Wikipedia talk:Manual of Style

RfC: Inconsistent capitalization of eponym in same context

Should MoS clarify that capitalized eponyms do not lose their capitalization when used adjectivally? The test case for this is the very long-running dispute about using "Gram stain" but "gram-negative" (i.e. negative in a Gram stain test) when both of these are eponymous (of Hans Christian Gram) and refer to exactly the same dye-staining process in microscopy. (It has nothing to do with the metric unit gram/gramme.) The rationale offered for the inconsistency has been that some medical publishers/organizations (like the US CDC, and some medical dictionaries) like to lower-case eponyms in adjectival usage, a rule that WP's Manual of Style (and most other style guides) do not have. Doing that would require a special exemption to MOS:ARTCON, the overriding consistency guideline of MoS. Use of lower-case "gram-negative" style is not consistent in reliable sources in bacteriology, medicine, histology, microscopy, and related fields. It's purely a house style choice (as our own articles indicate, with sources).

This RfC does not address cases where an eponym's connection to its namesake has been effectively severed and the meaning has shifted (e.g., we would continue to capitalize in constructions like Platonic solid, Platonic love, and the Draconian constitution of Athens, but permit lowercase for figural usage like "His relationship with his roommate was platonic", "She said her parents' rules were draconian", though in encyclopedic writing we'd be better off avoiding such wording). Lower-case is also used in various other cases when virtually all sources agree on lower case (eustachian tube, caesarian or cesarean section), again due to loss of a clear connection to the namesake in the public mind (contrast degree Celsius and other units, Hodgkin's lymphoma and other diseases, Newtonian mechanics and other scientific principles, etc.)

 — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  00:12, 21 October 2017 (UTC)

Comments on the RfC

• Support. Our central MOS:ARTCON guideline already indicates we should not mix-and-match these styles ("Gram stain", "gram-negative"). Wikipedia capitalizes eponyms and other proper names (except when off-WP usage consistently uses lower case, as with caesarean section), and has no special "do not capitalize if adjectival" rule; we don't care if CDC or AMA does have one, since their house styles are not ours. The most obvious problem with this mixed usage is that it can result in "gram-negative" and "Gram stain" in the same article, even the same sentence, which confuses readers as to whether these are even related concepts, and leads to long-term editwarring (since 2004!). In this particular case, an additional major rationale is that "gram-negative/positive" strongly but wrongly implies to non-expert readers that this has something to do with the metric gram unit. Attempts to impose spelling (especially capitalization) quirks from specialist sources are something WP routinely rejects; this is known as the WP:Specialized-style fallacy, the notion that sources reliable for technical facts about a topic are somehow transubstantiated into the most reliable sources for how to write plain English for a general audience any time that subject comes up. We've had numerous RfCs on this before (including this huge one), and it's a common theme at WP:RM, with consensus consistently siding with WP's style guide and with internal WP consistency. The habit of medical (especially American medical) people of down-casing eponyms used adjectivally because their journal publisher does it is understandable, but the attempt to force it on WP as a "standard" is a WP:CONLEVEL policy problem. Furthermore, carving out a special one-topic exception to ARTCON would be WP:CREEP and would do nothing useful, only inspire more demands for special topical exceptions to every rule someone in some field doesn't use in their writing for other publishers.  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  00:16, 21 October 2017 (UTC)

• Support in principle but what is the proposed wording? Usually I argue that when a term is not consistently capped in sources, it's not to be treated as a proper name in WP. But with Gram, clearly a person's name and clearly capped in the "Gram stain" context, though only capped about half the time in sources in "Gram-positive" and such, I think I'd agree that downcasing it randomly sometimes is a bad idea, especially given the ambiguous interpretation here that is both an explanation for and a bad effect of the downcasing. So I'm unsure whether there's a general principle here, but for WP capping the name Gram should be the clear preference, since we are about clarity, consistently using caps to signify proper names and not otherwise. Dicklyon (talk) 05:00, 21 October 2017 (UTC)

  Oh, I didn't much care about specific wording. Even what I used above, "capitalized eponyms do not lose their capitalization when used adjectivally", ought to work. This wouldn't impose anything weird like "always capitalize eponyms even when sources do not: Caesarean section".  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  05:12, 21 October 2017 (UTC)

• Oppose I cannot offer any argument against the use of Gram for the staining procedure - perhaps if the page was changed back to Gram stain that would work. I only know that most books that I refer to all use lower case when describing the bacteria and upper case when referring to the Gram stain. Lower case reads better, especially in pages with many references and I'm all for easier reading. A clearer guideline would probably be helpful.--Iztwoz (talk) 09:18, 21 October 2017 (UTC)
  • But the page is at Gram staining, and consistently uses that spelling: "Gram staining or Gram stain, also called Gram's method, is a method of staining used to ...". We already know, absolutely, from previous discussions on this that the lower-casing of "gram-negative" and "gram-positive" is just something a few particular publishers' house styles do (following a general rule they have to lower-case eponyms is adjectives, a rule WP doesn't have and which is also not found in most other style guides; there's nothing special about Gram and bacteria in this). That the books you happen to read are from those publishers is just a WP:IKNOWIT coincidence.  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  06:55, 23 October 2017 (UTC)
• Oppose. This should be done on a case by case basis (does this topic use upper or lower case for this use of this name?) rather than trying to set the hard-and-fast rule that topics are required to be consistent in capitalization for different usages. Clearly, in some cases, the common usage is not consistently capitalized and Wikipedia should nevertheless follow that common usage rather than trying to become a trendsetter for consistency. —David Eppstein (talk) 17:37, 21 October 2017 (UTC)
  • But this isn't about "different usages", it's about the same usage, the namesake "Gram" in bacteriology. Why on earth would we write "Gram stain" in one sentence then "gram-stained" in the next, in the same article, just because the second is adjectival? This is not a "rule" that MoS entertains anywhere for any case, so why would we do it in this one case? What's magically special about adjectives as used by people with microscopes?  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  06:50, 23 October 2017 (UTC)
    • But it's clearly not the same usage; they are different in other ways (like one is followed by the word stain and the other by the compound part -positive). As for why: because those are the ways they are commonly written (if that's true; I have no opinion on the specific case of "Gram", only on the general position that we should follow the scientific literature, even when we think it's inconsistent, rather than trying to impose our own consistency on it). —David Eppstein (talk) 20:51, 23 October 2017 (UTC)

