Wikipedia talk:Manual of Style

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

Quotation marks: typewriter vs. typographical

Though this topic has seemingly been beaten to death, a review of the archives suggests that a few aspects either have been given short shrift or have not been addressed at all. So I'll add a few of my observations to the record.

The statement

“There have traditionally been two styles concerning the look of the quotation marks”

is at best disingenuous. Straight quotation marks have traditionally used with typewritten material, because there was no alternative; typeset material, however, has traditionally used, unsurprisingly, typographical quotation marks, and has done so for hundreds of years. With typewritten material, the choice of glyphs was limited to keys on a typewriter, but with the advent of laser printers in the mind 1980s, the choices expanded to include symbols, such as opening and closing quotation marks, varying horizontal punctuation (e.g., hyphen, en dash, em dash, minus sign, and similar).

One characteristic of any font is that the glyphs are designed to harmonize. On a typewriter, straight quotes were no more unattractive than other glyphs; with typeset material (in essence, anything using proportionally spaced fonts), the clash between straight quotation marks and the other glyphs is quite noticeable, especially with serif typefaces. Although the clash may be less obvious with the default WP sans-serif typeface, it's glaring if a user has specified a serif typeface for printing, display, or both in her .css file.

Various arguments have been advanced for recommending straight quotation marks in WP: that they're easier to type in both edits and searches, and that typographical quotation marks don't display properly in some browsers. There is some validity to the former, but is the latter really an issue in 2010? For the former: why are quotation marks any different from any other non-ASCII character? In particular, I note

“Do not use substitutes for em or en dashes, such as the combination of two hyphens (--). These were typewriter approximations.”

It seems to me that the same is true for straight quotation marks. If the objective is to avoid non-ASCII characters, then that policy should be applied uniformly. And WP might as well use a monospaced typeface. And forget mathematics.

Perhaps I would not go so far as to require the use of typographical quotation marks, but I certainly would would make them at least as acceptable as straight quotation marks, and I'd probably second many previous suggestions to having no objection to converting straight quotation marks to typographical quotation marks as long as it was done consistently to an entire article.

Again, I simply do not understand why WP seems to treat quotation marks differently from many other non-ASCII characters that are commonly (and rightfully) used in articles. JeffConrad (talk) 10:09, 3 December 2010 (UTC)

• Indeed. This has been discussed umpteen times, but the consensus as always been against curly quotes, though I can't remember anyone giving a reason against them which wouldn't also apply to dashes or other non-ASCII characters. A. di M. (talk) 12:31, 3 December 2010 (UTC)
• The idea that curly quotation marks wouldn't display properly in older browsers is the most valid reason for banning them. However, because this is a technological reason, it is right and proper to periodically reevaluate it and see if it is time for the ban to be removed. While browsers have come a long way, we must remember that many of Wikipedia's readers are not from wealthy first-world countries that are always using the newest software. And for the record, I prefer the look of straight quotes. Aesthetics are in the eye of the beholder. Darkfrog24 (talk) 14:04, 3 December 2010 (UTC)

  The curly quotes are in the WGL-4, MES-1 and MES-2, so they are likely to be found in the default fonts of any computer less than about a decade old. Any font lacking appropriate glyphs for ‘ ’ “ ” is likely to lack appropriate glyphs also for – — and lots of non-Latin characters. (I don't think the browser itself has any direct role in it, except the fact that some but not all browsers are able to choose a fallback font if the default font lacks the glyph for a character.) A. di M. (talk) 14:59, 3 December 2010 (UTC)

I prefer the look of curlies, but making them consistent within an article would be a pain. How do we determine whether it's viable to allow them in terms of font choice? Tony (talk) 15:13, 3 December 2010 (UTC)

Tony, I don’t quite understand what you mean by “whether it's viable to allow them in terms of font choice”. Are you saying that there may be problems if a user chooses a certain font?

Compatibility with older browsers/fonts: again, if the system can’t display opening and closing quotation marks, it probably can’t display most other non-ASCII characters, either, so an argument based solely on quotation marks is unpersuasive.

Consistency within an article: I’ll concede that this can be an issue, but I don’t think it’s any greater an issue than consistency of use for other non-ASCII characters, or for consistency in many other areas that have nothing to do with punctuation. I suppose the case could be made that, in many articles, quotation marks appear more frequently than other non-ASCII characters and consequently present a greater challenge, but I am nonetheless unpersuaded.

Personal preference: I happen to prefer the look of typographical quotation marks, especially with serif typefaces, but I think the issue is more one of conformance to practice that’s been established for hundreds of years than of my, Tony’s, Darkfrog24’s, or anyone else’s personal preference. The current wording in the MoS is misleading because it implies that there is a long-standing tradition of using straight and typographical quotation marks interchangeably, without regard to the typeface; this simply isn’t true. Straight quotes weren’t common (if they were used at all) until the arrival of the typewriter, and the only reason then was the limited number of keys. Although the typewriter was a useful device for a bit over a hundred years, I think it’s fairly safe to say that its era is essentially over. JeffConrad (talk) 00:03, 4 December 2010 (UTC)

In typesetting (traditional publishing), curly quotes are of course established and the best choice. However, on typical computer screens, curly quotes are highly dubious: it is only on really good displays that there is any clear difference, let alone benefit. I support straight quotes for clarity on typical screens, for simplicity and consistency, and for the searching factor. Another weak reason to keep straight quotes is to avoid the agonising disruption from bot-like editors zapping hundreds of articles, with fight back from some editors citing "don't change established style". Johnuniq (talk) 01:02, 4 December 2010 (UTC)

I suggest that—in regard to the abstract question "which should be used in an ideal world"?—relying on people's opinon as to which looks better will be problematic at Wikipedia. There will never be an agreement on this here. In my opinion, the answer to this question should be answered by what experts say on the matter. Miles Tinker tells us in his landmark work, Legibility of Print, that "Mere opinions are not always safe guides to legibility of print" (p. 50). However, there is also the question of what is realistically possible and best for Wikipedia users. If the technological issues with using (or changing to) typographically correct quotation marks are pressing, that should override the "ideal world" expert opinion. My two cents is to keep the straight quotes until it is no longer an issue to change to typographical quotes, and then to rely on experts as to what will create a professional product for Featured/Good articles in Wikipedia—our best articles. Consider that in 5–10 years, the number of FAs and GAs will (hopefully) be much increased. I, for one, would like readers to see "best practice" (professional-looking products) in the English language, as agreed on here by our editors, but informed by the experts. Just my thoughts. --Airborne84 (talk) 01:36, 4 December 2010 (UTC)

I edited the passages in question to a (hopefully) more neutral wording. I hope the new wording is acceptable to the editors here. --Airborne84 (talk) 01:44, 4 December 2010 (UTC)

Johnuniq, I strongly disagree that “it is only on really good displays that there is any clear difference”; the differences have been mighty obvious to me on every monitor I've had for the last 25 years. I’ll concede that the differences are less obvious with a sans-serif typeface like the WP default, but the differences are far more pronounced with a serif typeface, which any user can specify by choosing a different WP skin. And it’s conceivable that WP could someday change to use a serif typeface for the default. Choosing a glyph strictly by appearance is usually short sighted: an obvious example would be treating a lowercase l as interchangeable with the numeral 1, which works in some fonts but not others.

Airborne84, I completely agree that, for the most part, editors’ mere opinions are largely irrelevant. I tend to follow established practice because

1. The people who developed it usually weren’t blockheads, especially with regard to typography. To me, it would seem sensible to defer to hundreds of years of experience.
2. I have better things to do with my time than reinvent the wheel, especially when it’s to little or arguably adverse effect.

As for “technological issues”, it’s far from clear to me just what they are. I’m sure if we try hard enough, we can find combinations of browsers/fonts/whatever that cannot display typographical quotation marks. Does that mean that we should avoid anything not supported by Lynx?

I should add here that Lynx handles handles typographical quotation marks just fine, displaying them as ASCII quotes. JeffConrad (talk) 03:56, 4 December 2010 (UTC)

And once again, it makes absolutely no sense to raise the display issue for typographical quotation marks but not for other non-ASCII characters. If we want WP to display properly on a Teletype 33, we should deprecate all non-ASCII characters. JeffConrad (talk) 02:47, 4 December 2010 (UTC)

Tony says that making curly quotes consistent within each article would be a pain. Would it be appropriate to add a line to the MoS advising editors to only use curly quotes if willing to put in the effort to check the rest of the article? We would have to be careful not to make it sound like permission to go in and change established styles from straight to curly without discussion. Darkfrog24 (talk) 00:03, 6 December 2010 (UTC)

If we're going to do this, then yes, I think that would be appropriate.

There's one other argument against curly quotes that we should discuss: Searchability. It's harder to search for curly quotes because they're not as easy to type; the average person won't know how to type them at all. I prefer curly quotes even with this problem, but I think it's worth a mention. Ozob (talk) 00:08, 6 December 2010 (UTC)

Same applies to dashes, or indeed to most non-ASCII characters. A. di M. (talk) 00:14, 6 December 2010 (UTC)

I agree with Darkfrog24 that any editor using typographical quotes should ensure that they're used consistently and properly. Though I’ve suggested that it should not be objectionable to change straight quotes to typographical quotes, an argument can be made for not changing an established style without prior discussion (though I think the case for stare decisis here is much weaker than for most other stylistic issues).

As has been suggested in some of the archived discussions, an article would ideally be coded with typographical quotes (like any other non-ASCII characters), and if desired, converted to ASCII approximations (in this case, straight quotes), much as Lynx does. AT&T’s troff/nroff worked much the same way: the former generated output for a typesetter-type device, and the latter generated output for a typewriter-type device, using ASCII approximations wherever possible; in most cases, the same source could be sent through either formatter to produce results appropriate for the output device. The issue here is slightly different (aesthetic preference rather than device compatibility), but it should be amenable to the same approach.

The issue of searchability is real, though I honestly wonder how often it comes up (I’m not sure I’ve ever done a search within an article that included quotes). Again, of course, as A. di M. notes, there is the same issue with any non-ASCII character, including accented characters, so I can’t see making a big deal out of it solely for quotes.

It’s also true that entry requires a bit more effort; this is especially true for apostrophes because the current edit bar only allows entry of single-quote pairs. Perhaps the ultimate answer is some additional WP markup specifically for these characters, as has been suggested for nonbreaking spaces. I don't have much experience with TeX, but later versions of troff handled this by recognizing `` as “ and '' as ”; the latter obviously wouldn’t work here given the WP idiom for italics, but a similar approach might be OK (the markup could be converted to Unicode upon preview or saving). Another option would be to have opening and closing single quotes as additional entries under the edit window. I’m personally fine with HTML entities, but I’ve encountered objections from other editors, so I usually enter the Unicode characters.

I assume that, in most cases, font and browser compatibility is not an issue; I’ve used typographical quotes for a dozen years with no problem even in Netscape 4; the only issue with Netscape was that it didn’t recognize the mnemonic, so the Unicode character codes (e.g., “/”) had to be used. A quick glance at List of XML and HTML character entity references should indicate what displays and what doesn’t; the only problem I can find is that Opera doesn’t display ensp, emsp, or thinsp. I used the latest Windows versions; I’m using XP, so I can’t test IE9, but I assume it would be at least as good as IE8.

I think we should change the wording for quotes in markup to something like

Do not use curly quotes in HTML markup (such as <ref name="xxx"/>) because they usually will not work.

We also might want to mention that that straight quotes are not proper substitutes for prime marks (′ and ″) used to indicate feet/inches and minutes/seconds.

