Wikipedia talk:Manual of Style: Difference between revisions

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia
Content deleted Content added
→‎Arbitration motion regarding hyphens and dashes: Let’s go through current MOS:ENDASH item by item, using subitems if necessary.
Line 670: Line 670:
::I suggest that we go through the current [[MOS:ENDASH]] item by item, much as I tried to do above, and, if necessary, break items into subitems if we agree on some uses but disagree with others. I also suggest that we avoid pointless, capricious labels such as “Oxford dash” because they only serve to polarize.
::I suggest that we go through the current [[MOS:ENDASH]] item by item, much as I tried to do above, and, if necessary, break items into subitems if we agree on some uses but disagree with others. I also suggest that we avoid pointless, capricious labels such as “Oxford dash” because they only serve to polarize.
::This would at least serve as a start; it would remain to determine the effect to which whatever we could agree on might work: prescriptive, advisory, or merely descriptive, though I can’t see the point of the last. [[User:JeffConrad|JeffConrad]] ([[User talk:JeffConrad|talk]]) 02:20, 19 May 2011 (UTC)
::This would at least serve as a start; it would remain to determine the effect to which whatever we could agree on might work: prescriptive, advisory, or merely descriptive, though I can’t see the point of the last. [[User:JeffConrad|JeffConrad]] ([[User talk:JeffConrad|talk]]) 02:20, 19 May 2011 (UTC)

:::For me, it's not that I'm necessarily "tired of it" (although there is a bit of that). Arbcom is threatening to deal with things "severely" here, so I'm going to make sure that I'm not involved in this at all.<br/>—&nbsp;[[User:Ohms law|<span style="font-family: Courier New, monospace ;font-style:italic">V = IR</span>]] <span style="font-variant:small-caps">([[User talk:Ohms law|Talk]]&thinsp;&bull;&thinsp;[[Special:Contributions/Ohms law|Contribs]])</span> 02:26, 19 May 2011 (UTC)

Revision as of 02:26, 19 May 2011

Template:MOS/R

WikiProject iconManual of Style
WikiProject iconThis page falls within the scope of the Wikipedia:Manual of Style, a collaborative effort focused on enhancing clarity, consistency, and cohesiveness across the Manual of Style (MoS) guidelines by addressing inconsistencies, refining language, and integrating guidance effectively.
Note icon
This page falls under the contentious topics procedure and is given additional attention, as it closely associated to the English Wikipedia Manual of Style, and the article titles policy. Both areas are known to be subjects of debate.
Contributors are urged to review the awareness criteria carefully and exercise caution when editing.
Note icon
For information on Wikipedia's approach to the establishment of new policies and guidelines, refer to WP:PROPOSAL. Additionally, guidance on how to contribute to the development and revision of Wikipedia policies of Wikipedia's policy and guideline documents is available, offering valuable insights and recommendations.

See also
Wikipedia talk:Writing better articles
Wikipedia talk:Article titles
Wikipedia talk:Quotations
Wikipedia talk:Manual of Style (dates and numbers)
Wikipedia talk:Manual of Style/quotation and punctuation

Compromise on WP:REFPUNC?

Wikipedia talk:Manual of Style (footnotes)#Compromise on WP:REFPUNC? -- Jeandré, 2011-03-21t12:46z

Amended proposal

En dashes (–, &ndash;) have several distinct roles.

  1. As a stylistic alternative to em dashes.
  2. To separate items in a list—for example, in articles about music albums, en dashes are used between track titles and durations, and between musicians and their instruments. In this role, en dashes are always spaced.
  3. To stand for to or through in ranges (pp. 211–19, 64–75%, the 1939–45 war). Ranges expressed using prepositions (from 450 to 500 people or between 450 and 500 people) should not use dashes (not from 450–500 people or between 450–500 people). Number ranges must be spelled out if they involve a negative value or might be misconstrued as a subtraction (−10 to 10, not −10–10).
  4. Some style guides recommend an en dash to stand for to or versus (male–female ratio, 4–3 win, Lincoln–Douglas debate, France–Germany border).
  5. Some style guides recommend an en dash to stand for and between independent elements (diode–transistor logic, Michelson–Morley experiment). An en dash is not used for a hyphenated personal name (Lennard-Jones potential, named after John Lennard-Jones), nor a hyphenated place name (Guinea-Bissau), nor with an element that lacks lexical independence (the prefix Sino- in Sino-Japanese trade).
  6. In compounds whose elements themselves contain hyphens or spaces (the anti-conscription–pro-conscription debate); this can also be done when prefixing an element containing a space (pre–World War II technologies, ex–prime minister) – but usually not when prefixing an element containing a hyphen (non-government-owned corporations, semi-labor-intensive industries). However, recasting the phrase (the conscription debate, technologies prior to World War II) may be better style than compounding.
Whether to use a dash or some other form in cases 4, 5, 6 should be decided on the article's talk page; when a strong consensus among reliable sources for the subject of the article agree on a form, this should be considered in the discussion. [Mentioning such decisions on WT:MOS will be helpful in keeping this page in touch with community thinking on this matter.]

————————

This seems the minimum change in Tony's draft at all likely to achieve consensus. Those objections above which do not require MOS utterly ignoring usage have been answered as far as I can; nothing else has chsnged. Septentrionalis PMAnderson 20:02, 9 May 2011 (UTC)[reply]

Support as amended

  1. Support as drafter. Note that this is the minimum which any member of the dissentients of class D seems willing to accept, and that I have intentionally kept even the bracketted sentence above, although I think it useless; even if MOS includes it, I do not expect anybody to actually comply with it. Septentrionalis PMAnderson 20:02, 9 May 2011 (UTC)[reply]
  2. I can agree with this. It reflects reality... letting editors know that different style guides do not always agree on these points, and so we have decided not to choose between them. Yet it gives editors a way to settle their disputes (ie look at what sources do). I also like this because the situations discussed in points 4,5 and 6 are the ones where there is potential for conflict with WP:TITLE... PMA's wording resolves that potential conflict. The two pages would give the same advice (look to the sources). Blueboar (talk) 21:13, 9 May 2011 (UTC)[reply]
  3. My first choice is to recognize that there is no purpose, value or use of en-dashes (or em-dashes) in WP for anyone. Readers are entirely unaffected by their use, and the only effect of using dashes on editors, when there is any, is dispute and consternation. Just look at how complex these proposals are! For no benefit whatsoever? Ridiculous! Just say no to any use of any dashes, turn on a bot that converts all dashes to hyphens (except in an article about dashes), and we're done. The alternative is to keep on pointlessly arguing about something that has no right answers, no hope of resolution, and couldn't matter any less. Absurd.

    Having said that, and in the absence of any apparent support for the solution that I strongly favor, at least this proposal brings the wisdom of the ENGVAR approach to this arena, which should help cut down on the conflict significantly. --Born2cycle (talk) 04:19, 10 May 2011 (UTC)[reply]

    It is my understanding that even people who cannot look at a piece of paper and find every last misplaced comma still get general impressions from the punctuation. Good punctuation gives the impression that the piece is finished and professional and poor punctuation gives the impression that it is slapdash and amateurish. It's like the temperature in a room: Some readers will notice that it's too stuffy or too chilly and some won't, but they'll all sweat or shiver. Darkfrog24 (talk) 13:58, 10 May 2011 (UTC)[reply]
    The problem is that different guides give different rules as to what "good punctuation" is when it comes to dashes. Just as different dictionaries will give different spellings for words depending on the variety of English. To an American, "honour" looks "wrong"... but to a Brit it looks "right". The same can be said for dash usage. Using a dash in a given situation may look "wrong" to some editors, but it will look "right" to others. As long as we are consistent within each article, we should be fine... the average reader won't even notice, and few who do (and care) will just have to accept that we allow variation. Blueboar (talk) 14:55, 10 May 2011 (UTC)[reply]
    The lack of standards about "good punctuation" when it comes to dashes is indeed a big part of the problem. But the, other issue, is that, most people, can, tell when commas are not being, used properly. The difference between -, – and — is not nearly so universally recognized. I hasten to add it's more often overlooked than noticed.

    Darkfrog24, to characterize this opposition to trying to distinguish between the nearly indistinguishable dashes and hyphens as opposition to use of proper punctuation in general is disingenous, and entirely missing the point. --Born2cycle (talk) 00:55, 11 May 2011 (UTC)[reply]

  4. Support but I'd replace Some style guides recommend an en dash to with An en dash can. A. di M.plédréachtaí 12:30, 10 May 2011 (UTC)[reply]
    I have no problem with that amendment. Septentrionalis PMAnderson 17:20, 10 May 2011 (UTC)[reply]
    This still would leave a lack of parallelism among the list items; I’d much rather see the same introductory phrase as for the others, and perhaps a statement in the paragraph following the list that not all style guides agree on items 4 and 5 (actually, not all guides agree on mixing hyphens and en dashes as in the first example in 6; one that does not support this is OSM. JeffConrad (talk) 17:45, 10 May 2011 (UTC)[reply]
    If this is a purely stylistic quibble, please let us discuss it after the substance is agreed upon. If it is substantive, I should add that the difference in expression, like the may in 6, represents a real difference in the standing of these rules. But I am perfectly willing to do a draft which omits 4, 5, and 6 and says in the final paragraph that some writers use en dashes instead of hyphens. Would you support that? Septentrionalis PMAnderson 02:06, 13 May 2011 (UTC)[reply]
    But this difference of standings in other style guides should not be used to make our own style guide more wishy-washy. We should stick with the MOS we have, which is designed to support the greatest possible clarity of writing, and consistency across WP. This amendment goes way beyond Tony's in muddying those waters. Dicklyon (talk) 02:11, 13 May 2011 (UTC)[reply]
    Why of course, how ever could I have failed to recognize the basic principle here? "Everybody should do what Dicklyon wants; the recommendations and cautions of reliable style guides be -er- disregarded." We should use the MOS we have, even where its distinctions make absolutely no difference in practice and are not used consistently by any respectable publisher; even though they are not consensus, because Dicklyon and his friends like them. I sit corrected.

    Quite seriously: I do not believe or trust anyone who could write this post. Of course we should listen to what reliable sources - such as style guides - say; it is our hypothesis that they know more than we do. When they contradict each other, and we must decide, then we must weigh them against each other; but this is not an issue on which we need to decide, except for those who agree with Tony that not doing so means the Decline and Fall of MOS.

    We already know that Dicklyon is opposed to any compromise in the guideline he owns, and wishes to drive on to the "yawning heights" of punctuation reform; we can allow for his opposition in considering whether any proposal is consensus; he's already said so for this one. Now can those of us considering actual compromise get on with it? Septentrionalis PMAnderson 02:23, 13 May 2011 (UTC)[reply]

  5. Gives pretty clear guidence, but still allows for enough flexibility when a specific (typically specialized) variation is needed. Some "terms of art", as it were, should be consistently and accurately represented in articles, even if it feels "wrong" to those who are attentive to typography. Of there's a question about it, that's what article talk pages are for. oknazevad (talk) 20:54, 10 May 2011 (UTC)[reply]
  6. Weak support as more sensible than the other proposal. If academic publishers don't enforce 4, 5, and 6 on their authors, and when significant variation is noted in practice, it would be quite unreasonable for a massive project like Wikipedia to impose a single style in those respects. We don't even do that for reference formatting, and publishers usually are more strict about enforcing their house style in that area. I think however that something should be said about page-level consistency like we have for citation styles. If find the oppose below claiming that not imposing a rule for 4-6 is equivalent to deleting the MOS to be quite misguided, and almost made me chuckle. Tijfo098 (talk) 07:12, 11 May 2011 (UTC)[reply]
    If somebody wants a link to MOS:CONSISTENCY, I have no objection; I didn't add it myself because I'm not convinced we need to repeat it for every item. Septentrionalis PMAnderson 17:43, 11 May 2011 (UTC)[reply]
  7. Support (and would also support the change suggested by A. di M. above), per my comments on the earlier proposal. With the "disputed" usage cases of the dash, the MoS should either leave the choice open, as this proposal does, or opt explicitly for the technically simpler, more commonly used form (i.e. decide against the dashes in favour of hyphens). What it should not do is force a rare stylistic affectation on users when there is no objective need to do so. Fut.Perf. 13:42, 11 May 2011 (UTC)[reply]
  8. Support. I am happy with this caveat, with extension to WikiProject consensus regarding similar usage. Axl ¤ [Talk] 08:25, 13 May 2011 (UTC)[reply]

Oppose as amended

  1. This has the same problems as the previous wording: it advocates multiple debates across the wiki instead of a centralized discussion, and this proposal would lead to articles with different styles. The point of having a manual of style is precisely so editors don't have to come to a consensus about style on the article talk page. All they have to do is look to the MOS for best practices. While I'm aware that there are critics of the MOS itself, I think any failings here should be addressed at the source rather than numerous article-specific arguments to fill in the cracks. As long as this proposal mixes two separate issues—the hyphen–dash debate and the MOS–article consensus debate—the results from supports and opposes is muddied. These two issues should probably be split up, with the latter debate concluding first, as its outcome may negate the need for the former. — Bility (talk) 20:40, 9 May 2011 (UTC)\[reply]
  2. The MOS should answer questions like this to provide a consistent, professional look and so we don't have to have this debate a thousand times. It's too much work. This is worse than Tony's, since it doesn't even seem to take a stand on whether to use hyphen/dash/nothing in several cases. We already decided up above to have an MOS, I don't think the point there was to have an MOS that just presented some options to choose from. ErikHaugen (talk | contribs) 20:55, 9 May 2011 (UTC)[reply]
  3. The MOS lead says "The Manual of Style is a style guide for Wikipedia articles that encourages editors to follow consistent usage and formatting." To divert the editor by discussing other style guides that differ from it can only confuse and lead to less consistency, and in the case of the deviations that are effectively encouraged here, less clarity and more ambiguity for the reader. Dicklyon (talk) 02:07, 10 May 2011 (UTC)[reply]
    Ah, but if PMA's proposal is implemented, those other style guides will no longer differ from our style guide... because our style guide will have been changed to effectively say "we don't mandate one particular style on these points"... our guide will thus incorporate and include what all the other guides say. Blueboar (talk) 02:18, 10 May 2011 (UTC)[reply]
    Just such a discussion takes place as to whether to write "US" or "U.S." and it comes off rather well. Darkfrog24 (talk) 13:46, 10 May 2011 (UTC)[reply]
  4. Oppose—This is just too radical. We may as well delete the MoS, and along with it all of its sister pages. Why set out a guideline and then say do what anyone else does? A primary problem is that usage differs significantly out there, so there's no one "authority" you can defer to. That is exactly why every decent publishing house has a style manual. On WP, of course, the MoS serves not only to coordinate the style on this sprawling project, but to minimise disputes in articles. Who wants to promote continual brawling about many many stylistic issues all over the place? Tony (talk) 02:55, 10 May 2011 (UTC)[reply]
    • If this were equivalent to deleting the MOS, I would not have written eight paragraphs of proposed text for inclusion in the MOS. WP:COMMONNAME says "follow the sources" - and nobody has proposed deleting it as equivalent to a blank space.
    • The gentle editors who have conducted four pages of article disputes to "preserve the authority of MOS" need to work harder to claim it minimizes disputes.
    • There are occasions where usage differs. We already offer guidance for that case: Be consistent within an article. This works - and actually does minimize disputes; whenever anybody insists that honor (or honour) is just plain "wrong", somebody introduces the good soul to WP:ENGVAR - and the dispute goes away. The brawling is produced by exactly what WP:DASH does - attempting to insist that literate editors defy usage because some piece of MOScruft says so. Septentrionalis PMAnderson 03:50, 10 May 2011 (UTC)[reply]
  5. I can agree that some variation is OK, but this is... too far in the other direction. We need to allow for exceptions to be made, but we should have a preference for how en dashes are used. External style guides are written for their own publications; we have our own distinct needs. I agree that we're not here to "correct English" or any such thing, but there's also nothing wrong with deciding "for Wikipedia, we're going to do it this way".
    — V = IR (Talk • Contribs) 18:31, 10 May 2011 (UTC)[reply]
    Then please do a draft. This seems to me a straightforward application of WP:CONSENSUS, and the minimum that will attract any of those who oppose Tony's draft - and it still states a preference on which way to go in points 4 and 5 (by this numbering). Septentrionalis PMAnderson 18:35, 10 May 2011 (UTC)[reply]
    I'm thinking about it.
    — V = IR (Talk • Contribs) 18:40, 10 May 2011 (UTC)[reply]
  6. I recall my old friend Tennis expert was fond of arguing "local consensus" overriding general consensus. However, it is rare indeed, and I have never seen it in practice. In any event, I believe notification to WT:MOS should not be an ex-post fait accompli but be intended as a notification of a re-opening of discussion about a specific case which may influence the evolution of the MOS. --Ohconfucius ¡digame! 07:22, 11 May 2011 (UTC)[reply]
    The actual consensus of English-speakers is to use hyphens in cases 4 and 5 (proper names sometimes excepted) and reword in case 6; most style guides agree. But let us see what local consensus does do, instead of guessing about it in advance. Septentrionalis PMAnderson 17:40, 11 May 2011 (UTC)[reply]
    If by English-speakers [sic] you mean people who don't know how to use en dashes, then yes. I'm pretty sure that the guides that describe this usage in British and American English were written by and for English writers and editors. Dicklyon (talk) 20:22, 11 May 2011 (UTC)[reply]
    Dicklyon, when you say "people who don't know how to use en dashes", do you really mean "people who don't know the correct way to use en dashes" (or "people who disagree with how I think en dashes should be used")... or do you really mean "people who don't use en dashes because they are unfamiliar with what they are"? Blueboar (talk) 01:15, 12 May 2011 (UTC)[reply]
    Ohconfucius, you probably didn't pay attention to ANI in recent weeks if you think the current en dash MOS rules have "general consensus". Tijfo098 (talk) 05:06, 12 May 2011 (UTC)[reply]

Other

Discussion

Typos: unbalanced "{" and "desisions". Art LaPella (talk) 21:33, 9 May 2011 (UTC)[reply]

Thank you; done. Septentrionalis PMAnderson 04:20, 10 May 2011 (UTC)[reply]

I’d rather see the added commentary covered once (with reference to the item numbers) after the list. In addition to reducing the repetition, this would maintain parallel structure of the list items. A couple of thoughts with either version:

  1. The list items include both an unspaced em dash and a spaced en dash; one or the other should be chosen. I’d opt for the former because it’s more common, and even OUP recommend it.
  2. Item one doesn’t mention that the alternative to an em dash is a spaced en dash; though this is covered in the more expansive treatment that follows, not every reader may get that far.

