[reply]
    I agree with Beland and Gawaon. Readers who recognize gaol likely recognize it as old fashioned. It's an extraordinary claim that they would be "confused" by the standard spelling. Readers unfamiliar with gaol, who represent the majority, will encounter at least some friction every time they read it, even if it's been defined for them. Some will encounter gaol as a significant barrier to understanding. The behavior Beland describes is common—many readers skim and skip around. We have many practices that anticipate this behavior and aim to improve the experience for such readers. For example, linking to article sections directly, guidance bolding selected terms in sections following the lead, and guidance on redefining and repeating wikilinks to key terms in subsequent sections. --MYCETEAE 🍄‍🟫—talk 16:45, 25 August 2025 (UTC)[reply]
    On that basis we can dispense with the whole MOS principle of internal consistency within an article. No need - they'll work it out! DeCausa (talk) 17:27, 25 August 2025 (UTC)[reply]
    Just because comprehension gets priority over consistency when there's a substantial conflict doesn't mean that consistency should be abandoned. For example, choosing British-style spellings for an article without strong ties might be a good idea if it mentions a proper noun like "Foo Correction Centre". Both "center" and "centre" are comprehensible regardless of the reader's national dialect. -- Beland (talk) 17:53, 25 August 2025 (UTC)[reply]
    I think everyone knows where you stand on his and what your arguments are. DeCausa (talk) 18:01, 25 August 2025 (UTC)[reply]
    OK, that was not clear from your comment that implied we think consistency should be abandoned. -- Beland (talk) 18:14, 25 August 2025 (UTC)[reply]
    You might want to consider whether it's wise to post a response in this thread to every comment that's adverse to your view. You are an admin, after all. DeCausa (talk) 20:42, 25 August 2025 (UTC)[reply]
    Well, I don't see how we can reach consensus without conversing with people who we disagree with; we just need to keep it respectful. I am also frustrated that this conversation lacks focus on comparing rationales and reaching actionable compromise, and there are a lot of easily refuted objections coming out of left field. (You're not at fault for that, and of course helpful points have also been made on both sides.) Instead of complaining about personal jabs or the conversational style of other editors, I've been ignoring personalities and trying to ask questions that get us back to the substance of the question.
    If you truly feel I've engaged in misbehavior here, please report me at WP:AN/I. I don't think people who care about the substance of this question are interested in personality clash drama. -- Beland (talk) 01:32, 26 August 2025 (UTC)[reply]
    Whoa! Steady there. I was just making a suggestion, not saying you've misbehaved! I didn't even link to the B-word. Oops I just did. DeCausa (talk) 07:49, 26 August 2025 (UTC)[reply]
    It's a misapplication of "internal consistency" to use unusual words or spellings in the article body just because they appear in the title. We use "cemetery" in Pet Sematary and Pet Sematary (1989 film); we use "vampire" throughout Nosferatu and Nosferatu the Vampyre; we use "the" in Ye Olde Curiosity Shop and it would be bizarre to use "olde" outside of the store's name, if the adjective were needed in running text. Even a strong reading of internal consistency must be balanced with COMMONALITY and comprehension. --MYCETEAE 🍄‍🟫—talk 20:56, 25 August 2025 (UTC)[reply]
    Oh hang on. I wasn't going to bother posting any further on this thread but that's just bollocks. There's no way in British English gaol is remotely comparable to "Pet Sematary" or "olde". It's standard British English even if less used than jail. DeCausa (talk) 21:47, 25 August 2025 (UTC)[reply]
    Alright, those were on the extreme. These are on a continuum and I maintain that gaol is too nonstandard for internal consistency to overcome the objections. Perhaps examples like The Vitamin Shoppe, The Pop Shoppe, Grand Canal Shoppes are more reasonable. The articles use "shop", "store", etc. in running text. --MYCETEAE 🍄‍🟫—talk 01:12, 26 August 2025 (UTC)[reply]
    That's equally ridiculous. If you think "shoppe" is comparable to "gaol" in British English then you've miscalibrated this badly. Gaol will appear in standard British English dictionaries. DeCausa (talk) 07:15, 26 August 2025 (UTC)[reply]
    They only list is as a less preferred variant, though. So why should we prefer it in some articles just because it happens to show up elsewhere in those article, just in the name of "consistency"? What about consistency with the rest of Wikipedia, and with what the reader is used to? Gawaon (talk) 10:13, 26 August 2025 (UTC)[reply]
    • "There will be no churn because My Way Is Right" is why we have policies against churn. Andy Dingley (talk) 11:14, 25 August 2025 (UTC)[reply]
      Of course. That's why I say people need to make edits in the direction the MOS points, not their own direction, and if there is any disagreement about which direction the MOS points (as there is in this case), that needs to be resolved with discussion, as we are doing. -- Beland (talk) 16:55, 25 August 2025 (UTC)[reply]
    • MOS:RETAIN prevents changing an existing choice of dialect, but it does not apply to articles with MOS:TIES
    You are literally trying to quote a policy supporting stability to support a change in spelling against another policy supporting stability. On the basis of using a web snippet about modern Australian use to change descriptions of 18th century Ireland. This makes no sense, it's just text wallpapering until you've bludgeoned away anyone who disagrees. Andy Dingley (talk) 11:19, 25 August 2025 (UTC)[reply]
    Well, sorry for the long posts, but sometimes it takes more words to explain why an idea is wrong than it does to state the idea.
    I don't follow your logic about stability. Are you saying we're supposed to leave "defense" in articles with UK ties, for example, or are we supposed to change existing text that violates MOS:TIES? -- Beland (talk) 16:54, 25 August 2025 (UTC)[reply]
    1. If the article has a {{Use British English}} tag, then you change "defense" to "defence" without further ado.
    2. If yours is the first non-stub edit, you can tag it, and then change it.
    3. Otherwise, you should establish consensus on the talk page to change the English variant.
    Hawkeye7 (discuss) 00:53, 26 August 2025 (UTC)[reply]
    I agree with (1), but I'm asking Andy if that kind of change violates the sort of stability they are advocating for, if that spelling which apparently violates ENGVAR is long-standing. -- Beland (talk) 03:21, 26 August 2025 (UTC)[reply]
    • Use "jail" throughout the article body. Use "gaol" only where it occurs in proper names or, sparingly, in quotes or other instances where it is beneficial to highlight this usage. In an article about "X Gaol", call it the jail, etc. in prose. Explain or gloss "gaol" on first usage or early in articles that use this word in the title. Obviously, use "prison" instead of "jail" if this is the more appropriate term. --MYCETEAE 🍄‍🟫—talk 14:03, 24 August 2025 (UTC)[reply]
      Also: the (partial) proper name should not be overused. Cork County Gaol, for example, contains phrases like: The Gaol was designed in the Greek Revival style and The County Gaol was the scene of executions by hanging, which took place in public outside the Gaol until the 1860s. We would not typically write the Theatre was built or the School opened in articles about institutions with 'Theatre' or 'School' as part of the name. This isn't an ENGVAR or COMMONALITY issue but it does add to the awkwardness of these articles. --MYCETEAE 🍄‍🟫—talk 18:10, 24 August 2025 (UTC)[reply]
      For the reason I just posted above, I think it will just increase puzzlement to talk about "jail" in an article where there's "Gaol" in the title. DeCausa (talk) 09:19, 25 August 2025 (UTC)[reply]
      I've responded above to keep the thread together. --MYCETEAE 🍄‍🟫—talk 16:46, 25 August 2025 (UTC)[reply]
    What is being argued for by some is that we should have a style guide that applies to British English articles that jail always be used in everything but proper names. But although rarer, gaol remains standard British English, and what is instructive is that when British publications and institutions write about places that are named x gaol, they will use use the same spelling throughout. I presume they do so because this is simply clearer. So should we. Examples: Derby Telegraph [17], or the UK parliament [18], [19], [20]. Here's the National Archive using the term [21]. Sirfurboy🏄 (talk) 07:50, 26 August 2025 (UTC)[reply]
    What's the problem with "gaol" anyway? as Hawkeye7 said above. Both words are current English so both can be used. Just because one is commoner does mean we automatically discard the other. If an article is about a place called a gal then we use gaol throughout the article. It's the same as the UK/US spelling rule about choosing one and sticking with it. The preference for jail should only apply to new article where no actual place name is mentioned. Just because some people don't know what a gaol is doesn't mean we should dumb down. We should assume a certain level of intelligence. Roger 8 Roger (talk) 08:01, 26 August 2025 (UTC)[reply]
    My post above crossed with sirfurboy's. Coincidentally, the same argument. Roger 8 Roger (talk) 08:20, 26 August 2025 (UTC)[reply]
    Yes, I think there's quite a few non-British editors in this thread that have an erroneous impression of how "antiquated" gaol is in British English. One (above) has even equated it with "shoppe" and "olde"! DeCausa (talk) 08:46, 26 August 2025 (UTC)[reply]
    Not if you keep MOS:COMMONALITY in mind, though. Gawaon (talk) 10:19, 26 August 2025 (UTC)[reply]
    Writing "jail" instead of "gaol" is not "dumbing down". It conveys exactly the same information, except with a spelling that more readers can successfully interpret. The point of MOS:COMMONALITY is exactly to avoid readers having to run to the dictionary when simply picking a different spelling would make a sentence comprehensible on its face.
    Readers who don't know how to pronounce "gaol" are not less intelligent; they simply have not been exposed to it because they live in the US or Canada or Australia or NZ, where Wikipedia:Manual of Style/Spelling tells us this spelling is not part of the standard language. (And haven't happened to read either specialized or foreign publications, perhaps because it's never come up, or perhaps because they are in elementary school.)
    I don't think it's reasonable to assume that the Derby Telegraph thinks that "gaol" is clearer for its American readers. It is not writing for an international audience, it is writing for the people of Derbyshire and their neighbors - at most England or the UK. -- Beland (talk) 18:11, 26 August 2025 (UTC)[reply]
    Yes, writing "jail" instead of "gaol" is 'dumbing down' because it to underestimate our readers and assume that they can't cope with the original, still-correct, word. This is Wikipedia, not Simple Wikipedia. Andy Dingley (talk) 00:56, 27 August 2025 (UTC)[reply]
    I'm not sure I would say "can't cope"; nearly anyone who can read the encyclopedia could look a word up in a dictionary. But isn't just factually accurate to estimate that many if not most of our readers would in fact have to look this word up in a dictionary to understand it? -- Beland (talk) 04:03, 27 August 2025 (UTC)[reply]
    Readers from Australia should be familiar familiar with the "gaol" spelling. Hawkeye7 (discuss) 01:15, 27 August 2025 (UTC)[reply]
    That probably depends a lot on how old they are and whether or not they read about certain topics, and we've heard a personal experience above that even with passing familiarity it's problematic. We could research the comprehensibility in Australia in more depth, but it hardly seems to matter as the vast majority of Americans certainly aren't familiar with this spelling. -- Beland (talk) 04:00, 27 August 2025 (UTC)[reply]
    And we can find British, Irish, and Australian sources that use jail in articles about a named Gaol.[22][23][24][25][26] --MYCETEAE 🍄‍🟫—talk 16:27, 26 August 2025 (UTC)[reply]
    And, indeed, American universities publishing papers using gaol [27]. But given that it is clearly a valid current alternative spelling, in use in the British parliament, the British National Archive interpretive information, and academia, as well as newspaper sources, there is simply no reason to impose a style restriction on pages, especially when we note in your examples that when the article uses jail it may find the need to explain itself. It is clearer just to use a consistent spelling, although nothing decided here should over-ride a local consensus relevant for a specific page. Sirfurboy🏄 (talk) 17:34, 26 August 2025 (UTC)[reply]
    Using "gaol" instead of "jail" does not obviate the need for the article to explain itself, as many if not most readers won't know what "gaol" means. The publishers you mention above are for a British national audience, or a specialized audience, not a general international audience.
    When you say "there is simply no reason to impose a style restriction on pages" to avoid using valid spellings, that sounds like an argument for repealing the "most commonly used current variant" of MOS:COMMONALITY. Editors obviously had reasons for adding that in the first place, and I doubt there would be consensus to remove it. But it sounds like you're specifically concerned about the tradeoff between consistency and clarity. Would you support adding an exception like: "except where the article must use a specific spelling, for example in a proper name or direct quotation"? -- Beland (talk) 17:55, 26 August 2025 (UTC)[reply]
    Using "gaol" instead of "jail" does not obviate the need for the article to explain itself - sure, but whereas you can just wikilink gaol, if you start using jail instead, you need to include a sentence of explanation as to why you did so. The publishers mentioned are not just writing to a British audience, because we have seen examples in Australia, Ireland and the USA. You just did not reply on those, nor on the use by Parliament nor TNA. as for "Would you support adding an exception like..." - no. I don't support the multiplication of intricate stylistic rules. Gaol is a variant spelling. It is still in use, including, as we have seen, by academics, parliamentarians and information professionals as well as journalists. I'd support dropping the stick on this one. Sirfurboy🏄 (talk) 20:39, 26 August 2025 (UTC)[reply]
    MOS:COMMONALITY says When more than one variant spelling exists within a national standard variety of English, the most commonly used current variant (across all varieties of Standard English) should usually be preferred, except where the less common spelling has a specific usage in a specialized context, e.g., connexion in Methodist connexionalism. That means if the article is about a proper noun article that includess gaol, stick with gaol throughout. Otherwise jail is 'usually' preferred, ie it's not set in stone. So, why were these mass changes ever made, there's no need?Roger 8 Roger (talk) 22:09, 26 August 2025 (UTC)[reply]
    Roger, if you're saying this falls under the specialist meaning exception, I'm not quite seeing that. The word "gaol" does not refer to a specialized concept or only to jails in one place or period. The most common way to refer to institutions with "Gaol" in the name using that term is "jail", including in British English.
    I do think the specialized meaning applies to something like "gaol fever", where that is actually still the most common spelling (perhaps because this concept is no longer present in modern medicine) and I support retaining that.
    Sirfurboy, sure, some publishers do use "gaol" in the UK, including Parliament and the BNA. As MYCETEAE points out, there are other British publishers that use "jail", and based on the Ngrams graph it appears they are a minority maybe 1/7th of usage) and extrapolating long-term trends, that spelling may just go away entirely in the UK. Wikipedia:Manual of Style/Spelling tells us "gaol" is so little used in the US, NZ, Australia, and Canada, that it's not considered a proper spelling.
    I don't think wikilinking is sufficient; it doesn't help people using print editions or seeing the word in search engine results or AI summaries or excerpts on other web sites. It's just as much of an interruption as going off and looking up the word in a dictionary. If we expect people to go off and do their own research to answer questions raised by the spelling in the article, I don't see why we'd put in a note that "gaol is an old-fashioned spelling of jail", either.
    If your rationale for not changing "gaol" to "jail" is that "gaol is a valid word in British English", that implies to me that we should always ignore the "most commonly used current variant" clause of MOS:COMMONALITY. Would you be in favor of removing that to avoid this sort of issue coming up again, or are there some circumstances in which that advice should be followed? -- Beland (talk) 03:52, 27 August 2025 (UTC)[reply]
    But unlike connexion (as in Connexionalism), gaol has no specialized meaning. It's defined everywhere as a spelling variant of jail. You yourself describe it this way. --MYCETEAE 🍄‍🟫—talk 15:16, 27 August 2025 (UTC)[reply]
    sure, but whereas you can just wikilink gaol, if you start using jail instead, you need to include a sentence of explanation as to why you did so. A wikilink is not a substitute for an explanation (see: MOS:NOFORCELINK) and at a minimum significant ENGVAR differences need to be glossed (MOS:COMMONALITY). Even if we didn't have these guidelines, I'm baffled by the notion that using a less common regional variant is somehow less confusing or requires less explanation than using a universally-recognized variant. --MYCETEAE 🍄‍🟫—talk 14:57, 27 August 2025 (UTC)[reply]