Ultimately, the HTML <Q> element might be the best approach. Unfortunately, present support in most browsers is very poor, and WP apparently doesn’t even recognize it. JeffConrad (talk) 02:41, 6 December 2010 (UTC)

Nit: "straight quotes are not proper substitutes for prime marks (′ and ″) used to indicate feet/inches and minutes/seconds". But such prime marks shouldn't be used in the first place, according to MOS:NUM#Unit symbols. Art LaPella (talk) 02:57, 6 December 2010 (UTC)

Oops, not for feet and inches anyway. Art LaPella (talk) 03:37, 6 December 2010 (UTC)

In general, I agree. But I think there are some exceptions that weren’t considered when that section was written. For example, giving a measurement as 8′ 3″ rather than 8 ft 3 in and perhaps perhaps describing a photographic format as 8″ ×10″ rather than 8 in × 10 in. Personally, I’d do the latter as simply 8×10, but some editors have objected. But I certainly agree that in most cases, 8′ is not a proper substitute for 8 ft. And for a conventional measurement, I’d go with 8 in × 10 in. JeffConrad (talk) 04:45, 6 December 2010 (UTC)

On further thought, in most cases I’d probably go with 13 ft 6 in rather than 13′ 6″ (and in most cases, I’d prefer not to mix units). So while there may be a few exceptions for feet and inches, they’re probably few and far between. JeffConrad (talk) 05:19, 6 December 2010 (UTC)

Though I’ve suggested that it should not be objectionable to change straight quotes to typographical quotes...

HA HA HA HA HA HA HA HA HA!! Holy heck, you're killing me. This is Wikipedia. Someone somewhere will object to it. Yes, I agree with your follow-up that we should keep to the established rule of not changing styles without at least a talk page heads-up first. Darkfrog24 (talk) 05:16, 6 December 2010 (UTC)

I’ve been to Tweakipedia and back . . . but honestly, someone will object to almost anything almost anywhere. I certainly did not mean to equate objectionable with “no one would object”. JeffConrad (talk) 05:27, 6 December 2010 (UTC)

• IMHO, the use of typographers quotes should be allowed for those shepherding editors who want to take the time to use them—which is only too easy for Mac users. They look far superior to straight quotes. That means that editors would not be required to use them. That would mean that when editors who eschew the complexity of typographers quotes want to type some text that includes quotes, they are perfectly free to use straight quotes. Sometimes, articles have shepherding authors who sweat such details and ensure their articles are refined and harmonious with consistent use of typographers quotes. A proper guideline, IMO, would also mean that editors should not go through articles and change a bunch of typographers quotes to straight quotes just because they added a set of straight quotes. Greg L (talk) 21:47, 6 December 2010 (UTC)

Quotation mark glyph straw poll

Since no-one seems to strongly object to curly quotes any more (except for reasons which would apply for pretty much any non-ASCII character), I'd propose to replace:

The exclusive use of straight quotation marks and apostrophes (see preceding section) is recommended. They are easier to type in reliably, and to edit. Mixed use interferes with some searches, such as those using the browser's search facility (a search for Alzheimer's disease could fail to find Alzheimer’s disease and vice versa). Furthermore, HTML elements (such as <ref name="xxx"/>) may not always work if curly quotation marks are used.

with:

Straight and curly quotation marks should not be both used in the same article. (This applies to quotation marks displayed in the text of the article: in HTML elements such as <ref name="xxx"/> only straight quotation marks can be used.)

Since this is a significant change, I'm going to advertise this discussion to the Village Pump. A. di M. (talk) 11:58, 7 December 2010 (UTC)

• The MoS should indicate which is preferred for new articles. It should also have something to say about changing existing articles. My vote is to prefer typographical, per almost all quality sources. Typographical characters are made easily available to all via the Insert bar which WP renders below the edit pane. — Wrapped in Grey (talk) 12:43, 7 December 2010 (UTC)
• Let me be the first to oppose. As the current MOS already states; typability is the main concern. How do you even type them? It's easy enough in Word, where they are converted automatically. But here I would need to look up the character codes, copy paste them from another page, or use charmap.exe to look them up, only to find out that there are even more variants available (just look in the edit toolbar; I count 10 possibilities), and wondering which one to choose because none of them look like the ones on the article. Add to that that curly quotes are less eligable on screen then they do on paper, and they are bound to spark editors (like me) into "correcting" them purely for readability. Basically, this is editing hell. — Edokter • Talk • 13:00, 7 December 2010 (UTC)

  ‘How do you even type them?’ Well, how do you even type en dashes and em dashes? A. di M. (talk) 14:13, 7 December 2010 (UTC)

• I also have to oppose this. Curly quotes are easy to type on a Mac – “like” ‘so’ – but are impossible read in the edit window and I don't know how many editors will now the key combinations (option-[, option-shift-[, etc.). They are similarly illegible in the edittools, especially an issue as there are a lot of them there. I would also raise the concern that as well as breaking Wiki markup and search it also could break links to article titles and section headings with quotation marks in, such as to this section of this article. This could happen either if a heading were changed from one style to another, or perhaps more likely, if an article used curly quotes but the editor linking to it didn't notice or was unable to use curly quotes so used straight (an editor can always copy and paste but not everyone thinks to do that). As for "Straight and curly quotation marks should not be both used in the same article" this is redundant as it's covered by the guideline on internal consistency, while changing existing articles is already covered by the next guideline.--JohnBlackburne^words_deeds 13:54, 7 December 2010 (UTC)

  If you can’t type them, just click the handy ‘buttons’ (I just did) that WP conveniently renders below the edit pane. The reduced readability argument seems spurious since other commonly-used glyphs such as commas and semi-colons contain the same elements: , ; ” ’. There’d be no need to require compliance; bots/scripts can be used to fix things up in most cases. And btw, they render fine in the edit window for me. — Wrapped in Grey (talk) 14:05, 7 December 2010 (UTC)

  Ah, I think you referred to my “handy ‘buttons’ ” as the edittools and complained of their illegibility—that’s not the fault of the glyphs therein; the edittools could be rendered in a larger font (or you could zoom your browser window a little). — Wrapped in Grey (talk) 14:27, 7 December 2010 (UTC)

  (edit conflict)They render fine but are too small to be readable. This is not an issue with the comma or semicolon as there's no other similarly shaped glyph used there. They are very unlikely to be confused with the similarly sized and placed full stop or colon as apart from being different shapes they are used in very different ways. To my eyes even in an article ” and " are very difficult to tell apart: it's only the opening “ that makes it clear which sort of quotes are being used. I suspect it's a display issue: a computer screen at 100dpi is no match for printed text at 1000dpi or more. And, yes, I could zoom the window but that makes the text too large for either reading or editing.--JohnBlackburne^words_deeds 14:37, 7 December 2010 (UTC)

• Oppose. Curly quotes are a pain in the neck to enter. They will not be used by most editors even in articles that have supposedly been converted to use that style. 86.173.169.220 (talk) 21:04, 7 December 2010 (UTC)
• Strongest possible oppose No, absolutely not. We have a uniform, or very near uniform style, when it comes to quotes. Allowing them will create zillions of problems and incredible amounts of both inter- and intra-article inconsistencies, none of which correctable through semi-automated editing means like AutoWikiBrowser, or automated editing (BOTS). They're even more of a pain in the ass to input than endashes and emdashes (for most editors at least), and, unlike emdashes and endashes, serve no different purpose than straight quotes. Bob said "Hi Jim". is the same as Bob said “Hi Jim”., but Bob was thinking-as bewildering as the thought of Bob thinking might be-that it just might be possible is not the same as Bob was thinking—as bewildering as the thought of Bob thinking might be—that it just might be possible. The former is just plain wrong, as using periods instead of commas. Curly quotes would be an utter blight on Wikipedia. NO! Headbomb {talk / contribs / physics / books} 21:22, 7 December 2010 (UTC)

• Support, for several reasons:

• I first note, like Wrapped in Grey, that typographical quotes are used in almost all quality sources, and were used exclusively for hundreds of years until the invention of the typewriter in 1870. Like any other glyph in a font, they are designed to visually harmonize with the set; that just isn’t the case with ASCII quotes.
• A few editors have indicated that they prefer straight quotes. While that's certainly a preference to which they’re entitled, it’s hard for me to believe that it represents a consensus of most editors. Perhaps clues can be found in templates such a {{Cquote2}}, which displays

  like this

If straight quotes were really preferable, templates such as this would use them.

• The example above shows quotes from a serif typeface, in which the differences between typographical quotes and straight quotes are more pronounced. The current default typeface for WP is sans-serif, but a user can specify a serif typeface for printing, display or both. And it’s conceivable that at some time WP could change to a serif face as the default, so judging solely by current appearance would seem shortsighted.
• I simply do not understand the purported difficulties in entering typographical quotes; I've used them in many articles, and it’s simply a non-issue (I usually highlight a quoted passage an then enter the quotes using the link below the edit box). Incidentally, I honor the existing style of an article that already uses straight quotes. There are several ways of entering typographical quotes

1. Using the links below the edit box, as I mentioned above.
2. Entering the character codes with either the Windows or Macintosh facilities. I think it should be borne in mind that we’re talking about four codes to handle quotes, so I don’t really see the need to resort to charmap.exe or something similar. I’ll concede that I sometimes do need to use that program (or the special links below the edit box in WP) for some of the more esoteric characters, but not for common ones like quotes or dashes.
3. Using the HTML entities (e.g., “).

• There probably is some validity to concerns about intra-article consistency; in articles I’ve worked on that use typographical quotes, I’ve sometimes had to change a few straight quotes to typographical quotes, but this really isn’t a big deal. And the problem is hardly confined to quotes; I see it with other symbols such as multiplication signs, minus signs, primes, and so on. And such fixes are usually trivial compared to fixing a reference that doesn’t match the prevailing format.
• Perhaps I’m missing something, but I don’t seem to experience the display problems, either in the edit box or in the article text, mentioned by some editors.
• I had thought that the main reason to have devices like laser printers and graphical monitors was to have output that looked typeset rather than just typed. There is no free lunch, of course; achieving this does sometimes require a few extra keystrokes, but I’ve just not found it to be a big deal. When laser printers first became commonly available about 25 years ago, the company I was with faced a similar transition from pure ASCII output to essentially typeset output, and getting people to use typographical quotes, proper dashes, minus and multiplication signs was simply a non-issue, so I’m somewhat baffled that it’s a problem 25 years later. Perhaps it’s just me . . .
• I’ll return to an observation that several of us have made repeatedly—it makes little sense to treat quotes any differently from any other non-ASCII character of which WP articles use many. It seems to me that if we want to stick to characters on a QWERTY keyboard, we should do so consistently, using two hyphens (“--”) for em dashes, x for multiplication, a hyphen for a minus sign, (R) rather than ®. and so on. I doubt we’d seriously consider doing this; yet these characters are just as difficult to enter as typographical quotes. I’ll concede that quotes are entered more frequently than most of these other character; the most bothersome for me is the apostrophe, because I use it so frequently.

Ultimately, I think the best argument for at least allowing typographical quotes is that it follows standard practice for typeset material. And again, it makes no sense to insist on directly keyable ASCII characters for quotes yet also insist on the proper characters for everything else. I cannot see any reason to return to the days of line printers and alphanumeric terminals, but if that’s what’s wanted, it should be conscious decision that’s applied consistently.

