Wikipedia talk:Manual of Style/Capital letters/Archive 21

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia
Archive 15 ← Archive 19 Archive 20 Archive 21 Archive 22 Archive 23 → Archive 25

The word "like"

Should the word "like" be capitalized within titles (excluding situations where it is the first word of the title)? I came across Template:Did you know nominations/Love Me like You where an editor cited MOS:CT for the capitalization of "like", though this page currently doesn't say whether it should generally be capitalized or not. Snuggums (talk / edits) 03:37, 10 February 2016 (UTC)

For further clarification, obviously the verb, noun, adverbial, adjectival, and conjunctive senses should be capitalized and that's covered. The problem is that like followed by a noun is usually glossed as a four-letter preposition (Wiktionary/OED). This page suggests that the prep. should not be capitalized and that was picked up by "Love Me like You" [sic] and "Love Me Like You Do", two songs whose likes are universally capitalized elsewhere but are being treated oddly here (WP:OR) because of this guideline. Are we just going to go it alone? or follow WP:ENGLISH WP:COMMON names? or is this just a four-letter preposition that should always be capitalized? — LlywelynII 14:55, 10 February 2016 (UTC)

My own opinion, fwiw, is that we can make it a general policy to always capitalize it or follow WP:COMMON usage, but we should not have a guideline here that encourages editors to park pages that ignore the formatting used by every WP:RS on their subject. — LlywelynII 15:19, 10 February 2016 (UTC)
  • This is already answered in the guideline: "The words that are not capitalized (unless they are the first or last word of the title) are: ... Prepositions containing four letters or fewer (of, to, in, on, for, with, from, etc.; but see below for instances where these words are not used as prepositions)." Hint: like in the constructions in which you are seeing it lower cased in titles is a preposition; it would be capitalized in a construction like "That's the Way I Like It" (verb).

    All major style guides off Wikipedia also advise this rule, but set varying cutoffs (3 letters, 5 letters, or vaguer constructions like "short" or "common" versus "long" or "uncommon", etc.). MOS picked a reasonable middle ground. Like all compromises of all sorts, not 100% of people will be happy with it, but the the compromise must be accepted for dispute to end and productivity to resume. Otherwise, those in favor of a 3-letter cutoff will be forever fighting with those in favor of a 5-letter one. This "issue" is now obviously WP:PERENNIAL. We do not second-guess every single line item in guidelines. Just accept them as compromises we agree to work under, and move on. MOS is internal documentation for how to write here; it is not an an article. It reflects internal consensus on what is best for the project and its readership, taking into account external conflicting sources, internal sources of conflict, the nature of writing in an encyclopedic tone and Register (socio-linguistics), and the needs and expectations of the broadest audience. This necessarily means avoiding jargonistic style from particular camps, since every camp's style preferences will conflict with those of the next camp over. Wikipedia is not a platform for advocacy of particular preferred style nitpicks found in a single off-WP style guide or in-crowd. Whether the entertainment press likes to capitalize these things is irrelevant; WP is not journalism much less entertainment journalism, and such publications usually derive the majority of their income from entertainment industry advertising, so they are not independent sources on this question. We accept rare style-rule exceptions, like "Deadmau5", "iPod", but only when independent, high-quality, reliable sources do so with near-total consistency.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  21:09, 11 February 2016 (UTC)