    I mean for example, about specialised meanings, if the article is about melbourne gaol, then use the spelling gaol whenever the word is used in that article, not just when referring to the actual building called melbourne gaol. Gaol is a normal contemporary spelling of the word, not an archaic spelling, even if used less than jail. Also, the mos guidelines say usually use the commoner version, meaning there will be exceptions and allowing for some flexibility and common sense by editors. Roger 8 Roger (talk) 08:25, 27 August 2025 (UTC)[reply]

    This reminds me of when Daniel Lambert was TFA way back in December 2010, we had all manner of people wanting to change "gaol keeper" to either "jail keeper" or "goalkeeper". If the soure says "gaol", we should stick with that spelling. --Redrose64 🌹 (talk) 19:16, 6 September 2025 (UTC)[reply]
    What about articles that use more than one source? (Hopefully, the huge majority!) Gawaon (talk) 20:50, 6 September 2025 (UTC)[reply]
    Well, "gaol keeper" is very hard not to read as "goal keeper", because "gaol" is rare or unknown in a reader's dialect, and "goal keeper" is common. I'm not sure why the right response to complaints about that article was to blame people for not reading carefully, instead of using the common spelling so it's easy to read in a single pass. -- Beland (talk) 23:23, 6 September 2025 (UTC)[reply]
    Good grief. The lead at Daniel Lambert is an example of what not to do. Readers should should not be forced to click a link or read an explanatory note to understand a basic vocabulary word. --MYCETEAE 🍄‍🟫—talk 17:05, 10 September 2025 (UTC)[reply]
    Indeed! Gawaon (talk) 17:50, 10 September 2025 (UTC)[reply]
    Tried to fix it and was swiftly reverted. *sigh* --MYCETEAE 🍄‍🟫—talk 21:41, 10 September 2025 (UTC)[reply]
    I suppose one who would argue to retain this usage might also be described as a gaol keeper, of sorts… --MYCETEAE 🍄‍🟫—talk 21:46, 10 September 2025 (UTC)[reply]
    On the contrary. That FA had a good solution which had gone through our featured article process). You tried "fix" something that didn't need fixing by introducing pointless repetition. Good revert. DeCausa (talk) 21:51, 10 September 2025 (UTC)[reply]

    Host/website names in text

    [edit]

    Where's the guideline, if there is one, for displaying in text a website name that takes the form of its domain name? When I came across the diff at [28], where an editor changed In August 2023, Namecheap launched Spaceship.com to In August 2023, Namecheap launched Spaceship.com, I realized I didn't know what the approved treatment is. Which of the following?

    • Spaceship.com
    • Spaceship.com
    • "Spaceship.com"
    • SpaceShip.com
    • SpaceShip.com
    • "SpaceShip.com"
    • spaceship.com
    • spaceship.com
    • "spaceship.com"
    • spaceship.com

    I looked up MOS:URL and MOS:DOMAINNAME, but those lead to guidelines for links. Largoplazo (talk) 17:56, 29 August 2025 (UTC)[reply]