      Already pointed this out, but will do so again: the style you're advocating would have "Gram stain" in one sentence and "gram-stained" in the very next one, simply because the latter is adjectival. We just don't do that on Wikipedia.  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  21:25, 23 October 2017 (UTC)

      "We just don't do that" is correct, if by "that" you mean making up new capitalization conventions because we don't feel the commonly used ones obey a consistency rule that we are making up ourselves. —David Eppstein (talk) 22:07, 23 October 2017 (UTC)

• Support: there seems to be no clear reason for inconsistency. –Sb2001 18:06, 23 October 2017 (UTC)
• Further consideration: See also Giemsa stain, Leishman stain, Papanicolaou stain, Wright's stain, Romanowsky stain, May–Grünwald stain. The "gram-negative" style (decapitalize eponym if used adjectivally) would result in "may–grünwald-negative", "papanicolaou-negative", "wright's-negative", etc.  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  06:43, 27 October 2017 (UTC)
• Dictionaries are inconsistent but favor uppercase: Uppercase in Oxford, Random House, and Collins; lowercase in Merriam Webster; both in American Heritage; not listed in Cambridge or Macmillan. Of these major online dictionary publishers, only a minority even provide the lower-case version at all, and only one rejects the capitalized one, while three reject the lower-case version. A more recent commercial site is WordWebOnline (also powers one of the most popular mobile-app dictionaries); it gives only the capitalized version [1].  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  07:17, 27 October 2017 (UTC)

Extended discussion of the RfC

Background material:

Dispute about this, especially "Gram-" vs. "gram-", on Wikipedia dates back to at least 2004 [2], and has never stopped, though with more editors in favor of consistently using "Gram-", and citing Wikipedia rationales for doing so, with a minority of editors insisting on "gram-" for the sole reason that CDC or some other entitity spells it that way. To use the history of Gram-positive bacteria as an example: consistent capitalization efforts for several years [3] [4] [5], followed by a sudden de-capping [6], later reverted [7]; reimposition of lower case [8], then upper [9], lower [10], upper [11], lower [12], upper (among other cleanup) [13], mass-revert back to lower [14]. For the last several years, this mixture of "Gram" and "gram-" has been "enforced" by a single editor, as the later diffs show.

Some time during this "slow editwar", editors began to use the article text itself as a battleground to falsely advance assertions that lowercase "gram-" is a scientific standard (which of course was challenged [15]). There's a similar history at Gram-positive bacteria and various articles on specific bacteria and other bacteriological subjects.

Previous inconclusive discussion has happened at:

• Talk:Gram-positive bacteria#Standardized spelling & punctuation
• Talk:Gram-positive bacteria#Gram positive
• Talk:Gram-negative_bacteria#Standardized spelling & punctuation
• Talk:Gram-negative bacteria#spelling instructions

The results of these discussions have been:

1. A helpful short section in the articles, on the conflicting orthography in off-WP sources; this removed the PoV/OR assertions in favor of "gram-" as some kind of standardized requirement.
2. A single user imposing "gram-" style on Wikipedia without consulting our own MOS [16] [17] [18], and without feedback from anyone; this was on the basis that some medical works prefer this style, but WP is not a medical work.
3. Reversion of "Gram-" at various bacteria articles, such as E. coli [19] (by the same editor present "enforcing" the "gram-" style at the main articles)
4. Recent revertwarring (same editor again) against "Gram-" on the basis of the above discussions (which are not about WP usage at all), even after MOS:ARTCON is cited both in talk and in edit summaries.

Despite the WP:POLICY position being obvious (from MOS:ARTCON to WP:CONLEVEL), an RfC seems warranted given the 13 years or so this dispute has been running.
 — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  00:12, 21 October 2017 (UTC)

I would just dispute the claims of this being a case of continued edit warring - since June 2014 with the addition of the orthographic note there has been an acceptance for this among the usual editors (myself included). On occasion someone has reverted the use without comment or just changed a few instances, and in line with ususal editing practices the previous version has simply been restored. This has happened on very few occasions since June 2014. Editors seem happy with using lower-case. --Iztwoz (talk) 09:02, 21 October 2017 (UTC)

I think using "Gram" in "Gram stain" and "gram" in "gram-positive" and "gram-negative" isn't a problem. The first is (if I understand this correctly) a procedure named after a person, and the latter two are adjectives describing bacteria and/or the results of a test using the procedure. I'm sure at some point in the past, before Wikipedia, there was a period when both "Caesarean section" and "caesarean section" were used. It was a transitional period with the changeover to lower-case. I think in English generally, the preference is for lower-case; I agree with Iztwoz that lower-case is easier to read. The fact that "gram" could suggest to non-experts that it has something to do with the weight unit I do not find persuasive. There are many homonyms in English; also, the real connection between "gram" and the person is just something that one needs to learn if one is interested in the topic. The only way I would support "Gram-positive" and "Gram-negative" would be if the great preponderance of sources spell them that way.  – Corinne (talk) 17:45, 21 October 2017 (UTC)

They're both named after the same person. Lower-casing "gram-negative" is exactly the same thing as writing "Gram stain" then "gram-stained", or writing "newtonian mechanics", or "shakespearean theatre". While there are style guides in the world that call for this, WP's MoS is definitely not one of them.  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  06:47, 23 October 2017 (UTC)

For whatever it might be worth, as a biology teacher I can tell you that upper case "Gram stain" for the process and lower case "gram-stained", "gram-positive", and "gram-negative" are what I am used to seeing in the texts we use. --Khajidha (talk) 14:41, 23 October 2017 (UTC)

We know that various (especially American) biology and medical publishers like to lower-case eponyms in adjectival constructions; that's explicit in the whole discussion.  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  20:35, 23 October 2017 (UTC)

