This is an archive of past discussions. Do not edit the contents of this page. If you wish to start a new discussion or revive an old one, please do so on the current talk page.

I've looked and just seen last year's RfC on quotation templates. Has there been any update specifically on the practice of using block quotations for illustrative purposes (i.e. in a box set off to the side, similarly to how we use images), as opposed to narrative quotations (the usual blockquotes supposed to fit into the flow of the text)? Since quote boxes have been quite extensively used it seems confusing that the MOS doesn't directly address them. --Paul_012 (talk) 11:12, 18 December 2017 (UTC)

@Paul 012: It used to. Then someone had the idea that if we don't mention them and just say to use ((Quote)) for block quotes, and to not use pull quotes, that they'd just go away. And of course that didn't work. May have to revert to the "don't abuse pull quote templates for things that are not actually pull quotes – and pull quotes are virtually never used in articles anyway" sort of instructions we had. The experiment is being less explicit has been a failure. Just like the experiment in using decorative quote templates all over the place has been a failure.

What's an example of an "illustrative" quotation in an article? Last year I went through over100 articles with pull quote templates in them, back-to-back, and in 100% of the cases, they were either block quotations mis-marked-up in decorative fashion, very short quotes that should have simply been given inline, or PoV-pushing exercises where an editors was drawing grossly WP:UNDUE weight to a particular statement by someone. I find it difficult to imagine that this has changed.
 — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  03:30, 20 December 2017 (UTC)

SMcCandlish, browsing the beginning of the list of Featured Articles on literature and theatre, I found quote boxes used in All God's Children Need Traveling Shoes, The Autobiography of Malcolm X, Big Two-Hearted River, Burger's Daughter, Candide, Casino Royale (novel), and Cento Vergilianus de laudibus Christi (about 22% of the ones I checked). Do you see these as examples of misuse? I think in literature-related articles at least, short excerpts of the work can help give the reader an idea of what the work is like, much like how images are used in articles about paintings. --Paul_012 (talk) 17:34, 20 December 2017 (UTC)

I don't think short literary excerpts are inappropriate in literary-topic articles, though the WP:COPYVIO crowd have divergent opinions on what is short enough and what is or isn't fair use. I even suggested to the FAC regulars that documentary excerpts might be a good case for sidebar templates. However, in many cases what's going on is highlighting of quotations for other reasons; those are a serious PoV problem, and should be put back into ((quote)), inline in their context. I forget the page that sparked that debate; some article on a gay UK politician from back when being a gay politician was a scandal. The bulk of the article is about a legal dispute between two parties, and multiple quotations of one party are used in decorative templates, including a sidebar, which of course grossly overbalances the article in that party's favour. There are a lot of problems like this.

There are many old FAs that have not been updated to current MoS standards because we're mostly directing our attention to crappy articles that need a lot of work. Bringing FAs into current compliance can actually be occasionally difficult, as certain editors feel excessively proprietary about "their" articles, and some of the WP:FAC crowd are overtly hostile to MoS-based cleanup and even to MoS's existence, despite "* It [the FA-candidate article] follows the style guidelines, including [three bullet points here]" being in the FA Criteria. This list of three things includes one which is actually not an MoS matter, but WP:CITE stuff. The handful of people strangely convinced that MoS doesn't apply to FAs are misreading it as "follows the style guidelines on the following enumerated points", when of course "including" is not exclusive and doesn't mean that at all.

The central issue is probably that all templates that could be used for documentary excerpts are mis-documented as quotation templates (formerly as block-quotation templates, but since changed to distinguish between block and pull quotes and to say not to use pull quotes in articles). People use them for decoration in any article because they think it looks cool, and they don't read MoS. When WP:MOS- and WP:UNDUE-cognizant editors object, the would-be owners of the article get angry, and point to an FA from 2006 as their justification, even though what they're doing isn't actually similar to what the FA is doing in most cases (literary excepts) – other than visually, in that they're both boxes of text attributed to someone. A previous commenter called them "attractive nuisances": as long as the templates like ((Rquote)) and ((Quote box)) named and documented as any kind of quote template, this "make Wikipedia look like a magazine or a blog" problem with continue. They should obviously be changed to refer only to documentary excerpts, and be documented as to when this is actually appropriate, with NPOV and COPYVIO pointers, and so on.
 — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  18:05, 20 December 2017 (UTC)

Please explain. Otherwise use wiki markup instead of HTML. Bright☀

It certainly prevents the browser (at least some of them) from properly repositioning at the edited section after a section edit, but my recollection is that there are other reasons. Perhaps you would like to research the reason and, if that reason no longer obtains, then propose changing the practice. But it's a bad idea, when you see that highly experienced editors are taking substantial trouble to go out of their way to do something you don't understand, to simply assume that everyone but you doesn't know what they're doing. EEng 18:03, 19 December 2017 (UTC)

It also clutters the page history, whereas I do not believe the <span> version does. --Izno (talk) 18:27, 19 December 2017 (UTC)

Just use Template:Anchor. -- Shyam Has Your Anomaly Mitigated (talk) 23:52, 19 December 2017 (UTC)

I'm with Shyam on this; use the anchor tag (or move the span) to directly below the heading. If coming to a section under instead of on the heading actually confused people, we would never use shortcuts that go to mid-section anchors. The clear fact is that no one's head asplode. There are no Web users unfamiliar with and unable to handle the fact a link can take them to some text in mid-page. Yes, there are long-term editors who insist on trying to shoehorn this markup into a heading; there are also just as long-term editors who oppose this practice. There is no established consensus either way. The majority of article-space usage is directly under, not in the heading. Beyond that current uneasy status quo, if people want to settle it then they can do an RfC, perhaps at WP:VPTECH. Most of us WP:DGAF, despite our vague distaste for one approach or the other. And there are actually a whole lot of approaches; see extensive test page I worked up with Peter coxhead a while back, at MOS:ANCHORHEAD. [And, yes, I do get bonus points for making a Star Wars pun.]  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  03:09, 20 December 2017 (UTC)

Sure, a link can take you anywhere on a page, but it helps the reader orient if the first thing he/she sees is a heading, or failing that a shortcut box -- "Ah, yes, I'm at...". Because MOS is so frequently linked-to we've always been super-careful, when changing a heading, to preserve validity and behavior of links to the prior heading. If you put together one of your patented comprehensive reform packages that addresses all issues, I'm behind it. EEng 04:00, 20 December 2017 (UTC)

I just outlined one: stop caring about the headings. If one changes, just create an anchor below it for the old name. It's the content that matters, people get linked directly to it, and putting anchors *in* headings is a pain the glutes for editors. Actually, we only really need to create these anchors when there are actual incoming links (do an insource search) and we don't update them to point to the new heading. I'm lazy, so I just create an anchor. We could actually have a bot clean it up, probably. The idea of doing ==<span id="Old name"></span>New name== obviously doesn't scale: if a long heading that was renamed to another long heading, you end up with an edit summary pre-filled with span crap that is actually longer than the edit summary window can handle, much less having any room for you to provide an actual edit summary. People who favor this "solution" haven't really thought it through very carefully.

The test-cases page I linked to above shows what we have to work with. It's not much. The proper solution would be for MW to be upraded to handle this is some other way. E.g, interpreting an empty, id-bearing span (or template outputting one) immediately before a heading as part of the section to be headed when doing section editing. But getting practical features added to MW is a decade-long process.
 — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  18:14, 20 December 2017 (UTC)

FYI

I moved WP:Alternative text for images to WP:Manual of Style/Accessibility/Alternative text for images, where the rest of the MOS:ACCESS supplement pages are. All links and shortcuts to it work, it just now isn't lost in the vast sea of the "Wikipedia:" namespace, but is consolidated with the related material. Also fixed the old WP:ALTPDI and MOS:ALTPDI shortcuts to work (they'd lost their anchor point when the "Purely decorative images" heading went away).  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  23:53, 20 December 2017 (UTC)

At MOS:PMC, an errant phrase needs to be corrected where it is written: "When a vulgarity or obscenity is quoted, it should appear exactly as it does in the cited source; unless faithfully reproducing quoted text, Wikipedians should never bowdlerize words by replacing letters with dashes, asterisks, or other symbols." I don't know what was intended by its inclusion and have temporarily commented it out. Does anyone know what sentiments were meant to be carried?--John Cline (talk) 19:34, 20 December 2017 (UTC)

It might mean don't fix the quote if the frickin quote spelled it wrong. --SarekOfVulcan (talk) 19:42, 20 December 2017 (UTC)

(edit conflict) :I see the meaning, now. I think it is a confusing preposition, as written, however, and suggest it should be reworked for clarity. I'll post a suggestion momentarily, unless it is sooner fixed by another. Yes, SarekOfVulcan, I believe that is correct. Thanks.--John Cline (talk) 19:51, 20 December 2017 (UTC)

• I offer this example for consideration:

"When a vulgarity is quoted, it should appear as it does in the cited source. Never replace letters with dashes, asterisks, or other symbols, nor spell-out an obscenity that is bowdlerized in the source; always, instead: faithfully transcribe quoted text exactly as it appears in the source."

I believe this retains the intended sentiments in less ambiguous prose. I give it for consideration, and possible use.--John Cline (talk) 21:28, 20 December 2017 (UTC)

You know, the more I look at it, the more I think it was phrased correctly in the first place. Your version isn't wrong, but it's more redundant than the original phrasing. --SarekOfVulcan (talk) 21:35, 20 December 2017 (UTC)

(edit conflict) I agree the last part is unnecessarily repetitive and emphatic, but the proposed change does add advice against the (possibly rarer) converse type of misquotation, where a word that was suppressed in the original is spelt out or supplied in the quoted version. (WP:CREEP perhaps but not redundant.) At any rate, either Never should be replaced by Neither, or nor by or, and the phrasal verb spell out doesn’t want a hyphen.—Odysseus1479 23:58, 20 December 2017 (UTC)

• It's means if the printed statement we're quoting literally reads, character-for-character: "I told the President of Elbonia to go f--k himself", then we quote it that way; we don't "correct" it back to fuck through some total misunderstanding of WP:NOTCENSORED. No one seemed confused by this before, but if it can be reworded more clearly, have at it. The instruction should remain, and should probably also advise wrapping such a thing in ((notatypo|...)) or ((sic|hide=y|...)), because many editors upon encountering a quoted "f--k" are probably apt to think an editor censored "fuck" out of the original [not to be confused with censoring the fuck out of the original, which might go further than that. >;-]  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  23:41, 20 December 2017 (UTC)

  I moved the words around and added only enough to make it parse, and this should work fine "When a vulgarity or obscenity is quoted, it should appear exactly as it does in the cited source; Wikipedians should never bowdlerize words by replacing letters with dashes, asterisks, or other symbols, except when faithfully reproducing quoted text that did so." I still am not sure we should be advising a visible [sic]. The only purpose of it is to alert editors that the censorship is in the original; readers don't (usually) know we have a WP:NOTCENSORED policy and don't care.  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  23:47, 20 December 2017 (UTC)

Thank you, I appreciate every response; and profit by each.

SarekOfVulcan, I appreciate the kind manner that governed your critique. I vaguely remember an era bygone when it seemed as though you were far less concerned with how others felt, and far more candid in verbiage and tone. I commend, in the interim, whom you have become.

Odysseus1479, I am impressed by your apparently high level of skill. In as much as Wikipedia is a thing being built, I'd be an apprentice by specialized trade who today had the privilege of the chief architect's ear.

SMcCandlish, the changes your edit gave measurably improve the guidelines at hand in this discussion. It's rare that such gain could even come of a request born of a fallacious premise. You're a great colleague, and an asset of much value overall. It's too often that we, as a community, forget to say thanks; today, I'll not forget. Thank you, for all you have done for this project.--John Cline (talk) 14:16, 21 December 2017 (UTC)

And happy holidays to you too! :-) This page perhaps more than any other has to deal with and write around inconsistent interpretations of what it says and means.  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  16:21, 23 December 2017 (UTC)

Level 2 headings are called "level 2 headings", not "==-level headings". Using this ungrammatical terminology is not "saying exactly what's meant", it's making up a new term for an existing term. A bullet list isn't called a •-list, a parenthetical statement isn't called a ()-statement, and a levle 2 heading isn't called a ==-level heading.

Also, mere paragraphs from MOS:IMAGES there's MOS:PAREN, which asks editors to avoid constructs like "Each image should be inside the major section to which it relates (within the section defined by the most recent ==-level heading, or at the top of the lead section)". Bright☀ 16:08, 19 December 2017 (UTC)

• Most people don't know what a "level 2 heading" is, but they sure know what a "==-level heading is".
• MOS does not apply to MOS, or indeed to anything outside article space.

EEng 18:08, 19 December 2017 (UTC)

Experienced editors know what a "==-level heading" is. Users using only the VisualEditor only have the notion of a "level 2 heading" to go by. --Izno (talk) 18:29, 19 December 2017 (UTC)

Good point, even though, of course, people who use VE are abnormal and not to be trusted. I've modified it to use both terms. EEng 18:33, 19 December 2017 (UTC)

And we actually do apply most of MoS to MoS, or people use it as a weapon to avoid MoS compliance in mainspace. Using both is good; my usual approach to this is something like: a level-2 heading (== ... == (or "a second-level heading", though "level-2" probably works better, as a clearer reference to <h2> for the super-dorks who know all that HTML mumbo-jumbo. >;-)  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  02:55, 20 December 2017 (UTC)

Of course "people who use VE are abnormal and not to be trusted." That makes them just like the rest of us.  SchreiberBike | ⌨  03:09, 20 December 2017 (UTC)

This perception might change if they'd stop eating kittens, but this is unlikely. It's a cultural matter.  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  16:30, 23 December 2017 (UTC)

It's fine to adopt MOS practices where possible in MOS itself, but when someone starts citing MOS:PAREN it's time to make clear what's what. EEng 03:48, 20 December 2017 (UTC)

You should worry less about "making clear what's what" and more about actual readable style. If you don't want to follow MOS:PAREN then start a discussion to remove it from the manual. Bright☀ 06:25, 20 December 2017 (UTC)

Older discussions

(Linking here for posterity) Wikipedia_talk:Manual_of_Style/Accessibility/Archive_10#Image_position - the original discussion about images in the appropriate section; Wikipedia_talk:Manual_of_Style/Archive_114#Images_and_third_level_headings - subsequent alteration to level-2 heading. As originally discussed, images should be placed in their appropriate sections (of any heading level) for accessibility reasons. This was altered to level-2 headings to give editors more freedom in placing images. Bright☀ 07:01, 20 December 2017 (UTC)

I've been re-reading Pinker's The Sense of Style (one of the most highly regarded how-to-write-better manuals in generations), and I'm struck by how frequently an illustration's relevance to the text isn't instantly clear; one often has to read a little further. This seems pretty common, and I doubt many brains are melting, as long as the image and the context are pretty close to each other. It's only a problem when the material is separated by many paragraphs. Readers can't hold in mind the subject of a caption of unknown connection to the material indefinitely, especially if additional subjects are being introduced in the text in the interim. They can do it longer if all the intervening material is building up to the "reveal", but that's a fiction and journalism technique, a minor form of burying the lead combined with using a teaser).  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  16:30, 23 December 2017 (UTC)

In many articles, for example The Bible and violence, we name some/a lot of scholars in the text. This is sometimes done "Anderson says" [1], sometimes "Philosopher, professor and author Eleonore Stump says" and anything in between. Is there any WP-guidance on what is "best", or is it all local consensus?