    I'd keep it simple, using spaceship.com or possibly Spaceship.com. Gawaon (talk) 19:55, 29 August 2025 (UTC)[reply]
    I'd say it depends on how the name is used in reliable sources. When I go to https://www.spaceship.com/ it looks like they call the website sometimes "spaceship" and sometimes "Spaceship". A quick web search leads to mixed use. I'd go with "Spaceship" as that seems to be the name of the website, which is a proper noun and we generally capitalize proper nouns. Oddly we italicize website names in references, but I don't recall ever seeing them italicized in text. SchreiberBike | ⌨  21:22, 29 August 2025 (UTC)[reply]
    However, "Namecheap launched Spaceship" sounds confusingly similar to the generic "Namecheap launched a spaceship" (and lowercasing it as "Namecheap launched spaceship" would be even worse). Gawaon (talk) 07:30, 30 August 2025 (UTC)[reply]
    In reaction to your remark, I've just added further alternatives with quotes to the list above. In 1973, Parton released Jolene would imply that Jolene was some poor soul held in captivity by Parton for some period ending in 1973, but we actually write it In 1973, Parton released "Jolene" so we have no problem of comprehension. The only alternative I can think of to setting off the website name through some typographical convention is to be explicit, Namecheap launched the website Spaceship. (In that case, MOS:TMLOWER—registered or not, a website's name is its trademark, it seems to me—would be why we wouldn't write Namecheap launched the website spaceship even if the word were exclusively uncapitalized on the website.)
    Instituting special formatting to designate website names will never be as easy or as understandable as saying "a website named ...". We aren't short on space in this electronic encyclopedia. We can write well and be clear. Problem solved. If a website is named say "Wikipedia" there's no point in calling it "wikipedia.org", unless you need to talk about the website's domain name. SchreiberBike | ⌨  18:49, 30 August 2025 (UTC)[reply]
    There are websites that have no title other than their domain name (and, in some cases, whose parent companies have no other name): hotels.com, booking.com, cars.com, answers.com, the old pets.com, and crypto.com (which is even the namesake of a stadium, Crypto.com Arena in Los Angeles). Largoplazo (talk) 20:37, 30 August 2025 (UTC)[reply]
    Right. If the website name is the same as the domain name, you could say "a website named pets.com". That would be clear. SchreiberBike | ⌨ 
    The existing rules are at MOS:ITALICWEBSITE but only mention names of web sites that are not the domain name. I agree italics is not necessary in these cases, and capitalization is optional but seems to be the dominant practice. I've added a note and cross-reference. For citations, the citation style being followed determines the formatting, and that varies from article to article. -- Beland (talk) 05:45, 6 September 2025 (UTC)[reply]
    I would use Space.com and treat it as the name of the website in this case. There are some subtle distinctions one might make like "Facebook can be accessed via the app or online at facebook.com" where one might make a case for lowercase. --MYCETEAE 🍄‍🟫—talk 23:10, 10 September 2025 (UTC)[reply]

    MOSPOSS, Bayes' Theorem, and Olbers' Paradox

    [edit]

    So there's a clear issue which has been simmering around various pages for a while but doesn't seem to have been explicitly taken on here. Many editors over the years have been debating the proper apostrophe format for possessive S's. MOS:POSS explicitly mentions that, when not difficult to pronounce, singular possessives should always end with an ['s] unless followed by "sake", but deference should be given to official names.

    However, the MOS is frustratingly silent on how this intersects with WP:COMMONNAME in cases where there is no 'official' body dictating a phrase but popular use goes against POSS principles. This may seem like a trivial gripe, but it has resulted in several different move request discussions over the years:

    It's clear that many independent users through time have, with the intent of following MOS:POSS guidelines, tried to move articles of non-official common names to be consistent, in almost every case being overruled by others on COMMONNAME grounds, but still these move requests persist (and sometimes, as in Shays'[s] case, succeed) before the discussion crops up anew once again, with another editor reading MOS:POSS and assuming something needs to be fixed.

    So: can we please add a provision in MOS:POSS that, when a possessive name violating this is commonly used enough that it fulfills WP:COMMONNAME, that this overrules the policy and should be used as such in articles? I might propose something like this:

    For the possessive of singular nouns, including proper names and words ending in s, add 's (my daughter's achievement, my niece's wedding, Cortez's men, the boss's office, Illinois's largest employer, the US's partners, Descartes's philosophy, Verreaux's eagle). Exception: abstract nouns ending with an /s/ sound when followed by sake (for goodness' sake, for his conscience' sake). If a name ending in s or z would be difficult to pronounce with 's added (Jesus's teachings), consider rewording (the teachings of Jesus).

    If the established common usage of a term or phrase omits an 's (Olbers' Paradox, Achilles' Heel, Shays' Rebellion), use the WP:COMMONNAME instead.

    I'm also not inherently opposed to MOS taking precedence, and us reworking each of these titles to include an 's contrary to popular usage, but right now the actual usage in articles is contradictory and the MOS is silent either way. Some sort of official comment one way or the other appears necessary. Arguably adding it could be considered fairly uncontroversial, but considering the number of high-participation move requests over the years (linked above) I wanted to get a reality check before making such a far-reaching addition. exoplanetaryscience (talk) 07:20, 30 August 2025 (UTC)[reply]

    As another example, see the discussion on Archimedes: Talk:Archimedes#MOS:'S. This would affect quite a few articles relating to classics, e.g. Xerxes' pontoon bridges. Thunkii (talk) 07:29, 30 August 2025 (UTC)[reply]
    I don’t think that is another example of what is raised here. It’s just a phrase referring to some pontoon bridges belonging to Xerxes, for which Xerxes’s bridges works fine. MapReader (talk) 06:55, 5 September 2025 (UTC)[reply]
    We can add that, although it should be technically unnecessary, since policy WP:COMMONNAME trumps guideline MOS:POSS anyway. —Kusma (talk) 10:34, 30 August 2025 (UTC)[reply]

    In addition to the evidence above, I'll add some experience as someone who's been around here for a while, if arguing about these things floats your boat, go on and do it, but at this point making Wikipedia consistent isn't going to happen unless we change the requirement for consensus. Consensus won't happen. We could appoint a dictator to make a decision. We could do the equivalent of flipping a coin. There are probably other ways to solve the problem. Most likely is a lot of words and effort and no change. As I said, feel free, but I suggest letting it go. SchreiberBike | ⌨  18:42, 30 August 2025 (UTC)[reply]

    • I have tweaked the bold edit made while this discussion is ongoing, to pick up the point I made above by including a reference to proper names, for which there might be an established format. Bayes’ Theorem is one such; Bayes’s underpants is not. MapReader (talk) 07:02, 5 September 2025 (UTC)[reply]
      I apologise for not getting in earlier, but I've reverted the bold edit for now -- there are a couple of problems here. First and most importantly, whether a given publication writes Bayes' theorem or Bayes's theorem is going to be a matter of that publication's style guide on apostrophes in general, not on that individual item, so we can't meaningfully extrapolate from how many publications plump for one or the other into specific guidance on specific terms. Secondly, it is established at various points in the MoS that differences of punctuation do not make for a different lexeme (see in particular MOS:CONFORM), and so WP:COMMONNAME can't be applied here, since the MoS treats Bayes's theorem and Bayes' theorem as the same name, differently rendered. Finally, the guidance as written would mandate some pretty bizarre things -- such as a sentence like Bayes' theorem is Bayes's greatest contribution to mathematics. Just looking at this discussion, I can't see any real feeling, other than from the proposer, that this change is/was a good idea. UndercoverClassicist T·C 07:19, 5 September 2025 (UTC)[reply]
      That’s fair enough; I too felt the edit was very hasty, based on the minimum of real consideration. MapReader (talk) 19:17, 5 September 2025 (UTC)[reply]
      Knowingly tangential to the main discussion, but out of curiosity, what would the correct course of action have been here? My thought process was that it's been around a week since I raised the question, so given how prominent talk:MOS is, anyone who was liable to see it and have a strong opinion would have likely commented by now, and the lukewarm response signified that it was a pretty non-contentious edit. Should I have waited longer? Or would the lack of responses suggest not to make the edit by default? exoplanetaryscience (talk) 23:59, 6 September 2025 (UTC)[reply]
      Personally, I don't think any harm was done -- consensus can be tacit, so "nobody objects" can be taken as evidence of it; bold edits are generally encouraged, on the understanding that they can be reverted and discussed. Equally, the MoS is a contentious topic, so it may be considered wise to err on the side of caution and wait for positive consensus before making substantive changes -- after all, the group of editors who watch WT:MOS is much smaller than the group affected by what the MoS says. From another point of view, three editors had chipped in with (at best) lukewarm views of the proposal, which might have been taken as an indication to take things slowly. But we had a bold edit, it was reverted, we're discussing: that's the process working as it should. UndercoverClassicist T·C 06:34, 7 September 2025 (UTC)[reply]
      If it were a regular article, it would have been better to wait for more discussion, but with the MoS, it’s a heavily watched page, so it wasn’t unreasonable to go ahead and try a BOLD on the offchance that others would find it non-contentious. Equally, very shortly after the edit was made, I amended it, and another editor reverted it, both of which are reasonable actions and indicate that it doesn’t have consensus. I don’t see the edit as non-contentious, since it changes a clear and unambiguous guide to always use "’s" into one that requires debate and consensus over “established common usage” in potentially a very large number of instances. It’s also worth noting that - in a week now - you don’t really have any comments from other editors fully supporting it? MapReader (talk) 06:49, 7 September 2025 (UTC)[reply]
      And incidentally, just for interest, I went looking at some sources regarding Mr Bayes’s little theorem, and found that usage varies considerably. In particular, the unpunctuated Bayes Theorem format is pretty common, especially in more modern sources including some scientific ones, and Bayes’s Theorem isn’t unheard of, including in some reputable sources such as the Oxford University Press. So even in your core example, there may a debate to be had? I don’t think there’s any doubt that the single apostrophe without an "s" is the classical usage, but it’s certainly not universal usage, leaving WP free to adopt its own style. In written English generally, a hundred years ago the single apostrophe was much more common for the possessive of singular words ending in "s", whereas modern usage trends increasingly towards "’s", which as I recall was a consideration in WP adopting it for the MoS, with the added advantage that it reduces ambiguity (because with ‘ you don’t know whether the word is singular or plural, except from context) and editor conflicts. MapReader (talk) 07:12, 7 September 2025 (UTC)[reply]
      Your last point doesn't apply to names, though. Nobody would read Bayes' and wonder whether it were singular or plural. And on the rare occasions when a plural possessive would be used on a name, the definite article will almost always precede the name (e.g. the Kennedies' assassins). 76.20.114.184 (talk) 18:33, 7 September 2025 (UTC)[reply]
      You have obviously never heard of the Baye brothers, Jim and Bob, who came up with a very interesting theorem… MapReader (talk) 22:01, 7 September 2025 (UTC)[reply]

    Redirects to/from ampersands: discussion at WT:WikiProject Redirect

    [edit]

     You are invited to join the discussion at Wikipedia talk:WikiProject Redirect § Redirects to/from ampersands. Pineapple Storage (talk) 13:34, 2 September 2025 (UTC)[reply]

    Discussion that might be of interest to the watchers of this page

    [edit]

    Find the discussion here: Wikipedia:Village pump (idea lab)#Firearm brands and models in articles about mass shooting events. Thank you. TurboSuperA+[talk] 04:46, 3 September 2025 (UTC)[reply]

    Should these sub-sub-headings be removed?

