Wikipedia talk:Manual of Style/Archive 193

Proposed demotion MOS:NOTUSA 23 May 2017

 * Please note that this thread follows on immediately to the OP's prior thread on the same subject, Wikipedia talk:Manual of Style/Archive 191.  E Eng  00:27, 24 May 2017 (UTC)

The exact reason for this policy remains unclear.LakeKayak (talk) 22:16, 23 May 2017 (UTC)
 * It's a matter of style. I think most style guides (at least the ones I see) advise 'US' or occasionally 'U.S.'. -Sb2001 (talk) 22:27, 23 May 2017 (UTC)
 * The reason for having a manual style is to create greater consistency and coherence in our work than would happen if we allowed editors to write in whatever style they chose. What is difficult to understand about this? —David Eppstein (talk) 22:32, 23 May 2017 (UTC)
 * Over which abbreviation to use is like which spelling to use. And WP:ENGVAR says that no spelling is preferred over another.  That's what's hard to understand.  You are going to be open-minded when it comes to spelling but not an abbreviation?LakeKayak (talk) 22:35, 23 May 2017 (UTC)
 * That's a grotesque oversimplification of ENGVAR, which actually says that we *should* use a specific national variety of English spelling on articles associated with that nation, and should stick to a consistent (but unspecified) spelling preference otherwise. —David Eppstein (talk) 23:26, 23 May 2017 (UTC)


 * LakeKayak, we're not going to excise something just because you, personally, don't see its justification within 6 hours of posting your query. In fact, MOSUSA is one of the more self-explanatory bits of the guideline. From your edit history it appears you just don't understand what a manual of style is. Please, don't make a Federal case out of this. Every publication has manual of style, and this is Wikipedia's. While exceptions are possible, I've reviewed the conflict you're having and that's not one of them.  E  Eng  23:29, 23 May 2017 (UTC)
 * No ad hominem attacks are allowed.LakeKayak (talk) 23:33, 23 May 2017 (UTC)
 * Correction: You don't know what a manual of style is, and you don't know what an ad hominem attack is. It looks like you have valuable expertise in a special topic area, and Wikipedia could really use your contributions. Please, be guided by the advice of the several experienced editors who have commented (here or on your talk page) and are unanimous: you're tilting at windmills. Don't waste your time (and -- if that doesn't convince you -- others' time) on something so, so trivial like this.


 * And since we're on the question, by the OP's logic we shouldn't write UK but rather UKGBNI all the time (or, at least, leave that up to editors on each article). Absurd.  E Eng  23:45, 23 May 2017 (UTC)


 * Oppose The reason for the policy is crystal clear. One person's WP:NOTGETTINGIT is hardly a reason to demote it. I also don't like to think of the edit warring that would ensue if it was demoted. MarnetteD&#124;Talk 23:52, 23 May 2017 (UTC)


 * For one, could we avoid straw man arguments? I only want someone to explicitly state why this policy is in effect.  Not why we have a manual of style.  And also the difference between whether UKGBNI and whether or not USA should be allowed is that USA is in common use.  I have never seen UKGBNI in use at all.LakeKayak (talk) 00:12, 24 May 2017 (UTC)
 * There's something very unpleasant about your demand that everyone else drop everything to satisfy your curiosity. You go search the 200 pages of WPT:MOS archives, analyze the discussions there, and then, if you have something new to offer, come back here. Or you could just go back to improving articles.  E Eng  00:31, 24 May 2017 (UTC)


 * This page has seen plenty of silly arguments but fortunately this one can be ignored. As has been explained, one person not understanding something does not mean that everyone else has to fall in line. The current MOS:NOTUSA is fine. Johnuniq (talk) 03:46, 24 May 2017 (UTC)

Then can you explain why that specific policy is in effect? I am still waiting for an answer.LakeKayak (talk) 16:49, 24 May 2017 (UTC)
 * What part of There's something very unpleasant about your demand that everyone else drop everything to satisfy your curiosity. You go search the 200 pages of WPT:MOS archives did you not understand?  E Eng  18:46, 24 May 2017 (UTC)


 * Oppose IDONTLIKEIT is no argument. Andy Dingley (talk) 17:02, 24 May 2017 (UTC)
 * Oppose I too have never seen a crystal clear argument for US rather than USA: however consistency of style is important, even if the choice were arbitrary.
 * It has been noted that "USA" is the United States Army.
 * It has been said that US is more popular in the um.. US, than USA.
 * US is shorter.
 * A search of the archives will reveal many additional arguments on both sides, but the guideline hasn't been overturned on over a decade.
 * All the best: Rich Farmbrough, 18:32, 24 May 2017 (UTC).


 * Thank you. That was the argument I was looking for.LakeKayak (talk) 19:49, 24 May 2017 (UTC)
 * For the record, that's not the "the reason" the guideline is what it is -- it's just Rich's guesses as to the reasons. What we're trying to get you to understand is that the reasons for many of these arbitrary decisions are lost to history. Only searching the archives will tell (and often not even then).  E Eng  21:36, 24 May 2017 (UTC)
 * The short version is that "USA" is disused in high-quality publications, and the style guides MoS is largely based on recommend against it.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  03:07, 2 June 2017 (UTC)

If we were to change any advice relating to this at all, it would be to deprecate "U.S." in favor of "US". Several years ago here I presented evidence of a declining use of the dots, and this decline has increased since then. That style is also used very inconsistently when used at all: Some publishers use it only in headlines, not in regular prose; some never; some always; and many leave it up to individual writers/editors and are not consistent about it across their publication. The only real rationale for it is that "US" and "us" are hard to distinguish in all-caps headlines. WP does not use all-caps, so we have no reason to use "U.S.", when we avoid the periods/stops in all other acronyms and initialisms. Use of the dots version frequently produces inconsistent results in our prose, e.g. "a summit attended by representatives of France, the U.S., and the UK". — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  03:07, 2 June 2017 (UTC)
 * Oppose - There is no country "United States of America"; its name is "United States" and there's no sense in putting an extra letter on the acronym. I also feel there are tone issues: for Wikipedia's voice, US is more formal and professional than USA. Reidgreg (talk) 16:07, 2 June 2017 (UTC)
 * Comment That is factually incorrect. The full name of the country is "the United States of America". It's is stated as such in the constitution. And on every piece of currency produced in the country. The short form "United States" is widely used, especially in international contexts, but the full name remains "United States of America". oknazevad (talk) 00:46, 3 June 2017 (UTC)
 * Retracted. I guess I never got past the first line of the constitution, though after checking a transcript at archives.gov I notice that it has United States of America 3 times and United States alone 51 times. Then again, how often does one hear United States of Mexico/United Mexican States/Estados Unidos Mexicanos. Reidgreg (talk) 10:47, 3 June 2017 (UTC)
 * Oppose  demotion and Support US over yoo dot es dot. d.g. L3X1  (distænt write)   )evidence(  02:07, 26 June 2017 (UTC)
 * Support US over U.S. – Reidgreg (talk) 16:07, 2 June 2017 (UTC)
 * Support: I certainly agree. Although 'we avoid the periods/stops in all other acronyms and initialisms' is not true - some editors demand the full stops in eg, ie and etc. In fact, you said specifically WP should 'definitely keep the periods/dots/stops, whatever you like to call them', describing the practice as 'lazy and sloppy'. -Sb2001 (talk) 00:23, 3 June 2017 (UTC)
 * I've addressed your side complaint in a separate sub-thread below; you're talking about a different class of objects in the language (and etc. is a contraction).  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  22:42, 6 June 2017 (UTC)
 * Oppose all changes No reason to drop the current guideline, nor to depreciate the dotted form, as the latter remains common enough in outside writing. oknazevad (talk) 00:46, 3 June 2017 (UTC)
 * All kinds of inconsistent stuff remains "common" and we don't use it here.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  17:22, 6 June 2017 (UTC)
 * Note This wasn't meant to be a !voting proposal. Rather, I'm suggesting that the matter should be re-examined again with 2017 research, and be an RfC. Given some of the emotionality attached to U.S., I would suggest RfCing this at WP:VPPOL, where we have successfully resolved other contentious style matters with a more site-wide audience. A weak apparent consensus to use US emerging from a subthread of a thread about USA on the MoS talk page is unlikely to be accepted as a legit consensus by fans of U.S., so we'd just have to RfC it anyway. Since I'm seen as "partisan" on the matter, someone else should write the RfC (I have a broken wrist, anyway; typing more than 10 minutes is painful). I'm happy to contribute style guide and other source research. As in my last round of it, I would include material in favor of U.S. as well – also as with the similar research I did for the MOS:JR debate about "Xerxes Y. Zounds Jr." versus "Xerxes Y. Zounds, Jr.," style.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  22:42, 6 June 2017 (UTC)
 * Support changing "U.S." to simply "US". Having both permissible doesn't seem useful to me. If all it requires is that a given article maintain consistency within it; any editor could just change all the US's to "U.S." in a given article or vice versa, simply based on one's individual attachment to either. A potential WP:EDITWAR or quarrel all over nothing really. Its those small "pet peeves" that tend to get people to make changes; which is why I appreciate that wikipedia has set outlines for the British vs American English spelling debate, instead of allowing a "free for fall" with the disclaimer that there be "consistency" within the article. Well, who decides which article gets to maintain all "US" and which ones "U.S."? DA1 (talk) 06:38, 10 June 2017 (UTC)
 * "British vs English"! Perhaps you mean "British English vs US English". :-) Martin of Sheffield (talk) 10:46, 10 June 2017 (UTC)
 * Yes! I mean't to say "British vs American English spelling", must have been a typo. DA1 (talk) 19:42, 11 June 2017 (UTC)


 * Support US over U.S. – The dots are archaic, vestigial leftovers from generations ago when acronyms and initialisms were uncommon in print and conversation. The inconsistent use between articles, and often within a single article, is inconsistent and unencyclopedic. Also, as others have pointed out: we don't afford the same treatment to the UK, NZ and SA. — GS ⋙ ☎ 07:51, 14 June 2017 (UTC)
 * "generations ago" - stop and think before going off in hyperbole please. I'm not yet of pensionable age and was certainly taught to use full stops in abbreviations.  You assertion about other countries is also questionable.  Call then "us", "uck", "nuz" and "sar" if you want, that is consistent, but historically they were "U.K.", "N.Z." and "S.A.". Martin of Sheffield (talk) 10:47, 14 June 2017 (UTC)
 * User:Martin of Sheffield: the important word there is 'historically'. In education more recently (ie the last 10-15 years at least), you have been taught to omit the points from initialisms (and - for that matter - other abbreviations (although somewhat more leniently)). Please stop with this way of thinking that we should stick to what people did however many years ago. It is 2017. I change my language to get with this fact (accepting '4:30pm' instead of '4.30pm', and writing 'v' instead of 'vs' are just two examples). I think everyone on WP should. -Sb2001 (talk) 16:30, 14 June 2017 (UTC)
 * Sb2001, I think you are mixing up a number of issues here. GrapefruitSculpin was being downright offensive by implying anyone over 50 was archaic from generations ago.  Probably thoughtless ignorance, but offensive nevertheless.  I didn't actually state whether dots should or should not be used, only that the usage is consistent.  As you observed I did use the word "historically".  I suggest you read Orwell's 1984.  Telling people to stop thinking in ways you don't like and seeking to change the language to cover it up is not a new idea. Anyhow, before you or I say something we (or the admins) regret I suggest this subthread is closed. Martin of Sheffield (talk) 17:12, 14 June 2017 (UTC)
 * User:Martin of Sheffield: I don't like to make a habit of saying regrettable things. Anyway, I have read Nineteen Eighty-Four - an excellent novel. Comparing me to BB is perhaps a little extreme. My previous comment actually stated that I was the one changing my views. The example I gave re time notation is an example of my open-mindedness (although I would never employ the colon myself), and the versus example shows where I have changed my opinion completely. Unfortunately the MoS is effectively an opinionated instruction book. Unless something is actually wrong, what it says should be effected, and reflect one view of language. There will always be people who disagree. Believe me, there is plenty about it I do not like. And I must note, I cannot see what is offensive about Grapefruit's comments. The hyperbole 'archaic' is just used for effect. I interpret 'generations ago' to mean around thirty of forty years. -Sb2001 (talk) 18:27, 14 June 2017 (UTC)


 * So, this is my first time participating in a community discussion here, and I didn't think it'd be possible to deeply offend someone by sharing an opinion on punctuation. No need for the snark, telling me to "stop and think" before I join the conversation, even if you figure my contribution is "thoughtless ignorance." That's awfully WP:UNCIVIL. If you feel I'm "implying anyone over 50 was archaic from generations ago," well that's on you. You're reading way too deep and have extracted an interpretation so far out of left field, I had to double-check and see if you were actually replying to me. I'm not slighting an age group, I'm simply saying dropping periods between capital letters is a holdover from the past and has become increasingly aberrant in modern writing. And while we're offering anecdotes, I'll add that I received all of my formal education in the US, and dot-free is my normal.  Didn't have an issue understanding. I appreciate you filling in the gaps for me. — GS ⋙ ☎ 08:32, 15 June 2017 (UTC)


 * Strongly oppose all changes: The USA is either the United States Army or the Union of South Africa. As for the dots, just use them in US-Eng articles and omit them in UK-Eng ones. – Illegitimate Barrister (talk • contribs), 02:42, 16 June 2017 (UTC)
 * Oppose all (with fire) – "USA" is just clumsy, and "U.S." is properly abbreviated "U.S." Both MOS:NOTUSA and MOS:U.S. should be left as is. --IJBall (contribs • talk) 19:19, 23 June 2017 (UTC)
 * Strongly oppose removing the guideline. The current guideline is useful and correct. It should be kept. However, I support deprecating the use of U.S., and changing it to US in all articles. LK (talk) 19:34, 23 June 2017 (UTC)
 * Example: United Kingdom–United States relations is a good case study in why to get rid of "U.S." on Wikipedia. Otherwise-sensible editors will freely mix "U.S." and "UK" in the same article.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  07:58, 11 July 2017 (UTC)

USA in sports results
While this is up for discussion, I've been having concerns regarding sports results. There's an exception in the guideline for FIFA codes, and I understand that if the article is actually discussing the FIFA codes themselves. However, I see USA being frequently used in tables giving results of international athletic competitions, where it seems to me that US should be used. (Results are given by country, not team name, so it isn't referring to Team USA. Wikipedia prefers US, so it shouldn't matter if results are reported in sources as USA.)  It's possible this might be the result of local consensus among sports editors, but I haven't found any record of discussion on talk page archives. I've also been wondering about the validity of USA being output from some of the templates. I'd appreciate any clarification, especially if anyone is aware of past discussion on this. – Reidgreg (talk) 16:07, 2 June 2017 (UTC)
 * The IOC and it's sport-specific member associations also use "USA", as, again, the full name of the country is "the United States of America". oknazevad (talk) 00:46, 3 June 2017 (UTC)
 * Right. We should use "USA" as a code used by a sports federation (or other entity with a three-letter country code system with "USA" for that country), when a table or list is using those codes, but not use "USA" otherwise.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  17:22, 6 June 2017 (UTC)
 * As already mentioned more than once, the official name of the country is "the United States of America". That's in the Constitution, which is sufficient, but also in ISO 3166-1, which has 2-letter country code US and 3-letter country code USA. Therefore, USA is a valid code or initialism in any context. — Stanning (talk) 21:04, 6 June 2017 (UTC)
 * Consensus for years and years here disagrees with this interpretation. The official name of Rhode Island is actually the State of Rhode Island and Providence Plantations, and the official name of the UK is actually the United Kingdom of Great Britain and Ireland, but WGaF? WP has no policy that we use ISO country codes, so WGaF?  All we do with them in any reader-facing manner is provide them as details in the infobox on the country. The ISO rationale is bogus for other reasons; many of the ISO codes are based on French not English names, and several are just made-up character strings that don't correspond to anything. We do not use them as abbreviations in encyclopedic text.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  23:01, 6 June 2017 (UTC)
 * So USA should be used as a code when using those codes? I get what you're saying, but it sounds a bit circular, and the MOS:USA wording could be improved from "in certain technical/formal uses".  Is there another section of the MOS which discusses use of those codes, to which the reader could be directed?  It still seems problematic to me, that use of USA in tables could easily leak into the prose.  FIFA World Cup uses USA throughout the prose (and, somewhat ironically in terms of this discussion, uses "United States" and other common country names rather than FIFA codes in the numerous tables). – Reidgreg (talk) 15:29, 12 June 2017 (UTC)


 * I much prefer "US"—no dots, no olde-worlde "USA". So does usage outside the US; and increasinly Americans prefer it, although it's baked into so many organisational titles, sadly. But IMO neither version should be banned: just article consistency (with exceptions for titles, refs, and quotes, of course). Tony   (talk)  13:20, 7 July 2017 (UTC)

US, UK, NATO acronyms versus etc., i.e., e.g. abbreviations
above raised a side point about abbreviations of Latin expressions not being treated consistently with acronyms like UK. They're not treated consistently with such acronyms by any style guides, because they are fundamentally dissimilar, with a history predating the more general abbreviation and acronym-formation trend by centuries. The Latin group (which includes dozens more such abbreviations) are symbols (as we'll see below), not acronyms or initialisms in the sense under discussion, which are capitalized. Many of these Latinisms – etc., c. or ca., fl., id. and ibid., op. cit., cf. and so on – are contractions or truncations, not acronyms/initialisms of any kind at all. Acronyms and initialisms (a distinction some sources do not draw, or may draw inconsistently) usually refer to proper nouns – US, UK, NASA, etc. There are many exceptions, especially in jargon – AIDS, DNA, TRO, APB, NAT, OS – which are capitalized but actually refer to common-noun phrases, as it's become the norm to apply the ABC not A.B.C. or a.b.c. or abc style to new ones, whether common or proper. There are also proper-name exceptions to the fully-capped rule – Amway, Amtrak, Nabisco – by convention, i.e. by the attestation of the majority of source usage. Most of those include partial words not just initials, are usually trademarks, and often involve camel case rather than sentence case (ECMAScript, ConAgra). Acronyms which have been re-assimilated as words, like laser, take all-lowercase.