JeffConrad (talk) 21:13, 9 May 2011 (UTC)[reply]

The individual bullets are as far as possible the same points. On the whole, it seems more sensible to polish phrasing after we have agreed on substance. Septentrionalis PMAnderson 04:20, 10 May 2011 (UTC)[reply]

The RfC immediately above this section doesn't address hyphens, which have caused similar conflicts

E.g. small cell carcinoma vs small-cell carcinoma, high dynamic range imaging vs high-dynamic-range imaging. Hyphens are problematic when certain technical/scientific terms don't commonly contain them, but when the matching MOS rules recommend usage. Tijfo098 (talk) 11:44, 8 May 2011 (UTC)[reply]

The common usage form should be used on Wikipedia. Gerardw (talk) 11:55, 8 May 2011 (UTC)[reply]
So it should. All forms of hyphenation should at base come down to "consult a good dictionary"; usage is changing. Septentrionalis PMAnderson 18:32, 8 May 2011 (UTC)[reply]
The form used should depend on what the goal is. If you want to emulate the usage in the specialist medical community, where the compound term is so familiar that nobody will misread it, then you omit the hyphen like they do. If you want to communicate to a general audience that understands English but is not familiar with the idioms of the oncologists, then you employ grammar rules more strictly, to convey the meaning more clearly (like this book for a general audience does). What you really want to avoid is the meaningless mixed up use of hyphens that you also find sometimes in otherwise reliable sources, like this one; the hyphens in high-dynamic-range imaging were more common around 1991, when the field was less well known. Which should be our goal here? Copy the specialist communities, or write for a general audience? Dicklyon (talk) 18:44, 8 May 2011 (UTC)[reply]
Prove that writers for a general audience point the term differently, and this would be an argument. Without that this is conjecture, and the result will neither equip our readers to understand popularizers nor communicate with the oncologists. Septentrionalis PMAnderson 19:35, 8 May 2011 (UTC)[reply]
Write for a general audience by copying the usage of the specialist communities. Small cell carcinoma is spelled that way on the designed for the general public NIH website.[[1]], and the first three pages of Google search results for "high-dynamic-range-imaging" almost exclusively are not hyphenated. Gerardw (talk) 20:34, 8 May 2011 (UTC)[reply]
As pointed out by this medical communication authority (Ellen Drake, CMT, FAAMT, Development Editor for Health Professions Institute, in "A Quick Primer on Hyphens"), the usage around things like "small cell carcinoma" is an inconsistent mess; hopefully we should aim to do better; she gives clear rules that say to hyphenate it. Here's another (Edie Schwager's "Dear Edie" column in American Medical Writers Association Journal . VOL. 20, NO. 3, 2005), who includes a good discussion of how hyphens tend to disappear when the compound becomes commonplace, but should still be used (in small-cell carcinoma in particular) when there is possibility of the phrase being misleading to the reader without it. That's why I said we should choose whether we want to write for the common reader; it's not about what most writers do, but about our aspirations to do as well as we can, as recommened even within the medical community by their experts on writing. Dicklyon (talk) 22:24, 8 May 2011 (UTC)[reply]
Disagree. We should do what most writers do because that's what most of our readers will see elsewhere. Analogous to WP not being about truth but verifiability. Gerardw (talk) 22:50, 8 May 2011 (UTC)[reply]
I guess we don't really need an MOS then. Wait, didn't we just conclude an RFC on that? Dicklyon (talk) 05:56, 9 May 2011 (UTC)[reply]
That's an illogical response. The necessity of MOS and what the MOS guidelines are separate topics. Gerardw (talk) 17:45, 9 May 2011 (UTC)[reply]
As for “you employ grammar rules more strictly”, this is not so much about grammar as about lexicon. I don't think there's any way to reliably predict whether a given compound is spelled with a hyphen, a space, or closed-up short of using a dictionary or a language corpus: this is no more of a grammar issue than whether the denonym for a given country should end is -an, -ish, -ese, or -i. (Also, with some compounds usage varies both with space and time.[2][3]) Insisting that the “clearer” version rather than the idiomatic one should be used is akin to insisting on lighted rather than lit even if the latter is much more common. A. di M.plédréachtaí 14:39, 9 May 2011 (UTC)[reply]
As a general principle, I'm saying that aiming for the more clear one is better than aiming to match majority usage. Hyphens are very underused in most publications, making them hard to interpret at first reading. The grammar rules about things like hyphenating two word compounds when used as adjectives are really quite simple, clear, and easy to apply, and they make the reading easier even for people unaware of the rules, by visually tying things together that need to be read as one item, in a context where they might otherwise be ambiguous, like "small cell carcinoma". Why would one recommend inflicting that ambiguity on a reader unfamiliar with the intended reading? Dicklyon (talk) 14:55, 9 May 2011 (UTC)[reply]
Can you elaborate on that? I can't see how any simple rule can correctly predict that blackbody radiation, black hole candidate and black-tie dinner are spelled this way. I think it's more a matter of lexical incidence, like the choice of which suffix to use to form a denonym. A. di M.plédréachtaí 15:56, 9 May 2011 (UTC)[reply]
Pretty straightforward, really. A black tie is a two-word compound noun, so you hyphenate it when using it as an adjective. A blackbody is a one-word compound, so it stays that way. A hole candidate who is black is unambiguous as you wrote it. If you had used a hyphen, it would have implied a candidate related to a black hole. Dicklyon (talk) 18:38, 9 May 2011 (UTC)[reply]
Well, let's see:
  • a black body is a two-word phrase; the OED and all of its quotations so spell it.
  • You have misunderstood black hole candidate, an object which is a candidate to be a (actual, as opposed to theoretical) black hole. In the context in which this phrase is actually found, it is unambiguous and needs no hyphen. Pmanderson (18:59, 9 May 2011 (UTC)), — (continues after insertion below.)[reply]
    No, DL did not misunderstand you, you might want to re-read his reply. I'll rephrase without the snark: "black hole candidate" is a "hole candidate" that is black. A "black-hole candidate" is a star that is maybe/probably a black hole. ErikHaugen (talk | contribs) 19:56, 9 May 2011 (UTC)[reply]
    In brief, "Words don't mean what they are actually used to mean; they mean what my theories hold they ought to mean." Sorry, that's not the way English works. Septentrionalis PMAnderson 18:30, 10 May 2011 (UTC)[reply]
    Except that "black+hole+candidate" black hole candidate actually means what PMA said, and it's semantically the same structure as black-tie dinner. Black body is commonly spelled as two words when a nominal of its own[4] and as one when a modifier.[5] Also, probably black holes are black bodies, but that's irrelevant, isn't it? :-) —Preceding unsigned comment added by A. di M. (talkcontribs) 22:29, 9 May 2011
I don't know how relevant a black-hole blackbody would be; it is interesting how variable the usage of blackbody has become; less interesting how often people omit the hyphen in compounds used as adjectives, to allow snarky interpretations. Dicklyon (talk) 02:42, 10 May 2011 (UTC)[reply]
In short, all these are matters of idiom. Several grammarians have proposed ways to change idiom, but we are not authorized to do so, which may be just as well. Septentrionalis PMAnderson 18:59, 9 May 2011 (UTC)[reply]
Wikipedia should reflect the usage in reliable sources not be dictated by grammar rules. We are not here to change the world. --Doc James (talk · contribs · email) 17:34, 9 May 2011 (UTC)[reply]
The American Medical Association manual of style uses "small-cell carninoma" and articulates the rules that say why. Dicklyon (talk) 18:38, 9 May 2011 (UTC)[reply]
The reference is to a 22 year edition. There's a 2009 edition [[6]]. Using the amazon Search inside this book feature shows the current usage is "small cell carcinoma" without hyphens. Gerardw (talk) 19:33, 9 May 2011 (UTC)[reply]
They also use a dash as in "non–small cell". :P — Bility (talk) 20:48, 9 May 2011 (UTC)[reply]
Yes, if you're not going to hyphenate the first, you need a dash for the second. More straightforward just to have small-cell c. and non-small-cell c.. Also, if we're going to follow the AMA style guide, we'd have Serbo Croatian ! Usage is all over the place. Whether we write black-body radiation or blackbody radiation we'll be unambiguous, so there's no reason not to go with the more common (and now well established) form. But when even med style guides disagree on small cell carcinoma, and intro ref books often hyphenate, we do our readers a disservice by choosing the misleading form.— kwami (talk) 20:53, 9 May 2011 (UTC)[reply]
That 2009 edition on Amazon says exactly what I was saying: "use a reader's perspective and the context as guide". They did include "small cell carcinoma" in the "list of examples that can usually be presented without hyphens" due to having become well known. For the general wikipedia readership perspective, are we to assume that such terms are so well known that it would be a bad idea for us to write them unambiguously? I don't think so. Dicklyon (talk) 03:07, 10 May 2011 (UTC)[reply]
I respectfully agree with Doc James. The vast majority of the literature and other sources use "non-small cell carcinoma" and "small cell carcinoma", and does not hyphenate "large cell", "basal cell", "squamous cell", "giant cell", etc. We should follow the style of the literature in the relevant area, and should not be out to change the world.Cliff L. Knickerbocker, MS (talk) 12:05, 10 May 2011 (UTC)[reply]
I don't understand this, really. How is Wikipedia deciding that a style choice for Wikipedia is the equivalent of being "out to change the world"?
— V = IR (Talk • Contribs) 18:39, 10 May 2011 (UTC)[reply]
Make that "out to defy the world" if you like; but several of the dash enthusiasts have expressed themselves as campaigning for a restoration of "good typography", in their own private understanding of the phrase. Septentrionalis PMAnderson 02:35, 13 May 2011 (UTC)[reply]
Wikipedia's articles should reflect common usage. Moreover, WikiProject consensus should overrule MoS. Axl ¤ [Talk] 08:35, 13 May 2011 (UTC)[reply]

Data

What I did come to do is to suggest some data on how rare most of the uses of dashes recommended by the present text are, and how rarely they are recommended by style guides; the discussion above seems to evidence some confusion on this point. I believe most of our examples of dashes above fall under the case that "a strong majority of reliable sources" do not use them. At least one or two of these provisions (including the proposed French–German border) are also omitted by a strong majority of style guides, for those who think this more important than usage. Let me begin with one or two of each kind; please feel free to join in. Septentrionalis PMAnderson 02:55, 7 May 2011 (UTC) Septentrionalis PMAnderson 00:03, 7 May 2011 (UTC)[reply]

The en dash in French–German border is not that unusual in sources, as we've shown (wasn't it about 15%?). In style guides, it's in accord with those you call "Oxford", that is the ones the suggest en dash in equal pairs. You find the same with Mexican–American War for the same reason, but not very frequently (I think we found about 5%). Dicklyon (talk) 18:57, 8 May 2011 (UTC)[reply]
15% is about the level which A. di M. sees as minimal for there to be a sound case to permit the usage; I tend to agree, and would permit it. To require it, however, against the other 85%, is defying usage, our only Academy. Septentrionalis PMAnderson 22:52, 8 May 2011 (UTC)[reply]

Examples

While I agree with Tony's renumbering, it may be clearer to do this using the present numbers. Septentrionalis PMAnderson 00:03, 7 May 2011 (UTC)[reply]

  1. (Tony's # 3): 1939~1945: almost entirely dash, as I would expect. Septentrionalis PMAnderson 00:39, 7 May 2011 (UTC)[reply]
  2. (Tony's # 4):
    • Lincoln~Douglas debates; the first ten results are three dashes to seven hyphens (and the Library of Congress listing for all those which have it in the title uses a hyphen.) Septentrionalis PMAnderson 00:39, 7 May 2011 (UTC)[reply]
    • France~Germany border; only 13 hits; but they are 8 hyphens and 5 slashes.
    • French~German border; of the first ten hits, 5 hyphens, 4 slashes, and a dash. Septentrionalis PMAnderson 02:43, 7 May 2011 (UTC) I make it 6 hyphens, 3 slashes, and a dash for the first ten. In the next ten hits: 5 hyphens, 1 slash, 3 dashes, one monospaced font not allowing distinction. The next ten hits: 8 hyphens, 2 dashes. The next ten hits: 7 hyphens, 3 dashes. The next ten: 9 hyphens, 1 dash. This brings the sample size up to 50, with results as follows: 35 hyphens, 10 dashes, 4 slashes, 1 monospaced font. --JN466 22:30, 7 May 2011 (UTC)[reply]
    • French~German border in University Press publications: 30 hyphens, 9 dashes, 1 slash. About half the dashes are from Cambridge University Press. --JN466 22:30, 7 May 2011 (UTC)[reply]
      And even Cambridge is not consistent; this is the fourth hit on the phrase, published by CUP. Septentrionalis PMAnderson 22:50, 7 May 2011 (UTC)[reply]
      No, they're not consistent. 5 dashes, 3 hyphens, 1 slash. :) --JN466 00:34, 8 May 2011 (UTC)[reply]
  • If you have patience, you can go through "male~female differences" in Routledge, and you'll probably die of old age before you finish. [7] En dashes, slashes, and hyphens can be found, so they don't seem to impose any standard on their authors. Tijfo098 (talk) 07:05, 8 May 2011 (UTC)[reply]
    • Like most publishers, they don't seem to hold their authors tightly to a standard. But they do appear to be improving, with about 50% en dash in male–female for book in the current century, and hyphens in a definite minority. The virgule, though less preferred, serves the same purpose in indicating the disjunction. Dicklyon (talk) 21:49, 8 May 2011 (UTC)[reply]

Style guides

Let's start with the ones in WP:MOS#Further reading.)