Finally, I think it should be noted that we’re talking about allowing (which technically we currently do anyway), rather than mandating (except for intra-article consistency), the use of typographical quotes, and forbidding, rather than simply discouraging the use of typographical quotes in HTML markup. For consistency toward that end, I would remove the Not recommended wording as well. JeffConrad (talk) 00:59, 8 December 2010 (UTC)

In response to your point about fonts: that's irrelevant. It is not shortsighted to make a decision based on the current Wikipedia font; it is silly to make a decision based on the possible situation of Wikipedia changing fonts. If Wikipedia does change fonts, then the quotation mark conversations can be started again. Till then, it is an irrelevant argument. McLerristarr | Mclay1 10:41, 8 December 2010 (UTC)

But different browsers choose different fonts based on what’s available, so what looks right on one user’s display may look entirely different on another’s. Moreover, a user can specify a different default font for display, printing, or both. So I’ll stick by my point: coding solely on the basis of appearance (bet it quotes or anything else is nearly always a mistake). Content-based coding will usually adjust to something that’s at least reasonable under most conditions. JeffConrad (talk) 18:29, 8 December 2010 (UTC)

• Oppose Wikipedia is not Microsoft Word. --Cybercobra (talk) 02:48, 8 December 2010 (UTC)

  I'm sure that many would agree with you; however, no-one has suggested that it is. — Wrapped in Grey (talk) 09:34, 8 December 2010 (UTC)

• Oppose They are a lot of hassle for editors for almost no gain to readers. On most screens they are virtually indistinguishable. −Woodstone (talk) 07:02, 8 December 2010 (UTC)

  What are you opposing—that you would be forced to use them? No-one has suggested this either. It seems that we’re back in discussion but without a clear proposal; what is needed now is an update (with detail in light of the above comments) to the proposed MoS change which can then be further discussed and/or voted on. — Wrapped in Grey (talk) 09:34, 8 December 2010 (UTC)

• Oppose I'm not very good at following long discussions, but it seems to be that the only advantage gained is aesthetics. The main reason I am opposing is there will be a lot of internal inconsistencies creeping in, which just creates a truckload of trivial, tedious work. I would, however, support a move to change to curly quote marks, but not an endorsement of both/personal preference. Adabow (talk · contribs) 09:50, 8 December 2010 (UTC)
• Oppose – Curly quotation marks are far more difficult to enter. Articles that use curly quotes will constantly have people, especially people who do not know MOS, entering straight quotes, which will be a maintenance nightmare. Also, there is no actual point to using any particular shape of quotation marks; they both mean the same thing and neither is ambiguous so why break consistency? If the guidelines are changed, I think it should be worded a bit better than how the proposition words it. McLerristarr | Mclay1 10:35, 8 December 2010 (UTC)
• Comment. Would it require the changing of the font used on Wikipedia to have a curly quote be the default when that key is struck on standard keyboards? I suspect that this may ignite a discussion regarding serif vs. sans serif fonts, but I thought it was an obvious possibility (although I'm not suggesting that it's the easiest one or necessarily a viable choice) that hasn't been addressed—at least as far as I can see. I'll admit that I haven't read through all the comments though. --Airborne84 (talk) 13:55, 8 December 2010 (UTC)

  No, the default (sans-serif) font contains curly quotes, but they are hardly distinguishable form normal quotes. — Edokter • Talk • 14:26, 8 December 2010 (UTC)

  Actually the Wikipedia style sheet only says "sans-serif", so it's the browser that chooses which sans-serif font to use, among the ones the operating system provides. For example, on my laptop " and ” look obviously different, but on the terminals at university they look completely identical, even after zooming in. A. di M. (talk) 14:45, 8 December 2010 (UTC)

• That won't work. It requires the software, e.g. MS Word, to be smart enough to do the right thing – 'smart' being a relative thing as it would be precisely what you don't want when editing reference tags, CSS elements, etc. Though they are used the same way ASCII quotes and curly quotes are different characters and appear in fonts in different places.--JohnBlackburne^words_deeds 14:22, 8 December 2010 (UTC)
• Oppose. (I should read VP more often.) Ugly (IMHO), prevents search engines from finding the material, difficult to type. "Difficult to type" could be fixed by adding it to the character insert list, (I see it's already there, but appears to be normal upright quote characters on my screen at the default size(!).) Identical appearance in some fonts. {{Cquote2}} is a special case, as the quotes are a different size, alignment, and font from the text. — Arthur Rubin (talk) 15:27, 8 December 2010 (UTC)
• Support so long as we make it clear that authors using curly quotes must be willing to put in the work to do them right. I would also support adding the words, "Wikipedia prefers but does not require straight/typographical quotes" (whichever we end up preferring). Darkfrog24 (talk) 16:23, 8 December 2010 (UTC)
• Oppose. I don't understand how the second straw poll below differs, but I oppose here and there, because I'd prefer consistency in how we write quotation marks. SlimVirgin ^{talk|contribs} 21:42, 11 December 2010 (UTC)
• Oppose. This would require an enormous amount of effort, cause numerous problems and provide very little benefit. I don't understand the logic behind switching from a near-consistent, easy-to-maintain state to an inconsistent, difficult-to-maintain battleground. [I oppose any implementation of the idea, so I've posted this comment under both the original straw poll and "straw poll within a straw poll."] —David Levy 01:49, 12 December 2010 (UTC)

Straw poll within a straw poll regarding curly quotation marks

• Support the idea but oppose the wording here that is under consideration to implement it. I would propose something more sophisticated to handle the issue.

  I couldn’t disagree more with those who think typographers quotes are ugly; the truth is they are the defining hallmark of fine typography. If most people found them to be ugly, they wouldn’t exist. The real problem, as many editors pointed out above, is that they are hard to type for those running Barbarian-OS. It’s just a fact that for those articles that A) are no longer stubs and are in a mature state; and B) are of a nature where they just don’t have many quotes; and C) have a shepherding author or authors who regularly clean up this sort of thing, there is nothing wrong with typographers quotes.

  I would propose instead wording that absolutely does not require that any editor have to horse around figuring out how to type typographers quotes. It could be something as simple as this:

Typographers (curly) quotes: No editor is required to use typographers quotes (He said “I gave him a solid ‘no’ to his question.”) in place of the easier-to-type straight quotes (He said "I gave him a solid 'no' to his question.") For those articles that consistently have typographers quotes, if editors must add new text containing quotes, they may use straight quotes and allow the shepherding authors of the article to later convert them. Alternatively, they may chose to copy a pair of typographers quotes from elsewhere in the article and use them for the newly added material. In such articles, do not make edits that accomplish nothing else than to convert typographers quotes to straight quotes.

The above avoids burdening editors who find it inconvenient to type typographers quotes while allowing those editors who are shepherding mature, well-done articles to clean the quotes up. For most articles that are no longer stubs and have reached a certain level of maturity, the simple reality is that while there will always be a continual dribble of various kinds of edits, those that entail the addition of a new quote are very rare. So this wording would allow a live-and-let-live relationship. Greg L (talk) 17:47, 8 December 2010 (UTC)

• Comment. I largely agree with Greg L, though I might add a sentence or two explaining the difference between typographical and typewriter quotes rather than assume that everyone knows it. I might encourage, but not require, new edits to articles with typographical quotes to also use typographical quotes. I might also suggest several alternatives for entering typographical quotes (with Greg’s suggestion, there now are at least four ways). JeffConrad (talk) 18:37, 8 December 2010 (UTC)

  The difference between the two types of quotes is explained immediately before the text which I've proposed to replace. A. di M. (talk) 19:33, 8 December 2010 (UTC)

I suppose there is an explanation of sorts, but the current wording isn’t really correct; the distinction isn’t between two methods of rendering (which is handled by the OS) but between two different styles of quotes that be entered. The explanations given in Quotation marks and Quotation mark glyphs are better but still not perfect. Perhaps something to the effect of

Two types of quotation marks can be used:

• Straight quotes, typically used with typewritten material; the opening and closing marks are identical.
• Typographical, or curly, quotes, typically used with typeset material; the opening and closing marks are different.

Because the common current meanings of “typewritten” and “typeset” are somewhat broader than the strict definitions, perhaps examples of each type of display could be given.

Perhaps just including links to one or both of the articles I mention above would suffice. The first article is interesting in that it shows different characters for opening and closing quotes in the box at the upper right and then uses typewriter quotes throughout the text of the article, which at least to me, is confusing. JeffConrad (talk) 20:32, 8 December 2010 (UTC)

• Comment. If fact it seems that " are quotes and ' is an apostrophe (all though several current computer programming languages treat and apostrophe as type of quote) not "double quotes or single quotes". unicode.org unicode.org says in regard to x'0027' "2019 ’ is preferred for apostrophe" as well as for x'0022' "preferred characters in English for paired quotation marks are 201C “ & 201D ” " . PS an example where both curly and straight quotes should be permitted in the same article. DG12 (talk) 17:32, 12 December 2010 (UTC)

  The exception of articles talking about quotation marks themselves (as opposed to using them to enclose quotations) is so obvious it didn't even occur to me to mention it. A. di M. (talk) 21:24, 12 December 2010 (UTC)

• Oppose. There's no good reason for using curly quotes. We're an online encyclopedia and subject to online rules which make using fancier typographic elements like curly quotes a pain to deal with. I find User:JeffConrad's reasoning unpersuasive: so some people made giant curly quotes in a rather superfluous template? That doesn't mean we should abandon common sense. Der Wohltemperierte Fuchs^(talk) 18:40, 8 December 2010 (UTC)
  • To your first sentence, the “good reason” is typographers quotes look better; if they didn’t, type foundries wouldn’t have ever bothered to make them in the first place for pretty much all typefaces on the planet. To your second sentence, if you find them a pain to deal with, then under this proposal, there is zero requirement that you deal with them. Greg L (talk) 03:49, 9 December 2010 (UTC)
• I support the idea, and the suggestions of Greg L and JeffConrad. Disallowing proper quotes is an odd artifact of obsolete technological shortcomings amidst the otherwise sane MOS guidelines on punctuation, and Greg's proposal would make it painless to adopt a style conforming to hundreds of years of typographic tradition without imposing any burden on those who can't be bothered with it.—Emil J. 19:12, 8 December 2010 (UTC)
• Comment. “We’re an online encyclopedia”. Yes, just like Britannica, which uses typographic quotes. As do almost all quality sources and the on-line content of The Times and The New York Times. That we should encourage (but not mandate) their use hardly seems in question; we just need to make sure that guideline is clear and does not introduce problems with, for example, wikilinks. I don't see that it should: if an apostrophe is changed in an article title then handling the consequences are no more onerous than when renaming for any other reason. — Wrapped in Grey (talk) 20:01, 8 December 2010 (UTC)
  • Aye, and I've enjoyed the prohibition of curly quotes because it made spotting copyvios from these news sources far easier to find. Der Wohltemperierte Fuchs^(talk) 00:15, 9 December 2010 (UTC)

  BTW, the German, Spanish, French and Japanese Wikipedias (and probably many others) all use their respective languages' version of typographical quotes, at least on today's featured articles. A. di M. (talk)

• I loathe curly quotes, but will be neutral toward a proposal that allows them without requiring or preferring them. ——chaos5023 (talk) 22:49, 8 December 2010 (UTC)
• Support. Having finally read through all the points, I think allowing editors to use typographic quotes without requiring them is reasonable. I don't think the Wikipedia community will ever allow a comprehensive change on this matter all at once. It's probably better to do this small steps at a time and see how it is received at each step. --Airborne84 (talk) 03:27, 9 December 2010 (UTC)

  What would be a smaller step than is being proposed here? A. di M. (talk) 12:54, 9 December 2010 (UTC)

• I haven't seen any comment that curly quote marks provide any advantage except that "they look better". Is this it, or can someone enlighten me further? Adabow (talk · contribs) 03:35, 9 December 2010 (UTC)
  • It’s purely an issue of they look better. Typographers quotes are simply part of fine typography. Typographers quote are indeed hard for people running Windows to type. But they are super-natural for many Mac people (keyboard-accessible and it’s the same for all programs). The above proposal basically just says “If you don’t want to use them, great; then there is zero requirement for you to have to horse around with them one iota. All editors are asked to do is to refrain from wading into highly refined articles where editors have taken the time to use them and do nothing but convert curlies to straight.” That’s all. Greg L (talk) 03:57, 9 December 2010 (UTC)
    • Thank you for that. I would still prefer either keeping straight, or a total switchover to curly, for the sake of uniformity. If this were to be the case I would support it, but a mixture of straight and curly doesn't really appeal to me. Adabow (talk · contribs) 07:15, 9 December 2010 (UTC)
      • If this poll were about requiring or preferring curly quotes, then we would need to say how they provide an advantage. However, this poll is about lifting the ban on them. We need to say how they don't provide a disadvantage. Darkfrog24 (talk) 15:35, 9 December 2010 (UTC)
• Oppose Yes, some people on some systems can see the difference, and they know how to enter the correct characters. But vastly more people fail to see any benefit from curly quotes on typical screens. The main problem from allowing curly quotes would be the pointless edit countitis and back-and-forth trivial edits: proponents would eagerly convert articles, or add quotations using their favored format (irritating those editors with a different view). Simple, unambiguous and easily-entered wikitext is good; anything else is dubious and requires more than ILIKEIT justification. Having curly quotes would be one more barrier for new editors, with no significant benefit. Johnuniq (talk) 03:59, 9 December 2010 (UTC)

Vastly more people . . .