    [edit]

    I think the section Wikipedia:Manual of Style § Possessives is full of unnecessary headings. It has only 4 paragraphs of text, yet there are 4 sections, one for each paragraph. I think this makes reading the section unnecessarily harder and interrupts it. This is what it would look like with the headings removed.

    Courtesy pinging @Moxy to hear their opinion. FaviFake (talk) 13:23, 3 September 2025 (UTC)[reply]

    I prefer the version with the subheadings. The MOS is typically quickly scanned rather than read through. Frequent subheadings makes it easier to identify the correct content in the TOC and find it in the body.--Trystan (talk) 13:36, 3 September 2025 (UTC)[reply]
    What we are looking for in our protocol pages is ease of locating things over aesthetics. We really need you to slow down pls review Wikipedia:Policy writing is hard#Things to consider. Moxy🍁 14:48, 3 September 2025 (UTC)[reply]
    Thanks. It seems the essay you linked to is not related to this situation; it talks about how one should write the actual sentences that form policies, not how they should be formatted. Did you mean to link to a different page? FaviFake (talk) 14:56, 3 September 2025 (UTC)[reply]
    I think the current layout makes it easy to scan, and to link to a specific section of the MOS from a discussion. – Jonesey95 (talk) 15:18, 3 September 2025 (UTC)[reply]
    I also think the headings are fine. To me, they don't make reading the section harder. —Alalch E. 15:26, 3 September 2025 (UTC)[reply]
    If there is a shortcut, then there should be a heading to match. The MOS is a dense read with lots of information and so deserves a lot of headings to facilitate navigation and inbound links. Also, I would second the view that you need to slow down. Editing the global MOS for Wikipedia needs much more care and contemplation. If you have changes to offer, then good, but start by proposing them in this talk page. This gives other editors the chance to comment, so any change can be polished. — GhostInTheMachine talk to me 16:03, 3 September 2025 (UTC)[reply]
    The subheadings are fine here. This is a style guide, not article content (where they would likely be excessive). Gawaon (talk) 20:42, 3 September 2025 (UTC)[reply]
    Keep the subheadings. They are more beneficial than not. The MOS is still easy to read as a whole and the subheadings make it easy to quickly locate specific guidance. --MYCETEAE 🍄‍🟫—talk 21:44, 4 September 2025 (UTC)[reply]

    Section on Scrolling lists and a link to a non-consensus help page

    [edit]

    I removed a link that went to a help page that was more negative than MOS on scrolling lists. That help page is not even essay status or consensus and could leave editors thinking it was part of MOS from the link. I did me. The link in this article should directly point out it is going to a non-consensus help page or it should be delinked so other editors do not make the same mistake I did. Hence this added discussion. Fyunck(click) (talk) 04:07, 8 September 2025 (UTC)[reply]

    Links to how-to pages from policies and guidelines are often fine, and are not generally treated as problematic. How-to pages are described in WP:HOWTOPAGES, and it isn't reasonable to say that links to these pages should be removed because they can be confused with guidelines. The linked page is not an opinionated essay and is not supposed to be interpreted as "stricter" than the MoS. It is only linked to explain to a reader who does not know what "scrolling list" refers to what these words mean. If you think the how-to page overreaches in some way and does not represent the actual practice, you should edit it. In any case it is not normative for the MoS.—Alalch E. 06:36, 8 September 2025 (UTC)[reply]
    Actually we have seen that some of these "how-to" pages are a problem with little vetting. To have a page like MOS link to one of these seems wrong. Perhaps if these pages said in giant bold red letters that it has no real meaning per MOS, then editors wouldn't be bamboozled into thinking they were special. It's like finding articles with no sources. And it is completely reasonable to say that links to these pages should be removed because they can be confused with guidelines, if it's true. I have seen it happen multiple times and it's time someone said something about it. We get folks telling us to do things a particular way based on these non-vetted musings and we have to explain to them that these pages are not MOS or policy. You might as well have Wikipedia MOS link to an outside blog... and that isn't right either. Browsers, phones, and screen readers have gotten smarter and smarter, and with ai probably even more so. Our MOS has been brought together with lots of hard work and elbow grease... kudos to those that helped build it. But linking our MOS to some of these non-consensus pages without spelling that fact out prior to the link, seems to weaken what others have created by consensus. Fyunck(click) (talk) 07:29, 8 September 2025 (UTC)[reply]
    I would agree in principle with removing it if it were an opinionated essay, and I've probably removed a few of those myself from PAGs, but this is a how-to page that intends to be objective and is not supposed to contradict the MoS. It primarily serves to explain what the thing is, and if it is seemingly "stricter" than the MoS and reads as if the MoS restricts the use of such lists more than it actually does, then it's a factual error in the how-to that needs to be fixed. My idea of how to proceed is you making your desired changes to Help:Scrolling list. I recommend that you leave the link to the page here alone; it has been here since 2012 (Special:Diff/526474166), and it is a useful link that explains what "scrolling list" refers to. —Alalch E. 07:49, 8 September 2025 (UTC)[reply]
    I will leave it alone here now because you reverted me. But we have a big difference of opinion on MOS and the way it should work. You say fix the other page and I say MOS should not be linked to such a page unless it is fixed or unless MOS explains what the page is that it is linked to. I don't think we will agree on protocol and what we feel is best for our readers, which is always my first priority. It is what it is I guess. Cheers. Fyunck(click) (talk) 08:30, 8 September 2025 (UTC)[reply]

    Use of main article template

    [edit]

    Can you please tell me which is preferred?

    1. A section heading followed by a {{main}} template, followed by prose that might wikilink to that main article.
    2. A section heading containing prose with a wikilink to the main article.

    My feeling is that a {{main}} template follows Wikipedia style and is easiest for a reader who may wish more information. The opposing view thinks {{main}} is redundant when the prose wikilinks to more. I looked at a featured article, and found an example. Earth §Size and shape, first has a {{main}} and then in prose, a wikilink to the same article. I chose #1 because it avoids us sending readers on a scavenger hunt for more information. MOS:LAYOUT, WP:SUMMARYHATNOTE, and WP:HATNOTERULES all appear to agree. Thank you. -SusanLesch (talk) 19:35, 10 September 2025 (UTC)[reply]

    I prefer #1 and I agree with your reasoning. Hatnotes are occasionally overkill but they serve a distinct purpose and should be used accordingly. {{main}} signals something important to readers that is not immediately obvious from the mere inclusion of wikilinks in prose. --MYCETEAE 🍄‍🟫—talk 22:00, 10 September 2025 (UTC)[reply]
    I believe the use of {{main}} means that we don't wikilink that article in prose. I prefer that over no {{main}} and linking it in prose. – Muboshgu (talk) 17:00, 11 September 2025 (UTC)[reply]
    When, instead of having the other article, its content could logically have been included directly in the section, and the reason for having it as a freestanding article instead is a combination of size and level of detail, then the other article is the main article for the section, and it can be seen as a direct offshoot of the article containing the section. Indeed, it may have been created by splitting the longer, more detailed text from the higher-level article. In that case the {{main}} template is virtually necessary, the other article is a component of the one containing the section. This is irrespective of whether the section also links to it. Or maybe it means the link is somewhat contrived and shouldn't be there. Largoplazo (talk) 15:26, 17 September 2025 (UTC)[reply]

    Myceteae, the inline link was suggested in two rounds of discussions on Talk:Donald Trump involving SusanLesch and Muboshgu. The links were then discussed in Talk:Donald_Trump#Follow-up to Trump-Epstein which would have been the place to discuss a hatnote. As for the section itself, it's not a summary of the Relationship of Donald Trump and Jeffrey Epstein, so IMO WP:SUMMARYHATNOTE doesn't apply. The two sentences are the lowest common denominator the editors could agree upon after more than a month of discussion and two rounds of !voting. Option 2 (section heading with a wikilink) wasn't proposed by anyone. Editors agreed on an inline link in the first sentence of the text. Space4TCatHerder🖖 18:05, 11 September 2025 (UTC)[reply]

    Thanks for the background. As always, context matters for these discussions and there will be reasonable editorial decisions to take a different approach based on the article. The example given in the discussion here was related to Earth#Size and shape. The usage there makes sense and any concern about redundancy is overridden by the fact that this is likely very useful to readers, based on the use of piped links and numerous wikilinks in hatnotes and in the section body. I haven't read through the discussion you linked, but it appears the situation was sufficiently different and was subject to a lengthy discussion, so a different approach is appropriate there based on the particulars. --MYCETEAE 🍄‍🟫—talk 18:41, 11 September 2025 (UTC)[reply]
    Space4TCatHerder, thank you for accepting my invitation to post here. Your version of facts are false and stymied the discussion.
    Myceteae, thank you for your replies. Multiple {{main}}s can cause a log jam and be overkill as you say. Nevertheless, may I remind everyone.
    A controversial context is exactly where editors should follow Wikipedia precedent and execute within the MOS shared wisdom we are lucky to have. Wikipedia has hundreds, maybe thousands, of contentious articles on wars and world conflicts. Each one observes MOS—otherwise they'd have splintered into diverging encyclopedias. We should treat this subject like every other to help our readers. -SusanLesch (talk) 19:17, 12 September 2025 (UTC)[reply]

    Will anyone stand up for MOS?

    "Editors should structure articles with consistent, reader-friendly layouts and formatting"

    . -SusanLesch (talk) 14:38, 17 September 2025 (UTC)[reply]

    [edit]

    Since it's common for editors to make the mistake of putting 's to pluralize abbreviations, I think there should be a shortcut link to § Plural forms for the sake of edit summaries correcting such errors.