  • ABC Radio national style guide. Date ranges only (WP:DASH #1. Septentrionalis PMAnderson 01:01, 7 May 2011 (UTC)[reply]
  • Chicago Manual of Style (16th edition; §6:78-80).
    • Endorses use as to-from (DASH #2).
    • says of DASH 5: The en dash can be used in place of a hyphen in a compound adjective when one of its elements consists of an open compound or when both elements consist of hyphenated compounds (see 7.78). This editorial nicety may go unnoticed by the majority of readers; nonetheless, it is intended to signal a more comprehensive link than a hyphen would. It should be used sparingly, and only when a more elegant solution is unavailable. It recommends avoiding several of its own examples, as clumsy.
    • Endorses US-Canadian relations [hyphen]. If United States were spelled out, it would use a dash, but only because United States has an internal space. Septentrionalis PMAnderson 18:56, 7 May 2011 (UTC)[reply]
    • Guardian style guide: Dashes (in full) "Beware sentences – such as this one – that dash about all over the place – commas (or even, very occasionally, brackets) are often better; semicolons also have their uses. Dashes should be n-dashes rather than m-dashes or hyphens." This is only the use as punctuation (original WP:ENDASH, and even there it urges caution (and opposes the em dash). Septentrionalis PMAnderson 19:14, 13 May 2011 (UTC)[reply]
    • I have no access to the other Australian guide. Tony, could you check it and strike this note? Septentrionalis PMAnderson 01:01, 7 May 2011 (UTC)[reply]
      • Reading the page quickly (aaaa! my eyes!), and checking the top 3 US style guides (which no one has done, except for these comments by Sept), there are clear differences between American and non-American usage. The most commonly followed US style guide is AP Stylebook, which pointedly ignores en-dashes completely. (Caveat: it's most commonly followed because it covers journalistic writing, and everyone acknowledges that this sometimes is and sometimes isn't good enough for scholarly writing ... nevertheless, nothing succeeds like excess, particularly in orthography wars.) Second (same caveat, but it's a little more high-brow) is the New York Times Manual of Style and Usage, which says not to use en-dashes at all. Third (and now we're solidly in "scholarly" style) is Chicago. Sept isn't wrong, but I read Chicago to be more limiting and more leery of the en-dash than Sept's summary. So: "The principal use of the en dash is to connect numbers and, less often, words." And: "In [non-numerical] contexts, such as with scores and directions, the en dash signifies, more simply, to." When it's taking the place of a hyphen where one or both of the elements being joined has a space: "... only when a more elegant solution is unavailable." And: "Chicago's sense of the en dash does not extend to between."
      • Comparing the top 3 US style guides to Garner's (an influential style guide published by Oxford University Press) and to many of the other style guides mentioned here, there's a clear split in usage. I've always been on board with MOS's advice on dashes simply because no one that I copyedited for cared, and when no one cares, it's nice to be able to stick with one style. Now that the issue is at Arbcom, it's pretty apparent that people care, and in this environment, it's time to get back to basics and follow consensus and style guides. - Dank (push to talk) 16:47, 13 May 2011 (UTC)[reply]
      • Inserting: My MLA Style Manual, latest (3rd) edition, commonly followed for books and articles in English, the arts, and philosophy, says "dash" as if there's only one kind (an em-dash). And here's everything relevant from APA style, latest (6th) edition, commonly followed in the social sciences: "An en dash ... is used between words of equal weight in a compound adjective (e.g., Chicago–London flight). Type as an en dash or, if the en dash is not available on your keyboard, as a single hyphen." - Dank (push to talk) 17:50, 13 May 2011 (UTC)[reply]
        Thanks Dank, that was interesting. Can you clarify, though; are you suggesting a particular change to the style guide, or whether we should use the same style on each article, or any of the other issues being discussed here? Thanks, ErikHaugen (talk | contribs) 16:54, 13 May 2011 (UTC)[reply]
        Clearly, there's a split between American and non-American usage. OTOH, I was never on board with the idea that people (including me) who argued in favor of our MOS were obviously up to no good just because our MOS didn't follow American usage, for the simple reason that I don't recall ever copyediting for a WPian who cared. And as long as no one cared, I thought there was something elegant about having one set of rules that's simpler than the rules in most of the style guides. But now, for at least as long as Arbcom cares about this issue, I'm going to have to be sensitive to what writers and wikiprojects want and (for AmEng articles) what American style guides say. The great majority of recent reliable sources of American origin will have been copyedited by someone who was trying to follow at least one of the style guides I just quoted. - Dank (push to talk) 18:17, 13 May 2011 (UTC)[reply]
        And a split within Commonwealth usage as well. Septentrionalis PMAnderson 19:14, 13 May 2011 (UTC)[reply]
        Sure – these regional variations complicate things. But specifically what are you suggesting? How can we "be sensitive"? Should the MOS not take a stand and instead defer to "local consensus" at the project level? Article level? Should we just do it ENGVAR-style and say whoever writes it first just leave it that way? I think Arbcom is just trying to get us to answer these kinds of questions and/or get consensus on what the MOS should say. ErikHaugen (talk | contribs) 20:49, 13 May 2011 (UTC)[reply]
        ENGVAR style would actually be "Leave it alone unless there is consensus that there is good reason to change." I'll support that; I would also support consistency within an article; I hope that it will not be abused to generalize one isolated dash into a whole system. A MOS which described actual English writing would be more useful, but there seems to be strong opposition to that. Septentrionalis PMAnderson 22:05, 13 May 2011 (UTC)[reply]
        Although I make uncontroversial changes to the MOS sometimes, I generally consider questions like those where there's real tension in the air to be above my pay grade, Erik. Discussions like these take a lot of time and research. You guys change the MOS however you like, and I'll continue to consult it, along with WP:MILMOS, relevant style guides, and what I pick up from discussions during FAC and A-class reviews. - Dank (push to talk) 23:59, 13 May 2011 (UTC)[reply]
      • What is your basis for “top 3 US style guides”? JeffConrad (talk) 20:33, 13 May 2011 (UTC)[reply]
        • What would you put higher? Septentrionalis PMAnderson 20:50, 13 May 2011 (UTC)[reply]
          “Top” is pretty subjective; I’d usually start with CMoS, simply because it’s so widely acknowledged and comprehensive. The APA Style Guide has consistently outsold it, however; the price (just under half that of CMoS, could well be a factor. JeffConrad (talk) 22:19, 13 May 2011 (UTC)[reply]
          Good question ... it's often asserted in J-schools and by the publishers, but I'm not aware of any data ... I don't think it's too much to ask them to supply it. I'll email the publishers. - Dank (push to talk) 21:06, 13 May 2011 (UTC)[reply]
          Though I don’t suggest that it’s necessarily an answer to the question, I just got the following Amazon Bestsellers Rank numbers:
          1. Publication Manual of the American Psychological Association, Sixth Edition: #19
          2. The Chicago Manual of Style, 16th Edition: #449
          3. The Elements of Style (4th Edition): #526
          4. A Manual for Writers of Research Papers, Theses, and Dissertations, Seventh Edition: Chicago Style for Students and Researchers: #696
          5. Garner’s Modern American Usage, 3rd ed.: #2,074
          6. Merriam-Webster’s Guide to Punctuation and Style: #12,396
          7. The Chicago Manual of Style, 15th Edition: #14,386
          8. MLA Style Manual and Guide to Scholarly Publishing, 3rd Edition: #20,161
          9. The New York Times Manual of Style and Usage: #27,743
          10. Words into Type, Third Edition: #34,474
          11. The Associated Press Stylebook: #111,881
          12. The Cambridge Grammar of the English Language: #122,589
          13. The Chicago Manual of Style, 14th Edition: #151,441
          14. Fowler’s Modern English Usage, 3rd rev. ed. #191,507
          15. Fowler’s Modern English Usage: the Classic First Edition, ed. by David Crystal (2010): #194,870
          16. Scientific Style And Format: The CSE Manual for Authors, Editors, And Publishers: #238,629
          17. The Oxford Style Manual: #297,890
          18. New Hart’s Rules: The Handbook of Style for Writers and Editors: #406,432
          19. Merriam-Webster’s Manual for Writers and Editors: #796,570
          JeffConrad (talk) 21:47, 13 May 2011 (UTC)[reply]
          I’ve added New Hart’s Rules and the Cambridge Grammar, and changed to a numerical list for easier reference. It should be kept in mind that these are the rankings on Amazon’s US site. JeffConrad (talk) 17:39, 14 May 2011 (UTC)[reply]
          (edit conflict) While I'm waiting to see if I hear back, here's how these ranked among the top 20 in (US) Amazon's most popular items in Writing Reference: #1 APA Style (see article for actual title), #3 Chicago, #6 Turabian's A Manual for Writers of Research Papers ... (a shorter and more digestible version of Chicago), #8 MLA Handbook for Writers of Research Papers, 7th Edition (overlaps the other MLA guide), #14 AP Stylebook (2009 ... probably #14 because most people either access it online or get the updated spiral-bound versions ... this year's version comes out this month), #16 Concise Rules of APA Style. NYTM is generally accessed online, although I admit I still rely on the 2002 book. - Dank (push to talk) 21:50, 13 May 2011 (UTC)[reply]
The Cambridge Grammar of the English Language also has some discussion of punctuation.
It may be of interest for several reasons:
  1. it is not intended as a prescriptive work;
  2. it is not intended to define the house style of a particular publisher.
  3. it is not published by the Oxford University Press;
  4. it is written by linguists—who might be presumed to have a different perspective from, say, teachers, lawyers, or printers.
On the subject of "long hyphens" (referred to in this debate as en dashes, their usual typographical representation), they write

This is used instead of an ordinary syntactic hyphen with modifiers consisting of nouns or proper names where the semantic relationship is "between X and Y" or "from X to Y":

  • a parent–teacher meeting, a French–English dictionary, the 1914–18 war
It can be used with more than two components, as in the London–Paris–Bonn axis. It is also found with adjectives derived from proper names: French–German relations. There is potentially a semantic contrast between the two hyphens – compare, for example, the Llewelyn–Jones company (a partnership) and the Llewelyn-Jones company (with a single compound proper name). This hyphen [i.e. the en dash] is also used in giving spans of page numbers, dates, or the like: pages 23–64; Franz Schubert (1797–1828).
--Boson (talk) 13:38, 14 May 2011 (UTC)[reply]
Right, and if the question is "Is there a split between American and non-American style guides on the question?", you've just given more evidence on the pro side; that's quite different from what the American guides say. - Dank (push to talk) 14:33, 14 May 2011 (UTC)[reply]
But there are a fair number of US guides that support essentially the same usage; admittedly they don’t all top our “bestsellers” list. Even among US guides, I don’t think there’s much dispute about using the en dash for inclusive ranges of numbers (and in most cases, also dates and times), and perhaps also in open compounds. Many of the US guides actually use the en dash in more applications than they discuss (e.g., APA and MLA use them for ranges of numbers). And as nearly as I can tell, CMos is the only one that specifically disavows the sense of between. If we’re making a big deal of the difference between US and non-US guides, I find it curious that several proponents of the former, including Pmanderson, use a spaced single hyphen for a dash, presumably representing a spaced en dash, which is far more common outside the US than within the US.
In any event, unless I’m really missing something, I don’t think we have significant disagreement on more than the sense of between (save perhaps the hyphenists). If that’s indeed the case, I think the discussion would be more productive if we were to focus on that rather than bouncing all over the map. JeffConrad (talk) 18:06, 14 May 2011 (UTC)[reply]
I don't agree. I understand that Arbcom is going to be pushing for discussion in a larger forum, and I'll make the case there. - Dank (push to talk) 18:42, 14 May 2011 (UTC)[reply]
Don’t agree with what? That we disagree only on use of the en dash in the sense of between? If the disagreement is indeed broader, perhaps we should at least try to agree on the matters on which we disagree. Because we seem nowhere near consensus on either of the proposed rewordings for MOS:ENDASH, let’s examine the current wording item by item, and see where we agree and where we disagree. I’ll indicate where I think we stand; if others disagree, perhaps they can so indicate.
  1. Number ranges. No disagreement. This usage is supported by most US and non-US guides alike; do we really take issue?
  2. To or versus (and essentially, between as well). Partial disagreement. This includes at least two distinct uses: between/versus and to. It seems to me that we disagree on perhaps all but the latter sense (e.g., 4–3 win).
  3. And. Disagreement
  4. Separation of items in a list. No disagreement. If I understand this correctly, it’s really just a special case of item 6, a stylistic alternative to an em dash.
  5. In place of a hyphen in open compounds. No disagreement? It’s tough for me to tell.
  6. A stylistic alternative to an en dash. No disagreement
Comment on ENGVAR (or whatever it is)
Many seem to want to see this as “US” vs. “non-US”. For sake of argument, let’s take CMoS as the “top” US reference. CMoS appears to support 1, 2 in the sense of to, and 5; lack of support for 4 and 6 seems to be simply that, for these purposes, CMoS uses the em dash exclusively. The CMoS position on some of the examples in 2 isn’t completely clear; it gives the impression that it would support male–female ratio (to) but that it would not support male–female relations (between). Raising even more questions is 6.81 (16th ed.), which, for certain disciplines, refers the reader to Scientific Style and Format: The CSE Manual for Authors, Editors, And Publishers, which supports almost every use of the en dash in the OSM. In that context, I’m not sure the transatlantic schism is nearly as great as some would maintain. That said, we still clearly disagree on some items. My question then would be “On what basis?” JeffConrad (talk) 20:03, 14 May 2011 (UTC)[reply]
I'm waiting for replies to emails, and I've also got a trip planned to my alma mater, UNC-Chapel Hill, which has a fine journalism school. I'll try to get some answers there, and report back. - Dank (push to talk) 21:23, 14 May 2011 (UTC)[reply]

Guides that prescribe en dashes to connect equal pairs

Many guides support the points in our MOS, including the points numbered 4 and 5 above, which have been the source of the most contention. Some sources reduce them to "equal pairs" and other such simplifications. Dicklyon (talk) 06:16, 8 May 2011 (UTC)[reply]

The section above intentionally surveys a sample of style guides chosen long ago, independently of this question. Presenting these, searched for because they support one position, as parallel is at best misleading; these are a handful of the hundreds of style guides published, not many. And even these show a remarkable predominance of those published by the Oxford University Press, which does not consistently follow them. Septentrionalis PMAnderson 18:01, 8 May 2011 (UTC)[reply]
Yes, these were intentionally picked to show examples that support the en-dash in pairs. You had said there were very few such, and had suggested that those of us who used such en dashes have made up the rules ("invented" I believe you said). These are listed as a sample of where these rules that we grew up on have been taught. Dicklyon (talk) 19:00, 8 May 2011 (UTC)[reply]
And it fails to do so; these are "very few" as compared to the whole number of guides published - and some of them don't support "pairs". Septentrionalis PMAnderson 04:30, 10 May 2011 (UTC)[reply]
  • New Hart's Rules (I just got my copy in the mail) as Pmanderson pointed out before, supports all of the "pair" relationships where "it means roughly to or and", including Permian–Carboniferous boundary, Marxism–Leninism and Marxist–Lenininst theory ("although for adjectives of this sort a hyphen is sometimes used"), and Greek–American negotations vs. Greek-American wife, and specifically French–German vs. Franco-German. Dicklyon (talk) 13:52, 15 May 2011 (UTC)[reply]
  • The various guides written by Bryan A. Garner from 1995 to 2010, including Garner's Modern American Usage and including the editions that Pmanderson refers to as "Oxford" to make them sound less American, which are like our MOS: "Use an en-dash as an equivalent of to (as when showing a span of pages), to express tension or difference, or to denote a pairing in which the elements carry equal weight" and "... wherever movement or tension, rather than cooperation or unity, is felt." Examples include current–voltage characteristic, Marxist–Trotskyite split. Says en dash is preferable to virgule (slash) in disjunctions, situations where a hyphen is clearly wrong, as in possessive–genitive dichotomy. A great review discusses the tension between prescriptive and descriptive and how Garner deals with it: "...descriptive scientific endeavours, investigating and recording the state of the language. That’s essential if we are to know what’s going on in the engine room of linguistic change and invention. But the results often don’t meet the day-to-day needs of those users of English who want to speak and write in a way that is acceptable to educated opinion. To give advice in that situation must be to lay down rules and to say that some common usages are simply wrong. Mr Garner does this. However, he is not a believer in worn-out shibboleths or language superstitions (indeed, he has a section with that heading in which he demolishes the most egregious of them). His article on the split infinitive, for example, the most notorious example of the type, is magisterially even-handed while at the same time practical; he states firmly that no rule exists that says they can’t be split, but that the decision to do so or not depends on the need for clarity, which has to be coupled with a keen ear to avoid clumsy phrasing. He dismisses the canard that you must not start a sentence with a conjunction (which is a good thing for me, since I do it often). He describes the rule that a sentence may not end with a preposition as “spurious”. He is in favour of the serial (or Oxford, or Harvard) comma as an aid to clarity." (my bold). Dicklyon (talk) 00:50, 8 May 2011 (UTC)[reply]
  • Bugs in Writing (1995) – "use en dashes when you have an equal-weighted pair serving as an adjective, such as love–hate relationship." Agrees with our MOS. Dicklyon (talk) 01:09, 8 May 2011 (UTC)[reply]
  • Text, type and style: A compendium of Atlantic usage (1921) – "The en-dash...may stand for the word 'and' or 'to' in such phrases as 'the Radical–Unionist Coalition,' 'the Boston–Hartford Air Line'; 'the period of Republican supremacy, 1860–84'; 'pp. 224–30.'" Agrees with our MOS. Dicklyon (talk) 01:09, 8 May 2011 (UTC)[reply]
    Sigh. I suppose Dicklyon will object if I call this false; but it is.
    • Ivey has one paragraph, describing the use (the only one Ivey recognizes) of an en dash: to represent "and" or "to"; that's only WP:ENDASH 1, although the examples with nouns show where the flakier uses of the Oxford dash come from.
    • This is, therefore, only a small part of what "our MOS" - or rather the present text - says. A footnote recommends not using a dash in New York–Boston Railroad, therefore contradicting WP:ENDASH 5 expressly, for the usual reason: grouping York and Boston is misleading.
    • More importantly, may stand for is not prescribing anything; it's permissive. That being the major difference between "our MOS" (I will gladly concede that it is WP:OWNed by a small body of enthusiasts) says and the reality of the English language, it is the point at issue. Septentrionalis PMAnderson 17:11, 9 May 2011 (UTC)[reply]
  • Fowler's Modern English Usage (1926–2004 editions) (at Amazon) – includes pairs like our MOS (or at least "place pairs in a political context): "the Rome–Berlin axis" and joint-author pairs "the Temple–Hardcastle project". Dicklyon (talk) 01:42, 8 May 2011 (UTC)[reply]
    • As Richard Bentley said to Alexander Pope: "A very pretty poem, Mr. Pope; but you must not call it Homer." That Oxford, as the copyright holders, chose at one point to issue somebody else's style guide under Fowler's name is their betise; but it gives no particular authority to the result. They almost immediately redeemed themselves by issuing a Fowler's written by -well- Fowler (with an updating supplement); we should let their folly rest in merciful oblivion. Septentrionalis PMAnderson 18:01, 8 May 2011 (UTC)[reply]
You’ll find much the same in
  • The Oxford Style Manual – "Use the en rule closed up to express the meaning of to or and between words of equal importance. In these cases the words can be reversed in order without altering the meaning. The hyphen must be used when the first element cannot stand on its own. Dover–Calais crossing, on–off switch, editor–author relationship, Permian–Carboniferous boundary, Ali–Foreman match, dose–response curve, cost–benefit analysis, wave–particle duality"
    • Not "pairs". This is expressly the use of an en dash for "to", which is (as an idea) not seriously disputed save for those who would not use en dashes at all - whether this stretches the idiom beyond its elastic limit is another question. Septentrionalis PMAnderson 04:30, 10 May 2011 (UTC)[reply]
  • The Publication Manual of the American Psychological Association, 6th. ed. – "items of equal weight (e.g., test–retest, male–female, the Chicago–London flight)"
  • Merriam-Webster’s Guide to Punctuation and Style (I have the 1st ed.).
  • Merriam-Webster’s Manual for Writers & Editors – "used to indicate linkages such as boundaries, treaties, and oppositions" (though it also allows hyphen as an alternative in male–female differences)
JeffConrad (talk) 04:45, 8 May 2011 (UTC)[reply]
This is a good example why ArbCom seemed inevitable: WP:TE. See for instance [8],
Emphasis in original. Tijfo098 (talk) 06:33, 8 May 2011 (UTC)[reply]
Did you not see that I already pointed out "(though it also allows hyphen as an alternative in male–female differences)"? I believe that I correctly characterized the "or" in that section about en dash usage, but feel free to elaborate if you think that "as an alternative" doesn't do it. Are you joining with PMA to call me dishonest or contentious every time I try to present information? Sheesh. Dicklyon (talk) 07:10, 8 May 2011 (UTC)[reply]
Then it does not support what MOS now says at all, for that allows no variant. It supports Tony's text very weakly, for the same reason: Tony only permits hyphens under the most stringent conditions: that there be a strong agreement in the sources and consensus to listen to them. Septentrionalis PMAnderson 04:54, 10 May 2011 (UTC)[reply]
Does the latter refer to differences among shemales? Incidentally, the or isn’t emphasis—the book uses italic font in all set-off examples to distinguish the explanatory text from the literal examples; Chicago do the same thing. Joe Ossanna famously described a “pseudo-page transition” rather than a “pseudo–page transition”; I wasted several years of my life trying to figure out what a “pseudo page” was . . . JeffConrad (talk) 07:13, 8 May 2011 (UTC)[reply]
Usually these grammatical ambiguities only trip up the reader for a second or so; are you going for a record? Dicklyon (talk) 07:20, 8 May 2011 (UTC)[reply]
I guess those who never used troff won’t understand the allusion, which was made only in jest. Ossanna’s description would have been equally incomprehensible with an en dash (it apparently threw off the author of every reference book that I ever read, and did the same for the HP-UX support people). To be honest, though, an en dash might have provided a clue to deciphering the explanation. But we′re getting off the topic, because this isn′t an equal-pair example.
As for tripping up the reader, most similar ambiguities are quickly resolvable, but the reader must often backtrack, and this makes for more difficult reading, especially when there are many instances to resolve. JeffConrad (talk) 07:52, 8 May 2011 (UTC)[reply]