I've just seen an edit summary of yours on Escherichia coli. When making the same revert you state that it (Gram) is a common name proper name etc. and should be treated in the same way as Kelvin and Ohm - nobody ever uses these terms capitalised - there is the ohm and the kelvin. ? --Iztwoz (talk) 10:57, 25 October 2017 (UTC)

I was misremembering and misstating; the Kelvin scale is capitalized, but the unit is not; that also seems to be the case with ohms and amperes. Celsius and Fahrenheit get the caps. So, it's an inconsistent system. The rationale I gave was faulty; "Gram-negative" isn't a unit or unit symbol anyway.  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  22:53, 25 October 2017 (UTC)

In mathematics, it is standard to capitalize the name Abel when it modifies another word as a noun adjunct ("Abel equation") but not when it is used adjectivally ("abelian group"). However, there are other names that remain capitalized even when used adjectivally (Euclidean, Eulerian, etc). It would be incorrect to change the capitalization in these cases. —David Eppstein (talk) 23:07, 25 October 2017 (UTC)

We just need to get entirely away from this idea that WP is going to mimic this mathematics publisher, and that medical style guide, and this train-spotting website, and that news publisher's stylesheet, when they don't even consistently apply their own "rules" (not within a field, and often not even within a publication). We don't have any issue with standardized ISO units being lower case even if often named after a person, but this "sometimes use lower case just because it's an adjective" is fiddly nonsense that inspires never-ending editwarring here, and it only exists off-WP in the house styles of particular publishers, so it has no place here.  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  08:09, 26 October 2017 (UTC)

On unit capitalization: ISO's rule on the matter is "lowercase the first letter and capitalize anything else derived from a proper noun". In this case, your error is that the unit is not "Celsius" but instead "degree" with the modifier "Celsius"; the same with "Fahrenheit" and "degree Fahrenheit", and also the only-used-in-freedom-loving-country-while-engineering-units "degree Rankine" (though occasionally you'll hear or see "rankines"...). The unit "kelvin" is consistent with the general rule, as are amperes, ohms, and any other unit. Our own article covers this at International System of Units#Unit names. The NIST follows that particular rule. --Izno (talk) 13:34, 26 October 2017 (UTC)

Fair enough; my point is they remain "Celsius" and "Fahrenheit", despite technically being adjectival (they're modifiers of "degree[s]" in such constructions).  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  08:50, 27 October 2017 (UTC)

Using {{clear}} to prevent images bleeding into other sections

Does the MOS have any recommendation for or against using the {{clear}} template to prevent images from spilling into subsequent sections? It appears to be the main use of the template, according to its documentation, but I couldn't find any official guideline mention. Obviously, the best course would be not to use too many images, but this is not easy to achieve consistently across all displays. With today's huge display resolutions, an article that follows MOS:LAYIM's benchmark of not spilling over at 1024×768 might still break down on wider screens. --Paul_012 (talk) 09:03, 24 October 2017 (UTC)

It's mentioned at Wikipedia:Extended image syntax. DMacks (talk) 09:16, 24 October 2017 (UTC)

Thanks. It's not really part of the MOS, though... --Paul_012 (talk) 14:49, 24 October 2017 (UTC)

Perhaps more specific recommendations could be made at MOS:IMGLOC and/or MOS:LAYIM? Something along the lines of "try not to have images spill into the next section, and optionally, use {{clear}} to prevent this happening on wider screens." --Paul_012 (talk) 14:49, 24 October 2017 (UTC)

I agree it should be mentioned in the layout MOS somewhere. I'm not sure simply spilling into later sections is a problem (the tradeoff is whitespace that interferes with flow of text). But if the later section(s) have images, there's an image-stacking problem that makes those other sections' images not located where their content is, and "put images where their content is" is a explicit MOS guideline (Wikipedia:Manual of Style/Images#Vertical placement. DMacks (talk) 16:43, 24 October 2017 (UTC)

Sorry, I think this is best left unsaid. If images are stacking (in windows of reasonable width – there's always going to be some nerd with a super-big monitor telling us that articles don't look good in super-wide windows) then resize them, juggle them to different sections, left-right alternate them, or whatever. You really shouldn't be using {clear} to solve that problem, because as already mentioned it substitutes one unattractive thing (excess vertical whitespace, which can REALLY look awful) for another. Pretty much the only time I use it is when there's an image near the end of the last section of the article, just before ==References== etc., and maybe on a wide screen the image will intrude into the references, distorting the column layout or whatever; plus I've used it in a few very well-considered situations in articles that are very stable, with a lot of images carefully laid out. (Open Phineas Gage for editing and search {clear}.) But this something best left to experienced editors, and it's best taught by their seeing it in actual use. If we start talking about {clear} in MOS some zealot will start running around adding it everywhere an image spills into the following section, and next thing you know there will be edit wars, angry words, and we'll all be at Arbcom. So let's just let this lie.

Wikipedia:Picture tutorial being burned by angry Wikipedians

BTW, I think I wrote the text at IMGLOC, An image should generally be placed in the most relevant article section; if this is not possible, try not to place an image "too early" i.e. far ahead of the point in the text discussing what the image illustrates, if this could puzzle the reader. Notice that it says an image shouldn't be too "early" – it doesn't talk about being too "late". I worded it that way consciously to avoid making editors feel every image must be neither too early nor too late, and must therefore be "just right" i.e. in the exact section. If an image of John Smith comes a paragraph later than where he's discussed in the text, that's not so bad, because the reader has presumably already read about him; but if the image comes before where he's discussed, then the reader may be puzzled.