My personal preference is towards the latter. As a reader, I want some in-text info on who this is and why I´m listening to them, even if there´s a wikilink (though I´m less annoyed if there´s a wikilink). It may make the text more clunky and less scholarly, but still. Gråbergs Gråa Sång (talk) 12:46, 23 December 2017 (UTC)

What's all this I hear about violins in the Bible? EEng 13:37, 23 December 2017 (UTC)

I don´t know about violins, but there´s a lot of cattle in the Bible so they might need more cowbell. Gråbergs Gråa Sång (talk) 14:04, 23 December 2017 (UTC)

A lot of cattle. —David Eppstein (talk) 02:54, 27 December 2017 (UTC)

...that is a lot, yes. Gråbergs Gråa Sång (talk) 08:13, 27 December 2017 (UTC)

Just "philosopher Eleonore Stump" (or "philosophy professor Eleonore Stump" if there's some need to distinguish Stump from one of "the" philosophers like Nietzsche or Wittgenstein) is sufficient. Total lack of context is wrong for this medium/audience; it mostly happens in science articles, when people used to galloping attribution of the form "Jackson 2013 demonstrated that the blasting of vent doors by flux capacitor cycling is 7% less efficient" in their professional writing, with no reference at anywhere to who this is except in a citation. (It's better that they include that much than nothing, since it's still citing a source, but it's a crappy way to do it in an encyclopedia unless we've already recently cited Jackson, by full name, and Jackson(s) has/have more than one work being cited in our article).

But "philosopher, professor and author Eleonore Stump" is pointlessly redundant. There are virtually no professional philosophers who are not academics (those that are are better known as authors than philosophers), being a professor isn't special or contextually important anyway (even being the Alfred B. Ceesdale Professor of Philosophical Inquiry, or some other fancy-named endowment position, isn't important for our purposes, except in a bio on that person), all academics "publish or perish", so we would not be quoting this person if not an author of at least a bunch of peer-reviewed papers (there'd be no way for us to ascertain their in-field reliability as a source if they had no publication history).  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  16:17, 23 December 2017 (UTC)

There are virtually no professional philosophers who are not academics – what about the many professional philosophers working in private industry? EEng 22:51, 23 December 2017 (UTC)

They're professional something-elses. I don't know of a company with "philosopher" or "VP of philosophy" in its org chart anywhere. That is, if they're notable, they'll be notable for something way more specific – as CEOs, as human–computer interface design experts, as university management consultants, or whatever the case may be. (Or maybe you're being funny. I can't tell when I've just woken up.)  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  16:49, 24 December 2017 (UTC)

I do know of a former company that hired professional philosophers. (More specifically ontologists. My wife worked there, as a programmer not as a philosopher.) It lasted surprisingly long before folding. —David Eppstein (talk) 01:50, 26 December 2017 (UTC)

A month? EEng 01:58, 26 December 2017 (UTC)

Good guess, but I think it was around 16 years. —David Eppstein (talk) 02:09, 26 December 2017 (UTC)

So it would seem. EEng 01:37, 26 December 2017 (UTC)

Me too. Full name on first use seems respectful to the reader and the person being mentioned. The, especially older, scientific standard of "Jackson 1913 said ..." feels like there's a club that's all on a last name only basis with each other, and if you aren't well known enough to be known only by last name, then you don't count. It's especially a mess with the last name is "Li" or "Smith".  SchreiberBike | ⌨  17:49, 23 December 2017 (UTC)

Even aside from that psychology, the context-free "Jackson 1913"-on-first-use style – and similar ones like "Jackson (1913)", or a "(Jackson 1913)" at the end of the statement – have been favored in journals to save space, but WP isn't paper; we favor clarity over compression. The academic journal reader may already know which Jackson is intended, and may well have already read what's being cited, while they don't care much until they decide to dig into the citations in earnest; they're more apt to care most about the overall theoretical and methodological framework of the piece they're reading, which is by a specific author or team. Our readers (when they care about citations) tend to care on a point-by-point basis (our articles are palimpsests stirred together by a global assortment of geniuses, crackpots, and everyone in between, sometimes citing great stuff, sometimes poor stuff, and sometimes nothing), and they aren't likely to know any author from another.  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  16:49, 24 December 2017 (UTC)

You might be interested in reading the article False title.  – Corinne (talk) 00:31, 26 December 2017 (UTC)

"False title" is pure BS predicated on the false logic that anything that precedes a name is a "title". I was waiting for someone to link to this—I'm surprised it took more than three or four comments. Curly "JFC" Turkey 🍁 ¡gobble! 01:15, 26 December 2017 (UTC)

Ever notice how the "false title" snobs are always the same people freaking out about infoboxes? But don't tar Corinne with the same brush – no doubt she was merely edifying her fellow editors. EEng 01:34, 26 December 2017 (UTC)

Well, our article on it seems well sourced enough. This is a "teach the controversy" matter. It's clear that some people "out there" are writing about (and objecting to) that style of writing, and have been for around a century. It's good that we have an article on it. The prescriptivism latent in it, however, isn't something MoS should adopt or editors here should try to enforce against. It's normal English these days when writing in tight prose. E.g., I would happily use it in a lead section when describing a peripheral figure, but use the "the" form in the longer article, unless it resulted in some kind of awkwardness. I recall a discussion about two years ago (in article talk) about describing someone as married "to the painter John Doe" (whatever his name was) rather than "to painter John Doe", with the conclusion being that the former was awkward implying, "to the painter John Doe" like he was extra-special or something. So, depends on context. A good solution is to provide more of it (e.g. "to John Doe, a painter from Saskatoon").  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  05:32, 30 December 2017 (UTC)

My sense is that the professional encyclopedic tone is "the American philosopher Eleonore Stump..." in the first instance, followed by "Stump" if she needs to be referenced again. If appropriate, something like "and Biblical scholar" could be inserted before her name, as well, in the first instance. I personally dislike the "Stump (2016)" thing in the flow of prose and feel that such should be consigned to footnotes, but I realize it's still common in certain fields.  White Whirlwind  咨  10:00, 28 December 2017 (UTC)

Agree with "followed by" of course, but why bother with nationality in general? There may be places where it matters, but in general/here? IMO it´d be more relevant to note, in a religous context, if they are a professor from say Pensacola Christian College. Gråbergs Gråa Sång (talk) 11:47, 28 December 2017 (UTC)

Agreed with both of you in this case– i.e., yes to WW's general approach to introducing Stump, but also yes to GGS's quibble about nationality, unless there's a particular contextual need for it; there might be in, e.g., an article on Canadian politics and an American commentator on it.  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  05:32, 30 December 2017 (UTC)

Please see: Wikipedia talk:Categorizing redirects#Request for comments on MoS shortcut redirect categorization.
 — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  08:30, 23 December 2017 (UTC)

Please direct some attention to this, in particular the idea of tagging all the MoS guidelines, except WP:Manual of Style itself, with ((R to subpage)), which has potentially non-trivial consequences [2].  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  11:11, 30 December 2017 (UTC)

• Honestly I don't understand the issue, but I'll be happy to support whatever you come up with, because you're usually right on stuff like that. EEng 18:43, 31 December 2017 (UTC)

Should we add in and display the year that a marriage ends when the subject of the biography has died. Or should we not display, or remove the end date for the marriage. This was originally presented at the template page and the RFC was closed and the instruction were to reopen it here. The issue arose because some editors are mass removing the end date for marriages, when the end date is when the subject of the biography has died.

Option 1 with displayed end date

• Y This should be the way to do it. It removes the ambiguity that the marriage is still a legal entity beyond death, or that the template has not been updated with the information that one partner has died. See below: With no end date it leaves the impression that they may have been divorced before death, or that the other partner died first, but that no one has done the research. Option 2 is inherently ambiguous. --RAN (talk) 18:20, 17 October 2017 (UTC)
• N Option 1 can be misread that Salma Rogers is a man who died in 1960, or that "his death" is a typo for "her death", or that Salma may have died first but we don't know her death date, or that Salma and Albert were divorced in 1956 but that no-one has added the date or done the research, etc., etc. The arguments for favoring option 1 are logically flawed: the format does not address the problem and provides no more clarity than the other option. DrKay (talk) 19:24, 25 October 2017 (UTC) Amended 21:28, 27 October 2017

The wording in the template can always be changed later if that is your problem. If you can think of better wording that provides more clarity, let us all know what that is. The current wording was from a previous RFC, no one declared it the ultimate super best, just the most appropriate offered at the time. It can just as easily appear as "end=Brock's death" with an additional RFC, this RFC is about adding the end of marriage date. --RAN (talk) 22:45, 2 January 2018 (UTC)

• Y This option has the greatest clarity. - PKM (talk) 17:54, 18 October 2017 (UTC)
• After thinking about it, I'll throw my support here. I really don't know if it should be hardcoded, though. Nohomersryan (talk) 19:32, 22 October 2017 (UTC)
• Support As argued above, the benefits of clarity to all readers, regardless of their experience with such conventions, provides the greatest benefit here. Alansohn (talk) 19:34, 25 October 2017 (UTC)
• Y ChristianKl (talk) 16:49, 27 October 2017 (UTC)

• Support This has the most clarity and explains remarriages more easily. --Trödel 03:37, 28 October 2017 (UTC)
• Support This option is the clearest for readers. Jklamo (talk) 10:24, 31 October 2017 (UTC)

• Support When we have an infobox, we must have an infobox. Meaning that the typical, tired argument about "duplicating info already in the article" is null and void. This option goes for the necessary infobox-information in a short and easy manner. Clearly preferable. -The Gnome (talk) 09:10, 3 November 2017 (UTC)

• It's not just about duplicating info already in the article, but already in the infobox. Per MOS:INFOBOX, we should avoid such duplication. Nikkimaria (talk) 14:46, 3 November 2017 (UTC)

• Please quote the sentence you are referring to in MOS:INFOBOX. Asking me to read the bible to find the answer is not useful in a discussion. The word "duplication" does not appear in the text. You have a habit of doing this during discussions. --RAN (talk) 19:37, 3 December 2017 (UTC)

• Referencing relevant policies/guidelines that support a position is appropriate in consensus-building discussions. Per the particular page I cited: "The less information [the infobox] contains, the more effectively it serves that purpose, allowing readers to identify key facts at a glance... wherever possible, present information in short form, and exclude any unnecessary content." Nikkimaria (talk) 19:45, 3 December 2017 (UTC)

So when the infobox is empty, it has reached perfection. --RAN (talk) 05:31, 5 December 2017 (UTC)

"the less information" is a misfortunate wording. "The more concise the information" better aligns with the spirit. Curly "JFC" Turkey 🍁 ¡gobble! 05:40, 3 January 2018 (UTC)

• Support gives most clarity this way. --SarekOfVulcan (talk) 21:39, 2 January 2018 (UTC)
• Comment. The 'his'/'her' text, especially if wired into the template, assumes that partners in a marriage can be distinguished by gender. How will this work when George Takei dies, assuming his husband outlives him? Pburka (talk) 22:15, 2 January 2018 (UTC)

I told you this gay marriage thing would lead to trouble. EEng 22:21, 2 January 2018 (UTC)

"end=Takei's death" or "end=Altman's death", the template accepts any text. It is just a matter of adding some text to the template guide for same sex marriages to explain what to do, would you like to volunteer? The end= parameter and the wording was added after a previous debate, this debate is about adding the date that a marriage ends, we can always go back and relitigate the text portion. If George Takei dies in 2018, won't you feel a little bit responsible for using him as an example? -RAN (talk) 22:39, 2 January 2018 (UTC)

What happened to widowed? (e.g. Jane Jones (m. 1930; wid. 1935)) Is it not gender neutral enough? AngusWOOF (bark • sniff) 23:09, 2 January 2018 (UTC)

Another good wording, I do not remember why it did not become the default. --RAN (talk) 00:22, 3 January 2018 (UTC)

Option 2 with no end date

• Y Since the death date of the article subject will be given already in the infobox and marriages obviously end on the death of the subject, there is no need to repeat the death date of the subject. Infoboxes should be simple and should not unnecessarily duplicate information. DrKay (talk) 19:21, 17 October 2017 (UTC)
• This seems a nice, simple approach. Nikkimaria (talk) 18:31, 19 October 2017 (UTC)
• Support. The version above this one is needlessly long-winded.  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  11:14, 20 October 2017 (UTC)
• Agree This seems simpler than option 1, and any missed detail will be contained in the article anyway. Sam Wilson 00:46, 28 October 2017 (UTC)
• Support; simpler version, clearly preferable. Happy days, Lindsay^Hello 18:13, 8 November 2017 (UTC)
• Support marriages in almost all cultures are understood to be until death. Adding d. 1940 would imply the spouse died the same year, perhaps before the person's death. AngusWOOF (bark • sniff) 23:12, 2 January 2018 (UTC)

Well, not Mormons. EEng 23:15, 2 January 2018 (UTC)

• Support; In the case that the surviving (if any) spouse remarries, there needs to be an end date. But otherwise, some (not necessarily me) might consider the marriage continuing into (religiously preferred afterlife). With a date, it looks like someone divorced on the day of death. If a surviving spouse does remarry, I suppose the day of death makes sense. Gah4 (talk) 23:17, 2 January 2018 (UTC)
• Support: there does not seem to be any need to specify why a marriage ended in an infobox. The 'death' line would imply the termination of such partnership. Their spouse's death should be covered elsewhere in the article (usually a 'personal life' section). –Sb2001 00:26, 3 January 2018 (UTC)
• Support: Un-necessary repetition otherwise. Only in death does duty end (talk) 10:39, 3 January 2018 (UTC)
• Support per others above. Ealdgyth - Talk 14:53, 4 January 2018 (UTC)

Option 3 - allow flexibility

Neither mandate nor disallow end dates... neither encourage nor discourage not having end dates. Let editors decide what is best on a case by case, article by article basis. Blueboar (talk) 23:32, 2 January 2018 (UTC)

• Problem then is, if a halfway observant reader, going from bio article to bio article, notices the differences in infobox info, they'll be apt to think they're missing something, attempt to figure out what the difference is, and realize that it's simply because a different person wrote the article and that, through some seriously misappropriated terminology, templates are not uniform here. No interest in the specific outcome beyond its occurrence. Primergrey (talk) 04:49, 4 January 2018 (UTC)

• Support - as well as whether or not to use the ((marriage)) template at all, and which parameters to use if so. These choices do not belong in our MoS, they belong in the template documentation; this should not be changed. Notwithstanding the OP's decision to omit the majority of the template's functionality, in choices given, remaining flexible allows many more possibilities, like, for example, the ones shown at right.--John Cline (talk) 04:25, 3 January 2018 (UTC)
• Support—let's keep the MoS out of this ridiculous micromanagement—this is almost always best left to editorial judgement, and when not, it should be left to talk page consensus. Curly "JFC" Turkey 🍁 ¡gobble! 05:34, 3 January 2018 (UTC)

Discussion

• Question: who actually started this thread? I think a signature to close the opening statement would be beneficial. –Sb2001 00:29, 3 January 2018 (UTC)
• In most biographies, the marriage is completely irrelevant to the notability of the subject and should not be included in the infobox at all. —David Eppstein (talk) 02:05, 3 January 2018 (UTC)

  In nearly all biographies, the date of birth is also "completely irrelevant to the notability of the subject" but it is certainly useful information, and therefor, rightly included. There is no reason to assume a notable subject's marriage can not be informative as well, and, IMO, no excuse for teaching that it can not. Respectfully--John Cline (talk) 05:20, 3 January 2018 (UTC)

  For dead people it is not an issue. For living people, non-notable third parties should not be named in the article (let alone the infobox) unless it can be shown they had a significant impact on the subject. (Granted all of the examples above are about marriages ending due to death, so unlikely to be relevant) Only in death does duty end (talk) 10:37, 3 January 2018 (UTC)

  And in others (e.g. Art Spiegelman and Françoise Mouly) it is central to their biography. But the discussion is not about forcing or disallowing marriage data in infoboxes, so the comment is irrelevant. Curly "JFC" Turkey 🍁 ¡gobble! 05:31, 3 January 2018 (UTC)

Is there a guideline on how to format interwiki hatnotes, and should they exist? The article Handegg (Swiss village) contains such a hatnote linking to to wikt:handegg (humorous term for American football, etc.). Did I format it properly? WP:HATNOTE says not to pipe non-disambiguation links, though I think it would be better to auto-format these links in Module:Hatnote (e.g. "For the English word, see handegg at [the English] Wiktionary"). Jc86035 (talk) 11:05, 5 January 2018 (UTC)

Seems entirely reasonable to me. Get people where they want to be, with clarity. This is way preferable to using one of those interwiki box templates (which would only be appropriate if it were the same topic, and the Wiktionary material on it were substantive). Some are apt to object to such a hatnote at all, though. I think they should reconsider. Things like this will help prevent people from creating WP:NOTDICT-violating micro-stubs on non-notable slang and neologisms.  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  05:18, 6 January 2018 (UTC)

I hope that you folks can help (have searched the archive). I regularly create aircraft engine articles, the subjects are often museum exhibits only or no longer exist at all. In the lead first sentence I always use the past tense as this seems natural to me. This is 'corrected' to present tense as if I have made a mistake. If the subject is still in service or even limited use in retirement I will use present tense.