  • That's just patently false. I've looked into all of these "like" cases, and it takes less than 10 seconds to find RS that don't capitalize it, in those titles, and in reference to those works.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  21:54, 12 February 2016 (UTC)
  • No, with respect, that is patently false. Cursory scanning shows every source on those pages uses a capitalized L. On pages like "Moves like Jagger", every source is given with a lower-case L because of a bad CTRL+H job. Actually clicking through to the sources show they actually use a capitalized L. If you have any actual data rather than unfounded smears, kindly present it. — LlywelynII 02:27, 13 February 2016 (UTC)
Pick any case. I'll find you RS using "like" not "Like" in under 1 minute, and then I'll point you to WP:You can search, too. I'm talking about RS that exist, not the RS that fans of "Like" vs. "like" have already cherry-picked to use in the extant copy of the article, which is meaningless; all of our articles are moving targets and may change at any time. If you want to talk about sources, we're talking about what exists in the real world, not what someone picked temporarily to run with for now in hopes of getting their way. In the real world, the "conventions" on this vary widely by style guide, even more widely by particular publications' house styles. The result is that there is no consistency in actual off-WP publication in the handling of either a) short but uncommon prepositions (especially words that often are not prepositional in other constructions), or b) deciding what constitutes "short". There's not even consistency, especially when it comes to the low-brow entertainment press, with regard to lower-casing short, common prepositions, and lately not even the definite and in-definite article, in mid-title. That is not style at an encyclopedic register, it's just random, lazy chaos. MOS:CAPS is clear: If independent, secondary reliable sources aren't consistently capitalizing something, don't capitalize it on WP. That's the entire matter in a nutshell.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  16:53, 14 February 2016 (UTC)
So I shouldn't "cherry-pick" Van Morrison's own website for the title of the song he wrote? This is "random, lazy chaos", yes? Martinevans123 (talk) 17:32, 14 February 2016 (UTC)
I'm not sure which song title you have in mind, but whatever stylization appears on Morrison's album cover is a MOS:TM matter as well as a MOS:CAPS one. Aside from falsely crying "original research!" when people are actually relying on major English style guides, and then engaging in really obvious OR like imagining that Van Morrison gives a damn about typography (or even that he had any input into the matter – it was almost certainly decided by functionaries at a record label), several of you are approaching this from radically different angles that really have nothing to do with MOS or AT. You're either making a WP:OFFICIALNAME argument that is counter to MOS:TM - "we must carefully mimic the stylization of the owner of the trademark, no matter what"; or a WP:SSF argument that turns the notion of reliable sourcing on its year – "we know that music magazines are RS for things like album sales records, tour dates, and who puked on the tour bus, so that automatically means they're the most reliable sources for everything that ever has anything to do with music articles on WP, and they dictate how we are able to use the English language about this topic, even if all actual style guides on English disagree." This amounts to the same argument as "if Spin magazine contradicts a physics journal on how the acoustics of guitar amps work, then the music magazine is right." Others yet are convinced that whatever the most common style is must be adopted per WP:COMMONNAME, even though common name says nothing about style, WP:AT policy and the naming conventions (including for music, companies, titles, capitalization), etc. repeatedly defer to MOS on style matters, and MOS:CAPS and MOS:TM, etc., would not even exist if this "COMMONNAME is about style" mistake were actually true. It's a confusion of a thing with the style that is applied to it, like mistaking your spouses clothes for your spouse. See also Separation of content and presentation; names are content; how they are styled is presentation. Only in ultra-rare cases do this merge in the public mind, as with "Deadmau5" and "iPod". It's not lost on me or anyone else that virtually the only time people make overcapitalization arguments like what's going on here, it's when they're focused either on pop-culture fandom material or the deep guts of some esoteric discipline, and always basing their desire to do so on what publications specific to those disciplines are doing. See WP:JARGON; the insider styles of entertainment industry publications are just as jargonistic as any other, and they also have a WP:INDY problem; most of their income is from entertainment industry advertisement, giving them a strong incentive to follow trademark stylization in ways that no mainstream style guide would ever accept.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  18:09, 14 February 2016 (UTC)
I have no idea who puked on the tour bus, thanks. I linked the song title. The title of a song isn't "pop-culture fandom material". Thanks. Martinevans123 (talk) 18:25, 14 February 2016 (UTC)
I note that you addressed zero substantive points I raised, so I'm just moving on.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  02:05, 16 February 2016 (UTC)
  • "Like" is a very complex case, being variously a preposition, conjunction, noun, adjective, adverb, or part of a phrasal verb. No, it should not always be capitalized, just almost always. No MOS rule can cover every case concisely and clearly enough for typical editors, and people who think typical editors should defer to expert editors misunderstand the reason for Wikipedia's phenomenal early success. If relying on reputable sources (not necessarily merely "reliable" sources), assuming reputable sources adhere to consistent styling, then relying on their styling should achieve the same end. Relying on "important words are capitalized, unimportant words are not" almost always gives the same result. Sometimes not, but in these cases, I think the MOSers should let exceptions go if subject-interested editors citing sources disagree, because MOSers not particularly interested in the subject at hand arguing with editors who are, and discouraging them from following source, is a very bad idea. --SmokeyJoe (talk) 11:07, 12 February 2016 (UTC)
  • It's not "complex", it's "confusing to some people"; these are distinct matters.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  17:11, 14 February 2016 (UTC)
  • Because "Like" is complex, it is subject to use as a double entendre. A title like "People Like Us" or "Love Me Like You" can mean both senses, and should be treated as if they sometimes intend the other sense. Because of this, we would probably best avoid conflicts by generally capitalizing "Like". bd2412 T 13:07, 12 February 2016 (UTC)
  • yes. Certainly it is subject to double entendre. That makes it very hard for editors to decide, and BDD's point is correct. In this case it is not a double entendre. It is a fragment of "[They can't] love me like you [love me]". I think that makes it a conjunction. Going against *all* sources is pretty bad. --SmokeyJoe (talk) 14:48, 12 February 2016 (UTC)
  • (edit conflict) The obvious solution is to capitalize it in any case where sources tell us it's a double entendre. Not difficult. There are complaints above that MoS following virtually every style guide in print in not capitalizing short prepositions but capitalizing long ones is "OR" and "PoV", and now this new one suggesting that making up our minds about what is or isn't a double entendre (i.e. engaging in obvious OR) is the way to go, and that all the RS on how to use the English language – the RS that matter here – should be thrown out the window to advance the language activism PoV that all prepositions should be capitalized if anyone could ever have doubts about them. Well, obviously our interpretations of policy are widely divergent, but I'd bet a fortune that mine is correct. It's a moot point anyway, because WP:POLICY pages are not subject to WP:CORE. We create rules in policies and guidelines based on internal editorial consensus; it is informed by what external sources are doing (e.g. on style, at a style guideline) but it is not dictated by them, because these are not encyclopedia articles, they're internal procedural documentation (in this case on how to write here to produce a consistent encyclopedia and avoid editorial strife) based primarily on WP's own needs. All of the present (and rather perennial) strife about this WP:LAME "I want 'Like A Dude'" stuff has come up because people are simply refusing to follow the guidelines, in the mistaken beliefs that CCPOL applies to POLICY and that COMMONAME applies to style. It's like showing up to an American football game and trying to play Canadian football, while also trying to follow some rugby-specific rules, then getting angry that everyone else isn't playing it your way.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  17:11, 14 February 2016 (UTC)
  • No matter how large the "almost always", there should always be tolerance for exceptions. A good condition for an exception is if every source does it the other way. --SmokeyJoe (talk) 22:03, 12 February 2016 (UTC)
  • That's necessarily the case, since MOS is just a guideline. If sources tell us that a particular case of [l|L]ike in a title is a double entendre, then it's not simply a preposition, and it should be capitalized here. "Do It like a Dude", for example, fails that test. The only argument for it is that a lot of off-WP sources, mostly entertainment journalism ones, write it inconsistently as "Do It Like a Dude", "Do It Like A Dude", etc. WP doesn't care, because WP:NOT#NEWS and we do not follow Rolling Stone house style. Even all mainstream journalism style guides would not support "Do It Like A Dude", if MoS were based on them, which it's not (NOT#NEWS again). Its just demonstrably substandard writing, in any register.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  17:17, 14 February 2016 (UTC)
  • In the OP's second example—"Love Me like You Do"—like serves as a conjunction, not a preposition (you is the subject in the clause "you do", not the object of a preposition), so like should be capitalized in that instance. Deor (talk) 11:18, 12 February 2016 (UTC)
    • It's not a conjunction here. Try substituting another conjunction: "Love Me and You Do". It's similar to "Love Me as You Do". Walter Görlitz (talk) 16:55, 12 February 2016 (UTC)
  • The main point here is Love Me like You, as it is causing an issue at DYK.  — Calvin999 11:38, 12 February 2016 (UTC)
    • Yeah, that's not the main point here any more but it does look like this will take awhile. I'll go ahead and remove my objections there since it was obviously a WP:GOODFAITH attempt to uphold our style rules instead of accidental error. — LlywelynII 04:55, 13 February 2016 (UTC)
  • Generally capitalize, and for that matter, just change "Prepositions that contain five letters or more" to "Prepositions that contain four letters or more". While I oppose just saying capitalize like other sources do—we have a MOS for a reason—this does conform with others' practices more often. I want us to follow our own rules, but I also don't want us to stick out in ways like this. I don't want to see editors bickering and pulling apart phrases to tell what's a preposition or not. --BDD (talk) 13:29, 12 February 2016 (UTC)
If we change it, we would uppercase all of such prepositions and then disregard the RMs in talk pages. I like "into" lowercased; same for "past" (unless double entendre obstructs). --George Ho (talk) 17:43, 12 February 2016 (UTC)
Regardless of what we like, the RS generally leave "over" and "into" lower case but (very famously in the case of Wikipedia) not always. A general fix here is going to have to avoid any bright line fixes. Capitalized "Like" in titles, however, does seems like a bright line: it seems like it's almost never left lower case outside of our pages. — LlywelynII 05:07, 13 February 2016 (UTC)
Why are we encouraging ourselves to disregard parts of speeches and assume that four letters are long enough to uppercase it within titles of works? Should we just basically follow the pre-discussion rules? As for Star Trek, not a good example mainly due to fandom and pop culture. Of course, we are discussing also pop media. George Ho (talk) 18:01, 13 February 2016 (UTC)
Huh? We're following all the major reliable sources on written English usage, which unanimously agree do not capitalize short prepositions or the indefinite and definite articles in the middle of a title. It's been done this way for many generations, because capitalizing things like "a" and "from" in mid-title make the title harder to parse for meaningful words. You can even hear the difference if you read it out loud: "Do It like a Dude" is accented on do and dude, the other words are just getting us from the verb to the object; "I'm Not Sure You Like Me" would be accented on sure and like, as the most meaning-conveying words in the title; "like" is a main word in the latter case, not some preposition we mentally blur over But this is not the place for a psycholinguistics lecture.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  17:23, 14 February 2016 (UTC)
  • I agree with BDD, 5 --> 4, a simple solution. Rothorpe (talk) 15:57, 12 February 2016 (UTC)
  • Agree with @BDD:, if the official capitalisation is unclear, generally capitalise, I could not find any RS supporting the uncapitalised 'like' in Feels Like Home (Linda Ronstadt song) (which is why I came here, and if you are wondering what brought me to that page it is Qantas again for they use that song in an advertising campaign). - Champion (talk) (contribs) (Formerly TheChampionMan1234) 02:51, 25 February 2016 (UTC)
  • What am I missing? "Love Me Like You Do" is a proper noun used to describe a thing, in this case a song, which has a name. Why wouldn't "like" be capitalized as a being a proper noun. Mkdwtalk 16:55, 12 February 2016 (UTC).
    • No. Well, yes but no. It is a particular type of proper name treated specifically, it is a composition title, with "rules" set out at Wikipedia:Manual_of_Style/Capital_letters#Composition_titles. --SmokeyJoe (talk) 22:00, 12 February 2016 (UTC)
    • Wouldn't that argument apply to any/every word in a title? There are other words that clearly should not be capitalized when they occur in titles and are not the first or last word of the title. mwalimu59 (talk) 17:58, 12 February 2016 (UTC)
  • Yes and that would be up to the person who created the name as to whether the capital is part of the proper name or it should be treated like a preposition in a sentence. Not the community on aesthetic preference. By the same logic, does that mean someone who is named John Like Smith, their name would appear as John like Smith? What about someone who named their company Like Us? Would we lowercase both to "like us"? The arguments about prepositions are superficial because a company name can be a sentence and include prepositions. Mkdwtalk 20:59, 12 February 2016 (UTC)
  • Agree with BDD, I don't have a problem with 4-lettered prepositions like "with" or "from" being capitalized either as long as it's done consistently. Timmyshin (talk) 17:13, 12 February 2016 (UTC)
  • *Sigh* I did my best to do case-by-case method, but that led to repetitive arguments and endless cycle of one side outbalancing the other. I tried centralizing the discussion at WP:VPP, but there wasn't enough discussion over there. I didn't know where else to discuss without RFC tag, like above. Moving to the central issue, "common sense" in Wikipedia contradicts "common sense" in the real world. Clearly, almost no one outside Wikipedia is aware of the "like" issue. We can't ignore the (upper/lower)casing rule to our own benefit without knowing whether it improves Wikipedia. Too bad "common sense" Wikipedians use their interpretation on rules without knowing whether they share the same senses. Recently, I started a separate discussion at Talk:Love Me like You, where double entendre and repetitive arguments were discussed. George Ho (talk) 17:14, 12 February 2016 (UTC)
  • The Chicago Manual of Style and many other major style guides seem to advocate the capitalization of Like in titles. I see no reason to deviate from what is pretty much accepted grammar. - SanAnMan (talk) 17:26, 12 February 2016 (UTC)
  • The Wikipedia Manual of Style should not be in the business of re-styling the titles of other people's works. It's fine to state a default rule, but we should be clear that where the sources overwhelmingly differ from that rule, we follow the sources. The current exception for songs named after their first line was based on this principle: why not just generalize it? – Smyth\talk 17:53, 12 February 2016 (UTC)
  • Notwithstanding what style guidelines say, can anyone name an example of a song, movie, or other work with the word 'like' in the title where the word is generally not capitalized in reputable publications and webpages? The example that led me here was Fly Like an Eagle, which I don't recall ever seeing with Like not capitalized, and it looks very odd to my eye when it isn't. The same is true of some of the other examples others have given in this discussion. mwalimu59 (talk) 17:58, 12 February 2016 (UTC)
  • I'd strongly support this change, and I've argued for it at Someone like You (Van Morrison song), where the contrast between title and content looks quite ridiculous, and at Someone like You (Adele song). It's time this section of MoS looked a little less like a world of its own. Martinevans123 (talk) 18:30, 12 February 2016 (UTC) p.s. are these two examples conjunctions or prepositions? I keep forgetting.
Both of them use "like" as prepositions, Martin. If there are two clauses, "like" would have been used as a conjunction, but there isn't another clause in each title. George Ho (talk) 19:02, 12 February 2016 (UTC)
Thanks (but I'll probably still try to keep forgetting). Martinevans123 (talk) 19:05, 12 February 2016 (UTC)
Almost forgot; if the s comes after "like", that becomes "likes", a present-tense verb in a third-person narrative. George Ho (talk) 19:07, 12 February 2016 (UTC)
... and if it comes before? Martinevans123 (talk) 19:38, 12 February 2016 (UTC)
... you don't want to know. — LlywelynII 05:16, 13 February 2016 (UTC)
  • 'like' isn't an isolated four letter preposition; we either change the rule to always capitalise all of them or we don't and it continues to be done on a case by case basis. I don't accept either of the arguments: "it causes endless debate" nor "it is difficult to determine the part of speech in a double entendre". If we endlessly debate then we are not harming the output in any way only ourselves and that is our choice. That creative writing plays on the multiple meanings of words is well-known and we should support it by being careful in our presentation of others' works. I see nuances of importance of words in all titles and I intensely dislike the common publishing practice of Treating Every Word In A Title Like It Is Of Equal Importance, which is the graphic equivalent of speaking in a monotone. The question we should ask is: "what is added or taken away when we choose to treat all instances of a word the same regardless of their usage?" I think we lose meaning: 'I Like It like That' has a rhythm to it which is lacking in 'I Like It Like That' - the drop in emphasis in speaking the phrase is mirrored by the drop from upper to lower case. Consider the difference in spoken emphasis in "Why don't you like me?" and "Why aren't you like me?" As a general point, we are moving further and further from a humanist medium where the appearance of text is bound up in its use to a point where we simply accept the banal mass-produced default on our devices. It pains me that, while I consider Wikipedia to be one of the great creations of this century, compared to earlier printed reference works it is as a shopping list is to Book of Kells. I vote for continuing to debate each and every instance of the word 'like' to ensure we pass on the maximum amount of information and knowledge to posterity. If this means that 'Someone Like You' is presented thus, because the whole point of the song is that the singer can't have the person but only a facsimile, and is sung with the emphasis firmly on the middle word (unusually - and that's part of the reason it's brilliant) or that 'Love Me like You' differs from 'Love Me Like You Do' for similar reasons, then that is OK as long as we arrive at it for reasons of conserving original intent and not just because the current Zeitgeist is to do one thing or another. Btljs (talk) 21:18, 12 February 2016 (UTC)
"'I Like It like That'" is a special case where the verb can be contrasted with the preposition. In that case, one could argue, having the lower case letter aids comprehension. I'm concerned with all the song and book titles that simply don't need a lower case letter - and that's why the authors and songwriters don't give them one. Martinevans123 (talk) 21:57, 12 February 2016 (UTC)
That's my point: they're all special cases and need to be considered individually. It's not just 'like': what about 'over' as in Bridge over Troubled Water versus Can't Get Over You? Btljs (talk) 22:07, 12 February 2016 (UTC)
"Bridge over" uses it as a preposition. "Get over" is a phrasal verb. George Ho (talk) 17:53, 13 February 2016 (UTC)
Although I agree with Btljs's conclusion, I think encouraging editors to get into artistic debates on "original intent" and "the whole point of the song" is as misguided as the current debates on whether something's a preposition or not. If all sources capitalize the title a certain way, then that should be decisive. It's not our place to second-guess the way it "should" have been written. – Smyth\talk 23:00, 12 February 2016 (UTC)
(edit conflict)If RS do not tell us what the "original intent" or "whole point" of the song is, it's original research and PoV. If there's clear evidence that a preposition in a title is not actually intended as one (or intended as one only in half of a double-entendre), we would not lower case it, so the whole "intent/meaning" thing is a red herring.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  02:05, 16 February 2016 (UTC)
'Bridge Over Troubled Water' looks better, as does 'I Like it Like That'. Bring in a four-letter rule. (And no tedious arguing over parts of speech.) Rothorpe (talk) 23:13, 12 February 2016 (UTC)
"Looks better" is not a valid reason to uppercase or lowercase a word. George Ho (talk) 17:53, 13 February 2016 (UTC)
Why not? I thought that's what the discussion was about, at least in part. Also (which may be equally unimportant) "I Like it like That" doesn't echo the equal stress given to the two Likes in both songs of that name. Rothorpe (talk) 18:43, 13 February 2016 (UTC)
Because it's your personal PoV subjectivity. What "looks better" to you look terrible to someone else. All style matters are subjective. What we care about is what is practical and what is conventional. There is no universal convention on prepositions in titles. There are three: Never capitalize any of them ("high academic" style), capitalize them if five characters or longer (the typical style in general publishing), or capitalize them if four characters are longer (a style represented almost nowhere but in journalism style guides, which writ in news style, which is not WP's style. MOS uses the middle ground approach. This is an entirely reasonable compromise position. Not 100% of people will be happy with it, but about 1000% more will be happy with it than either of the more extreme choices. If we make up own own bullshit rule, like "follow the 5-character rule, except for like because a handful of people will never stop fighting about it", or "follow the 5-char. rule except for pop music", or whatever, just to please one wikiproject, everyone else on WP will shout "original research!" and just start another wave of conflict.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  02:05, 16 February 2016 (UTC)
  • "Like" should be excluded from the words we lowercase by default. As others have pointed out, sources nearly universally uppercase the word and other manuals of style (in particular the CMoS) capitalize it. Numerous requested moves have resulted in an uppercased Like, despite the guidance at MoS (in fact, the only argument for the lowercase at those has been the MoS). Given that Wikipedia is not a bureaucracy and written rules "document already existing community consensus," it's clear that the change should be made. Calidum ¤ 22:18, 12 February 2016 (UTC)
    • There is no list of specific words we lowercase by default, and no basis in any sources anywhere for making a special exception to a general class for this one word just because you "Like" it that way. We follow the same under-five-letters rule for propositions as most style guides on general English usage (and you really wouldn't like the "never capitalize any propositions at all" rule of the more academic style guides, some of which, like Chicago Manual of Style are not very academic but quite mainstream, and some of which specific to music, like Writing About Music: A Style Sheet (3rd ed. 2014).  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  02:05, 16 February 2016 (UTC)
  • Not closely related, but what about Four past Midnight and dot the i? Why is lowercasing the i okay, while lowercasing like is not? --George Ho (talk) 18:05, 13 February 2016 (UTC)
I feel like this is a lost battle really. To me this is 'dumbing down' to achieve the same level as the sources which are either lazy, ignorant or just don't care. If we can't see why 'Dot the I' would be stupid then we're in real trouble. Do I care what the media think about the pedantry of these discussions? No more than they care about what I think of them. What the hell - just capitalise everything, then we'll have nothing more to discuss. Or- or we could do our job which is to look at each page on its own merit and treat MOS as a starting point and not a straight jacket. This is not second guessing - it's research - and even if it were, we'd still be right some of the time which is exactly what will happen if we just set a rule and walk away. Btljs (talk) 19:57, 13 February 2016 (UTC)
Many songs and books have been published with titles in which "like" is capitalised. This is not a question of "'dumbing down' to achieve the same level as the sources which are either lazy, ignorant or just don't care." It's about accurately reflecting what song writers and authors actually intended. Martinevans123 (talk) 20:03, 13 February 2016 (UTC)
If we want to reflect writers' or authors' intentions, we should search for past interviews, self-published sources, or primary sources and then interview them. Shall we ask Stephen King or someone else in-person? Otherwise, let's stick to rules without changing them. George Ho (talk) 03:38, 14 February 2016 (UTC)
You're suggesting, perhaps, that publishers simply decide on behalf of writers or authors? Maybe they do, but that's still "normal usage", isn't it. Otherwise, you mean "let's stick to some arbitrary rules that have no basis in normal usage"?? Quite happy to see a general rule of "three letter prepositions" instead of four, but with the possibility of exceptions if they can be established on a case-by-case basis. Quite happy to see a separate wholesale exception for "Like." Martinevans123 (talk) 11:14, 14 February 2016 (UTC)
Actually, in my opinion, having an arbitrary rule whereby proper names are sometimes not capitalized, is in fact itself a dumbing down of the language. Rather than editorial oversight to refer to research and citation, editors are using their own personal preferences for a look. It's not based in grammar or research, it alienates the encyclopedia from common usage language, and it really seems to serve no other purpose than to relieve editors from having to do this comparative work -- i.e. a dumbing down of process whereby editors refer to a rule rather than refer and judge based upon published material. Mkdwtalk 01:37, 14 February 2016 (UTC)
Um, Mkdw, all style and grammar rules are arbitrary. Thats why they're rules, with some exceptions, instead of total random chaos. If they were not arbitrary, everything would be an exception. There is nothing intrinsic, like the physical properties of matter and energy, to anything at all about style. Written communication could be written all run-together with no case and no punctuation if that how people wanted it (if you don't believe that, see any of thousands of ancient manuscripts).  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  17:45, 14 February 2016 (UTC)
"Common" language is not exactly accurate and not similar to standard or formal language. Also, sticking to rules is not preference; too bad "sources" are interpreted for preferences by advocates of uppercasing. --George Ho (talk) 11:25, 14 February 2016 (UTC)
That is, in point of fact, the precise opposite of what is occurring here. SM'C is jumping through hoops trying to avoid accepting Billboard, professional style guides, and other sources (along with core policies about RS, RF, and CONSENSUS along the way) in favor of a house style supported by quite literally no one outside of Wiki. (And, as linked and shown here, by only a minority of Wiki's own editors.) I've linked well over a hundred pages above. Kindly find any of them where prepositional "like" is lower case in the majority of their sources. I have looked and I haven't seen a one. This isn't about "dumbing down" anything. On at least this particular word, the style guide is simply wrong, advocating formatting at odds with standard English, with real-world consequences in the form of rolling years of edit wars which continuously decide against this policy on page after page. — LlywelynII 14:57, 14 February 2016 (UTC)
Cite even one style guide that says to capitalize short propositions, or the definite or indefinite article, in mid-title. I'll provide 20 or more that say not to. (But not here; I don't source MoS discussions, because MoS is not subject to WP:CCPOL; I'll do it at Talk:Capitalization in English; that article is in a terrible state – which probably has a lot to do with this silly thread – and badly needs additional sources, and a lot of additional information.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  17:45, 14 February 2016 (UTC)
That's all strawman and fluff. We aren't talking about a; we're talking about like. An editor above referred you to the Chicago Manual. You're welcome to cite any of them you like and we still might argue with them. But any evidence at all is going to be more productive and persuasive than your attempts to shut people down by trying to use policy to override OR, COMMON, and RS. No one is going to listen to that, nor should they. — LlywelynII 18:59, 14 February 2016 (UTC)
As to your examples, you can't possibly be serious. Based on your own comments at "Love Me like You", you're perfectly well aware why the editors at "dot the i" styled the page that way and why that WP:IAR situation has nothing to do with this conversation. — LlywelynII 15:00, 14 February 2016 (UTC)