No one writes ETC and EG for etc. and e.g. (even if they would write AMTRAK!); this category of abbreviated Latinisms are a different class of convention, though the truncation and contraction means of their derivation may be similar, and some of them technically are initialisms (q.v., i.e., e.g.). They've also not been assimilated as words; no one pronounces i.e. as something like "ee" or "ay", nor e.g. as if it were the word egg, nor etc. as "ehtk". [Aside: Same goes for the non-Latinism a.k.a. or AKA, as you prefer, which should never be written as aka on Wikipedia; it has not been assimilated as a word pronounced "ahkah" or "acka".]

Some more evidence these Latinisms are a different class – a more symbolic one – of convention is that those that take pluralization usually do so by a special letter-doubling practice (which dates back to Latin manuscripts and isn't an English imposition); it doesn't actually correspond to doubling of the abbreviated word, or to the spelling of an abbreviated word: q.v. = quod vide, plural qq.v. (sometimes written q.q.v.) = quae vide not "quod-quod vide"). This is also used in pp. which is actually a symbol for Latin paginis (singular pagina, symbol: p.) not English pages, much less "page-page"; pp. is used in other languages than English. See also mss. for "manuscripts"; again, it is used outside English, and is actually a symbol for the Latin plural manuscripta, singular manuscriptum (symbol: ms.). The doubling convention evolved by analogy from the Roman numeral system (X = 10, XX = 20). Clearly, this entire class of primarily literary Latinisms are which (like various mathematical, chemical element, unit, and other symbols) sometimes incidentally look exactly like English abbreviations and sometimes do not; they are not just normal abbreviations. This is even clearer when one remembers that etc. was formerly commonly given in the more symbolic form &c., and that & itself is a symbolicized Latin et formed by fusing and bending the Latin letters. Similarly, what we represent in ASCII as No. for numero does not correspond (having an extraneous capitalization) to the word spelling; more to the point, in offline typography it's conventionally given with a raised, underlined o – № – not with a normal lower-case "o" and period/point. So, it's clearly not a regular abbreviation. See also the prescription symbol, which in ASCII is often miscalled the "Rx" symbol, from Latin recipe, 'you take': ℞ (it has no x in it, and is actually a crosshatch added to the R, which again does not correspond case-wise to the r in the word). There are others, including the paragraph symbol or pilcrow: ¶; it derives from adding vertical marks to the c of the Latin abbreviation for capitulum and has nothing in origin to do with anything starting with p. Next, see the section sign, §, which derives from Latin signum sectionis (and which is also doubled, §§, to pluralize for 'sections'). Once in a while, such a Latinism conventionally given in all-caps, as is QED in mathematical and logical proofs, though outside that context it's often given as q.e.d, and appears as Q.E.D. in older maths works.

It's essentially just accidental that some of the symbolic compressions of stock Latinisms have remained essentially plain-text abbreviations and subject to some-but-not-all rules for abbreviations in our language (mostly having instead become fixed in style since the 19th century when orthography was first normalized to any real extent), while others have become less-alphabetically symbolic, to the point of obscuring their origins. The same is true of various unit and mathematical symbols. Current English orthography rules about modern acronyms and initialisms of the usual sort don't really apply. It doesn't matter that if i.e. were a brand new acronym we'd spell it IE; it doesn't matter, for the same reason that we don't re-spell the unit symbol dB as DB, or rewrite the mathematical sum sign, ∑ (majuscule sigma, the Greek capital S) as a big English-alphabet s to match the word sum: Convention simply has not evolved in that "super-conformity" direction, and likely never will. We intuitively understand that these are symbols, not expediency abbreviations like Prof. for Professor or Dr[.] for Doctor before someone's name, or shortening Tuesday to Tues.

If anyone is looking for perfect consistency and logic in the use and form of a natural language, they will be disappointed everywhere they look. The simple fact is that most English-language style guides call for UK and NATO, for scuba and radar, and for i.e., e.g., and etc., and for cm and ft, so MoS does too, absent a WP-specific reason not to. Clarity, along with reader expectations of formal prose, are WP-specific reasons to so, and to avoid clarity-reducing and inconsistently applied fads, like willy-nilly dropping of punctuation marks without a convention or standard requiring their absence. "Follow the sources" and "go with the flow" are sufficient reasoning for us to keep going in this direction instead of trying to act as language change advocates in favor of the hyper-simplification proffered by some British news publishers (or the "keep writing like it's 1920" hyper-traditionalism of some American publishers, like the New York Times on certain style points, such as prepending "Mr.", "Mrs.", or "Ms" before any parties mentioned by surname). We have a strong disincentive (even aside from WP:NOR, WP:NPOV and WP:SOAPBOX policies) to pursue such change activism when it would actually negatively impact WP's own output for its readers.

The desire to drop the dots from i.e. and etc. is a habit picked up from British journalism, which drops a lot of punctuation (including commas) for expediency reasons at the expense of clarity, and does so in ways that violate the norms of mainstream British punctuation. For abbreviations, those well-sourced norms are to drop the point (period) from, and only from, acronyms like UK and NATO and abbreviations which begin and end with the same letters as the full word, thus St for Saint or Street, but Prof. for Professor. You'll find these conventions in Hart's Rules and its successors, and Fowler's [Dictionary of] Modern English Usage in successive editions, and other British style guides that are not the house stylebooks of specific newspapers, which are all wildly inconsistent with each other on innumerable points, and have little to do with an encyclopedic register of writing.

A supposed trend toward rewriting pronounceable "word acronyms" (just "acronyms" in the nomenclature system that distinguishes acronyms from initialisms) like NATO and AIDS as if they were words, Nato and Aids (which is already an unrelated word), is not well-evidenced outside of British journalism either, and by no means are all British news publishers in favor of it. It's the lowest-common-denominator publications like The Guardian doing it; the style books of higher-end publications like The Economist call for NATO and AIDS, as do the more academic style guides like Fowler's and New Hart's in the Commonwealth English sphere, along with virtually all North American style guides. The down-casing is, honestly, a patently stupid practice – an intentional reduction of communication effectiveness by people trying to communicate – since it results in obvious understandability problems; when The Guardian writes about ISIS (more properly IS or ISIL) as "Isis", it appears they're referring to an Egyptian mythological figure, or perhaps a modern band or actress by this monicker; no one unfamiliar with the topic (or familiar with it as ISIL or IS, as many non-native English speakers will be) will understand that an acronym is meant. (The practice originated in broadcast journalism, as a cue for pronunciation on teleprompters, and spread by reuse of material for broadcast and print release without bothering to reformat for the latter.) A trend toward expediency over precision has long existed in news writing, due to deadline pressure both for writers and for actual typesetting in the pre-digital era, and has become exacerbated by the Internet for an unrelated reason: people used to peruse the paper over coffee and on the bus at a fairly leisurely pace, and had few TV channels to choose from; but now a news organization has only seconds to grab a reader's attention on their mobile device, and less likelihood of retaining it very long on such devices or when we have hundreds of TV channels to change to.

None of these expediency pressures apply to encyclopedia writing or reading. We are a long-term reference work, not an ephemeral, click-bait infotainment source. We have a duty to write clearly and precisely (including for children, ESL learners, screen readers, and easy machine translation), and there is a reader expectation that we will do so. We fail to do so when we drop conventional punctuation and/or capitalization that helps distinguish between regular words and abbreviations (of whatever kind, other than unit symbols that have been formally standardized without them). Finally, as a matter of policy, WP is not written in news style, so we really just DGaF what British (or American, or whatever) news publishers are doing with words and names anyway.

Digression, to forestall any "what about...?" objections: An exception to that DGaF is for uncommonly fast-moving matters (like pronouns and the transgendered) that have been evolving slightly too quickly for the slow publication cycles of academic style guides, but are addressed consistently in more frequently updated journalism ones, if they actually reflect common practice outside newspapers as well. It's difficult to remember any example other than the TG one, where sources like the AP Stylebook were in fact useful in the MOS:IDENTITY debates. A conceivable additional example would be terminology should some country split and generate new nationality-indicating adjectives; news stylebooks would be updated within the year to account for it. Another exception, outside the MoS sphere, is that newspapers are often key sources for WP:COMMONNAME determination (which is not a style question), though academic book sources are preferred, when available, as higher-quality sources. For example, we can rely entirely on news coverage to determine that most reliable sources refer to The Oatmeal cartoonist Matthew Inman by that name, even though he actually signs the cartoons as "The Oatmeal". While his real name presently redirects to the article on the cartoon, if a proper WP:BLP article were split from it (and this could arguably be done, since he's also published several books and is notable for some other things, like saving the Nikola Tesla Museum project) it would be at Matthew Inman not The Oatmeal (cartoonist) or Oatmeal (cartoonist), on the strength of journalistic sources alone, since he's too recent a public figure to be mentioned in many if any books. — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  22:42, 6 June 2017 (UTC)

I will say thank you for your response. You raise some good points. I am slightly disturbed by the inclusion of the phrase 'It's the lowest-common-denominator publications like The Guardian': The Guardian is a very well respected publication. I myself purchase it several times a week on my way to studying English Language and English Literature at an educational establishment. I will note that here you are advised to drop much of the punctuation for abbreviations, acronyms, etc (including page references 'p' and 'pp'). This leads me onto the issue of 'We have a duty to write clearly and precisely ... for children': today's youth are exposed to all sorts of styles. I can say with complete confidence, what they are taught at school/college/university is what is followed in the vast majority of cases. The Classics Department at the establishment I attend generally avoids using full stops to show initials/abbreviations. I am not always keen on pronounceable acronyms' capitals being dropped, unless they either do not need to be recognised as a series of contracted words, or are such common knowledge that it is not necessary to keep the caps. I am somewhat pleased to see British media is having such an influence on the way people write rather than American. It worries me to see people writing a colon to separate hours and minutes in 12-hour time, but that is another matter. Whilst you may not see British media influence as a positive, I most certainly do. Going back to the 'lowest-common-denominator' point you raised earlier, I carry with me a copy of the Guardian style guide and the University of Oxford style guide. I do not think the latter can be described so dismissively. Its English Department is fantastic. Change is important in language, as it is in everything else. When it becomes clear that general language is swinging against the well-established versions we see in use at any one moment, we should be the ones to change. Whether that be me accepting that some people will write a colon instead of a full stop to separate hours and minutes, or you accepting that some people prefer non-dotted abbreviations. Those versions that we prefer do not have to become obsolete and unseen, but should not stay as law and the only correct version of something. -Sb2001 (talk) 23:48, 6 June 2017 (UTC) — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  05:15, 29 June 2017 (UTC)
 * Fine, then "a low-common-denominator publication", like any other mainstream newspaper when it comes to style matters, if you want to consider tabloid trash to be lowest (I wasn't including them in the analysis because they fail WP:RS and we don't use them as sources at all.)  I'm not criticizing the The Guardian 's reporting, but the entire manner in which the news industry presents text, which is sharply at odds with how more academic prose is written, in almost innumerable ways.  I mentioned that newspaper as high-profile, that's all, like mentioning The New York Times in the context of American newswriting.  Youth are, yes, exposed to all sorts of styles. Most of them are poor in clarity, consistency, lack of ambiguity, and other virtues of encyclopedic style. (They may have features that are virtues in other forms of writing, such as dynamism and "color", avoidance of passive voice, etc.)  Your classics department, and various other academic circles, are increasingly importing conventions from citation styles (a particularly encoded form of text that is not normal prose) into their normal prose; this is another example of expediency over clarity.  I don't see British media per se as a positive influence, but rather British, and American, and Canadian, and etc., increasingly commingling, with a kind of de facto "International English" slowly evolving over the Internet.  The only thing that distresses me about it is a tendency (due to the influence of bloggers, texting/IM, and comments sections and webboards – and much earlier unmediated channels like Usenet, mailing lists, and BBSes) toward the sloppier side of expediency and informalism. Fortunately this is not having much of an effect on professional writing that is subject to any editorial control. Anyway, MoS is not here to say what "the only correct version of something" is, but it is here to establish a consistent set of on-WP practices to the extent practical, so that a) the output is reasonably consistent and doesn't look like it was written by 10-year-old Reddit users, and b) the frequency and heat of inter-editor dispute over style trivia is curtailed. MoS is necessarily a bit prescriptivist in action (prefer this, avoid that), though much less so than most style guides, while not being even faintly prescriptivist with regard to rationales (i.e., it doesn't assert anything about what is more correct, proper, better, true, high-class, etc., only what is likely to be clearer for readers and what is the norm in style guides for a fairly formal register of writing.  Clarity sometimes comes at the expense of features that some readers consider not strictly necessary; natural languages are chock full of redundancy of many kinds, and the reason is increased likelihood of the message being parsed correctly. That's a Linguistics 101 point, but an important one that many people don't consciously think about. WP's goal never has been and never will be to aim for the shortest possible way of expressing something, but rather the most easily understood to the largest number of readers (short of Simple English dumbing down, of course; that's a separate project).

Here's an abbreviation: TLDR. Seriously, Wikipedia talk pages aren't your personal blog, SMc, and not the place for your personal analysis and judgement. Even so, some of your analysis is based on misunderstanding of the reason behind usage (for example, dB would never be written with a capital d because it's a metric prefix for "deci", and an international standard regardless of language). oknazevad (talk) 02:14, 7 June 2017 (UTC)
 * Actually review WP:TLDR, please. You'll note that it's about article text not talk page discussions, and has multiple warnings that calling "TL;DR" is apt to be considered rude, unhelpful, even intentionally stifling of views one disagrees with but can't mount an argument against. No one is forcing you to read what I posted. If you cannot handle a few paragraphs of well-reasoned prose, why are you at an encyclopedia project, which consists, aside from a few pictures and tables, entirely of paragraphs of well-reasoned prose?  I write detailed style analysis stuff like this because it often forestalls re-re-re-debating the same crap endlessly, something that is an order of magnitude more time-consuming, for many more people, than reading one long post. All one has to do is refer back to the analysis, and challenge anyone to refute any of it.  This strategy has served me well personally, and served MoS and the entire WP project well. It has directly settled many recurrent and temper-raising "style fights" – from whether a name like Robert Downey Jr. "must" have a comma in it (and whether a second should follow), to whether it's okay to give a song title as "Do It Like A Dude" to mimic the marketing stylization on the cover, and many many more.  The research work I put into things like this (often much more detailed, with fully formatted source citations) also doubles as research for article improvement. PS: Talk pages absolutely  the place for our personal analyses and judgement; it sure doesn't belong in articles. My views, yours, and those of everyone else here, based on sources and reasoning, all of them intermixing and percolating in editor discussion, is the very process by which consensus forms. Anyway, if you have a refutation for any argument I've presented, please put it up. If you won't read and respond to something, only vent about it not being convenient for you, why should anyone read ?  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  05:15, 29 June 2017 (UTC)
 * I found the post very valuable and reasonably long and detailed, given its subject matter. I was about to go click the link in the history to thank SMcCandlish for it. This will do instead. Also, SMcC was discussing the function of such terms (and how that in turn affected their current form), not their entire etymology.

— SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  16:57, 5 July 2017 (UTC)
 * The only wish I'd have is that SMcC would in the future break up long paragraphs for better readability for those of us with pain-related limitations, but that's certainly not mandatory on talk pages. I also would never intentionally scold anyone for not doing so, in the hope of encouraging editor retention. Well, such crankiness also would discourage people from agreeing with me. Heh. —Geekdiva (talk) 10:34, 4 July 2017 (UTC)
 * Will keep that more in mind. The paragraphs look smaller to me because I have a huge monitor.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  16:57, 5 July 2017 (UTC)
 * This is more just a request that SMc be more succinct, which is easily possible. It's hardly the first time he's been so asked. The biggest problem with the long-windedness is it makes it less likely for other users to bother reading, which is itself rather rude, especially considering his style of writing tends towards a tone of "authoritative pronouncement", leading some to inherently expect it to be factually correct. This combination has a strong potential to create misconceptions in others' understanding when there are errors of fact, such as the aforementioned dB symbol or that "Amtrak" is not an abbreviation of anything (the actual legal name of the carrier is the National Rail Passenger Corporation; the common name is a portmanteau word coined to be used purely as branding). This is particularly frustrating when there is good material in the posts, such as explaining the origin of British journalistic practice of capitalizing only the first letter of abbreviations pronounced as one word, such as NASA (though I would like to verify it, it is a plausible explanation). Again, it comes down to this being a Wikipedia talk page, not a university lecture hall, nor a personal blog. Brief, effective writing is far more appropriate here. oknazevad (talk) 14:27, 4 July 2017 (UTC)
 * Nothing I said was incorrect about dB. Nothing you said about its nature is incorrect either, but your assumption that it would "never" be miswritten as db or DB is demonstrably false with a few seconds on Google. The existence of standards doesn't mean people will follow them, otherwise we wouldn't need MOS:NUM to contain any details, and we wouldn't ever need to correct editors writing "CM" for centimetres in our articles, but of course we do. The fact that there are good reasons to continue writing the decibel symbol as dB is a sound argument – as is mine that there are good reasons to continue writing conventionalized Latinisms like e.g. in the conventional format. The entire point of comparing them is "good reasons over here, good reasons over there, and the reasons are similar: these are conventionalized symbols, not typical abbreviations"; so your pointing out that the good reasons over here are valid is in no way harming my argument. Next, Amtrak most certainly and obviously is an abbreviation, of "American" (by truncation) and "track" (by contraction).  You seem to have imagined that I said it's an abbreviation of a pre-existing trademark, but of course I did not. That the name was formed by a portmanteau process rather than by pure truncation or pure contraction is orthogonal; there is no contradiction in play here. Many abbreviations and acronyms are formed this way, e.g. NORAD (which some might prefer as NorAD), and radar. The supposition that something must be either an abbreviation or a portmanteau and never the twain shall meet is a false dichotomy; the portmanteaux process necessarily involves abbreviation (otherwise you get a compound, like teacup).  Some would disagree that Amtrak is actually a portmanteau, because it involves no blending; the elements could be hyphenated or camelcased as Am-trak or AmTrak (the latter has actually been used in Amtrak logos) without confusion, while this cannot be done with "classic" portmanteaux like smog, Brexit, Jeepster, and spork.  In a different view, extreme in the other direction, every "word acronym" like NASA and AIDS is also a portmanteau. I consider such arguments (language mailing lists are full of them) a form of omphaloskepsis; words mean what we use them to mean successfully in communicating, and definitional hair-splitting is largely a waste of time.  Your criticism of my tone seems hypocritical, since yours is the same but combined with an ad hominem streak; the latter aspect of it is off-topic here – user talk exists for a reason. (But "shut up this one particular person" positioning is anti-collaborative anyway.)  Note that concision did not serve you here; you could perhaps have made longer, more nuanced arguments; brief does not always equate to effective. I don't feel very refuted.  Verification of changes in news style is "easy" but expensive and tedious.  Start collecting style guides from every English-speaking nation and institution, via eBay, Amazon, and every other outlet, and actually study them. Broadcast journalism has been using "spell it the way you want it read aloud" teleprompter tricks for decades before anyone in print journalism starting printing abominations like "Aids" for AIDS, a trend that began after a sharp uptick in media company conglomeration and the increased repurposing of news for multiple media as a cost-saving measure.  Is this original synthesis on my part? Sure, and you won't find me trying to insert it in an article here, because a secondary source on language change or mass-media history hasn't written about it yet (that I know of).  But it's more sound than most as an observation in a free-form discussion, especially on a talk page usually dominated by irrational pronouncements of the One True Way to write English.

"Sir" is an honorific prefix, not a forename
"Sir" is an honorific prefix, not a forename. As such, it belongs in the honorific prefix section of an infobox, not immediately precedent to and on the same line as the article subject's name. It functions exactly as other honorific prefixes — "Mr", "Mrs", "Ms", "Dr", "Rev.", and so on.

--Vabadus91 (talk) 00:37, 11 July 2017 (UTC)
 * Yes, but why state that here? This is already covered at MOS:BIO.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  03:32, 11 July 2017 (UTC)


 * "Sir" is exceptional in that Sirs are called "Sir" at Deaths in 20xx articles when they die, no matter their article name. I tried to argue they should be treated like a Doctor or Reverend, but was utterly destroyed. Perhaps this means the normal rules also don't apply to them in infoboxes (and elsewhere). InedibleHulk (talk) 05:00, 12 July 2017 (UTC)


 * As SMcCandlish points out, this is is covered at MOS:BIO, specifically MOS:HONORIFIC which makes a distinction between honorific prefixes and the honorific titles Sir, Dame, Lord and Lady. The latter therefore are not honorific prefixes and do not belong in the honorific prefix section of an infobox. If you want to discuss this and/or propose a change to the MOS, you're welcome to do so at WT:MOSBIO. — Stanning (talk) 09:35, 12 July 2017 (UTC)


 * With respect, I still disagree. Honorific prefixes vs honorific titles are a semantic distinction without a difference. Generally, both are simply called "honorifics" in English (see Wikipedia's own page on English honorifics). Mr, Mrs, and Ms are in fact titles, just as Dr, Rev'd, and Sir are. The most accurate term for all of them is "pre-nominal honorific" (it might be an idea for honorific prefix to be changed to pre-nominal honorific for infoboxes). Even if there were a meaningful difference (the only difference is what the title signifies — marital status, knighthood, academic achievement, etc. — not its linguistic categorisation as a pre-nominal honorific in all cases), Sir remains an honorific of some kind, so wherever it belongs, it isn't as part of a name as though it were a forename.


 * I have posted this here because here is where I was directed following a previous disagreement over an edit.


 * Vabadus91 (talk) 14:25, 12 July 2017 (UTC)


 * You're welcome to disagree with the MOS – lots of people do – but not to ignore it. On this point, it's quite clear: MOS:HONORIFIC says honorific prefixes vs honorific titles is a distinction with a difference! Either comply with it or, since you don't agree with it, propose a change on the relevant talk page. — Stanning (talk) 14:48, 12 July 2017 (UTC)


 * And I shall do exactly that. Thank you for the discussion.


 * Vabadus91 (talk) 22:36, 12 July 2017 (UTC)

Mixing "series" and "season" in the same article?
Is this an ENGVAR issue? Specifically, Thronecast is a British programme discussing an American programme, and each "series" of the former coincides with a "season" on the latter.

I think For the third series [of Thronecast], Thronecast became a 15-minute on-air show that [followed] each episode of the third season [of Game of Thrones] looks particularly weird, because if the British terminology was changed to American or vice versa, it would read like the third series was the same thing as the third season; it appears that the different terms are being used to disambiguate inline.

I'm actually not sure if this is a problem, though, as I have heard Americans (specifically Chuck Sonnenburg) use the word "series" when describing British shows like Red Dwarf and Doctor Who, which leads me to believe that the "terminology" is mutually intelligible, but it still seems weird...

Hijiri 88 ( 聖やや ) 01:07, 13 July 2017 (UTC)


 * I don't think there can be much of a case for mixing the terms. They are both television shows (programs) and using different terms creates an impression of a substantive difference. In your specific case, the article is about a British show so BrEng seems obvious (to me). Primergrey (talk) 04:44, 13 July 2017 (UTC)
 * Agreed. They're synonyms in this context; as long as it's clear that "series" here means "series of episodes presented during a broadcasting season" not "entire TV show", use "series" consistently. By way of direct comparison, an article on autos (cars) would not compare the "boot" size of a BMW four-door sedan to the "trunk" size of comparable type of American-manufacturer car, much less in the same sentence.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  23:14, 13 July 2017 (UTC)

Roman Catholic
I'm not sure this is really significant enough, but I can't think where would be better to ask. Periodically I see articles which say "Fred McSubject is a Roman Catholic". I usually alter this to say "Catholic", given that the word "Roman" seems entirely redundant. I appreciate that - well, essentially, the stuff in the lead of Catholicism is so - but AFAIK in ordinary usage the word "Catholic" alone is always interpreted as referring to the Catholic Church, and as such "Roman" doesn't alleviate any actual confusion on the part of the reader. Comments? Pinkbeast (talk) 16:57, 10 July 2017 (UTC)
 * There are two issues here.
 * The attempt by the Roman Catholic Church to appropriate the term "Catholic" is deeply offensive to some other Christians. It's a biassed claim to universality. As an Anglican, I consider myself a member of the "holy Catholic Church" as per the creed I say weekly. Wikipedia should be neutral as to claims of being "the" Catholic Church, just as it should be when some branches of Islam claim that other branches cannot be called Islamic. The correct neutral description is "Roman Catholic".
 * Even within its own usage, "Roman Catholic" is not the same as "Catholic". There are non-Roman churches recognized by the Roman Catholic Church as "Catholic", such as the Eastern Catholic Churches. An example of the confusion that comes from eliding usage is in the statement at the head of that article: This article is about Eastern churches in full communion with the Catholic Church The Eastern churches that are in full communion with the Roman Catholic Church form part of what they call the "Catholic Church"; to say they are in communion with the "Catholic Church" is to say they are in communion with themselves.
 * So, no, "Catholic" is not the same as "Roman Catholic". Peter coxhead (talk) 17:49, 10 July 2017 (UTC)


 * If I remember correctly from my GCSE RS, you can have Catholic Christianity, but not Roman Catholic Christianity. Roman Catholicism exists as something else. So it is almost like they are entirely separate ideas/belief systems. – Sb2001 talk page 18:02, 10 July 2017 (UTC)


 * Agreed with these comments; the distinction is encyclopedically important. Lots of "how people say stuff in daily life" matters are not how we write at Wikipedia, usually for similar precision, disambiguation, and other clarity reasons.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  19:11, 10 July 2017 (UTC)

I'm afraid I don't quite agree with the replies above. The salient point about the "attempt by the Roman Catholic Church to appropriate the term "Catholic"" is that it has been entirely successful. In every context save discussion about flavours of Christianity, "Catholic" is clearly and unambiguously understood to mean Roman Catholic. It has been for hundreds of years (eg Pepys and his contemporaries write about "Catholiques", when they don't write about "Papists") and it is today.

There's no issue of precision or clarity. If you read "X is a Catholic" there is no actual doubt as to what is meant. Wikipedia is already littered with unqualified "Catholics" none of which produces the slightest confusion in the reader. Pinkbeast (talk) 14:18, 12 July 2017 (UTC)


 * Nonsense. There are many churches which use "Catholic" in their names. And I have been to enough funerals to know that the Roman Catholic Church itself often refers to itself that way ("Roman Catholic"), as shown by the use of that term in hymnals and prayer books.  Kablammo (talk) 20:18, 12 July 2017 (UTC)
 * And in response to a comment further up: The Roman Catholic Church is a Christian church, full stop.  Kablammo (talk) 20:25, 12 July 2017 (UTC)

Agree with Peter. "Roman Catholic" is a member of the group "catholic", so the two are not interchangeable in an encyclopaedia. This is complicated, though, by the existence of our article Catholic Church. Now there's a can of worms (or is it "diet"?). Still, I don't see that it's necessary to address that in order to do this in the right way. Formerip (talk) 20:59, 12 July 2017 (UTC)


 * One of many articles which use an unqualified "Catholic" without producing any confusion whatsoever. Pinkbeast (talk) 13:08, 13 July 2017 (UTC)
 * If there is more than one catholic church, and we have an article with that title dealing with only one of them. then that's potentially confusing. Formerip (talk) 01:50, 14 July 2017 (UTC)

The Vatican itself uses "Roman Catholic". Kablammo (talk) 21:12, 12 July 2017 (UTC)


 * Of course "Roman Catholic" is itself correct. My argument is that one word in it is unnecessary. Pinkbeast (talk) 13:08, 13 July 2017 (UTC)

Possessive of singular nouns ending with a SILENT s?


This -- the use of "Louis' father", etc. in the Louis XIV of France article -- seems like a violation of MOS:POSS. Specifically Add 's if the possessive has an additional /ᵻz/ at the end. But it's been a long time since I read any professional work on French history in English, so I may be missing some way in which it could be covered under the following Some possessives have two possible pronunciations.

Thoughts?

Hijiri 88 ( 聖やや ) 08:33, 12 July 2017 (UTC)
 * Nice one. Basically, of course, follow MOS:POSS which says to follow the pronunciation. In the case of Louis, I guess it depends whether you pronounce the name with a silent s as in French and as in British English, or whether you pronounce it "Lewis" as in American English (don't know how it's pronounced in other flavours of English). In British English the possessive form of Louis is pronounced with only one s, not two as in Morris's or Lewis's, certainly when the Louis in question is French, so when writing British English I'd write Louis' as I think most Brits do, although some British writers have used Louis's. I'd suggest that any article where the question arises should specify its flavour of English (with ) and use the approproiate possessive for that flavour. — Stanning (talk) 10:16, 12 July 2017 (UTC)
 * My experience of American English is somewhat biased in this case, since I grew up in Ireland learning French since primary school (my mother was also a French teacher), and so when I heard "incorrect" pronunciations of French names my mind might have just zoned them out, but... do Americans actually pronounce the king's name as "Lewis"? The only American source I can think of off the top of my head is two lines spoken by David Hyde Pierce in episodes (2:17 and 4:22) of the sitcom Frasier, and he didn't pronounce the s, which leads me to believe that if it had been possessive he would have pronounce it with an additional /ᵻz/ at the end. Hijiri 88 ( 聖やや ) 08:11, 13 July 2017 (UTC)
 * I'm a native speaker of British English. In my experience of American English, they pronounce Louis inconsistently! For example, the s is sounded in St Louis, Missouri, and Louisville, Ohio, but it's silent in Louisville, Kentucky, and other places. Go figure, as our friends over there say. But "Lewis" seems to be the more usual prounciation, so much so that the other pronunciation is often spelled "Louie". The only Americans that I've heard mention the kings were (dare I say) well-educated Americans in Europe, and they of course said "Louis" reasonably close to the French pronunciation. I think I was mistaken. I've been trawling my bookshelves for examples, and although I only found two, they were both Louis's. And I see that MOS:POSS does cover the case of a silent s implicitly, where it gives the example Descartes's philosophy, and indeed a silent x in Verreaux's eagle (though it's not clear from the context how the writer expects Descartes or Verreaux to be pronounced). Following those examples, Louis's is correct in Wikipedia. MOS:POSS does perhaps need a clarification, though. — Stanning (talk) 10:20, 13 July 2017 (UTC)
 * We borrowed that directly from New Hart's Rules and Chicago Manual of Style, the main sources of inspiration (and of specifics) for MoS. Since there's not a different rule for silent-s/x/z, there isn't any need to say anything about the matter. Those paper style guides include a separate segment on it because they're huge and attempt to be comprehensive (perhaps to a fault), and because some guides (including CMoS) have historically had different rules for silent-s/x/z (which is why anyone even asks about it).  We're trying to keep MoS short, at least on the main page (there's really no way to make MOS:NUM short!), so the examples illustrating the silent-s/x endings are sufficient.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  09:27, 14 July 2017 (UTC)
 * In American English, it is pronounced both ways. I think the pronunciation that should be considered is the French one with the silent 's', so Louis's should be used. AHeneen (talk) 15:17, 13 July 2017 (UTC)
 * I agree, at least for the name of the king. In case you're interested, there is further discussion at Apostrophe. – Corinne (talk) 15:41, 13 July 2017 (UTC)

— SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  09:27, 14 July 2017 (UTC)
 * Louis pronounced /LOO-iss/, and French /Loo-EE/ are two different names with the same spelling (the formerly clearly being an Anglicization of the latter, influenced by Lewis). In the Southwest and other areas with a lot of Spanish-speakers, we're also familiar with Spanish Luis, /Loo-EES/. (No, I'm not going to IPA all that; too much hassle.) The name Louie, pron. /LOO-ee/, is a diminutive of the /LOO-iss/ version of Louis, though many mispronounce the French name this way. As an American, I've never heard an American (or anyone else) above an elementary-school education level refer to any "King /LOO-iss/ of France", except in jest. The usual pronunciations, English-wide, of Louis in reference to a French monarch, are /Loo-EE/ or /LOO-ee/ depending on French familiarity. The possessive is Louis's,  here and in most other styles. (Chicago Manual of Style tried switching to Louis'  in the 15th ed. for unknown reasons, and after getting slammed for it, switched back in the 16th and current edition, with an almost apologetic note.)  Other major, modern style guides back this, including New Hart's.  But there's no WP-related rationale to use Louis'  in the possessive of St. Louis - /LOO-iss/ – Missouri, either, or similar constructions like Kanasas's, Marx's, showbiz's. The various and contradictory "punctuate based on the sound of the word/name where you grew up" approaches are a [bad] joke, abandoned by more style guides as time goes on.  It's untenable except in local-to-regional publishing, because pronunciations change from area to area, and there is a lot of dialectal variation in the UK, US, Canada, Ireland, etc.  I firmly predict that the "huge pile of style guides" sourcing run I've started will show that there's an overall preference in academic and general-audience guides for "Jones's" style as a consistent approach to all the "what if it ends with ...?" questions; that in journalism, marketing, and business stylesheets there's a preference for "Jones as yet another typical, niggling big of compression for its own sake from that sector; and that newer works are less apt to try to justify complicated exceptions.  MoS almost always lands on the academic-through-general side of this broad rift in English style.  I'm doing the source research mainly for Apostrophe, which isn't bare, but has too few sources, is missing obvious major ones, uses some old and unreliable ones, and has clear original research, like unsupportable generalizations about English usage from nothing but a single newspaper's internal house style sheet.  MoS's main sources, CMoS and New Hart's Rules, both back an apostrophe-s standard, the first explicitly and the second by default.  NHR is more permissive of optional exception-making (for persnickety reasons no one is ever going to remember, about multisyllabic stress which only matters for reading out loud – precisely the irrelevant, editor-unfriendly WP:CREEP stuff we avoid here). Fowler's is even worse for our purposes in this regard; but both mostly treat their declared exceptions as optional. CMoS no longer goes there at all; since 2010, it sticks with apostrophe-s, including for Jesus's and Camus's, and explicitly declares in two places that it is overruling previous editions on these matters, for increased consistency and logic. Plus ça change ...