Finally, Wikipedia:Picture tutorial is the most godawful hypertechnical overcomplicated help page we have, and should be ripped to shreds and the pieces ritually burned, if we could figure some way of doing that over the internet. It's useless – horrible. EEng 02:27, 25 October 2017 (UTC)

Okay, fair points. I've also mostly used it to stop images breaking the references columns, but was kinda worried it might be regarded as an inappropriate lazy hack or something. I'll stop worrying. (Since you mentioned WP:Picture tutorial, might I add that I've never understood why there are so many image policy/guideline/instruction pages, nor been able to easily locate information in any of them?) --Paul_012 (talk) 17:43, 25 October 2017 (UTC)

Archaic -st words

Please see Wikipedia talk:Manual of Style/Words to watch#Archaic 'st' words – it's more of a MOS:COMMONALITY vs. MOS:ENGVAR matter than a MOS:WTW one.  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  10:12, 25 October 2017 (UTC)

The title of this section is biased. -- PBS (talk) 13:15, 26 October 2017 (UTC)

I just copy-pasted it from the original (and fixed the markup). This is just a pointer to the discussion; the place to object to the title is at the actual discussion.  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  08:48, 27 October 2017 (UTC)

Regardless of the bias in the title (it could have been "Chiefly UK 'st' words") the subject still merits discussion. Sorry if you think the title is biased, but the fact remains that outside of the UK (particularly in America), those words are considered archaic. Various UK style guides seem to agree. ~Anachronist (talk) 20:47, 27 October 2017 (UTC)

The trouble is, if you are wanting to start an open RfC-style discussion about something, beginning by creating an impression that these words are archaic is not particularly good. You do not need to say anything about them. –Sb2001 21:24, 27 October 2017 (UTC)

RfC on quotations within links

There is an RfC in progress at Wikipedia_talk:Manual_of_Style/Linking#RfC_about_linking_in_quotations about MOS:LINKSTYLE & MOS:LWQ. NPalgan2 (talk) 18:17, 26 October 2017 (UTC)

When more than one variant spelling for a word exists within a national variety of English

I often despair of Wikipedia and its growing list of does and don'ts. I am reverting the Revision as of 06:28, 2 October 2017 by user:Scribolt because it is not only unnecessary instruction creep it is also potentially harmful to the goals of the section in which it resides.

The problem is that if someone adds a bullet point to this page like this then it is only a matter of time before someone runs AWB or a bot script over articles that the user of the bot or AWB has never manually touched. If there is what you think is an archaic spelling on a page then be bold page and fix it. If it is reverted then follow BRD. That is standard practice and it does not need additional guidance here.

This looks to me like an addition that was not thought through. The bullet point that I am removing says is in a section called "[MOS:COMMONALITY |Opportunities for commonality!]: When more than one variant spelling for a word exists within a national variety of English, the most commonly used current variant should be preferred unless there is a consensus at the talk page of the article to use the less common alternative.

MOS:COMMONALITY a complicated issues because it is often used as a method to ride rough-shot over national varieties of English, particularly those of minority English dialects -- which in practice means everything but American English. Just look at how "fixed-winged aircraft" article was usurped by "Airplane". If that additional article had been created with "Aeroplane" as its title it would probably have been speedily deleted.

The point is the "there is a consensus at the talk page of the article to use the less common alternative" is an inversion of the usual Wikipedia way per BRD where changes are only made if there is consensus on the talk page to make them. I do not believe this is an issue where the usual proof of consensus ought to be changed, because if the spelling is really archaic then most editing in good faith will accept it, and if not then the usual dispute resolution will end up with change.

Or user:Scribolt was it your intention to create a bludgeon so that those running bots and AWB script could force through changes on multiple pages without the tedium of engaging in BRD to gain a consensus for such changes?

As phrased this rule would allow someone could go through any topic with close connection with Britain and "rationalise" [sic.] the spelling of words ending in "ize" to the usual British ending of "ise". That might be desirable for consistency across British English articles, but how does that improve MOS:COMMONALITY in general? -- PBS (talk) 08:31, 27 October 2017 (UTC)

I've restored a variant of it (since the line-item was added after extensive consensus discussion), but without the "unless ..." caveat that is the subject of this dispute, and modifying it to say "usually", which is consistent with our general approach to such matters (advise, permit exceptions, don't dump a huge list of examples on people). PS: The Oxford -ize worry above is not actually applicable, because Oxford spelling is itself explicitly recognized on WP as an legit ENGVAR, and we even have templates for it along with the "use American English", etc., templates.  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  08:46, 27 October 2017 (UTC)

Thank you SMcCandlish for your rewrite, that is an improvement. PBS, you might want to consider whether speculating on my motivations for an edit without evidence in such a manner is appropriate. Indeed, if you'd taken even a cursory amount of time to read both the RFC and discussion I referred to in my edit summary I think you would have been unlikely to conclude that an intent to facilitate bludgeoning of any kind was not very likely. For the record, I have never contributed to Wikipedias growing list of female deer, but even if I had I would see nothing to apologise for. Scribolt (talk) 10:30, 27 October 2017 (UTC)

At User:Scribolt I did not speculate on your motives, ask asked you if it was you intention to. There is such a thing as unforeseen consequences. -- PBS (talk) 10:48, 27 October 2017 (UTC)

User:SMcCandlish you write The PS: The Oxford -ize worry above is not actually applicable, because Oxford spelling is itself explicitly recognized on WP as an legit ENGVAR, I know about Oxford spelling and its acceptability to date (which is why I used it as an example), but according to the wording you restored it contradicts that for British English because "When more than one variant spelling for a word exists within a national variety of English, the most commonly used current variant should usually be preferred." as "ise" is more common in British English that negates the use of "ize". If on the other hand you are going to argue that "ize "should be used because it is the most common thanks to its use in American English then that negates ENGVAR. -- PBS (talk) 10:48, 27 October 2017 (UTC)

Except, again, WP treats Oxford English as its own ENGVAR. If you can't replace New Zealand English with Canadian English (without a legit reason), you can't swap out Oxford and non-Oxford British English either. If you think it's necessary, we could add a footnote about it or something. On the side matter, I wouldn't personally mind at all if we just settled on -ize across the board for commonality reasons, since the British have no lack of familiarity with the spelling, but some just prefer -ise. (There are lots of similar conflicts in Canadian English, some arising as "big deal" matters only since the 1990s.) However, there appear to be more British (and other Commonwealth English) users who prefer -ise (as with -st, e.g. amongst – see other thread) despite Oxford and other British publishers advising to avoid that style in formal writing. So, I would not expect such a proposal to gain consensus, and it would piss people off to advance one seriously.  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  11:16, 27 October 2017 (UTC)