Have you done a survey or is this just speculation? Absent a perceptible benefit, it seems doubtful that type designers would have used typographical quotes for hundreds of years. Whether they’re “right” or “wrong” is in the eye of the beholder, but it’s difficult to argue dispute that they aren’t considered good typographical practice. A simple search will yield many articles such as this one; I don’t suggest that Ms. Strizver is the last word on the subject, but she does have considerable experience with digital typography, and has been published by a reliable third-party source, which is likely more than most WP editors can claim. I think many of us would just as strongly assert that a prohibition requires more than IDONTLIKEIT, which is about all that I’ve seen so far. As for ambiguity, I simply don’t understand what you mean. As for the difficulty of entry, I’ll repeat once again a question that many have asked but to which no one has yet responded: why are quotes more difficult to enter than any other non-ASCII character, and why should they be given different treatment? It seems absurd to bar typographical quotes while simultaneously barring the use of a double hyphen for an em dash; the latter strikes me as unambiguous, if perhaps not as pretty, as the real thing. JeffConrad (talk) 07:47, 9 December 2010 (UTC)

My "vastly more people" is referring to typical screens. I have used Compugraphic typesetting equipment and need no convincing that curly quotes and unspaced em dashes are best in high-resolution publishing. They don't work so well here, although I am happy with unspaced em dashes. If curly quotes are required, MediaWiki should be enhanced to render straight quotes as curlies (with some tricks/special markup to enter quotes that must be straight). Johnuniq (talk) 02:42, 10 December 2010 (UTC)

Two questions come to mind: “What is a ‘typical’ screen?” and “What is a typical viewing distance?” I’ve easily seen the difference with any graphical display I’ve ever used. Additionally, is the output viewed at the default size or increased or decreased (increasing size makes sense for a greater viewing distance). And yest again I ask the question, Why are quotation marks any different from any other non-ASCII character, either in discernability of differences from ASCII approximations (where there are such things) or in difficulty of entry? If it makes no sense to use

Bob was thinking—as bewildering as the thought of Bob “thinking” might be—that it just might be possible to use a 4×5,

why is is not OK to use

Bob was thinking--as bewildering as the thought of Bob "thinking" might be--that it just might be possible to use a 4x5?

Surely the latter, though not quite as pretty as the former, is equally clear in meaning, and it’s easier to enter. It would seem to me that we either worry about typography (as the other-language WP pages mentioned apparently do) or we do not.

It might be nice to have quote transformation done by MediaWiki, but the transformation would need to handle most foreseeable special cases, and would need to ensure that ASCII quotes in markup were not altered. I’ve written several programs that worked most of the time, but there were usually a few cases that did not work quite right. Another option might be simplified markup (similar to what was discussed under En dashes above). As I’ve mentioned, I’ve used typographical quotes and other non-ASCII characters for 25 years without giving it much thought, but full disclosure: before software with “smart” quotes, I used software that recognized `` . . . '' to provide opening and closing double quotes, so the extra effort wasn’t much. As mentioned, this idiom would not work here because of the existing markup for italics, but perhaps a similar concept would work. Of course, it would become yet another WP-specific think to learn, so I really wonder how much it would save over Alt-0147 . . . Alt-0148 on Windows or the slightly simpler sequences with a Macintosh. Or simply using the character links below the edit box. JeffConrad (talk) 05:16, 10 December 2010 (UTC)

Why is it not OK? Because "--" is a half-assed emdash done on typewriters with a limited set of characters, and that "x" is a half-assed times symbol. x is a letter, denoting a character, × is a symbol, denoting multiplication (or other operations, such as the cross product). x x x is typographical unclear in meaning. Does it stand for "x × x = x²? Or does it stand for xxx = x³? In the case of 4x + 4x5 does it stand for 4x + 4 × 5 = 4x + 20, 4 × +(4 × 5) = 80, 4x + 4x5 = 4x + 20x = 24x. In your example, does "4x5" stand for 4 × 5 = 20, 4x5 = 20x or some abbreviation such as "4(letter x)5" which would stand for something like "Quad x-5 model engines"?

Straight quotes are not half-assed quotations symbols, they are quotations symbols. Bob was thinking—as bewildering as the thought of Bob "thinking" might be—that it just might be possible to use a plunger. is no less clear than Bob was thinking—as bewildering as the thought of Bob “thinking” might be—that it just might be possible to use a plunger., and the allowing the latter will makes our editing life utter hell (see my above post for why). Headbomb {talk / contribs / physics / books} 01:38, 12 December 2010 (UTC)

If you italicize x as a variable and stick it to the previous stuff, whereas you keep x as a multiplication sign upright and space it on both sides, it's unambiguous: Template:J, xxx, Template:J, Template:J, Template:J, and so on. I'm not saying it wouldn't be a terrible idea to do that, only that there are reasons other than ambiguity to avoid it. A. di M. (talk) 11:27, 12 December 2010 (UTC)

• In case this is a separate poll, still Oppose. If this suggested version were to be included, I would suggest that if the so-called (not in the proposed version) "shepherding editor" makes one mistake, a monitoring editor would be justified in removing all typographical quotes, rather than locating the specific mistake(s). Ease of searching seems adequate justification for forbidding those quotes, and ease of editing (even if the quotes were to appear in the text). As an alternative, I would suggest that editors be encouraged to use {{smartsinglequote|1=quotation}} or {{smartdoublequote|1=quotation}}, which would properly expand to the desired form, but that the quote characters, themselves, should not be in the Wikitext. This would still damage searches of Wikipedia, but would allow editors without a specialized Wiki-editor to edit articles. — Arthur Rubin (talk) 10:03, 9 December 2010 (UTC)

  Would some of the people repeating the mantra about "ease of search" please educate me what on earth do they mean? In typical search engines like Google, it is impossible to search for text including typewriter quotes, because they are special characters (they denote a literal phrase). Thus, if anything, replacing typewriter quotes with proper quotes will make searching for them easier. Why would anyone want to search for text including quotation signs in the first place is escaping me.—Emil J. 11:28, 9 December 2010 (UTC)

  It is possible, although not easy, to search for text containing a quotation mark in google. More important, though, is that if you search for foo bar (as a quoted string), and the actual text is foo “bar or foo, “bar, it will not match. — Arthur Rubin (talk) 11:37, 9 December 2010 (UTC)

  It sure will. Here's a test: [1]. Two of the top ten results (for me; Google search notoriously depends on location, sun spots, and other stuff) contain typographical quotes, one of them being a Wikipedia article.—Emil J. 12:58, 9 December 2010 (UTC)

  And to preempt another point: in WP internal search, it will not match. However, this is no different from the behaviour of typewriter quotes, they also will not match:[2]. I conclude that this whole thing about searching is pure FUD.—Emil J. 13:13, 9 December 2010 (UTC)

  You don't need a specialized Wiki-editor. You can insert “” in the special character palette below the edit box. Also, I don't see what such templates would accomplish (and they'd need far shorter names than that). A. di M. (talk) 13:01, 9 December 2010 (UTC)

• Oppose increasing the burden on editors (even those who freely accept it) when the burden ought to be placed upon the maintainers of the editing software (I know, they're volunteers too). Jc3s5h (talk) 13:21, 9 December 2010 (UTC)
• Comment: The wording "Straight quotes, typically used with typewritten material; the opening and closing marks are identical" is factually incorrect. Typically, typewriters are no longer in use. Seriously, I tried years ago to explain to a neice who is now in her 20s how the enter key on a computer was like the carriage return key on a typewriter, and she replied "but I've never used a typewriter." Explaining anything in terms of a typewriter will fail to communicate to younger editors. Jc3s5h (talk) 13:21, 9 December 2010 (UTC)

“typewriters are no longer in use”. Kinda my point, in a way . . . Seems to me that a WL would solve it; a more strictly correct description could be given, but it would be more complicated, to the effect of “monospaced output”, or “devices that generate typewriter-like output (character printers, alphanumeric terminals)”, or “pure ASCII output”. I think most readers would find the typewriter description easier to follow. I suppose we could say that straight quotes “were used in typewritten material, and some people still like them”, but I think most editors would find this pejorative. JeffConrad (talk) 22:48, 9 December 2010 (UTC)

• Comment: I’d have to agree that templates would be unlikely to be seen as superior to the simple ‘point-and-click’ that we already have. — Wrapped in Grey (talk) 13:29, 9 December 2010 (UTC)
• Support. Editors should be free to choose. New wording looks great. SteveB67 (talk) 17:27, 9 December 2010 (UTC)
• Comments. I agree with Emil J that the search argument is largely FUD, an argument looking for a basis. I also agree with Wrapped in Grey that templates, even with more compact names, would be more work than any of the four alternatives for entering non-ASCII characters that we’ve discussed. The only justification might be if a template could handle the tasks (choosing quotation marks, including those for nested quotations, based on the language setting) the <Q> element is supposed to handle. JeffConrad (talk) 22:48, 9 December 2010 (UTC)
• Support the idea, per JeffConrad. While I rarely if ever use them myself, and am in general not a "fan" of their use in either-or situations, I see no reason to prohibit others from doing so.--Epeefleche (talk) 00:18, 10 December 2010 (UTC)
• Oppose -- curly quotes look good in rendering, but not good enough to make them worth the effort to maintain. --SarekOfVulcan (talk) 21:38, 11 December 2010 (UTC)
• Oppose, I would prefer consistency. SlimVirgin ^{talk|contribs} 21:41, 11 December 2010 (UTC)
• Oppose I don't really know how this differs from the above section, so I'm gonna oppose here per the same reason as above as well just so it's clear that this is a horrible idea, and that whatever "variation of it" is proposed is equally horrible. Headbomb {talk / contribs / physics / books} 01:15, 12 December 2010 (UTC)
• Oppose. This would require an enormous amount of effort, cause numerous problems and provide very little benefit. I don't understand the logic behind switching from a near-consistent, easy-to-maintain state to an inconsistent, difficult-to-maintain battleground. [I oppose any implementation of the idea, so I've posted this comment under both the original straw poll and "straw poll within a straw poll."] —David Levy 01:49, 12 December 2010 (UTC)
• Support. Generally speaking if an author wants to format an article in this manner I see no reason to reject. But, the curly-quotes are significantly more burdensome to maintain and editors shouldn't be required to follow that style when adding to existing articles, so this proposal seems like a happy media. Christopher Parham (talk) 02:38, 12 December 2010 (UTC)
• Oppose both. The relative difficulty of maintaining consistent use of curly quotes in an article outweighs any minor aesthetic benefits. And the consistency of always using the one kind across every article is beneficial.