Arbitary break

This listing of two camps of editors, which looks like calling up the armies, is senseless and a false dichotomy. Anyone "voting" for capitalizing or lowercasing "like" across the board is making a mistake, and I'm skeptical that more than handful of the people you're trying to pigeonhole in this manner are mkaing it. Whether to capitalize it or not is dependent on whether it's being used as a preposition or not, which is a case-by-case matter. "Do It like a Dude" is an 'In the manner of' prepositional use. In the sentence "Just like I suspected, the cats did it", like is acting as a conjunction; no actual comparison is happening. "I dressed like a pirate" is prepositional. A large number of these articles need down-casing.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  02:15, 17 February 2016 (UTC)
Indeed it does. Many thanks, Llywelyn. Rothorpe (talk) 04:06, 13 February 2016 (UTC)
Yes, style guide is wrong. Martinevans123 (talk) 18:38, 13 February 2016 (UTC)
I you need to go actually read WP:PERENNIAL. The meaning of the term here is that people keep bringing something up again and again yet consensus never changes because their reasoning is flawed. They need to drop it and go do something productive instead of disruptively rehashing the same campaign over and over again.  — SMcCandlish ☺ ☏¢ ≽ʌⱷ҅ᴥⱷʌ≼  18:13, 14 February 2016 (UTC)
The fact that there's been confusion between how to apply WP:COMMONNAME and MOS:CAPS / MOS:TM at the same time, leading to some incorrect decisions at WP:RM, doesn't really tell us anything other than that it needs to be clarified better than that COMMONNAME is not a style policy. It never has been, it's just that incoming waves of editors seem to take an inordinately long time to figure this out.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  17:45, 14 February 2016 (UTC)
Ah yeah, right, I'm confused. Thanks. We can all look forward to more inordinately long time-wasting discussions on individual article talk pages, then. Martinevans123 (talk) 17:56, 14 February 2016 (UTC)
Can't have it both ways. The argument that there should not be a general, and necessarily arbitrary, rule about some point of capitalization that is followed by default, and rare exception made to it, is inherently an argument for treating everything on a case-by-case basis, which will multiply, not reduce RM disputes. The reason we typically have fewer than 100 open RMs across millions of articles is precisely because we have general rules and almost always apply them, even if some minority of editors are not perfectly pleased with every result. They recognize that it's more important for dispute to stop and for productivity to resume, than to fight and fight and fight until they get their way (otherwise none of our dispute resolution processes would ever do anything, and WP would have died shortly after being launched). PS: I was referring to wrong-headed "vote-counting" closes at RM that did not apply policy correctly when I said "there's been confusion between how to apply WP:COMMONNAME and MOS:CAPS / MOS:TM at the same time", and this was very clear, since I made that clarification in the same sentence. If you choose to self-identify with a particular word in that statement on the basis of arguments you're making here, that's on you, not me; please do not project hurt-feelings sarcasm at me for insulting yourself.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  18:23, 14 February 2016 (UTC)
It was hard to tell if I was in one of the "incoming waves of editors" or not. No, it wasn't very clear. And, no I don't have any "hurt feelings", thanks. If we "can't have it both ways", why have we got so many exceptions to this increasingly pointless rule? I'm not concerned with 100 open RMs across millions of articles, just with whether Like should have a capital L. Martinevans123 (talk) 18:32, 14 February 2016 (UTC)
I'm not sure what is being proposed: a) capitalise all 4 letter prepositions; b)capitalise only 'like' and leave other 4 letter prep.s alone; c)Some other rule change? If (a) or (b) then is there any style guide which supports either of these rules? I'm simply not interested in counting occurrences - although I know a lot of editors think this is reflecting the real world which is what we should do. I disagree: we should report the real world in a house style. And replacing a case-by-case decision process with a rule is always dumbing down in my book. Expedience over quality. Btljs (talk) 15:13, 14 February 2016 (UTC)
It's clearly being proposed that "like" should be a special exception. It's a specialist style fallacy, as usual. The notion is that because the entertainment press, and some mainstream press, who happen to be following their own house styles like to expediently capitalize "like" no matter what, that these are reliable sources for how WP "must" write English any time it's talking about entertainment topics. This is, of course, not tenable. WP is not journalism much less a specific genre of it, and MOS is based on WP consensus about what's best for WP editing and reading, which is informed by actual reliable sources for how to write formal English. The FAQ at the top of WT:MOS has covered this for years.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  17:45, 14 February 2016 (UTC)
You don't see an album or an artist's website as giving some kind of indication as to how a song title should be written? Sorry that my reply is not laden with jargon and policy links. Martinevans123 (talk) 18:36, 14 February 2016 (UTC)
[Moved to reflect unconnected to other comments in the thread.] The real world capitalizes prepositional "like" in titles, as does most of Wikipedia and most Wikipedians. That's the scope of this discussion, at the very least; some editors above want to expand that into a rule that would cover "into" and "upon". I'm not sure that would be helpful; I'd need to see stats on the general usage. In any case, it's not expedient when a house rule so completely contradicts common English usage that we have unending move wars in favor of stylization unsupported by secondary sources, as above. — LlywelynII 15:40, 14 February 2016 (UTC)
MoS isn't based on song titles written in inconsistent journalism house styles, it's based on a) WP's need for consistent rules that people can follow without having to memorize a thousand random exceptions, and b) actual mainstream style guides for the English language, which do not support what you want to do here. This is taking on a strong character of WP:GREATWRONGS campaigning.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  17:45, 14 February 2016 (UTC)
The bottom line is wikipedia is built upon reliable sources; when reliable sources capitalise or don't capitalise like in a title they should be relied upon and followed.Changing the capitalization to make it differ from reliable sources is original research and very inaccurate and should always be reverted.Atlantic306 (talk) 18:31, 14 February 2016 (UTC)
Why is a song title on an album or on the artist's website "journalism"? Martinevans123 (talk) 18:37, 14 February 2016 (UTC)
Again, kindly take a break from your attempts to assume bad faith and lawyer a defense of this policy and post any evidence supporting your position. As copiously documented above, this is a single instance where the MOS is simply wrong as a matter of English style. There is no sensible reason to defend it on the grounds that since others arbitrarily did it the wrong way at first we should soldier on. — LlywelynII 18:46, 14 February 2016 (UTC)
Please don't reframe the argument to fit your case. This is not about the MOS being wrong: a style guide cannot be wrong, only at odds with other information. The words 'wrong' and 'arbitrary' should be avoided if you are serious about resolving this rather than simply restating your view point with no extra information. We already know what you think. Can you (as I have previously asked) provide ANY style guide which supports your case? If not then there is no foundation for changing the MOS, other than to clarify the reasoning behind it, which SMcCandlish has already stated is already in place. Btljs (talk) 19:11, 14 February 2016 (UTC)
Thanks to JezGrove for reminding us: "New Hart's Rules says: "Exactly which words should be capitalized in a particular title is a matter for individual judgement, which may take account of the sense, emphasis, structure, and length of the title. Thus a short title may look best with capitals on words that might be left lower case in a longer title." (2005 edn, page 133) For what it’s worth, WP:NCCAPS recommends Fowler’s Modern English Usage (3rd edn), which itself acknowledges its debt to, and generally follows, the rules set out in Hart's." Martinevans123 (talk) 22:08, 14 February 2016 (UTC)

That's cherry-picking and OR, though probably accidental. New Hart's Rules is not Hart's Rules. They're different works. HR, and its later revisions as The Oxford Guide to Style and Oxford Style Manual say: "Capitalize the first word and all nouns, pronouns, adjectives, verbs, and adverbs, but generally not articles, conjunctions, or short prepositions." (Oxford Style Manual, 2003 [2002]; R. M. Ritter, Oxford U. Press; ISBN 978-0-860564 Parameter error in {{ISBN}}: checksum-5. At "Capitalization and treatment of names: Titles and subtitles of works" § 4.1.8, pp. 77–79, in this OSM printing). This is what modern editions of Fowler's are based on. The NHR is too new to have had any noteworthy influence, and it simply does not offer any guidance, leaving it even up to "individual judgement", which is obviously not acceptable in any context in which consistency is desired, e.g. in a work with multiple writer/editors.

But you're even selectively quoting from NHR to only show it saying what you want it to say. What it really say, above your partial quotation, is: "The traditional style is to give maximal capitalization to the titles of works published in English, capitalizing the first letter of the first word and of all other important words (for works in other languages ...). Nouns, adjectives (other than possessives), and verbs are usually capitalized; articles, conjunctions, and prepositions are usually left uncapitalized." [Emphasis added.] Later, at "Musical works: Popular music and traditional songs", &sect. 8.6.2, pp. 153–4: "Song titles in English are set in roman type with quotation marks, capitalized according to the style adopted for titles in general"; it proceeds to show album titles done just like book titles, traditional songs (named for their first lines) done in sentence case, then complicated rules for classical. So, it does in fact give advice, against capitalization of short prepositions, and then it basically contradicts itself and says it's up to individual whim, and then goes back to asserting that there are rules and that songs/albums should follow those of books/journals, and even more complicated ones for classical.
This is clearly a result of failures to coordinate between the old staff of Hart's, and whoever was working with Ritter, and the incoming editor, Waddingham, plus rushed editing that failed to reconcile the goals of recommending academic usage (the main purpose of the Oxford guide in every edition) with the secondary one of describing observed use in other contexts, such as journalism and informal writing. They've mixed prescriptive and descriptive roles in a way that needs copyediting on their end for the passage to make much sense. For an overview of the problems inherent in OUP's decision to mix-and-match prescriptive and descriptive approaches in the same volume, see this detailed New York Times piece on the matter [1] (it's about the Burchfield edition of Fowler's, but it was produced by Burchfield [now also deceased, and replaced with a new editor, Butterfield] during the same time as Ritter's work on what had been Hart's, with the exact same goal, outlook, mixed intent, and mixed results).

If anyone is thinking Fowler's is going to back them on things like "Do It Like A Dude", they'd be wrong. The current Fowler's (ed. Butterfield, 2015) cites (at § "capitals", p. 132) the NHR but it cites NHR own section on capitalization in general (pp. 88–100), not the titles of works section, and Butterfield does not have a separate entry on capitalization of titles of works. So it is original research to claim that Fowler's agrees with NHR on how to capitalize work titles. Neither did the Burchfield ed. of Fowler's (pp. 128–129), which cited the original Hart's pp. 8–14 – that's pre-OGS/OSM. Both editions of Fowler's present titles of cited works with lower-cased short (4-letter or short) prepositions and indefinite/definite articles throughout their entire content.
 — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  02:13, 16 February 2016 (UTC)

Yes it says "generally", doesn't it? Martinevans123 (talk) 09:01, 16 February 2016 (UTC)

Arbitrary break 2

A few remarks:

  • To me it seems right and proper for Wikipedia to have a house style and to strive for some degree of professionalism and consistency across articles. As the guideline says, that is a practice that is universal among high-quality publishers.
  • I don't know the history of why the Wikipedia consensus resulted in four versus five letters as the cutoff for "short" versus "long" words. I assume that dividing line is used somewhere else, but I don't really know, and I'm not an expert on such matters. To me that dividing line seems reasonable, but I think reducing the cutoff point would also be OK, as long as we have some reasonably clear rule.
  • Whatever our general rule is, there should be a strong presumption that we will follow it. We should not just deviate from our guidelines in random localized ways for individual articles without some strong indication that a particular case should be an exception to the general rule.
  • There is a big difference between the name of a work and the styling of the name – especially when it comes to as small a detail as capitalization. Wikipedia's "common name" practice is properly about names, not the styling of the names – not the choice of capitalization or fonts. Publishers of creative works typically wouldn't even bother to check with the original author when writing a title on the cover art of a book or in their publishing catalog. Capitalization is absent when reading aloud or singing, and artists and publishers generally don't comment about the capitalization of titles. In fact I can't recall ever hearing an author, artist or publisher talk about whether a conjunction or preposition in the title of their work should begin with a capital letter or not (or a verb or noun, for that matter). Even if they did, I wouldn't necessarily think I needed to follow their suggestion – indeed, I might bristle at their attempt to promote their work by trying to control the typographical choices I make when discussing it. Trying to control the capitalization of the name of a work seems almost as silly as trying to control which font people can use when writing the name. We are writing words, not rendering logos.
  • The people who promote things generally tend to over-capitalize (or sometimes to undercapitalize as another way to make something stand out by looking unusual). Wikipedia generally eschews excess capitalization, and I'm glad it does. People often seem to just use capitalization whenever they think some word seems important or has a special meaning. Excess capitalization is annoying and ugly, and appears informal and promotional.
  • Most ostensibly independent sources that comment about popular creative works are not really so high in quality, not really so independent of the publishers, and are often generally somewhat promotional in tone.
  • So I basically don't think attempting to survey sources is very helpful or desirable for questions of capitalization styling. The need to do that would also seem to mean we would get into endless tiresome case-by-case arguments about relative frequencies in sources and which sources are adequately reliable and adequately independent in ways that would "blow with the wind".
  • Remember the capitalization disputes about the common names of birds and butterflies? Specialist literature sometimes has its own capitalization conventions, but Wikipedia is not specialist literature.
  • "Like" doesn't seem like such a unique word that it should be given explicitly different special treatment from others.

—BarrelProof (talk) 19:48, 14 February 2016 (UTC)

Yes, some good points,. But just take a look at "Someone like You (Van Morrison song)" where the "authority" of the article title is wholly undermined by the article's own 12 instances of "Somebody Like You" - the first of which is in the opening line. Am I one of those people who "generally tend to over-capitalize"? Hardly. Martinevans123 (talk) 21:54, 15 February 2016 (UTC)
That's a WP:OFFICIALNAME-style argument, and the design staff at the record label are not "the artist"; it's original research to suppose that because an album was printed a certain way that this reflects the exact intent of a singer-songwriter. And ultimately, why should we care? Sony's exactly intent in writing SONY in all caps is to scream their company name at people to get attention, and we just don't care to help them do that. We (like almost everyone else) have a standardized way of approaching such things, so that the results are consistent, not a random hodgepodge no one can predict.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  20:18, 25 February 2016 (UTC)
  • To me, it is less important what the rules of the Wikipedia MOS actually are and more important that the rules are applied consistently, because that's the main purpose of a MOS: providing (internal) consistency. While I see nothing wrong with the current rules, it seems changing the rules would lead to a higher acceptance of MOS:CT and thus to an improved consistency, therefore I'm not opposed to changing the rules. However, as was mentioned before, implementing a special rule for "like" would be arbitrary. After all, there are other prepositions that should be lowercased according to MOS:CT, but are almost always capitalized outside Wikipedia (One Flew over the Cuckoo's Nest, Once upon a Time in the West). Therefore I'd prefer changing the cutoff from 5 to 4 letters for all prepositions (incidentally, this would match AP Style). Darkday (talk) 20:48, 14 February 2016 (UTC)
    • Changing the cutoff from 5 to 4 might even resolve that nagging feeling about Star Trek Into Darkness! —BarrelProof (talk) 21:44, 14 February 2016 (UTC)
      • That would disrupt the consistency of all other prepositions. As for into, I will try to propose lowercasing it in all other titles. Star Trek fandom triumph the rule. George Ho (talk) 22:45, 14 February 2016 (UTC)
  • Agree with BarrelProof – It is OK for WP to have a house style; the style currently recommended by the MOS works fine, in spite of some projects who Like to over-capitalize their composition titles. No particular need to treat "like" as special; lowercase when it's a preposition is less confusing than the opposite. It is a commonly recommended style; see many grammar and style books: [2], [3], [4], [5]. Dicklyon (talk) 01:03, 15 February 2016 (UTC)
    • I think it was very helpful for Dicklyon to link to a few style guides, and I do see that one of them mentions the use of four versus five letters as one dividing line that is sometimes used. One thing I notice is that our MOS:CT guideline not only has a distinction between four and five letters, and a distinction between conjunctions and prepositions, but also a distinction between coordinating conjunctions and subordinating conjunctions. A substantial number of the title discussions in question that involve "like" versus "Like" seem to involve that distinction between coordinating conjunctions and subordinating conjunctions (which mostly hasn't been pointed out in the discussion above – e.g., there hasn't been a single mention of subordinating conjunctions in this discussion so far, although our current guidelines require identifying whether "like" is being used that way or not). I wonder if our rule is getting a bit esoteric by making a distinction on that basis. Is there a good reason to draw that distinction? I notice that none of the style guides that Dicklyon linked appear to discuss that question. Although I consider myself above average in my ability to write in grammatically correct English, I find it hard to identify the differences between these three uses of "like" (preposition, coordinating conjunction, subordinating conjunction – not to mention the other ways that "like" can be used). —BarrelProof (talk) 06:39, 16 February 2016 (UTC)
When used as a conjunction, like is always a subordinating, not a coordinating, conjunction (and therefore should be capped). Deor (talk) 06:45, 16 February 2016 (UTC)
  • I don't have much sympathy for this itch to cap everything in sight just because the band did it. Must we also use the original font and font-size? Curly glyph for apostrophe, where our house style is straight? Tony (talk) 01:24, 15 February 2016 (UTC)
  • No, I don't think we need to do that at all. And I don't have much sympathy for that argument either. The question is: "Should the word "like" be capitalized within titles...?" Martinevans123 (talk) 08:49, 15 February 2016 (UTC)

...and just to be totally contrary - should the MOS cover it at all? If the back of the CD/Vinyl/Cassette(ye gods) or whatever lists it with a capital L or not, then surely that is what should happen. The direction should be from the release from the group/artist. Otherwise the leading track on Marillion's second album Fugazi, would be Assassin and not Assassing if we decide to tinker with it and remove the 'G' at the end because it doesn't belong on the word as spelt in British English. I know that some bands list all tracks with all words as lower case (oh how funny they are) and yes, that will be a sticking point - buy my point is essentially the same; how it is written by the release should be the guide and not in how we think the architecture of the grammar fits in with MOS. The joy of all things (talk) 20:59, 15 February 2016 (UTC)