This[20] should do it.  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  11:41, 27 October 2017 (UTC)   And that's been sagely compressed by EEng and DrKay [21]; I was sleepy when I cobbled the original together. :-)  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  22:42, 27 October 2017 (UTC)

Note that there has been a related discussion about exactly this bit of the guideline over at Wikipedia talk:Manual of Style/Words to watch#Archaic 'st' words, which may have been the basis for starting this thread here. For my part, I don't see the problem that PBS describes unless there needs to be a clearer distinction between common and universal vocabulary. So we use regional spellings that are common to a region, but for cases where both universal and regional synonyms are used in the same region (such as among and amongst in the UK), we prefer the universal term. ~Anachronist (talk) 19:29, 27 October 2017 (UTC)

At first I thought you said savagely, and the funny thing is that I didn't bat an eye. EEng 23:00, 27 October 2017 (UTC)

FWIW, I share PBS's concern that the recently added "unless there is a consensus at the talk page of the article to use the less common alternative" is an inversion of normal editing and BRD; we've had this issue before with ENGVAR, DATEVAR, etc., being written as if a consensus discussion has to happen before anyone's allowed to edit. Which reminds me we had a consensus discussion on one of these (DATEVAR, I think) that agreed to clarify it, but the clarification hasn't been made yet (archiver bot hid it away a few weeks ago). As for -st, I'm skeptical we'd get consensus to always prefer "among" over "amongst" right now (though going that direction is inevitable in the long term because English itself is moving that way).  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  21:27, 27 October 2017 (UTC)

That bit was actually your suggestion... But yeah, I agree that it's better now, and thanks to everyone involved. Scribolt (talk) 11:05, 30 October 2017 (UTC)

Bystander comment

I never cease to be amazed at the things people manage to find worth arguing about. EEng 11:27, 27 October 2017 (UTC)

I feel I need to take issue with such a parochial view. Martinevans123 (talk) 11:33, 27 October 2017 (UTC)

What?! That word is abused! No church parish is at issue!  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  22:45, 27 October 2017 (UTC)

Gesundheit. --A D Monroe III^(talk) 22:58, 27 October 2017 (UTC)

Should MOS cover "data is" vs. "data are"?

It's been suggested at Talk:Disk storage#RfC on "data are" or "data is" that this issue should be decided by MOS. Is this a MOS issue? Specifically (assuming both "data are" and "data is" are proper English, and the difference is use as either a count noun plural of "datum" or a mass noun) is use choice a national variety, like "colour/color"? Is "data is/are" already covered by WP:ENGVAR? If so, should ENGVAR be amended to make that coverage obvious? If not, should ENGVAR be expanded to cover similar variations that don't follow national boundaries? Should other English "boundaries" be identified by common usage within a field, industry, subject, time, etc., or some combination? How? --A D Monroe III^(talk) 15:44, 29 October 2017 (UTC)

I commented there. I see no reason to think this is ENGVAR related, and I don't think the MOS needs to cover such narrow cases. Dicklyon (talk) 21:48, 29 October 2017 (UTC)

Nothing to do with ENGVAR, no need for special treatment at MOS. The answer to the underlying question depends on the context (whether the data is counable) , not on a rigid rule. "All of that data is now stored on the backup server". "These three specific pieces or data are now stored on the backup server". --Guy Macon (talk) 22:05, 29 October 2017 (UTC)

Concur with Guy Macon and Dicklyon.  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  01:43, 30 October 2017 (UTC)

Concur. EEng 02:15, 30 October 2017 (UTC)

Perplexing preposition problem

Recently I've run into problems with some editors who insist that the preposition at should not be used with respect to cities, but should always be changed to in, of, from, or some other word. Attempts to point out that at is perfectly appropriate and idiomatic in this context, and also expresses the intended meaning better than the alternatives, usually result in the same changes being made repeatedly. It looks as though a few editors are actively seeking out such phrases and changing them to their preferences, and resisting any attempt to convince them that at a city, town, or other geographical location is perfectly acceptable. I have asked for some authority supporting the claim that this usage is wrong or should be avoided, but haven't been shown any, and haven't found any on my own. Most grammar books and style guides are silent on the issue, or seem to support using at with cities and towns. The Oxford English Dictionary specifically says that this is one of the primary uses of at, and gives examples from the thirteenth century to the present (mostly limited to English cities such as Winchester and London, in my edition).

I'm not sure there's a Wikipedia policy that applies here. Is it simply an English variant? Or just personal preference? That's how I see it when people substitute other prepositions for the intended one. I've written a lot of articles about ancient Romans and Roman families, and over time I've come to prefer the phrase at Rome, because it conveys location without adding unintended or inaccurate meanings. In Rome implies "within the territorial boundaries", which is too specific and not necessarily accurate; of Rome and from Rome imply origin, which is often somewhere other than Rome, or at best uncertain. In any case, if at is perfectly appropriate, and more accurately expresses the intended meaning, is there any policy to point to when reverting changes to other prepositions based on another editor's preferences? Is it relevant if such a change is the only involvement that an editor has with certain articles? It's a little annoying when it appears that editors are simply searching for examples of phrases they dislike, in order to change them to ones they prefer, if they have no other interest in the articles, or understanding of the reason why one choice of words is preferable to another in a given instance. Or is even asking the question displaying "ownership behaviour"? The situation is becoming quite frustrating. P Aculeius (talk) 03:53, 30 October 2017 (UTC)

Some examples would help, but I think you're way off base here. Looking at your recent edits, I see for example "The gens Flavia was a plebeian family at Rome." (wikilinks omitted). This is horribly awkward; in or of would be a lot better here. --Deacon Vorbis (talk) 04:17, 30 October 2017 (UTC)