There is a huge difference, by the way, between curly quotes and other non-ASCII characters like the dashes. Curly quotes give no additional difference of meaning from straight quotes, whereas the dashes have distinct roles that cannot be replicated by a hyphen (unless one is careless). That's why we need to allow them. oknazevad (talk) 03:01, 12 December 2010 (UTC)

In some cases, there is no substitute for a non-ASCII character; however, for some, ASCII approximations (e.g., two hyphens for an em dash, x for a multiplication symbol) are unambiguous, and it makes no sense to disallow these approximations while disallowing (technically, not recommending) typographical quotes. JeffConrad (talk) 04:46, 12 December 2010 (UTC)

As others have said, those are poor substitutes that were only used when they were needed; while it could be argued that also describes straight quotes, there's no loss of correctness of punctuation when using straight quotes. The same can't be said for using a hyphen for an endash, and if endashes are introduced, then proper emdashes also should be used.

None of this addresses the other (and I would say more important) aspect of ny objection: the increase difficulty in maintaining article consistency. Or that we should be encouraging site-wide consistency where it doesn't conflict with known (usually national/regional) variations in language usage. As far as I know, there's no distinction in any variation of English between the meaning of straight and curly quotes; they are fully equivalent. oknazevad (talk) 13:23, 12 December 2010 (UTC)

• Support. 1) The proposal is to "allow" not require. 2) Comments related to "in font fff" or "displayed on device ddd" should consider what "may be" available now or in the future. 3) I object to the restriction that straight and curly NOT be used in the same article which is especially useful with code samples. DG12 (talk) 17:10, 12 December 2010 (UTC)

  I see no compelling reason to "allow" the introduction of unnecessary inconsistency (along with the various difficulties discussed above), far outweighing the minuscule benefit of having arguably prettier quotation marks in some articles.

  It's inaccurate to claim that such a change would require nothing of editors; it would require us to either actively participate in the style replacement or stand back and allow the encyclopedia's quality to decline. "You're not obligated to clean up the resultant mess" (scare quotes) is of little solace to those of us who see no need to create such a mess in the first place.

  If you believe that there are specialized cases in which the two different types of quotation marks should be combined to draw an actual distinction, please draft a proposal. But that's unrelated to this proposal, which is based upon the assumption of interchangeability. —David Levy 17:46, 12 December 2010 (UTC)

  Code samples are a special case, much like one should write 2 * 2, 2 - 2 and 2^2 (or 2**2 instead of 2 × 2, 2 − 2 and 22. A. di M. (talk) 21:28, 12 December 2010 (UTC)

• Support. But it would be preferable to take the present text, drown exclusive, and include the arguments for curly quotes (which I have no intention of using, btw). No rule should be understood so as to be prevent the listing of a font including both kinds of quotes; but if this can be covered by in text or some such phrase. Septentrionalis PMAnderson 17:56, 12 December 2010 (UTC)

To clarify, do you support the allowance of curly quotation marks in general prose, or do you merely advocate that we explicitly note an exception (already existent in practice) for cases in which we seek to illustrate such a glyph? —David Levy 18:10, 12 December 2010 (UTC)

Both. Note that the first is allowance, not requirement, which would be worse than what we have now. Opinions against requiring curly quotes should be noted for when that proposal comes up. Septentrionalis PMAnderson 18:24, 12 December 2010 (UTC)

Thanks for clarifying. I oppose the allowance of curly quotation marks in general prose (for reasons cited above), but I have no objection to the notation of an exception for cases in which we seek to illustrate such a glyph (a matter of common sense, really). —David Levy 18:32, 12 December 2010 (UTC)

This has really been illuminating. Far too many editors had a knee-jerk reaction to what they thought the above-proposed guideline said something it didn’t and responded anyway with shear nonsense like This would require an enormous amount of effort. The “effort” is currently borne by shepherding authors who may decide for themselves how they want to spend their time contributing to Wikipedia. All this would have done is sanctified the real-life reality we see on Wikipedia and asked for some common, reciprocal curtesy. But that was apparently too God-damned much to ask.

All it says is that if editors don’t want to use typographers quotes, they don’t have to—not even if they add text containing quotes. There are a number of such fine articles on Wikipedia that consistenly use typographers quotes.

No; I won’t point them out for fear of a WP:DICK acting like a dick.

And these articles are maintained by shepherding editors who understand that typographers quotes are part of a suite of fine typographic practices that include the proper use of em-dashes (—) instead of a double-hyphen (--) and the use of the en-dash to denote a range (pages 20–24). That’s why fine word processing programs like Word automatically generate typographers quotes.

All this guideline would have done is sanctified proper etiquette and asked of editors who find typographers quotes tedious and who want to stick with typewriter quotes is to abide by one simple curtesy: just don’t wade into articles that consistently use typographers quotes and do nothing but change them to typewriter quotes.

For shepherding authors who go though the extra effort, that was not too much ask and the result here simply shows that editors who don’t take the damned time to actually read and comprehend the proposal have as much say as do people who take the time to read the proposal. It reminds me of people who vote for candidates in local elections based on who has the most *American*-sounding name: these people’s votes count as much as anyone else’s. There are editors flitting about who create problems on Wikipedia and write inane crap like how the degree Celsius is not one of the International System of Units. It takes so damned much time (*sigh*) and respond to this stuff to prevent Wikipedia from being taken over by the 9th graders and people who have no flying clue about what they are talking about (other than they have an opinion and Wikipedia gives them a voice).

So…

(Quotation Marks) Or just let others do the heavy lifting and simply not go flat out of your way to undo that effort

So I’ll propose this guideline since to help better identify the editors who are just plain ornery:

Typographers quotes No one is required to use typographers quotes. But for those articles that consistently use typographers quotes (because the article has shepherding authors who address quotes as a superset of fine typographic practices that include the use of en‑dashes em‑dashes), don’t be WP:DICK where you wade in and change them all to typewriter quotes.

There… Clearly, the wording with the “WP:DICK” isn’t seriously intended to be part of a real guideline. But the principle being conveyed here is spot on precisely what I mean. Ample electronic white space is provided below so we can now see editors’ real spots here. Greg L (talk) 19:53, 12 December 2010 (UTC)

For someone stressing the importance of "curtesy" [sic], the above comments are extraordinarily discourteous. I don't appreciate having my views dismissed as "shear" [sic] nonsense" stemming from not bothering to read and comprehend the proposal.

I took the time to read every word of this discussion before (and after) weighing in. I don't know whether you've done the same (and I won't hurl accusations to the contrary), but you obviously don't intend to take seriously the opinions of those with whom you disagree (or "dicks," as you prefer to label us). —David Levy 21:06, 12 December 2010 (UTC)

Just where in the world, Mr. Levy, did you get the idea that being discourteous means that others must profess to admire your reasoning and arguments as much as you do?? You and every other wikipedian and I.P. editor has an entitlement to free from ridicule for being what you are (your religion, your race, etcetera). Nothing on Wikipedia (including the lack of a “bright-line” rule for defining incivility) means that this and any other venue where vigorous discussion and debate occurs requires that others refrain from opining that the proposals and reasoning you put forth that seem to be shear nonsense are just that: nonsense. It’s an opinion and I spared your idea of one of my more onerous labels: that it is a metric ton of weapons‑grade bullonium. Now…

The proposed guideline to which you responded could not possibly be clearer that if YOU (I’m talking about David Levy now) don’t want to use typographer’s quotes, then you don’t have to. Yet you wrote This would require an enormous amount of effort… Well, just pardon me all over the place for pointing out that the guideline could also not have been clearer that for those articles that already consistently use typographer’s quotes, the only rule of the road you would be asked to abide by is to not exert an *enormous amount of effort* to undo others’ work since it is they who are perfectly free to exert such effort—and obviously already did. M‘kay??

Now please desist with that tired old Wikipedia tactic of posturing for an advantage by hiding behind the apron strings of how I was somehow discourteous and uncivil by having directed *hate speech* towards your clearly splendid ideas. Why don’t you just ante up with a “support” or “oppose” vote, below so we can all see whether we can agree on that little basic element of courtesy of simply not going out of one’s way to piss in other editors’ corn flakes. That’s not too much to ask. Greg L (talk) 21:38, 12 December 2010 (UTC)

Just where in the world did you get the idea that I was criticising you for failing to agree with me or express admiration?

I disagree with you. You disagree with me. That's fine. Assuming that anyone who disagrees with you didn't bother to read the proposal (or was too stupid to grasp it) is not. Calling us "dicks" (and "dick-holes" in the original revision) is not.

I don't seek to "hide behind" these issues or use them to my advantage. (Note that I haven't reported your incivility.) However, I don't care to waste my time with a futile attempt to explain my position (in a manner largely redundant to what I've already written) to someone who has dismissed my views as the nonsensical ravings of an ill-informed "dick."

If you wish to engage in respectful discourse (instead of dismissing dissenting opinions out of hand and assailing their authors), let me know. —David Levy 22:25, 12 December 2010 (UTC)

Well, I’ve been on a crash diet so I wasn’t in a mood to engage in the standard Wikipedia beat-around-the-bush horse crap that wastes time. See my 22:35 post, below, which I wrote before your response here. I see glimmers of a foundation upon which we have some common ground. Greg L (talk) 22:40, 12 December 2010 (UTC)

I've responded to that message. I'd also be glad to address your above points (regarding perceived incongruities between your proposal and my objections thereto), provided that you're interested in hearing me out. —David Levy 22:45, 12 December 2010 (UTC)

• Support Because I don’t go to Microsoft Word’s preferences to disable automatic generation of “smart quotes” based on the notion that typewriter quotes (“dumb quotes”?) are somehow better looking before I send my documents to clients. And—most importantly—I like being a good wikipedian citizen rather than someone who enjoys running around on Wikipedia and devoting extra effort to do nothing more than simply undo others’ work. Greg L (talk) 20:20, 12 December 2010 (UTC)
• Support. The MoS directs us to “follow the sources” and the """sources use typographic quotes—clearly, their use is desirable on that basis. As discussed by others above, the previous objections (per the current MoS) on technical grounds are no longer applicable (or no more significant than with other, currently recommended, non-ASCII characters). — Wrapped in Grey (talk) 22:06, 12 December 2010 (UTC)

  If our goal were to abide by external style conventions, we wouldn't need the MoS at all. We'd simply consult the AP Stylebook (or a similar publication) and call it a day. But it isn't feasible for a wiki to adopt 100% of the conventions used in another medium.
  It's been noted above that unlike the use of dashes instead of hyphens, there is absolutely no difference in meaning between the two types of quotation marks. It's reasonable to prefer one type's appearance, but neither is more correct than the other. —David Levy 22:25, 12 December 2010 (UTC)

Now we’re seeing some convergence in views. I agree 100% with you, David. Greg L (talk) 22:35, 12 December 2010 (UTC)

That neither type of quotation mark is more correct? —David Levy 22:38, 12 December 2010 (UTC)

Yes. Neither is more “correct.” Clearly, one doesn’t make a Word-doc incorrect by disabling “smart quotes” in the preferences. But by enabling it, one makes a Word-doc better. But typographer’s quotes a just as clearly a pain for editors stuck in front of computers running Barbarian-OS. Even some Mac people don’t want to horse around with them even though they are easily typed straight from the keyboard. This proposal would not change the principle that no editor is required to use typographer’s quotes. Proportional-spaced typefaces, and en‑dashes and em‑dashes (as distinct from simple hyphens) are all part of fine typography. The art of typography is all about making text as easy as possible for the eye (the “mind”, really) to follow. There is clearly a limit though; no one is advocating a flourish like the “fl” ligature (the word “flourish” uses that ligature). This is just about curly quotes in those articles that consistently use them. Some shepherding editors who’ve grown stubs into fine articles consistently use them and they look wonderful. The above proposal would just endorse a bit of etiquette that has long been observed on Wikipedia. Greg L (talk) 23:01, 12 December 2010 (UTC)

Indeed, your proposal is based upon the assumption that curly quotation marks are "better." Presumably, this is why it includes no prohibition of a mass edit (perhaps on the part of someone who has decided to "shepherd" an article) to replace all instances of straight quotation marks with curly ones.