Apparently the derogatory comments about my brain and grammar are due the fact that I remind those highly involved here that push-pull (or is that push–pull?) activities like this are WP:TE according to my confused brain. Maybe you should also read what the Arbs said. Anyway, I'll let you guys finish the pinnacle of style & grammar RfC and vote on it when it's done. Tijfo098 (talk) 08:00, 8 May 2011 (UTC)[reply]

OK, I guess we'll just to agree to treat your interjection as a nasty non-sequitur. I thought we were trying to collect data on style guides here, for information purposes; we weren't arguing about it much, were we? Well, yes, PMA did accidentally say that some of the info was a falsehood, but we're used to that. Dicklyon (talk) 08:09, 8 May 2011 (UTC)[reply]


This entire section takes any favorable reference to using a dash for compounding - no matter which class of compounds is under consideration in the guide - and mashes them all together to prove something about "pairs"; which they do not. These guides do not support the same instances of compounding with dashes - and some of them do not support dashes between "pairs" at all.

In short, this is a random assembly of "pro-dash" style guides - a concept almost devoid of meaning, since actual style guides aren't written to be pro-dash or pro-hyphen, but to reocmmend different sets of specific idioms. The only function this could possibly serve in this discussion only to answer the small minority which would remove the en dash from Wikipedia. As for me, I would consider that suggestion if it would bring peace and consensus; but it won't.

Please read (and understand) what your source says - preferably before you cite it. Septentrionalis PMAnderson 04:42, 10 May 2011 (UTC)[reply]

This is a more general problem with the RfC of Tony1 above. It takes the maximal coverage of the cases where en dashes could be used, and makes it the default choice for all of those, asking for extraordinary evidence for alternative typography. Tijfo098 (talk) 06:55, 11 May 2011 (UTC)[reply]
Yes. I support it because with good faith and restraint it might work in practice, not because it is well-put. Septentrionalis PMAnderson 20:12, 11 May 2011 (UTC)[reply]

Oxford dashes

(comments moved out of data section above by Dicklyon; section title by PMAnderson)

The following above have been cherry-picked as the guides, perhaps the only guides, which endorse the Oxford dash. The reason I began with the section of MOS was that it is a selection of style guides not chosen by what system of dash they support. Septentrionalis PMAnderson 05:25, 8 May 2011 (UTC)[reply]

  • (in reference to the comments on Fowler's showing en dash in pairs:)
    • I believe this to be another falsehood; whether it is a mere error or not is not clear to me. I read through several editions of Fowler on this point at the beginning of this discussion.
    • I therefore request page numbers, year, and edition please; I believe you are confounding particularly visible hyphens with dashes; although it is also possible, since Fowler quotes actual news stories and other extracts at length, that it is his sources and not Fowler who is responsible for this.
      • The edition linked above "at Amazon", p.197, the entry for "dash", which says: "dash. Hart's Rules is an excellent guide to the use of this mark of punctuation. What follows is a revised and extended version of Hart. The en-rule (–) is used to: (a) ...numbers and dates... (c) between separate places or areas linked, for example, in a political context, e.g., 'the Rome–Berlin axis'; (d) between the names of joint authors to avoid confusion with the hyphen of a single double-barrelled name...". Dicklyon (talk) 06:16, 8 May 2011 (UTC)[reply]
        • Ah; that is a third text: Burchfield's rewriting of 1996, substituting his own opinions for Fowler's, whereas Gowers had limited his rewriting and did make some effort to indicate whose was whose. That is not the work we refer to, but I know a university library which has a copy buried in the stacks. Referring to an incomplete on-line version of a book avaoible in print is not really verifiability. Septentrionalis PMAnderson 17:46, 8 May 2011 (UTC)[reply]
    • Fowler exists in two different texts (the first edition, which is Fowler's, of 1926, and the second edition, re-edited by Sir Ernest Gowers, of 1965); the first edition was reprinted, photographically, in 1998. Gowers completely rewrote the relevant article; but both editions agree that, except for compounded compounds (WP:ENDASH 5; I'll get back to that), there are three forms for compounds: blackbird, black bird, black-bird; both articles discuss at length where each of the three is used, and come up with different advice. For compounded compounds (Lloyd-George-Winston-Churchill Government), Fowler proposes (what he himself calls "an innovation") that some new symbol be chosen as a sort of super-hyphen (he does not suggest the en dash, although one of his possibilities is a long hyphen), but recommends recasting the sentence; Gowers suggests recasting, a dash, or a virgule [/] about evenly. Septentrionalis PMAnderson 05:25, 8 May 2011 (UTC)[reply]
      • If you "believe [Dicklyon is] confounding particularly visible hyphens with dashes" then you believe he made an error. The word "falsehood" is ambiguous about the cause of why it is false, but if particularly visible hyphens is the cause, then the cause is an error. Art LaPella (talk) 05:44, 8 May 2011 (UTC)[reply]
        • But no such error has been made, and there is no falsehood (and I'm quite accustomed to hearing that from him). OK, so maybe the range of edition dates was wrong; I only checked the current one linked on Amazon and wasn't aware of the edition differences; I have now ordered copies so I can see what the older ones say. PMA says they don't support this stuff about pairs, which is fine. Dicklyon (talk) 06:16, 8 May 2011 (UTC)[reply]
      • It's interesting that the early Fowler's would propose a new mark; the American George Burnham Ives had already published his guide describing the en dash to connect pairs in 1921; he credits Fowler's The King's English for some of his stuff but specifically for this. Anyway, it's good to know that Fowler at least recognized the folly of using hyphen for too many different meanings, introducing ambiguity to inflict on readers. It is fairly to common the see the virgule; one book I looked at (don't recall which now) characterized that as more informal than the en dash; others just say don't do that. Dicklyon (talk) 06:52, 8 May 2011 (UTC)[reply]
        • If I understand the nonce-word pairs correctly, Fowler is not talking about them; he is talking about compounding phrases which are already compounds (and so already have spaces or hyphens). Septentrionalis PMAnderson 22:52, 8 May 2011 (UTC)[reply]
          • That sounds likely for your Fowler. For the 2004 one on Amazon, it's definitely pairs per the quote above. Dicklyon (talk) 23:03, 8 May 2011 (UTC)[reply]
          • That's not Fowler; calling it so is obfuscation. (That OUP started it is another matter.) Septentrionalis PMAnderson 23:41, 8 May 2011 (UTC)[reply]
            • Perhaps one should use the full title of the book, as catalogued: The New Fowler's Modern English Usage, 3rd ed. ". --Boson (talk) 08:42, 9 May 2011 (UTC)[reply]
            • We have previously debated who wrote "Fowler", so does that mean that most "reliable sources" call it "Fowler", like "Webster's" Dictionary and "Roget's" Thesaurus? That sounds like the opposite of the reliable sources argument used for the Mexican-American War. Art LaPella (talk) 14:02, 9 May 2011 (UTC)[reply]
                • Fowler wrote "Fowler's"; most reliable sources don't speak of Burchfield's edition at all (which is why I forgot about it above). OUP has replaced it with Fowler, revived in full, with notes to update it. Septentrionalis PMAnderson 17:40, 9 May 2011 (UTC)[reply]

Discussion

You can start with those at en dash. Tijfo098 (talk) 03:14, 7 May 2011 (UTC)[reply]

I've been looking at those and cleaning up the sourcing and stuff a bit there. You say that "To stand for and between independent elements" is rare, yet it's what I've always been taught; I find it in the various guides by Bryan A. Garner (Oxford and others), except that in the punctuation section that he did for the 16th CMOS he wasn't allowed to go that far, so he added the explanation that "Chicago's sense of the en dash does not extend to between" (to rule out US–Canadian relations in their example of using hyphens instead), like he was apologizing for them being the outlier. I also found my Bugs in Writing and quoted its succinct advice on this, where they're called equal-weighted pairs. Dicklyon (talk) 03:58, 7 May 2011 (UTC)[reply]
Yes, it is found in the minority of style guides that teach Oxford English, although they were advising fewer dashes in the first edition of Hart's Rules in 1893. But that is a minority - and a style guide routinely ignored by its own publisher is a poor guide for general English usage. Some people will have been taught by those for whom Oxford English is the only real English - but most have not been or it would indeed be normal usage. Septentrionalis PMAnderson 17:34, 7 May 2011 (UTC)[reply]
What is this Oxford English that you keep referring to, and why does it appear to be favored by the American writers of guides to American usage? Dicklyon (talk) 19:07, 8 May 2011 (UTC)[reply]
See Oxford English; you wish to have the Manual of Style insist on one of its odder features, where it demonstrably disagrees with both English and American usage. Bryan Garner may well be an American, as our article says; but he was published by Oxford University Press. Septentrionalis PMAnderson 19:28, 8 May 2011 (UTC)[reply]
That's a redirect to Oxford spelling; nothing but spelling there, as far as I can see. As for Garner's, the cited review says "It has a characteristically American directness that contrasts favourably with another work from the same publisher, Oxford University Press — Robert Burchfield’s Third Edition of Fowler’s Modern English Usage, whose comments are more reserved and balanced and which sometimes leave the reader unsure of what is regarded as correct and what isn’t." It doesn't sound like it represents Oxford English, but rather American English. His publisher was the US branch of OUP, in case that lessens your xenophobia. Dicklyon (talk) 20:50, 8 May 2011 (UTC)[reply]
That's a review of style, not substance - and if Garner is more opinionated than Burchfield, who has a note justifying his expressions of opinion, it must be truly abysmal. Septentrionalis PMAnderson 23:43, 8 May 2011 (UTC)[reply]
So who would say is more opinionated and wrong about en dash usage? Me? Or one of these authors of style guides? Or someone else who teaches the use of en dashes in a way that you have missed, or perhaps rejected? Dicklyon (talk) 15:58, 14 May 2011 (UTC)[reply]
As for Dicklyon's recurrent personal attack: No; this isn't xenophobia. I like and practice much of Oxford English; but WP:ENGVAR says that this aspect of it is inappropriate for American subjects. Septentrionalis PMAnderson 17:53, 10 May 2011 (UTC)[reply]
The American usage guides are not really Oxford, was my point, and you calling them that seems to be an attempt to portray them as un-American (even though being American has nothing to do with wikipedia). In this case, I don't see any substantive different between what I was taught in American schools and what the Oxford and Cambridge guides say, though; so I think we ought to stop calling it Oxford English, which is just FUD, if not xenophobia. Dicklyon (talk) 15:58, 14 May 2011 (UTC)[reply]

Has anyone looked at The Elements Of International English Style? would it be useful here (in general, not just limited to the DASH issue)?
— V = IR (Talk • Contribs) 19:35, 8 May 2011 (UTC)[reply]

An interesting book, but geared to writing for the not quite fluent; more use to the Simple English wikipedia than to us. Some of its advice may be helpful there: but Always use short sentences and use hyphens wherever possible (it mentions dashes once; don't use them between sentences) are not helpful here. Septentrionalis PMAnderson 18:23, 9 May 2011 (UTC)[reply]
That book definite does over-simplify, but has one useful concept: when writing for a broad audience that includes non-native-English readers, one should strive for maximum clarity; a step that the recommend for that is liberal hyphenation (not omitting optional clarifying hyphens). That's the same reason that signifying meaning with the proper choice of hyphens and en dashes is recommended in our MOS, as opposed to the common practice of omitting hyphens in fields where compound terms are so well known that insiders will have no chance of being misled by their absence (e.g. small-cell carcinoma). Dicklyon (talk) 15:58, 14 May 2011 (UTC)[reply]

Descriptive accounts of usage

There is no shortage of style guides. I'd like to see some descriptive accounts of how often this or that en dash rule is followed. You might as well add them to our dash article. I for one was only able to find a study accounting for the differences in usage of hyphen and en dash (as a group) versus em dash in Russian vs. English. (ref on Talk:dash; If you're curious: the Russian corpus uses more em dashes, and then English one more hyphens/en dashes.) Tijfo098 (talk) 03:18, 7 May 2011 (UTC)[reply]

Is the Oxford dash useful?

I'm sure it would be illuminating to see how rarely the en dash is used. Nonetheless, there are lots of places where, if it were used, meaning would be greatly clarified, and places where if a hyphen were substituted, meaning would suffer. I just was looking at uses of love–hate relationship and spotted this book with a paper about a "flower-ant love-hate relationship". If you understand what a hyphen means, you'll be wondering what a flower ant is. That's why they actually said a "flower–ant love–hate relationship". If you see the difference, it's immediately clear that it's not about a flower ant, but about a flower–ant relationship. The en dash that they also use in love–hate follows the same grammatical rule, even though in this case the construction is common enough that you probably won't be misled to ponder what kind of hate love hate is. The fact that many writers and even publishers these days ignore this level of English grammar and inflict all manner of bad ambiguity on their readers to figure out is unfortunate. For us to sink to that level would be a real shame, given that there are no real hurdles to doing it right. When I see the number of people who claim they don't care, can't see the difference, never heard of it, etc., it is hard to reconcile that with the energy being poured into doing it wrong. Dicklyon (talk) 06:03, 7 May 2011 (UTC)[reply]
The only places where the Oxford dash clarifies things are where it offers a different meaning than a hyphen: Michelson–Morley and Lloyd-George–Winston-Churchill government; those are also the only cases where its usage does not verge on the insignificant.
Of these, the first is actually useful and used only in some cases; the OED so spells the example, but attests that Morley himself did not. If readers may reasonably be expected to know that the conjoint discoverers are two persons, its utility becomes marginal. The second should be, as we now advise, rewritten where possible; the government headed by Lloyd George and Winston Churchill.
In other cases, it does not add to clarity; it's a typographical tic. Septentrionalis PMAnderson 17:48, 7 May 2011 (UTC)[reply]
You guys are still debating #Style guides here. Tijfo098 (talk) 23:40, 7 May 2011 (UTC)[reply]
We have veered off from the proposed topic (so I've made a new section), but the subject so far has been whether the Oxford dash really clarifies meaning. I think there are two isolated cases where it does; Dicklyon finds it clarifying across the board. Septentrionalis PMAnderson 05:39, 8 May 2011 (UTC)[reply]

Note; so does the following paragraph from Oxford Reference Online:

The shorter en-rule [i.e. shorter than the em-rule] has two principal uses: (1) to separate a range of dates, as in pages 34–6 and the 1939–45 war, and (2) to join the names of joint authors and suchlike, as in the Temple–Hardcastle project and Lloyd–Jones, 1939 (as a citation; Lloyd-Jones, with a hyphen, would be a single double-barrelled name).