"The internet" and "the web"
Somewhere I thought we were already addressing this, but I don't find it in MOS:MAIN, MOS:CAPS, or MOS:PN. We really should address this. The Internet and the Web (short for the World Wide Web, which no one says any more) are proper names; they are specific, unique things like the Pacific Ocean and the language French. Journalism style guides (Wikipedia is not written in news style) have been pushing hard to lowercase these, but they're factually wrong to do so when the do it to the Internet and the Web.

Lower-case internet as a noun means "a (i.e. any) network of networks". Various internets combined to form the Internet, and there are internets that were not and still are not part of it (though various internets are also part of the Internet today). Due to ambiguity, internetwork or inter-network – or more recently a completely different term, wide-area network – has usually been used instead.

Lower-case web as a noun is something a spider makes, with various metaphoric applications of the word ("a web of trust", etc.).

Lower-case as adjectives, both words refer to technologies (protocols, standards, applications, etc.) that can be used for intranets as well as the Internet. These are thus properly lower-cased most of the time; it is hard to think of any such technology that cannot be used for an intranet. Both can also be capitalized as adjectives when used to refer specifically to the Internet or the Web ("he had no Internet access while on the island"). — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  03:31, 11 July 2017 (UTC) — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  10:01, 14 July 2017 (UTC)
 * You probably know we have Capitalization of "Internet". I recently said the above on its talk but I have a feeling there is an enwiki recommendation to use lowercase per everyone does it. Johnuniq (talk) 04:27, 12 July 2017 (UTC)
 * Everyone's been wrong before. I recommend caps. InedibleHulk (talk) 04:54, 12 July 2017 (UTC)
 * If you recommend caps then I suggest hatting this discussion.  E Eng  05:03, 12 July 2017 (UTC) That's a joke.
 * I almost toque it seriously. InedibleHulk (talk) 05:18, 12 July 2017 (UTC)
 * Well played. You are cordially invited to join my glittering salon of talk page stalkers.  E Eng  05:48, 12 July 2017 (UTC)
 * Sometimes the only thing on 's mind is a hat. Atsme 📞📧 12:26, 13 July 2017 (UTC)
 * Ya'll are brimming with humor. Now off to EEng's talk page if you want to continue this. --Izno (talk) 12:33, 13 July 2017 (UTC)
 * "Everyone's been wrong before" indeed, and "everyone" doesn't write the internet; it's something news journalists do (only for about the last ten years, and not universally – New York Times didn't make the switch until 2014), and something that (non-tech-industry) people writing business memos do, and something that people who send "hi jane, CU at 7" in SMS messages do. People in tech professions know better, and care. Just like doctors know and care about the difference between a canker sore (a bacterial lesion from biting the inside of your lip or cheek) and a chancre sore (a viral lesion caused by a sexually transmitted disease). In both cases, they sound the same, are hard to distinguish for the non-expert, and have a similar linguistic origin, but a provably different meaning.  To the average schmoe, both sores are a similar "ouch, this hurts" thing that is alarming but not life-threatening.  To the average schmoe, an internet[work] and the Internet are "ooh, I can get my computer to do something, if I'm lucky, and it will probably be in a browser or a web app" things, that work in ways they don't understand and can't tell apart. But the "distinction" argument isn't even necessary; "the Internet" is a proper name/proper noun in both the philosophy and linguistic senses, so it's capitalized per our style guide (which is not NYT 's or The Guardian 's), no different from the Pacific Coast Highway and the International Space Station – unique, monolithic human creations. I chose these examples carefully, because there's more than one major thoroughfare on the US Pacific Coast and more than one orbital habitat/lab created through multinational cooperation. "A Pacific Coast highway" and "an international space station" are valid expressions, referring to real categorizations, just like "an internet" (or internetwork or WAN).  PCH, the ISS, and the Internet all are the most notable but not exclusives entries in their respective categories.

Scrolling lists?
Hello, as I think I might be interpreting MOS:SCROLL incorrectly, could somebody tell me if the usage I have made here (the six images) is acceptable or not? Without it, the last image overflows outside of the white frame and into the gray space. I would've put them into a gallery, but the fact that they are templates doesn't let me do that easily. — Anakimi talk  18:44, 9 July 2017 (UTC) — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  22:29, 9 July 2017 (UTC)
 * Not a robust approach. Output is going to be highly variable by device and browser, and I suspect the results are going to be negative on mobile. In Chrome on MacOS with a big monitor, I see a pointless vertical scroll bar appearing and that is all.  At any window width, the vert. scroller remains.  If I markedly reduce window width, I start getting a horizontal scroller, but only briefly, when the field images are inlined with text between the Di Stéfano and Amarmo images.  If I make it smaller still, the field images all snap to below the two player pictures, and the horiz. scroll goes away.  At a very narrow width I get the horiz. scroller back, only when the viewport is finally too small to hold the field images on one line.  Which of course raises the question: Why are you trying to force them all to be on one line?  It would be a more usable and accessible approach to just let them flow naturally, and appear on two or more lines as needed (especially on mobile browsers). It's said that ~60% of our users are on mobile devices now (of which probably 4/5 are phones and phablets). See Real_Madrid_C.F./sandbox for a robust approach that should work regardless of OS and browser. On a big monitor, the images are all on one line. They'll auto wrap and re-center as needed on smaller devices, the way  works.  I've applied this fix to both sets of field (pitch) layouts, and to the uniform (home kit) material. The CSS can be put into templates; you need a wrapper div with  ; created that as, since this isn't really footy-specific.  The image templates themselves can be edited to have a parameter to use   for such layouts.  That would probably save time in implementing this solution on other articles. PS: I also fixed a typo and an venue infobox parameter error.
 * That was very helpful! Thank you very much! — Anakimi</b> <i style="color:grey; font-family:Courier New">talk</i>  02:25, 10 July 2017 (UTC)
 * You're welcome. If you need help integrating the  into templates, let me know.  It should work regardless what the surrounding element of the template is, so for some templates it can just be passed in a style or css parameter to a parent template. For templates that aren't meta-templated, the template's own code may need to be changed to be wrapped in a  or  to which this CSS is applied when requested, e.g. by a y being passed to the template.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  19:14, 10 July 2017 (UTC)
 * Update: Manual CSS no longer needed, due to .  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  02:48, 16 July 2017 (UTC)

Can't seem to find this...
There's an editor at Futurama that believes that when talking a possessive of the show's title (eg, that the  should fall within the italics eg , but best I've seen that standard practice is to keep those out of the italacs implicitly set by the docs for  and friends . But the MOS does not lay out this type of case. The style guides that I can see online agree with the latter but I think if we don't have that already we should spell it out. As well as the case to avoid possessive forms of quoted titles (which the style guides also agree on). --M ASEM  (t) 05:46, 16 July 2017 (UTC)
 * Ha! As it happens, I've got a Museum on that very subject!  E Eng  05:50, 16 July 2017 (UTC)
 * That's perfect, and the AP one I could find agrees too, but we don't list that here at MOS, hence why we might want to 1) make sure we agree that this is right, and 2) add that style to be clear to the MOS. --M ASEM (t) 06:02, 16 July 2017 (UTC)
 * It's not a big deal, but I'm wondering about the logic of excluding from italics the bit that changes the grammatical function of a titular item. In languages that signal some of their grammatical changes by morphology, I'm pretty sure they include the pre- or post- addition in italics. And what about that in English? "How many Futuramas are there? Just one. Tony   (talk)  06:11, 16 July 2017 (UTC)
 * a titular item: Tony, you're not channeling Neelix, are you?  E Eng  10:18, 16 July 2017 (UTC)

Overcapitalization of "vol." and "no."
Presently we have text in WP:MOS that says "write volume two, number seven or Vol. 2, No. 7." This appears to conflict with common practice, and with MOS:CAPS. Under no other circumstance would we permit capitalization of a common noun simply because it's abbreviated (and "no." is an abbreviation, of Latin numero, which is not written Numero; see also "etc.", "cf.", "e.g.", and a zillion other Latinisms in English, which are uniformly lowercased). I would suggest that this be changed to "vol. 2, no. 7".

Regardless, this section should probably also be moved to MOS:NUM, and only summarized here in compressed form. MOS:NUM doesn't have anything on this, but it is the obvious place to look, and we're trying to move nit-picks out of the main MOS to the, well, nit-pick pages. — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  22:06, 4 July 2017 (UTC)


 * Is it true to say that most academic reference styles don't use Vol and No at all any more, but just use the order of the two numbers, typically separated by a colon or comma, to denote those? Just sayin'. Martinevans123 (talk) 22:16, 4 July 2017 (UTC)
 * Sure, and our citation templates also do this, but this has nothing to do with usage in prose. We should not be advocating overcapitalization of abbreviations of non-proper nouns. It's just baseless and without any apparent precedent. What I think has happened here is that someone was thinking of how something like this is presented in subtitle style on a book cover or frontispiece ("My Adventures with in Elbonia, Vol. II") and has overgeneralized.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  01:42, 5 July 2017 (UTC)

I would always favour 'No' and 'Vol'. Caps seems to be well-established, and recommended by most style guides I see. I am not sure of the reasoning, though. Maybe it has something to do with their origins in titles. There is possibly an argument for dropping them (as well as the full stops!).-Sb2001 (talk) 22:22, 4 July 2017 (UTC)
 * I would tend to agree. If we have to use those abbreviations, I think they should be capitalised. Would also be happy to lose the full stops. Martinevans123 (talk) 22:30, 4 July 2017 (UTC)
 * Fullstops are grammatically correct for abbreviations, and are especially useful when as in this case the abbreviations could easily be mistaken for words.-- JohnBlackburne words<sub style="margin-left:-2.0ex;">deeds 23:42, 4 July 2017 (UTC)


 * JohnBlackburne: Americans seem to dot everything in sight, though there's a glacial move toward easing up on that count. Outside North America, it is usual to dot an abbreviation that doesn't end with the last letter of the expanded version, but not otherwise ("St", "Dr"). Acronyms and initialisms are now not normally dotted. I think "No." and "Vol." should be dotted in references; but conventional "22(3)" is fine if consistently used in a reference list. Tony   (talk)  06:04, 6 July 2017 (UTC)


 * "I would always favour..." – on what basis? "I am not sure of the reasoning, though."  Okay, so just WP:ILIKEIT.  "I would tend to agree" – Okay, you also just like it and don't have a rationale? "I think they should be capitalised" – on what basis? This is not a discussion about ".", either. We all already know that some indeterminate percentage of British and Commonwealth editors prefer dropping a bunch of punctuation, and other editors from the same places disagree, while editors everywhere else disagree, so that discussion is not going to reopen or go anywhere if it did.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  01:42, 5 July 2017 (UTC)
 * User:SMcCandlish: Please do not try to patronise me. I think we should avoid this discussion. It is turning aggressive, and not on my part. I will not partake in a discussion when the other editor's words and dismissive attitude affect my work in real life. Yes, they act as a distraction - they are harsh and somewhat personal. I have been thinking about what you - and other editors - have said in response to my innocent and somewhat friendly contributions all day. I cannot deal with that. I contributed here to be helpful, and add my opinion. I will refrain from partaking in your discussions in future. I will not let WP have an effect on my real emotional state. Well done. -Sb2001 (talk) 16:05, 5 July 2017 (UTC)
 * No idea what "aggressive" stuff you're referring to. I'm not patronizing anyone, I'm simply expecting an actual rationale to be provided, not just more WP:ILIKEIT or WP:IKNOWIT.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  03:49, 6 July 2017 (UTC)
 * Stanton, in your opening statement you say "This appears to conflict with common practice". I wonder could you elucidate what you mean by "common practice" here? Or perhaps there are prescriptive guidelines for usage, in such publications as The Chicago Manual of Style? Thanks. Martinevans123 (talk) 07:31, 5 July 2017 (UTC)
 * I mean that it is not typically done; "[not] common practice" seems pretty clear to me. :-) This is not an article and we need not provide source citations to come to consensus. But to look at a few anyway: Chicago Manual of Style has "vol." (which means Turabian and Scientific Style and Format should as well, since they follow it on such matters closely, but someone can go look if they want.) MLA Handbook has "vol." AMA Manual of Style (the medical AMA) has "vol" (its citation style eschews "." after an abbreviation in cites), except after a period (full stop), where it is necessarily "Vol".  The Red Book (legal), AP Stylebook, New Century Handbook, AMA Handbook of Business Writing (the marketing AMA), and Bedford Handbook do not address this directly, but certainly do not advise capitalizing abbreviations of common nouns, and provide examples of them in lower case. ACS Style Guide uses "Vol." but only in formatted reference citations; that cite style seems to capitalize all cite elements. Publication Manual of the APA doesn't use "vol." or "Vol.", just a bold number (in formatted citations; they don't address use in running prose, but I see no evidence of support for upper-casing abbreviations of common nouns). The APA pattern is common in other academic/scientific fields' guides; the issue doesn't arise in them because they just use numbers.  However, if you examine them closely, they also use "vols." to indicate multi-volume works, and this is lower-case in the instances I've found so far (e.g., in APA and in MHRA Style Guide, as well as in CMoS).  All I've got time for right now.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  03:49, 6 July 2017 (UTC)
 * Ah yes, those style guides again. What I meant was - what's that corpus of English usage on which you base your conclusion about "common practice"? Or is it just your own personal subjective view of common practice. You know, along the lines of WP:ILIKEIT or WP:IKNOWIT? Thanks. Martinevans123 (talk) 07:37, 6 July 2017 (UTC)
 * My assessment is based on style guides; I have the rules of many of them on many points memorized, and am willing to cite from them directly when others request it, as above. There's no ILIKEIT or IKNOWIT in this, and it's pretty silly for you to make that claim right after I dumped a bunch of reliable sources in your lap.  Your turn: provide sources in favor of "Vol.", not inside a title or subtitle, not after a stop/period, and not just in formatted citations, but in mid-sentence in running text. Good luck.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  23:13, 6 July 2017 (UTC)
 * I'm not sure that any number of prescriptive style guides, many of which may have parts simply copied from other earlier style guides, always translate into "common practice". Much in the same way that the content of the MoS guide for Wikipedia doesn't always translate into "common practice" here. Martinevans123 (talk) 12:51, 7 July 2017 (UTC) p.s. please don't dump things in my lap. Thanks.