    What should the link's name be? I got a few ideas: either MOS:SNOT'S, MOS:NO', MOS:NOAPOSTROPHE, or MOS:PLURALFORM. CheckNineEight (talk) 14:17, 12 September 2025 (UTC)[reply]

    Superseded by topic below Should MOS:COLOUR permit multiple colors in some situations?

    [edit]

    Please see discussion about MOS:COLOUR at Wikipedia_talk:WikiProject_Military_history#Using_colors_in_military_maps_to_convey_information:_MOS:COLOUR_issues? Noleander (talk) 22:24, 15 September 2025 (UTC)[reply]

    This discussion in WikiProject Military History has been moved to the MOS Talk page, in the topic immediately below. Noleander (talk) 13:58, 16 September 2025 (UTC)[reply]

    Should MOS:COLOR be updated to clarify if and when color hues may be used to convey information in maps?

    [edit]
    Red and blue distinguishing armies
    Red, blue and green distinguishing voyages
    Blue, green, grey distinguishing range of an animal
    Many colors distinguishing railways

    I have seen a lot of confusion in WP:Featured Articles and WP:Good Articles about how to apply the MOS:COLOR accessibility guideline to maps. That guideline says:

    "When using color, editors should keep accessibility for users with low vision impairments and color blindness in mind:... Color should not be used as the sole visual means of conveying information, or for distinguishing elements such as links, templates, or table rows. Always provide an alternative method [such as icons, symbols, etc]" [emphasis added].

    On the right are four maps that illustrate the confusion I observed in FA and GA. The confusion revolves around the question: When does MOS:COLOR permit the use of hue (red/blue/green/yellow/orange) as the sole means of conveying information in maps?

    Notably, everyone agrees that color lightness  – e.g dark blue/ medium blue/ light blue/white – is consistent with MOS:COLOR. Likewise, using dotted/dashed/solid line styles is acceptable for conveying information in maps.

    WP:MOS says "New content added to [WP:MOS] should directly address a persistently recurring style issue." I think this color issue is "persistently recurring" because I've seen it four times in the past six weeks, so I suspect that hundreds of editors are facing this issue monthly as they add or review maps within articles. Therefore, keeping MOS:COLOR unchanged is, in my opinion, not wise.

    Personally, I have no opinion on how to apply MOS:COLOR to maps. I raise the question here only because it is clear that many editors are confused about it.

    Therefore, I pose the question:

    Should MOS:COLOR be updated to clarify if and when color hues (such as red/blue/green/yellow/orange) may be used to convey information in maps?

    Noleander (talk) 13:56, 16 September 2025 (UTC)[reply]