The assertion that curly quotation marks are "better" is very much in dispute. Unlike en dashes, em dashes and hyphens, there is absolutely no difference in meaning between a curly quotation mark and a straight quotation mark. Therefore, the issue is one of aesthetics. And even if we assume that the appearance of curly quotation marks is more widely preferred (which might or might not be true), they factually are substantially more difficult for most people to type.

Yes, I realize that you haven't proposed requiring that anyone type them. But you advocate a setup in which unnecessary inconsistency (i.e. inconsistency not required to avoid favoring a particular English variety) among articles is permitted and encouraged. I (and others) regard a consistent style as vastly more aesthetically pleasing.

Most of us agree that it's undesirable for a single article to contain mixed usage (apart from some specific exceptions). When I wrote that "this would require an enormous amount of effort," I was referring to effort on the part of editors monitoring articles to correct the never-ending introduction of straight quotation marks. You note that such a responsibility would belong to "shepherding authors" (an entirely optional role). But as I wrote in my reply to DGerman, users encountering such flaws would be forced to either actively participate in their correction (effectively becoming "shepherding editors") or stand back and allow the encyclopedia's quality to decline. "You're not obligated to clean up the resultant mess" (scare quotes) is of little solace to those of us who see no need to create such a mess in the first place. And what happens when an article's "shepherding author" stops editing it?

You've asserted that this change would merely formalize an accepted bit of Wikipedia etiquette, but I strongly disagree that there ever has been consensus for such a practice. Deviations from the MoS (including some that are considerably more obvious than this one is) frequently go uncorrected. The same is true of misspellings and flagrant factual inaccuracies. It's unreasonable to interpret these entities' retention as implicit approval. —David Levy 23:57, 12 December 2010 (UTC)

• Oppose. The community resoundingly does not want typographic quotes for several reasons listed above. This is simply an attempting to bypass consensus. All citation template would have to be forked into "straight quotes" vs. "curly quotes", as well as several other templates. The issue goes beyond a simple a lack of will from certain editors to "do the work". Headbomb {talk / contribs / physics / books} 23:10, 12 December 2010 (UTC)

Please dismount, kind sir, from your high horse as you blocketh the sunlight for the minions down below who don’t own MOS (as you apparently think you do). The only thing that is clear is you don’t like this proposal “for the reasons listed above”. User:Wrapped in Grey, above, should be quite disappointed that you have undertaken to declare that his opinion apparently doesn’t matter in this discussion because you’ve spoken for the community. And I’ll thank you very much to assume good faith by not declaring that my motives here are to bypass consensus. Do you think you own this place??? Try familiarizing yourself with WP:Consensus can change before displaying a metric butt-load of failure to assume good faith and the breathtaking display of arrogance you just exhibited. Bypassing consensus is done by editing in artriclespace contrary to consensus. One does not “bypass consensus” by proposing guidelines on the talk pages a style guide like WP:MOS. The only thing about your vote that is clear as glass is that you reserve the right to wade into any article you can find and do nothing but replace smart quotes with typewriter quotes; an attitude I find utterly reprehensible. Editors should be drummed out of Wikipedia for that sort of attitude. Greg L (talk) 23:23, 12 December 2010 (UTC)

Greg L is being a little extreme here, Headbomb, but I must agree that starting a discussion to seek a new consensus does not constitute trying to bypass consensus. How else should we find out what the community wants than by asking? Darkfrog24 (talk) 14:49, 13 December 2010 (UTC)

If you expect others to assume good faith, perhaps you should refrain from accusing editors with whom you disagree of commenting on a proposal that they haven't bothered to read.

And if you're going to accuse someone of "declar[ing] that [someone else's] opinion apparently doesn't matter," I suggest that you avoid equating your opponents with "people who vote for candidates in local elections based on who has the most *American*-sounding name" and lamenting the fact that "these people's votes count as much as anyone else's." —David Levy 23:57, 12 December 2010 (UTC)

…And not one twit of what you wrote changes the fact that what Headbomb wrote is pure nonsense. Greg L (talk) 00:28, 13 December 2010 (UTC)

I disagree with Headbomb's assertion that your proposal constitutes an attempt to bypass consensus, but it's a stone thrown from a domicile constructed of non-glass materials. —David Levy 00:50, 13 December 2010 (UTC)

• Oppose. For now, curly quotes should be removed on sight. Actually that applies to unnecessary Unicode in general. Unicode rendering is insufficiently supported and standardized — what looks good on one editor's screen may be appalling on another's. --Trovatore (talk) 23:37, 12 December 2010 (UTC)

Thanks for your charming post. Greg L (talk) 23:41, 12 December 2010 (UTC)

• Comment. Though I agree with the concept, I might change the wording. I’m not sure I quite understand “a superset of fine typographic practices”, and I’d probably leave WP:DICK out of it (even though my actual thoughts are far stronger). I disagree with David Levy that the community “resoundingly does not want typographic quotes for several reasons listed”. The responders seem to be about 2–1 against them; whether or not this is resounding, I can’t say, but it certainly is not consensus. I also question whether the opposition is for the reasons stated, because honestly, most of them just don’t hold water, and some seem akin to “I want a regime change”. This isn’t to question anyone’s sincerity, but simply to suggest that most of the reasons seem half thought out.

• Typographical quotes look no better than straight quotes, or even typographical quotes are ugly. Perhaps a matter of opinion, but one that is at odds with hundreds of years of considered judgment of type designers and professional publishers. From CMoS, 16th ed., 6.112:

. . . These marks, which are available on any modern word processor, generally match the surrounding typeface. For a variety of reasons, including the limitations of typewriter-based keyboards, these marks are often rendered incorrectly. . . All software also includes a “default” mark ("); in published prose, this unidirectional mark, far more portable than typographer’s marks, nonetheless signals a lack of typographical sophistication. [emphasis added]

I don’t suggest that CMoS or any other guide is dispositive, but I do think it speaks with greater authority than the average WP editor.

• Typographical quotes look no different from straight quotes. Perhaps with some fonts on some monitors with certain browsers under certain conditions, but hardly universally true. They look far different under some conditions, and at worst cause no harm for others.
• Typographical quotes are too difficult to enter. There’s no disagreement that they require a slight extra effort, but it’s hardly a big deal. And the effort is no greater than for any other non-ASCII characters. As has been noted by several others, editors of other-language WP pages seem to have no problem using them; are those editors smarter than we are? I’ve used typographers’ quotes for almost 25 years, far longer than MS Word has had “smart” quotes, and I just don’t see the problem.
• Typographical quotes place too much burden on the average editor. We haven’t proposed that editors be forced to use them, though we do force (or at least demand) that editors use an em dash rather than two hyphens, so this argument makes no sense. As for possibly making lack of fine typographical skills a barrier to entry: we have articles for which the editors know little about simple punctuation, spelling, writing, the subject matter, reliable sourcing, reference formatting, and so on, so the lack of typographical skills seems a minor issue.
• Unlike dashes, typographical quotes do not alter the meaning. This simply is untrue; an em dash has always been represented on a typewriter by two hyphens, so there is no confusion. In many cases (perhaps except for compounds), a single hyphen used in place of an en dash doesn’t interfere with correct interpretation. There are characters for which there is no good typewriter approximation, but they’re comparatively infrequent. If we really think that non-ASCII characters are too difficult to enter, we should allow any typewriter approximations that don’t hinder interpretation.
• I simply do not understand Trovatore’s assertion that “Unicode rendering is insufficiently supported and standardized”; could we have some examples? Except for em, en, and thin spaces with Opera, I have no problem displaying anything in List of XML and HTML character entity references; although that list doesn't include all Unicode characters, it’s the vast preponderance of what might be needed for WP articles.
• Allowing typographical quotes would lead to inconsistency. Article-to-article, probably true, but we already allow considerable inconsistency (e.g., unspaced em dashes vs. spaced en dashes, reference styles, treatment of some mathematical symbols, American vs. British English), so concern about the impact of typographical quotes is a red herring. Intra-article consistency is another matter, which would be addressed by some of the proposed changes.
• Typographical quotes interfere with searches. Just enter the typographical quote character in the search box. And how often to people really search for quotes, anyway?

I don’t suggest that my thoughts are the last word on the topic, and I’m sure that some of those opposed to typographical quotes can (and will) rebut them, but in aggregate, I find most of the sometimes vitriolic opposition utterly unpersuasive. There’s one other thing to note: the MoS currently recommends but does not mandate straight quotes, and does not even require intra-article consistency. The various proposed changes at least address the latter; though the opponents in the recent discussions are in the majority, I see nothing even approaching consensus to change the recommendation to a mandate. JeffConrad (talk) 03:15, 13 December 2010 (UTC)

1. You've attributed a comment by Headbomb to me.

2. As noted above, pointing out that editors wouldn't be forced to use the curly quotation marks is the same as saying "You're not required to clean up the resultant mess." Some of us don't want an unnecessary mess to be made (which would require us to either assist in its cleanup or stand back and tolerate it).

3. The use of two hyphens to represent an em dash is a makeshift solution in which two instances of a mark are combined to approximate the appearance of one instance of a different mark. Conversely, a straight quotation mark is a single mark whose correct meaning is identical to that of a curly quotation mark. The two situations simply aren't comparable.

4. Most of Wikipedia's inconsistencies stem from differences among the various English varieties. Where this is not an issue, consistency is preferred (hence the existence of a style guide). Heck, we even use logical punctuation throughout the encyclopedia, despite its absence from conventional American English.

Does this mean that we're required to favor consistency when feasible? No. But some of us want to. You're pointing out that we don't have to, thereby ignoring the fact that arguments in favor of consistency stem from actual preference, not a sense of obligation.

5. I disagree with your interpretation of the MoS. We seldom use ironclad wording (and most rules have exceptions), but if you're suggesting that editors are welcome to ignore the recommendation because they feel like it, you're mistaken. —David Levy 03:57, 13 December 2010 (UTC)

1. My apologies for the misattribution—sometimes these threads are tough to follow. Quite honestly, the comment does seem to reflect your position. I still strongly disagree with the comment, regardless of who made it.

2. I guess I don’t see “the resultant mess” that would need cleanup; all of the proposed revisions mandate intra-article consistency, and place the onus for maintaining it on those who would use typographical quotes. I think this is much ado about nothing.

3. So what? The case can be made (as in CMoS) that because opening and closing quotes are different (something I learned in grade school), a single straight quote is not the same character. The case also can be made that a straight quote is a nonexistent character in typeset material; it really depends on how nutty we want to get.

4. I guess I see far more inconsistencies than just varieties of English. I’m very much a believer in consistency, but take issue with those who would wield the sword of consistency capriciously and inconsistently. Compared with the variations in citation style, the difference in quotes is invisible. And there are some of us who don’t care for amateurish typography. Again, I won’t say that straight quotes are “wrong”, but I can cite many sources that support the use of typographical quotes; I doubt that you could cite many to the contrary. I honestly wouldn’t have such strong objections if it were decided that we would avoid all non-ASCII characters (including those represented in HTML entities) as much as practical; my objection is to the capriciously differential treatment.

5. My interpretation is based on what the MoS says; we seem to have no problem saying no when that’s what we mean, e.g., “Do not use substitutes for em or en dashes, such as the combination of two hyphens (--). These were typewriter approximations.” Quite ironic given that the straight quote was a typewriter approximation as well . . . As to “you're mistaken”, is this a threat and are you the chief enforcer? What a way to encourage people to contribute to Wikipedia . . . I think we should bear in mind, as the project page register points out, the original admonition to use straight quotes was a capricious change by one editor, with no discussion, and it has apparently been used to justify the “status quo” ever since. I don’t dispute that many editors, perhaps a majority of those who’ve participated in the endless discussions of the topic, oppose typographical quotes, but there also are a significant number of editors who disagree. If that represents consensus for banning typographical quotes, then I guess I don’t understand what the word means. JeffConrad (talk) 06:46, 13 December 2010 (UTC)

1. I never doubted that the misattribution was accidental.

No, such a claim does not reflect my position. While there appears to be more opposition, the proposals also have drawn a significant amount of support (and I've stated in no uncertain terms that I disapprove of out-of-hand dismissal of viewpoints contrary to one's own).