I don't see why you think it's "horribly awkward". Could you explain? In either case, in would be completely wrong, since it implies spatial relationships, rather than location, and therefore territorial limitation. The Flavii were a Roman family, no matter where individual members went or lived. While they were chiefly associated with Rome, they didn't have to live within the city limits, or cease to be part of the family when they went elsewhere. Of would be better than in, but it implies origin, which is not necessarily true of many Roman families. But your point seems to be that at should not be used of cities. I don't understand why anybody thinks this; it's flatly contradicted by the best authorities. The Oxford English Dictionary says, "2. With proper names of places: Particularly used of all towns . . ." with examples following: at Winchester, at London, at Jerusalem, at Edinburgh. It's not easy to search for specific prepositional phrases in literature, but I found a site to search Shakespeare, and found numerous examples: at Marseilles, at Rome, at Ephesus, at Antium, at London, at Harfleur, at York, at Venice, at Antioch, at Tyre, at Pentapolis, at Mantua, at Verona. On point, my copy of Reading Latin describes the locative case: "it is used to express 'at' with names of towns and one-town islands", with examples at Rome, at Corinth, at Athens, at Carthage, at Sardes. So what is the authority for this being incorrect? Is it just that some editors don't like it? And if that's all it is, then what's the relevant policy? P Aculeius (talk) 05:07, 30 October 2017 (UTC)

Maybe it's just me, but to me "at" is something you would use for an instantaneous action, while "in" implies a longer duration. So, "our train stopped at Rome", because stopping is a quick event, but "I stayed in Rome for a week last year" because staying is not. In the case of the gens Flavia, they were in Rome (that's where they lived), not at Rome (on their way through from somewhere else to somewhere else). This web site gives a different distinction, based on the size of the place (one that I'm not sure I agree with) but it ends up with the same result in this example. This other web site comes closer to the distinction in my mind: "at" is for specific points in space or time (time, as I explained it above) while "in" is for a broader location or time period. —David Eppstein (talk) 05:34, 30 October 2017 (UTC)

If you think for a second that The gens Flavia was a plebeian family at Rome is acceptable 21st-century English, you need to get your nose out of Gibbon. EEng 05:58, 30 October 2017 (UTC)

Concur with Deacon Vorbis and David Eppstein. To answer P Aculeius's question, The gens Flavia was a plebeian family at Rome is horribly awkward, as Vorbis put it, because it's not idiomatic in formal English. It may be idiomatic in colloquial English somewhere, but WP isn't written in colloquial English. That "at [city name]" usage is virtually unseen in writing we'd find in sources we'd consider reliable, except for things like David's "the train stopped at Rome", which is still a construction most people would probably avoid. That kind of case is actually a shorthand for something more specific, either a particular station actually named "Rome", or one named something else and just serving Rome, but called "Rome" for short by insiders to that transit system. The station is not the city of Rome (present or past) or vice versa. By way of comparison, I might get off at a bus stop named Foo Street, at the corner of Foo and First, and say "I got off at Foo Street". But if I got in a car wreck on that street, I'd say it happened "on" (or maybe in British English, "in") Foo Street, not "at" it. The street and the station named after it are sharing a name, but are not the same thing or the same kind of thing, ergo different prepositions are liable to apply in such cases, and even to the same case in different contexts. The Flavian family did not live "at" Rome any more than I live "on" the city of Oakland, or I'm going "for" the grocery store, or your cousin grew up "of" Boston. (Yet a sign may point at Rome, a rain can fall on Oakland, I could work for a grocery store, and your cousin might be of Boston).  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  06:16, 30 October 2017 (UTC)

More on "on" vs "in" for streets —David Eppstein (talk) 06:26, 30 October 2017 (UTC)

Well, I clearly seem to be in the minority among Wikipedians, but at least I've been able to think clearly enough to check some reliable sources, which flatly contradict the argument that this is somehow wrong or archaic. My OED is from the 1970's, so I guess it's not a reliable source for language anymore; Merriam-Webster has for decades defined at first and foremost as "a function word indicating presence in, on, or near", which clearly covers this usage; and with respect to Rome in particular, I think the argument that "sources we'd consider reliably" would seldom or never use at Rome to indicate location, rather than spatial relationship or point of origin is clearly wrong. I found an abundance of what I would like to think everyone would acknowledge as reliable sources, in terms of Roman scholarship, formal English (whether American or British), and in the case of Mary Beard, the vernacular.

T. J. Cornell: The Beginnings of Rome (Routledge History of the Ancient World) (1995)

• "This seems to have happened at Rome at the end of Phase IIB."
• "What happened at Rome at the end of the sixth century . . ."
• ". . . to my mind the traditional accounts imply a change of precisely this kind at Rome."
• "At least one chamber tomb has been identified at Rome itself . . ."
• ". . . the fall of the monarchy at Rome was part of this wider picture."
• ". . . the Esquiline necropolis at Rome . . ."
• "The Tarquins were not the only outsiders to rule at Rome."
• ". . . the rule was already established that Roman citizens could not be enslaved at Rome."
• ". . . such luxuries as were to be found at Rome must have been imported . . ."
• ". . . may indeed have ruled at Rome."

Gary Forsythe, A Critical History of Early Rome: from Prehistory to the First Punic War, University of California (2005)

• "These simple graffiti constitute some of the earliest samples of writing discovered at Rome."
• "Furthermore, although examples of writing at Rome are quite rare for this period . . ."
• ". . . J. C. Meyer has combined both these concepts to explain the history of human habitation at Rome during the early Iron Age."
• ". . . public performances of some sort existed at Rome much earlier than is generally supposed . . ."
• "Laws ameliorating the conditions of indebtedness were not forthcoming at Rome until the fourth century B.C."