An "I'm dismissive of style guides" argument equates to "I don't have an argument" here, since what MoS says is largely based on what other style guides are doing (with priority generally in decreasing order from academic through general-audience to journalism and marketing guides). You wrote: "Caps seems to be ... recommended by most style guides I see". Really? I've cited plenty that don't. I ask the same of both of you, again: Please provide sources in favor of "Vol[.]" and "No[.], not inside a title or subtitle, not after a full stop (period), and not just in formatted citations, but in mid-sentence in running text. Let's count them and compare both their number and their reputability to what I've cited already and continue to cite below.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  22:00, 11 July 2017 (UTC)
 * An assertion that "style guides don't dictate real usage" equates to a "style guides don't dictate real usage" assertion. Nothing more, nothing less. By all means demonstrate the corpus of written English on which your assertion of "common practice" rests. Not what "common style guides" dictate, but what common usage demonstrates. Thanks. Martinevans123 (talk)
 * My sources trump your lack of sources and your handwaving at the sources provided. Cf. the fallacy of moving the goalposts (a.k.a. raising the bar): an argument in which the evidence presented, in response or upon demand, is dismissed and some other sort of, or greater volume of, evidence is demanded; it's a variation on shifting the burden of proof.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  01:58, 12 July 2017 (UTC)
 * PS: I'm collapse-boxing this sub-thread since its continuance isn't going to be of interest to anyone but you and me, if that.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  02:37, 12 July 2017 (UTC)
 * I'm sure this discussion might be of interest to many other editors. I've waived no hands, I've moved no goalposts, especially the bar. I suspect that if I wrote "The cat sat on the mat" you would accuse me of ignoring the sedentary habits of canines and of perpetuating the unfair mistreatment of floor-based household products. Martinevans123 (talk) 09:55, 12 July 2017 (UTC)
 * It was of interest to me, so I read the whole thing. It seems to me that Martinevans specifically requested examples of "prescriptive guidelines for usage, in such publications as 'The Chicago Manual of Style'" and then complained, in a rather biting and unhelpful tone, that prescriptive style guides actually aren't useful after all and all we care about is popular usage when SMcCandlish did precisely as asked.  If that's not moving the goalposts, I'm not sure what is.  I am glad that editors with such impressive command of style guides persist notwithstanding this sort of unfair and unproductive acrimony.   AgnosticAphid  talk 06:07, 21 July 2017 (UTC)


 *  Lowercase What David Eppstein and Trappist said (below). And I'm all for moving as much detail from main MOS into subsidiary pages; the main MOS is grossly bloated for what ought to be its function, which is to give the high points for a newcomer who cares, but doesn't care that much.  E Eng  23:49, 4 July 2017 (UTC)
 * Lowercase: As the first sentence of MOS:CAPS says, "Wikipedia avoids unnecessary capitalization." Let's stick with that philosophy. It might be good to put the advice into MOS:CAPS or WP:CITE rather than (or in addition to) MOS:NUM. —BarrelProof (talk) 23:52, 4 July 2017 (UTC)
 * Mixed. Capitalized when it is in the middle of a title-case title: "The Art of Computer Programming, Vol. III: Sorting and Searching", or when it's after a full stop (as used to separate things in Citation Style 1, except that CS1 doesn't use the Vol. abbreviation ). Lowercase in other situations, including in the example sentence at the start of this discussion section. —David Eppstein (talk) 00:13, 5 July 2017 (UTC)
 * cs1:
 * lower case numero because there isn't a terminal stop after volume number.
 * cs2:
 * —Trappist the monk (talk) 00:35, 5 July 2017 (UTC)
 * Agreed it can be capitalized when used in a subtitle like "Vol. III: Sorting and Searching"; that's already covered by MOS:TITLES at least in theory if not in specifics. No objection to explicitly clarifying that.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  01:42, 5 July 2017 (UTC)
 * If the idea is to follow a full stop with an uppercase letter, doesn't that cs1 example break down a bit when " " becomes ""? —BarrelProof (talk) 02:30, 5 July 2017 (UTC)
 * MOS and MOSNUM cover the body of the article, infoboxes, etc. If you want to discuss citations, please discuss at WT:Citing sources. If you don't want to discuss citations in general, only the ones produced by Citation Style, please discuss at Help talk:Citation style 1. Jc3s5h (talk) 04:45, 5 July 2017 (UTC)
 * If the idea is to follow a full stop with an uppercase letter, doesn't that cs1 example break down a bit when " " becomes ""? —BarrelProof (talk) 02:30, 5 July 2017 (UTC)
 * MOS and MOSNUM cover the body of the article, infoboxes, etc. If you want to discuss citations, please discuss at WT:Citing sources. If you don't want to discuss citations in general, only the ones produced by Citation Style, please discuss at Help talk:Citation style 1. Jc3s5h (talk) 04:45, 5 July 2017 (UTC)


 * How does the example break down? We would expect "... blah blah blah. Vol. 1, blah blah blah ..." to have "Vol." because it follows "." as terminal punctuation.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  04:24, 6 July 2017 (UTC)


 * Grammar point The phrases "Volume 3" or "Page 74" are proper noun phrases in English – for example, they cannot be preceded by a determiner (you say Look at Volume 3 not Look at the volume 3). The equivalent common noun phrase uses the ordinal (Look at the third volume). Traditionally, proper noun phrases are capitalized, although the modern trend seems to be to de-capitalize almost everything except proper nouns and noun phrases used purely as labels rather than for any descriptive element. Peter coxhead (talk) 06:36, 5 July 2017 (UTC)
 * Does that point apply if we have abbreviations? And to my eye Page 74 looks stranger than Volume 3. But that's just my unsourced subjective UK opinion, of course. Martinevans123 (talk) 07:37, 5 July 2017 (UTC)
 * it may be an age thing (I'm British too). I was only making a grammar point. There's never been a 1:1 relationship between being grammatically a proper noun or noun phrase and being capitalized. Once there was much more capitalization (see e.g. the 1662 Prayer Book); then there was a trend to capitalize only proper nouns and noun phrases (I was taught to write "in Chapter 3" or "on Page 4" as opposed to "in this chapter" or "on the following page", and I taught the postgraduate students I supervised to do the same in their theses); now there's a trend to decapitalize further. Who am I to stand against such trends? :-) Peter coxhead (talk) 14:36, 5 July 2017 (UTC)
 * I'm happy to admit that the 1662 Prayer Book was a little before my time. :-) Martinevans123 (talk) 15:33, 5 July 2017 (UTC)
 * That's just one philosophical approach to the meaning of "proper name" and it's highly dependent on language (some languages would use a definite article there, but it doesn't change the underlying nature of the term "vol."/"volume" or its meaning in any way). For purposes of style discussions. "proper name" and "proper noun" are essentially interchangeable.  The "Mixed" argument presented by Eppstein and others is cognizant of this; the "Vol." in The Art of Computer Programming, Vol. III: Sorting and Searching should be capitalized because it's part of the title (or subtitle, more specifically), and thus part of a proper name (proper noun phrase). Lots of things do not take a definite article in English by convention, and are not capitalized, even if some philosophers want to classify them as proper names. E.g., "entry 2349499393 in my database", "See example [number] 32 in the list below", "I cracked thoracic vertebra 4" (or "I cracked vertebra T-4"), "at 6 o'clock", etc., etc.  Most of these can be replaced by ordinals, at least when given numeric designations; that breaks when they aren't ("See example F in the list below").  No doubt some would write "Example F", but few would do this upper-casing to all the examples here. It's no longer conventional.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  04:24, 6 July 2017 (UTC)


 * Some languages, eh? I suggest we stick to English for now. Martinevans123 (talk) 07:30, 6 July 2017 (UTC)
 * Missed the point. If different languages treat the same construction with or without a definite article, then the presence or absence of the definite article cannot logically determine whether something is or is not a proper name, since that concept isn't language-dependent (Pacific Ocean and Stephen Hawking do not magically become common nouns in some languages, and milk and eggs don't transmogrify into proper names in any, either). You'll find article variance across many languages (e.g., "death", "life", "war", and the like as concepts, states, or processes are without an article in English, but require a definite one in many languages: der Tod, la vida, le guerre. Same goes for presence or absence of capitalization; e.g., nouns are capitalized in German, but Katze ('cat') is not a proper name (or proper noun) in German any more than "cat" is in English.  PS: Your renewed habit of stalking the MoS talk pages to pop in with sarcastic, non sequitur one-liners, to show us how clever and above it all you feel, is not helpful.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  23:24, 6 July 2017 (UTC)
 * I'm sorry if you think commenting on a topic of interest to me for a page that is on my Watchlist is "stalking". I'm sorry if you think "one-liners" are in some way worthless. I'm very sorry is you feel any of my genuine observations here are "sarcastic". I certainly don't feel "clever and above it all"; perhaps you know better than me how I feel? I have avoided making any personal attacks against you and I'd respectfully request that you do the same for me. But I'm not sorry to reiterate my point that I think your observations about other languages here have rather limited value. Martinevans123 (talk) 12:45, 7 July 2017 (UTC) p.s. your reply was not helpful.
 * Re: "I think your observations about other languages here have rather limited value" – Then you still didn't understand the point; I suppose I must have explained it poorly, but the discussion has moved past this and I don't think a re-re-explanation is needed. (As for the side matter: I'm not raising any issue with your normal engagement in these discussions, just your periodic habit of dotting them with one-liner snarky commentary that appears to be done for your own amusement and which doesn't further the discussion. These tend to come in bunches, e.g. this other recent "gadfly" comment.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  22:55, 9 July 2017 (UTC)
 * You say If different languages treat the same construction with or without a definite article, then the presence or absence of the definite article cannot logically determine whether something is or is not a proper name, since that concept isn't language-dependent.. But I think your argument is fallacious, as it denies the possibility that the logic that can be applied to rules of grammatical construction can be language dependent. Just because different rules apply in other languages, these don't invalidate general rules that may apply to written English. We wouldn't expect the grammatical rules that apply to written Jaopanese, or Hindi, or Telugu, to apply to English, would we? And please don't characterise my contributions as those that one might expect from an insect in the order Diptera. Please do not feel obliged to reply. Just look for a chowrie. Martinevans123 (talk) 21:30, 10 July 2017 (UTC)
 * Very very definitely not getting it. The very fact that grammar (broadly defined) is language-dependent is to the problem of trying to pin a broad philosophical concept on an accident of English orthography in particular.  I'm not saying anything at all about the grammatical nature of proper nouns and adjectives in English, much less applying it to other languages; we already know that this varies by language; español in Spanish itself is lower-case. Which again comes right back to the central point: if things like capitalization and use or absence of a definite article are not cross-linguistic constants, then they cannot be used to determine a cross-cultural philosophical absolute like "what really is a proper name?"  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  03:10, 11 July 2017 (UTC)
 * Whether or not there is a valid notion of "proper noun" that transcends individual languages is an interesting question. But I don't see how the answer to this question informs our knowledge of how words should be capitalised in English. The way in which some specific languages use capital letters might throw light on capitalisation in English, I guess, at least from a historical perspective. But this doesn't seem to be your point. If another editor can explain your argument to me in a way I could understand I'd be more than happy to hear from them. As no-one else has taken issue with you over this point, I can only assume they don't see it as important, assuming they do understand what you mean. Martinevans123 (talk) 07:41, 11 July 2017 (UTC)


 * Mixed as with 's comments. I think the proper nouns can be used for the citation section and for the title if it is part of the title as with the Vol. 3: Sorting and Searching, but that when referring to "vol. 3, no. 15" in the text body, "volume 3, issue number 15", "part 2", or "chapter 5", it can be lower-case. This is comparable to cite episode template where they use "Season 1. Episode 15". in the citation but refer to season 1, episode 15 in the text body. With Episode, however, it's in capitals because the cite episode can be done without specifying a season number for television series that has a single season. <strong style="color:#606060;">AngusWOOF ( bark  •  sniff ) 16:24, 5 July 2017 (UTC)


 * Mixed: Citations, including parenthetical references, should continue to follow WP:CITESTYLE/WP:CITEVAR. I do not believe there is a compelling reason to have a policy for citations that would force capitalization in citations to go against the citation style being used (unlike WP:IBID). For example, the Bluebook citation style used by the US legal profession doesn't use a full stop until the end of a citation, but uses capitalized "No." in certain cases where the docket number of a case needs to be included, eg. "Matal v. Tam, No. 15-1293, slip op. at 5 (June 19, 2017)." As for prose, they should be capitalized as part of a title or at the start of a sentence. If used as a proper noun, except as a title or in a quote, then the word should be written, eg. "Smith's writing in volume 3 was regarded by his contemporary Johnson as 'abysmal.'" In other cases, they should be lowercase. AHeneen (talk) 17:06, 5 July 2017 (UTC)


 * Update: Additional sources (UK, Australian, Canadian):


 * .uk:
 * New Hart's Rules (2nd ed.) gives "no.", "nos.", "vol.", "vols.", with that punctuation, though it advises avoiding them entirely in reference citations.
 * The original edition (under various titles) is consistent with this on "vol.", but eschews "no." or "No." entirely (except as part of a proper name such as a work title). It recommends the № symbol only for French. Its sections "Lower case abbreviations" and "e.g., i.e., etc." (which covers similar constructions like "viz." and "et al.") both consistently give Latin and bibliographic abbreviations in lower-case form, with points; there is no basis on which to suppose this would not apply to "no.", which is just one of numerous such cases.
 * Fowler's (4th ed.) has no entry for either; it inconsistently uses both "vols." and "vols" on the same bibliography page (and exhibits other editing problems throughout). This edition cites New Hart's (2nd) in hundreds of entries, so a resonable presumption is that it would follow that work on these matters. Its advice on the full point is that it should be retained when ambiguity would result from dropping it (clearly the case with "no."/"No."), and that a consistent approach should be used (thus, if using "no.", then also use "nos.", "vol.", "vols.").
 * The previous edition likewise lacks entries, but consistently uses "vols." in its own bibliography. Its advice on full points and consistency matches that of the current edition.
 * The Oxford Dictionary for Writers and Editors gives both "No." and "no.", plus "vols." (not "Vols.), without addressing "vol." Does not advise use of the № symbol except in French material.


 * .au:
 * The Australian government's Style Manual for Authors, Editors and Printers uses "no.", except when the string appears in a proper name such as the title of a work (an exception all style guides would make if they bothered commenting on it).


 * .ca:
 * Editing Canadian English by the Editors' Association of Canada uses "no." and "vol." (except where capitalization would normally occur, e.g. at the beginning of a sentence).
 * The government publication The Canadian Style uses "vol." and "vols." but "No." and "Nos." The rationale offered for this is that "No." is a symbol. However, that applies to №, the actual numero sign. The "no." form is a normal abbreviated Latinism, of the same character as "etc." and "q.v."  We similarly distinguish et ... Latin abbreviations like "etc." and "et al." from the et symbol, & (ampersand).
 * A Canadian Writer's Reference cites the government guide as a source and follows it on most matters, including this one.


 * Other:
 * The UPI Stylebook uses "No." No rationale is offered, and its section on abbreviations and acronyms is internally inconsistent.


 * The combined two blocks of sourcing I've provided here (plus that done yesterday in the footnotes at Exempli gratia) indicate an overwhelming preference for "vol." and "no." (lower-case and punctuated) in professionally published material today, regardless of English variety. MoS should follow this practice; it is not just better supported in the real world, it's also more consistent with our treatment of other abbreviations and other Latinisms, and is less ambiguous (at least in the case of "no"/"No").  Style guides in favor of "No." and "Vol." are mostly one of: only addressing a particular citation format, not recent, of low to middling reputability, or some particular entity's house style. Those in favor of "No", "no", "vol", etc., are primarily the house stylesheets of particular news publishers. The main guides for writing fairly formal English – Chicago Manual and New Hart's, on which MoS is largely built – both use "no.", "nos.", "vol.", "vols." Again, this would not lowercase either abbreviation in a) the title of a work or b) a citation in a format that capitalizes either or both. These are the concerns the Mixed comments above have in mind.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  22:00, 11 July 2017 (UTC)

Just use №. — Jon C.  ॐ  10:21, 12 July 2017 (UTC)

Related matters: dropping the dot, and MOSABBR
Some side issues: Despite MOS and WP:MOSABBR discouraging dropping of dots (points, periods, stops) from abbreviations except where utterly conventional and non-ambiguous (e.g., "Dr Smith" and "St Stephen" in British ENGVAR), I keep running into "No 1" in various articles (e.g. here). This is intolerably ambiguous, especially for users of screen readers, though it produces confusing gibberish for everyone, like "No of discs" in infoboxes. Worst of all (so far): table header labeled "No". Also saw a "No total" in one of these. We need to explicitly state that No. (whether we keep that capitalized or not) must retain the dot and is not an ENGVAR matter (sourcing above demonstrates this; both New Hart's and Fowler's retain the dots).

Second, neither vol. nor no. appear at MOS:ABBR but of course should be listed, since we use them frequently. I'll wait until the discussion above concludes before adding them, so that what gets added doesn't have to be changed later. — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  02:55, 11 July 2017 (UTC) PS: Ha ha, I rickrolled Wikipedia.


 * If you keep running into No 1 in various articles, why don't you try frequenting articles where there are people around? Then you wouldn't be so lonely. At Christmastime, try the No L articles.  E Eng  03:09, 11 July 2017 (UTC)
 * LOL.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  03:36, 11 July 2017 (UTC)
 * I do the best I can with the material available.  E Eng  03:40, 11 July 2017 (UTC)`
 * As a screen reader user, it doesn't really bother me that much that the full stop is omitted. "no." is read as no with a pause after it, rather than "number", and all educated blind people should know what "no." stands for. (IIRC the local talking book catalogue says things like "Book no. 12345".) Even without the full stop I can figure out what's meant. Yes it's less ambiguous and more correct with the full stop, but I've never thought about it before as an accessibility issue. Graham 87 07:38, 11 July 2017 (UTC)
 * Good to know, though I'm surprised that's still true today; I would think that "No."or "no.", followed by zero spaces or a single space, then a numeral or string of numerals or a spelled out number word, would be read as "number". Maybe the logic to pick it out is more complicated by false positives than I imagine.  Even if so, it seem that more cases would be correct that false matches, and that this would be preferable to every case being read as the word "no".  Hmm.  Well, I've never tried writing a screen reader so I guess this is kind of akin to an "I am no a lawyer, but ..." musing.   — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  10:09, 14 July 2017 (UTC)

RfC: Red links in infoboxes
Opinions are needed on the following matter: Wikipedia talk:Manual of Style/Infoboxes. A WP:Permalink for it is here. Flyer22 Reborn (talk) 09:31, 22 July 2017 (UTC)

Disagreement on comma punctuation when parentheses occur in a sentence
Users Adûnâi and Dank are in disagreement with me over how MOS:COMMA should be applied.

It started when I pointed out an error (my opinion) on the Main Page, adding this remark:


 * Per WP:Copyedit#Punctuation, a comma is missing in "The British North America Act of July 1, 1867 (now celebrated as Canada Day) united the colonies of ... ". A comma is required after mdy dates, unless it's followed by other punctuation, for instance period/full stop at the end of sentence. Parentheses are not punctuation, and so there should be a comma after the end of the parentheses.

Adûnâi disagreed, writing:


 * Wouldn't it be a comma between a subject and a verb?

Dank agreed with Adûnâi:


 * That comma rule is overridden by comma conventions that avoid a comma both before and after those parentheses.