Reading the guideline wording of do not use past tense except for dead subjects, past events, and subjects that no longer meaningfully exist as such I would regard an aircraft engine in a museum as not meaningfully existing as it is no longer running, turning a propeller or producing thrust and powering an aircraft.

Does this wording need adjusting for clarity? I've looked at some other random museum items (retired/defunct etc) and they all seem to use the past tense in the first sentence, e.g. Stephenson's Rocket, North American XB-70 Valkyrie, Space Shuttle Enterprise, RRS Discovery and the Wright Flyer, all subjects that no longer operate as they were originally designed to do ('retired' in less words!). Cheers Nimbus (Cumulus nimbus floats by) 07:27, 3 January 2018 (UTC)

• It seems a little odd, but what if there were an article about the Foo Exhibit? An article that would focus, at least secondarily, on a thing that obviously exists, no matter how tenuously, would have to refer to it in the past tense. I do a lot of tense-related copyediting and I see the fairly strong language in the current wording as a bright line. If there's one around, anywhere, it gets present tense. Primergrey (talk) 02:04, 4 January 2018 (UTC)

  So the articles I linked are incorrect then? There are clearly other editors that think the same way as me. On the Foo exhibit I would agree that many objects could always be referred to in the present tense, a sculpture or painting was intended to be looked at by its creator so it is still 'meaningfully existing'. A piece of engineering machinery (ship, aircraft, car, engine etc) once it is in a museum with none left working is not 'meaningfully existing'. Perhaps it's an engineer's/operator's thing along the lines of vessels having female gender, if a device is broken or stops working it immediately becomes 'was' in my eyes, its 'soul' has gone like a person passing away. There's a nuance in the grammar that is being overlooked. Nimbus (Cumulus nimbus floats by) 12:11, 4 January 2018 (UTC)

  Really, the whole "meaningfully" thing is problematic anyway, as it is in the editing guideline, but the end result, in the article, is "The Foo was this and that. A well-preserved example is housed at..." This reads, correctly, as contradictory. There are nuances here, but if they don't get conveyed through article prose, then we are simply navel-gazing. Primergrey (talk) 15:06, 4 January 2018 (UTC)

  The "Foo Exhibit" is not the same as "Foo" itself. You could say that "the Foo Exhibit is a museum exhibit focusing on foos. Foos were an item used for bar." (bar being the function of the foo) This would make it clear that the foos themselves no longer meaningfully exist, but that relics are available for viewing.--Khajidha (talk) 13:23, 4 January 2018 (UTC)

  I see your point, but I think that if something has existing examples and a museum exhibit, it has a meaningful existence. Primergrey (talk) 14:51, 4 January 2018 (UTC)

  Not sure how you figure that non-used examples and a museum exhibit equals a "meaningful existence". By those criteria I could claim that Vladimir Lenin has a meaningful existence because his unused body is still on display. --Khajidha (talk) 15:04, 4 January 2018 (UTC)

  Vladimir Lenin's body has a meaningful existence. Primergrey (talk) 15:09, 4 January 2018 (UTC)

• "meaningfully exist" is more reasonably interpreted as "destroyed", not "decommissioned and now used only in an exhibit". "X is an aircraft engine that was available in Y timeframes used in Z aircraft" is a fairly easy rewording for the case where you could walk into a museum and find the engine. If you have a reasonable belief that no engines of the model/class remain (where "reasonable belief" might be translated as "reliable sources tell you all of them were scrapped"), then I don't see an issue with "X was an aircraft engine (... available in Y timeframe)". I would suggest similar rewordings for a number of those example articles. --Izno (talk) 14:20, 4 January 2018 (UTC)

  Sorry, but ""meaningfully exist" is more reasonably interpreted as "destroyed", not "decommissioned and now used only in an exhibit"" is the exact point being discussed. Your pronouncement that this is so is no more valid than Nimbus227's pronouncement of the the converse. To me, the idea that a disused piece of technology no longer "meaningfully exists" is much more reasonable than your point of view. --Khajidha (talk) 14:44, 4 January 2018 (UTC)

  Let's look at another sphere: A computer program which is still known to exist, but which I can't use because I don't have the correct technology, does not cease to "meaningfully" exist just because I can't install and use it on my modern-day machine. It ceases to meaningfully exist when the only record we have of it is in writings rather than some potentially installable form. An engine model is no different: it only ceases to exist meaningfully when all of the class of engines no longer exist. Just because they were "retired" (i.e. no longer installed in an aircraft and no longer intended to be installed in an aircraft [because X technology deprecated it, etc.]) doesn't mean they don't work (I bet any number of "retired" engines could be tinkered with a bit and subsequently function as originally intended). --Izno (talk) 14:51, 4 January 2018 (UTC)

  Is your inability to install the program indicative of the fact that machines that could run the program are no longer used or only that you, yourself, do not have such a machine? The first seems not to "meaningfully exist" to me, while the latter might mean only that you have the most up to date technology while the program is still commonly used by many others with less up to date technology. --Khajidha (talk) 15:00, 4 January 2018 (UTC)

  Thought provoking stuff. I've had a look at the 'factsheets' from several museums including the Royal Air Force Museum, London Science Museum and the National Museum of the United States Air Force. They tend toward using the past tense for their retired aircraft and engine exhibits, which is my preferred method. In some cases they are using past tense for types that are still in limited use which I also wouldn't disagree with. Quite cleverly they avoid the use of tense at times by constructing their object descriptions in a different way. The MOS:TENSE guideline either needs clarifying considerably or amending to say that either style is acceptable. The nutshell box at the top of the page indicates that exceptions may apply (IAR) if all else fails. I have two more articles in the pipeline and would be reluctant to create them if they are going to be 'corrected per the guideline' the instant they are published, I find it very annoying unfortunately. Nimbus (Cumulus nimbus floats by) 17:41, 4 January 2018 (UTC)

  Writers always dislike it when the editorial staff of their publisher “corrects” their writing. It shouldn’t keep you from publishing. Granted, our “editorial staff” is self appointed... but then ... so are our writers. Blueboar (talk) 20:14, 4 January 2018 (UTC)

  Anyone who won't write an article because it is likely to get copyedited in line with our MOS (or for a multitude of other reasons) lacks the collaborative spirit this website relies on for success. Primergrey (talk) 00:59, 5 January 2018 (UTC)

  And they wouldn't get published by traditional publishers either. Heh.  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  01:20, 5 January 2018 (UTC)

"Lenin's body has a meaningful existence." Exactly. It's perfectly fine to say that Challenger was a space shuttle and that parts of it are in a museum. If the entire something is in a museum, and still in working or potentially working condition (not just a shell or a fragment), though, it should probably remain in present tense. Cognitive dissonance about this can be resolved by sensible rewriting. Wright Flyer is a perfect example: The Wright Flyer (often retrospectively referred to as Flyer I or 1903 Flyer) was the first successful heavier-than-air powered aircraft. It was designed and built by the Wright brothers. They flew it four times on December 17, 1903, near Kill Devil Hills, about four miles south of Kitty Hawk, North Carolina, US. Today, the airplane is exhibited in the National Air and Space Museum in Washington D.C.. Analyze it: Past-tense claim about something it did. Past-tense claim about something they did. Ditto. Present tense claim about what the plane is up to right now. There is no problem there. PS: The style on that needs fixing; it should be '''''Flyer'' I''' or '''1903 ''Flyer'''''; the name of the aircraft did not change.  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  01:29, 5 January 2018 (UTC)

I am very open to copy editing of my actual mistakes, I strive very hard not to introduce them. I would disagree that I lack collaborative spirit, it should be apparent that I have created many articles and happily left them to be edited, indeed they are often short on text and I hope that they will get expanded, I am very aware of WP:OWN having been accused of it several times. I have also uploaded many photographs for free use and am only too happy if someone improves them through editing. This tense guideline is a guideline but appears to be being enforced as a policy which is worrying considering the wording is open to interpretation (editors commenting above clearly read it in different ways). I'm afraid that most retired engine types are so dull and boring that it is very difficult to write the first sentences in the manner suggested for the Wright Flyer. Nimbus (Cumulus nimbus floats by) 20:30, 5 January 2018 (UTC)

So write it however you want and let someone else edit it in that manner. Nothing is being "enforced as a policy" unless "enforced" means "someone coming along and making style-related changes to an article to bring it in line with this site's style guidelines". No one is blocked for writing in a different house style, but people have been blocked for disruptive reversions to MOS-compliant edits. Primergrey (talk) 21:39, 5 January 2018 (UTC)

To comply is to acquiesce. Wikipedia has an editor retention problem, blanking the educated foot soldiers' opinions and experience might be related. Nimbus (Cumulus nimbus floats by) 22:02, 7 January 2018 (UTC)

Unless you are speaking literally, I don't know what you mean by "educated foot soldier". Regardless, as I said before, any editor that cannot work collaboratively is one not worth retaining. No article worth a damn has ever not been proofed and copyedited. Here or anywhere else. Primergrey (talk) 22:22, 7 January 2018 (UTC)

Just to be clear, you would rather I leave the project than discuss a problematic guideline? Nimbus (Cumulus nimbus floats by) 23:38, 7 January 2018 (UTC)

I was addressing the "editor retention problem" that you had mentioned during this discussion. But I wouldn't want to be the cause of anyone's martyrdom so, no, I would rather you did not leave the project. In fact, if it's causing you this much consternation, I would suggest, again, that you not comply with what is looking like the consensus interpretation of MOSTENSE. Just let other, equally well-meaning, editors tweak it (not correct it) to this site's house style. Primergrey (talk) 00:00, 8 January 2018 (UTC)

Some of the most frequent comprehension-impeding errors I encounter and correct could be addressed in a small section. I think it would ease some copyediting load in the long run (and of course make the encyclopedia text better for readers in innumerable places); probably goes right after MOS:SERIAL, but it could also live in MOS:LISTS, and just be summarized in the main MoS page without the example table:

[End proposed material.]

Quite short as to rulemaking, though the examples are rich. These are just two real examples from about an hour ago (some details compressed for brevity) – in otherwise very well-written articles. The journalistic and fiction-writing aversion to the serial comma is poisonous. While no one is confused by the lack of one in a very simple construction like "My dog strangely likes pistachios, potatoes and eggplant", there's no question that it's needed in these kinds of constructions.

Yet only two days ago, I had someone tendentiously revert-warring against a comma despite patently obvious ambiguity (that was a case of "So-and-so guitarist was a member of the Chicken and Dumplings, Doodah and the Snorkel Weasels" (whatever the actual band names were); only someone familiar with the bands would be certain this meant three bands (C&D, D, and SW). The reverter's excuse? As always: "MoS doesn't say it's required."

— Preceding unsigned comment added by SMcCandlish (talk • contribs)

Oh boy, Oxford comma wars. *gets popcorn* —David Eppstein (talk) 17:39, 6 January 2018 (UTC)

Did you say "comma wars"? Will go dust off my uniform from the past skirmishes. Better yet, did you say "popcorn"! (p.s. nice work, would happily support this proposal.) Randy Kryn (talk) 20:45, 6 January 2018 (UTC)

Could be worse – could be an Oxford coma. EEng 15:08, 7 January 2018 (UTC)

• SM, does the lowly semicolon have no place in your toolbox of disambiguators? EEng 22:35, 7 January 2018 (UTC)

  Argh! Yes, you know I wallow in the semicolon. I left it out by accident due to copy-pasting and whatnot while building that table. I've put it back in.  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  20:07, 9 January 2018 (UTC)

When a list of areas served is present in an article's infobox, should Puerto Rico be listed if the United States is already on the list? Daylen (talk) 00:02, 7 January 2018 (UTC)

Can someone activate our team of rabbis, please? EEng 22:28, 7 January 2018 (UTC)

I would say “yes”... most readers don’t think of Puerto Rico (or other US Territories) when they see a reference to th US... they only think of the States. Alternatively, the listing can read: “United States and its Territories”... or something similar. Blueboar (talk) 23:46, 7 January 2018 (UTC)

Agreed. It'll depend on context.  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  20:08, 9 January 2018 (UTC)

Can we get a clarification on how MOS:LQ should be interpreted with regards to including terminating punctuation without quotemarks? For example, given two source sentences:

(1) "Curly Turkey can really get under people's skin when he gets going."

(2) "I really hate it when Curly Turkey opens his beak; he gets so annoying."

Would we quote them in a Wikpedia™ article in the following manner:

(A) Critic A stated Curly Turkey could "really get under people's skin". Critic B concurred, stating that Curly Turkey "gets so annoying".

or:

(B) Critic A stated Curly Turkey could "really get under people's skin". Critic B concurred, stating that Curly Turkey "gets so annoying."

—Curly "JFC" Turkey 🍁 ¡gobble! 11:42, 4 January 2018 (UTC)

Discussion

• Obviously, LQ requires the first quote fragment to have its terminating punctuation outside the quotemarks. Logically, the period punctuates the enclosing sentence, not the quoted fragment. Following that logic, the second sentence would also require the period outside, but some^[who?] argue that, because the original had a period there, the quoted fragment should also include the period inside the quotemarks. Problems with that:
  • It ignores the "logic" in "logical quotation"—the period terminates a complete sentence, not a sentence fragment.
  • It results in a hodge-podge of in-and-out styles throughout an article that:
    • makes no sense from a reader's perspective
    • can be verified or resolved only by consulting all the sources
    • may motivate reader/editors to "fix" the article by putting all the terminating punctuation inside (breaking text–source integrity), or placing them all outside (a violation of MOS:LQ if MOS:LQ truly requires the original terminating punctuation to be included within the quotemarks).