I've based my arguments upon reasons why I (along with others) regard the consistent use of straight quotation marks as preferable.

2. "The resultant mess" refers to the intra-article inconsistency that would arise via the never-ending introduction of straight quotation marks. Again, the explanation that only willing editors would be responsible for the corrections simply doesn't suffice. The rest of us would encounter these flaws, forcing us to either participate in the cleanup efforts or tolerate their presence while awaiting intervention by the "shepherding authors" (assuming that they still monitor and edit the affected articles).

3. Indeed, an opening quotation mark and closing quotation mark serve separate purposes. But a straight quotation mark serves exactly the same purpose as one or the other (depending on its placement). In normal usage, a pair of straight quotation marks functions identically to a pair of curly quotation marks (without any difference in meaning or correctness).

4a. I agree that Wikipedia already contains avoidable inconsistencies. We should be working to eliminate them, not introducing new ones.

4b. While straight quotation marks certainly are simpler than their curly counterparts are, I disagree that they're "amateurish." I regard them as appropriate for use in the wiki medium, which varies from other media in significant respects.

4c. Several of us have explained why we regard curly quotation marks as contextually incomparable to other non-ASCII characters in use within the English Wikipedia.

5a. No, I assure you that I'm not issuing any sort of threat. I neither possess nor claim to possess greater authority on this matter than that of any editor in good standing.

5b. Most of our rules are descriptive, not prescriptive. The addition of an MoS entry without advance discussion can be perfectly appropriate and uncontroversial, provided that it reflects actual practice. For the same reason, "the rules say so" is not a valid argument. (In other words, this rule has remained because the practice that it documents has remained, not the reverse.)

We generally require consensus for article elements' inclusion, not for their exclusion. —David Levy 07:57, 13 December 2010 (UTC)

Why you “regard curly quotation marks as contextually incomparable to other non-ASCII characters” is of little relevance: we are directed to “follow the sources”—something that you choose to ignore but which is a guiding principle of the MoS. “Straight quotation marks ... [are] appropriate”—observing quality on-line publishers such as The Times, The New York Times, and Encyclopaedia Britannica suggests otherwise. — Wrapped in Grey (talk) 08:58, 13 December 2010 (UTC)

I responded directly to this statement when you wrote it previously, so I'm pather ruzzled by your claim that I "choose to ignore" it. —David Levy 11:59, 13 December 2010 (UTC)

You responded to the statement ‘the MoS directs us to “follow the sources” ’ with “if our goal were to abide by external style conventions, we wouldn't need the MoS”—an apparent attempt to contradict/ignore the guideline. You went on to mention the difficulties of “adopt[ing] 100% of the conventions used in another medium” (which had not been suggested) and argued the case against typographic characters based on your personal opinion of their merit—clearly at odds with “follow the sources”. — Wrapped in Grey (talk) 13:07, 13 December 2010 (UTC)

1. You seem to be interpreting "follow the sources" to mean "laboriously and indiscriminately emulate style conventions employed by others." That isn't what it means. We look to high-quality sources for guidance, but we're permitted to weigh various considerations and determine what works best for Wikipedia.

2. I'm not implying that anyone actually advocated adopting 100% of the conventions used in another medium. My point is that you're failing to consider the possibility that the needs and realistic goals of a wiki might differ from those of a professionally edited publication in this instance.

3. No, I made the factual statement that neither type of quotation mark is more correct than the other is (irrespective of personal preferences or those of external sources).

4. As noted by JohnBlackburne, many (perhaps most) high-quality online sources favor straight quotation marks. —David Levy 14:11, 13 December 2010 (UTC)

I did not notice this the first time, but 'such as' implyies such usage is typical, while I find when reading "quality" online sources that straight quotes are almost always used, by e.g. the BBC, The Economist, The Guardian, CNN and The Telegraph. The only news site I visit which uses curly quotes is the NYT. And that is the exception: all other sites including many more less quality ones use straight quotes. So if we are to as you suggest "follow the sources" we should use straight quotes.--JohnBlackburne^words_deeds 12:54, 13 December 2010 (UTC)

I did not suggest “follow the sources”; the MoS suggests it. However, the publishers I listed were not an attempt to list typical sources (did you not spot Britannica in the list?); they were listed to show the feasibility/appropriateness (denied by others) of using these characters in an on-line medium. — Wrapped in Grey (talk) 13:28, 13 December 2010 (UTC)

I see. Then it should be pointed out that the efforts expended by such publications to maintain and enforce such style choices are quite significant. What newspapers generally have is a dedicated staff of copy editors to turn raw copy into newsprint. The last newspaper office I worked in had a dozen people working full time on this, and it was a quite small circulation (though high quality) newspaper. They had to fix errors, make sure the copy fitted the layout and followed the paper's editorial policy, and enforce the house style. And it is enforced, so it is consistently one style, not a mix of two which mean exactly the same but are just visually different as some are advocating here, which will be a lot more work to deal with and generate a lot more pointless edits as quotes are added then changed across many thousands of articles.--JohnBlackburne^words_deeds 14:05, 13 December 2010 (UTC)

Who has claimed that it's unfeasible or inappropriate to use curly quotation marks in an online medium? And if many (perhaps most) high-quality online publications use straight quotation marks, how are we failing to "follow the sources"? —David Levy 14:11, 13 December 2010 (UTC)

A number of editors made comments along the lines of “how do you even type them”; you used the words “isn't feasible”, one would assume in reference to the subject; you also suggested that “straight quotation marks ... [are] appropriate for use in the wiki medium, which varies from other media in significant respects” which would suggest that you believe typographic quotation marks to be inappropriate—apologies if this is not the case. There seems to be some confusion regarding the MoS’s use here of the words “high-quality sources”; my assumption is that (as with other sections of the MoS where it is stated explicitly) this refers to (the) article sources (not sources in general), which, in my experience, tend to be printed works (where typographic characters are used almost exclusively). — Wrapped in Grey (talk) 15:12, 13 December 2010 (UTC)

I stated that "it isn't feasible for a wiki to adopt 100% of the conventions used in another medium," by which I meant that one must expect a wiki's style to deviate from those of other media in some respects. I didn't intend to imply that it wouldn't be feasible for us to use curly quotation marks.

Indeed, I regard straight quotation marks as appropriate for use in the wiki medium (my point being that while one would expect to see curly quotation marks in print media and even on some websites, the circumstances under which a wiki is written justify the use of straight quotation marks). That doesn't mean that curly quotation marks are inappropriate for use on a website (including a wiki), but I've explained why I regard the exclusive use of straight quotation marks within our articles (apart from specific exceptions) as advantageous.

We refer to sources relevant to the matter at hand. In this case, the typographical conventions employed by print media, particularly where they differ from those of online media, are of little relevance (hence our lack of paragraph indentation, for example). —David Levy 15:44, 13 December 2010 (UTC)

In response to the assertion that "straight quotation marks ... [are] appropriate," you stated that "observing quality on-line publishers such as The Times, The New York Times, and Encyclopaedia Britannica suggests otherwise." How so, and how does this jibe with your above explanation? —David Levy 14:11, 13 December 2010 (UTC)

Simply that the 1st 3 sites that I checked to see if whether-or-not they used typographic quotes, did use them (though not always exclusively) and that this may indicate that the FUD (“can’t be typed”, “can’t be seen”, “won’t render correctly”, etc.) is unwarranted. — Wrapped in Grey (talk) 15:12, 13 December 2010 (UTC)

I personally disagree with the "can't be seen" and "won't render correctly" arguments. It isn't literally true that the marks "can't be typed," but they certainly are substantially more difficult for most wiki editors (as opposed to professional publications' authors) to generate. —David Levy 15:44, 13 December 2010 (UTC)

1. I then apologize for ass-u-m(e)ing your position. I shall henceforth take greater care to avoid doing so.

2. You’re probably correct that many (most?) editors would not bother to read the guideline, let alone practice it. And some “shepherds” would probably default on their obligations. But if I were to look at areas of Wikipedia that need attention, even intra-article, inconsistencies would not even register in comparison with variations in citation/reference style, and even more, unsourced nonsense (of which there is all too much).

3. Yes and no; typographical quotes are immediately be recognized as beginning and ending quotations (or “so-called” material in the case of “scare quotes”). And if it’s really that simple to do the same thing with straight quotes, a bot could probably be written that would do the cleanup and get it right as often as most current bots do with other things. If indeed the mess were to materialize.

4b. Perhaps “amateurish” is overly pejorative, but as CMoS observe, it’s long been one of the most obvious indicators of typographical unsophistication, and a reminder of one down side of the demise of typesetters. Few people who know much about typography use straight quotes; but perhaps few people know (or care) much about typography. I also guess I don’t see why the “wiki medium” is so different in this regard. To be fair, the use of straight quotes in online versions of major newspapers seems to be the rule rather than the exception, but online content there often uses only ASCII characters, including either single or double hyphens for em dashes. I wish some of the folks who say that that straight quotes look fine would generate a PDF using a serif typeface (still used overwhelmingly in professionally published material); the straight quotes stand out like a sore thumb, and because they clash with the patterns of the rest of the characters, even if most readers don’t quite realize it.

I went through this exercise almost 25 years ago. I had read AT&T’s reports claiming that typeset material could be read 27% faster, and was skeptical to say the least. When we did some actual testing, we were somewhat surprised to find, to a person, that if anything, AT&T had understated the case. We also had the same reaction to straight quotes, and strongly discouraged their use. I won’t suggest that typographical quotes made a 27% difference, but the glaring distraction of straight quotes was clearly evident to everyone involved. We used serif typefaces for most running text, so this undoubtedly influenced the perception. Now perhaps we were unusual (or I put the fear of dog into everyone), but type designers and professional publishers seem to share the perception we had.

4c. And those who see little difference between quotes and many other non-ASCII characters have explained why we feel as we do. I’ll be the first to say that different is never the same, but there is such a thing as reasonable equivalence. For the life of me, I cannot see why an em dash is not reasonably equivalent to two hyphens; yes, the latter are two marks, but so what? They’ve been recognized as an em dash since before any of us was born, so it seems to me we’re nitpicking. And an x is a different character from ×, but in the appropriate context, the former is easily recognized as representing the latter, so again, what is the problem? I guess on this issue, the two groups agree to disagree.

5a. Perhaps I overreacted, but the comment did seem a bit confrontational. And again, the MoS reads as it reads. My approach is to say what’s meant and interpret according to what’s said, much as judges (or at least many of them) require of legislators and contract authors; vague laws and contracts are invalid for good reason.

5b. I hardly suggest that every addition to the MoS (or an article) needs prior discussion—were that required, nothing would ever get done. In this case, however, though it undoubtedly reflected the most common practice, it’s obvious that many editors have disagreed almost from the beginning, so it’s difficult to maintain that it reflects consensus. And that is the basis of my objection. JeffConrad (talk) 09:50, 13 December 2010 (UTC)

1. Thanks for that.  :)

2. I agree that Wikipedia already has issues significantly worse than any that I foresee arising as a result of the proposed changes. I believe that said changes would be harmful, but I don't mean to imply that the harm would be catastrophic.

3. Do you mean that curly quotation marks result in superior communication?

Regarding the deployment of bots, I regard this as a good idea (irrespective of what style[s] we ultimately settle upon).

4b. It's true that such perceptions historically have prevailed, but times change and language evolves. Just as it no longer is widely regarded as unsophisticated to say "you are" instead of "thou art," the stigma attached to straight quotation marks has faded and continues to do so. Content is conforming to the new media (instead of the reverse).