Quite agree. A sensible observation. The use of unconventional styling, c/o graphic artists on album covers, and that record labels often employ only block capitals, need not be a sticking point. Conventional forms are very often found on the titles of published scores, at the artist's own website and in sleeve notes / CD-booklets. Martinevans123 (talk) 21:45, 15 February 2016 (UTC)
That will never work. To respond to be Martinevans123 and The joy of all things, this idea is untenable for numerous reasons:
  1. We have real work to do here, and it's not agonizing for days or months over the case of a single character in every over song title. No one is going to research this kind of stuff. We have a general approach to capitalization because it's practical to have one.
  2. How it's printed may vary from issue to issue (even simultaneous releases in different markets)
  3. How it's printed can even vary in the same issue (spine vs. liner notes, etc.)
  4. Except for self-releases, which are rarely notable, "the release from the group/artists" does not actually exist. The releases are from the label, who often do not do what the artist wanted in the first place. It's WP:OR and WP:POV to insist that what is printed on the cover or liner represents some pure artistic intent. If you think that a particular pop star even cares about the difference between "from" and "From" on an album title, that's more OR, short of a RS quoting them saying they care about it.
  5. Cover style is often either excessive ("GET WITH IT!!!") or intentionally ultraminimal ("one morning"); these are stylistic visual effects for marketing purposes, not names. This stylizations, however, are frequently also found in artist websites, sheet music, and in liner notes, so "the use of unconventional styling" concern does not magically go away just because cover art is excluded.
  6. We have a long-standing guideline at MOS:TM saying not to try to emulate logos or other trademark stylization; this is not just a MOS:CAPS matter.
  7. The vast majority of people with any writing and serious reading experience do not accept the idea that writers for professional-quality publications are hostages to the style whims of their self-promoting coverage subjects.
  8. Making an exception for pop music will necessarily and immediately lead to demands for exceptions of all sorts, from every other topical "camp" on Wikipedia, and we may as well have no manual of style at all if every topic gets to make up it's own random "rules" for how to write English in "their" articles. We have policies (like the one just linked to, and WP:LOCALCONSENSUS) against that for good reasons.
  9. The vocal camp who insist on trying to push content naming conventions as style matters, too (WP:COMMONNAME) are never going to accept something that amounts to a "WP:OFFICIALNAME is rewritten, and the official name is now the only acceptable title" argument. (Never mind that the official title, as noted above, often cannot reliable be determined anyway, due to difference between releases, between what's on the release and on the website, etc., etc.).
  10. See WP:SSF for why the entire motivation to do this is not appropriate here. WP exists to consistently present digestible, verifiable information to the most general audience of all time. It does not exist as a platform for pushing specialists' style peccadilloes that make sense to them in their topical circles but cause problems for others outside of them. MOS's under-five-characters rule is very well attested in sources for the kind of writing we do, and it's far less restrictive than the "never capitalize any prepositions" rules of some of them. What we have is workable, and consensus does not require unanimity. It is entirely sufficient that ours is the lowest-conflict route of the three available that have support in sources on how English works.
  11. "Follow the sources" necessarily entails following the sources that actually matter for WP's purposes. The marketing desires of bands and labels are not a reliable source for how to write English in an encyclopedia. Even journalistic style preferences are not. Formal and general English style guides are. The applicability of WP:ABOUTSELF is very, very limited, and it does not in any way dictate to WP how it has to style titles of works by that "self".
PS: The point about deleting characters from the name of a Marillion title is a non sequitur; it doesn't relate in any way to how WP capitalizes.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  02:05, 16 February 2016 (UTC)
Whether or not L is a capital letter is "just a MOS:CAPS matter". Yes, WP:ABOUTSELF is about "self-published and questionable sources", so how is it relevant here? And if you don't want to "agonize for days or months about a single character", feel free to remain silent here, except it's not really "every other song title" is it? Martinevans123 (talk) 08:17, 16 February 2016 (UTC)
To respond to SMcCandlish - I can see your point, but Wikipedia is supposed to NPOV - if we change things, no matter how small where are we? If should be as directed by the artist, otherwise we start retitling things that someone has created and they will be upset. Authors, modern art sculpture type people and other creative types will throw their teddies out of their respective cots if we unilaterally decide to change the way something is written on a piece of art (that's how they see what they create). I do see your point about future releases being different etc, but we should not interfere with the title's of things, because that invalidates our neutrality. What next? 08:37, 16 February 2016 (UTC)
We change all sorts of thing, otherwise we'd have to delete WP:COMMONNAME policy. Where are we? In the same boat as every other publisher. It's not WP's job to make people happy about the naming decisions, it is to be a helpful encyclopedia, that people can edit without having to split hairs with people who will argue for years about some style point no one else cares about, including the artist (show me any case of a newspaper or other publisher being hounded by an angry artist or publisher for not capitalizing "correctly"). We don't follow the "official" names of anything when it conflict with the project's priorities (WP:OFFICIALNAME). And this is just how the name is styled. All that said, we already have a tendency to support a particular stylization (if it's just a case change or letter substitution) when there are WP:RS indicating that the stylization was intended by the artist (hell, we even accept Deadmau5, k.d. lang and some non-artist, purely commercial output like iPod). You're concerned about a problem that has not manifested in 15 years of WP being around. "Five people won't stop arguing about song titles" does not equate to "artists are angry with Wikipedia all over the world". :-)  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  09:33, 16 February 2016 (UTC)

A heads up that this overwrought debate is attracting off-wiki attention. Why don't you just put RULE OF REASON vs. RIGID RULE 1 vs. RIGID RULE 2 up for community consensus via a publicized RFC and then we can agree by a large margin that capitalization of "Like" should happen on a case-by-case basis determined by common use, with common use generally capitalizing the word. I'll start fishing for a couple trout, they will come in handy after the vote... Carrite (talk) 17:10, 16 February 2016 (UTC)

  • Get rid of MOS rule, leave spelling open. This is a typical case of MOS over-reach. Where two typographic or spelling variants are in competition, the Wikipedia MOS should never prescribe one over the other except in the following situations: (a) where one variant is clearly and universally agreed to be the correct one, as manifested in near-universal practice in carefully edited prose and/or agreement of multiple reliable style guides; (b) where a fundamental, highly visible and salient aspect of the format of Wikipedia pages is affected, such that consistency is necessary to maintain the visual "face" of the project; (c) where consistency across Wikipedia is necessary for technical reasons such as accessibility/editability. None of this is the case here. Beyond what is needed according to these criteria, Wikipedia should never impose an arbitrary "house style" but accept and embrace variability as a matter of principle. Fut.Perf. ☼ 18:06, 18 February 2016 (UTC)
How is MoS adopting the same rule favored by other mainstream publishers "over-reach"? Can't have it both ways or please everyone all the time. People either complain that MoS doesn't follow sourceable rules (which is almost always a false claim, and has been rectified in the rare cases it was not; I doubt there are any instances of this left), or they complain when MoS does follow sourceable rules and it's not their pet one. How could it possibly serve WP's interests to do what you suggest, and abandon every MoS line-item that doesn't rise to the level of your made-up criteria (though they might be good for determining what to move into a style policy instead of guideline...)? All it would do is lead to never-ending strife. Every single title of an article here about a published work with a preposition in it, even a tiny short one like "of", would be available for new WP:RM warring because someone can find a source somewhere that prefers to capitalize all words in work titles (I actually cited one below, already), and by your reasoning there is no call to prefer one RS or collection of RS over any other, only a binary choice between unanimity or utter chaos. This means most such debates would be necessarily be inconclusive and recurrent, and there would be many thousands of them. Preventing that kind of anger-increasing and time-sucking WP:LAME disputation is why MOS exists, why it's enjoyed broad consensus, and why the majority of it has remained so stable for so long. If we tore down every rule that WP has on the basis of half-a-dozen people not getting what they want, or some subset of the rest of the world not doing things the same way WP does, then WP would and could have no rules of any kind. MOS already has WP:ENGVAR to account for actual dialectal variation. This is not a dialectal variation; it is a difference between informal, formal, and ultra-formal writing styles cutting across all dialects (as I've reliably sourced), and WP uses the formal one, throughout. It's not even a matter of this minor capitalization thing, it's our general approach to all writing. PS: Titles of articles about compositions having a consistent, predictable, and not made-up-out-of-nowhere approach to handling prepositions already qualifies under your point (b), anyway.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  00:31, 21 February 2016 (UTC)

Style guide sources on this matter

This is a vexing matter, and policies vary. ...

Pick a policy and be consistent.

— Jane Straus,
The Blue Book of Grammar and Punctuation
(11th ed., 2007, Wiley)

In reality (I've been doing a sourcing run on this for our article on capitalization in English, which needs a lot if work), there are three competing would-be-standards for title capitalization in English, using mixed case:

  1. Never capitalize prepositions, even long or unusual ones [except if first or last word in title or after a colon] (Chicago Manual of Style, MLA Style Manual and Guide to Scholarly Publishing and its derived MLA Handbook for Writers of Research Papers, MHRA Style Guide, Canadian Writer's Reference, United Nations Editorial Manual Online, Berghahn US Style Guide [major journal publisher; it's also in their UK version], and even some music-specific sources like Writing About Music: A Style Sheet [3rd ed, 2014]).
  2. Capitalize all of them that are four letters or longer (AP Stylebook and, with variances, The New York Times Manual of Style and Usage, along with some other journalism sources, such as the style guides of National Geographic and The Wall Street Journal in the US, and, with a deviation, Guardian and Observer style guide, in the UK; only two non-journo sources so far, both of them obsolete: The Canadian Style: A Guide to Writing and Editing of 1997, and an decade-old Gregg Reference Manual – I have the current ed. on order).
  3. The broadly supported middle ground, capitalize all of them five letters or longer (numerous sources, though few have been mentioned in the debate above – they're general-audience style guides, neither journo-speak nor ivory tower, including at least Garner's Modern American Usage, Oxford Dictionary of American Usage and Style [closely related to Garner's but more prescriptive], McGraw-Hill Handbook, New York Public Library Writer's Guide, and Webster's New World English Grammar Handbook; also in this camp are some more specialized guides, e.g. The Redbook: A Manual on Legal Style).

There are/were also two styles not relevant here:

  • It is common in certain disciplines (for all titles or only some kinds of works) and across many of them (for subtitles, headlines, chapters, sections) to use sentence case: "Annual report on the incidence of flying monkeys". This style predominates in non-English usage generally, though French has odd rules we need not get into.
  • The old style of capitalizing literally every word ("Poultry And The Fowl Of The Field, Being A Treatise On Birds") is entirely obsolete, jarring, and difficult to read; a few style guides comment on its [former] existence, but zero have recommended it for generations, as far as I can determine. Today, it can only ever be seen in reputable print as a cover and title-page stylization for typographic effect, on a par with using other font tricks like changing font mid-word, or overlapping words.

MOS uses the third, generalist, compromise approach among the three relevant options, and has been stable on this for a long time. The WP:GREATWRONGS demands to make an exception for "Like" are the actual original research at work here; there is no reliable source of any kind for such a notion. The attempt to use Google to "prove" that "Do It Like A Dude", etc., is the "most common name" is a confusion of names and how to style them (and also more original research - it's a methodologically flawed personal attempt to build a corpus using weak tools and then engage in a novel analysis and synthesis of that data; even if it weren't OR, it would still do nothing but show us what the journalism approach is, which we already know from journalism style guides like AP Stylebook, and about which WP does not care much, per WP:NOT#NEWS.

A reasoned argument could potentially be made to switch to the four-letters-or-more approach of journalism, but no cogent arugment has presented, just lots of fist-shaking, finger-pointing, and hand-waving. This would be no shoo-in at all: It would mean capitalizing "With" and "From", and I'm dead certain more editors and readers would revolt at seeing that that all over the place, in literally millions of instances on our pages, than will ever object to a handful of cases of "Do It like a Dude" (where a few simply don't understand that it's a preposition there, and not intrinsically special).

While, in spirit, I'm sympathetic to the view that there could be some rule like "capitalize common prepositions of five characters or longer, and any word being used as a preposition that is commonly not used as one", there is no real-world, RS support for such a rule (that I have found yet, and I have more style guides than God); it would be precisely the kind of "making up bullshit rules" that people wrongly accuse MoS of all the time (simply because the rules we have don't always match what they learned in school). Even journalism does not employ such a rule; the news sources that use "Like" in titles also use "With" and "From"; the handful of exceptions I've seen were from blogs (WP:SPS), wikis and forums (WP:UGC), and music-industry specialist publications mimicking the style of products from their advertiser base and pandering to a fan base (WP:INDY, MOS:TM, WP:JARGON).

Anyway, I'm not done with the sourcing run yet (if you want page numbers and quotes, you'll have to wait until I post it all to article talk; it takes a long time to construct and organize the citations). I have a stack of a dozen or so more paper ones to go through, then directories full of electronic ones. So far, I can only find two non-journo style guides written for the public (i.e., not private house style) that support the AP four-letter view without reservation, and they're both old. Many style guides do not address the question, either using vague wording ("short", "important", "usually" – New Hart's Rules is in this camp, along with European Commission English Style Guide, U.S. Government Printing Office Style Manual, et al.), or never mentioning the matter (Elements of International English, Webster's Dictionary of English Usage, Writing for Scholarly Journals, etc.). The Copyeditor's Handbook is divided between the never rule and five-letters rule, while never even mentioning any four-letters rule. Even some journalism sources do not cover it, including (so far as I can find in them) The BBC News Styleguide (professional version from BBC Training & Development), BBC News style guide (student version at BBC Academy website), The Economist Style Guide (online and print editions), Telegraph Style Book (online), and The Times Style and Usage Guide (online until 2015). The Blue Book of Grammar and Punctuation is strictly neutral (it's a thin, summary volume). In noting the three styles (which it delineates in detail, as do Garner's and some others), it observes: "This is a vexing matter, and policies vary. ... Pick a policy and be consistent." Best advice I've encountered on this or any other style issue to date. The Redbook echoes this in two places, too.

Wikipedia has picked a rule, and we've been stuck with it for years – except when people present OR-based and anti-policy arguments at RM, and inexperienced admins don't notice, and make incorrect closes. Consensus can change but it doesn't do so because a handful of people are angry and don't understand. If there's an understanding issue, we just need to concisely explain there are three potential rules, two of which are extreme, and WP uses the middle one, because it's the one that is used by and useful for general-audience works, and add that to MOS:FAQ.
 — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  09:33, 16 February 2016 (UTC)

Update: I found one academic style guide that prefers the four-letter rule, the Publication Manual of the American Psychological Association. On the other hand, the UPI Stylebook, a journalism guide, has jumped ship to the five-letter rule, so it evens out.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  11:10, 25 February 2016 (UTC)

I’m not angry, just disappointed. Yes, general style guides are relevant. This thread is about the word "like", remember? Why exactly would any change have to "mean capitalizing "With" and "From""?Martinevans123 (talk) 09:46, 16 February 2016 (UTC)
  • It seems undesirable to have style guidance that would differ based on which specific word is being used (and perhaps also differing for how the word is being used) rather than based on categories of words such as parts of speech (as in the current guidance). Personally, I think the survey of style guidance used elsewhere is very helpful. —BarrelProof (talk) 19:28, 16 February 2016 (UTC)
I agree it's all very useful stuff. But they are all just "guides", including our own. Your desire for a guide based rigorously and solely on parts of speech, sounds much easier than one with exceptions, but is I feel overly prescriptive and very likely to continue to provoke the sort of dissatisfaction expressed so robustly in this current discussion. Martinevans123 (talk) 19:38, 16 February 2016 (UTC) p.s. I had assumed this entire discussion was just about "titles" as in titles of Wikipedia articles. Or is all this style guide straight-jacketing also meant to apply to the titles of songs, films and books etc within articles? I'm not sure everyone is clear on this.
Certainly the capitalization guidelines apply just as much to the content within articles as they do to article titles. We are having a discussion about MOS:CAPS, not MOS:AT. —BarrelProof (talk) 21:42, 16 February 2016 (UTC)
Goodness me. So I really was am confused. And poor User:SMcCandlish was concerned about me "projecting hurt-feelings sarcasm" at him for "insulting myself". But thanks for the smiley face. I may try and return when I have a better idea of what's being debated here, Like. Martinevans123 (talk) 22:11, 16 February 2016 (UTC)
Touché; I shouldn't have been sarcastic in complaining about sarcasm.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  23:52, 16 February 2016 (UTC)
Already addressed this and don't want to put up another list of bullet points. To try to get to heart of the matter in prose: There is no reliably published source that anyone has turned up that suggests singling out the word "like" for special treatment when it's used as a preposition. For MoS to just randomly make one up and impose it would be precisely the kind of prescriptiveness, unsourceable whim, and PoV-pushing that people sometimes falsely accuse MoS of when trying to get their own prescriptive, unsourceable viewpoint enshrined in MoS so that others will have to write the way that editor likes. There is no special "Like" style; this is false pattern recognition. There is, rather, a style to capitalize all four-letter or longer propositions, "like" among them, and it is particular to journalism, which WP is not. It would be potentially possible for WP to adopt the journalism rule instead of the mainstream five-letter rule, but plenty of people would prefer we had the never-capitalize-prepositions academic rule, so this present compromise position – the one supported by most general English-usage sources – is stable and likely to remain that way. "Just capitalize exactly how the artist did" isn't workable for reasons already outlined, most importantly because it leads to ratholes of original research on the part of people who really have better things to do here than fight over a one-character nit-pick. Style rules are general because it's the only way they're practical; otherwise every case is always "special" and has to be researched, for style trivia no one really cares about.