Is there really any plausible reason to prefer (or even allow) (B)? Regardless, how can MOS:LQ be amended to make whichever interpretation clear? Curly "JFC" Turkey 🍁 ¡gobble! 11:42, 4 January 2018 (UTC)

• LQ starts by saying, without qualification, Include terminal punctuation within the quotation marks only if it was present in the original material, and otherwise place it after the closing quotation mark, then goes on to give a bunch of examples which in many cases contradict that. If you LQ zealots would get your act together and come up with a guideline that actually makes sense, is self-consistent, and is not too far off from they way normal people here on earth actually write, I'm sure most editors will do their best to follow it. In the meantime I, for one, will just do what that first sentence clearly tells me to do. EEng 12:08, 4 January 2018 (UTC)

  Now I understand: it wasn't a dispute over interpretation of LQ, but your opposition to LQ that led to the dispute. Curly "JFC" Turkey 🍁 ¡gobble! 22:53, 4 January 2018 (UTC)

  No, you don't understand. It was about the impossibility of making sense of LQ's contradictory provisions. EEng 23:26, 4 January 2018 (UTC)

  It's awfully straightforward. I'm sorry you're having difficulty with it. Curly "JFC" Turkey 🍁 ¡gobble! 00:09, 5 January 2018 (UTC)

• Seems to me that we could make it clearer just by removing the sentence EEng quoted. --Khajidha (talk) 16:05, 4 January 2018 (UTC)

  Then no one would understand it it all. It would just be a bunch of examples without any principle to follow. If we removed or changed anything, it would be an allegedly confusing example, but none have been identified. What seems to be happening here is that EEng is extrapolating beyond what it actually says and arriving at something like "Include terminal punctuation within the quotation marks if it was present in the original material, no matter what, and place all other punctuation after the closing quotation mark", or something to that effect, though it's not entirely clear, since he's just objecting without being specific. The only likely clarification I can see to make is to reverse the order of these two sentences one one line, about full versus fragmentary quotes, so that we have: If the quotation is a single word or fragment, place the terminal punctuation outside the closing quotation mark. If the quotation is a full sentence and it coincides with the end of the sentence containing it, place terminal punctuation inside the closing quotation mark. We should do that because we more often quote fragments than full sentences, so people should see and absorb that part first and foremost.  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  02:08, 5 January 2018 (UTC)

• My interpretation of LQ is that we put inside the quotes the things that we intend to quote, and not the punctuation that is part of our framing sentence. In this example, what we intend to quote is "gets so annoying". In most contexts it would not make sense for what we quote to be extended by one character to "gets so annoying." (this sentence being an exception to the "most contexts") because that longer quote is less-grammatical, having a period after something that is not a complete sentence. So in this case the clear choice is A over B. The weird part is that when a sentence ends on a quoted sentence, we omit the period from the framing sentence and let the quoted period do the job. Instead, we should have both; for instance, "This is a quoted sentence.". It looks uglier that way but it would be more logical. —David Eppstein (talk) 19:03, 4 January 2018 (UTC)

  There was actually one (and only one, as far as I've been able to determine, with almost every style guide ever) that actually advised this. So, it's not totally unattested, but incredibly rare, presumably because it's too pedantic for most people to tolerate. Technical writers would still do this when quoting code in which the . at the end wasn't a full stop/period but a different meaning of the glyph, and that's fine. But it would probably be better to rewrite.  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  02:00, 5 January 2018 (UTC)

  Unless it can be demonstrated to solve an actual, real-world problem, there's no way it will gain traction at Wikipedia. Curly "JFC" Turkey 🍁 ¡gobble! 02:23, 5 January 2018 (UTC)

  • Well, lets keep the discussion to that of sentence fragments, as the quoting of compete sentences is a different issue; your solution doesn't apply to this one. Curly "JFC" Turkey 🍁 ¡gobble! 22:53, 4 January 2018 (UTC)
    • Ignore the last two sentences of my comment if you wish, but all the rest is directly relevant. —David Eppstein (talk) 01:14, 5 January 2018 (UTC)
      • Sure. Curly "JFC" Turkey 🍁 ¡gobble! 02:09, 5 January 2018 (UTC)
• Both of them would go outside, since they're just fragments, and our terminal punctuation belongs to our entire sentence and is only meaningful in that fuller context; the string "really get under people's skin." by itself is meaningless. LQ forbids insertion of punctuation not in the original, as blatant misquotation; it does not require retention of terminal punctuation that was in the original when it doesn't make sense to do so. Otherwise no excerpting would be possible at all. I.e., there is no difference between taking an original statement, "Janet's cat is orange and ugly.", and quoting it as "He said her cat is "orange and ugly", which I though was funny.", versus "He said her cat is 'orange and ugly'." But include it inside the quote here: "He said: 'Janet's cat is orange and ugly.'" Leaving it out when quoting a full sentence strongly implies we're quoting a fragment. But if we're quoting a fragment, the exact same principle that allows us to partially quote without all the words also allows us to partially quote without terminal punctuation that doesn't make sense inside the fragment when the fragment is viewed on its own. A third way of putting it: If I write "My cat is loud, ha ha", you would not partial-quote it as "McCandlish said 'My cat is loud,'." – desperately retaining the comma as part of the original construction.

  Re: the claims that MOS:LQ "then goes on to give a bunch of examples which in many cases contradict that" we "[i]nclude terminal punctuation within the quotation marks only if it was present in the original material, and otherwise place it after the closing quotation mark" – I've reviewed it line by line and it does not do that at all. All the examples are correct. (This has not always been the case; people have monkeyed with it before, especially to make it match one of the dozen+ British styles that were inspired by LQ but have diverged from it in inconsistent ways.)
   — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  02:00, 5 January 2018 (UTC)

  Which is how I interpret things, but is this what MOS:LQ actually says? EEng believes it does not, and has placed the punctuation at Sacred Cod back inside the quotemarks, citing MOS:LQ. I've (occasinally) had others interpret MOS:LQ this way (which is why I opened this discussion), although not to the point of reverting me. Curly "JFC" Turkey 🍁 ¡gobble! 02:09, 5 January 2018 (UTC)

  After just re-reviewing it, yes that is what it says. I think flipping the order of two sentences and examples [3] will resolve it.  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  02:54, 5 January 2018 (UTC)

  (edit conflict against SM's post just above, and maybe that fixes it, I don't know) Look, LQ says Include terminal punctuation within the quotation marks only if it was present in the original material, and otherwise place it after the closing quotation mark, period (so to speak). It says if X, do A, otherwise do B. It purports by its wording to dispose of every case that arises. The later examples don't call themselves exceptions to this Prime Directive; they're just there, telling you to do things, some of which (depending on circumstance) contradict the opening instruction. For example, let's say the source says, What Marlin said was, "I need to find Nemo.", and for whatever reason I want to cast this as either Marlin needed, he said, "to find Nemo". or Marlin needed, he said, "to find Nemo." According to the later examples given by LQ, I'm supposed to write the first, but according to the LQ's unambiguous, here's-how-it's-done, no-exceptions-provided-for Prime Directive, I'm supposed to use the second. Thus the contradiction. EEng 03:17, 5 January 2018 (UTC)

  It's a question of what you mean by "the original material" that you're quoting. If you intend the original quoted material to be "to find Nemo." as a sentence fragment with a terminal period, then by all means include the period within the quote. But why would you want to do that? If, as seems much more likely, that what you intend to quote is the unpunctuated fragment "to find Nemo", then the period is not part of the original material that you're quoting and should not be put there merely because it syntactically matches a period in the quoting sentence. —David Eppstein (talk) 03:26, 5 January 2018 (UTC)

  You're not understanding what you're quoting, EEng. The guidline does not say to include punctuation if it's in the original, but only that you can include punctuation only if it appears in the original—meaning: don't through a comma or period in there that wasn't in the original. The "to find Nemo" example at MOS:LQ is totally correct—why do you think it's not? Which part of the guideline do think think makes it ambiguous? Curly "JFC" Turkey 🍁 ¡gobble! 03:35, 5 January 2018 (UTC)

  Yep. I had the same perception, though I sandwiched into the above material: 'What seems to be happening here is that EEng is extrapolating beyond what it actually says and arriving at something like "Include terminal punctuation within the quotation marks if it was present in the original material, no matter what, and place all other punctuation after the closing quotation mark", or something to that effect, though it's not entirely clear, since he's just objecting without being specific.' [4]  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  03:39, 5 January 2018 (UTC)

  The Prime Directive absolutely says If X do A, otherwise do the opposite of A. There is no middle. It does say or imply If X do A, otherwise you don't have to do A; if it's trying to say the latter it does an incompetent job of it. David Eppstein, I appoint you my proxy for the rest of this discussion, to vote my shares for me as if I were personally present. Perhaps SM's changes fixed it. I'll rely on your judgment. EEng 03:49, 5 January 2018 (UTC)

  But where does it say that? None of us sees that. Curly "JFC" Turkey 🍁 ¡gobble! 04:18, 5 January 2018 (UTC)

  What part of Include terminal punctuation within the quotation marks only if it was present in the original material, and otherwise place it after the closing quotation mark don't you understand? David Eppstein, can you toss me a lifeline here? EEng 05:26, 5 January 2018 (UTC)

  Maybe it's the meaning of "only"? To me it means that you cannot add punctuation that was not already there in the quote, but it says nothing about whether you should or should not take as part of the quote the punctuation at the beginning or ending of the quote. You are trying to add extra punctuation at the end of your quote. I don't understand why that string, and not the shorter string without the punctuation, is what you want to quote. —David Eppstein (talk) 05:34, 5 January 2018 (UTC)

  You keep quoting that, EEng, and the rest of us keep telling you you're misreading it. The sentence does not convey the meaning you insist it does—it would if the word "only" were not in it. Are you making sure you're reading the whole sentence before you copy-&-paste it? Curly "JFC" Turkey 🍁 ¡gobble! 07:15, 5 January 2018 (UTC)

  Exactly. The "prime directive" does not say "If X do A, otherwise do B [the opposite of A]" at all. It says "Do A only if X; default to B", and "doing A when X" is not even required, just permissible. See first revision proposal below. It will just end this confusion (which is rare, but apparently hard to shake when it happens) without further question or ado.  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  08:53, 5 January 2018 (UTC)

Potential clarification

This following semi-recent addition is not actually very helpful, and may be the primary source of confusion:

Aside from being repetitive of much of the sentence that precedes it, and vaguely worded, this presupposes that the reader understands all about the niceties of question mark usage, when this actually varies between style guides, and we have a section here instructing editors on our version, so we know they didn't all arrive here with this in their brains already. Worse, question marks themselves are subject to an LQ rule of their own, rendering this statement confusing and meaningless.

It would probably make more sense to just say this:

It's shorter and doesn't presuppose anything, uses clearer wording, and it covers both the He said "Ouch, that hurts." case and the "Ouch," he said, "that hurts." case (it's not all about terminal punctuation), while ruling out "Why," he asked, "did you hit me?" (must be "Why", he asked, "did you hit me?") and "Ouch," was what he said. (it's "Ouch" was what he said.) It also doesn't exclude retention of "!" and "?", which we cover in a later point. We already provide conforming examples; this isn't a substantive change in any way, just a better clarification than that "same way as question marks" attempt.

However, we should ditch the insipid fictional dialogue examples we have. Replace them with structurally identical material that actually pertains to how we write encyclopedia articles, e.g. quoting from non-fiction sources. It would make this all much easier to relate to and apply.
 — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  03:39, 5 January 2018 (UTC)

• I don't think that'll satisfy EEng if the current wording doesn't. Should there not be something explicit about sentence fragments vs complete sentences? Curly "JFC" Turkey 🍁 ¡gobble! 04:21, 5 January 2018 (UTC)

  There already is ("If the quotation is a single word or a sentence fragment, place the terminal punctuation outside the closing quotation mark. When quoting a full sentence the end of which coincides with the end of the sentence containing it, place terminal punctuation inside the closing quotation mark."), and this wouldn't touch it, just prepare people for it. I.e., there is no "exception" as EEng's been conceiving of it; the existing language about fragments doesn't diverge from or contradict the main LQ rule, it's just so clouded with "same way as question marks" blather – before we even get to question marks in LQ at all – that at least a few people aren't getting it. This nutshell replacement sentence should fix that.  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  06:05, 5 January 2018 (UTC); revised 06:16, 5 January 2018 (UTC)

Apology

You're all correct. I was somehow parsing the only if as if, even when it was rubbed in my face. In my (very pale) defense I'll say that the second half of Include terminal punctuation within the quotation marks only if it was present in the original material, and otherwise place it after the closing quotation mark is redundant to the first, and it's natural for the brain to look for two different imports i.e. an implication and its converse. Laugh at me if you will.

Nonetheless, I'm glad to see there's recognition that the presentation needs improvement. May I suggest that the Prime Directive be restated simply as something like Any terminal punctuation inside the quotation marks should be marks present in the material quoted. EEng 18:32, 6 January 2018 (UTC)

Knew you'd come around. Coffee is your friend. I'm thinking on the revision idea. I'm not sure it can be stressed enough how particular we have to be in this section of MoS or people won't understand or will, secretly, but will wikilawyer and look for ways to drive wedges. I don't dig the "marks should be marks" bit. And "the material quoted" is apt to be misinterpreted; this could be misread (willfully) as "If you're quoting something, and using quotation marks (instead of a block quote), then always put the terminal punctuation inside." I.e. use typographer's quotes a.k.a. "American" style.