User Khajidha agreed with me:


 * What conventions are those? By what I've always been taught, User:Handsome Fella is correct.

The discussion was removed as a new day arrived.

I then contacted both editors on their talkpages, but was unable to convince them, despite pointing out the explicit "Burke and Wills" example:


 * Do not be fooled by other punctuation, which can distract from the need for a comma, especially when it collides with a bracket or parenthesis, as in this example:
 * {| style="background:transparent"


 * : || Burke and Wills, fed by locals (on beans, fish, and ngardu) survived for a few months.
 * :   || Burke and Wills, fed by locals (on beans, fish, and ngardu), survived for a few months.
 * }
 * :   || Burke and Wills, fed by locals (on beans, fish, and ngardu), survived for a few months.
 * }

Adûnâi claimed this was "a completely unrelated example", while Dank re-iterated that "a comma separates each element and follows the last element unless followed by other punctuation", ignoring the explicit part I had pointed out: "Do not be fooled by other punctuation, which can distract from the need for a comma, especially when it collides with a bracket or parenthesis, as in this example (which appears above the "Burke and Wills" example).

Let's look at another example:


 * In the article Mary Jo Kopechne, this sentence is found in the first section: "Her father, Joseph Kopechne, was an insurance salesman, and her mother, Gwen (née Jennings), was a homemaker".

In it, "and her mother, Gwen, was a homemaker" is interrupted by the parenthetical expression "(née Jennings)", resulting in the above sentence. Do Adûnâi and/or Dank mean that the comma after "Gwen" should be removed? I don't know. I have not understood how either of them explains why the comma shouldn't be there, which is why I'm starting this discussion.

Could someone please clarify this?

Thanks.

HandsomeFella (talk) 12:12, 19 July 2017 (UTC)


 * I completely agree with all of your examples here. But they are different than the sentences in question. On my talk page, you wrote this.


 * The British North America Act of July 1, 1867[,] united the colonies of ...


 * In the current version of the article, there are three points which are comparable (searching by "July 1, 1867"). The third one looks like the following.


 * The enactment of the British North America Act, 1867 (today called the Constitution Act, 1867), which confederated Canada, was celebrated on July 1, 1867[,] with the ringing of the bells...


 * I'm not sure about the comma because I haven't seen a clear example that proves it should always follow dates such as July 1, 1867.--Adûnâi (talk) 12:49, 19 July 2017 (UTC)


 * Luckily it's in the MOS, right under the Burke Willis example. Primergrey (talk) 13:27, 19 July 2017 (UTC)


 * Ok, it finally dawned on me what Adûnâi means. He means (I think) that "[The] British North America Act of July 1, 1867" is the subject, and that the comma in question would separate the grammatical subject from the verb (right?). Ok, I guess that could be argued – it could also be discussed – but when in a full sentence, punctuation rules still apply, in my opinion. So we still disagree. HandsomeFella (talk) 13:42, 19 July 2017 (UTC)
 * It would also mean that the comma is superfluous even when there's no parenthetical. I think that's wrong too. HandsomeFella (talk) 14:03, 19 July 2017 (UTC)

I'm going to guess from the lack of response that people are happy with our understanding of the MOS guideline. Please ping me if not. - Dank (push to talk) 21:11, 22 July 2017 (UTC)


 *  What MOS guideline are you referring to? If you mean what you wrote above,


 * That comma rule is overridden by comma conventions that avoid a comma both before and after those parentheses.


 * Where is this taken from? I never heard anything like this. Though I know some editors think the comma after the year in m-d-y dates is optional, traditionally it is required, and I think it often prevents confusion, so I use it, and it does belong after the parenthetical phrase. One way to avoid having to place a comma after a parenthesis (which doesn't bother me, but apparently bothers some) is to remove the parentheses around "now celebrated as Canada Day" and instead use a pair of commas or a pair of en-dashes:


 * The British North America Act of July 1, 1867, now celebrated as Canada Day, united the colonies of ...


 * The British North America Act of July 1, 1867 – now celebrated as Canada Day – united the colonies of ...


 * As an additional note, I think I was told that parentheses slow down readers who use screen readers more than other punctuation. – Corinne (talk) 21:40, 22 July 2017 (UTC)


 * WP:MOS says "Dates in month–day–year format require a comma after the day, as well as after the year, unless followed by other punctuation." Agree, disagree? That word "Dates" is linked to this section at WP:MOSNUM, which gives an example: "Everyone remembers July 21, 1969 – when man first landed on the Moon". The example illustrates that that sentence from MOS is meant literally. This corresponds to what I've seen at FAC. - Dank (push to talk) 21:58, 22 July 2017 (UTC)

The comma does not belong this time. Yes, we have a rule for "July 1, 1867, ..." with two commas (so do other style guides). Commas that serve a parenthetical purpose are always in pairs, unless replaced by: I.e. we don't use pointless double punctuation, like a comma before a period/stop or a dash.
 * Terminal punctuation: "It happened on July 1, 1867."
 * Other punctuation that introduces another parenthetical that pertains to the surrounding material not to the first parenthetical: "The murder happened on July 1, 1867 – the day my great-great-great-grandmother was born." The parenthetical introduced by the dash pertains to the entire subject, not to "1867".

Much of the above has been red-herring discussion. Yes, this case differs from the guideline's and the Kopechne article's examples. The beginning of "(now celebrated as Canada Day)" is a "(" introducing another parenthetical, which applies to the subject, not to the year, so it cancels the need for a comma to balance the first one in ", 1867". Without the date, this would read "The British North America Act (now celebrated as Canada Day) united the colonies of ...", with no commas. Contrast the other examples: "her mother, Gwen (née Jennings)," has a parenthetical that applies to the material in another parenthetical (the ", Gwen ...," structure) and is part of it, all of it ultimately referring to "her mother". Without the second parenthetical, it would read "her mother, Gwen, ..." with both commas. Without the first parenthetical, it is a non-sequitur: "her mother (née Jennings)"; makes little sense without the first name also included.  "Burke and Wills, fed by locals (on beans, fish, and ngardu), survived for a few months" is essentially the same as the mother example, and would read "Burke and Willis, fed by locals, survived ..." without the second parenthetical clause, which pertains to the feeding not to B&W. Without the first parenthetical, it is again broken: "Burke and willis (on beans, fish, and ngardu) survived ...". Both are broken when their first parentheticals are removed because the second parenthetical is an internal reference to the first, not to the subject directly. [Aside: Both could be unbroken with a rewrite if information was genuinely missing: "her mother, née Jennings, first name unknown," and "Burke and Willis survived for a few months on beans, etc., but it is not known if they locals provided this food or they gathered it themselves."] In the case at issue, the holiday name pertains to the entire subject, not to the "1867" fragment, and it works regardless what you do to it, because the parentheticals are independent of each other: "The British North America Act of July 1, 1867, united ...; and "The British North America Act (now celebrated as Canada Day) united ...".

Another way of putting it: The Canada thing could very reasonably be rewritten as "The British North America Act of July 1, 1867 – now celebrated as Canada Day – united ...". But no sane writer would do "... her mother, Gwen – née Jennings ...", or "Burke and Wills, fed by locals – on beans, fish, and ngardu ...". They're a different kind of structure from the Canada case.

PS: Technically, the "1867" is not a parenthetical, but is conventionally formatted as one. The nature of this kind of confusion/dispute is one of the reasons (source: Garner's Modern American Usage) that another pseudo-parenthetical, the "Firstname Lastname, Jr, ..." style, has gone out of fashion, in favor of "Firstname Lastname Jr ...". It's also probably (my speculation) got something to do with why "July 1, 1867, ..." date formatting has been abandoned in most of the world, when it was neck-and-neck with "1 July 1867" style in the 19th century (source: Google N-Grams). 05:04, 28 July 2017 (UTC) — Preceding unsigned comment added by SMcCandlish (talk • contribs)

Bolding components of article titles
Hello everyone, I stumbled across Special:Contributions/5.107.140.121, who has made numerous edits of this form. Am I correct in reverting these, or am I missing something? . Tazerdadog (talk) 19:12, 22 July 2017 (UTC)
 * I would say your reverts are correct. Blueboar (talk) 19:25, 22 July 2017 (UTC)
 * Definitely. We have a rule about this somewhere, but I didn't find a version of it in MOS:ABBR, so I added it. People will expect that page to have every acronyms-related rule in it, and we can't figure that people are going to dig up out of some other page.  I see the anon doing those "treat our readers like morons" edits has received a pack of warnings about inappropriate editing.  And did at least one after the warnings, but it was a day ago and maybe too stale for action.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  06:52, 28 July 2017 (UTC)

WP:NBSP
The above redirect points to a very short section here. That section seems to be missing links to where we would otherwise use non-breaking spaces (at least a reference to MOSNUM). Maybe it would be a good idea to add some pointer links? --Izno (talk) 11:56, 25 July 2017 (UTC)
 * MOSNUM itself keeps pointing to another page for more info on non-breaking spaces (or nonbreaking, whatever spelling you like; I note that MOSNUM is not using a consistent spelling of it, either).  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  06:54, 28 July 2017 (UTC)

non-breaking space before per cent/percent
The MoS asks for a space before per cent/percent, but does not make clear whether this should be non-breaking. Do you know? – Sb2001 talk page 14:07, 18 July 2017 (UTC)
 * Clarified . Reverted myself. I must be losing my mind.  E Eng  16:08, 18 July 2017 (UTC)
 * EEng, I read the question as asking whether the space before the words "per cent" or "percent" should be non-breaking, that is, whether one should write  or  . (IMHO a nbsp might be a good idea.) — Stanning (talk) 17:11, 18 July 2017 (UTC)
 * Yes, User:Stanning: that is exactly what I meant. Although, User:EEng: your edit was still more than worthwhile, and does prevent ambiguity. – Sb2001 talk page 17:17, 18 July 2017 (UTC)
 * IMO, there should be a non-breaking space between a numeral and the first word of a unit or "percent/"per cent", eg. "133nbspper cent" or "348nbspsquare kilometers" but not "more than onenbsphundrednbspsquarenbspkilometers", which is too long for no breaks (keep in mind that some users read WP articles on cell phones with small screens!). This would be consistent with MOS:NBSP (discussed below). AHeneen (talk) 18:09, 18 July 2017 (UTC)
 * [edit conflict] Actually, the MoS says NOT to use a space before the percent sign. Before the previous edit, it said: Write 3%, three percent, or three per cent, but not 3 % (with a space) or three %. "Percent" is American usage, and "per cent" is British usage (see § National varieties of English, above). In ranges of percentages written with an en dash, write only a single percent sign: 3–14%. AHeneen (talk) 17:14, 18 July 2017 (UTC)
 * OK, now that I've had some coffee... I would say regular space, not nbsp, by analogy to the fact that we code a regular space in 5 kilograms but nbsp in 5kg. See the table at WP:Manual_of_Style/Dates_and_numbers, in the section of the table labeled Numeric values.  E Eng  17:30, 18 July 2017 (UTC)
 * Yes - that does make sense. – Sb2001 talk page 17:58, 18 July 2017 (UTC)
 * That page only mentions an NBSP before a unit symbol: "Use a nonbreaking space (nbsp or ) between a number and a unit symbol, or use nowrap". MOS:NBSP says "It is desirable to prevent line breaks where breaking across lines might be confusing or awkward. ... Whether a non-breaking space is appropriate depends on context: whereas it is appropriate to use   in prose, it may be counterproductive in a table (where horizontal space is precious) and unnecessary in a short parameter value in an infobox (where a break would never occur anyway)." I think usage of an NBSP before percent or per cent should follow that guideline. I don't have time right now, but talk page archives should be checked to make sure this issue hasn't been discussed before. AHeneen (talk) 18:09, 18 July 2017 (UTC)
 * I'd say % is analogous to a unit symbol, while per cent and percent are analogous to a unit name. Coding  creates an overlong unbreakable string.  E  Eng  19:37, 22 July 2017 (UTC)

My take on this: 1. There is little need for site-wide consistency on this. If some cases break "5 percent" and other cases don't, that's really ok. Nobody will notice except a few obsessive Wikipedia editors. We are here to serve our readers' needs, not our own. WP:CREEP. 2. Where there is disagreement in a specific case, MoS is not a substitute for discussion and consensus, the cornerstones of Wikipedia editing. &#8213; Mandruss  &#9742;  04:39, 28 July 2017 (UTC)
 * Any time MoS is advising that a numeric figure be followed by a space and a something, it means a non-breaking one.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  04:02, 28 July 2017 (UTC) I was wrong on this; we only require it when what follows the number is a symbol or abbreviation (50 dB, 4:29 p.m., 1252 BCE, etc.).  Using a nonbreaking space it is harmless when what follows is a regular word that is a unit or something comparable.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  06:16, 28 July 2017 (UTC) Corrected 06:18, 28 July 2017 (UTC)
 * I take it you feel it's reasonable to expect editors to somehow intuit that? If not, you feel it's reasonable to educate the community about that one misled editor at a time, forever? Besides, how do you know what MoS intent is?
 * Not sure what you're on about. I am was agreeing that the page needs to be updated to say to use a nonbreaking space, not just a space.  I said nothing about anyone intuiting anything, or doing anything editor by editor.  There is a need for site-wide consistency on this, or we would have no nonbreaking space rules of any kind at all; we use them because it keeps the digit meaningfully grouped with what it refers to. [Correction: The rule is narrower than I remembered, and applies to symbols/abbreviations only.]  If every other time you post to an MoS talk page you cannot refrain from aspersion-casting about other editors' mentalities, you need to stop posting to them (really). There is no disagreement on this; MOS:NUM applies this specific spacing style to all such cases.  If you think the nonbreaking space rules are WP:CREEP, go start an RfC at WT:MOSNUM. I tend to agree, but while they are the rule, they are, and we should do it consistently.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  06:07, 28 July 2017 (UTC) Corrected 06:18, 28 July 2017 (UTC)
 * I'm not taking the bait, but I would be very interested to see your evidence - on my talk page - any evidence at all - that I cast aspersions "every other time [I] post to an MoS talk page" (speaking of aspersion-casting). &#8213; Mandruss  &#9742;  06:26, 28 July 2017 (UTC)
 * I'm being hyperbolic about the frequency, of course. Please don't label MoS editors, and editors at articles doing MoS compliance, as "obsessive". It's a direct insult and an aspersion about their mentality. Just don't.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  06:58, 28 July 2017 (UTC)
 * Ah, so "every other time" is your "hyperbole" for "once". Wow. Ok. Anyway, as much as I'd love to debate the point with you, it would be too tangential and distracting, as well as being pointless. My comments stand, and I wouldn't make a fuss if somebody collapsed the whole thing. &#8213; Mandruss  &#9742;  07:08, 28 July 2017 (UTC)
 * Anyway, what SM said isn't true. We don't call for an nbsp in .  E  Eng  04:57, 28 July 2017 (UTC)
 * I stand corrected. I thought MOSNUM was explicit that all numeric measurements/values are followed by a nonbreaking space then the unit name or symbol or whatever; it's only symbols and abbreviations.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  06:07, 28 July 2017 (UTC)

Blogs and italics
Should the names of blogs be italicised? Most of the articles for entries on List of blogs do not have italic titles (although some do). I was about to italicise the title of a blog article but was quite surprised to find this was not the general practice. <b style="background:#FAFAD2;color:#C08000">Spinning</b><b style="color:#4840A0">Spark</b> 17:17, 19 July 2017 (UTC)
 * From MOS:ITALICTITLE:

"Website titles may or may not be italicized depending on the type of site and what kind of content it features. Online magazines, newspapers, and news sites with original content should generally be italicized (Salon or HuffPost). Online encyclopedias and dictionaries should also be italicized (Scholarpedia or Merriam-Webster Online). Other types of websites should be decided on a case-by-case basis."


 * I think the titles of blogs should be italicized. Be sure the article is about the blog and not primarily about a company who publishes a blog of the same name. For example (I could only think of examples of websites): BuzzFeed (about the company) and Politico (also about company) versus The New York Times (about newspaper/website, as company has separate article: The New York Times Company). AHeneen (talk) 01:29, 21 July 2017 (UTC)


 * Yes, they should be italicized, unless they're part of a larger work, in which case quotation marks. The same rules apply regardless of medium: major, stand-alone works get italics, minor works and sub-works get quotes. The New York Times doesn't become The New York Times or "The New York Times" if you're referring to an online article instead of one printed on dead trees. Our citation formatting templates (and various external citation style) are going to apply italics to titles of blogs as the work or website or journal parameter value, so deciding on your own not to use italics is going to cause a conflicting style in the same article. If you're dealing with something like a newspaper's official blog, use Newspaper Title "Blog Title" (official blog) Article title; resulting in a cite like:  Publications get italics except when referred to as a business (use the company name), or a computer system (no styling, use domain name): "He writes for Salon"; "She took over as the CEO of Salon Media Group"; "The DDoS attack also affected www.salon.com for two hours." For online publications that use their domain name as their title, complete with the ".com" (or whatever), it still goes in italics when discussed as a publication rather than a legal entity or a stack of hardware. This, too, is what our cite system will do, so just deal with it and don't introduce a conflict style, please. :-) PS: Abusing parameters of the cite templates to get an alternative style (e.g. using at for the work title) will emit incorrect COinS metadata and is to be avoided. Anyone may revert/fix that on sight.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  05:12, 28 July 2017 (UTC)
 * I think I would prefer to see the blog in the work parameter (if cited at all) since it is also a major work of many articles of content. But this is a quibble with your comment. --Izno (talk) 12:47, 28 July 2017 (UTC)

"A UART vs "An UART"
I thought that we have an existing rule about this, but cannot find it.