The dissatisfaction expressed in the above debate has not been "robust" in any meaningful sense, just noisy. A robust argument would present a cogent, thoroughly reasoned, and fact-based position, and this has not happened. This failure to recognize that a difference observed between a subset of external sources versus WP's (and various other publishers') style on something is part of a systemic difference – distinct schemata with many rules, not one particular rule about one word/name – is the primary generator of unclear attempts to get WP to make case-specific "exceptions" based on what turns up in Google searches. It's not unique to this song-titles dispute. E.g., the common belief that it's "wrong" to start a sentence with "IPods are..." is not supportable with any mainstream style and writing sources, in any writing style/market; they all say the opposite when they address this question, and say to upper-case the first letter, trademark be damned. The Wikipedian misbelief is based on observation of how willing bloggers are to violate basic English rules like "begin a sentence with a capital letter" before they'll "violate" the "sanctity" of some brand name stylization.
PS: I'm tempted to go write a song called "Like, I Like A like I Like a Like, But U & I Alike Like B like U Like a Lick (Alight like a Lack of a Light and Alike Luck Lock All Alight, Alright?)", and then feign indignance if anyone doesn't capitalize it exactly that way. Even if I were as famous as Bono, I would expect people to tell me to go soak my head about my artistic intent!
 — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  23:52, 16 February 2016 (UTC)

You're intending to change (and in many cases change back) all examples of upper case preposition Like, when used in a title, at all locations at Wikipedia, to lower case l, because that's what the MoS says? Martinevans123 (talk) 21:18, 17 February 2016 (UTC)
More generally, many wikignomes intend, and invest time and effort in, fixing everything in WP that doesn't agree with policies and/or guidelines. This is not like special. Dicklyon (talk) 21:27, 17 February 2016 (UTC)
Surely that would be a job for a bot, with community sanction first? I don't see how such a proposal would ever gain consensus. I think there are many editors who are happy to see MoS rules applied more rigorously to article titles than to titles across the whole of article space. You'd be happy to see every title here changed to lower case l? Martinevans123 (talk) 21:34, 17 February 2016 (UTC)
Some such things are done by bots, but many are not. It's hard to make bots smart enough to be correct enough to trust them with styling decisions that depend on parsing the English. Dicklyon (talk) 22:19, 17 February 2016 (UTC)
Yes, bots can't make judgements about how things "look" (can they?) It's a bit subjective - to me it looks like a deliberate mistake, a bit like those "Bak 2 Skool" supermarket posters with some letters written backwards to look like "kid's spelling". But you didn't answer my subjective question - would you be happy? Martinevans123 (talk) 22:42, 17 February 2016 (UTC)
Probably can't be done by bot, per WP:BFDB. If you were asking do I personally intend to launch a bunch of RMs about this – well, no. My only interest in this has been the level of linguistic error, and of fact-denial about differing real-world style systems, that have been displayed in this and related discussions, and the threat that activism on such faulty bases poses to title and style stability across the encyclopedia. I really don't care if it takes a month or 5 years to clean up the "like"-related song titles. What's important here is that editors understand that MoS has a system, it is not a system MoS editors made up (it's called down style, and you can find it described in Garner's, for example), it is the system best supported in mainstream style guides, and it is a happy medium between two other systems that are extremes that would never gain consensus. This system exists to prevent dispute; disputes like the one above generally only arise when people deny that the system is real and try to impose their own pseudo-system based on personal or inside-group preferences, and misunderstanding of what they're seeing, like imagining that the press is treating "like" as some kind of special exception. They're not.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  09:27, 19 February 2016 (UTC)
Thanks for clarifying. I still think it is "a system MoS editors made up", it's not a set of facts. But, as long as it accurately reflects commonly-accepted style guides in use today in English-speaking domains, it may be perfectly valid. You seem to be claiming that the reason it "looks wrong" to many people is because of ignorance. But I'm not sure about that. Martinevans123 (talk) 10:13, 19 February 2016 (UTC)
I've sourced in considerable detail how it's a set of facts, so, QED. I didn't say it looks wrong to some people because of ignorance. Things look wrong to people because of their own subjective expectations, which are formed from wide-ranging and highly variable influences; the fact that they're idiosyncratic is the very reason we need MoS and that it will sometimes (like all style guides) set an arbitrary rule just to have a rule that stops the fight. The ignorance is in insisting that "like" is being capitalized in some journo sources because it's a special exception; this is a misperception based on failure to do the homework to find out that it's part of a very well-documented journalistic capitalization system that applies to four-letter-or-longer prepositions generally, and which is not the other, also well-documented, system that WP uses.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  22:00, 19 February 2016 (UTC)
Sorry, I just don't see the four in "four-letter-or-longer prepositions generally" as the same sort of fact as the four in Beryllium, or even the four in the Four Horsemen of the Apocalypse. Martinevans123 (talk) 23:28, 19 February 2016 (UTC)
Ah, I see what you mean. Well, sure, it's not. It's just a system, for dealing with a pretty trivial matter. The other systems did not appeal here because the one extreme means not capitalizing "toward" or "amongst" in a title, and the other means capitalizing "With" or "From" in one. If MoS made up some new WP-only standard, like "capitalize prepositions of four letters or more, except 'with' and 'from'" or "lowercase for four or shorter except uncommon prepositions including 'like'" we'd never hear the end of complaints about it, and general condemnation of MoS as a bunch of made-up nonsense with no basis in real-world writing standards. Our happy medium isn't happy for everyone all the time, but it's workable. The nature of compromise.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  03:51, 20 February 2016 (UTC)
You're saying there are just two systems of capitalisation and we simply have to choose between them, lock, stock and barrel, with no "exceptions" allowed. The fact that the title "Someone Like You" is used all over the place, including by the artists themselves (or at least as near as we can get to them through their commercial product), is irrelevant. In fact we have to deliberately ignore these because they are part of the "wrong" system? Martinevans123 (talk) 09:38, 20 February 2016 (UTC)
I said nothing of the sort; as I noted below, I don't feel like you're paying attention to anything being said here unless is supports your overcapitalization goals. I've documented in detail that there are three well-sourced capitalization systems, a journalistic one (that many of its adherents randomly modify for house style), the high academic style against all preposition capitalization, and the five-character rule used in most other publishing, the middle-ground compromise approach WP uses. I've also already addressed this "artists themselves" plea: What the label does (which, very often is ALL CAPS) on a release cannot be tied, except in imagined original research to artistic intent (absent, say, a reputably published interview with the artist stating something about it), and we wouldn't necessarily care anyway, per WP:OFFICIALNAME. We almost certainly would care if it were a double entendre in which a word like "like" was being used both as a preposition and as something else, e.g. a verb, in which case we'd capitalize. I.e., there is no dispute about that point, so it's a red herring.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  23:42, 20 February 2016 (UTC)

Please don't accuse me, yet again, of "not paying attention". And please don't characterise me as having "overcapitalization goals". My interest here was prompted purely by wanting upper case L in the song title "Someone Like You". I now see there are likely to be thousands or tens of thousands of similar instances throughout the encyclopedia. You tell us "that there are three well-sourced capitalization systems". Can you provide a WP:RS source for that claim? Or explain to us why there are not four, or five, or whatever number of well-sourced capitalization systems? And when did I claim there is any possibility of double entendre for that song title? And I've never suggested using label or album cover titles, which are very frequently block capitals. The contradiction between article titles and article content is immense, and even extends, of course, to first letters. Martinevans123 (talk) 11:59, 21 February 2016 (UTC)

"Can you provide a WP:RS source for that claim?" Please proceed to top of this subjection, at #Style guide sources on this matter and start over. Neither I nor anyone else has any need to explain why there are not five well-sourced capitalization systems; it's self-explaining: The sources show three of them, and outline them in detail. See The Blue Book of Grammar and punctuation and Garner's Modern American Usage for two sources that describe the three systems in series. Even if there were 20 different capitalization systems, it wouldn't matter. WP still follows the best-supported one, absent some important WP-specific reason not to, which no one has advanced. I didn't say you personally were making an argument about double entendres; rather, I was providing an example of how fears that the rule would be applied in anti-WP:COMMONSENSE ways are unfounded, and addressing previous respondents' claims that such a particular case would be an issue, and their claims that WP should follow the style on the cover (I'm glad that you also realize that suggestion isn't workable, for the same reason I pointed out: They're often just given in all-caps, which has nothing to do with "artistic intent", it's just marketing noise).  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  00:02, 22 February 2016 (UTC)
  • The further away we can get from the execrable NYT headline mindset Of Capping Everything In Sight, the better. Tony (talk) 07:18, 17 February 2016 (UTC)
Tony, I respect the fact that you dislike that so many reliable sources "over-capitalize"... But Wikipedia is not the place to right great wrongs. Blueboar (talk) 14:12, 20 February 2016 (UTC)
Or even write great rongs. Martinevans123 (talk) 14:39, 20 February 2016 (UTC)
On the flip side... to those who argue that we should capitalize because "that's what the artist intended"... no. That is essentially an official name argument. I am completely in favor of adopting a COMMONSTYLE approach to these works... but if we do so we will also have to adopt an OFFICALSTYLE caveat.
As I see it, it does not matter how a fairly small group of Wikipedians working on an MOS page think a work should be capitalized... nor does it matter how the artist thinks the work should be capitalized ... What should matter is how reliable sources that talk about the work routinely capitalize it (in the real world). Yes... adopting a COMMONSTYLE approach will give inconsistent end results (with words like "like" being capitalized in some cases, but not in others)... but that is OK. Adopting a COMMONSTYLE approach would give us a consistent "rule" to follow... one that would better mesh MOSCAPS with WP:AT. Blueboar (talk) 15:04, 20 February 2016 (UTC)
However, "how reliable sources that talk about the work routinely capitalize it" is a soft rule: While this may not be true for all titles, I think there are many titles for which sources can be found to support either capitalization. Then each side will try to outnumber the sources provided by the other side, or discredit the other sources (not reliable, not independent etc.), and there will still be endless discussions. Darkday (talk) 15:34, 20 February 2016 (UTC)
Yes, I'm all for the real world. I don't believe there is one single "journalistic style" that has to be eschewed at all costs. But I can see what Darkday says. This leaves us with arbitrary rules which exist, not to "decide" anything, but merely for the convenience of "stopping endless discussions." The list of examples presented by LlywelynII (~170 vs 16) suggests that, for whatever reason, capitalised prepositional Like seems to be the preferred style in song titles. Why can't this be accepted as a "wiki consensus exception"? Prescriptive arguments from external guidelines are informative, but why do they have to be binding here? 15:53, 20 February 2016 (UTC) Martinevans123 (talk)
It's not the preferred style in song titles. It's the preferred style in journalistic writing for titles of all works. The preferred style for titles in formal, general-purpose prose is to not capitalize four-letter or shorter prepositions. The preferred style in exclusively academic writing is to never capitalize prepositions at all. It has nothing to do with song titles in particular and nothing to do with "like" in particular.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  22:07, 20 February 2016 (UTC)
I'm saying that, based on the evidence, it's currently preferred at Wikipedia. I still agree with what Blueboar has said above. I quite agree that "The preferred style in exclusively academic writing ... has nothing to do with song titles in particular and nothing to do with "like" in particular." That's why it's less relevant. I don't see this encyclopedia as "exclusively academic". But I think you do. Martinevans123 (talk) 22:16, 20 February 2016 (UTC)
To the contrary. If we were taking a purely academic approach, we'd use the academic standard for this, which is never capitalize any preposition, even long ones like "throughout" and rare ones like "whither". This has already been explained half a dozen times, so please forgive me for feeling that you are not paying much attention to anything except that which you think will get you closer to capitalizing things your personal way. We have a system. It's the most standardized, most source-supported system outside of purely academic writing. There is no consensus to abandon it for another system, nor for no-system-at-all chaos. What else is there to cover? I'll rebut BlueBoar's position (for the second time – it's already been addressed at WT:AT and just moved here as it it wasn't addressed there already), in a moment.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  23:34, 20 February 2016 (UTC)
I've paid plenty of attention, thanks. I certainly won't forgive you in the slightest. Where is your WP:RS for "three systems of stylization"? That's you own personal invention. So it's complete WP:OR. Thanks. Martinevans123 (talk) 23:52, 20 February 2016 (UTC)
You're replying to a thread full of sources about three systems of stylization with a demand for sources about three different systems of stylization, and wrongly accusing someone of doing OR and making stuff up, after they've already listed the sources that say what they claim they say. So, yes, I do think it's accurate to say that you are not paying attention to the thread. It's not a matter of trying to be insulting, it's just observable fact.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  00:02, 22 February 2016 (UTC)

Music-specific style guide sources on this matter

Summary of sourcing thus far on this question in music-specific style guides (full cite details I'm still compiling; will post those with the rest at the talk page of Capitalization in English when I'm done with them).

All of the following either explicitly state a rule to not capitalize prepositions (at all) in the titles of musical works [except when first or last word in a title or subtitle], or defer to Chicago/Turabian, MHRA, and/or MLA, all of which have the same rule:

  • Writing About Music: A Style Sheet, 3rd ed. (2014); D. Kern Holoman
  • "Journal of Music Theory Style Guide" (April 2014); Duke U. Press
  • "SIUC School of Music Guide to Some Music-Specific Issues in Writing" (January 2009; Southern Illinois U. – Carbondale)
  • "Style Sheet for Sacred Music" [journal] (13 May, 2008, Church Music Association of America
  • "JSCM Style Sheet" (Bruce Gustafson, 8 November 2012, Journal of Seventeenth-Century Music)
  • "American Music Style Sheet" (27 July 2006, U. of Illinois Pr.)
  • "Cataloging Sheet Music: Guidelines for Use with AACR2 and the MARC Format" (Lois Schultz, Sarah Jean Shaw, Working Group Sheet Music Cataloging Guidelines; 2003; Music Library Association / Scarecrow Press)
  • "Notes Style Sheet (8 January 2016; ed. Deborah Campana; Music Library Association)
  • "Citing Music Sources in Your Essay and Bibliography" (Lisa Rae Philpott, 20 October 2010; London, Ontario: The University of Western Ontario; no rule stated or guide deferred to, but illustrates lowercase four-letter prepositions)
  • Music Theory Spectrum, "Instructions to Authors" (2016, Oxford Journals) – surprisingly defers to Chicago not Oxford/Hart's.
  • "A Guide to Completing the Music Education Thesis" (Frank Abrahams, Patrick K. Schmidt; June 2007; Lawrence, New Jersey: Rider University)
  • "A Guide for the Preparation of Music Papers" (eds. J. Kent Williams, Elizabeth L. Keathley; 2011; Greensboro: University of North Carolina)
  • "Writing for Music" (2011; University of North Carolina – Wilmington)
  • "Style: A Brief Guide for Music Students" (2010; School of Music, Trinity College Dublin)

I could only find one that seemed to be in firmly in support of the four-letter rule

However, it a) contradicts itself directly with regard to how to do titles (p. 10, p. 20), and b) it's only a would-be standard for metadata in digital music systems like iTunes, anyway, and has nothing to do with regular prose.