Why can't we leave this where the en dash serves some actual purpose? Septentrionalis PMAnderson 04:24, 10 May 2011 (UTC)[reply]

Agreed, with a suitably liberal interpretation of "and suchlike" to mean pairs of interchangeable or equivalent-functioning names or terms, as opposed to the kind of things for which hyphens are used, it serves an actual purpose in signifying that distinction, and there's not any more to it than that. We could combine points 4 and 5 to make that more clear. Like this:
To stand for and, to, versus or suchlike relationship between independent elements in a compound disjunctive or attributive modifier (male–female ratio, 4–3 win, Lincoln–Douglas debate, French–German border, diode–transistor logic, Michelson–Morley experiment, London–Zürich flight, Mexican–American War).
Yes, we could leave as such a shorter description of its actual purpose. Dicklyon (talk) 04:16, 13 May 2011 (UTC)[reply]
As usual, Dicklyon is perfectly willing to agree to have his own way, against consensus and English usage. We are not in the business of indulging his taste for affectations, especially when they serve no encyclopedic purpose; was it Fut. Perf. who found that mot juste?
What he has neglected to do, however, is to show that the same compounds mean anything different with dashes than they would with hyphens.
This lacuna can indeed be filled; the first could mean, by point 6, something that was both a female ratio and a male. What that would be boggles the mind (nor does it require a separate point to justify it); but it would at least be something distinct from a hyphen. Septentrionalis PMAnderson 04:32, 13 May 2011 (UTC)[reply]

A real paraphrase of the Oxford passage quoted would be:

An en dash can be used

  1. [Like an em dash, to indicate an abrupt change of meaning in a sentence, or a parenthetical expression – like this one – in the midst of a sentence. For example, it can be used to format a list of items, like the songs in an album. ]
  2. To separate a range of dates (1939–45) or numbers pp. 34–6; in this construction, it is preferable to a hyphen. By analogy, it can be used to replace "to" in other compounds: Lincoln–Douglas debate, but also Lincoln-Douglas debate.
  3. To join two or more authors or sponsors: Michelson–Morley experiment, Temple–Hardcastle project. Using a hyphen here could be read as a single author with a hyphenated name: Lennard-Jones effect. Again, it can be used to mean "and" in general.
  4. [To join words which are already compounded into a compound: Lloyd-George–Winston-Churchill Government. Lloyd-George-Winston-Churchill Government can be read as four people, or one with a very long name; Lloyd George–Winston Churchill Government suggests three leaders, including the imaginary George Winston. But it is usually preferable to avoid such towering compounds (the Government led by Lloyd George and Winston Churchill).]

I include two more clauses to include the whole range of possibilities in the present text. Septentrionalis PMAnderson 05:14, 13 May 2011 (UTC)[reply]

But if you want to recognize the actual useful purpose of the en dash, you'll want to extend to more "suchlike" relationships, rather than limiting yourself to that abbreviated treatment. E.g.
male–female ratio is a ratio of male to female, whereas male-female ratio suggests a ratio about male females (shemales?).
French–German border represents a border between French and German regions, whereas French-German border a border that is French German.
Lincoln–Douglas debate was between Lincoln and Douglas, but Lincoln-Douglas debate seems to be by a guy named Lincoln-Douglas.
diode–transistor logic represents logic implemented with a combination of diodes and transistor, whereas diode-transistor logic suggests logic using diode transistors, whatever those are.
Mexican–American War suggests a war that was symmetrically Mexican and America, whereas Mexican-American War looks like a war about Mexican Americans.
Leaving these out misses the opportunity to represent the full scope of the usefulness of the en dash to clarify relationships between words that are misleading when set with a hyphen. There is no cost to set these in the clear way, the way suggested by our MOS. Why do you resist clarity so? I've never claimed that such a clear style is widely or uniformly practiced, but it still seems like a good goal for wikipedia, as it has been for the last four or five years. Dicklyon (talk) 06:49, 13 May 2011 (UTC)[reply]
Dicklyon is inventing his own idioms, instead of reading English, again.
  • male-female ratio is not a ratio of male to female; it is a ratio of male population to female population. More importantly, so is male-female ratio [hyphen], which is why most people use it.
French–German border represents a border between French and German regions, whereas French-German border a border that is French German.
  • No. The French Germans (if history acknowledged such people) would be gradually losing their hyphen, as the Asian Americans are. Their border would be the French German border, which will be unambiguous in practice.
Lincoln–Douglas debate was between Lincoln and Douglas, but Lincoln-Douglas debate seems to be by a guy named Lincoln-Douglas.
  • Only in the imagination of a pedant, not in practice.
diode–transistor logic represents logic implemented with a combination of diodes and transistor, whereas diode-transistor logic suggests logic using diode transistors, whatever those are.
  • No, the second would be "diode transistor logic", as (outside the moves of another self-appointed defender of MOS) high dynamic range imaging is imaging with a high dynamic range; noun phrases group to the left, adjectives don't.
Mexican–American War suggests a war that was symmetrically Mexican and America, whereas Mexican-American War looks like a war about Mexican Americans.
  • No, that last (in addition to being a desperate anachronism) would be Mexican American War.
But all of these fantasies beg one enormous question: why don't the vast majority of printed books see it that way? Septentrionalis PMAnderson 17:36, 13 May 2011 (UTC)[reply]
Indeed the French-German border is at once both a French border and a German border, and the Mexican-American War was at once a Mexican war and an American war; on the other hand “border between/of/from (the) French and/to German(s)” can only plausibly refer to a border between languages or ethnicities, not between political entities, and “war between/of/from Mexican(s) and/to American(s)” can plausibly refer to a war between ethnicities but that's not what wars are usually called (cf Polish-Swedish wars, *Pole–Swede wars). A. di M.plédréachtaí 13:43, 13 May 2011 (UTC)[reply]
I think the border means between French territory and German territory, more than referring to the political entities France and Germany per se. In any case, it's commonly done that way, with territories being implicit in border. Similar, people, governments, or armies are implicit in war. Dicklyon (talk) 15:18, 13 May 2011 (UTC)[reply]
The point of the alternate interpretations is that that's how the hyphenated form should be interpreted, but it is the sort of interpretation that is suggested by the hyphen, since the standard use of the hyphen is in converting a compound noun to an adjacent form. A di M doesn't seem to understand this, or he would realize that Asian Americans pick up a hyphen in things like "Asian-American culture"; that's why "French-German border" with a hyphen makes the reader trip over a wrong reading before getting to the right reading. Same with all the other examples; I'm just pointing the kind of wrong readings that the reader will have to discard before getting to the right one, whereas see the less-tight coupling with the en dash would lead the reader directly the right reading. Even readers unfamiliar with the rules benefit from the cue in this evolved typographic style. Dicklyon (talk) 02:30, 14 May 2011 (UTC)[reply]
If I benefited from the cue of dashes before I learned the rules, then I assure you that benefit is repressed deeply into my subconscious (except where hyphens and dashes are mixed together in the same compound word, which is almost unknown in real text). Art LaPella (talk) 03:26, 14 May 2011 (UTC)[reply]
To be honest, I can’t remember whether I even noticed en dashes before I learned the rules. But I learned the rules a long, long time ago, and have been well aware of the distinctions ever since. A “pseudo-page transition” just isn’t the same as a “pseudo–page transition” . . . but then I suppose many would not care. I guess this is to say that in some circumstances, it is useful to some people; to those who don’t notice, it certainly does no harm. JeffConrad (talk) 04:28, 14 May 2011 (UTC)[reply]
My pre-Wikipedia punctuation education was more like this (no dashes even mentioned). A pseudo-page transition is just the same if you don't notice the difference. But dashes cause no harm other than WP:Instruction creep, which wouldn't be so bad if we could make the rules consistent and easier to find. Art LaPella (talk) 05:06, 14 May 2011 (UTC)[reply]
And my pre-CMoS (13th ed.) education was mainly The Elements of Style and whatever I managed to remember from elementary school. Even for many who notice the difference, a “pseudo–page transition” is incomprehensible (if you must know the basis for this esoteric humor, see troff User’s Manual, and search for “pseudo-page transition”). But for those who do know the difference, the en dash arguably provides a clue.
Making the rules easier to find would make big difference, for editors new and experienced alike. With most style guides, it’s often a matter of looking in the index or the TOC; with Wikipedia (and probably any other wiki), this doesn’t necessarily lead the editor to the desired information. But I’m not sure we can fix that here. JeffConrad (talk) 06:05, 14 May 2011 (UTC)[reply]
Better use a more plausible example for 4. I can't see why people wouldn't just leave first names out and say Lloyd George–Churchill Government (with the same typography as in Born–van Karman boundary condition). A. di M.plédréachtaí 14:10, 13 May 2011 (UTC)[reply]
I don't insist on including 4 at all; but it is, very rarely, a useful dodge, as with Hay–Bunau-Varilla Treaty. (Bunau-Varilla, unlike van Karman, is hyphenated.) In general, it should be avoided.
Some comments on the example (which can go, but I think it more reasonable than you suppose):
  • I have that example from Fowler, who is (as usual) commenting on a real example he found in print; so somebody did think Winston worth specifying. (This was when the Conservatives were about to send Churchill to the wilderness; Churchill was not as pellucid then.)
  • More importantly, I look at Lloyd George–Churchill Government and see George Churchill; Born–van Karman boundary condition is a special case: since van is lower case, Born–van is not a plausible compound. Septentrionalis PMAnderson 17:36, 13 May 2011 (UTC)[reply]
Should someone ever name something after someone and me, I'll stop complaining about my last name starting with a small letter. :-) (I've had a few problems with that, as few Italian last names–even if starting with a preposition–do, and I've long stopped caring when other people use a capital letter.) A. di M.plédréachtaí 12:16, 14 May 2011 (UTC)[reply]

No more MOS?

According to Pmanderson here, editing and moving according to the specifications of WP:MOS and WP:TITLE, if it involves en dashes, is now called "POV pushing". I thought we had pretty broad agreement on at least the use of en dashes between two names (discounting those who want to do away with en dashes altogether); is this not the "actual purpose" that Pmanderson said above that en dashes serve? Was I wrong? Or is it now POV pushing to follow the MOS in routine editing on stuff that's non-controversial? Is there someone that thinks putting a hyphen between two names, as in Manley–Rowe relations and Shack–Hartmann is in any sense preferable? Did we abandon the MOS when I wasn't paying attention? Dicklyon (talk) 02:19, 14 May 2011 (UTC)[reply]

I didn't think we had pretty broad agreement; in particular we had no such agreement with Pmanderson. He doesn't like dashes in disjunctions, and the paragraph previous to the "actual purpose" quote is also consistent with not liking dashes in disjunctions. However, "No more MOS" and "abandon the MOS" doesn't represent his opinion either, as both my references to Pmanderson's opinions show he likes dashes in some situations. Art LaPella (talk) 03:17, 14 May 2011 (UTC)[reply]
Only Tony and Dicklyon talk about ending the MOS; I should like to have one. (I also would support Art's idea of making it an essay, if it continues to express solely the opinions of an outspoken minority; but that's accuracy in labelling - and is no more desirable than {{POV}} tags on articles.)
Dicklyon's moves were inappropriate while ArbCom is in the act of a moratorium on dash-based moves; especially since the range of possibilities under discussion include spelling neither with a dash (if B2C's blanket elimination wins consensus) and spelling only one so.
I do support dashes where English usage and utility to the encyclopedia recommend them; not elsewhere. For my part, therefore, a rational section on the subject would begin with the ancient differentiation – hyphens make compounds, dashes are punctuation marking an abrupt shift in the sentence.
One additional sentence would then mention the only function of the en dash which is consensus among style guides: use in a numerical range, as pp. 34–6; this may be a variant of the use as punctuation. Then I would say that some writers use an en-dash to identify multiple authors at Michelson–Morley effect (with the present reason), and to form compounds from compounded words (with example, and warning to recast where possible). I would say no more; many authors and copyeditors manifestly do not use them. This is permission; it does not authorize the trench warfare, or the mass "correction" to a rare and pretentious style, now being employed by the faction of dash enthusiasts. Septentrionalis PMAnderson 06:07, 16 May 2011 (UTC)[reply]

Foreign words in IPA instead of italics

I notice this in several books where words of foreign origin are written in IPA instead of italics. They are written in between slashes. This gives the reader a more detailed romanization that focuses on the standard pronunciation rather than an ad-hoc romanization that may have focused on transliteration. Is such a thing suitable on Wikipedia? --Dara (talk) 18:48, 12 May 2011 (UTC)[reply]