The original isn't really redundant so much as giving both sides of the coin for extra clarity, and is fixable by juggling words a little, maybe Place terminal punctuation after the closing quotation mark by default; only include it within the quotation marks if it was present in the original material. That seems totally unambiguous to me, especially if we combine it with #Potential clarification above and get rid of the confusing "like questions marks" stuff.
 — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  20:48, 9 January 2018 (UTC)

Potential clarification 2

Moot

 – Withdrawn.  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  20:48, 9 January 2018 (UTC)

Things like this have (not for the first time) inspired me to consider adding something like the following after the fragments and full sentences part: "However, for long fragments that form most of a complete sentence and simply omit an introductory clause, the terminal punctuation may be retained inside the quotation marks." This a) would not be inconsistent with general LQ principles, and b) would agree better with both typical North American and most (not all) British/Commonwealth quotation styles. While it verges on WP:CREEP at first glance, I think it would have a tension-reduction effect that, while subtle, would be widespread and long lasting. Similar to no longer requiring a colon before a quote of a full sentence if it's short. (People just don't like writing XYZCorp's response was: "Not yet." We don't really have a reason to demand it, even if a colon is better as an introduction of longer material. Same principle applies here.)  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  06:16, 5 January 2018 (UTC)

The revised wording looks OK to me. Personally I would prefer to choose my quoted 'fragment' to end before any terminal punctuation, so that the punctuation sits outside, where it belongs. The 'parent' of the sentence in the article is, after all, the quoting editor, who should also give birth to its terminal punctuation; whether the quoted author happened to take breath at the end of his or her contribution isn't material to the meaning of the sentence. To distinguish between he said "this rule is crap." and he said "this rule is crap". on the basis of whether the quoted author concluded his sentence there is somewhat bizarre. MapReader (talk) 10:57, 5 January 2018 (UTC)

I share that view, especially for short quotes, but when the material being quoted is, e.g., "Our study's methodology is [30 more words]." – I mean that there are 30 more words we're actually quoting, not a bracketed thing saying "30 more words" – and we quote it as The study's methodology was "[30 more words]". , then people are apt to fuss about having to keep the period outside the quote. The same thing could be done as "[The] study's methodology is [30 more words]."  I'm just looking for ways to avoid people grinding their teeth, especially when it's about something we don't actually have a strong reason to care about. That said, if no one after a week or so is enthusiastic about this one, then there's no reason to add it, since it is a minor consistency variance. I'm just going by what I see people do fairly often, and what I've seen people revert about in confusion or disagreement.  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  05:32, 6 January 2018 (UTC)

How could the length of the fragment play into it at all? Sitting punctutation inside the quotes based on the length of the quoted fragment (a) is arbitrary, (b) is illogical, and (c) complicates the rule. What is going through these people's heads? Curly "JFC" Turkey 🍁 ¡gobble! 07:42, 6 January 2018 (UTC)

I was thinking the same. It is not misquoting someone to use their exact words, simply leaving the full stop off the end and sticking it quite properly outside the quotation. The only purpose of such edits is as a leisure activity for editors otherwise occupied in scouring articles for double spaces to delete. MapReader (talk) 08:23, 6 January 2018 (UTC)

Plays into it the same way that dropping of the serial comma does in short constructions like "I like metal, punk and industrial music". And also how it plays into dropping of introductory-phrase commas when the phrase is short, as in "In 1998 Smith moved to Botswana." It's not logical from a writing structure perspective, but it is from a mental processing one. I'm entirely happy to not make such exceptions, mind you, I'm just not particularly gung-ho about opposing them; bigger fish to fry. I'll abandon this "Potential clarification 2" idea, though, since the response has been negative so far. My effort to appease some and reduce their tooth-grinding is causing others to grind their own.  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  20:32, 9 January 2018 (UTC)

I often see at Wikipedia that quotes are provided inline without any inline attribution, which is really bad IMHO unless the speaker is clear from the context. Am I correct that this MOS merely requires attribution, rather than inline attribution? If so, I support editing the MOS to require not just attribution but inline attribution, unless the context makes it obvious. Often the attribution isn’t even explicit in the footnote, and so you have to click on a link in the footnote to figure out who made the quoted statement. I don’t think readers should have to go look at a footnote at all, because the body of the Wikipedia article should only include quotes with inline attribution unless the attribution is otherwise obvious. Here’s an example, which I now promise to never edit since I’m seeking a policy change that would affect it. Anythingyouwant (talk) 18:51, 29 December 2017 (UTC)

MOS:QUOTE says you should use in-line attribution to name the speaker of a quote, unless you're talking about a statement made by the person/entity whom is the topic of the article. WP:WHYCITE says any quote or close-paraphrase should have an immediate citation to that source at the point the quote is used. In the diff example, that does need to be attributed as I cannot tell from context who said it. --Masem (t) 19:06, 29 December 2017 (UTC)

Thanks User:Masem, can you please quote the part of MOS:QUOTE? Thanks. Anythingyouwant (talk) 19:46, 29 December 2017 (UTC)

It's the Attribution section "The author of a quote of a full sentence or more should be named; this is done in the main text and not in a footnote. However, attribution is unnecessary with quotations that are clearly from the person discussed in the article or section" --Masem (t) 20:04, 29 December 2017 (UTC)

Hmm, that wouldn’t seem to apply to the example I gave because it’s not a full sentence. Do you know why we require it to be a full sentence or more? Anythingyouwant (talk) 20:40, 29 December 2017 (UTC)

Would anyone object if I change this MOS to say: "The author of a quote of a full sentence or more should be named unless their identity is already obvious in context; this is done in the main text and not in a footnote. However, attribution is unnecessary with quotations that are clearly from the person discussed in the article or section"? Anythingyouwant (talk) 21:47, 29 December 2017 (UTC)

I suspect the present text might have been intended as a bright line to take account of things like

• "She is widely described as 'the Leader of the Free World'." [with a footnote referring to several articles]
• "She met the Dalai Lama for 'private and informal talks'." [with a footnote referring to a press release]
• "She is sometimes referred to as 'the decider'." [with no attribution for the inner quotation]

--Boson (talk) 23:47, 29 December 2017 (UTC)

Thanks Boson, good examples. I think we can take account of stuff like that by saying, "The author of a quote of a full sentence or more should be named unless their identity is already obvious in context, or the quote is not attributable to any particular named author; this is done in the main text and not in a footnote. However, attribution is unnecessary with quotations that are clearly from the person discussed in the article or section"? Does that take care of your examples? Anythingyouwant (talk) 00:41, 30 December 2017 (UTC)Edited01:46, 30 December 2017 (UTC)

Wait... the author should be named unless scare quotes are being used? And when do we use scare quotes, except when they occur inside a larger quote? I'm... confused. EEng 00:53, 30 December 2017 (UTC)

Sorry, I have just removed the bit about scare quotes, it’s unnecessary. Anythingyouwant (talk) 01:46, 30 December 2017 (UTC)

OK, so here's the proposal:

The author of a quote of a full sentence or more should be named unless their identity is already obvious in context, or the quote is not attributable to any particular named author; this is done in the main text and not in a footnote. However, attribution is unnecessary with quotations that are clearly from the person discussed in the article or section

Any comments, objections, snide remarks, glowing praise? Anythingyouwant (talk) 03:50, 30 December 2017 (UTC)

The "not attributable to a named author" bit isn't going to work. Everything we quote is attributable (unless it's quoted in a source we attribute that doesn't attribute who it's quoting, in which case it would be dubious to include it without some special context for it). It's just that sometimes it's not important to attribute a quote inline, usually when it's a string of descriptive material from multiple, similar sources. E.g. 'The film was praised by critics as "edgy and creative"^[1], "darkly humorous while seriously thought-provoking"[2], and "the best thing Jackson has done since My Life with the Weasels",^[3] though also reviewed less favorably for "its self-consciously shaky camera work"^[4] and "overuse of pop music, making some scenes feel more like music videos".^[5]' No one cares which non-notable but reputably published reviewer said what, unless they're digging into the citations.

There needs to be a better way to get at "attribute the quote when it needs to be attributed", basically. What we don't want to see is "She is 'gunning for the governorship in 2020, on a platform that amounts to center-left backlash against ignorant populism'.^[1]" without this opinionated encapsulation being attributed, right there in the sentence, to someone whose opinion our readers might GaF about.
 — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  05:17, 30 December 2017 (UTC)

Well, if the source is one particular author, then that author always needs to be named inline, right? So:

The author of a quote of a full sentence or more should be named, and for a shorter quote that is sourced to only one author or source we also should provide that name; this is done in the main text and not in a footnote. However, attribution is unnecessary with quotations that are clearly from the person discussed in the article or section

Anythingyouwant (talk) 05:35, 30 December 2017 (UTC)

I don't think that quite works for quoted short phrases that should be attributed, but where the decision whether this should be in a footnote or in the main text is left to editorial judgement (e.g. theatre critics, press officers, reporters). --Boson (talk) 13:48, 30 December 2017 (UTC)

One issue in one of Boson's examples: "She met the Dalai Lama for 'private and informal talks'." might need attribution if the context is related to criticism of this hypothetical person. Say this person is a member of Free Tibet or a similar organization but known to support militaristic action against China or the like, so that any time she travels to Tibet, her critics suggest she's plotting some type of violent action (again, hypothetical case here). If that is the context where that phrase is used, that needs inline attribution , like "She met the Dalai Lama for 'private and informal talks', according to her press agency." or something like that to avoid . On the other hand, if we're talking a non-controversial person, like a country's ambassador, the lack of attribution is probably okay, though even there, the need to quote that may be unnecessary.

A lot of this comes down to context. Editors should step back, consider the context, and make sure that the phrase being quoted makes it clear who the speaker(s) may be or immediate sources where they can be found if there are far too many to list, and avoid having the text within the quote appear as controversial statements spoken in Wikipedia's own voice (hence the need for attribution to id the speaker). But it still is all about context; I don't think it is very easy to make exacting rules here, but instead remind editors what the reason for inline attribution is, and methods of how to include it when needed. --Masem (t) 14:06, 30 December 2017 (UTC)

Okay well, User:Boson and User:Masem, here is my last attempt:

The author of a quote of a full sentence or more should be named, and for a shorter quote that is sourced to only one author or source we also should seriously consider providing that name; this is normally done in the main text and not in a footnote. However, attribution is unnecessary with quotations that are clearly from the person discussed in the article or section

That’s more flexible. Personally, I don’t see why it’s not better to name theatre critics inline; likewise for press officers and reporters, the name of their organization (i.e. the name of the source if not the person) ought to be provided inline, IMHO. But if you disagree with this newest proposal, please suggest an alternative. Thanks and Happy New Year. Anythingyouwant (talk) 16:46, 30 December 2017 (UTC)

• I appreciate you're trying to improve the guideline, but we're certainly not going to say anything like we also should seriously consider; that can be fixed of course. My major worry, though, is that we need to address the points SM made: there are times when it really is appropriate to quote material which is either uncontroversial (but expresses a point better than any our paraphrase would) or, even if potentially "opinion", is just one of many similar such opinions whose authors need not be specifically called out. In these cases there's no need for in-text attribution, and the guideline as it is doesn't allow for these cases. That's what needs to be fixed. EEng 01:37, 31 December 2017 (UTC)

User:EEng#s, I find 15 (fifteen) occurrences of the word “consider” already in this guideline, so it wouldn’t be the first such instance. Anyway, if the quote marks in SM’s example were removed it would still not be in wikivoice, so maybe that’s the distinction we need to make. If a sentence uses a quotation and would be in wikivoice without the quote marks, then a source needs to be named. I don't think it's wise for this guideline to suggest that no inline attribution is needed for quotations that are less than a sentence. Anythingyouwant (talk) 02:19, 31 December 2017 (UTC)

Consider, yes – we should seriously consider, no; but like I said that can be probably be fixed. I don't know what you mean by "would [would not] still be in wikivoice". EEng 02:32, 31 December 2017 (UTC)

User:EEng#s, if we remove the inner quote marks from SM’s example, it becomes this: “The film was praised by critics as edgy and creative[1], darkly humorous while seriously thought-provoking[2], and the best thing Jackson has done since My Life with the Weasels,[3] though also reviewed less favorably for its self-consciously shaky camera work.” That is not in wikivoice so I agree no inline attribution is needed (with or without quote marks). But consider this example: “The film was ‘edgy and creative’[1], and ‘darkly humorous’[2], and also ‘the best thing Jackson has done’.[3]” That’s in wikivoice, if the quote marks are removed, so it needs inline attribution of some sort (with or without the inner quote marks). Anythingyouwant (talk) 02:44, 31 December 2017 (UTC)

That's a distinction without a difference. The inner quotes make it clear that it's not wikipedia saying this, but someone else; rewritten slightly as

The film was reviewed as ‘edgy and creative’[1], ‘darkly humorous’[2], and ‘the best thing Jackson has done’.[3]

-- there's nothing wrong with it. Or maybe I'm misunderstanding what you're saying. Please give exact examples of article text, enclosed in the ((tq)) template and set off on their own lines with :, so there's no confusion about the function of the various quote marks. In the meantime, you'll find numerous examples of appropriate use of unattributed quotations in Sacred Cod and Widener Library. EEng 03:10, 31 December 2017 (UTC)

SM said, “There needs to be a better way to get at ‘attribute the quote when it needs to be attributed’”? So there’s no way to do that more comprehensively than this guideline’s present statement about quotes longer than a sentence? Anythingyouwant (talk) 03:18, 31 December 2017 (UTC)

I now have no idea what you're talking about? EEng 03:20, 31 December 2017 (UTC)

User:EEng#s, do you have an idea what SM was saying when he wrote, “There needs to be a better way to get at ‘attribute the quote when it needs to be attributed’”? I think there needs to be a better way than how this guideline does it now. The present guideline only refers to inline attribution of quotes over a sentence in length, whereas shorter quotes often need it too. Are you saying there’s no way we can edit this guideline to address quotes shorter than a sentence? Anythingyouwant (talk) 03:25, 31 December 2017 (UTC)

I'm not sure exactly what he meant there; I'm talking about his prior paragraph. The sentence-length cutoff is obviously arbitrary and shouldn't be there, but as I've tried to show (and, again, I point you to the two articles I linked) it already overprescribes where in-text attribution is needed. EEng 03:30, 31 December 2017 (UTC)

Please take a crack at scaling back the overprescription. If you do so, then maybe you will have come up with a way to address attribution of quotes of any length. Anythingyouwant (talk) 03:34, 31 December 2017 (UTC)

Against my better judgment, I'll try

Here's the old:

The author of a quote of a full sentence or more should be named; this is done in the main text and not in a footnote. However, attribution is unnecessary with quotations that are clearly from the person discussed in the article or section.

I've scrapped the inexplicable "full sentence" criterion. And the "clearly from the person" criterion. In fact, I scrapped the whole thing and just made it refer to NPOV's attribution requirement, which was a pretty clever thing to do, even if I do say so myself. Here we go:

As with all article content, the reader must be able to determine the source of any quotation, at the very least via a footnote. But the source should be additionally attributed in article text if, were it recast as paraphrase, it would need to be attributed per WP:Neutral_point_of_view#Attributing_and_specifying_biased_statements.

I'll point out right now that this is in conflict with weaker than WP:Citing_sources#In-text_attribution, which (in my opinion) greatly overprescribes what should be in-text attributed – it says all direct speech, indirect speech, or close paraphrasing "should" be in-text attributed, and that's clearly wrong. (At least it doesn't say "must" be in-text attributed.) EEng 06:12, 31 December 2017 (UTC)

• Support. Looks like a big improvement to my eye. And I don’t think it contradicts any other guideline or policy, because it doesn’t say that “the source should be additionally attributed in article text ONLY if, were it recast as paraphrase, it would need to be attributed....” Anythingyouwant (talk) 18:33, 31 December 2017 (UTC)

Good point. I've modified my post just above. EEng 18:47, 31 December 2017 (UTC)

Doesn’t anyone want to disagree with me? Anythingyouwant (talk) 00:18, 3 January 2018 (UTC)

Negative. EEng 00:30, 3 January 2018 (UTC)

No disagreement with what you are trying to do here, but I have a few quick copy-edit ideas for you:

1) "As with all content" is not necessary (it's implied).

2) "if, were it recast as paraphrase, it would need to be" isn't quite grammatically correct. The second "it" can be removed, or moved before the word "were".

3) Could we say "were it to be paraphrased" instead of "were it recast as paraphrase"?

4) How about "it requires attribution" instead of "would need to be attributed"?