Three were neutral on the matter (by virtue of citing Chicago and APA at the same time, when APA, along among academic style guides, does not use the academic style)

  • Music Research: A Handbook (Laurie Sampsel; 2013): Cites Chicago and MLA (academic style) and APA (four-letter rule).
  • "A Guide to Completing the Music Education Thesis" (Frank Abrahams, Patrick K. Schmidt; June 2007; Lawrence, New Jersey: Rider University): Defers to Chicago/Turabian, or Form and Style (covers the same plus MLA), but also APA, depending on content/focus.
  • A Handbook for Preparing Graduate Papers in Music (J. David Boyle, Richard K. Fiese, Nancy Zavac; Halcyon Press Ltd.; 2nd ed., 2004); cites both Turabian/Chicago but also APA.

Two did not address the issue at all:

  • A Guide to Library Research in Music (Pauline Shaw Bayne; 2008; Scarecrow Press)
  • New Zealand School of Music Composition & Orchestration Style Guide (Michael Norris; February 2013)

Some paper publications I don't have immediate access to, several of which are obsolete anyway:

  • How to Write About Music (Marc Woodworth and Ally-Jane Grossan; 2015; Bloomsbury Academic Press)
  • A Style and Usage Guide to Writing About Music, Thomas Donahue, Scarecrow Press, 2014 (2010))
  • A Short Guide to Writing about Music (Jonathan D. Bellman, 2007
  • How to Write About Music: The RILM Manual of Style (ed. James R. Cowdery; 2nd ed., 2006; Répertoire Internationale de Littérature Musicale)
  • Irvine's Writing About Music (Demar B. Irvine; 3rd ed; 2003; ed. Mark A. Radice; Amadeus Press)
  • Writing About Music: An Introductory Guide (Richard Wingell; 3rd ed., 2002; Prentice-Hall)
  • A Handbook for Preparing Graduate Papers in Music (J. David Boyle, Richard K. Fiese, Nancy Zarac; 2001)
  • Introduction to Research in Music (Robert Wingell and Silvia Herzog; 2000; Pearson)
  • Words and Music (Eugene Helm, Albert T. Luper; Joseph Boonin Press, 1971)

Given all the general style guides (see parent subsection of this sub-subsection) in favor of no-caps or the five-letter rule, and virtually zero support for the four-letter rule outside the AP Stylebook and its journalism clones (Gregg, a business-English i.e. marketing style guide being the only notable not-quite-journalism holdout, aside from one academic guide, APA, counterbalanced by one journo guide, UPI, favoring the five-letter rule), fans of capitalization in style titles should be really glad that WP uses this compromise style, or many thousands of song and album titles would be moved to have a whole lost more prepositions in lowercase. Speaking of cases, I rest mine for now.
 — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  10:55, 25 February 2016 (UTC)

And some "ground truth" from UGC

There's an undercurrent to this entire discussion, one that runs "WP is pissing off and alienating fans!". But it simply isn't true. Let's see what user-generated content and other "Interwebs" material is doing (this is just some ground truth data for internal discussion; obviously we couldn't use many of these as article sources). The blogosphere doesn't reflect any reality behind the proposal above that lowercasing of only three-letter-or-shorter prepositions is some kind of de facto standard for song titles, that WP is somehow failing to represent sourceable usage.

  • Standard Midi Files on the Net: "Prepositions are sometimes capitalized: ... Prepositions that are over four letters long. (across, after, among, beyond, ...)" [7] (external discussions, e.g. at Snopes [8], Yahoo! Answers [9], and Absolute Punk [10], cite that one like it's authoritative in their ambit).
  • SongCase - the song title capitalizer: "'Do It like a Dude' ... The word "like" should be capitalized if used as a verb." [11].
  • The similar tool Capitalize My Title produces the same result for "do it like a dude" input [12].
  • The do-it-yourself tool for coders at "Capitalization for Song Titles" at Pseudorandom agrees: "These are lower-case, unless they are the first word or last word: ... prepositions that are less than five letters long" [13].
  • English.StackExchange.com has lots on this (even a "How should I capitalize album titles and band names?" section); samples: "Do not capitalize ... short prepositions (as, ..., from)"; "In Titles: Do Not Capitalize: ... Prepositions (fewer than five letters)"; "This heavily depends on the style guide in use; they usually have a fairly exact specification. In a formal publication you should perhaps inquire what the recommended style is. ... Compare newspaper headlines from today: New York Times: 'Senate Votes to Confirm Elena Kagan for U.S. Supreme Court', Washington Post: 'Senate confirms Elena Kagan to Supreme Court'"; "It's all a matter of style and consistency"; "There is no universal standard"; "capitalization of a title depends on the medium"; "There is no hard rule on that, you need to refer to the style guide for your target audience (newspaper, academia, etc.)"; [14].
  • Your Dictionary (which seems to be professionally edited for the high-school student market; UGC is limited to a Disqus comments section I've ignored [actually, I don't even see it, as my Ad Block Plus nukes that stuff]: "The rules for capitalizing titles can vary according to a particular style guide, such as Associated Press Stylebook (AP), Chicago Manual of Style, and MLA style. They all have different rules for how to capitalize titles. ... [i]t all depends if a certain style is required by your teacher, course, or subject/field. ... General Rule: ... In Titles: Do Not Capitalize: Prepositions (fewer than five letters)" [15]; and, at same source, "Capitalization of Songs ... For prepositions, words that are four or less letters should not be capitalized." [16].
  • TuneCore, "How do I format my album title, track title and artist name for stores?" (advice page on formatting for iTunes releases): Applies the five-letter rule for prepositions [17].
  • "Capitalization: The Major Words in the Titles of Books, Articles, and Songs" at Grammarly: "Prepositions, articles, and conjunctions aren't capitalized (unless they’re the first letter of the first word)" [18].
  • Another title capitalization tool, TitleCap: Automatically Capitalize Your Title, at first seems to support "Like", but it's greyed out; when you mouse over it, it says "Like can be used in multiple ways. Do capitalize 'I Like You' (verb). Don't capitalize 'Run like the Wind' (preposition)". [19].
  • GrammarBook.com (not really a blog; this is the online supplement to Straus's The Blue Book of Grammar and Punctuation): "A major bone of contention is prepositions. The Associated Press Stylebook recommends capitalizing all prepositions of more than three letters (e.g., with, about, across). Other authorities advise lowercase until a preposition reaches five or more letters. Still others say not to capitalize any preposition, even big words like regarding or underneath."
  • Finally, one site that prefers the journalistic approach: MusicBrainz: "Capitalize all words except ... Short prepositions (three letters or less)" [20].
  • At English Forums, some random respondent named Michael West also gives this style among other advice about capitalization; the rest of the forum participants took apart everything he said, before the thread wandered off into Emily Dickson discussion. The take-away message was: "This is a matter of style, so you might want to consult a Style Manual." [sic], as one would expect [21].
  • The student-use house style sheets of various universities and colleges tend toward the academic style of all lowercase for all prepositions (e.g. "do not capitalize minor words like articles, and prepositions, and coordinating conjunctions ... with the only exception if one of these minor words come first or last in the title.", from the amazingly redundantly named University of Maryland University College [22]. I probably needn't list out a zillion of these, since we all know that universities lean toward academic style. (The university marketing department internal style sheets – i.e., how employees of the institution market the institution – are almost always based on AP style, since it's marketing. There's an almost perfects split between AP and Chicago/Turabian style when it comes to PR vs. academics and the same institutions. The message is pretty clear.
  • Discogs uses the 19th-century rule (probably so their software can automate it) to capitalize the first letter of all words [23].
  • A poll at MediaMonkey demonstrates that 36% of respondents also follow this pattern (and for the same reason: "it is impossible to automagically format every title using mixed capitalisation") [it is impossible to automagically format every title using mixed capitalisation].
  • IMDb isn't clear on the matter. At Submission Guide: Title Formats, "Capitalization and character sets" [24], they provide a list of words that "must" (in their house style) be lowercase, and it includes the four-letter prepositions with and from, but this is all they say. They suggest none of the following explicitly: that other four-letter prepositions should be lowercased; that all four-letter prepositions besides those two should be uppercased; than all other words must be uppercased aside from those on the "must lowercase" list; or even that any words at all must be uppercased other than the first and last. Their general rule at the top of this page is "We use the original title of a movie/show in its original language as it appears on screen (on the title card) in the opening credits." All variations must coded as "alternative titles". These instructions appear to be self-contradictory, since if I name my short from "That Guy From Oakland", their rules "do not compute". [Entry added to original post, 08:01, 25 February 2016 (UTC)]

General conclusion: There is no support in real-world usage among music fans for the "four-letter" rule of journalism. People either follow the five-letter rule, say to follow whichever of the three rules you like, or they just can't be bothered with any rule and capitalize every single word.

I have not cherry-picked these results; I just looked at the non-false-positive sites Google coughed up in the first 1–3 pages of results for various searches relating to capitalization and song titles (I obviously excluded things like "song titles that mention capital cities" that were returned when I used the keyword "capital"; I also excluded MusicBee at Wikia because is said it was using en.WP's MoS as a source, and excluded WMF results like WikiSource and WikiBooks).
 — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼ , 23:36, 20 February 2016‎ (UTC) [updated: 08:01, 25 February 2016 (UTC)]

So what do professional music publishers do, as opposed to "music fans"? Do contemporary "rules" apply equally to songs published many years ago? But thanks so much for quoting "some random respondent named Michael West" there. You characterise all of the above as user-generated content. But I suspect most users of say discogs don't really see how they have a means of influence on how titles are styled. They submit content, certainly, but they are not part of the process that decides the style of presentation on the web page. Martinevans123 (talk) 00:00, 21 February 2016 (UTC)
I'm not sure who "professional music publishers" refers to. We know that record labels and such do whatever their marketing department decides will look cool, and this is often ALL-CAPS, or Small Caps, or wEirD RanDOM !!!stuff that often cannot even be done with CSS 3 no matter how hard you tried [25]. Sheet music publishers seem to use the same cover art as the original release. I'm not sure what they do on the inside in plain text, but as Blueboar, I think, pointed out that would be a WP:OFFICIALNAME fallacy, anyway. We care about what sources that write about music do, not what its promoters do. The research I've done on that so far, digging up music-specific style guides (forthcoming in an another section posted at #Music-specific style guide sources on this matter) is almost entirely in favor of high-academic style, using capitalization for no prepositions at all unless they are the first or last word in a title or subtitle. Yeah, I thought the "random Michael West" part was pretty amusing.

Re: Discogs: I have no idea who determines their titling policies. The point of the above list isn't that the policies are UGC, it's that the sites are arguably unreliable as sources to use in a WP article, because they're UGC or primary. I've listed them here in addition to the RS, as discussion data points, demonstrating that the four-letter rule of marketing/journalism writing is not actually followed by music aficionados and the sites they build and use. I.e., the impression some have that it's the music fans and the music press against WP siding with a bunch of dusty academics is demonstrably false. The press is one extreme, the academics are the opposite extreme, and everyone else, including WP and fans, are in the middle where we be belong. (Or in the case of Discogs, doing their own technical thing without any regard to the style question at all.)  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  08:01, 25 February 2016 (UTC)

Arbitrary break III

Not really centralizable. It's a matter of whether the usage in question is conjunctive (capitalize), prepositional (lowercase), adverbial (capitalize), or questionable (lowercase by default), and that will be a case-by-case determination.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  09:04, 19 February 2016 (UTC)
  • Are you familiar with the conventions of the titles used in published music scores? I don't know if you count this as "journalism" or "academia", but I would have thought it would provide an excellent "real life" source on which to base a specific style guide for song titles. Martinevans123 (talk) 22:38, 20 February 2016 (UTC)
If you don't mind, I've refactored this into the thread about this, and out of the the merge discussion, where it was off topic. What convention are you citing? I've looked for one, and cannot find any such publication. Given that sheet music is controlled by copyright holder/publisher of the musical release in most cases, we can expect what appears on the cover of it to match what appears on the album for which it is sheet music. I haven't seen any evidence to the contrary so far. Note also that Writing About Music: A Style Sheet (already cited above) follows the high-academic style of Chicago (do not ever capitalize prepositions unless the first or last word in the title), which is really, really not what you want. The more I look into actual music-specific style guides, the more I see this "down style" being supported. I'll create a new sourcing section for that when I get around to it.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  00:44, 21 February 2016 (UTC)
Thank you. Martinevans123 (talk) 10:02, 21 February 2016 (UTC)
Heh. Sorry, I doubled my "smalls" instead of closing one. Fixed.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  13:38, 21 February 2016 (UTC)
"Follow the capitalization found on the sheet music" is essentially a flawed OFFICIALSTYLE argument (see my comments about that above). A true COMMONSTYLE capitalization guideline would say to follow reliable sources that talk about the work (Not unreliable fan sites, but reliable published music/entertainment magazines such as Rolling Stone). Now, in some cases it may be that there is no COMMONSTYLE capitalization (ie the sources are clearly mixed in their usage)... I would have no problem with MOS preferring non-capitalization as a "default" proposition in such cases. My concern is focused on the few titles where the clear majority of reliable sources do capitalize. Blueboar (talk) 01:48, 21 February 2016 (UTC)

While I agree that the "OFFICIALSTYLE" notion, as you called it, is flawed, there is no "COMMONSTYLE", and such an idea is just as unworkable, if not more so. How external publishers capitalize something is going to depend entirely upon a) which of the three general capitalization conventions for titles they generally follow, and b) whether their own house style injects any idiosyncratic rule changes. A "common style" approach is not tenable for numerous reasons:

  • We already have the closest thing to a "common style" provision that we'll ever need, in both MOS:CAPS and MOS:TM: If the independent RS, and not just in one field like journalism, almost universally style a particular proper name a particular way ("eBay" instead of "Ebay"), then so does WP. Observation clearly shows that journalism sources mostly but not entirely prefer to capitalize four-letter prepositions, and that other writing genres and registers do not, including most writing specifically about music (big source pile on that coming tomorrow; I'm almost done with it). There's a sharp distinction between the unusual case of reliable sources across the board (not just journalism ones, but reputable books, academic journals, etc.) overwhelmingly agreeing to go along with an unusual trademark stylization, e.g. Deadmau5 or iPhone or Left 4 Dead, and what you're proposing. These are almost always either a letter substitution (using normal English alphanumerics), or intercaps a.k.a. camelcase, for a particular effect in a specific case, and have nothing to do with conflicts between journalistic and formal style across a whole sector of English writing. Such cross-genre agreement is simply not true of treatment of prepositions in titles of works, and it never will be within our lifetimes.
Lots of others, including OR, NPOV, UNDUE, and RS problems with this idea:
  • It's still commingling the name with how different kinds of writers style it for their house rules, and pretending that the popularity of a particular topic with a particular subset of writers concretely dictates how WP must write – exactly like that subset does, no matter what any other concerns might be. It would mean (or be WP:GAMEd to mean) that WP would have no choice but to use ridiculous gee-whiz attempts to ape the contorted stylization of some product, any time some undefined number of RS in one particular market segment, that happened to write about it more than others, incidentally went along with the stylization for any reason, like "LaTeX", which can be found, exactly that way, in lots of books about typesetting systems, and Linux journals, etc., etc. Geeks pride themselves on being able to pull off that sort of thing with CSS. Well, no thanks. "LaTeX 2e" is perfectly adequate for encyclopedic writing, and it's even fine for us to note this stylization with a "styled as" comment in the article at LaTeX. Can do the same with song titles.
  • All of this is skirting the elephant-in-the-room problem that unusual stylization, especially excessive capitalization, is just attention-grabbing by use of typography for promotional, undue emphasis. There is no other explanation for things like "Do It Like A Dude". Is there anyone on the planet who thinks that capitalizing the indefinite article "a" in mid-title like that is correct English, in any form of writing other than title stylization for marketing purposes? It's fine if someone wants to do this on their CD cover. It is not fine to expect WP and everyone else to help you market it that way, even if much of the press sector – often with shared corporate ownership – will blithely join in that Hollywood circle-jerk.
  • Even for things like treatment of prepositions, it's going to vary randomly by topic even in the same general subject area, depending upon which sources happen to show up in search results. For pop music, the results will be skewed toward journalism and (more to the point) promotional marketing via stylization quirks like "Do It Like A Dude". For classical, it'll be skewed toward humanities academic writing (MLA and MHRA). For general nonfiction books, it will be a mixture of journalistic and academic writing, and some specialist (AMA, APA, CSE, etc.) style. And so on. And it will vary depending on which search engines you use, which what search constraints – Google? Google News? HighBeam?
  • It's going to vary across broader topics, depending on how academic-to-lowbrow they are, and whether money is changing hands. The more a subject has coverage in academic sources, the more sources are going to incidentally be using never-capitalize-any-prepositions academic style when referring to them (and, incidentally, virtually all actual style guides on music say to follow that convention, which I don't see anyone here favoring. The more a subject appeals to youth culture, the more it's going to be written in weird styles including in journalistic coverage pandering to that audience. The more a subject is tied to the entertainment industry, the more sources are going to ape the style of the publisher as closely as possible to keep their advertisers/owners happy and sending money. WP has the same consistent approach used in non-extreme style guides for a reason.
  • Did you know that a music industry trade group is pushing a draft standard that would, in essence, just do everything that the labels/studios demand? Even the regular press is balking at and mocking this (see, e.g. "Grammar Rocks: These New Punctuation Rules Are fo' Realz", Wall Street Journal, 10 June 2013 [subscription reqd.] [26]). It is absolutely not WP's job (per WP:NPOV, and WP:NOT in about 5 places) to help some external faction win a "style war". We should continue to do what mainstream style guides recommend. We have no WP-specific reason to prefer either the academic "no caps for any prepositions" extreme or the journalistic "capitalize virtually all prepositions" extreme, both of which would incite reader and editor rebellion. Our adoption of the well-documented middle ground on this has been stable for years. A desire to do what Rolling Stone does, just because, isn't a rationale to change that. I've looked over this Music Metadata Style Guide [27] from the Music Business Association. It's directly self-contradictory in multiple places, including title handling, and was clearly written by committee, without any sanity-checking. More on its nonsense later. It is not a reliable source, nor an actual standard in practice, it's just a WP:PRIMARY proposal that no one is buying into, from the evidence I've seen so far.
  • Back to the general topic: It would require subjective, methodologically worthless original research to even attempt to establish what the "common style" of something is, almost always based on Google searches, which we know are affected by paid placement. These searches would be conducted mostly by someone with an axe to grind, usually an overcapitalization mission; everyone else just WP:DGAFs and is content to follow our mainstream capitalization conventions. Because the result of this OR would always be a matter of whoever obsessed more about a particular title and was willing to waste hours building cherry-picked lists of usage, 99% of it from the exact same writing sphere, just to WP:WIN, it would do nothing useful – we'd still end up with some titles with a word capitalized that others did not, and people continuing to fight about it. The proposed "common style" solution is not a solution, just a recipe for constant conflict.
  • Journalists do not actually follow any particular quasi-convention closely at all; there's a house style regarding prepositions and other short words at the NYT that differs from that of The Guardian, in turn from that of AP Stylebook, and it from the UPI Stylebook, etc. It's very much like "British quotation" – there is no single system for us to adopt. So, no true "common style" is going to emerge with regard to the entire class of composition titles.
  • The "common style" notion gives WP:UNDUE weight to low-quality, inconsistent journalistic style, the purpose of which is expediency at all costs, including clarity or consistency. Journalism style conventions evolved for the sole purpose of producing something basically readable in the shortest typesetting time they could do it in. Today, modern typography mostly obviates such concerns, but the habits have become ingrained in that writing microcosm. Per WP:NOTPAPER as well as NOTNEWS, WP has no reason to allow an external camp to force a particular stylistic approach on our own writing, just because they like it for "tradition" reasons. We've already had this argument for 10 years about typesetters' quotation, and consensus never changes about it, despite long-term, focused activism to see it changed in favor of the "traditional" way. In both cases, there are multiple source-attested styles, we picked one that works best for our needs, 99.999999999% of WP editors are readers have no issue with it, yet a particular handful (notably overlapping) refuse to let the matter go, and rant about it for years. We have pages about this, like WP:TE.
  • It's the same WP:SSF stuff as usual: The mistaken belief that because some newspapers or music magazines may be reliable for facts about a band or an album, that they mystically transubstantiate into the most reliable sources for how the rest of the world must use English when addressing the topic, even in a totally different class of writing. WP:RS / WP:V (and just WP:COMMONSENSE) simply don't work that way.
  • We already had the "what if sources in one narrow field near-universally do it a particular way, but broader publishing circles don't, shouldn't we do it the specialist way?" discussion many times, and the answer has consistently been "not if it conflicts with mainstream usage". But it's a moot point. I've already built up a huge source list showing that, aside from the badly broken "Music Metadata Style Guide", music-related style guides don't follow the four-letters rule (the actually mostly follow the capitalize-no-prepositions rule. That is the actual specialist style. But that's just the opposite extreme from the capitalize-like-mad journo style, and WP has no reason to go along with either of them, when the well-sourced middle road has served us well. This is closely akin to WP:FRINGE debates: Just because something can commonly be seen in writing doesn't make it what WP should present to our readers. News writing it's fringey by nature of course, but the relationship of news style, like bureaucratese buzzwords, to formal, encyclopedic writing has strong parallels to that between pseudoscience or faith-based argument and rigorous scientific inquiry, in superficial similarity of outward form and the deep disjoints between them at the intent and methodological levels.
  • Following only the journo sources, and capitalizing "Like" willy-nilly, does not even match what people do anyway (see "Ground truth" subsection, above). The idea that what the press are doing to titles is what everyone is doing and expecting is provably false.
  • "My concern is focused on the few titles where the clear majority of reliable sources do capitalize" – This is the same flawed approach as accepting that astrology is real science because a majority of people believe in it. You're counting heads instead of distinguishable positions and the sources that support those positions as such. It doesn't matter how many journalism sources write "Do It Like A Dude". For purposes of any useful analysis they're all one source, one viewpoint being advanced about title capitalization, that of the AP Stylebook and its clones. All other usage is consistently against it (having either a 5-letter rule or a no-caps-for-preps rule), including virtually all music-specific style guides. This is part of why WP:COMMONNAME is a name policy, not a style policy. What the largest number of sources say the name is (regardless of style, which most people are never going to bother to memorize) directly affects people's ability to find an article here and be sure that they've arrived at the right one. Whether that name is capitalized by AP rules, Chicago rules, or the middle-ground (I guess we can call it Garner) rules does not. If we get "Do It like a Dude" wrong, as "Did It like a Dude" or "Do It like a Gal", or maybe even "Do It (Like a Dude)", we have a failure situation. There is no such screw-up if have "Do It like a Dude", "Do It Like a Dude" or "Do It Like A Dude", other than that the third one will make us look collectively illiterate to a large number of people; the other two are common and perfectly understandable approaches to work-title capitalization in the real world, that no reasonable person will mistake for a different title referring to some completely different work. (Note that WP:DIFFCAPS only works when there's an obvious semantic difference, immediately apparent to everyone, between the lowercase and capitalized title; it would never work to distinguish two different songs, one supposedly titled "Do It like a Dude" and the other said to be "Do It Like a Dude"; we would very obviously use parenthetic disambiguation.) At any rate, when it comes to a title where a majority of sources regardless of field/style/register/genre did capitalize, e.g. because RS indicated a reason for it, like a double entendre or because it's really an acronym, when you read the lyrics, then WP would also capitalize, per the already extant rules about this. No new rule of any kind is needed on this.

Anyway, I'm about 90% done with a sourcing run on music-specific style guides, but am too tired to complete it tonight (well, this morning). Will post the cites when I've had some coffee tomorrow. The productive thing coming out of all this is that the forthcoming section on title capitalization at our miserable stub article at Capitalization in English will be sourced beyond belief. It's shameful that we have almost no Featured Articles about our own language.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  13:38, 21 February 2016 (UTC)

SMC, You say: If the independent RS, and not just in one field like journalism, almost universally style a particular proper name a particular way ("eBay" instead of "Ebay"), then so does WP. This is exactly what I am calling for... following the capitalization in independent RS's... which includes (but is not limited to) journalism sources.
If there are non-journalism sources that discuss a particular work (the song, movie, book, work of art, etc.)... of course these sources should be factored into any COMMONSTYLE determination. However, we can not simply dismiss the journalism sources that discuss the work. These also need to be factored in. The key principles behind my stance on this are those of Recognizability and Naturalness (as outlined at WP:AT). To determine what is Recognizable and Natural, we need to examine all sources that discuss the topic (not just the ones that give us the outcome we personally prefer).
Where MOS and AT seem to conflict is that MOS calls for the source usage to be "almost universal" while AT sets its standard at "significant majority". When I call for a COMMONSTYLE, all I am really asking for is to change MOS guidance so that it is more in line with AT. Blueboar (talk) 14:14, 21 February 2016 (UTC)
If MoS is already doing exactly what you're calling for, then what is there to call for? We certainly can and always have dismissed the programmatic style decisions of publishers who follow a different stylistic program from ours, most especially if it directly conflicts with what is done in mainstream, influential style guides that are not limited to a specific writing niche (like journalism, or gaming, or one subset of zoology). This is not a new discussion, and it always resolves out the same way: We write for our readers, we do not pander to editorial whims in particular publishing camps. What you want to implement would result in WP having no consistent style at all on anything, just jumbled incompatible style choices imported from unrelated publishing frameworks and commingled in a stew, veering from one to another on an article-by-article basis, determined by nothing but temporary, accidental Google search results. It would be a WP "style system" that would not be systematic in any sense, not predictable, practical, or usable, but utterly random, and which would conflict with every style guide on earth. Each case would require massive amounts of PoV-pushing OR, by people not capable of producing an accurate statistical analysis, from data not capable of producing one either.

It's confusing the reliability of sources of facts relating to a topic, with the authority to dictate how everyone else must write about them. It's exactly the same as supposing that because most physics journals use a serif font, that WP must use a serif font not only in all physics articles but every time a physics concept is mentioned by name.

Ask yourself: What happens, under your system, the next time a "SONY" comes along, some company or product with a logo reading "ZORBO", something that is not actually an acronym but which is written in all caps simply to shout for attention, and it turns out that due to the obscurity of the barely notable topic, there are hardly any sources about it yet but a handful that just happen to uncritically repeat the capitalization style because they didn't figure out it's not an acronym? [This has actually happened recently, with regard to some Linux thing, with some parties to the discussion making the same argument you are, and not being persuasive.] You would force WP to accept "ZORBO" in all caps on the basis of those half-dozen sources, instead of allowing WP the editorial independence to say "no, we have a consistent system that follows mainstream style and we don't all-cap things that are not acronyms/initialisms". What you've outlined and the approach you are taking to it is missing the reliable sourcing forest (how style works systematically in well-documented ways in particular registers of writing) for particular source trees (what Google coughed up this time): "Wow, 50 newspapers do it this way, so this must be the best or only way to do it!" (cf. "50 physics journals ....") No, it's not. It's the one line that all the ants from one particular anthill march along. Capitalization of four-letter prepositions is the chemical signature trail of the AP ants. There are other, unrelated ants going to different places, not following that trail, even if you can't see them because they're not at the patch of ground you're examining.

The core practical problem with the idea is that news publishing has a short cycle and wallows in trivia; it may be years before a particular composition, like "Do It like a Dude" is mentioned (if ever) in non-journalistic sources, like a musicology journal (or a language one, which is actually more likely, remarking upon the capitalization of "A" in so many journalism articles, as if a bunch of record labels and news organization are – shocker! – owned by the same corporations and making cross-promotional decisions). Ergo, all topics that just happened to be of passing popular interest but no [known, yet] lasting significance would, under your system, automatically receive journo style treatment, even though it is incompatible with WP's style system and WP is not not news-writing, and then if academic sources later addressed that subject for some reason, a whole new fight would erupt about how to treat the prepositions in the title of it, because new sources would appear using a different capitalization system (probably the other extreme that WP doesn't use, either).

This would be silly and impractical, like assuming that an immortal mouse lives under your fridge because it keeps coming out and you keep killing it, when the obvious answer is that there's a mouse hole behind your fridge and a nest of breeding mice back there, a consistent source of mice. There is a consistent source of four-letter-prepositions-capitalized titles (PR and journalism – AP style and its knock-offs), there is a consistent source of no-prepositions-capitalized titles (academic journals), and there is a consistent source of five-letter-prepositions-capitalized titles (the whole rest of the publishing world between the two extremes). The different systems do not affect different titles (different names or different classes of names). An AP source will capitalize "With" in the title of an academic journal article; an academic journal will lower-case "behind" in a newspaper article's title, and all other publishers will use the system everyone is already familiar with (consciously or not), and give the title as "Mouse Hole with Nest Found Behind Blueboar's Refrigerator".  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  21:40, 21 February 2016 (UTC)

Wow, "Capitalization of four-letter prepositions is the chemical signature trail of the AP ants," yes? Thanks so much. But very pleased you want to rule out "temporary, accidental Google search results." Martinevans123 (talk) 21:47, 21 February 2016 (UTC)
If your principal objection is that you don't like one of my metaphors, I'm good with that.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  22:16, 21 February 2016 (UTC)
PS: COMMONNAME and MOS are not in conflict on this, because a name and how different publishers style it are not the same thing; you are not the suit you are wearing. Virtually all of our major policies and guidelines involve WP:RS in some way, and apply it as needed to the context at hand; that doesn't mean they're "in conflict" just because they're not copy-pasting the same RS-related wording in every place. WP:N uses RS differently from WP:COMMONNAME, too.

There is no recognizability or naturalness problem following the style recommended by mainstream style guides instead of specialist ones; the opposite it the case. It's not a plausible scenario that someone won't recognize that when they get to Do It Like a Dude they have arrived at the song article they're looking for. Wouldn't it make an order of magnitude more sense to explain "this usage is prepositional" on the talk page, and move on, than capitalize all four-letter prepositions and piss off a million readers just to save a handful of them from having to learn something about parts of speech? It's a non-problem that we can't reasonably "fix" at the very high cost of forcing the capitalization of "with", "from" and other four-letter propositions in titles, something that, while recognizable, will be angrily resisted by the majority of editors and treated as inappropriate, unnatural journo-speak by the majority of our readers. It would be a stake through the heart of our credibility, and inspire incoming editors to write articles here in a journalistic style, following a beacon-like signal that we now suddenly adhere to the AP Stylebook, that we should start titling articles and headings within them like news headlines, etc. We've already had to go out of our way for years to stop people from writing WP:LEADs as if they were journalistic "ledes". We already have a massive problem across something like 30,000 articles, with people abusing pull quote templates just to improperly draw undue, PoV-pushing attention to block quotations, mimicking PR and news style. We must learn from these experiences. Especially when it comes to topics of pop-culture interest, it's already a daily struggle to marshal an encyclopedic instead of news-style tone; this problem badly affects fiction-, sports-, and entertainment-related articles. Pandering to that urge would be disastrous.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  22:16, 21 February 2016 (UTC)

Gosh, a million readers, you say? And I'm certainly not wearing a suit. Martinevans123 (talk) 22:50, 21 February 2016 (UTC)
Way more than a million; I was using that as a shorthand for "a whole bunch". En.WP gets over 8 billion (with a b) page views per month from a global Internet-using population of about 3 bil people (16723 mil pageviews, of which 48.6% are for en.wp = ~8,127,378,000 [28]. [29]). Given that the four-letters-and-under rule only exists in journalism style guides (with one exception, Gregg, and that's a 10-year-old edition), and the never-capitalize-prepositions rule is only used in academic publishing (also with one exception, Chicago), it's a certainty that way more that 1 mil people find capitalizing "with" and "from", per the journo extreme, to be jarring and weird, and that they also find lowercasing "amongst" and "throughout", per the ivory tower extreme, jarring and weird.