    Two relatvely recent discussions/comments at FAC are here (re. a graph) and in this review (re. a diagram). —Fortuna, imperatrix 15:12, 16 September 2025 (UTC)[reply]
    • There's MOS:ACCESS/MOS:COLORS (Note: to be distinguished from MOS:COLOR, mentioned above) to consider, perhaps, per WP:FA?#1F. —Fortuna, imperatrix 14:46, 16 September 2025 (UTC)[reply]
    • Only somewhat related to the original question, but I do find it frustrating that while MOS:ALT is part of the MOS, folks at WP:FAC seem to interpret It follows the style guidelines as including a carve out making MOS:ALT optional. In general, I think we give accessibility issues less emphasis than we should; we should not just be the encyclopedia anybody can edit, we should also be the encyclopedia anybody can read. And I'm still doing a bit of a slow burn that when I recently pointed out to somebody that their user page was formatted with so much fancy styling that it difficult for me to read, they blew me off with (paraphrasing) "I can read it just fine, you must have vision problems". To be sure, my vision isn't as sharp as it was 10 (or 20, 30, 40, 50, or even 60) years ago, but you'll all get here too if you survive long enough. And while I still see well enough to qualify for a driver's license, I guess I'm quickly reaching the point where I can't read an encyclopedia. RoySmith (talk) 15:18, 16 September 2025 (UTC)[reply]
      folks at WP:FAC seem to interpret It follows the style guidelines as including a carve out making MOS:ALT optional. I share your frustration here. The usual point that gets made is that we don't have good guidance on what makes good alt text, and external sources of wisdom often disagree -- but then I still think there ought to be general agreement that you can't follow MOS:ALT without at least having a go at writing some alt text, and therefore that an article with no alt text at all can't pass FAC. But that may be a discussion for another venue. UndercoverClassicist T·C 15:24, 16 September 2025 (UTC)[reply]
      I'd appreciate a ping if this gets raised elsewhere. I hadn't encountered this issue before, and it boggles the mind. Firefangledfeathers (talk / contribs) 15:25, 16 September 2025 (UTC)[reply]
      +1 Absolutely fundamental. We're effectively excluding a chunk of our readers from "seeing" the same article as others. —Fortuna, imperatrix 15:42, 16 September 2025 (UTC)[reply]
      MOS:ALT sucks and should be replaced by a useful guideline about writing helpful ALT text. (I raised this at Wikipedia talk:Manual of Style/Accessibility/Alternative text for images#Rewrite but we didn't get anywhere). —Kusma (talk) 16:08, 16 September 2025 (UTC)[reply]
      This is somewhat off topic here, but what would be a good practice for the alt text in Lexell's theorem? I just set the alt for every diagram to alt=Refer to adjacent text on the recommendation of MOS:ALT because I can't think up a way of describing these images in a way that is independently valuable, but would it be better to leave alt out or blank? In general, a diagram of a geometric proof conveys roughly the same formal information as the text of the proof, but the text is often pretty hard to follow without the image as an overview and mental anchor. I frankly don't know how to make an article like this accessible to someone who can't see (especially since most screen readers mangle mathematical formulas, which are the primary non-image content; I wouldn't necessarily mind providing explicit alt text for every formula, but we don't currently have a way of accomplishing that). –jacobolus (t) 19:54, 16 September 2025 (UTC)[reply]
    • I am colour blind and I often find maps useless because I cannot tell apart the colours. I do not have a problem if they are poster colours, that is pure red, blue, etc, with no shade of another colour. I do not know how typical I am. Dudley Miles (talk) 16:57, 16 September 2025 (UTC)[reply]
    • user:GeogSage said at WT:MIL § Using colors ... "The average person making maps for Wikipedia is not qualified." And pointed to the work of Cynthia Brewer, especially her ColorBrewer tool, as a quick shortcut to getting a better result than whatever a non-expert might naïvely first try. Quoting Brewer, GeogSage pointed out that "Hue variations are used to allow more contrast between categories, but lightness should still be the primary differentiating characteristic."
      Brewer (and GeogSage) is right here, but that's not the end of the story.
      Lightness contrast is completely essential for making anything that will be placed directly side-by-side clearly distinguishable, whether in a photograph, user interface, diagram, map, or whatever. If you place two large filled objects with very limited lightness contrast next to each-other, the result will be that they blend together; if the two are of similar color, then they will tend to just look like a uniform blob, and if they are both highly colorful but of substantially different hue they will be extremely distracting and "clashing", while also simultaneously being hard to distinguish. It is completely essential that any colors representing different data values be split by lightness (say in a choropleth map or the like).
      But in some contexts the lightness contrast between objects of one type and objects of some other type is more important than contrast between various objects of the same type, so we sometimes have a relatively limited range of contrast to work with for, e.g., several fill colors that need to all contrast with foreground text, or for several line colors or text colors that need to contrast with the same fill colors.
      Just like writing decisions, color design takes some taste, knowledge, and careful trade-offs, based on the content, output medium, intended viewership and their expectations, etc. It's hard to boil these decisions down to any kind of ironclad rules, but only general guidelines, like: Try to make the most contrast you can, while also being non-distracting in the context of a text article people are trying to read (this limits how colorful, busy, or high-contrast any image should be, and argues against very active animation), keeping labels and symbols legible, trying to support readers with non-standard color vision as best as possible, etc.
      I think more useful than trying to force any kind of universal requirements (or e.g. blocking articles from being featured if the images have issues) would be to have a place where image authors could consult some more experienced Wikipedians who have thought about these aspects of design and offer advice, or maybe some kind of team effort to produce some examples (perhaps drawn from existing images on Wikipedia/Commons) showing specific problems and possible ways of improving them. –jacobolus (t) 06:58, 16 September 2025 (UTC)[reply]
      That is an excellent suggestion: "... have a place where image authors could consult some more experienced Wikipedians who have thought about these aspects of design and offer advice" If that were implemented, those advisors would inevitably create a FAQ that summarizes the common issues that arise. Such an FAQ would be required, for example, when the advisors are busy IRL, or retire from WP, etc. Can you sketch-out the advice that hypothetical FAQ would provide to editors regarding the use of color in maps (in the context of accessibility)? Noleander (talk) 18:46, 16 September 2025 (UTC)[reply]
      Is this the kind of request that could be added to the Wikipedia:Graphics Lab/Map workshop? Sariel Xilo (talk) 19:26, 16 September 2025 (UTC)[reply]
      As an example and to follow up on my statement quoted above, I would like to use the opportunity to point out the category on Wikicommons Category:Choropleth maps using absolute numbers.
      I can list several sources if anyone is interested, but this entire category is filled with maps that the literature would consider misleading. Attempts to outright delete them have been blocked, and instead we've been told to just replace them as we go to avoid disrupting the project. Bad maps are worse then no maps, there is no defense for the maps in this category (I've done the literature review). This is ONE example of how bad the problem is with Wikipedia's cartography. Many maps have no sources, including many of the boundaries on country pages. The projections are often not documented, but when they are are often inappropriate to the map type. Color is a huge issue, but we have a lot of other glaring problems that need to be addressed. GeogSage (⚔Chat?⚔) 20:03, 16 September 2025 (UTC)[reply]
      Next thing you will be saying that bad images are worse than no images. Hawkeye7 (discuss) 21:21, 16 September 2025 (UTC)[reply]
      Maps are not the same thing as an image. Maps are models of reality, bad maps should be viewed the same as an inaccurate or misleading piece of text. All maps lie, however there are conventions and norms that help us to ensure they are white lies. If a bar chart or line graph is improperly used, it is worse then no graph at all as it might paint an inaccurate view of the underlying data. GeogSage (⚔Chat?⚔) 03:52, 17 September 2025 (UTC)[reply]
      To be fair to the Wikipedian mapmakers, most other aspects of Wikipedia have the same kinds of problems: poor sources or no sources, poor organization, inaccurate factual claims, logical fallacies, misleading interpretations, claims not reflective of the scholarly mainstream consensus, poor writing, bad API design of templates, etc. etc. But despite all its problems, it remains a marvel, and generally better than I would hope it to be when just imagining the concept. –jacobolus (t) 04:43, 17 September 2025 (UTC)[reply]
      The issue is that if I point out an article is making inaccurate claim, or if material is poorly sourced, I can remove it without much of a fuss. When it comes to maps, people demand that they must be replaced with something better, rather then deleted. That entire category I linked is filled with maps that are worse then no maps at all, not only do they present data with inappropriate symbolization leading to misleading representations, they set an example that other amateur cartographers follow. Bad maps out number good maps online, and we are part of the problem. GeogSage (⚔Chat?⚔) 04:56, 17 September 2025 (UTC)[reply]
      Bringing this thread back to the topic of this section :-) .... I'm perceiving that there is no appetite to make the MOS:COLOR hue guidelines mandatory for maps. As I read the comments above, there appears to be a strong feeling that some maps can and should use hues to convey information. Of course, there are many, many variables, and every map must be considered on a case-by-case basis. But I don't see any editor that says "maps must use colors in a in a manner that permits visually impaired readers to clearly grasp the data". So perhaps we could add text or a footnote to MOS:COLOR that clarifies that the color guidelines are recommended as an aspirational goal for maps; but may be waived if there is a valid reason. That should put an end to the frequent confusion that has been observed in FA nominations over the past decade about how MOS:COLOR applies to maps Noleander (talk) 19:34, 17 September 2025 (UTC)[reply]
      "Guidelines may be waived" also seems like the wrong framing (and our current guidelines are pretty weak). More like: people should try to do the best they can within available constraints, and most of the criteria involved should not be interpreted in a black-and-white fashion.
      If there are obvious improvements that can be made to particular maps folks should feel free to make those or offer specific critiques. In many cases a bad map is still better than no map, and in my opinion a map being ugly or amateur isn't, by itself, sufficient reason to throw it out.
      But sometimes a bad map is worse than nothing: If maps have more serious issues violating core Wikipedia policies (e.g. are inaccurate, non-neutral, not supported by reliable sources, link to sources but mischaracterize their content, etc. they should probably be removed from articles.
      There are many resources available (including free resources online) for learning about graphical design, data visualization, cartography, human vision, etc. Wikipedia's articles about these topics are often mediocre, so if someone wants to work on this topic, such articles are an obvious target for incremental (or radical) improvements. If anyone feels like making more specific advice or guidelines for Wikipedia maps per se, that would also be a good project. –jacobolus (t) 20:48, 17 September 2025 (UTC)[reply]
      Thanks, that text you suggest is useful: "People should try to do the best they can within available constraints, and most of the criteria involved should not be interpreted in a black-and-white fashion." That looks like it would address the primary goal of this topic (helping to minimize confusion about usage of colors in maps); and it appears to be consistent with the comments other editors have made above. What do you think of this wording: "Applying these color guidelines to maps is recommended, but they should be balanced with other cartographic goals and constraints. [This would be added to MOS:COLOR as either text or a footnote]. Does that sound sensible? Noleander (talk) 22:33, 17 September 2025 (UTC)[reply]
      a bad map is still better than no map I can not stress enough how much I disagree with this statement, although my definition of "bad map" might be a bit different then yours, as being ugly or amateur does not inherently make a map "bad." Bad maps break cartographic conventions that misrepresent the underlying information to the user, and the situation on Wikipedia is REALLY bad. Another example, outside color, can be seen in our locator maps, like the one of South Sudan I'm including.
      South Sudan (orthographic projection) highlighted
      Try digging into the sources for the boundary files on Wikimedia. I've found that most of the time, it is completely impossible to find out where a user is getting these from, and several list the source as "Own work," or that they are derivatives of other maps in the commons. This is the exact same thing as original research, and would not be tolerated anywhere else on the project.
      In terms of improving the state of cartography pages on Wikipedia, I can not agree more. My general mission on Wikipedia is to improve the geography literature where ever possible; I've nominated several for Wikipedia:Articles for improvement/Nominations, and Cartography is high on my list. I've done some work on Computer cartography, GIS, Distributed GIS, and created pages for Web GIS and Internet GIS, and want to take on the page for Map, but it really needs a lot of editors eyes on the problem. If you're interested in helping, I'd personally appreciate others involvement. My comments here have mostly focused on the broader situation involving maps as a whole because I want attention called to the problem, it is difficult to address the finer points of the MOS like color choice when we have so many maps that aren't following basic conventions like proper projection, thematic map type, and citing sources. Some fun pages to add to the list you have are Cartographic design, Cartographic generalization, and Map communication model. GeogSage (⚔Chat?⚔) 00:26, 18 September 2025 (UTC)[reply]
      Your example map of south sudan is a very helpful basic locator, and still would be with moderately inaccurate borders for any reader who is not trying to carefully parse every pixel of the border. It is in my opinion clearly better than having no map at all. –jacobolus (t) 00:49, 18 September 2025 (UTC)[reply]
      If I write a sentence that says, "North and South Korea are divided 39th Parallel," it would be immediately deleted. If I make a map that shows North and South Korea divided on the 39th Parallel, we would keep it until someone made something better. This is stuff wars are fought over, and different groups have different sets of boundaries. If you do not provide the source for the border you are using, it is as useful as an uncited textual claim describing the the land holdings of Genovia in the article History of Europe. We are endorsing disputed territorial claims in Wikivoice, and people with political agendas are actively exploiting this to push versions of the world map that fits their world view. GeogSage (⚔Chat?⚔) 03:48, 18 September 2025 (UTC)[reply]
      You are far over-interpreting the map here, and fundamentally misunderstanding/mischaracterizing its purpose. There are a variety of detailed maps later in the article which show various more precise claims about territory and explain more about their sources (I have no idea if those are accurate; I didn't investigate), but the basic country locator thumbnail is showing the globe-scale context, and does not make any of the precise border details even visible; we're talking about differences of 1 or 2 pixels on someone's screen, which are not even going to be distinguishable unless someone is extremely knowledgeable and peeping the thumbnail with a magnifying glass.
      A better analogy would be the sentences that say: "South Korea [...] is a country in East Asia. It constitutes the southern half of the Korean Peninsula and borders North Korea along the Korean Demilitarized Zone, with the Yellow Sea to the west and the Sea of Japan to the east." While it's true that this description is not incredibly precise, it's also still very useful to someone who is trying to figure out where in the world the place is that we are talking about. If the South Korea locator map accidentally drew the border in the wrong place by 50 kilometers (2 pixels at the scale of the locator map in South Korea), it would be a regrettable error and worth fixing, but it would basically not matter in any practical sense. –jacobolus (t) 05:02, 18 September 2025 (UTC)[reply]
      You are far over-interpreting the map here, and fundamentally misunderstanding/mischaracterizing its purpose. From the perspective of the outside literature involving cartography, I do not believe I am. Look at the map of South Sudan again, in that there are boundaries showing Western Sahara as a separate country without any indication of border disputes. This IS a political stance, and without a source it is Wikipedia taking a stance on that issue, without a citation, the Wiki IS the source. Different maps have different levels of cartographic generalization, which is why it is important to document the source. It will tell you the level of accuracy you can expect from the boundary files. Another way to look at the problem, if I drew the border of Egypt 50 kilometers in the wrong place, I could move ALL of Gaza into Egypt. GeogSage (⚔Chat?⚔) 16:56, 18 September 2025 (UTC)[reply]
      At the scale of this map, Gaza isn't even visible at all. It's like 1 pixel or something. –jacobolus (t) 17:21, 18 September 2025 (UTC)[reply]
      But Western Sahara and Morocco are visible. If a set of borders are not cited, it is original research, plagiarism, or in some cases a combination of both. Borders should be viewed as an exact quote, the South Sudan example, like many, omits the person or organization that is responsible for the base map. I suspect there are several maps in the commons that are using copyrighted material inappropriately. GeogSage (⚔Chat?⚔) 20:24, 18 September 2025 (UTC)[reply]
      I don't see what Western Sahara has to do with anything. In any case, map borders are not protected creative works, and copying them is not "plagiarism". The person responsible for the base map (User:Martin23230, who originated the file File:Africa (orthographic projection).svg) was a Wikipedian in 2009. They didn't cite a source and haven't made any contributions here since 2015.
      If someone wants to make (and keep updated) better base maps with better explained sources, that would be fine, and I'm sure widely appreciated. I'd recommend using https://www.naturalearthdata.com/ as a source. –jacobolus (t) 20:33, 18 September 2025 (UTC)[reply]
      Western Sahara is a disputed territory and the United States now recognizes Morocco has sovereignty over it, which is disputed by several other countries. The choice to include the line dividing Morocco and Western Sahara is a political stance, which is one reason for requiring citations. As I've said, for some reason, things that would not be acceptable in text, like uncited direct quotes and misinformation, are kept in map form unless someone takes the time to make something better. Per Wikipedia:Verifiability and Wikipedia:No original research, unsourced or user generated map boundaries should be removed, not kept until we find something better. This is a problem across the project. GeogSage (⚔Chat?⚔) 21:04, 18 September 2025 (UTC)[reply]
      That sounds extremely off-topic. This discussion is about colours, not borders. Gawaon (talk) 21:21, 18 September 2025 (UTC)[reply]
      I stated above, Color is a huge issue, but we have a lot of other glaring problems that need to be addressed. The discussion is about cartography on Wikipedia. On Wikipedia, we have a lot of maps that are violating basic cartographic conventions, that are missing sources for their data, that are complete user fabrications, and that are spreading misinformation either intentionally or by the ignorance of the map makers. Discussing color choices is great, but we have absolutely no consistently upheld standards when it comes to maps. Whenever these issues are brought up, it is dismissed. The current discussion is kind of like an HOA discussing how to paint houses that are violating every possible housing regulation imaginable, on fire, and actively spreading that fire to other parts of the city. GeogSage (⚔Chat?⚔) 01:05, 19 September 2025 (UTC)[reply]
      Sorry to get a bit off topic in the above.
      I meant to mention before, but one of the basic problems here is that MOS:COLOR is (a) not a very complete guideline, and (b) doesn't really try to make itself applicable to graphical content like images or maps. Its focus is on content consisting of primarily text, e.g. in lists, tables, image captions, warning banners, or the like. It has a few useful advice tidbits in this context, such as: don't rely on color alone to convey information, make sure that links are obviously clickable, and don't go overboard with colors. But it isn't an adequate replacement for serious resources on this topic (I second GeogSage's recommendation to look up Maureen Stone's advice). It leans heavily on the WCAG 2.0 text contrast recommendations, which are unfortunately fundamentally broken because they are not based on an accurate model of human perception, and as a result provide extremely misleading guidance to designers. (User:Myndex was so dissatisfied with these that he spent years making up a better text contrast metric.) I don't think anyone should be over-interpreting MOS:COLOR; I don't know how it achieved its current form, but I doubt there was too much expert effort put into crafting it. –jacobolus (t) 01:27, 18 September 2025 (UTC)[reply]
    • I think that MOS ACCESS is overly restrictive for maps. It ignores the role of alt text to remedy some visual issues. --Guerillero Parlez Moi 08:27, 17 September 2025 (UTC)[reply]
      Unfortunately writing alt text that could reasonably substitute for a complex map is not easy to do. Nikkimaria (talk) 23:48, 17 September 2025 (UTC)[reply]
      A personal anecdote to illustrate this, in a map use class I took the final assignment was to write a 10 page paper that could communicate everything contained on a topographic map, with the main point was to prove how futile such a task is. Even in a simple map, how do you write all of the spatial relationships between the objects of interest out? GeogSage (⚔Chat?⚔) 00:28, 18 September 2025 (UTC)[reply]
      I'm very doubtful that any wording about maps would work, as it would need to be so hedged about with ifs and buts that it would be worthless.
      I have instead a small suggestion: add a guideline that use of colour on maps and diagrams should be redundant with a different cue, such as text labels, tone (black/grey/white), type of shading, etc. So, you don't just colour an area blue, you label or mark it as well. Chiswick Chap (talk) 15:23, 18 September 2025 (UTC)[reply]
      @Chiswick Chap - Thanks for the suggestion " ... use of colour on maps and diagrams should be redundant with a different cue, such as text labels, tone (black/grey/white), type of shading, etc..." That wording seems to be consistent with the existing guidance in MOS:COLOR, which says "Color should not be used as the sole visual means of conveying information.... Always provide an alternative method—such as an accessible symbol and/or text [or icons or labels, etc]" [emphasis added to "sole"]. Can you clarify: are you suggesting that maps that solely rely on hue (red/blue/green/yellow/orange) are never acceptable? or are merely discouraged? What do you think about notion, mentioned above by other editors, that there are there some situations where hues may be used as the sole means of conveying information in maps (after balancing accessibility goals with other map requirements)? Noleander (talk) 16:19, 18 September 2025 (UTC)[reply]
      My own recommendation would be for authors to try to add a moderate amount of lightness contrast to lines or fills of different colors (not always feasible), try to choose colors such that the map can still be accurately read when various "color-blindness simulator" type filters are applied, depending on the context consider also using alternate graphical differences in addition to color, and ask for help if they are having trouble making a map accessible. –jacobolus (t) 16:33, 18 September 2025 (UTC)[reply]
      A web search turned up this paper about the topic of map colors specifically:
      Jenny, B.; Kelso, N. V. (2007). "Color Design for the Color Vision Impaired" (PDF). Cartographic Perspectives (58): 61–67.
      –jacobolus (t) 16:40, 18 September 2025 (UTC)[reply]
      I don't think, as I implied already, that we can use a "shall" for any of this. Hue alone is unlikely to be any better; given that light, eyesight, and computer monitors all vary, it must be a dead duck as a sole method. In particular, when there is some sort of scale, say from warm brown for mountain tops down to dark blue for ocean depths, distinguishing particular intermediate levels can be quite difficult even if there are contour lines to assist, and is pretty much hopeless on its own. Chiswick Chap (talk) 17:32, 18 September 2025 (UTC)[reply]
      @Chiswick Chap - You wrote: "Hue alone is unlikely to be any better; given that light, eyesight, and computer monitors all vary, it must be a dead duck as a sole method." That sounds like it applies primarily to gradient-type maps such as File:Alicella Projected Distribution.jpg (third map at top of this topic). Referring to the two topmost maps at the top of this topic (File:Vicksburg Campaign April-July 1863.pdf and File:Cook Three Voyages 59.png) those use red and blue to distinguish graphical overlay lines. Thousands of maps in Wikipedia use hues for graphical overlay lines that way. For those sorts of overlay situations, it is not clear to me what you are suggesting. Do you suggest that those kinds of red/blue overlays be prohibited? Permitted? Discouraged? Avoided (but okay as a last resort when balanced against other map-design factors)? I apologize for pestering you, maybe I'm too dense to grasp your intention :-) Noleander (talk) 18:41, 18 September 2025 (UTC)[reply]
      Hue, Saturation, Lightness: I supposed you were suggesting hue as the key variable, and I said it wouldn't do. Maps do indeed often use such scales: I'm saying the scales are very difficult without additional cues. Please stop pinging me. Thanks. Chiswick Chap (talk) 19:00, 18 September 2025 (UTC)[reply]
      It applies to all kinds of map elements including lines, discrete symbols, etc.
      • File:Alicella Projected Distribution.jpg has a badly chosen depth gradient because it does not have enough lightness contrast. The legend explaining the depths is extremely confusing, and the distance scale is distracting, unnecessary, and somewhat misleading for this projection. The map uses a not very effective map projection, the tectonic plate boundaries are confusing, the gray color has way too little contrast with the green/blue, etc. This map is certainly better than nothing, but was made by cartographic amateurs (biologists) and leaves a lot to be desired.
      • File:Vicksburg Campaign April-July 1863.pdf is very hard to read because the background shading is too dark and the text and pattern are much too busy. Much of the text is nearly illegible, the foreground elements are easily confused with background elements, the symbols for battles are not effectively shaped or colored, etc. The description page says it's the creator's own work, but it's not clear what that means (how was the background map generated, from what sources?) The red and blue colors for lines are okay though; they have sufficient lightness contrast and well enough chosen colors to be pretty clearly distinguishable by color blind readers.
      • File:Cook Three Voyages 59.png has very problematically chosen colors. The red and green are almost completely indistinguishable to deuteranopes and hard to distinguish for protanope. This is a significant accessibility problem, and should make this map a priority for fixing. A basic fix (choosing better colors) would be easy. A more complete fix might also involve using different line styles for the three voyages. (While we're at it, the Eckert IV projection cut near the prime meridian is probably not the best choice for this map, since a lot of the action of the map takes place at the edges where there is a lot of distortion.)
      –jacobolus (t) 19:07, 18 September 2025 (UTC)[reply]