No. I assume that Wikipedia uses IPA because of dogma. I find it meaningless and incomprehensible. I am sure that I am not the only one. The original word in the native script can be useful, and a romanisation that would be recognised by native inhabitants is useful. Please do not sacrifice genuinely useful things on the altar of IPA dogmatism.--Toddy1 (talk) 21:26, 12 May 2011 (UTC)[reply]
Toddy, WP uses IPA for the same reason it use kg and km rather than pounds and miles. We can't accommodate all of our readers' lack of education. It's simply not practical. What would you suggest instead?
Dara, it depends on the context. If we're listing Dostoyevsky's novels or Khayyam's poems—anyplace we'd use the national orthography if the language were French or German—then simple transliteration is appropriate, or italics in the case of a language with a roman orthography. The names of the articles on Dostoyevsky's novels, or Miyazaki's films, are going to be transliterated, not transcribed. If however we have reason to give the pronunciation, of words foreign or English, even if the language is written in the roman alphabet (say, how French eau is pronounced), then the IPA is appropriate and really the only encyclopedic choice. That is, French eau is pronounced [o], not "oh". Generally we'll give roman orthography/transliteration only, or both transliteration/orthography and IPA transcription. See WP:Pronunciation. — kwami (talk) 22:08, 12 May 2011 (UTC)[reply]
(edit conflict) IPA is not used because of "dogma", it is the international standard for pronunciation transcription, and thus we should continue to use it for transcription of pronunciations. However it is not designed to be an all-purpose transcription or translation of foreign words into English and I don't see why we should want to use it as such. I think that there are only three places IPA should normally be encountered:
  • In the leading paragraph of an article/section about a foreign word/subject (e.g. Ouagadougou (/[invalid input: 'icon']ˌwɑːɡəˈdɡ/, Zeitgeist (German pronunciation: [ˈtsaɪtɡaɪst] ))
  • In other parts of an article where the pronunciation is important (e.g. in ough (orthography), articles about languages). Most, but not all of these, of these will be in articles about language and linguistics topics.
  • Where terms are used that have a pronunciation that is not likely to be obvious to native English speakers (or where the obvious pronunciation would be wrong). (e..g. ʻAʻā (also spelled aa, aʻa, ʻaʻa, and a-aa; Template:Pron-en, from Hawaiian meaning "stony rough lava").
In all these cases the IPA should not be a replacement for a romanization but a supplement to it. It should not be used more than once for the same word/term, except where pronunciation is the subject or an essential part of it. There will be exceptions to all these though. Thryduulf (talk) 22:33, 12 May 2011 (UTC)[reply]
Yes, I think that pretty much covers it. Exceptions will be rare, such as unwritten languages where everything is given in IPA. — kwami (talk) 22:38, 12 May 2011 (UTC)[reply]
As often, the use of IPA is well-intentioned, but has become dogma, while not being particularly useful. There was a discussion of this on pi - a word of foreign origin - which missed the endemic problem: IPA is sensitive enough, as well it should be, to represent the difference between British and American pronunciation, so using either IPA will be wrong for half our readers, and unintelligible for many. Pronounce it like "pie" is much more useful, and true everywhere; although less posh. Septentrionalis PMAnderson 01:56, 13 May 2011 (UTC)[reply]
Transcription between slashes are broad, so that /aɪ/ just means “the vowel of eye” regardless of how it's realized; also, I don't think the difference between the realizations of it between North America and the British Isles are larger than than those within North America or within the British Isles, and indeed most dictionaries would use /aɪ/ for any variety of English. What are you thinking about exactly? (Problems of English dialects do occur when the dialects have different sets of phonemes–such as the non-prevocalic R maintained in North America or the father-bother distinction maintained in the British Isles–but our transcription scheme is diaphonemic so that's only an issue with exceptional cases such as tomato, where two transcriptions would be given.) On the other hand, I agree with the removal of the IPA in pi on the grounds that it was unnecessary (redundant with the mention of pie). A. di M.plédréachtaí 02:43, 13 May 2011 (UTC)[reply]
The OED gives /pʌɪ/ as the British, /paɪ/ as the American, pronunciation; I presume they know more about IPA than I do. (Their guide distinguishes this as the difference between fly and buy. Septentrionalis PMAnderson 02:54, 13 May 2011 (UTC)[reply]
The OED used to give /aɪ/, as most other dictionaries still do, and apparently not even John C. Wells has any idea of why they changed.[9] A. di M.plédréachtaí 03:13, 13 May 2011 (UTC)[reply]
I can understand the new IPA for buy, although the distinction seems very faint in my own speech. Come to think of it, that is how I might describe a Southern-accented pi. Noting the difference everywhere it applies would be Too Much Information; but we should not "decide where doctors disagree." Septentrionalis PMAnderson 03:30, 13 May 2011 (UTC)[reply]
Southern US or southern England? Anyway, of course a series of narrow transcriptions is TMI (even within Dublin you can hear [aɪ], [ɑɪ] or [ɒɪ] depending on the social class of the speakers), but in a broad transcription /aɪ/ is perfectly reasonable, being what the overwhelming majority of dictionaries use. A. di M.plédréachtaí 04:26, 14 May 2011 (UTC)[reply]
Darkest Alabama; sorry for the confusion. Septentrionalis PMAnderson 21:14, 14 May 2011 (UTC)[reply]
Sep, there is a nascent distinction between /aɪ/ and /ʌɪ/ for some speakers, for example in spied vs spider. If you're American, you probably rely on this distinction for rider and writer. This comes from [ʌɪ] being used before voiceless consonants. We ignore this, both because it's minor and because we don't have a credible way of documenting the distinction for most of the names we transcribe. I actually agree with you on the Pronounce it like "pie" approach in many cases, and have removed the IPA where I don't see any point to it (for example at C++ and various initialisms). But Pronounce it like "pie" is simply unworkable for all but the simplest situations; beyond that, we need a phonetic alphabet, and on an international stage that means the IPA. — kwami (talk) 07:35, 15 May 2011 (UTC)[reply]
This may seem like a heretical question... but why are we telling readers how to pronounce words at all. Isn't that the job of a dictionary? (and Wikipedia is not a dictionary).
As for the various pronunciation formats, they can be more confusing than helpful... they only help a reader who is already familiar with the format - For example, it does not help the reader to be told that a syllable of a word is pronounced "ʌɪ" if the reader doesn't know how how the IPA symbol "ʌɪ" is pronounced. Blueboar (talk) 13:03, 15 May 2011 (UTC)[reply]
What about words not normally found in dictionaries, such as proper names and technical terms? Also, I can't see the harm in including pronunciations for words with very counter-intuitive pronunciations (say colonel). Also, a template was being designed which causes a tooltip text reading “aɪ as ie in pie” when you hoover on aɪ in a transcription. (What became of it?) A. di M.plédréachtaí 13:23, 15 May 2011 (UTC)[reply]
That does not answer the question... why give pronunciations? Our job is to tell the reader about the significance of the person, place, thing or term. You don't need to know how to pronounce something to find out its significance. So why are we doing so? Blueboar (talk) 22:26, 15 May 2011 (UTC)[reply]
One purpose for an encyclopedia is to provide an introduction, which will allow the reader to pursue the subject in other sources. Failure to provide a pronunciation makes it hard for the reader to pursue the subject in audio sources. Jc3s5h (talk) 22:43, 15 May 2011 (UTC)[reply]
So what?... they can pursue it in written sources. Or give one of those "listen" icons. It doesn't help to give a bunch of symbols the reader does not understand. Blueboar (talk) 00:14, 16 May 2011 (UTC)[reply]
Well, have a look at today's featured article, USS President (1800). You don't need to know how many guns it had, or which day of the month the Little Belt Affair was, to find out its significance, do you? That's a pretty restrictive criterion. I also disagree with the premise: I think that how something is called is pretty much one of the most important things about the topic, and I'd consider pronunciation to be at least as important as spelling. (As for “listen” icons, they have the same problem as narrow transcriptions, that they only cover one accent; a Briton hearing an American pronounce a word rhyming with merry won't be able to know whether they'd pronounce the same word rhyming with merry, Mary or marry in their own accent. Hell, when an American vet student talked about /dɛri kaʊz/ it took me more than a day to realize she wasn't referring to a breed of cattle from Derry, Northern Ireland.) The fact that some readers don't understand IPA is a reason to also give other indications such as rhymes with whatever or Template:IPAc-en or Wikipedia:Pronunciation respelling key, much like the fact that most readers don't use kilometres (55% of them being from the US or the UK) is not a reason to use miles exclusively. A. di M.plédréachtaí 11:27, 16 May 2011 (UTC)[reply]
Feh... I still think adding all that is confusing and crufty... better placed in a dictionary rather than an encyclopedia. If people think it necessary to include pronunciation (and obviously they do) I think it would be better to place it in an infobox or some other unobtrusive place off to the side where it doesn't clutter up the lede. Blueboar (talk) 12:11, 16 May 2011 (UTC)[reply]
I think the pronunciation guides are extremely helpful; I wouldn't object too much to preferring to move them out of the way like you suggest, but I hope we don't try to remove them. ErikHaugen (talk | contribs) 13:09, 16 May 2011 (UTC)[reply]
Until User A. di M mentioned it, I wasn't aware that readers had difficulty with pronouncing words like "president" and "colonel". If users have difficulty with such simple words, they need to go on a language course. I don't think IPA pronunciations will help them. If you can't guess how to pronounce "president" and "colonel" correctly, what hope is there of guessing how to pronounce "tsaɪtɡaɪst"? (Hint: it is pronounced "zeitgeist".)--Toddy1 (talk) 13:15, 16 May 2011 (UTC)[reply]
Who mentioned the word president? I agree that its pronunciation is obvious, the article President, as millions of other articles, doesn't include pronunciation guides, and I don't think anyone would seriously argue for such a guide to be added. As for colonel, I don't remember ever encountering that word in the ten years I studied English at school or in the two advanced English courses I did here in Ireland, and the only reason why I know how that word is pronounced is that by chance I saw a dictionary using colonel in its entry for ɜː in its pronunciation guide and that surprised me, so I'd expect a large majority of non-native English speaker to be unaware of its pronunciation; as for guessing, how the hell is one supposed to guess that, considering that I don't know any other English word where ‹olo› is pronounced that way? As for Zeitgeist, you really think that it's obvious to everybody how the Z should be pronounced, whereas ts wouldn't be? A. di M.plédréachtaí 15:12, 16 May 2011 (UTC)[reply]

My understanding of the conclusion of this debate. Nobody (not even fans of IPA) agrees with the original 'proposal' that Wikipedia should copy the example of some "books where words of foreign origin are written in IPA instead of italics". Everybody who has responded, seems to agree that the actual spelling is necessary, as is a 'romanised' version of words not written in Latin script. Nobody is saying that Wikipedia should stop putting IPA into the introduction of articles; some people can understand IPA, some can't. There may also be benefit in using non-IPA methods of telling the reader how to pronounce a word - a link to Wiktionary might be the answer.--Toddy1 (talk) 13:15, 16 May 2011 (UTC)[reply]

Re: "Nobody is saying that Wikipedia should stop putting IPA into the introduction of articles" ... I am. My feeling is that we shouldn't include pronunciation at all... but if we do, it should go in an infobox or side bar, not in the introduction. However, I recognize that my opinion may not be consensus. Blueboar (talk) 13:24, 16 May 2011 (UTC) Blueboar (talk) 13:24, 16 May 2011 (UTC)[reply]
Yes, you are right. Sorry.--Toddy1 (talk) 13:26, 16 May 2011 (UTC)[reply]
As for “a link to Wiktionary”, again, what about proper names and very technical terms? Do you expect a dictionary to have an entry for Dún Laoghaire or Aegir? A. di M.plédréachtaí 15:12, 16 May 2011 (UTC)[reply]
Whether to include place names (at all/which of several subsets) or not has been a long-running debate in Wiktionary, gradually tending towards more inclusiveness. I've not been active there for a wile so I don't know what the current state of play is. A few, but not many, brand names are included and people's names almost never are, so you cannot reliably link to Wiktionary for these. Additionally the pronunciation given here is just a single one designed to give information about the encyclopaedic topic. At Wiktionary the goal is to give readers information about the lexical unit. To that end the pronunciation sections are sometimes quite extensive, giving sometimes two or three different pronunciation transcriptions (occasionally up to five) with information on syllables, stress, rhymes and homophones; some or all of these are repeated where pronunciations differ by region, context, part of speech, etc. We would do our readers here a disservice if we forced them to look elsewhere for a pronunciation - and how would they know whether something was non-obvious (and thus they need to look it up elsewhere) or not (in which case they don't)? In short, a simple overview of the pronunciation is as much encyclopaedic as it is dictionaric. Thryduulf (talk) 22:08, 18 May 2011 (UTC)[reply]

Odd punctuation issue

I'd like to get some advice on how to deal with the following issue. I am building a table for the British magazine New Worlds, and one row needs to show the range of issues from "December 1967 – January 1968" (that's a single issue) to "July 1968". Normally I'd do a range with an en dash:

December 1967 – January 1968 – July 1968

but that looks pretty weird and is ambiguous in any case. I thought about changing to a slash:

December 1967 / January 1968 – July 1968

but the slash looks odd with the spaces. Eliminating the spaces doesn't help much:

December 1967/January 1968 – July 1968

because "1967/January" is more closely tied together than "December 1967" and "January 1968", which isn't what is needed. This last option is probably the best I can come up with, though. Any better ideas? Mike Christie (talk - contribs - library) 00:24, 13 May 2011 (UTC)[reply]

Parentheses or brackets, to make evident which is the stronger grouping?
[December 1967–January 1968] – July 1968
The lack of spacing is IAR; nobody is likely to see 1967-January as a group. Septentrionalis PMAnderson 01:46, 13 May 2011 (UTC)[reply]
Good point; IAR does seem to apply here. I'm not sure about parens or brackets since it would look out of sync with the rest of the list, but I agree that avoiding spacing on the left side is the answer. Thanks. Mike Christie (talk - contribs - library) 02:33, 13 May 2011 (UTC)[reply]
Personally, I'd go with the unspaced slash. Using a different mark than the dash-for-range makes it clear that it's a different relationship. Definitely an IAR case, though. oknazevad (talk) 03:20, 13 May 2011 (UTC)[reply]
Other solvers should note (as I had not) that [] and () both already have meanings in the table. Septentrionalis PMAnderson 03:22, 13 May 2011 (UTC)[reply]
If the unspaced slash is adopted for December 1967/January 1968, then it I think it should also be adopted for other double-monthed issues such as May–June 1964 and September–October 1969. It's jarring to see both unspaced en dashes and a slash in the same table. Ozob (talk) 11:19, 13 May 2011 (UTC)[reply]
What about “May–June 1964 to March 1967 (142–172)”, “July 1967 to November 1967 (173–177)”, “Dec. 1967 – Jan. 1968 to July 1968 (178–182)”? The abbreviations are so that the ‘names’ of the issues take each roughly the same space. A. di M.plédréachtaí 14:07, 13 May 2011 (UTC)[reply]
I don't know about Mr. Christie, but I find that ambiguous. I would work out that the first is {{May–June 1964} to March 1967}, not {May <1964>–{June 1964 to March 1967}}, but I would have to hesitate. An extra column (first/last/publisher) may be too crowded. Septentrionalis PMAnderson 17:04, 13 May 2011 (UTC)[reply]
I agree on both counts. I think I will go with Ozob's suggestion, above, to use the slash throughout the article to indicate single issues named with multiple months. If I thought there was a lesson to be drawn from this example that could be added to MOSDASH I would suggest it, but I don't see one beyond IAR. Thanks for the help. Mike Christie (talk - contribs - library) 13:15, 14 May 2011 (UTC)[reply]

A need to indicate regional spelling differences within quoted material

I have an article that clearly falls into the US side of the pond in terms of origin, content, and the like, (plus I'm the primary editor on it) so it's fully compliant with US spelling and dating. In the case one specific quote which is coming from a English European source, a critical word of the quote is the UK version, ("marvellous" vs "marvelous") and in context, it stands out as wrong, but completely correct for a UK writer.

If the word or phrase it was in wasn't so critical (obviously here using to express outside opinion) I would rewrite it away to not worry about it. But since I feel it needs to be included as a quote, do I need to add any specific markers (of the {{sic}} variety, but obviously not "sic" itself) to indicate this, or is it ok to leave it as such? Leaving it unadorned as I feel other editors may not recognize the source as foreign and change it, or readers will also ignore the source and see it as an obvious misspelling, given the US-dict used throughout the rest of the article. --MASEM (t) 12:38, 14 May 2011 (UTC)[reply]

Just add a hidden comment such as <!-- This is a direct quotation from a source in British English where "marvellous" is the correct spelling, so leave it alone. --> A. di M.plédréachtaí 13:06, 14 May 2011 (UTC)[reply]
Better to use {{lang}}, as bots and the like will leave it alone. Example: {{lang|en-GB|instalment}}instalment. You colud also use {{not a typo}}. ---— Gadget850 (Ed) talk 13:20, 14 May 2011 (UTC)[reply]
A. di M. has it exactly. Spelling differences like this happen all the time, and while you are not required to give any explanation for them, if you feel someone might "correct" it, go ahead and leave a note. Darkfrog24 (talk) 13:22, 14 May 2011 (UTC)[reply]
Thanks - I marked up the wikitext with both the lang template and a short invisicomment for this. --MASEM (t) 13:25, 14 May 2011 (UTC)[reply]

Arbitration committee motion

Hi folks, as everyone probably realises by now, this motion - Wikipedia:Arbitration/Requests/Case#Motion - is passing, so what we're thinking is setting up a clear page detailing succinctly the remaining bones of contention. I am pleased to see some areas of agreement above, but the lack of structure means that I am confused as to what we have consensus on and what we don't. The idea is some uninvolved admins to act as facilitators in setting up and framing this debate so we can get some quantifiable consensus. Those familiar with the debate thus far could help out preliminarily by listing succinctly and neutrally outstanding issues below (please don't repeat an issue someone else has mentioned!). This will help frame where we go from here over the next 8 weeks. Casliber (talk · contribs) 14:29, 14 May 2011 (UTC)[reply]

I'm involved, obviously, so I won't volunteer to facilitate. But I'll try to help list; please feel free to edit the list or reformat as appropriate! ErikHaugen (talk | contribs) 19:38, 14 May 2011 (UTC)[reply]

Issues to discuss

  • How prescriptive should the MOS be?
    • Can exceptions be made only when subject-specific style guides conflict? eg. perhaps something like Wikipedia:Manual of Style (chemistry) can say a hyphen or dash should be used in such-and-such context if the relevant sources support the guideline. Other examples might be whether medical style guides address how to write "basal-cell carcinoma".
    • Can exceptions be made if sources about a subject tend to use a different style? Should such exceptions be discussed on this page first?
    • Should the MOS be silent about style or just provide exceptions, and as with wp:ENGVAR, editors just be consistent with the style already in the article and only change when there is consensus to do so at that article?
  • Assuming the MOS should prescribe a style, what should it say if different from what it says now?
    • We should probably at least reword a little bit to make it clear what to do at Mexican-American War.
    • <hopefully others can fill in the various issues here>
      • With regard to the current MOS (presumably, MOS:ENDASH), where do we agree and where do we disagree, and why? JeffConrad (talk) 20:10, 14 May 2011 (UTC)[reply]
      • If the MOS should prescribe a style, what is the basis for deciding what should be prescribed?
        • Preponderance of major style guides?
        • Preponderance of high-quality sources?“
        • If either of the above (or combination thereof), how would “preponderance”, “major”, and ”high-quality be determined?
        • If indeed stylistic differences turn out to be a matter of practice in the US vs. practice elsewhere, should the prescribed (or permitted) style treated as simply another example of WP:ENGVAR?
JeffConrad (talk) 20:42, 14 May 2011 (UTC)[reply]


I suggest "A foolish consistency is the hobgolin of little minds/ adored by little statesmen and philosophers and divines." (Emerson). Collect (talk) 21:05, 14 May 2011 (UTC)[reply]

Indeed. Septentrionalis PMAnderson 21:54, 14 May 2011 (UTC)[reply]

Some suggestions:

  • Does MOS have any authority over article titles, given that its chief mention of them is expressly an (outdated) summary of the governing policy, WP:TITLE?
    • Should MOS ever require an article to use a different style than its title?
    • Under what circumstances, if any?
  • What is the value (if any) of encyclopedia-wide consistency in style, given the limitations in WP:CONSISTENCY and the (entirely successful) abdication from consistency in WP:ENGVAR and the discussion of the Oxford, or serial, comma?
  • When should MOS require a style?
    • Should this require consensus of editors?
    • Should this require predominance in published English?
    • Should it require consensus of style guides?
    • Should MOS ever require a style when all three are absent?
  • When should MOS permit a style?
  • When should MOS forbid a style?
More follow. Septentrionalis PMAnderson 21:54, 14 May 2011 (UTC)[reply]