Warren -talk- 00:32, 3 January 2018 (UTC)

I support Warren's proposed amendment (especially No 2—it was a little clunky to my eyes). –Sb2001 00:44, 3 January 2018 (UTC)

• Oppose. The length of the material is not the issue, it's whether it's clear who's speaking/writing in the context and whether it's something that actually needs individual attribution. I've covered this in detail above. PS: I agree with Warren's copyedits, but it's putting a suit on a scarecrow and expecting it to go to a business meeting.  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  05:10, 3 January 2018 (UTC)

  Thanks... I think?? 😁 Warren -talk- 05:28, 3 January 2018 (UTC)

• Here's a new version taking Warren's points into account:

The reader must be able to determine the source of any quotation, at the very least via a footnote. But the source should be additionally attributed in article text if a paraphrase of it would require attribution under WP:Neutral_point_of_view#Attributing_and_specifying_biased_statements.

SM, I can't understand you. If you agree the length doesn't matter, then surely you agree something needs to change in the guideline. Can you suggest new text of your own? EEng 05:35, 3 January 2018 (UTC)

SM's example was this: “The film was praised by critics as ‘edgy and creative’[1], ‘darkly humorous while seriously thought-provoking,[2] and ‘the best thing Jackson has done since My Life with the Weasels’,[3] though also reviewed less favorably for ‘its self-consciously shaky camera work’[4] and ‘overuse of pop music, making some scenes feel more like music videos’.[5]” Recast as paraphrase it would be something like, “The film was praised by critics for its creativity and edginess,[1] as well as thought-provoking dark humor,[2] surpassing anything Jackson has done since My Life with the Weasels,[3] though the film was also panned by reviewers who complained about jittery and self-conscious camera work,[4] along with excessive pop music.[5]” So now we go to WP:Neutral_point_of_view#Attributing_and_specifying_biased_statements to examine whether the paraphrase includes any biased statement. I don’t think it does, because it’s completely factual. So, no in-text attribution would be required in the original version before we paraphrased it. Anythingyouwant (talk) 05:37, 3 January 2018 (UTC)

No, there's no difference in "bias" between "The film was praised as 'edgy and creative'" and "The film was praised for its edginess and creativity". That a film was as "praised as 'edgy and creative'" is as precisely as much a "statement of fact" as its being "praised for its edginess and creativity"—both wordings require attribution, for the exact same reasons. Curly "JFC" Turkey 🍁 ¡gobble! 05:48, 3 January 2018 (UTC)

They both require attribution via footnotes, but neither requires additional attribution by name in article text. Anythingyouwant (talk) 05:52, 3 January 2018 (UTC)

(edit conflict) If there's no in-text attribution, there's no attribution, and the quote lacks sufficient context for the reader to evaluate it. Curly "JFC" Turkey 🍁 ¡gobble! 06:07, 3 January 2018 (UTC)

Per policy, attribution can be other than in-text: ”Attribute all quotations and any material whose verifiability is challenged or likely to be challenged to a reliable, published source using an inline citation.” Anythingyouwant (talk) 06:27, 3 January 2018 (UTC)

You've (a) ignored the point, and (b) disregarded WP:INTEXT. Curly "JFC" Turkey 🍁 ¡gobble! 07:03, 3 January 2018 (UTC)

I'm sorry if I did not correctly understand you. Regarding WP:INTEXT, I don't know whether you agree or disagree with the comment above that it "greatly overprescribes what should be in-text attributed". You say I've disregarded WP:INTEXT but I don't know what part of WP:INTEXT you're referring to. Anythingyouwant (talk) 07:14, 3 January 2018 (UTC)

It might help understanding your position if you gave us some clue as to why you think attribution should not be given in-text. As a reader, I put less trust in text that does not give proper in-text attribution. It feels almost manipulative, rather than strictly informative. Curly "JFC" Turkey 🍁 ¡gobble! 09:26, 3 January 2018 (UTC)

No one's saying there should never be in-text attribution, just that there are times it's required and times it's best omitted. There are many examples at Sacred Cod and Widener Library. EEng 09:32, 3 January 2018 (UTC)

I can't imagine why anyone would think the attributions in those articles are "best omitted". Why are "indignant" and "scores" even quoted in the first place? What is the point of an unattributed "painted to the life"? If it's only meant to mean it's realistic—why not just say so? Please explain? I honestly don't get this at all. Curly "JFC" Turkey 🍁 ¡gobble! 09:58, 3 January 2018 (UTC)

"Scores" and "indignant" are quoted to put some distance between WP's voice and the New York Times' slightly facetious tone; to pass on as straight fact the idea that police personnel were individually indignant, or that literally scores of leads were followed, would be to misrepresent the source. "Painted to the life" is, as you say, a way of saying "realistically painted", but a way that engages the reader instead of simply reciting a cold fact; in particular it prepares the reader for the style in which most commentary on the subject, quoted later in the article, is written. But whether or not this particular quotation should be used directly, or paraphrased as "realistically painted", isn't important here. What's important is that if it's quoted, it doesn't need in-text attribution because if it was paraphrased as "realistically painted" that wouldn't need in-text attribution either. EEng 10:28, 3 January 2018 (UTC)

Reading this response made me go crosseyed. A facetious tone is the last thing we need in a Wikipedia article, and your last sentence is a non sequitur. Those quotes draw attention to themselves by the act of quoting—that alone requires in-text attribution to make sense of why they are being quoted. This is an extraordiarily poor example of how quotation would be "appropriate", let alone non-attribution. These are encyclopaedia articles, not articles in the Lifestyle section of the local newspaper. Curly "JFC" Turkey 🍁 ¡gobble! 22:30, 3 January 2018 (UTC)

The article doesn't take a facetious tone; as I said it merely passes on the sources' words so readers can make of them what they will. As for the rest, a friend was kind enough once to pass on Teresa Nielsen Hayden's evaluation of the article; she called it Wikipedia as art, deft, beautiful, possibly even perfect (after following the link, hover over the words Sacred Cod at lower left). Perhaps you take refuge in the idea that articles are supposed to be grey and lifeless, but some of us aim higher.

Anyway, none of this has anything to do with the question at hand. Do you have any comment on the proposed text? As you say, a given statement either requires in-text attribution, or doesn't, whether it's a paraphrase or a quote – and that's exactly what my proposed text says. EEng 03:50, 4 January 2018 (UTC)

If the proposal can be interpreted as allowing attribution to be shoved into an endnote, I'm opposed to it, for the reasons I've already given. Quotation needs in-text contextualization. Without attribution it's not even clear if a quotation is a quotation or merely scare quotes, or something else. Curly "JFC" Turkey 🍁 ¡gobble! 08:13, 4 January 2018 (UTC)

Why do quotations need "in-text contextualization" any more than do any other content? EEng 11:08, 4 January 2018 (UTC)

All text needs sufficient contextualization for the reader to make sense of it. That's why we avoid unexplained technical jargon, include background sections, etc, etc, etc. Why would quotations be an exception? Curly "JFC" Turkey 🍁 ¡gobble! 22:57, 4 January 2018 (UTC)

All text needs sufficient contextualization for the reader to make sense of it. That's why we avoid unexplained technical jargon, include background sections, etc, etc, etc. Why would quotations be an exception? Right. Exactly. Quotations are not an exception. They have exactly the same needs for "contextualization" as does anything else not a quotation. That's what this proposal precisely says, on the specific question of the aspect of contextualization known as attribution: the source should be additionally attributed in article text if, were it recast as paraphrase, it would need to be attributed. EEng 00:26, 5 January 2018 (UTC)

I'm not following where you're going with this at all. I don't see an argument in your comment against requiring attribution. Curly "JFC" Turkey 🍁 ¡gobble! 02:12, 5 January 2018 (UTC)

Arbor tree-ish break

Curly, this MOS says in-text attribution is not needed if a quote is less than a sentence long ("The author of a quote of a full sentence or more should be named; this is done in the main text and not in a footnote. However, attribution is unnecessary with quotations that are clearly from the person discussed in the article or section"). I believe shorter quotes often should have in-text attribution, and I understand you to feel the same way. But not every one, right? There are plenty of examples above, like "She is widely described as 'the Leader of the Free World'." [with a footnote referring to several articles] That doesn't need in-text attribution, right? Anythingyouwant (talk) 23:05, 4 January 2018 (UTC)

"widely described" gives the reader the context needed to make sense of the quotation; in a sense, that's a form of "attribution" (it's attributed "widely"). That's a whole world of difference from "The work was described as 'enlightening' and 'ground-breaking'." Both those quotes are less than a sentence and would absolutely require attribution. Curly "JFC" Turkey 🍁 ¡gobble! 23:16, 4 January 2018 (UTC)

Can that be easily cleared up by amending the proposal to say, "The reader must be able to determine the source of any quotation, at the very least via a footnote. But the source should be additionally attributed in article text, by name if a paraphrase of it would require attribution under WP:Neutral_point_of_view#Attributing_and_specifying_biased_statements."? I have inserted "by name". Incidentally, the current sentence in this MOS is already addressing when the source "should be named". Anythingyouwant (talk) 23:28, 4 January 2018 (UTC)

I'm not sure what advantage there is to the rewording besides dropping the "full sentence" thing, which never should have been there in the first place. Curly "JFC" Turkey 🍁 ¡gobble! 00:13, 5 January 2018 (UTC)

User:Curly Turkey, you said, "'widely described' gives the reader the context needed to make sense of the quotation; in a sense, that's a form of 'attribution'". But it's not a form of attribution by name. This MOS section is about when attribution is required by name. You don't think attribution by name is required for every quotation, do you? Anythingyouwant (talk) 00:18, 5 January 2018 (UTC)

The text should make clear where the quotation's coming from. In most cases, that would mean naming the source, and there shouldn't be any room in the wording to allow wikilawyers to arbitrarily worm out of that. In certain contexts, attribution should or must be handled without naming names, but should not be dropped into running text without contextualization along the lines of a "widely considered" or whatever. Curly "JFC" Turkey 🍁 ¡gobble! 00:32, 5 January 2018 (UTC)

The text should make clear where the quotation's coming from. No, see I'm not going to let that pass. Why should the text male clear where the quotation's coming from, always, any more than the text needs to make clear, always, where anything else comes from? I agree that in most circumstances quotes should be in-text attributed (usually definitely – "Smith said", or "his sister said" – but sometimes less so – "a witness later said"), but not always. To drag in another of my pet articles, when the lead of Phineas Gage says that the subject was once termed "the case which more than all others is cal­cu­lated to excite our wonder, impair the value of prognosis, and even to subvert our phys­i­o­log­i­cal doctrines", it doesn't help the reader at all to learn have it rubbed in his face, directly in the text that was written by H.F. Campbell in The Ohio Medical & Surgical Journal for 1851, though if he really wants to know he can check the citesfootnotes. Had we paraphrased as Gage's case attracted unprecedented interest, made physicians question their ability to predict whether a given injury would or would not be fatal, and brought many established ideas about human physiology into question, no one would dream of requiring in-text attribution, so why for the quotation, which gives precisely the same information (in a much longer and less vivid way)? EEng 01:47, 5 January 2018 (UTC) Later insertions and deletions for clarity. 03:30, 5 January 2018 (UTC)

Of course the Phineas Gage quote requires attribution. How could it not? Why is this your example? If that quote doesn't require attribution, then how could any? Curly "JFC" Turkey 🍁 ¡gobble! 02:15, 5 January 2018 (UTC)

Jeez, Louise! Of course it requires attribution – in a footnote at least. But does it require IN-TEXT attribution? It's the IN-TEXT question we're talking about. How is the reader helped or informed or contextualized or englightened or whatever it is we're doing, by telling him IN THE TEXT that H.F. Campbell was the specific writer? EEng 03:24, 5 January 2018 (UTC)

Do we really have to go around in circles with this? I've already given reasons. You don't appear to be reading them. Aside from all the regular reasons, a quotation draws attention to itself by the act of quotation. In-text attribution justifies and clears up for the reader why it's being quoted. Curly "JFC" Turkey 🍁 ¡gobble! 11:46, 6 January 2018 (UTC)

Do you think EEng's proposal leaves any room in the wording to allow wikilawyers to arbitrarily worm out of that? If so, can you please suggest how to improve EEng's proposal? If we take out the one-sentence minimum from the current MOS, then attribution by name would be required for every quotation, and I don't think you or me or EEng would agree with that. Thanks. Anythingyouwant (talk) 00:38, 5 January 2018 (UTC)

I suppose the consequences of merely dropping "by name" would be a lot of "a certain professor"-type "attributions", which obviously isn't acceptable. An eloquent and concise wording escapes me. Curly "JFC" Turkey 🍁 ¡gobble! 02:18, 5 January 2018 (UTC)

I like where Anythingyouwant is going with this. And agree with CT that the "one sentence" thing is nonsense.  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  01:15, 5 January 2018 (UTC)

I think we all agree the one-sentence thing is nonsense, so that's something. I hope you'll stay with this, because we need a brain that can keep track of all that's going on in a discussion like this. EEng 01:48, 5 January 2018 (UTC)

Resuming an earlier subthread

Remember I said I was doing this against my better judgment. At least as between Anythingyouwant and C.T., the disagreement seems to be on what the outcome of applying WP:Neutral_point_of_view#Attributing_and_specifying_biased_statements would be, not whether WP:Neutral_point_of_view#Attributing_and_specifying_biased_statements is the right standard. I'd still like to hear any alternative text from S.M., if he can. EEng 05:57, 3 January 2018 (UTC)

I don't have strong feelings about this one way or the other, but I flipped through several featured articles that have reviews or critical analysis (The Concert in Central Park, The Autobiography of Malcolm X, Z. Marcas, Ultima Underworld: The Stygian Abyss) and they all consistently identify the source of a quote, in the article text itself: The Swedish Datormagazin considered the game to be "in a class by itself".[29] In Germany, Power Play praised its "technical perfection" and "excellent" story,[31] while Play Time lauded its graphical and aural presentation, and awarded it Game of the Month.[15] .... This is pretty high-quality construction and seems to neatly take care of bias issues, but I also appreciate that getting there is quite a lot of work. Warren -talk- 06:17, 3 January 2018 (UTC)

We're getting bogged down debating individual examples instead of focusing on what the right standard is. Is, or is not, WP:Neutral_point_of_view#Attributing_and_specifying_biased_statements the right standard for in-text attribution of quotations (whether there's disagreement on how to apply it in the particular examples above)? S.M., help!!!! EEng 06:35, 3 January 2018 (UTC)

Current draft proposal

The current draft proposal is kind of buried above, so here it is (slightly prettified):

The reader must be able to determine the source of any quotation, at the very least via a footnote. But the source should be additionally attributed in article text, by name if a paraphrase of it would need attribution as a biased statement.

This drops the silly limitation to quotes that are a sentence or longer; merely dropping that limitation would mean every quotation would have to be attributed by name in the article text, which no one thinks would work, so this proposal loosens that requirement by only asking for attribution by name if it’s a biased statement. Anythingyouwant (talk) 04:32, 6 January 2018 (UTC)

• Support. This does not preclude in-text attribution by-name for any quotation, nor does it preclude in-text attribution not-by-name for quotations that do not need in-text attribution by-name. So it’s quite flexible. Anythingyouwant (talk) 00:32, 7 January 2018 (UTC)

If no objection within 24 hours, I think someone can go ahead and make this change. The current language in the MOS is “The author of a quote of a full sentence or more should be named; this is done in the main text and not merely in a footnote. However, attribution is unnecessary with quotations that are clearly from the person discussed in the article or section.” Anythingyouwant (talk) 21:33, 9 January 2018 (UTC)

• Anythingyouwant, to reduce the change from the current guideline, would you be OK with –

The reader must be able to determine the source of any quotation, at the very least via a footnote. But the source should additionally be named in article text if a paraphrase of it would need attribution as a biased statement.