The Roman Historical Tradition: Regal and Republican Rome, James H. Richardson & Federico Santangelo, eds, Oxford University Press (2014)

• "This child, who had been left at Rome, was the seed of the Fabian race . . ."
• ". . .Timaios attributes the first coinage at Rome to Servius."
• "What is far more perplexing to us is how a historian contemporary with the very first coinage at Rome. . ."
• "However, with the credentials of the story of the Tarquins at Rome now restored . . ."
• "Festus preserves the precious information that she changed her Etruscan name to Gaia Caecilia at Rome."
• "The possible identification of Servius rex at Rome and the seruus rex at Aricia gives an obvious origin and a terminus post quem for the tradition . . ."
• "According to one version, the child of the Fabii who owed his life to having been left at Rome had the praenomen Numerius."
• ". . . Claudius mentioned foreigners who had even attained the kingship at Rome."
• ". . . a time before the establishment of written historiography at Rome near the end of the third century . . ."
• "It seems, therefore, that Marcellus had a decisive impact on the tradition of the spolia opima at Rome, starting in his own lifetime."
• ". . . the fact that they never had, except at Rome, a nomen gentilicium . . ."
• ". . . it implies a fertile and creative narrative tradition existing long before the introduction of literary historiography at Rome."
• "On the Capitoline at Rome, then, there was an acropolis dedicated to the great deity of Olympus . . ."
• ". . . and his mother (unnamed) bore him in the palace at Rome."
• ". . . with interesting repercussions on the transition from monarchy to Republic at Rome."
• "The record of their women as priestesses and queens certainly fits their aristocratic origins and position at Rome."
• ". . . developments at Rome were influenced by Rome's relationship with both the Greek world and Etruria."
• "Even that single Fabius left at Rome, who later became the propagator of his race, has a parallel in the Greek story."
• ". . . they may be used as evidence by those who think the change at Rome from monarchy to Republic was more an evolutionary than a revolutionary process."
• "By traditional dating, this change takes place fifty years earlier at Rome than at Athens.
• ". . . although the three elements of monarchy, aristocracy, and democracy existed at Rome under the monarchy . . ."

Mary Beard, SPQR: a History of Ancient Rome, W. W. Norton (2015)

• ". . . after his detention at Rome and attempts at popular politics at home . . ."
• ". . . the snobbery that was another side of life at Rome . . ."
• "In the first civil war at Rome since the brief conflict after the death of Nero in 68 CE . . ."
• "Some of his rivals called him just a 'lodger' at Rome . . ."
• ". . . Polybius tries to shoehorn the political life that he witnessed at Rome into a Greek analytical model that does not entirely fit."
• "It was never a rallying cry at Rome, even in its limited ancient sense . . ."
• ". . . who would never have dreamt of standing for election at Rome . . ."

And just by titles,

• R. Develin, The Practice of Politics at Rome 366–167 B.C.", Ed. Latomus (1985)
• Sandra R. Joshel, Work, Identity, and Legal Status at Rome: a Study of the Occupational Inscriptions, University of Oklahoma Press (1992)
• Keith Bradley, Slavery and Society at Rome, Cambridge University Press (1994)
• Denis Feeney, Literature and Religion at Rome: Cultures, Contexts, and Beliefs, Cambridge University Press (1998)
• David Noy, Foreigners at Rome: Citizens and Strangers Duckworth, Classical Press of Wales (2000)
• Francisco Pina Polo, The Consul at Rome: the Civil Functions of the Consuls in the Roman Republic, Cambridge University Press (2011)

I'm sure I could come up with many more examples if I spent hours searching for them. But I think my point must be adequately demonstrated by now. At with the names of towns and cities is perfectly acceptable English, whether formal or familiar, and even recommended in certain contexts, of which at Rome is a prime example. And in the particular example that's being cited repeatedly, it's the preferable alternative because I do not mean in with the connotation of "inside, within the boundaries of", nor do I mean of or from, suggesting point of origin, which is frequently unclear or known to have been some other place. Indeed, as some of the passages quoted above discuss, it was not only possible for a family to hail from one town or region or people and yet come to be regarded as "Roman" in subsequent times, but also that it held a markedly different social status at Rome. A family might have been part of the local aristocracy at Tusculum, Antium, Praeneste, but enrolled among the plebeians at Rome; or like the Claudii, distinguished only by wealth and influence at Regillum, but accorded patrician status at Rome.

My question, however, remains. If a word is correct, perhaps more correct than the alternatives in a given context, and another editor makes it a crusade to change it to his or her personal preference wherever that usage occurs, is there any particular Wikipedia policy that supports reverting the change? P Aculeius (talk) 15:02, 30 October 2017 (UTC)

Without commenting on the merits of "at" vs "in/of", I'll just say that it's generally considered ok to revert such a change and then discuss on the talk page, per WP:BRD. Kendall-K1 (talk) 15:22, 30 October 2017 (UTC)

(edit conflict)

P Aculeius I appreciate your question, your gracious tone, and the time you spent looking for sources to support your view. I agree with the others, above, that at with a city is unusual today. In the course of reading and copy-editing many articles on Wikipedia, though, I have to say that I have seen the preposition at used with cities, and places in general, more than I ever had before. I think it must be more a British English usage (and perhaps, as EEng said, a usage in only certain parts of England, and perhaps it is a usage that was more common in the past but is fading) than American English usage. Americans would never say "at Rome", or "at London", except for something like Our train stopped at Rome. Instead, they would use in Rome, of Rome, or from Rome, depending upon the intended meaning. In that particular example you cited, I think the sentence could be re-worded so that the connection between the family and the city of Rome were made clearer. "At Rome" doesn't really explain much. I would write something like:

• The gens Flavia was a Roman plebeian family.
• The gens Flavia was a plebeian family centered in Rome.
• The gens Flavia was a plebeian family long connected with Rome.  – Corinne (talk) 15:34, 30 October 2017 (UTC)

Perhaps an important detail missing from this discussion is that the example provided is from the lead sentence of the article. It's true that at Rome doesn't really explain much; but that's precisely the point; the alternatives are too specific. The article, and others like it, already provide more detail about origin, location, and time frame. The natural course of an article is to move from the general to the specific detail, beginning with the most general description in the lead. With respect to the alternatives you suggest, the third seems to be the least appropriate, since it implies that the family had some independent status as plebeians divorced from Rome or the Roman state; which it did not, as the distinction between patricians and plebeians is specifically Roman. The first alternative also stumbles here, as there were no non-Roman plebeians; but I also think the phrasing is awkward. Meanwhile, "centered in" combines the notion of territorial limitation that I'm trying to avoid with in, and adds the suggestion that the family diffused outward as one traveled away from the city, which may or may not be true, but which is certainly not the intended meaning of the sentence.