Is "A UART or "An UART" correct? Likewise with HTML, ATM, etc? What is the general rule? --Guy Macon (talk) 19:59, 28 July 2017 (UTC)
 * Follow the pronunciation. Anyway, UARTS are so 70s.  E Eng  20:22, 28 July 2017 (UTC)


 * Most people who work with microcontrollers pronounce it "Yoo-Art" but a lot of people who work with the telephone system pronounce it "Ooo-Art", but there is a lot of overlap. Regional dialect differences?
 * https://books.google.com/books?id=I9FtDQAAQBAJ&pg=PA198&lpg=PA198&dq=%22an+UART%22+-%22a+UART%22&source=bl&ots=YDN08a39BK&sig=CBhJLVIucLv8Fd4h9o67D9W0r70&hl=en&sa=X&ved=0ahUKEwj_iKG9gq3VAhUpjlQKHfoHDn8Q6AEIcDAM#v=onepage&q=%22an%20UART%22
 * https://www.microchip.com/forums/m578823.aspx
 * https://www.ieeexplore.ieee.org/document/7342320/
 * https://forums.xilinx.com/t5/Welcome-Join/adding-an-uart-on-zc702-with-PMOD2/td-p/752074
 * https://stackoverflow.com/questions/40607345/how-to-communicate-with-an-uart
 * --Guy Macon (talk) 22:33, 28 July 2017 (UTC)

Intrinsic-Harmony.jpg If you combined an op amp with a UART, would it be op art? – E Eng 


 * Ooo-art? I give up.You're on your own.  E Eng  22:50, 28 July 2017 (UTC)
 * Pro tip: If you ever have a problem with someone using a voodoo curse of you, just use a standard hex inverter. -Guy Macon (talk) 13:37, 30 July 2017 (UTC)
 * I'd have to discount evidence from newbies on question forums, but there are clearly differences in practice. I have to say from personal experience since around 1980 dealing with relevant tech from the computer side (I never got a lot of exposure to the telecom side), I've never heard it pronounced as "ooo-art". If I were trying to make an argument, I might try looking at Google n-grams, but it claims not to be able to find instances of "an uart". Or I could say that, absent solid evidence to the contrary, the leading word of the acronym should govern: "universal". But I'm glad I asked the reasoning behind his edit; this has been enlightening!  &mdash; jmcgnh  (talk)  (contribs)  23:13, 28 July 2017 (UTC)


 * Saw these online which says if the U is a Y sound then treat it like a consonant.    This Merriam-Webster one talks about acronyms too:  <strong style="color:#606060;">AngusWOOF  ( bark  •  sniff ) 02:09, 29 July 2017 (UTC)


 * I think they might pronounce it as Ooo-Art in the West Country, where brand new combine harvesters come from. (Not really.)  nagual  design   13:10, 29 July 2017 (UTC)

Based on the feedback above, I am going to use "a UART" from now on. The interesting aspect here is determining how most people pronounce an acronym/initialism. I can look around me and see that most people I talk to pronounce "FAQ" as "fack" and not "eff-ay-cue" and that most people I talk to pronounce "RfC" as "Arr-Eff-Cee" and not "Refsee", but is this universal among English dialects? I recently worked with some Australians, and when they read something like "E-Z-Off" they read it as "Eee-Zed-Off". And what pronunciation should I assume for "FUQ (Frequently Unanswered Questions"? --Guy Macon (talk) 13:37, 30 July 2017 (UTC)
 * I'll never tell.  E Eng  14:08, 30 July 2017 (UTC)


 * Back when, the university computer lab people tended to spell it out F-A-Q, while the dialup crowd used /fak/ a lot. The Jargon File gives both, prefers the spelled-out version, and dates to the era. "RFC" or "RfC" is usually spelled out; the /ref-see/ thing is uncommon (barely attested in N-grams).  N-grams show a spike in "a FAQ" usage, but it fell off, and usage is now stably parallel. Because of the FUQ thing, I'd recommended going with the spelled-out version, thus "an FAQ".  Another option would be to use "a FAQ" except in a) technical contexts, or b) in the same article as FUQ appearing, since "a FAQ" appears to be more common outside geeky contexts, though no longer by much of a margin. Also agree it's "a UART"; the /oo-art/ pronunciation seems to be poorly attested.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  23:06, 30 July 2017 (UTC)

non-breaking space after page
I recently started placing nbsps after 'p.' and before the number. It then became clear that the citation insert tool does not do this. Other editors do not, either. Is it A/ bad/something to go back to undo, B/ something which is harmless, but unnecessary or C/ something which we should all be doing? – Sb2001 talk page 11:40, 28 July 2017 (UTC)
 * Inside a citation template, A: this corrupts the citation metadata emitted. Other places, probably B (unless we have guidance otherwise at present). --Izno (talk) 12:42, 28 July 2017 (UTC)
 * For the last two years I've been routinely adding a non-breaking space template nbsp after "p." (only where the line is long and might break after "p.") in references formatted with ...< /ref>, and the html non-breaking space if the reference is in the cite ref format with curly brackets (but EEng commented recently that the template nbsp should not mess up even that kind of citation), and no one has ever said it messes up anything.  – Corinne (talk) 13:33, 28 July 2017 (UTC)
 * What I've said for many years is that if the citation templates can't handle some of the ways of inserting the same thing (i.e. literal nbsp vs. &amp;nbsp; vs. nbsp) then the priests that control those templates should fix them, so that editors don't have to remember a bunch of extra rules for no apparent reason.  E Eng  16:27, 28 July 2017 (UTC)
 * You should not be using "p." at all within the citation templates; they add that themselves (with page instead of pages if you want p. instead of pp.) If you're formatting citations yourself instead of using a template, I would go somewhere between C and B: I think it's preferable to use a non-breaking space, but not actually harmful not to. —David Eppstein (talk) 21:29, 28 July 2017 (UTC)
 *  Since you say the citation templates are formed automatically, does the template somehow contain a hidden no-break space after "p." and "pp." – and, if it doesn't, could it be added? Doesn't it make sense to do so? It would save a lot of adding of no-break spaces. I think everyone would agree it would be preferable to avoid having "p." or "pp." end up alone at the end of a line. – Corinne (talk) 15:40, 29 July 2017 (UTC)
 * Those cs1|2 templates that prefix pagination with p. or pp., insert a  character between the prefix and the first page number.  It has been this way for a very long time.
 * Not all cs1|2 templates render pagination with the p. or pp. prefix. If you are writing p.&amp;nbsp;... or pp.&amp;nbsp;... in the cs1|2 templates that render page numbers with a colon prefix,  and  (when journal is set), then you should stop doing that.  Typical professional, academic, or scientific journals usually do not use p. or pp. prefixes so  and  follow that style.
 * —Trappist the monk (talk) 16:23, 29 July 2017 (UTC)
 * I never add a non-breaking space after "p." or "pp." if an equals sign or a colon follows either of those, and I usually don't touch any ref that's within curly brackets. It's mainly in the citations that use the ...< /ref> format that I would add the non-breaking space template. Is that safe to do? – Corinne (talk) 16:44, 29 July 2017 (UTC)
 * This is a false dichotomy: "any ref that's within curly brackets" versus "citations that use the <ref ></ref> format"; most uses of the citation templates are inside <ref ></ref>; they're only not when people are using the separate-refs-and-bibliography-sections system, which is mostly at longer articles, especially those repeatedly citing the same sources at different pages. Anyway, I agree with the gist above: The templates already handle this. For non-templated citations, C is probably the ideal answer, but B is fine, too (see elsewhere on this page for consternation about non-breaking spaces; not everyone agrees that the end result, for the reader, of using them is worth the instruction creep of inserting rules about them into MoS).  It's generally a safe bet to include one, in regular (non-templated) text any time you don't want a number or other tiny string to separate from something else.  As noted in the other thread, we don't actually have a rule to use, e.g., , just  , but many of us do the former as well for the same rationale, even in absence of it being demanded by MoS.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  23:22, 30 July 2017 (UTC)
 * Because of how rare it is, I had not even considered the refs in braces. As long as there is nothing wrong with me adding the nbsp, I will continue to do so. – Sb2001 talk page 23:29, 30 July 2017 (UTC)
 * Because of how rare it is, I had not even considered the refs in braces. As long as there is nothing wrong with me adding the nbsp, I will continue to do so. – Sb2001 talk page 23:29, 30 July 2017 (UTC)

MOS:NUMERO
Whenever I see 'No.' or 'No' (now you have all brainwashed me into meticulously eradicating all non-MoS styles of writing), I place the {abbr|No.|number} tag on it, for clarity. I understand that both versions, ie dotted and non-dotted may seem slightly confusing. Although, I maintain that most instances of it in the UK drop the full stop (or, full point as the MoS now likes to use). My question is whether this is actually worthwhile: very few editors seem to bother doing this outside infoboxes, which makes me wonder whether I should be doing it at all. If it is advisable, why does the MoS not require editors to add the abbr tag? It is the same with circa. I always tag that if I see it (and – as much as it hurts – put the full point on the end). The MoS is, as usual, rather ambiguous on this matter. – Sb2001 talk page 11:37, 28 July 2017 (UTC)
 * It would not be unreasonable to suggest that editors use abbr (or its HTML equivalent--perhaps we can be silent on that option) when using an abbreviation for the first time in a document, as this increases the accessibility of the abbreviation. My preference is usually to link a phrase first if the word is relevant to the topic, though. And perhaps an abbr isn't necessary when the phrases immediate context makes the definition clear (as in an infobox on a page about a taxonomic group). --Izno (talk) 12:40, 28 July 2017 (UTC)
 * And when used, it only need be done at first occurrence within the article, same as with linking it. If you do it a bunch of times in the same page, it's annoying; it browbeats the reader and implies they can't remember an abbreviation from one paragraph to the next. There effectively is not difference between <abbr ></abbr> markup and a wikilink, in the sense that a string has been marked up such that doing something with it with the pointer leads to additional contextual content; one goes to another page, the other results in a tooltip. Sso does  (or same on some other element that doesn't have a reserved use for  ). I mention that last bit, because I keep running into cases of people abusing <abbr ></abbr> to generate tooltips for things that are not abbreviations, which is an HTML spec violation (a well-formedness rather than validation error, in the extended definition of "well-formed" which includes "follows all the syntactic rules").  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  23:57, 30 July 2017 (UTC)

Guideline on using diacritics?
I'ld like to ask if there is a guideline on using diacritics in commonly used words like naïve and café. The reason I ask is that someone just did this:, and I'm not sure whether that's appropriate or not. LK (talk) 08:16, 22 July 2017 (UTC)
 * There doesn't seem to be, since it's one of those common-sense, follow-the-sources things. We don't use anything that dictionaries label obsolescent, and that includes diacritics on foreign words fully assimilated into English and consistently used without them, and words that used to use them to indicate pronunciation (e.g., no one writes "rôle-playing games", or "they decided to coöperate", or "he's a very learnéd man" in 21st-century English).  The "café" spellings is not obsolete, and preferred by many; "naîve" is getting rare.  A tiny handful of words have survived assimilation and retained the diacritics almost universally, e.g. "née" and the uncommon masculine form "né" (sometimes encountered without, but I've been in genealogy circles long enough to see it usually retained, especially after the effective demise of ASCII-only e-mail by the late 1990s). Another example is "façade", which retains the cedilla so people don't think it's said "fakade".  But we've lost the umlaut in the English [mis-]borrowing of "uber", and along with it the alternative German spelling "ueber", which is unrecognizable to most English speakers.  The acute accents are also retained (except in very sloppy writing) in "résumé", to avoid confusion with "resume" meaning "to continue". General rule of thumb: check some dictionaries, and use the spelling that is listed first by the majority of them.  In your cases, the results are "café" but "naive".  Regardless of the answer on the latter, the diacritic would be retained in non-assimilated words (e.g. naîf) and in a foreign word that has an English non-diacritic equivalent: "naîvete" but "naivety".  The very existence of "naivety", "naiver", "naivest", and other derived forms, which never have the umlaut, discourages "naïve" in English. Zero dictionaries of the eight or so I checked listed "naïve" as preferred, and three did not list it at all, only "naive". All of them preferred "café" over "cafe".  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  06:49, 28 July 2017 (UTC)
 * I wouldn't say no one uses "coöperate". Famously, The New Yorker still uses umlauts on double vowels to indicate separate pronunciation instead of a long vowel sound. But it's so rare as to be essentially obsolete. Though I vaguely remember an editing dispute not long ago with someone who insisted on it's use in a baseball article. I think the person was blocked as a series sockpuppeteer not long after, though. oknazevad (talk) 21:06, 28 July 2017 (UTC)
 * Yeah, The New Yorker and a few other publications (e.g. The New York Times) intentionally do a few obsolete things in their house style, as a branding mechanism. It seems to have no detectable impact on usage by others.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  00:13, 31 July 2017 (UTC)
 * In that related article, Café Royal is a proper noun place, so that accent can be retained. Also, if certain words are quoted without the diacritics, then you can leave those alone. <strong style="color:#606060;">AngusWOOF ( bark  •  sniff ) 21:27, 28 July 2017 (UTC)
 * It would definitely be retained in a proper name, regardless of anyone's spelling preferences. If the owner was quirky and had spelled it Kafé Royal, that's the spelling we'd use; we wouldn't rename the restaurant to "correct" spelling.  But MOS:TM limits apply; we're not going to mimic something even more quirky, like Kafé&middot;Яoyal, where the alterations are just typographic logo-design gimmicks.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  00:13, 31 July 2017 (UTC)

Capitalization of IUCN status
In the interest of the wider community, and by the solemn Wikipedia accord to endlessly fork around, I'd like to draw your attention to the following discussion. Regards,  nagual  design   00:13, 23 July 2017 (UTC)
 * We've been over this many times before. Time to add a rule about this to MOS:CAPS, since it comes up again and again; I can think of at least three previous discussion of this just in the last year or two.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  06:53, 28 July 2017 (UTC) Hmm. Its already in MOS:QUOTE, as examples under "Typographic conformity", which has been quoted in that discussion. The specialized-style fallacy is rearing its ugly head, with someone unhappily suggesting that the MoS is "bad" on this and should be changed, so he/ she can continue capitalizing "Endangered".  I explained why that can't happen.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  08:06, 28 July 2017 (UTC)
 * Begging your pardon, but my suggestion was not an unhappy one, nor do I wish to 'continue capitalizing "Endangered"'. In fact the only edit I have made to any article in this regard was to switch "Critically Endangered" to lowercase once, which was reverted, hence I began a discussion (per WP:BRD). As I have repeatedly stated, it doesn't matter to me one way or the other if we use caps or lowercase, I'm just trying to push for consistency, like yourself.
 * While I appreciate your input at WikiProject Tree of Life I think you ought to lay off being so antagonistic. You make a lot of good points but then keep slipping in comments about other editors' motivations and thought processes; all unnecessary assumptions on your part. The discussion may well smack of some previous discussion that you once took part in, but there's no need to take out your frustrations on those involved here.
 * When I read comments about "editorial resignations due to the heat generated over style trivia" I can't help thinking that you're probably part of the problem, and referring to such discussions as "trivia" or "pointless style disputes" doesn't help quell the heat. So please, try to be more like Sisyphus and less like the boulder. Regards,  nagual  design   12:53, 29 July 2017 (UTC)

A few more commenters on this issue would be much appreciated.  nagual  design   23:31, 1 August 2017 (UTC)

Formatting of earthquake magnitude scales
Does anyone have any pertinent information or strong preferences on whether the labels for the various earthquake magnitude scales should be italicized (like this ML, Mw, MS, mB, mb, mbLg, etc.) or not (like this ML, Mw, MS, mB, mb, mbLg)?