    Tools

    [edit]

    I found what looks like a really nice tool for visualizing what an image looks like to people with various sorts of color vision deficiencies: https://www.terrific.tools/color/color-blindness-simulator. If people know of other similar tools, please list them here. RoySmith (talk) 15:48, 16 September 2025 (UTC)[reply]

    If that tool (and its website) is reliable, accurate, and stable, it might be worth mentioning it in a footnote within MOS:COLOR ... especially if the outcome of this discussion encourages maps to use hues in a manner that promotes accessibility. Noleander (talk) 16:02, 16 September 2025 (UTC)[reply]
    Thanks for this. Brb after I go through all the articles I've worked on. ~~ AirshipJungleman29 (talk) 16:30, 16 September 2025 (UTC)[reply]
    If you use Firefox, hit Ctrl-Shift-I - the accessibility menu allows you to simulate various kinds of colour blindness as well as contrast loss. Nikkimaria (talk) 03:36, 17 September 2025 (UTC)[reply]
    @Nikkimaria: Thank you for pointing that out. That's a nice tool. Along the top bar of that screen, select "Accessibility", then there's a drop-down list next to "Simulate:" where various vision patterns can be examined. It's explained here as "Color vision simulation". There are some other useful tools too. Do other browsers have that?  SchreiberBike | ⌨  13:21, 17 September 2025 (UTC)[reply]
    • Remember that this and every MOS page starts with a disclaimer that says “exceptions may apply”. A map depicting a complicated situation can probably be justified as one of those “exceptions”. Blueboar (talk) 00:25, 18 September 2025 (UTC)[reply]
      Thanks for pointing out that "exception" statement .. I was not aware of it. It looks like most editors commenting above agree that many maps need an exception to the MOS:COLOR guidelines. WP:MOS says "New content added to this page should directly address a persistently recurring style issue." There have been scores of examples in FA nominations over the past decade where confusion ensued after editors suggested that maps were not in compliance with MOS:COLOR ... so this may be a "persistently recurring" situation. What do you think about the addition of a brief sentence or footnote to MOS:COLOR? Perhaps a footnote that references this Talk page topic? Or summarizes its themes? Noleander (talk) 01:01, 18 September 2025 (UTC)[reply]
    • In addition to this, I notice people making maps tend to use graphic design software, which has limitations. With GIS software freely available that is capable of doing almost anything you can do with the paid stuff, there really ins't an excuse for not using it. To add to the list a bit:
    1. ColorBrewer is a great tool to help with picking map colors that considers color vision impairment. You can see the website here.
    2. I recommend QGIS which is free and opensource to actually make maps, you can download here.
    3. Another free software that can be used to make maps is GeoDa, which you can download here. GeogSage (⚔Chat?⚔) 04:31, 18 September 2025 (UTC)[reply]

    Is this a "persistently recurring" issue?