Anyway, I'm going back to sourcing. Arguing with the same two people who cannot refute a single facet of the facts and arguments I present, and just react with denial and with repetition of their preference for a result that won't actually work here for reasons they decline to absorb, is a waste of everyone's time. We already know mainstream style guides for general English usage do not capitalize four-letter prepositions; that journalism publishers don't even follow general journalistic conventions but pride themselves on idiosyncratic house style; that music-specific style guides follow academic usage; that our compromise, like that of most publishing, between the two extremes is stable and reasonable; and that the ground truth of how the general, Internet-using public approaches title capitalization does not follow the marketing/journo style. That is more than sufficient.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  23:44, 21 February 2016 (UTC)

Hmm, sounds like My kind of Town.... it's Almost like Being in Love. Martinevans123 (talk) 23:10, 22 February 2016 (UTC)
  • Jeez, is this Wall of Text all about the word "like" and if it should be capitalized? Common sense would say that if an author or screenwriter or playwright or poet, or just a kid with a blog read by his mates, names one of their artworks using a capitalized "Like", then "Like" should be capitalized. I've read that when an organization is around long enough that those who make policy just keep on making policy, every day, round the clock, with walls of text like the walls of Jericho, after awhile this constant policy control seriously damages that organization. Well, as Marilyn swayed in Some Like It Hot, we know artwork when we see it. Randy Kryn 4:44, 23 February 2016 (UTC)
Per that poster artwork, don't you mean SOME LIKE IT HOT? Dicklyon (talk) 05:41, 23 February 2016 (UTC)
Cool, you're back! I was gone for the past couple of months, good to see ya. I'm sure people seeing the film on the big screen didn't care what it was named. I see wall of text redirects to paragraph, which doesn't really describe the phenomena of knowing an important question is being decided but that the time it takes to read the novel is beyond the average Wikipedian's endurance. This seems to be about if "like" should always be lower-cased, which doesn't make much sense to me given the number of major books and films named with the upper-cased "Like". Randy Kryn 6:00, 23 February 2016 (UTC)
Well, yes, I'm back, and it's clear you didn't read nearly enough to know what this is about. Nobody is suggesting that Like should always be lowercase; as a verb in that movie title it would be capitalized in most known styles including wikipedia's. Dicklyon (talk) 06:30, 23 February 2016 (UTC)
The issue is around the word "like" when used as a preposition, but excluding when it's a preposition being used as a subordinating (not a coordinating) conjunction. I think. Martinevans123 (talk) 10:46, 23 February 2016 (UTC)'
If it's being used a subordinating conjunction, it's not a preposition at all.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  08:17, 25 February 2016 (UTC)
@Randy Kryn:: It's a wall o' text because some people will not take "no" for an answer, and the only way to put the issue to bed is to source it to kingdom come, and show where their position if faulty. Yes, it's tedious, but as far as I've learned in a decade here, there is no other way to put out a specialized-style fallacy fire. Or in this case, a WP:Common-style fallacy one (page forthcoming done).  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  08:17, 25 February 2016 (UTC)
  • Random idea off the top of my head, which probably makes no sense because of special treatment for one word: Words of four letters or fewer unambiguously used as prepositions [are not capitalized], unless they are the first or last word of the title. The word "like" is excepted from this rule, and is lowercased only if reliable sources do so. Under this proposal, the word's usage as a preposition must be unambiguous for the MoS to require lowercasing, and the word "like" defers to reliable sources to avoid any complications of declaring it always upper- or lowercase.  ONR  (talk)  21:06, 24 February 2016 (UTC) — Preceding unsigned comment added by Old Naval Rooftops (talk • contribs)
That seems to be a solution to a different problem. Nobody is saying that like should be always uppercase or always lowercase. Dicklyon (talk) 05:17, 25 February 2016 (UTC)
A few people actually do seem to always want it uppercase! But it's pointless anyway, because any academic source will lowercase it as a preposition, so there will always be RS for it in lower case. Zero style guides anywhere have a special rule about "like", and even if one existed, others would not, so there will never in our lifetimes be a day in which there aren't RS for "Do It like a Dude". It doesn't really matter what the word is, it's all about the three well-established capitalization systems that operate in journalism at the four-letter extreme, academe at the capitalize-no-prepositions extreme, and everywhere else in the middle with the five-letter system. [shrug] If we made an exception for "like", then this would just start all over again with some other word. And everyone who is wrongly convinced that MoS is making up fake rules out of nowhere that have no basis in real-world English language usage (just because what we advise doesn't always match what they learned in school in 1989 in Wilmington, and they don't read style guides) would suddenly be correct. MoS really would be full of crap on something for a change.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  08:17, 25 February 2016 (UTC)
If we see a real life example of "Someone Like You" (and I think we all agree there are many), how do know if it's an example of the source applying the "journalism" four-letter rule or an example of the source making an exception to the "middle ground" five-letter rule because the following noun has only three letters? What if sources regularly ignore and adjust style-guides for whatever reasons? Where's the evidence that style guides are regularly followed? I guess you'd say these questions are all red herrings because common use is irrelevant and we should simply stick rigidly to what's been agreed as house style here. But I think that any claims that strict capitalization rules at Wikipedia "prevent arguments over capitalization" are a little wide of the mark, don't you? Martinevans123 (talk) 11:41, 25 February 2016 (UTC)
There is no rule in any known style guide to "capitalize short prepositions if they precede a three-letter noun", and we wouldn't care if someone else's house style had such a rule. One might as well ask "how do we know this publisher doesn't have a rule to mimic the title capitalization of albums and songs from record labels owned by the same parent company but not anyone else's?" or "how do we know this publisher wasn't ordered to capitalize this way by Zorkonn the Lord of Alpha Centauri and his invisible alien army?" It's just totally extraneous. We have a house style that matches most non-fiction publishers; just apply it, no more problem. And, no, our capitalization and other rules are not wide of the mark in preventing disputes. En.wikipedia has over 5 mil articles. The number of actual title disputes is some fraction of the number of pages listed at WP:RM, so on average around 25–100 at any given time, despite numerous ways to approach the titles of probably at least 1 mil of those articles. MoS and WP:CRITERIA combined work amazingly well. Most of the disputes that arise are actually caused by people misunderstanding the rules or trying to buck them on purpose.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  20:18, 25 February 2016 (UTC)
I find your suggestions about Zorkonn the Lord of Alpha Centauri and his invisible alien army less plausible than ones based on possible real world mechanisms. But you didn't answer my questions about how we know any sources follow any particular set of capitalisation rules. If they have no bearing on what's done here, you should admit that. We have your own personal claims about knowing about "how journalists work", based on your own personal work life experience, but that's not quite the same, is it. I wasn't claiming that capitalization and other rules are "wide of the mark", but that the claims that they prevent arguments are wide of the mark. I think you've have to agree, however, that such claims are wholly untestable hypotheses. Martinevans123 (talk) 23:41, 27 February 2016 (UTC)
It's irrelevant what some other source's house style on capitalization is; just follow Wikipedia's. Similarly, we don't care at all whether the Wall Street Journal handles unit and currency symbols exactly the same way WP does. There's nothing to "admit"; the entire point is that whether this source or that is following AP style or Oxford style or MLA style doesn't matter. My opinions and experience about journalism have nothing to do with the question at all; I offered that extraneous info in response to your extraneous question about knowledge of journalism; the entire matter is irrelevant chatter. You're mix-and-matching unrelated subjects. Another is your disbelief that style guides prevent arguments. It simply has nothing to with whether WP has a style guide, what it says, why it says it, and whether you should follow it as a WP guideline or continue resisting forever just to be a gadfly. We both have better things to do that pursue any further debate on this. There is not a consensus at WP to adopt AP Stylebook's rules, and that's enough to stop arguing and move on.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  03:05, 4 March 2016 (UTC)

Relevancy of part of speech

How relevant is part of speech relevant for the word "like"? Some Like It Hot uses "like" as a verb unless I missed the dialogue (above) of humor or sarcasm. The issue or problem is people's way of categorizing the word "like", leading to capitalization problems. An editor either disregards or ignores the part of speech and then does whatever s/he likes. Same with consensus in every RM. If the part of speech should be ignored, then what are we going to do next? --George Ho (talk) 18:48, 25 February 2016 (UTC)

That's what unabridged dictionaries are for. This most common dispute of this sort is over whether "as" is being used as a preposition or as a subordinating conjunction, and it's not hard to figure out, because between Oxford, Cambridge, and Dictionary.com, pretty much every possible use of such a word is catalogued, and identified as prepositional, conjunctive, or whatever. The fact that some people have difficulty figuring it out without consulting a dictionary is surely why a number of journalism house styles, including that of The New York Times, modified the AP system, and always lowercase the word "as" in a title except in initial or final position; this is one example among many of the journalism "system" of title treatment not even being consistent from publisher to publisher. Fans of the "we should do what newspapers do" idea seem as unaware of this problem (i.e. there wouldn't actually be anything systematic for us to follow) as they are of WP:NOT#NEWS.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  20:18, 25 February 2016 (UTC)
Unabridged dictionaries are very bulky and heavy. On which other editions can common people rely? George Ho (talk) 23:12, 27 February 2016 (UTC)
I mean the online ones, ha ha. That said, I now remember that the Cambridge one in particular is not always specific as to distinctions between prepositional and conjunctive use, so it's less useful for this. Oxford Dictionaries Online does not have this problem. As far as paper ones, I'm not sure. When I get home I can look through the collection and see if I have a collegiate-size one that does provide these distinctions clearly. (If you don't hear back on this in a few days, ping me; I probably forgot.)  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  02:54, 4 March 2016 (UTC)
@SMcCandlish: I forgot to ping you several days ago. Seems almost overdue. Find print dictionaries? George Ho (talk) 07:53, 14 March 2016 (UTC)
I have them around, but I have to go to bed now (more like an hour ago)! D'oh. I may have time to dig around in them tomorrow night or so.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  08:38, 14 March 2016 (UTC)
Pinging you again, just in case. George Ho (talk) 10:39, 17 March 2016 (UTC)
@George Ho: Thanks. I've recently started using my {{Todo}} again, but finding all the "stuff I mean to get around to" and adding will take time. Heh. Anyway, I have the "current" (all 2005–2010) editions of four college dictionaries around. They're all American; it appears to be an American concept. Oxford and Cambridge both produce varying editions of dictionaries (as do Chambers and some others), but I'm not sure which correspond, if any do, to the same concept (i.e. a dictionary compact enough for a desk reference, but comprehensive enough and in-depth enough for people in or beyond secondary education, and including most major science and humanities academic terms). Not sure it matters, since both Ox. and Cam. put their dictionary databases online for free, at least for the basic definitional material (full OED access costs almost US$300/yr!). The (American) Random House Dictionary's database (or most of it; it's not clear what "Based on the Random House Dictionary" really means) is online at Dictionary.com, which also included some material from the (British) Collins English Dictionary, but the brevity of the material from the latter suggests it's not the full dataset (I will know next month, as the hardcopy unabridged Collins in is my next batch of orders). Here is a review of what I have on hand, using the "like" entry, since disputes about that word are what led here ("as" also, but dictionaries have comparable definition sets for both words):
* The American Heritage College Dictionary, 4th ed. (2010, Houghton Mifflin Harcourt) – Good: Divides all usage by parts of speech, and provides definitions under those, and lots of them, for words like "as" and "like". It's comparable with Oxford in this. (Note: entries from the unabridged AH are online at AmericanHeritage.YouurDictionary.com)
* Merriam-Webster's Collegiate Dictionary, 11th ed. (2005, Merriam-Webster [Britannica]) – Very good: Completely separate entries by parts of speech, with more cases listed than AH, and some history of usage (but less information on current vernacular usage, which is hilarious if you know back-story how AH was created as negative reaction to the linguistic description vs. prescription approach of M-W's Webster's Third International).
* Webster's New College Dictionary [formerly Webster's II New College Dictionary], 3rd ed. (2008, Houghton Mifflin) – Good: Same approach as AH (and shares the same Houghton Mifflin lexical database), but is slightly less comprehensive.
* Random House Webster's College Dictionary, 2nd rev. ed. (2005, Random House [Penguin/Bertelsmann]) – Very good: Same structure as AH and WNCD, and has a few cases covered by none of the others, but little in the way of usage notes.
Among these volumes, I think RHWCD wins, with MWCD as a close second for depth of entries (but AH for number of them, book-wide). MWCD is much easier to use because of the layout. 'WNCD was in last place, but is still very serviceable. If you don't want a paper book, OxfordDictionaries.com, AmericanHeritage.YouurDictionary.com, and Dictionary.com are probably sufficient. As I noted before, the Cambridge site is less careful about grammatical distinctions. One paper college dictionary I do not have on hand is Webster's New World College Dictionary, 5th ed (2014), the most recent volume in this sub-field. It is on my to-get list for April. While some content from it is online as a database at Websters.YouurDictionary.com (in very poor form with no parts-of-speech info at all; I really hope the print edition isn't that poor). PS: Wiktionary is surprisingly deficient at wikt:like.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  14:08, 17 March 2016 (UTC)

Well, here's another colossal fucking waste of time

(Summoned by bot.) Anybody with a hint of self-respect will just walk away from this and leave it to those with nothing better to do. Every time they see the words like or Like in an article title (whether it's like, or not Like, the way they prefer it) it will be a reminder that they had the sense to spend their time elsewhere‍—‌productively. The remaining pedants can resolve it by flipping coins or any other method that leaves the rest of us in peace. EEng 01:39, 7 March 2016 (UTC)

Could we get an agreement on which currency the coin should be? --Richhoncho (talk) 03:40, 7 March 2016 (UTC)
You have captured perfectly the essence of this entire thread. EEng 04:24, 7 March 2016 (UTC)
If only people would stop trying to "re-legislate" every stylistic nit-pick they ever encounter on WP that doesn't agree with their personal preference, these kinds of threads would not arise. I'm not sure why it is that people who accept at face value that if they write an article for the Encyclopaedia Britannica, or The New York Times, or The Guardian that they will submit it in conformance with the house style guide at the publication in question, or that it will be edited to conform; yet when they come to Wikipedia and encounter a style guide here, their heads explode and they go on months- even years-long warpaths about trivia. It boggles the mind. People must accept that, like all other professional-grade publications, this one has a style guide, and like all others it is a compromise between innumerable available options, the point of which is to present content consistently. WP cannot support every possible style in the world simultaneously, or all anyone would ever do is editwar about every style for everything, in every article. I guess it's better that we periodically have these disputes on the MoS talk pages, since it at least centralizes the perennial noise, but even most of these brouhahas are a waste of time and energy. If someone doesn't like something in MoS, they are free to write new material not in conformance with it and others will clean it up later.  — SMcCandlish ☺ ☏ ¢ ≽ʌⱷ҅ᴥⱷʌ≼  23:48, 8 March 2016 (UTC)
Is that coin a little l, or a big L, or something more tropical? Did someone say Chicago? Martinevans123 (talk) 23:59, 8 March 2016 (UTC) I'm sure when this thread is finally put to bed, another will, sooner or later, "Rise like a Phoenix." Martinevans123 (talk) 22:20, 9 March 2016 (UTC)
Retrieved from "https://en.wikipedia.org/w/index.php?title=Wikipedia_talk:Manual_of_Style/Capital_letters/Archive_21&oldid=1137419954"