It might be noted that this is entirely a matter of style, not nomenclature, and that some seismological authorities italicize, and some do not. Also: these terms are not variables: they are labels, analogous to "F" and "C" for temperature scales. ~ J. Johnson (JJ) (talk) 21:43, 29 July 2017 (UTC)


 * I think italics are more difficult to read than Roman (regular) font, so I prefer Roman font, but I guess it's important to know what most experts in the field use. Pinging ' and '. – Corinne (talk) 20:20, 30 July 2017 (UTC)
 * You might be surprised, but earthquakes are not really my thing. That said: One of the problems I see is that some sources appear not to use italic font period, so it is not clear whether the absence of such around earthquake magnitudes means anything. Jo-Jo Eumerus (talk, contributions) 21:16, 30 July 2017 (UTC)


 * Our basic rule about all stylization is that if the sources don't do it consistently, we drop it, unless there's a good reason to retain it. This appears to be a vestige of italicizing mega- as a Latinism, which is something we no longer do in English since the assimilation of that and related prefixes into everyday language via the computer industry. It's inconsistent to apply the italics in this one case but not in MB, etc.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  23:12, 30 July 2017 (UTC)


 * That's good info. My take on this is: there are no strong preferences. For various reasons I will go with non-italic (unless something comes up). And we can let this discussion time-out. ~ J. Johnson (JJ) (talk) 21:26, 4 August 2017 (UTC)

question about verb tense in lists describing events
The article list of medical ethics cases has a mix of past and current tenses in the summaries of the cases. (Example of present tense: "A teenaged woman is declared brain-dead and her family wishes to maintain her body on mechanical ventilation perpetually.") I think the summaries should use past tense, since these all occurred in the past, and it reads better to me. At the same time, using the present tense makes sense because it describes the particular ethical dilemma at the time it was a dilemma. The article creator prefers present tense but doesn't have a particularly strong opinion about it, so I thought I would ask here whether or not there is a specific guideline in place (or if one could be made). Either way the summaries should consistently use the same tense. —Мандичка YO 😜 23:10, 4 August 2017 (UTC)

RFC announcement
I have started an RFC at Wikipedia talk:Manual of Style/Abbreviations. Anyone interested in this issue should comment there, please. DES (talk)DESiegel Contribs 22:20, 6 August 2017 (UTC)

Honorifics in MOS:ISLAM
In my opinion, MOS:ISLAM is missing a line about honorifics in non-Muhammad religious figures. Please comment on Wikipedia talk:Manual of Style/Islam-related articles. --HyperGaruda (talk) 11:49, 7 August 2017 (UTC)

MOS:TMRULES
Please take part in the following discussion at the MoS Trademarks talk page; Wikipedia talk:Manual of Style/Trademarks – Sb2001 talk page 23:15, 10 August 2017 (UTC)

MOS:DASH in infoboxes
I happened across this article, and tried to change the infobox, to remove the spaces either side of the en rule/dash, as per MOS:DASH. I was shocked to find that the template actually includes the spaces on its own. This means that it is very difficult to remove these unnecessary spaces. Does anyone know how it is done, and how the template can be amended to fix the clear error? – Sb2001 talk page 19:39, 1 August 2017 (UTC)
 * If you mean the spaces around the ndash in the in-office dates, they belong. MOS:DATERANGE.  E Eng  21:12, 1 August 2017 (UTC)
 * Right. I was not aware of that ... another sensible part of the MoS. – Sb2001 talk page 22:45, 1 August 2017 (UTC)
 * All parts of the MoS are sensible. --Izno (talk) 02:19, 2 August 2017 (UTC)
 * Well, MOS:DASH says the same thing, though maybe not in as organized a way.  E Eng  03:19, 2 August 2017 (UTC)
 * Yes, it's been an irritant for many years. This is why it happens: whoever first designed the field on the infobox template must have had in mind a dash between two full dates (3 February 1961 – 15 April 1962). Or they just didn't care and decided to slip-slop it. But they can't have it both ways. The field should not be automatised in that way. Is it so hard for editors to write the date range / year range manually? Tony   (talk)  03:21, 2 August 2017 (UTC)
 * What sort of justification is there behind having no spacing for some ranges, but spaces either side of the – for others? As I interpret it, Tony is saying that this rule is present because of the infobox templates. Is this the case? – Sb2001 talk page 15:07, 2 August 2017 (UTC)
 * As described both in MOS:DASH and MOS:DATERANGE, spaces are added around the en dash if either element in the range contains a space, presumably to flag that the en dash is linking at least one boundary indicated with multiple words. isaacl (talk) 16:31, 2 August 2017 (UTC)
 * 'The 1940 peak period was May–July': I assume that you meant to say that if both sides have spaces, there should be a space either side of the –. Well, I think that the policy is confusing. It should be either space everything, or nothing. But, if that is the way it is, I cannot argue. – Sb2001 talk page 16:45, 2 August 2017 (UTC)

 E Eng  01:27, 3 August 2017 (UTC)
 * I just don't have the energy to explain why the rules on when to space around the ndash, and when not to, are they way they are, but they make sense and are there for good reason; will explain it comprehensively I'm sure.
 * My speculation is that no one noticed that the template is inappropriately always spacing because it's so rare that both "sides", in this particular template, contain no space. Whatever the reason, I have little doubt our template wizards can add some magic to check that data and space, or not space, the ndash as needed.
 * From MOS:DASH: The en dash in a range is always unspaced, except when either or both elements of the range include at least one space. In the example you quote, the elements bounding the range are "May" and "July". isaacl (talk) 15:01, 6 August 2017 (UTC)
 * What example who quotes?  E Eng  16:26, 6 August 2017 (UTC)
 * I would imagine that it is my last comment, just before the outdent. – Sb2001 talk page 17:11, 6 August 2017 (UTC)
 * Got it. Are we done with this?  E Eng  17:35, 6 August 2017 (UTC)
 * I don't think I need to explain it comprehensively; in the example given, 'The 1940 peak period was May–July', the two linked elements are "May" and "July" which have no spaces, so the whole "if both sides have spaces" thing never arises with this example. It's about spaces being juxtaposed, not spaces that incidentally are on the outside of them, e.g. between "was" and "May".  If  side has spaces  it, then space the en dash: "the Australia – New Zealand trade agreement".  Otherwise you get a confusing mess like "Australia–New Zealand", which in many fonts looks almost (or even entirely) indistinguishable from "Australia-New Zealand", e.g. in the popular Lucida Sans Unicode, which has better Unicode support than most fonts.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  04:09, 13 August 2017 (UTC)

Serial comma
The example is attempting to show ambiguity. Chaning it to "The author thanked Sinéad O'Connor, Bob Marley, and her parents" is unambiguous unless you posit that Bob Marley was female. The comment "did she thank her own parents, or O'Connor's parents?" indicates that the writer of this example did in fact mean "The author thanked Bob Marley, Sinéad O'Connor, and her parents".

The next paragraph develops the theme: "The author thanked her mother, Bob Marley, and Sinéad O'Connor".

Personally I think the whole section needs a good going over, particularly the interaction of serial commas with appositives. Martin of Sheffield (talk) 13:45, 9 August 2017 (UTC)


 * The purpose of the example was to show how to remove ambiguity resulting from (in this case) a lack of the serial comma: "In such cases of ambiguity, there are three ways to clarify: [...] Recast the sentence (first example above):". The problem was that the proposed clarification added a  ambiguity, as the invisible comment in the wikicode even talked about.  Of course, you still have to know that Bob Marley was male, but that's another issue.  --Deacon Vorbis (talk) 14:02, 9 August 2017 (UTC)
 * So pick a better example. We clearly still need one, since one of the most common cleanup edits I do is inserting necessary serial commas into ambiguous sentences, without even actively looking for them.  It's a very frequent "poor writing" matter, mostly caused by people who habitually avoid all serial commas and just think its okay to force people to read something several times to understand it.  Well, it's not.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  04:11, 13 August 2017 (UTC)

Using two different surnames for a subject throughout the Mary Kay Letourneau article?
Opinions are needed on the following matter: Talk:Mary Kay Letourneau. A permalink for it is here. The discussion concerns whether or not to use "Letourneau" for some parts of the article and "Fualaau" for other parts of the article. Flyer22 Reborn (talk) 18:57, 20 August 2017 (UTC)

Correcting usage 3 of #Hyphens not to contradict usage 1
Okay, what is it that you guys don't understand about how the correction I'm trying to make to #3 usage of hyphens is to correct a self-contradiction? #3 says never to use a hyphen in a proper name, but #1 already said to use a hyphen in a personal name if it was designed that way, and personal names are proper names/proper nouns. So why should we supposedly not clarify the exception using a wording like I tried to use?

2600:100E:B149:8DAB:F80C:D7D2:437F:F3BC (talk) 15:41, 19 August 2017 (UTC)
 * The original text made perfect sense. There was no need to change it. You really should have brought this here before making the edit. Further to this, your edit directly contradicts MOS:NUMERO, which instructs editors to use No., rather than the hash. I will warn you now—certain editors around here like to see evidence of the MoS causing problems before it is changed. Also, you need to show that you have a good understanding of the MoS itself: not adhering to NUMERO is probably not the best idea when you want editors to accept your change. It proves that you have no experience making stylistic edits, or working with the MoS in other ways. – Sb2001 talk page 15:51, 19 August 2017 (UTC)


 * How do you figure "Never insert a hyphen in a proper name" isn't a direct contradiction of thee instruction to use hyphens in names that already have them? How can that possibly "make perfect sense" to you?
 * And I never said anything to anyone about any numbering format that I thought they should use. Where did you get that idea?
 * 2600:100E:B149:8DAB:F80C:D7D2:437F:F3BC (talk) —Preceding undated comment added 15:58, 19 August 2017 (UTC)
 * No, you use a hash instead of the numero sign. That is against the MoS. Drop the aggression; it really will to help your case. The word insert suggests to anyone with a basic understanding of English that it is an alteration. Come back with a draft of a more sensible edit, ie fix the hash issue, and we will consider allowing you to apply it. – Sb2001 talk page 16:09, 19 August 2017 (UTC)
 * Oh, so your whole "hash" thing--the number sign, as in "example #1"--is causing your "style" complaint, making you falsely think I "have NO experience" in stylistic edits? Wow. That by itself is not proof of that. But okay, I can easily change from using the so-called "hash" to just not using one, haha!
 * Drop what supposed "aggression:" defending myxelf? Where did you get the idea that simply defending my position constitutes "aggression"?
 * 2600:100E:B149:8DAB:F80C:D7D2:437F:F3BC (talk) 16:27, 19 August 2017 (UTC)
 * The hash is its common name in the UK, as you would know if you had any stylistic knowledge. You are supposed to come here to convince us, not mock our language. I assume that you make no stylistic edits from your contributions list: . You only started editing today, according to this. Another reason why you should make an account. My point regarding the hash is that MOS:NUMERO instructs editors to use the numero sign. User:EEng—thoughts on this matter? – Sb2001 talk page 17:16, 19 August 2017 (UTC)
 * Your edit is difficult to follow, and really just unnecessary. This is all in the section on compound modifiers, not personal names.  Do you have an example of where this might cause a problem?  --Deacon Vorbis (talk) 16:00, 19 August 2017 (UTC)
 * Exactly. No evidence of having to change it regularly, no evidence of edit-warring, etc. – Sb2001 talk page 16:09, 19 August 2017 (UTC)
 * How is plain English so "difficult to follow"? I know #3 is about modifiers, but without some exceptional language it contradicts the instruction from instruction #1. A personal name is a proper noun/name, is it not? So #1 says to use a hyphen there if the name has it already. OK, fine.
 * But then as-is, the "NEVER" from #3 turns right around and undoes the instruction from #1, because #1 gives a time to use it in a proper name, but then #3 takes it all away. That's why some kind of "unless" or "except" clause needs to be included: Never, except when..., etc.
 * 2600:100E:B149:8DAB:F80C:D7D2:437F:F3BC (talk) 16:27, 19 August 2017 (UTC)
 * No. 3 says don't insert a hyphen (i.e. add a hyphen where there wouldn't otherwise be one); it doesn't say you must remove a hyphen that's already there for some other reason. Maybe a solution would be to change "never insert a hyphen into a proper name" to "never insert a hyphen into an unhyphenated proper name" (but strictly speaking this change is unnecessary if you understand what "insert" means). --  Dr Greg   talk  17:21, 19 August 2017 (UTC)
 * Just as I think, and said. This was met with the IP criticising my use of the word 'hash'! Your proposed change makes sense, so you have my support. – Sb2001 talk page 17:28, 19 August 2017 (UTC)


 * How about, instead, changing never insert a hyphen into a proper name --> never add a hyphen to a proper name. Some intuition tells me that add a bit more connotes "putting one in where there wasn't one before" than does insert (though the denotation is the same).
 * Side note: MOS does not apply to MOS itself, with good reason. Trouting someone for using # in this context is uncalled-for.  E Eng  18:37, 19 August 2017 (UTC)
 * I would have said that a lot of your complaining about my comments was something for which there was no call. Why does MoS not apply within MoS? Your suggestion is good, though. – Sb2001 talk page 18:46, 19 August 2017 (UTC)
 * MOS doesn't apply to MOS for the same reason advice on writing a college essay isn't written like a college essay: they serve different purposes and need not – should not – be written the same way. MOS is written in a more relaxed style than is an article (e.g. uses contractions), sometimes refers to or addresses the reader, uses all kinds of formatting we'd never use in a articles, promiscuously mixes ENGVARS, and more.  E  Eng  19:05, 19 August 2017 (UTC)
 * As I previously discussed, the importance of following a common writing style is lessened on the manual of style pages themselves, and by social convention Wikipedia editors will typically be more lax on non-article pages, to avoid going down uncollaborative sidepaths. So whether or not the manual of style strictly applies, for less formal pages intended for community interaction, it's generally best to not get in a slap flight over things like using the number sign. isaacl (talk) 20:32, 19 August 2017 (UTC)
 * As I previously discussed, the importance of following a common writing style is lessened on the manual of style pages themselves, and by social convention Wikipedia editors will typically be more lax on non-article pages, to avoid going down uncollaborative sidepaths. So whether or not the manual of style strictly applies, for less formal pages intended for community interaction, it's generally best to not get in a slap flight over things like using the number sign. isaacl (talk) 20:32, 19 August 2017 (UTC)

I think some of the comments here are an example of biting the newbies. The original poster (OP), IP2006, may not have known that many editors watch the Manual of Style, are cautious about accepting changes, and prefer that things be discussed on the talk page first. S/He may not even have known there was a talk page. S/He may have been confused by the hash/pound sign/number symbol that appears in edit mode but creates a bullet. S/He may have become frustrated when, seeing what appeared to be an obvious contradiction, others did not see it – a normal reaction. If someone had invited or directed this editor to discuss the proposed change on the talk page, perhaps feelings would not have been ruffled. I happen to agree with the OP that some readers may see a contradiction in MOS:HYPHEN:

Hyphens (-) indicate conjunction. There are three main uses:


 * 1) In hyphenated personal names: John Lennard-Jones.

and the second sentence in the first bulleted item in 3.:

But never insert a hyphen into a proper name (Middle Eastern cuisine, not Middle-Eastern cuisine).

However, I think the OP's suggested wording is wordy and unclear. I'm not even sure we need either "insert" or "add". I suggest the following wording for item3:


 * Other than those personal names that already have a hyphen, such as in item1 above, do not hyphenate proper names: (Middle Eastern cuisine, not Middle-Eastern cuisine).

– Corinne (talk) 21:00, 19 August 2017 (UTC)
 * Biting the newbie would not have happened if they had not been rude. See the talk pages of a few involved editors (their contributions will be the easiest way to access these). It happens to everyone, anyway—I certainly got it, and I was nice in my questions! Even when there was a discussion ongoing, they chose to instate their version into the MoS. This combined with the 'I know best' attitude was met with some regrettable responses. I tried to help them, by advising account creation, and offering to exchange emails with them. They did not choose to do this. Your wording is okay, also. – Sb2001 talk page 21:30, 19 August 2017 (UTC)
 * I am sorry, Sb2001. I thought the above was the only thread on this. I didn't realize the OP had posted on various editors' talk pages. I agree that the OP was aggressive and rude, which would tend to irritate other editors, and I think in spite of that you were polite and helpful in your comments, which seemed to fall on deaf ears. – Corinne (talk) 21:48, 19 August 2017 (UTC)
 * Yep. Not every potential new editor is one we want, per WP:COMPETENCE. This probably goes double for an anon whose impetus to participate is to just rail against some WP policy or guideline until they get their way. The anon has no edits other than squabbling.  Ergo, either a troublemaker or someone logging out to be one.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  19:00, 21 August 2017 (UTC)
 * Corinne's rewording doesn't quite work, introducing an ambiguity ("contradiction") in another direction. We do in fact hyphenate proper names  other things all the time, and should continue to do so: "Italian-style sausage", etc. Try:
 * Other than those personal names that already have a hyphen, such as in item1 above, do not insert a hyphen into a multi-part proper name: (Middle Eastern cuisine, not Middle-Eastern cuisine).
 * The original wording seemed fine to me, honestly; a single user, ever, who doesn't get it and blows up about it isn't evidence of a real problem. John Lennard-Jones comes with a hyphen, it doesn't have one because we inserted/added it. But, yes, spelling it out more clearly can't hurt.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  19:00, 21 August 2017 (UTC)

Using "illicit" to describe Letourneau's interaction with Fualaau at Mary Kay Letourneau article
Yeah, this article again. Opinions are needed on the following matter: Talk:Mary Kay Letourneau. A permalink for it is here. Flyer22 Reborn (talk) 14:49, 21 August 2017 (UTC)
 * Seem like a MOS:WTW matter; this doesn't relate to the main MoS page. This talk page will get unmanageable if we start listing every vaguely "style"-related dispute here.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  21:13, 21 August 2017 (UTC)
 * There also seems to be a WP:FORUMSHOP issue here given the thread that F22 created on the same subject at WP:BLPN. —David Eppstein (talk) 21:18, 21 August 2017 (UTC)
 * Well, they're neutral notices pointing to the same discussion. Just not really a noticeboard item; better "advertised" at WT:BLP, probably.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  21:25, 22 August 2017 (UTC)