–? EEng 22:45, 9 January 2018 (UTC)

Sure. Anythingyouwant (talk) 23:37, 9 January 2018 (UTC)

• Support. EEng 00:57, 10 January 2018 (UTC)

See the disambig header at Mind Meld. I'm pretty sure that, technically, "mind meld" when used in running prose like that would be spelled with all lower-case letters, but are we supposed to refer to it in this context the same way it would appear as the title of a Wikipedia article? It's particularly weird in this case since the way it is linked it looks like a standalone article; should it rather read For the fictional practice, see Vulcan (Star Trek)#Mind melds.? Hijiri 88 (聖やや) 04:58, 10 January 2018 (UTC)

I would think lowercase as well, though there's no reason not to do For the fictional practice, see mind meld; I don't see the point of giving Vulcan (Star Trek) § Mind melds. Just makes the hatnote take longer to read and understand.  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  17:54, 10 January 2018 (UTC)

I would cap the initial letter of an article title in a hatnote, or in contexts like "see Mindmeld" and such, so it looks like an article title (sentence case). Dicklyon (talk) 02:44, 11 January 2018 (UTC)

Right, but mind meld isn't an article title, and redirects to a section with a different title. I don't feel strongly about it either way; it's kind of a link-style consistency versus link-target surprise thing. Hijiri's not arguing for something like "for the genre, see science fiction", to an actual article title. (At least I don't think that's the idea.)  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  09:32, 11 January 2018 (UTC)

It's already linked though. And not really a title of an article as SmCCandlish points out. IMHO it should be "mind meld"- it's a sentence, so this sort of capitalization looks strange. Galobtter (pingó mió) 17:03, 11 January 2018 (UTC)

One of Our Upper Case Letters is Missing? Martinevans123 (talk) 19:37, 11 January 2018 (UTC)

• When hat noting to a specific section of an article, I would indicate the article title ... just so the reader knows where he/she is being pointed to. This does not need to be done in the link, however... the text of the hatnote could read: “For the fictional usage (by Vulcans in Star Treck), see mind meld.” Or something similar. Blueboar (talk) 12:24, 11 January 2018 (UTC)

That's great for Star Treck, but what about Star Trek? EEng 19:19, 11 January 2018 (UTC)

blink... oh, yes... that too. 🙂 Blueboar (talk) 23:04, 11 January 2018 (UTC)

Is there any guideline concerning using complete sentences in the article prose? In particular, I am referring to this version of the lead sentence vs. this version. In the first version, the first paragraph ends with a sentence fragment. In the second version, the first paragraph ends with a complete sentence. Is there any guideline concerning which version is preferred? If there isn't, I have been told that I should start an RFC to discuss it [5]. Thank you. Frietjes (talk) 01:00, 12 January 2018 (UTC)

Somebody really thought that was worth an RfC? I think it goes without saying that the text of an article should be regular English, which means sentences, so I don't think you'll find any guidelines saying that.

The example here, "Population: 8,086 (2013 est.)", isn't even a sentence fragment. I don't know what you call it; it's not English. Bryan Henderson (giraffedata) (talk) 03:23, 12 January 2018 (UTC)

The idea is covered in WP:PROSE. I can see that some might prefer what is essentially a bullet point ("Population: whatever") because it shows the essentials with no fluff. However, it ain't English and prose is preferred. Johnuniq (talk) 06:43, 12 January 2018 (UTC)

thank you for the feedback, it appears it may not be an issue going forward[6], but it's good to have the feedback. My personal opinion is that the population figure should be moved out of the lead section and into the infobox. Frietjes (talk) 13:04, 12 January 2018 (UTC)

Yep, something like "Population: whatever" is an infobox or list or table entry, not regular article text. This is actually covered, in a sideways manner, at MOS:TRIVIA, which instructs us to avoid building up lists of factoids and to instead write in full sentences and paragraphs.  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  19:00, 12 January 2018 (UTC)

Examples like that belong in the infobox. I wouldn't go as far as others to say it's "not English", but it doesn't fit in the running prose of a Wikipedia article. Hijiri 88 (聖やや) 04:14, 13 January 2018 (UTC)

It could be used in a parenthetical, though, "Doodah, Iowa (population 12,345)"; no colon.  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  14:04, 15 January 2018 (UTC)

Keeping in mind that could draw the ire of those bent on removing prentheticals from the leads. Curly "JFC" Turkey 🍁 ¡gobble! 01:22, 16 January 2018 (UTC)

Wasn't thinking of leads; this sort of stuff is usually infoboxes and in the body, isn't it?  — SMcCandlish ☏ ¢ 😼  22:09, 19 January 2018 (UTC)

FYI

 – Pointer to relevant discussion elsewhere.

Please see Wikipedia talk:Naming conventions (people)#Conflict between WP:NCP and WP:MOS
 — SMcCandlish ☏ ¢ 😼  23:54, 19 January 2018 (UTC)

At time of writing, this page has 194236 bytes. I suspect that most of this is prose. Some of the subpages, such as Wikipedia:Manual of Style/Dates and numbers (121031 bytes), also seem too long. LaundryPizza03 (talk) 02:18, 20 January 2018 (UTC)

What? If it is actually a problem, the only thing we can do is have different pages for individual sections, for example MOS:QUOTE may occupy its own space. –Sb2001 02:23, 20 January 2018 (UTC)

I agree that this page is very long, but do you have any concrete proposals on what to do about it? — Warren. ‘ talk , 03:56, 20 January 2018 (UTC)

• This is, generally speaking, the summary page for all the other MOS pages. (I think there's some stuff that's only here.) The last thing we need is to break it up -- why is the size, per se, a problem? If anything, if it needs to be slimmed down we'd do that by moving less-important detail to one of the existing subpages. Making more fragmented pages is definitely not what we need. EEng 05:11, 20 January 2018 (UTC)
  • Well, one option would be to make it smaller by getting rid of a lot of rules. I would generally be in favor of that, though I wouldn't express it in terms of the length of the page. --Trovatore (talk) 06:08, 20 January 2018 (UTC)

No one's against slimming MOS down ... as a generality. EEng 06:20, 20 January 2018 (UTC)

There are a lot of areas where we could tighten up the language and presentation, but I don't think actually narrowing what it is trying to say is going to work in the long run. Almost everything that's there is a consequence of dealing with real situations that have come up. Remove something now, and it'll get added back in someday. — Warren. ‘ talk , 23:50, 20 January 2018 (UTC)

Actually, LaundryPizza03, what on earth do you mean by, 'I suspect that most of this is prose'? It seems you can't be bothered to read the page about which you are complaining, you can't be bothered to read about the tag you are placing—see Primergrey's comment, and you can't be bothered to provide a potential resolution; you just want us to do all the work for you. –Sb2001 20:09, 20 January 2018 (UTC)

FYI, the "too long" tag, that was applied and subsequently removed, links to a page that begins, "This page contains an overview of the key issues concerning article size" (bolding Wikipedia's). Nothing to say one way or another on the MOS's length, but this editing guideline does not apply in any case. Primergrey (talk) 07:45, 20 January 2018 (UTC)

• Given the page is not meant to be read in one sitting, but rather its parts consulted when the need arises, I have to wonder what "too long" is supposed to mean. Curly "JFC" Turkey 🍁 ¡gobble! 09:45, 20 January 2018 (UTC)
• Unless it's crashing browsers, it is not too long. This is a reference work, not an essay. We probably could move a few points to topical subpages, but as the main MoS page, the point of it is a WP:SUMMARY-style overview of the key points of the entire MoS tree.  — SMcCandlish ☏ ¢ 😼  08:55, 23 January 2018 (UTC)

FYI

 – Pointer to relevant discussion elsewhere.

This is only peripherally MoS-related, but I figure watchlisters of articles on the English language are a probably also overlapping a lot with MoS watchers.

I've made a request at at WT:Bot requests#Tag talk pages of articles about English with Template:WikiProject English language to have the articles within the project scope bot-tagged, since doing it by had or even with AWB might be an enormous amount of effort. I'm not sure if BOTREQ requires a showing of support before action is taken to implement a bot, but I get the sense that this might be the case.  — SMcCandlish ☏ ¢ 😼  09:58, 23 January 2018 (UTC)

Moved to #RfC Should the usage of the terms "Arab" and "Arabic" be guided by the Manual of Style?

Can we move WP:Manual of Style#Use of "Arab" and "Arabic" out of the MoS and into WP:List of commonly misused English words? Don't get me wrong, the advice in that MoS section is good, but I don't see it as an especially evil case.

The "Allows to" discussion above received some WP:CREEP complaints, "swat a fly with a nuclear missile" etc, this seems like the same kind of "fly". It would be nice to know that the MoS is not an ever-expanding tome. Very few pages link to the shortcuts to that MoS section. Batternut (talk) 13:02, 21 January 2018 (UTC)

I agree. Maybe this belongs in Wikipedia:List of commonly misused English words, but not here. EEng 21:06, 21 January 2018 (UTC)

Support move to that list. —David Eppstein (talk) 21:58, 21 January 2018 (UTC)

Concur in support of move. Very out of place. Galobtter (pingó mió) 13:14, 22 January 2018 (UTC)

• Done. EEng 21:52, 22 January 2018 (UTC)

Great. I changed the shortcuts to point to Wikipedia:List of commonly misused English words#Arab, though they could well be deleted. Batternut (talk) 22:20, 22 January 2018 (UTC)

The instruction is problematic in any case; read literally, it could result in people changing gum arabic to gum arab. I'm not quite sure how to fix it, but it would be nice to address it before forgetting about it. --Trovatore (talk) 22:08, 22 January 2018 (UTC)

Gum Arab? EEng 22:20, 22 January 2018 (UTC)

The fix is stating that it applies to proper-noun usage. Anyway, I agree that moving it is fine, though I would move it to WP:Manual of Style/Words to watch or to WP:Manual of Style/Islam-related articles (and tweak its lead to include Arabs as within its scope), so that it remains part of MoS. Moving it to WP:List of commonly misused English words is demotion of guideline material to an essay, which would require an RfC/proposal. And that page shouldn't exist anyway. It's a once (and possibly still) badly-written article that got project-spaced, and which has since been replaced in mainspace by List of English words with disputed usage. This content probably needs a merge/split operation, with the salvageable, sourced descriptive material moved into the real article, and the advice material squeezed down into a more compact list, and moved to WP:Manual of Style/Commonly misused English words or simply WP:Manual of Style/Vocabulary. That might be a viable home for the Arab/Arabic/Arabian material. However, going that route is a lot of work.  — SMcCandlish ☏ ¢ 😼  09:04, 23 January 2018 (UTC)

I'm not sure "proper noun" is sufficient — my intuition at least is to prefer medieval Arabic architecture. I'm not convinced it's limited to language.

I don't buy that removing a minuscule tidbit from the MoS requires a full RFC. Was it put there by RFC? It shouldn't be done boldly (neither should rules be added boldly) but I think some discussion on the talk page is reasonable for such a minor point. --Trovatore (talk) 09:19, 23 January 2018 (UTC)

Who says it's a minor point? Anyone who cares about cultural sensitivity won't consider it minuscule, and we're talking about importance not verbiage length. Lots of MoS consists of short points and many of them are in fact minor ones (that have nevertheless resulted in drama when we lack a clear rule about those particular points). That doesn't make them any less part of the guideline; they are not essay material to demote on a whim. The Arab/Arabic/Arabian material was the genesis of the MOS:IDENTITY section. I've notified the other mentioned MoS pages' talk pages to get some additional discussion. "I don't think this point is super-important, ergo just move it to some essay because a grand total of three people agreed with me over the course of barely a day" is not how we deal with policy and guideline material. PS: Fine, use "proper name" if you like; I thought it was obvious that "proper noun" in the context also includes proper adjectives derived from proper nouns, but not genericized adjectives as in "gum arabic" (like "french fries" and "fine china" and "his parents' draconian rules") that have lost all sense of connection to the ultimate namesake.  — SMcCandlish ☏ ¢ 😼  09:48, 23 January 2018 (UTC)

Stanton, I've seen you argue for changing guidelines and even policy on a BRD basis! Where is this need for an RFC coming from all of a sudden?

As for proper name/noun, I don't see how that distinction addresses the point I was making, which is that Arabic is sometimes the better word in non-linguistic situations, like medieval Arabic architecture. --Trovatore (talk) 19:26, 23 January 2018 (UTC)

There's nothing sudden about it. I and others have been using RfC to edit MoS more and more over the last ~2 years, after too-frequent complaints of it substantively changing too frequently and with too small a show of consensus. The antidote for that is, necessarily, a slower process with a larger show of consensus.  — SMcCandlish ☏ ¢ 😼  23:20, 23 January 2018 (UTC)

A certain non-definitive n-gram viewer would suggest "Arab architecture" is the popular choice among authors. Batternut (talk) 20:34, 23 January 2018 (UTC)

Did you add "medieval"? I had that word in there on purpose. --Trovatore (talk) 20:37, 23 January 2018 (UTC)

With medieval the n-gram thing finds nothing at all. Simple book searching gets 32 for "medieval Arab architecture", 11 for "medieval Arabic architecture". Batternut (talk) 22:53, 23 January 2018 (UTC)

FYI, the original addition to the MoS was in February 2005 - the pre-existence of the "identity" section would appear to contradict SMcCandlish's suggestion that this material was the "genesis of the MOS:IDENTITY section". A search of the MoS talk archives found the following relevant discussions, no RfC's, and none of which proposed the original entry:

• Wikipedia talk:Manual of Style/Archive 16#"Arab" etc., July 2005
• Wikipedia talk:Manual of Style/Archive 92#Identities, August 2007
• Wikipedia talk:Manual of Style/Archive 145#Alternative proposed wording for MOS:IDENTITY, August 2013
• Wikipedia talk:Manual of Style/Archive 160#Arab exception?, September 2014
• Wikipedia talk:Manual of Style/Archive 180#Shortcut proposals, February 2016

Batternut (talk) 11:08, 23 January 2018 (UTC)

I must've mis-remembered the order of the material's addition, then. But it doesn't matter; the issue is deleting substantive advice from MoS just because three people said so after a less than a day of discussion. We don't do that.  — SMcCandlish ☏ ¢ 😼  23:20, 23 January 2018 (UTC)

I'm pretty sure advice gets added with similar or lesser process. It should not be easier to add advice than remove it. If anything, it should be the opposite. --Trovatore (talk) 23:40, 23 January 2018 (UTC)

It shouldn't be easier, at this date, to add more rules without RfCs, and we mostly don't. After 16+ years MoS is pretty close to feature-complete for WP's purposes.  — SMcCandlish ☏ ¢ 😼  06:38, 24 January 2018 (UTC)

Having just raised an RfC on this subject, I suggest closing this discussion now, with a reason such as "superseded by RfC (below) - please discuss the case for MoS guidance upon Arab/Arabic usage there"? Batternut (talk) 10:02, 24 January 2018 (UTC)

Long-standing text

Exceptions: Ref tags are placed before dashes, not after. Where a footnote applies only to material within parentheses, the ref tags belong just before the closing parenthesis.

Someone removed the second sentence; I restored it, and it was removed again.