Comments

I would approach it differently. The MOS is a guide, specifying the style, and the principles of style, that wikipedia strives for. It neither forbids nor requires anything, but says what direction editors are encouraged to move in, toward compliance with wikipedia style. It should have little or no interaction with, and certainly no conflict with, WP:TITLE, except in that the title will be styled the same way as the article text, generally (I don't think there has been any actual contention on that point, though the RM process has introduced some annoying discrepancies). The point of the MOS is to lay out a style that will help articles become relatively consistent, and a style that will help articles become very clear and easy to read. That's why the en dash usage guidelines are in there, and why they have been followed in a great majority (I think) of articles to which they apply. The present question should be more narrow: should those usage guidelines be amended? If so, why? If a reason to not suggest en dashes is that these guidelines are not followed in a majority of reliable sources, then we might as well scrap the guideline altogether, and go back to writing on typewriters; but I think we already had an RfC conclude that that's not what we want, so if we can take that off the table, things will be simpler. We got here, I think, because some editors thought the en dashes are never used in "Mexican–American War"; then I pointed out that they are, in about 5% of books and scholarly papers, which is a bit low. So the question became what sort of process or test to use to make legitimate exceptions to the normal MOS usage guidelines when some editors say that the en dash is not right in some context. I (and many others) agreed to a compromise that would write an article-by-article exception into the MOS for this case, even though I didn't think it was needed, since article-by-article exceptions (known as consensus) are pretty standard everywhere. The compromise was about a tie, due both to people who saw no reason to pollute the MOS with such an odd statement, as well as by the hyphen-only contingent and the pilers-on who said they don't really care but wanted to oppose it anyway. OK, so I admit I'm baffled. Is there a larger question here that makes sense, given the years of stability and good service of the MOS, or is it just a question of what amendments, if any, to make to the en dash usage guidelines? I guess there's that philosophical question of whether we really want our guidelines to be based on the principle of clarity and readability, versus just copying what common style guides say. Dicklyon (talk) 22:24, 14 May 2011 (UTC)[reply]

  • MOS neither forbids nor requires anything Well, that is novel. Then your activities on Talk:Mexican-American War (and Tony's and Noetica's) calling for it to "comply with MOS" are unfounded and disruptive. If three or four of you will simply stop, we can drop this entirely; if not, we can return to a conduct discussion. Septentrionalis PMAnderson 18:55, 15 May 2011 (UTC)[reply]
  • So the question became what sort of process or test to use to make legitimate exceptions to the normal MOS usage guidelines when some editors say that the en dash is not right in some context.
    Not at all; the question becomes why we should have a MOS which encourages (or rather demands) a style used in 5% of the reliable sources. No credible reason for this other than WP:ILIKEIT has ever been given. Septentrionalis PMAnderson 18:55, 15 May 2011 (UTC)[reply]
  • to pollute the MOS with such an odd statement
    There are two objections here:
    • the use of un-called for language like pollute, for any purpose, let alone to describe a citation of policy. Please retract, or I will propose a simple solution to this dilemma, which means this page and its advocates need no longer fear the black touch of Wikipedia policy, consensus, nor English usage: let the erring pages depart in peace.
    • It is not an odd statement. If MOS indeed neither requires nor forbids, but encourages, it is for consensus on individual talk pages to decide whether they have been encouraged. Septentrionalis PMAnderson 18:55, 15 May 2011 (UTC)[reply]
  • Article-by-article differences from MOS are not "exemptions"; they are decisions of the community. It is for MOS to conform to them. Septentrionalis PMAnderson 18:55, 15 May 2011 (UTC)[reply]
  • the title will be styled the same way as the article text, generally (I don't think there has been any actual contention on that point, though the RM process has introduced some annoying discrepancies). In other words, Dicklyon and his allies have been able to revert war on the text of Mexican-American War, but not on its title; they are a minority. Since their minority is belligerent enough to have its way on the text, the majority of eight should give way to a factitious minority of four on the title. Septentrionalis PMAnderson 18:55, 15 May 2011 (UTC)[reply]
  • If a reason to not suggest en dashes is that these guidelines are not followed in a majority of reliable sources, then we might as well scrap the guideline altogether, and go back to writing on typewriters
    We are writing on keyboards. But if this is intended to say that the vast majority of reliable sources look as though they were written on typewriters, please look at the sources; they don't. Septentrionalis PMAnderson 18:55, 15 May 2011 (UTC)[reply]
  • we really want our guidelines to be based on the principle of clarity and readability, versus just copying what common style guides say.
    In other words, where Dicklyon and his few friends oppose the vast majority of style guides, it is because these self-appointed experts care about clarity and readability and "common style guides don't". In fact, it is exactly where dashes clarify, by differentiating from hyphens, that style guides recommend them . The dash enthusiasts demand or "encourage" dashes precisely where they make no difference in clarity - and where the result will be less readable to all those of us who have learned to read English, not Dicklyon's fantasy language. Septentrionalis PMAnderson 18:55, 15 May 2011 (UTC)[reply]
  • In fact, only two or three of the forty or so !votes on Tony's draft support only hyphens - as only three or four oppose it by demanding nothing but the Oxford dash. In the real Wikipedia, also, as A. di M. found, only 54% or so of articles use dashes, and most of those are the relatively uncontroversial uses.
  • There is indeed a general question: Why have we tolerated this nonsense and its advocates for five years? Septentrionalis PMAnderson 18:55, 15 May 2011 (UTC)[reply]


As I had implied above, I’m not sure the conflict between “clarity and readability” and extant style guides is nearly as great as some would maintain; if Scientific Format and Style is included for certain disciplines, CMoS and OSM (and except for the en dash as a stylistic alternative to the em dash, the current MOS) are almost in agreement. Those who would eliminate all but hyphens spend most of their time on another planet, and even those who would ban only the en dash (from what I can tell, primarily because they don’t like it) seem to ignore the majority of style guides. Practice, of course, is often another manner, but in many cases it may say more about the practitioner than the practice.
My suggestions immediately above are largely pro forma, primarily in response to what I think the ArbCom decision has called for (if indeed it has really called for anything). I think we should begin with trying to isolate the usages on which there is disagreement, so that we waste no further time on voluminous dissertations on usages for which there really isn’t significant disagreement, and in the process lose the forest in the trees. If it ultimately comes down to WP:IDONTLIKEIT, resolution will of course be difficult. JeffConrad (talk)
Good idea. If you want to start with specific points of disagreement, perhaps look at the articles I moved before the moratorium, and then moved back because they were said to be controversial; I don't understand why they are controversial, so they're a potential conversation starter: Shack-Hartmann (which also needs a move to a noun form), and Manley-Rowe relations. Or if you want to get beyond pairs of names, then take a look at this one that I came upon just now and would have fixed but for the moratorium: Time-frequency analysis for music signal (which should be for music signals, too, right?). Dicklyon (talk) 04:45, 15 May 2011 (UTC)[reply]
That discussion should be at Talk:Time–frequency analysis. But no, it is not obvious. Both time–frequency analysis and time-frequency analysis mean an analysis using time and frequency; but the dash is ambiguous (where the hyphen is not): it could mean ENDASH 5, which would be something which is both time and frequency analysis. I have not checked usage thoroughly, but it is clearly divided. Septentrionalis PMAnderson 17:17, 16 May 2011 (UTC)[reply]
I would put reliable style guides first and style-reliable sources second. (By "style-reliable," I mean that we should acknowledge that not all sources that are reliable for facts are written in good-quality formal English.) Darkfrog24 (talk) 05:11, 15 May 2011 (UTC)[reply]
  • I′ve previously listed points on which I think we disagree. The page moves mentioned would seem to fall under item 3 of the current MOS:ENDASH (with which they seem consistent); assuming the issue assumes that article text would be consistent with the title, the general topic of en-dash use would seem part of the discussion. Hopefully, we leave en dashes out of it, because they’re quite honestly a red herring.
  • I agree that style guides should normally come first, simply because to do otherwise would seem to invite endless exchanges of “my preeminent RS” vs. “your poopy-head RS”. Moreover, I agree that RS isn’t really the right term for published examples, for the reason Darkfrog24 states; that’s why I used “high-quality” rather than “reliable”. I’m not sure “reliable” is the right term for style guides, though, because I’m not sure what the term would mean in that context. Whatever the term, be it “reliable”, “widely accepted”, “authoritative”, or something else, resolution would seem to reduce to some means of agreement on which guides we consider controlling (or at least persuasively guiding). As I’ve said several times, I’d probably start with CMoS; although its position isn’t quite clear on several uses in the current MOS because it doesn’t give comparable examples (e.g., would it accept Michelson–Morely experiment?), but the only point that’s clearly different is the sense of between. As I mentioned, however, if we follow CMoS’s referral to Scientific Format and Style, even that the sense of between becomes acceptable, and there is essentially no conflict with our current MOS. Clearly, however, some participants in this discussion aren’t likely to accept that, for whatever reason, and that’s what we need to work out, preferably with some rational basis. For what it’s worth, a look at Amazon’s UK Web site showed the following rankings:
    1. New Hart’s Rules: The Handbook of Style for Writers and Editors: #8,687
    2. Publication Manual of the American Psychological Association, 6th ed.: #9,308
    3. The Chicago Manual of Style, 16th ed.: #18,273
    4. Modern English Usage, 3rd rev.ed. #20,023
    5. The Elements of Style, 4th ed.: #33,548
    6. The Oxford Style Manual: #74,154
If this has any significance, perhaps the AmE vs. BrE difference isn’t all that great (except that style guides seem to sell far better in the US).
  • If we arrive at an impasse (which I fully expect), we might consider the benefit vs. harm of particular uses. Though as Art noted, there is the risk of slight instruction creep, I honestly see little downside to using en dashes in most of the instances in our current MOS. For those who understand the difference, there often is at least a slight increase in clarity; for those who can’t see the difference, well, there isn’t any difference, so it’s tough to see how there could be any harm. Admittedly, there’s slight additional effort, but editors who insist that it’s too difficult can simply use hyphens, and hopefully more fastidious editors will eventually clean them up, as always has been the case, with other non-keyboard characters as well as with en dashes. Offhand, I can’t recall the last crucifixion for failure to perfectly follow the MOS. JeffConrad (talk) 06:32, 15 May 2011 (UTC)[reply]
Editors who insist on everybody following the few badly selling style guides which support - or sort of support - the Oxford dash (except for the 3rd edition of Fowler, which trades upon Fowler's name to impose styles which he did not support, the ones cited in the post are the bottom of the list of style guides) are no service to the encyclopedia.
We do not object to the Oxford dash because we are lazy - that is another personal attack substituting for an effort at persuasion; we object to it because it is a failed experiment in Language Reform, and it does not communicate with our audience; it is a pointless barrier. Septentrionalis PMAnderson 06:20, 16 May 2011 (UTC)[reply]
Yes, the point of a MoS should be making articles easier to read. But I think it is easier to read a style you'rethe reader is used to reading than a more ‘logical’ one with which you'rethe reader is unfamiliar–otherwise we'd better just switch to Lojban. A. di M.plédréachtaí 06:54, 15 May 2011 (UTC) [amended at 12:31, 15 May 2011 (UTC)][reply]
Be careful where you point that second person . . . I′m used to reading en dashes. In any event, for those for whom there is no discernible difference, there should be no extra difficulty. JeffConrad (talk) 07:30, 15 May 2011 (UTC)[reply]
Yes, en dashes are a distinction without a difference for those who are not familiar with them. And likely to not even be noticed.
One criterion which has not been raised: given that we're a general reference work, should we align with publications targeting an audience familiar with the jargon, such as medical journals, or should we align with publications targeting novices, such as student dictionaries? The latter tend to be more precise in their use of hyphens as well as dashes, but are often minority usage, as people tend to drop hyphens from familiar terms, and terms are used most often by those familiar with them.
As a general reference work, how much value should we place in precision? This is the argument I've seen for the preference of using logical quotation marks on WP. We could just drop out hyphens and dashes altogether, as some texts I've seen tend to do, and let the reader work it out. At what point do we decide it's clear enough and not bother to be precise, or do we value precision in general? — kwami (talk) 07:51, 15 May 2011 (UTC)[reply]
Is there any harm that results from precision? I concede that, except for ranges of numbers, many published works seem to ignore the en dash. But eliminating hyphens and dashes altogether (if I correctly understand that option) would put us at odds with every style guide I′ve ever seen; simply eliminating hyphens in compound adjectives would make for mighty confusing reading, as would using hyphens in place of em dashes. It′s not that the reader couldn’t eventually sort things out, but rather that doing so would entail unnecessary extra work.
Again, I don′t think anyone has suggested chastising editors who use hyphens rather than en dashes. But it truly amazes me that some would chastise editors who make the effort to use the various marks properly. JeffConrad (talk) 08:40, 15 May 2011 (UTC)[reply]
The problem is that there is disagreement over what the "proper" use is. Different style guides give us different advice... so it often comes down to "Style guide X says to use a hyphen in this situation" vs. "Style guide Y says to use a dash". What people are being chastised about is the attempt to force the use of one over the other, and edit warring over the issue. Blueboar (talk) 12:20, 15 May 2011 (UTC)[reply]
By “proper” I guess I simply meant using en dashes rather than hyphens in certain situations, which requires some additional knowledge and slight additional effort. The guides don’t differ nearly as much as some would maintain; the greatest disagreement seems to be on using the en dash in various situations between “equal pairs” of words, but there actually is much support for what the current MOS prescribes. Significantly, CMoS disapproves of the sense of between; however, as I’ve mentioned several times, if you include Scientific Format and Style, to which CMoS directs the reader for certain applications, that difference disappears. It’s clearly impossible for every aspect of the WP MOS to conform to every aspect of every other style guide, but it’s hardly necessary—choosing a reasonable consensus of the major guides should suffice. In some cases, this may entail some arbitrary choices; CMoS acknowledge as much, advising writers in Rome to do as the Romans (i.e., if writing for the New York Times, follow their style guide). JeffConrad (talk) 19:36, 15 May 2011 (UTC)[reply]
That is improper, by the only court which can decide propriety: the consensus of anglophones. We do not beg pardon for "breaching MOS"; we propose to change it - as not consensus, not English, and making Wikipedia look stupidly pretentious. Septentrionalis PMAnderson 06:23, 16 May 2011 (UTC)[reply]
The problem, of course, is in reaching consensus on what constitutes a “consensus of anglophones”; as I’ve repeatedly said, the style guides don’t really differ that much. Now perhaps practice largely ignores style guides, making them essentially irrelevant. Unfortunately, if we discard the guides in favor of a preponderance of the “quality” published sources, we make every decision require exhaustive research and discussion. That time could be far better spent; I thought avoiding such a waste of time was one purpose of a style guide. “Stupidly pretentious”? I′d much prefer it to just plain stupid . . . in any event, labels such as this don’t help discussion at all. I’ve repeatedly suggested that we at least try to agree on the elements of the current MOS:ENDASH with which we disagree, so that we waste no further time hurling broad-brush pejorative labels. So far, there hasn’t been a single comment, and at this rate, I don’t see anything close to a resolution anytime soon. JeffConrad (talk) 15:32, 16 May 2011 (UTC)[reply]
That's because the disagreements on what it should say are rooted in preposterous claims as to what this page should be. (And the rest of Wikipedia can be counted on to provide just plain stupid.) But since you ask, I disagree with
  • Any mention of article titles, which are not the business of this guideline. Keep on topic.
  • The mention of WP:ENDASH 2 and 3 as more than possible styles, used by some authors. (Adding that it may be useful to use Michelson–Morley to distinguish from Lennard-Jones.
  • That the warning on WP:ENDASH is much too weak. It should say to recast where possible.
But that's my disagreements; some people disagree with the whole thing - a minority almost as large as the minority that agrees with the whole thing. Septentrionalis PMAnderson 17:11, 16 May 2011 (UTC)[reply]
  • Would you then possibly have usage in the article text differ from that in the title?
  • With regard to 2 and 3: what about 4–3 win? It’s very common usage, though perhaps less so than ranges of numbers. Because there seems to be partial agreement, should we then separate 2 (and perhaps 3) into separate types of use? JeffConrad (talk) 17:27, 16 May 2011 (UTC)[reply]
  • I would not. That sort of refinement has already led to making six points out of (perhaps) four, and overall has led MOS into an unreadable wilderness. It would be enough to say that, by analogy, en dashes are sometimes used for other compounds.
  • As a writer, I would avoid 4–3 win as a false analogy; it looks too much like the standard use for ranges, and blurs the distinction of meaning. Septentrionalis PMAnderson 21:09, 16 May 2011 (UTC)[reply]
I deny the greater "precision" of logical quotation; the "imprecision" is an illusion. Aesthetic quotation sometimes ends in the compound mark ," or ." which mark the end of quotation and end of clause coming together. This is just as precise; it can confuse those who assume that "logical" quotation is the only method, but that is their inexperience showing. Septentrionalis PMAnderson 06:32, 16 May 2011 (UTC)[reply]
I personally don’t care for “logical” quotation, but recognize that the US (and perhaps Canada) are the only places that use “aesthetic” quotation; perhaps I prefer the latter (at least in most cases) simply because I’ve always done it that way. I’ve always suspected that the requirement for “logical” quotation was slipped in by some OUP agent, but perhaps it’s a necessary concession in recognizing that there actually are English speakers outside the US. JeffConrad (talk) 15:32, 16 May 2011 (UTC)[reply]
A rational Manual of Style would permit both forms of quotation; it might recommend considering a comment in the handful of cases where the difference in one character might make the meaning ambiguous. Septentrionalis PMAnderson 21:09, 16 May 2011 (UTC)[reply]
So that it’s not overlooked: it’s always about the reader. I assume that “precision” as you used it here was confined to dashes and hyphens (though I don’t quite see how jargon enters the picture). In the broader context, we must of course consider the audience and avoid losing them in (often gratuitous) jargon. In some cases, I might even sacrifice precision for accessibility, though I′d probably cover such a deviation with a substantive footnote. JeffConrad (talk) 09:26, 15 May 2011 (UTC)[reply]
(edit conflict)@JeffConrad: I wasn't referring to the dashes in particular, I was referring to the MOS in general (as, if I understand correctly, most of the comments in this sections are). Also, I wasn't addressing anyone in particular by you, it's just that one sounded too stilted and the reader too alliterative in that passage, but I'm changing that anyway. A. di M.plédréachtaí 12:31, 15 May 2011 (UTC)[reply]
My response was facetious—I never thought I was of sufficient importance to be specifically considered. My point was that some readers are used to reading en dashes, and in any event, to those who can’t tell the difference, en dashes and hyphens presumably look the same, and should read with equal ease, right? I had assumed (apparently incorrectly) that you were specifically referring to en dashes, because I thought this discussion was in response to the ArbCom notice; I’ll concede that some of the comments, mine included, do sound more general, perhaps misleadingly so. As far making WP accessible to the reader, I completely agree, as I indicated above. JeffConrad (talk) 19:16, 15 May 2011 (UTC)[reply]
There is a question here: which en-dashes are you used to? Those who are used to dashes in "pages 34–36" stand with the overwhelming majority.
But this issue began with dashes which are far less common. I would permit the 5% who are used to dashes in the unusual cases to write as they choose (being consistent in any given article), and the 95% to do so too. Why should the 95% be compelled (or "encouraged") to follow the tastes of the 5%? Septentrionalis PMAnderson 02:25, 16 May 2011 (UTC)[reply]
I′d say I’m used to en dashes in situations corresponding to items 1 and 5 of the current MOS:ENDASH, and in the sense of to for item 2, as I discuss further above. Style guides are largely in agreement; practice in published sources obviously varies, so numbers like 95% and 5% aren′t especially meaningful without mention of how and from where they were obtained. JeffConrad (talk) 02:49, 16 May 2011 (UTC)[reply]
It's Dicklyon's figure above, for the usage actually at issue; it's not unreasonable, and if anything high. He's looked for English books that employ his favorite dashes, and this is how often he finds them. The samples of #usage far above support this.
But this is a distraction, like the apples of Atalanta: if it were 10% (or even 20%), why should MOS pretend to bind - or "encourage" - the majority (who follow the vast majority of English writing) to follow the minority usage in these matters? Septentrionalis PMAnderson 05:53, 16 May 2011 (UTC)[reply]