In this specific context, I chose at Rome because it was the simplest, most straightforward way of indicating location without describing territorial limitation, point of origin, duration in time, spatial relationship, concentration, dilution, organization, or other more specific meanings. It's a phrasing that's been used from Middle English up to the present time, and continues to be used in formal and academic writing, both British and American, as well as less formal sources such as SPQR, in ways that cannot be clearly distinguished from the case at issue. I think I have my answer now: there's no particular policy about arbitrary changes to language, but they can be reverted, and if necessary discussed on an article's talk page so that the reasons for or against a particular wording can be debated. Thank you for taking the time to answer me civilly, as I very much appreciate the courtesy, and your suggestions, even though I wasn't convinced that any of them were better than the original wording. It's much easier to discuss issues when you're not being told that your choices are wrong, that you're wrong to ask the question, wrong to present your reasons, to support your point with reliable sources, and that you've just been wasting everybody's time. P Aculeius (talk) 17:52, 30 October 2017 (UTC)

• Aside from suggesting that he's got a bit too much invested in this, P Aculeius' list of examples is pointless. Yes, there are lots of places the phrase at [city] makes sense, but They were a family at Rome isn't one of them. Prepositions are funny that way. As usual, Corinne is infinitely patient, and her suggestions are good ones. This is not a MOS matter. EEng 15:54, 30 October 2017 (UTC)

  This is the gist of the entire thread right here: "there are lots of places the phrase at [city] makes sense, but They were a family at Rome isn't one of them." Things like "at Rome at the end of Phase IIB" and "possible identification of Servius rex at Rome and the seruus rex at Aricia" are references to an extended archaeological site and artifacts within it (e.g. inscriptions), not references to a city per se. Families are not artifacts or digs. The "in the palace at Rome" type of construction is different, and common for nested placenames ("University of Texas at Austin"). Families are not placenames. Similarly, "his detention at Rome" is the same kind of case as "The train stopped at Rome"; here "Rome" is referring to a facility (a governmental/military facility in the one case and a transit station in the other). Families are not facilities. And so on. The only kind-of-comparable case I'm seeing in that list (about names) is "she changed her Etruscan name to Gaia Caecilia at Rome." It's uncertain what the exact intent of the statement is without more context (given the legal nature of a name change among the Roman aristocracy, it probably refers to a institution not the entire city, and is thus another shorthand; compare "Alfred the Great died at Winchester"; this means at his royal facilities in Winchester, not "somewhere within the city"). But lets take it at face value and say that "in [city]" is at least attested in the sense that P Aculeius wants to use it, "The gens Flavia was a plebeian family at Rome." Is it the dominant usage? No. Is it common? No. Is it likely to be understood and interpreted as proper English by the average reader? No. Is it likely to result in later editors changing it, and thus in edit-warring over it? Yes (proven, since it already happened).  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  00:04, 31 October 2017 (UTC)

I'm sorry, but the distinction you're trying to make doesn't make any sense. The quotations given above cover a wide variety of circumstances including several referring to people. So you seem to be saying, "it's okay with respect to objects, buildings, institutions, and individual people, but not groups of people". And there's no basis for such a distinction other than that you think it sounds odd. Maybe several other people with similar backgrounds may think so too, but that doesn't make it bad English or in any way inappropriate if it happens to be the word that best conveys the intended meaning. I think your conclusion that "it's unlikley to be understood and interpreted as proper English by the average reader" is simply untrue. You've almost certainly passed over constructions of this type hundreds, if not thousands of times while reading various literature and not paid any attention to it because the meaning is perfectly transparent. And I'm even more certain that as an experienced editor in this field, you recognize that just because somebody, or lots of somebodies, prefer to say A rather than B, doesn't make it right, any more than the fact that edit wars get started over it. How many edit wars go on every single day due to people's personal preferences between different ways to say something that are both perfectly acceptable? The fact that one phrase is less common than another, or less common than it used to be, doesn't make it wrong, much less unintelligible. The question here was not whether this was good English, which I think the evidence clearly shows, even if you disagree. The question was how to respond to editors who make a crusade out of imposing one wording over another, if both are acceptable, but one may be preferred due to its meaning. And that question was answered already. P Aculeius (talk) 02:07, 31 October 2017 (UTC)

This is getting silly. It's apparent you're not a native speaker of English. And this isn't a MOS matter anyway, so work it out with the editors of the article. EEng 02:33, 31 October 2017 (UTC)

Markup for math variables

FYI

 – Pointer to relevant discussion elsewhere.

Please see Wikipedia talk:Manual of Style/Text formatting#Mathematics variables section is wrong and needs updating
 — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  05:41, 30 October 2017 (UTC)

Use of "died by suicide" at the David Reimer article

Can we get some opinions at Talk:David Reimer#"Committed suicide" vs. "died by suicide"? A permalink for it is here. I mentioned there that we have discussed "died by suicide" at this guideline's talk page before. There doesn't appear to be any consensus on Wikipedia about whether we should avoid "committed suicide" or use "died by suicide." And since it keeps coming up, maybe we should address it in the guideline? Flyer22 Reborn (talk) 23:38, 30 October 2017 (UTC)

Agree that this should be addressed below, so we can hopefully generate some consensus to guide us in the general case. Two articles already talk about the public debate concerning this point of terminology, namely the Suicide article, and the Suicide terminology article, the latter having an entire section devoted to it. Mathglot (talk) 01:58, 31 October 2017 (UTC)

Request for opinions posted at WT:LAW, WT:MED, WT:LGBT, WT:SOCIOLOGY, WT:PSYCH. Mathglot (talk) 02:19, 31 October 2017 (UTC)

Happy with which ever. We have had RfC on this in the past[22]Doc James (talk · contribs · email) 02:28, 31 October 2017 (UTC)