My rationale

This is long-standing practice. There's a huge difference between:

1. Claim (caveat<ref 1 />).
2. Claim (caveat).<ref 1 />
3. Claim (caveat<ref 1 />).<ref 2 />

In the first, the claim isn't even sourced, only the caveat is. In the second, the claim and caveat share the same source.

I think this should be restored. It maybe looking better, to some people, to change #1 into #2 is no excuse for sacrificing certainty about what is being attributed to which source (if any). If the claim in this case is not in fact attributable to the same source as the caveat, then doing this is outright falsification of the sourcing. We can't have MoS advising to do this on purpose.  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  10:52, 30 December 2017 (UTC)

• I'm the someone.
• This was added by a single editor [7] with no apparent discussion [8].
• Chicago (14.26) opposes this except "on rare occassions":

Though a note number normally follows a closing parenthesis, it may on rare occasion be more appropriate to place the number inside the closing parenthesis—if, for example, the note applies to a specific term within the parentheses.

(In an earlier book he had said quite the opposite.)[2]

Men and their unions, as they entered industrial work, negotiated two things: young women would be laid off once they married (the commonly acknowledged “marriage bar”[1]), and men would be paid a “family wage.”

-- but we don't do the kind of writing seen in that second example because in our work, everything needs to be sourced e.g.

Men and their unions, as they entered industrial work, negotiated two things: young women would be laid off once they married[2] (the commonly acknowledged “marriage bar”),[1] and men would be paid a “family wage.”[2]

-- or something like that. It's simple: each cite covers everything back to the prior cite.

• Similarly, I don't understand SM's example (1.) because he's leaving Claim unsourced, apparently. I'd write:

Claim[2] (caveat).[1]

His example (2.) is obvious -- everything's covered by a single ref:

Claim (caveat).[1]

His example (3.) I'm not sure I understand. If [1] applies only to Caveat, and [2] applies to both Claim and Caveat, then write

Claim[2] (caveat).[1][2]

English isn't a programming language with push-down stacks of subsidiary clauses. Like I said, it's simple: each cite covers everything back to the prior cite, parens or punctuation or whatever notwithstanding. Cites come after parens for the same reason they come after periods and commas: because otherwise they look stupid^[1], and awkward^[2], and (awful^[3]).

EEng 15:04, 30 December 2017 (UTC)

"This was added by a single editor" is not a rationale for anything. Everything is added by a single editor or an edit conflict will result. The question is whether it describes consensus practice, and the answer is yes. If you'd like to change that practice, please open an RfC. Given that it could affect untold numbers of articles, it should probably be advertised at WP:VPPOL. Yes, you are not understanding the examples. I'm not leaving the claim in the first examle unsourced, it simply is unsourced, as are millions of claims in our articles (and this is fine, per WP:V policy – claims must be sourceable not sourced unless/until they're controverted, though with special exceptions like WP:BLP and WP:MEDRS material, in which anything potentially counterfactual but unsourced should be deleted not tagged). The editor who added the caveat and source for it in example one is unlikely to be the same editor who added the original claim or have access to source material needed for the original claim (or no time or inclination to deal with it, being a volunteer). Adding the sourced caveat may be very important, e.g. if the claim is generally true but is not under particular circumstances, or whatever.  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  12:53, 31 December 2017 (UTC)

To me EEng's "stupid^[1], and awkward^[2], and (awful^[3])." looks perfectly normal. The other option (stupid,^[1] and awkward,^[2] and (awful)^[3].) would seem to mean that we are citing the punctuation as well as the text. --Khajidha (talk) 15:03, 31 December 2017 (UTC)

I wouldn't advocate any of those examples. EEng 18:54, 1 January 2018 (UTC)

The rest of the section regarding the location of refs and other punctuation (full stops, commas, and the like), is not in question. We're specifically speaking on parentheses. --Izno (talk) 15:24, 31 December 2017 (UTC)

And my point applies either way. --Khajidha (talk) 16:39, 31 December 2017 (UTC)

You could make all these same arguments for periods or commas: Where a footnote applies only to material within a single sentence, the ref tags belong just before the closing period. Where a footnote applies only to material within text set off by commas, the ref tags belong just before the "closing" comma. There's nothing special about parens in this regard, or with regard to any of the arguments that have been made.

Really, none of SM's examples are even on point to my quibble with the current guideline. I just want to be able to write

Claim.[1] (Additional.)[2] More.[3]

instead of being forced to write

Claim.[1] (Additional.[2]) More.[3]

How about if we follow Chicago's "rare occasions" recommendation and say something like this:

Where it is desired to emphasize that a footnote applies only to material within parentheses, the ref tags may be placed before the closing parenthesis.

EEng 18:54, 1 January 2018 (UTC)

Works for me.  — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  05:07, 3 January 2018 (UTC)

• Any dissent? EEng 05:11, 4 January 2018 (UTC)
  • I respectfully dissent on three grounds:
    1. Citation clarity: Putting the anchor within the parenthetical signals its scope with minimal hassle, and a modest gain in precision ought to outweigh minor stylistic quibbles. This crops up regularly in biographies:

       Alice Betty Carolie Smith (born January 1, 1990^[1]) is a Martian saxophone player and political candidate.

       References

       1. ^{^} Ref for birth date that doesn't give the middle names.

    2. Ease of reading: In my lay opinion, having the footnote before the closing parenthesis makes it easier for the eye to skip over the entire parenthetical.
    3. Stare decisis: That the guideline—in our main style guide—went unchallenged for more than five years is not insignificant, and, in my opinion, the lukewarm and indecisive discussion here is insufficient reason to overturn this longstanding practice.
    Rebbing 03:37, 5 January 2018 (UTC)

    Shit, just when there was agreement on something on this page.

    • 1. The proposed text allows you to do that if you want.
    • 2. In my lay opinion the superscipt [1] inside the paren catches the eye, and you trip on it instead of skipping over it.
    • 3. MOS is supposed to reflect good practice actually in use, not tell everyone to change an OK thing they're already doing. A quick sample of FAs shows about a 2:1 ratio (but let's just call it even); the proposal simply recognizes that.

    EEng 04:01, 5 January 2018 (UTC)

    SMcCandlish, what now? EEng 22:39, 7 January 2018 (UTC)

    Bueller? Bueller? EEng 20:31, 14 January 2018 (UTC)

    Then the long-standing original wording should be restored, since its removal is challenged by multiple editors and no one is agreeing with the removal, and it's widely followed (so appears to represent consensus practice). The numbered stuff:

  1. I almost agree with your first point, that the operational effect of both versions is the same, but it's not really, since your version couches it in terms of "desire for emphasis", when this is really about sourcing accuracy. So, the original wordig was fine. The effort to compress MoS wording is not much of an end in and of itself, especially if it leads to dispute or confusion in the process of doing it.
  2. The second point is just subjective opinion, and doesn't appear broadly shared, or we wouldn't have been doing it this way for years (and the "rule" wasn't inserted out of the blue, but to describe actual practice, which has been common the entire time I've been here). This alleged readability issue is the exact same sort of WP:ILIKEIT concern as "I want to put put all punctuation inside quotation marks [i.e. typographer's quotation] because it just looks more tidy, tucked in like that" and "Sammy Davis Jr. doesn't look right without the comma to me", and "I've always written it as Jesus' not Jesus's", and so on. Here, it's neither more nor less a reading impediment, on average, by being inside or outside the bracket. However, we lose precision – sometimes very misleadingly – when it's moved outside and the preceding material doesn't have its own inline citation. And it usually won't in leads – we discourage use of inline citations in leads, preferring them in the body text.
  3. I cannot agree with your third point at all; it's flatly impossible for you to prove that citations in the form "Alice Betty Carolie Smith (born January 1, 1990)^[1]" were intended to convey the same thing as "Alice Betty Carolie Smith (born January 1, 1990^[1])". If you encounter the former, I'll bet good money that at least 95% of the time, especially in FAs, that the cited source provides the full name not just the date, and that when the second is used it provides the date but not the full name (or a citation is already provided for the full name).

  No one else seems to have difficulty with this format, and we have a twin policy reason for using it (WP:V accuracy about what facts are sourced and WP:NOR avoidance of misrepresentation of sourcing, two sides of the same coin). In a case where there actually is a citation "^[1]", your preferred "Claim.^[1] (Additional.)^[2] More.^[3]" would actually be permissible, since MoS is a guideline not a law, and no confusion could result. "Claim. (Additional.)^[1] More.^[2]", where the initial claim has no source yet, is simply a wrong format (if "^[1]" is a source only for the parenthesized material) and would probably prevent anyone from sourcing the unsourced claim, perhaps for years, because it falsely already looks sourced.

  I'll ask you the same thing you ask everyone who wants a substantive rule change: Can you provide evidence there's an ongoing problem to solve, that editors are actually fighting about "(something^[x])" formatting? What I see is editors following it without incident.
   — SMcCandlish ☏ ¢ >^ʌⱷ҅_ᴥⱷ^ʌ<  21:30, 14 January 2018 (UTC)

The "what problem needs solving" challenge applies where it's proposed to add rules; here the proposal is to relax one. As I think I mentioned somewhere above a quick sample of FAs reveals this rule is widely ignored. EEng 23:00, 14 January 2018 (UTC)

Says who? Any substantive change should address an actual problem and that's how we approach it (not just here but in WP:POLICY pages generally). It's the status quo ante principle, a.k.a. "if it ain't broke, don't 'fix' it". PS: A quick sample of FA is generally always misleading because most of our FAs are not recent and FAs as a class are notoriously resistant to MoS (or other guideline) compliance edits – various would-be page "owners" are revert-happy and attempt to keep them in a more-or-less fixed state. It's the dark side of "ain't broke, don't 'fix' it taken too far.

See also point 3 above. I do not believe it possible to demonstrate that FAs are doing what you suppose that they might be. In every case I've looked at, the source either encompasses the material before the parenthetical as well as the parenthetical, or the material before it has its own source cited just before the parenthetical, ergo there is no possibility of confusion. That is, "Foo^[1] (bar^[2])" and "Foo^[1] (bar)^[2]" are exactly equivalent, while "Foo (bar^[2])" and "Foo (bar)^[2]" are not; meanwhile "Foo (bar^[2])" would not be found in an FA, because it indicates that the "Foo" part is unsourced. You will thus find "Foo (bar)^[2]" in FAs, but only when the source covered both "Foo" and "bar", which case that markup is exactly what we want. What we do not want, ever, is "Foo (bar)^[2]" when the source only provides "bar". We consequently have no reason whatsoever to use "Foo^[1] (bar)^[2]" when "Foo^[1] (bar^[2])" is clearer, because the former inspires a misuse, that of "Foo (bar)^[2]" when the source only provides "bar".
 — SMcCandlish ☏ ¢ 😼  22:25, 19 January 2018 (UTC)

Says me. When you're adding a rule, you should ask what problem is being solved; when you're removing one, you should ask if a problem will be created. (The problem that would be solved is MOSBLOAT, though of course only incrementally.) My quick survey of a bunch of FAs found, among the maybe 15 I found that had one or the other, about an even split between Foo.[1] (Bar.)[2] and Foo.[1] (Bar.[2]) As mentioned before all your arguments could be used for other punctuation like periods and commas. In any event I give up -- just not important enough, and it appears this is widely disregarded anyway. EEng 23:45, 19 January 2018 (UTC)

Except we don't have any rule about other punctuation to move ref tags inside that punctuation, so the argument couldn't actually be applied. We'll just have to agree to disagree; citation accuracy is why it's important. The fact that about 50% of cases are inside disproves that its rare or disused (a 50% compliance rate with any guideline line-item is quite good; far fewer than 50% of statements in our articles are sources, for example), while the half that are not are (we've already been over this) almost invariably going to be preceded immediately by a citation for the material in front of the parenthetical, or really do source both statements. I.e., It's not broken. The Foo.[1] (Bar.)[2] cases probably simply pre-date the rule in most cases, and they are harmless anyway, except to the extent they may inspire something that's not. What's not harmless is Foo. (Bar.)[2] when [2] doesn't also source Foo. Been over this so many times I can't think of any way to make it clearer.  — SMcCandlish ☏ ¢ 😼  19:41, 25 January 2018 (UTC)

Moved to Talk:Electric pump-fed engine#Requested move 24 January 2018

My edit corrected a venial mishyphenation (https://en.wikipedia.org/w/index.php?title=Electric_pump-fed_engine&diff=prev&oldid=821874204); The Rambling Man's edit un-fixed it to restore the venial error (https://en.wikipedia.org/w/index.php?title=Electric_pump-fed_engine&diff=next&oldid=821921671). The principle that hyphens do not bridge across open compounds is covered at Wikipedia:Manual of Style § Instead of a hyphen, when applying a prefix to a compound that includes a space. In this instance, there is no rewording/recasting of the term to be done, because the established name (of the type of engine) is the established name. But what does need to be done is professional punctuation rather than the amateur type. Who can move the page? In professionally edited work, "an engine that is fed by an electric pump" is an "electric pump–fed engine" or an "electric-pump-fed engine" but not an "electric pump-fed engine". — ¾-10 23:06, 23 January 2018 (UTC)"

Weirdly, I'm seeing "moved page Electric pump–fed engine to Electric pump-fed engine", which are the same. Whatever. It appears it should be electric-pump-fed (though another term is electric-feed, which would be more WP:CONCISE). It's an engine fed by an electric pump, not an electric engine fed by a pump (it's not an electric engine – properly a motor – at all, it's a rocket engine fed with rocket fuel from an electric pump). We don't care that some of the specialist literature uses "electric pump fed engine"; governmentese drops pretty much all such hyphens, and WP isn't written in governmentese. If we used Electric pump fed engine we'd be removing hyphens from every other compound adjective, but we're not.  — SMcCandlish ☏ ¢ 😼  23:27, 23 January 2018 (UTC)

Regarding "which are the same", no—the first contains an en dash; the second, a hyphen. The first is correct in professionally edited work; the second is what someone would write who doesn't care about the difference (for example, most scientists and engineers wouldn't bother). In mainstream typefaces the difference in glyph between the characters is clearly visible; if the glyphs look identical on your screen, it may be due to local font choice. You are right that "electric pump fed engine" is not optimally styled. It is not orthographically wrong but is inoptimally edited. So, in short, my original post here is still the operative point. Hyphens don't bridge across open compounds in well-edited work. — ¾-10 23:38, 23 January 2018 (UTC)

I found out per WP:MOVEREQ to take it to the article's talk page with a specific template, which I did (Talk:Electric pump-fed engine#Requested move 24 January 2018). — ¾-10 01:38, 24 January 2018 (UTC)

Ah! Yes, I'm trying out a new editing font, and it doesn't clearly enough distinguish between – and -. An en dash is wrong in that construction (anywhere); the term is not juxtaposing two essentialy things, as it is in US–Canada relations, that's a just a compound adjective, which takes a hyphen not a dash. MOS:DASH and MOS:HYPHEN cover this in detail.  — SMcCandlish ☏ ¢ 😼  07:22, 24 January 2018 (UTC)

Just clarifying that the en dash is DEFINITELY not wrong. CMOS 16th ed at 6.80 shows why. But the all-hyphen style is fine too. — ¾-10 00:44, 25 January 2018 (UTC)

Already addressed at Talk:Electric pump-fed engine#Requested move 24 January 2018.  — SMcCandlish ☏ ¢ 😼  20:01, 25 January 2018 (UTC)