Proposed change to Keep markup simple

I propose to change WP:MOS#Keep markup simple by adding

The use of HTML entities instead of the equivalent Unicode character when it would be difficult to visually identify the Unicode character in edit mode is an appropriate use of markup, for example, "&Alpha;" rather than "Α".

at the end of the subsection.

This has been discussed by one other editor (who wants to use a robot to change nearly all HTML entities into Unicode characters) and I at Wikipedia talk:Bot policy#Changing Greek letters from HTML entities to Unicode characters but the Bot Approval Group does not seem interested in addressing policy issues. Jc3s5h (talk) 13:59, 15 May 2011 (UTC)[reply]

I suggest that normal keyboard characters should never be regarded as against any MoS. Collect (talk) 15:38, 15 May 2011 (UTC)[reply]

Anybody who wants to use a bot to change styles is admitting that his preferences do not have consensus, and he wants to do it anyway. That should be handled by discipline; not by this page, which will be distorted by true believers anyway. Septentrionalis PMAnderson 17:51, 15 May 2011 (UTC)[reply]
Then I might be an object of such discipline, as "change styles" might be defined to include my AWB editing. Presumably "change styles" includes changing &ndash; to –, but not changing "kat" to "cat"; somewhere in between. Art LaPella (talk) 18:15, 15 May 2011 (UTC)[reply]
Collect, do you mean popular computer keyboards sold in English-speaking countries? Jc3s5h (talk) 16:38, 15 May 2011 (UTC)[reply]
Nope - I mean keyboards and mappings for keyboards which are normally used. And that means, AFAICT, the "en-dash" stuff is not normally found on any such common keyboards, while accents generally are available, and people whose keyboards have them should reasonably expect that they can be used. Just that nothing must be used which is not readily found on keyboards. Is this clear? Collect (talk) 18:36, 15 May 2011 (UTC)[reply]
En and em dashes have been on all English-language and many other Mac keyboards since 1984. And these symbols can be entered in several other ways by people who don't have a key for them. And nobody is proposing any requirement that an editor enter these characters, so it's not clear what problem you are considering here. Dicklyon (talk) 19:47, 15 May 2011 (UTC)[reply]
Although the MOS sometimes calls for characters not on a standard keyboard, I don’t think we chastise an editor for using a typewriter equivalent. That the MOS calls for an en dash in some situations simply means that it’s perfectly acceptable for another editor (or a bot) to replace a hyphen with an en dash (e.g., in a range of inclusive numbers).
I’m not aware of any popular keyboard that provides accents (or any other non-ASCII character) without some arcane keystroke combination, so they’re no easier to enter than dashes. JeffConrad (talk) 19:51, 15 May 2011 (UTC)[reply]
  • I'd like to see us move away from using HTML entities (or xHTML markup for that matter), as much as possible. The whole "Keep markup simple" idea is paramount, in my view; although, it's important that we don't try to "outlaw" the use of xHTML or HTML entities either. There are a couple of issues here:
    Legibility
    Page source is exactly that. It's a type of source code which generates the page content. That's the main reason that HTML entities and xHTML markup shouldn't be "outlawed", because judicious use of such code makes things easier to understand from a "self-documenting source code" point of view. For that reason I think that the current wording is perfect (in terms of legibility).
    Ease of use
    Using many different types of markup concurrently is a recipe for confusion. I've always felt that, since the platform is MediaWiki, we should strongly prefer wikitext markup over the use of anything else. The effect of that is simple: use templates as much as possible in order to hide other markup such as xHTML and HTML entities. This is the reason that templates such as {{reflist}} exist at all, for example. With this in mind, it's clear that the current guidance is lacking somewhat. I would be completely supportive of the creation of, and advocacy for the use of, templates for each HTML entity.
    — V = IR (Talk • Contribs) 19:38, 15 May 2011 (UTC)[reply]
I think that to extend the topic of the present discussion to source formatting style is likely to be a counter-productive excursion; I agree with you that seeing html junk in wiki source is somewhat annoying, and that templates are better, but I don't see why we should complicate the current style discussion to include source-level preferences. Dicklyon (talk) 19:43, 15 May 2011 (UTC)[reply]
I don't understand Dicklyon's comment; if we are not to consider source-level preferences then the "Keep markup simple" section should be deleted. Jc3s5h (talk) 19:55, 15 May 2011 (UTC)[reply]
I'm confused as well. "source formatting style" is exactly what we're discussing, isn't it?
— V = IR (Talk • Contribs) 20:05, 15 May 2011 (UTC)[reply]
Perhaps Dicklyon missed the talk page header. Art LaPella (talk) 20:36, 15 May 2011 (UTC)[reply]
Yes, sorry, I was confused, thinking this proposal was related to the section above, perhaps because of the mention of en dash. I withdraw my comment; striking it out above. Dicklyon (talk) 00:42, 16 May 2011 (UTC)[reply]

I have no problem with specifically allowing HTML name references to eliminate ambiguity, but I’m sure there are many who will object. There’s even a slight advantage in using such references; because they’re ASCII, they display properly in almost any browser, even if the user has mis-set the encoding, though I think this is usually a pretty minor issue. I agree that we should prefer wiki markup whenever it provides the required functionality. But sometimes, e.g., all but the simplest vertical lists, it’s necessary to use HTML. The same is true for tables; all but the simplest require a combination of wiki markup and HTML. But this is straying a bit from the original question. JeffConrad (talk) 20:47, 15 May 2011 (UTC)[reply]

Right, but it still brings up an important point. we should never try to say that people should not use HTML entities, or other non-wikitext markup. There are cases, such as the two you've mentioned here, where it's easier to use the "foreign" markup, for various (usually technical) reasons. The idea that I'm trying to advance is a prefernce for wiki markup where possible, which you said that you agree with. Something about preferring wikitext markup should be added to Wikipedia:Manual of Style#Keep markup simple I think, which would tangentially address the issue originally raised here (dealing with when to use HTML entities).
— V = IR (Talk • Contribs) 21:00, 15 May 2011 (UTC)[reply]

Allowing people that already know HTML to use some of that experience in the narrower context of Wikipedia editing is good thing in my opinion. In contrast, asking them to use only "Wikimedia markup"—which frankly is not even Wikimedia's because it's just some templates used in en.wp—is just another pointless artificial hurdle thrown at newbies by the regulars. Tijfo098 (talk) 20:54, 15 May 2011 (UTC)[reply]

Taking the flip side of Tijfo098's point, if an editor does not know about HTML entities and learns about them to overcome the problem of visually ambiguous characters, that knowledge can be used elsewhere. A future editor who discovers that {{ALPHA}} is available will have no use for that discovery outside the English Wikipedia. Jc3s5h (talk) 21:10, 15 May 2011 (UTC)[reply]
Eh... it's a nice train of thought, but there's been plenty of study to suggest that the exact opposite is true. There's been a significant amount of coverage on usability issues in The Signpost, for example. Heck, there's a whole WMF based "Usability initiative", and one of their goals is to simplify markup...
— V = IR (Talk • Contribs) 21:42, 15 May 2011 (UTC)[reply]
Wiki markup is nonetheless limited to Wikipedia. Although it and many similar efforts (e.g., BBC for online forums) represent attempts to simplify markup for newcomers, when the functionality approaches that of HTML, so invariably does the complexity, forcing the user to learn many different languages that all do essentially the same thing since they all ultimately map to HTML. Users could alternatively have spent the time learning how to enter non-ASCII characters :-) But I’m getting off the original topic. JeffConrad (talk) 22:17, 15 May 2011 (UTC)[reply]
I hear what you're saying, but... is Wikitext, or even Wikitext + extensions that add xHTML elements + the allowance to use HTML entities, actually approaching the complexity of HTML (to paraphrase what you said above)? More important, Wikitext markup is for use on MediaWiki, which is what Wikipedia is (currently, and for the forseeable future) hosted on. We're not here to "right great wrongs", or anything like that... that we use wikitext is simple reality. I don't see how arguing that we shouldn't prefer it's use when we're able to do so is helpful; indeed, there appears to be broad consensus that we should prefer wikitext (that's what the MoS currently says, after all...).
— V = IR (Talk • Contribs) 06:10, 16 May 2011 (UTC)[reply]
My comment was simply an aside in response to other comments rather than a suggestion that we abandon wiki markup. Does wiki markup actually approach the complexity of HTML? It depends; with judicious use of CSS, HTML can often be simpler. Questions about preference sometimes arise when considering the wiki markup for italics vs. HTML tags such as <em>, <cite>, <def>, <var>, and so on. The wiki markup is obviously simpler, but in being so unfortunately descends into one of the greatest evils ever to beset computer-based document processing: procedural markup, one of the main drivers of CSS. So in “preferring” wiki markup, should we deprecate the descriptive tags? I once used many of them, but gave up after edits with bots, AWB, and similar converted them all to consecutive apostrophes; I concede this makes the markup more consistent, but I’m not sure it’s better.
In any event, I’ve already agreed with preferring wiki markup when it provides essentially the same functionality as alternatives, but making it clear that there is no problem using an alternative that affords a significant advantage. JeffConrad (talk) 15:50, 16 May 2011 (UTC)[reply]

The discussion seems to have concluded, so I request addition of the following, changed slightly from the original to emphasize that html entities are not mandatory, to the end of the WP:MOS#Keep markup simple section:

The use of HTML entities instead of the equivalent Unicode character, when it would be difficult to visually identify the Unicode character in edit mode, is an allowable (but not mandatory) use of markup, for example, &Alpha; rather than Α.

Jc3s5h (talk) 23:51, 18 May 2011 (UTC)[reply]

Pronoun use for vessels

Do we have a section on this in the MoS? I couldn't find it, but I thought it might be in an off-shoot MoS page. I was reading the day's featured article about a U.S. ship, and the pronoun use was always "she". fdsTalk 17:04, 16 May 2011 (UTC)[reply]

IIRC, we do somewhere in a subpage. Both it and she are OK unless an article uses both to refer to the same ship, so if an article is consistent Wikipedia:Requests_for_arbitration/Jguk#Optional styles applies. A. di M.plédréachtaí 17:22, 16 May 2011 (UTC)[reply]
WP:SHE4SHIPS Art LaPella (talk) 20:16, 16 May 2011 (UTC)[reply]
See Wikipedia:WikiProject Ships/Guidelines#Pronouns (permanent link here).

Ships may be referred to either using feminine pronouns ("she", "her") or gender-neutral pronouns ("it", "its"). Either usage is acceptable, but each article should be internally consistent and employ one or the other exclusively. As with all optional styles, articles should not be changed from one style to another unless there is a substantial reason to do so.

Wavelength (talk) 20:55, 16 May 2011 (UTC)[reply]
I have corrected two grammatical errors (a dangling modifier and faulty parallelism).

Ships may be referred to by the use of either feminine pronouns ("she", "her") or gender-neutral pronouns ("it", "its"). Either usage is acceptable, but each article should be internally consistent and employ one or the other exclusively. As with all optional styles, articles should not be changed from one style to another unless there is a substantial reason to do so.

Wavelength (talk) 23:04, 16 May 2011 (UTC)[reply]
Well, I wouldn't consider using an error (dangling modifiers can be ambiguous, but they are not ungrammatical in English and there's no ambiguity in here), but anyway I've made it even simpler: “referred to by either ...”; also gender-neutral pronoun doesn't usually mean it. A. di M.plédréachtaí 16:18, 17 May 2011 (UTC)[reply]

Arbitration motion regarding hyphens and dashes

In relation to a request for arbitration on 5 May 2011, the Arbitration Committee has passed by motion these interim decisions:

  1. Temporary injunction on the article title disputes secondary to hyphen/endash issue:

    There is to be a moratorium on article title changes that are due to hyphen/endash exchange. The only edits allowed will be to create a redirect to the existing article title until the resolution of the debate below.

    All discussions on the subject of En dashes in article titles discussion (interpreted broadly) are subject to civility and 1RR restrictions. Administrators are urged to be proactive in monitoring and assertive in keeping debate civil. Actions requiring clarification can be raised with the Committee on the appropriate subpage.

  2. Motion of instruction to editors involved in dispute:

    Interested parties are instructed to spend from now until 30 May 2011 determining the structure of a discussion on En dashes in article titles to obtain consensus. Note that this can be the continuation of a current discussion or commencement anew. From 30 May 2011, a period of six weeks is granted for the gathering of consensus on the issue. The discussion should be of sufficient structure to allow easy quantification of consensus rather than a large amount of poorly-framed debate. If after two months, a determination isn't realised, a case will be opened and conduct violations will be dealt with severely.

Both provisions were passed 12 to 0 with 1 recusal.

For the Arbitration Committee
AGK [] 21:21, 16 May 2011 (UTC)[reply]

Discuss this
The Arbitrator iridesecent has said, behind the link, that he doesn't care whether we use
  • long rambling discussion in which you all thrash out something you can all live with;
  • an RFA style semi-headcount;
  • a formal Ireland/Republic of Ireland/Ireland (state) style Alternative Vote on all the various options;
  • the toss of a coin or
  • some other process
But we are expected to use one of them. Unless some other suggestion materializes in the next day or so, I will attempt to summarize what most people seem to positively agree with. Neither the people who want to get rid of en dashes nor those who want to always use the Oxford dash will find it their first choice; neither will I. But if nothing happens, the position that there is no consensus, and WP:ENDASH should simply be shelved as non-consensus, will be tempting.
The comments in Signpost suggest how little respect any of this is likely to get if it comes to Arbitration. Septentrionalis PMAnderson 00:59, 19 May 2011 (UTC)[reply]
Our previous attempt at #Issues to discuss and beyond, started out with organized lists of points of disagreement, then reverted to debating the points themselves, and then we got tired of it. I would think if we don't agree on what the points of disagreement are, we should pretty much include everything. To exclude a point of disagreement on the grounds that the other side is wrong, is self-contradictory. There will be plenty of time to re-argue the points. So I welcome another attempt by Septentrionalis or anyone else to get back to an "attempt to summarize what most people seem to positively agree with", and disagree with. That was the first step in the plan. Or are we hoping all this will blow over? I don't think arbitrations blow over. Art LaPella (talk) 02:07, 19 May 2011 (UTC)[reply]
I suggest that we go through the current MOS:ENDASH item by item, much as I tried to do above, and, if necessary, break items into subitems if we agree on some uses but disagree with others. I also suggest that we avoid pointless, capricious labels such as “Oxford dash” because they only serve to polarize.
This would at least serve as a start; it would remain to determine the effect to which whatever we could agree on might work: prescriptive, advisory, or merely descriptive, though I can’t see the point of the last. JeffConrad (talk) 02:20, 19 May 2011 (UTC)[reply]
For me, it's not that I'm necessarily "tired of it" (although there is a bit of that). Arbcom is threatening to deal with things "severely" here, so I'm going to make sure that I'm not involved in this at all.
— V = IR (Talk • Contribs) 02:26, 19 May 2011 (UTC)[reply]