    [edit]

    To gauge the magnitude of this issue, I looked in the "What links here" lists for MOS:COLOUR and MOS:COLOR. Limiting the search to the year 2022, and only in the FA nominations, I found about twenty (20) discussions where MOS:COLOR was applied to maps: Some examples (not limited to 2022): 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16.

    The typical pattern is: an editor improves an article and nominates it for FA; the article includes a map with color usage that is not consistent with MOS:COLOR. A reviewer points it out. The editor is puzzled, because the map was in WikiCommons, and there does not seem to be any remedy available. Sometimes the map is removed from the article. Sometimes a new map is created. Most often, the map is left in the article and the nomination continues.

    Contrasting maps with tables: Roughly the same number of FA nominations in 2022 (about 20) uncovered MOS:COLOR issues in tables, but those reviews took a different path: nearly always, the tables were immediately rectified. So, MOS:COLOR - on its face - is understandable. It is only when applied to maps that confusion arises.

    Extrapolating: if there were 20 COLOR-map issues in FAs in 2022, then there were roughly 200 in the past decade in FA. Extrapolating to the entirely of English Wikipedia, as a wild guess, there were 2,000. And that is only the discussions that involved multiple editors ... certainly there are also a large number of situations where isolated editors consulted MOS:COLOR for map guidance and were confused. Noleander (talk) 22:07, 18 September 2025 (UTC)[reply]

    Tables are usually trivial to fix, because the they are rendered by the browser and any text editor can be used to create or edit them (sometimes smarter software or small programs can be helpful, but is not necessary). They are usually graphically pretty simple, and their styling and assessment involves relatively straight-forward criteria. They are also hosted either in the page or in a template, here on English Wikipedia.
    Maps and diagrams are typically at least an order of magnitude harder to fix, because they are made using a variety of software running on various hardware platforms, usually involving at least some steps using (often expensive) commercial software, and the published version on Wikimedia commons is just a rendered output file, not an editable version which preserves the full chain of steps/materials for creation/modification; even when a fully editable source is provided, it is often nontrivial for someone else to work with it. Various map/diagram authors have different preferred tools and incompatible workflows. There has been limited effort building broadly accessible tools for creating maps and diagrams for Wikipedia per se. Making better ones would be a very valuable project, but would take a lot of effort and probably some funding from Wikimedia. –jacobolus (t) 22:28, 18 September 2025 (UTC)[reply]
    No one will fix the problem if we continue to allow the bad maps to exist. We need to purge them all, and then spend the next few years carefully monitoring new maps added to fill the void. GeogSage (⚔Chat?⚔) 01:08, 19 September 2025 (UTC)[reply]
    For your sake I hope you aren't serious. You are definitely not going to find any consensus for a position of «let's delete most of the maps from English Wikipedia and then reject any new ones unless they are made by professionals». If someone else made this post, I would say they were writing an reductio ad absurdum parody to discredit you. –jacobolus (t) 01:15, 19 September 2025 (UTC)[reply]

    "Anybody can publish any kind of a map, however bad, and get away with it. Ordinarily a field is subject to the law of natural selection the things that are bad or inadequate fail to survive. But in cartography this law does not operate effectively because the ability to discriminate among maps is not widespread in this country."

    — Richard Edes Harrison, Cartography in Art and Advertvertising, in The Professional Geographer 1950

    "If the wine is sour, throw it out!"

    — The Agony and the Ecstasy, 1965

    I'm absolutely serious, as someone who teaches cartography I believe it is my ethical responsibility to advocate for good maps. To clarify, I think we should absolutely delete maps that violate basic cartographic standards, that spread misinformation, and that are not cited in the same way we would delete text/articles that are plagiarized or spread misinformation. This not not necessitate that we only have professional cartographers in the same way insisting on proper grammar and factual accuracy does not necessitate English professors and professional researchers. Anybody can edit Wikipedia and publish anything if others don't check it. Misinformation is not excusable because it is contained in a map. POV pushing is not acceptable when done using cartography. We don't leave bad information until someone replaces it with something better. GeogSage (⚔Chat?⚔) 01:46, 19 September 2025 (UTC)[reply]

    Proposed update to MOS:COLOR

    [edit]

    Summarizing the discussion above:

    1. There were about 20 instances of confusion related to applying MOS:COLOR to maps in FA nominations during year 2022. Extrapolating: 200 in FA the past decade, and more than 1,000 across the encyclopedia.
    2. The number of incidents of confusion meets the WP:MOS requirement: "New content added to this page [MOS] should directly address a persistently recurring style issue." (emphasis added).
    3. Review of the FA nominations shows that MOS:COLOR is applied to tables without confusion, which suggests that MOS:COLOR, by itself, is understood by most editors.
    4. Ideally, all WP map data would be understandable to visually impaired readers. Where possible, maps should use the recommended MOS:COLOR techniques (lightness shading, labels, icons, dashed lines etc) to make map data accessible to those readers.
    5. There were no editors (in this Talk page discussion above) who suggested that MOS:COLOR guidelines must be strictly applied to all maps
    6. There are many competing design goals that a map must satisfy, including sophisticated cartographic design standards. Good map design is a complex process and must consider many factors.
    7. There are some maps which may not be able to meet the MOS:COLOR guidelines without adversely impacting other map design goals.

    Therefore, I propose adding the following to MOS:COLOR:

    a) "These color accessibility guidelines apply to maps, but they should be balanced with other cartographic goals and constraints."

    or

    b) ""Applying these color accessibility guidelines to maps is recommended, but they should be balanced with other cartographic goals and constraints."

    This could be text within MOS:COLOR or, less obtrusively, included as a footnote. The new text could include a wikilink pointing to this Talk page discussion. Thoughts? Noleander (talk) 17:19, 21 September 2025 (UTC)[reply]

    As written, I think both of those run the risk of creating an exception that ends up un-writing a rule. Everything in the MoS applies to everything, but should be balanced with other goals and constraints. But then I also see that this is a persistent issue with maps but not (as you note) with tables, so perhaps there's some room to draw attention to the fact that some uses of colour are more difficult to substitute than others. UndercoverClassicist T·C 20:50, 23 September 2025 (UTC)[reply]
    @UndercoverClassicist Good point. Do any of these options move in the right direction? (boldface emphasis added to indicate key differences between the proposals):
    c) Applying these color accessibility guidelines to images and maps sometimes requires judgement when balancing them with other design principles.
    d) Applying these color accessibility guidelines to maps is sometimes difficult due to other cartographic design principles.
    e) When applying these color accessibility guidelines to maps, balancing them with other cartographic design principles may be difficult in some cases.
    f) Applying these color accessibility guidelines to maps sometimes requires judgement when balancing them with other cartographic design principles.
    All of these would include a footnote referring the reader to this Talk page discussion, so they can find additional insight, examples, and background. Noleander (talk) 21:26, 23 September 2025 (UTC)[reply]
    I think c) is the best of those: being difficult isn't normally a problem for this kind of thing (you don't get a pass on MOS:PEOPLETITLES because the rules are complicated and confusing), and it's not just maps where this could be an issue: there are other types of images where colour is useful and can't easily be done away with. UndercoverClassicist T·C 06:30, 24 September 2025 (UTC)[reply]
    Okay, thanks for the feedback. I'll wait a couple of days, and if there are no objections, I'll implement the (c) proposal. Noleander (talk) 15:59, 24 September 2025 (UTC)[reply]
    It may be worth mentioning the technical difficulty of making significant changes. Just exchanging colors in an SVG file is usually not too hard (search/replace one set of color coordinates for another), but trying to change line styles, add patterns to fills, move labels around, etc., requires some skill, access to tools, and time to do the work. –jacobolus (t) 16:42, 24 September 2025 (UTC)[reply]
    That's definitely useful information. There are three places we can convey info to the reader:
    1. Text within MOS:COLOR (probably a single sentence, such as proposal (c) above)
    2. In a (new) footnote attached to the new text in MOS:COLOR; footnote would appear at bottom of MOS:COLOR
    3. Here in this Talk page discussion (new MOS:COLOR text's footnote will link to this discussion)
    @Jacobolus Would you be willing write a couple of paragraphs here in this Talk discussion? Describing "Best practices applying MOS:COLOR to maps". It would be readily available for future editors, and may even turn into something more significant (essay?) in the future. Noleander (talk) 17:06, 24 September 2025 (UTC)[reply]
    I think would be hard/time consuming enough to do well (especially with possible discussions around it) that I don't want to commit to doing it in the short term. I'd recommend again:
    Jenny, B.; Kelso, N. V. (2007). "Color Design for the Color Vision Impaired" (PDF). Cartographic Perspectives (58): 61–67.
    Feel free to ping me in a month or two and I might feel more motivated. –jacobolus (t) 17:14, 24 September 2025 (UTC)[reply]
    I understand. It is clear from the discussions above that is a very complex and subtle topic, with many nuances. That document you cite looks excellent. I propose to include a link to it in the new footnote (at bottom of the MOS:COLOR page). Noleander (talk) 18:01, 24 September 2025 (UTC)[reply]
    As discussed above, I added option (c) to MOS:COLOR. The change was made in this diff. Noleander (talk) 18:41, 27 September 2025 (UTC)[reply]
    [edit]

    Greetings, all. Have there been discussions about tolerating wikilinks more in articles whose subject is scientific? I searched but it appears there aren't any; and if that's indeed the case, I'd start a discussion about it. -The Gnome (talk) 09:09, 28 September 2025 (UTC)[reply]

    • Wikipedia:Manual_of_Style/Linking: Articles on highly technical subjects might demand a higher density of links than general-interest articles, because they are likely to contain terminology unfamiliar to the average reader. However, do not use links as a substitute for explanation; if a technical term can be simply explained in a few words, do so. EEng 09:15, 28 September 2025 (UTC)[reply]
    That's what I was looking for. Thanks! -The Gnome (talk) 11:30, 28 September 2025 (UTC)[reply]
    Retrieved from "https://en.wikipedia.org/w/index.php?title=Wikipedia_talk:Manual_of_Style&oldid=1314045939"