4c. In my view, there is a substantial difference between visual approximation and 1:1 equivalence. A hyphen is a distinct mark with its own meaning. Combining two of them is intended to simulate the appearance of (i.e. pretend to display) an em dash. Conversely, a pair of straight quotation marks is directly equivalent to a pair of curly quotation marks (despite its slightly different appearance). The characters possess no separate meaning and are used exclusively for this purpose (in ordinary prose, that is). And of course, we also use en dashes.

However, I understand your view, just as you understand mine. It largely comes down to a matter of where to draw the line (with mine drawn halfway between the two that you regard as most logical). So yes, let's agree to disagree.

5a. I didn't intend to be confrontational (and I apologize for coming across in such a manner).

When the MoS indicates that one style is "recommended" and another is "not recommended," this does not mean that said advice may optionally be ignored (beyond the usual degree of wiggle room) because someone happens to disagree with it. Such a setup is rendered unfeasible by the fact that no one possesses special editorial control of articles published here, including those that they create or dramatically improve. In other words, no one has the right to label him/herself a "shepherding author" (Greg L's terminology) and exert authority as to the style in which an article is written. Someone performing his/her first edit to an article has as much say as someone who's worked on it for years.

Therefore, if we interpret the advice in question as optional, it encourages users to engage in edit wars, flipping back and forth between the two styles (despite the fact that one is "recommended" and the other is "not recommended") according to their personal preferences. The prevention of such problems is one of the MoS's most important purposes.

5b. It might be true that there is no consensus to disallow the use of curly quotation marks in general prose. However, there certainly isn't consensus for their allowance. As noted above, we generally require consensus for article elements' inclusion, not for their exclusion. —David Levy 11:59, 13 December 2010 (UTC)

3. In some cases, yes. The following three lines are, in one sense, equivalent:

He said, "She said".

He said, "She said".

He said, “She said”.

But different is never the same; there’s a considerable difference in readability. The most obvious improvement is the change from monospace to proportional font, but there’s also a noticeable difference with quotes that match the surrounding typeface. Now these passages are so short that the differences are probably of little consequence, and the differences are definitely more apparent in a serif typeface. I’m sure not everyone will agree, but what I’ve said reflects the considered judgment of type designers for hundreds of years, so it’s hardly off the wall.

4b. Certainly, almost everything changes with time. I don’t necessarily agree with regard to typography; typographical quotes remain the norm in print, and many web publishers (and their readers) simply may not know better. When I started using proportional type, it was relatively new outside of professional publishing, and the only examples were printed sources from professional publishers, and of course they used typographical quotes. The use of typewriter quotes (along with a hyphen for em dash and a minus sign; few even knew what an en dash was) was a dead giveaway that the author was a newbie. With time, non-professional publishing has become far more common, and perhaps what originally stood out no longer does. Like several others, I’ve readily acknowledged that straight quotes are the norm in online versions of most of the mainstream press. Absent information from a reliable source, I would make no assumptions about why this is done. As I’ve also mentioned, much of this material is all over the map for other non-ASCII characters: hyphens and double hyphens (and sometimes actual dashes) for em dashes as one example. This hardly argues that online style is carefully scrutinized. One thing that has changed is that traditional news publishers have cut back to the bone, and this undoubtedly includes copy editors.

4c. I hear you but don’t agree with you. No disputing that one mark isn’t the same as two, but in this case, I’d say they’re functionally equivalent.

5a. When “not recommended” is used in one instance, and “do not” is used in another, by any reasonable standard the former is not mandatory. Who makes the official interpretation of the MoS? I’d say that if it even needs interpretation, we have a problem.

Though a first-time editor may have the same general rights as anyone else, this certainly is not license to disregard prevailing style, as we state many places in Wikipedia policy. The general rule, in many places express, in others implied, is stare decisis for matters of pure style; this reflects simple common sense as well as common courtesy. Were it otherwise, articles would be in constant turmoil. When I’m the first to add references, I usually use author-date; when I add references to an existing list, I honor the prevailing style. There are any number of options, many of which we’ve already discussed, that are far more susceptible to edit warring than quotes.

5b. In this instance, the inclusion of straight quotes effectively excludes typographical quotes, and I would disagree that the inclusion or the equivalent exclusion reflects consensus.

In responding to Wrapped in Grey, you said, “the circumstances under which a wiki is written justify the use of straight quotation marks”; what exactly does this mean? I’m tempted to suggest an answer, but recent experience suggests I’m better off not doing so. JeffConrad (talk) 00:47, 14 December 2010 (UTC)

Wow. --Airborne84 (talk) 01:50, 14 December 2010 (UTC)

3. I honestly perceive no difference in readability. Some have opined that straight quotation marks are more readable, another position with which I disagree. To me (and obviously, this is highly subjective), there is no difference in readability or clarity.

I am, of course, highly accustomed to seeing curly quotation marks in professional print media (to the extent that straight ones would appear sloppy). But online media incorporate different conventions. We also omit paragraph indentation, the absence of which arguably is more obvious.

4b. Indeed, straight quotation marks are the norm in online versions of most of the mainstream press. We need not make any assumptions as to why, let alone speculate that these publishers "may not know better." For whatever reason, in online media, straight quotation marks are not widely regarded as substandard.

Therefore, we can safely use them without fear of embarrassment. This, of course, doesn't mean that we must use them; it means that we can base our decision upon other factors.

No one has disputed that straight quotation marks are significantly easier for most editors to generate (so no matter what style[s] we select, most editors will continue to type them). There is disagreement regarding the acceptability of inter-article inconsistency (with some seeing no problem, while others regard it as highly undesirable), but there is agreement that intra-article consistency (apart from specific exceptions) is needed.

All of this points to the use of straight quotation marks. As far as I can tell, when the matter of professionalism is removed from the equation (in light of the fact that most professional online publications favor straight quotation marks), the argument in favor of curly ones boils down to "we think that they look better."

4c. Darkfrog24 explained it well: [Both] forms of quotation marks are correct and proper. "--" is not a real em dash but " is a real quotation mark.

Surely, you recognize the distinction (though it's perfectly reasonable to regard it as unimportant). We're agreeing to disagree, yes?

From a practical standpoint, another material difference is the fact that no representation of em dashes serves as a de facto standard among members of the general public (and by extension, this wiki's editors). Some people type two hyphens. Some people type one hyphen. Some people generate actual em dashes (via various means). Some people place spaces before and after, while others don't. So no matter what style[s] we favor, a considerable amount of cleanup (for the sake of intra-article consistency, at the very least) is necessary.

Conversely, most editors type straight quotation marks, so maintaining such a style is easier and less time-consuming than any alternative would be.

5a. If the advice isn't mandatory (apart from exceptions justified by specific circumstances), what purpose does it serve? What sense does the interpretation "x is recommended and y is not recommended, but feel free to do whatever you please" make? How does such an approach not encourage "constant turmoil" (i.e. edit warring)?

The stare decisis principal applies to situations in which multiple styles are equally valid. For example, no English variety would be uniquely appropriate for an article about a subject lacking strong ties to a particular nation. So if the first substantive revision happens to be written in Australian English, it probably shouldn't be changed to a different variety. (I say "probably" because legitimate reasons occasionally arise. For example, we moved Check (finance) to Cheque [and switched from American English to British English] because this enabled us to use a title lacking parenthetical disambiguation.) Conversely, a style not compliant with the MoS is not equally valid to one that is.

Furthermore, this proposal includes no concept of honoring the existing style of an article containing straight quotation marks. It allows a "shepherding author" to stake a claim, converting all straight quotation marks to curly ones (which would instantly become sacrosanct). It would effectively designate curly quotation marks as the preferred style, with straight ones permitted as temporary placeholders in articles that have not yet been "shepherded."

5b. "Consensus" means more than "agreement among persons discussing something." Among editors of this wiki (few of whom will ever see this page), straight quotation marks overwhelmingly predominate.

I assume (and please correct me if I'm mistaken) that you would question why (and perhaps speculate that most of our editors "may not know better"). I assert that we need not concern ourselves with the reason, which simply doesn't matter.

This is not an argumentum ad populum. An error should not be permitted to stand, regardless of the extent to which it predominates. But as discussed above, both straight and curly quotation marks are correct and proper. So if we're to have an encyclopedia-wide standard (as some of us advocate), it's sensible to select the style overwhelmingly favored by its editors (even if only because it's generated with greater convenience).

By "the circumstances under which a wiki is written justify the use of straight quotation marks," I was referring to the fact that a wiki, unlike a professional publication, is written primarily by persons to whom curly quotation marks are relatively inaccessible, with no editorial hierarchy through which content is standardized or point at which it's finalized. —David Levy 16:09, 14 December 2010 (UTC)

3. That’s an opinion to which you are certainly entitled, but as several of us have noted repeatedly: it’s at odds with centuries of opinion of type designers an publishers. Perhaps they have it wrong . . .

4b. With regard to inter-article consistency, the MoS reads as it reads; perhaps the current wording does not reflect consensus. Most encyclopedias disallow inter-article variations, including reference style and Engvar. I have difficulty seeing that fly here.

4c. Darkfrog24 stated, without support, that “both forms of quotation marks are correct and proper”. This again is a matter of opinion; the one authoritative guide I have that mentions it disagrees, and to the other guides the straight marks do not even exist. I remain unpersuaded.

5a. I can’t answer the question other to say advice is advice. I’ve had a fair amount of involvement with standards organizations, including style guides therefor, and have no problem distinguishing between advisory, permissive, and mandatory language. The interpretation you give is your opinion of what is meant, and is at odds with the plain words. Now perhaps the current wording does not say what is meant, but I suppose that the intended meaning is yet another issue for which it would be difficult to find consensus. JeffConrad (talk) 07:20, 15 December 2010 (UTC)

3. That's a false dilemma.

4b. Indeed, Wikipedia's style differs substantially from that of any professional guide (and I thank you for acknowledging this). But we generally regard inter-article consistency as beneficial, excepting difference related to English varieties. And we even use logical punctuation throughout the encyclopedia, despite its absence from conventional American English.

4c. Whether curly quotation marks are preferable to straight ones (and vice versa) is highly debatable. But the belief that straight quotation marks actually are incorrect (as opposed to merely stylistically inferior) clearly is not widely held by our editors or the participants in this discussion.

Professional style guides, contrary to our MoS, are prescriptive rather than descriptive. They've been slow to address the Internet (including its terminology) and acknowledge its stylistic differences. It made headlines when the AP Stylebook finally switched from "Web site" to "website" (reflecting longstanding common usage).

As you noted, most of the guides that you consulted don't even acknowledge straight quotation marks' existence. And yet, there they are, widely used on professional websites (including those of publishers whose print editions contain curly quotation marks).

You're exactly right. The passage was sloppily written. And if its wording had been clarified at the time, we wouldn't be discussing it in this context. But now that we're having this debate, such a change would be perceived as an attempt to modify the text's intent.

But again, if the "recommended" style isn't mandatory (apart from exceptions justified by specific circumstances), what purpose does it serve? What sense does the interpretation "x is recommended and y is not recommended, but feel free to do whatever you please" make? If that's so, why even bother recommending a particular style?

As previously noted, our MoS is descriptive, not prescriptive. If it doesn't accurately describe our conventions, it should be revised. And of course, consensus can change (which is why we're having this discussion). Users of varying opinions are providing arguments as to why their preferred outcomes are advantageous. "The rules [don't] say so" is a weak rationale for any of us to advance. —David Levy 12:33, 15 December 2010 (UTC)

But the MoS is proscriptive. It tells people what to do on Wikipedia. We occasionally mention something about the prevalence or origins of a given rule but only as that serves the document's main purpose. That's as it should be. We have source-citing articles like question mark and title case for handling the descriptive aspects of the language. Darkfrog24 (talk) 02:07, 16 December 2010 (UTC)

1. Do you mean "proscriptive" or "prescriptive"? (This is not a spelling nitpick, as either plausibly could be suggested).

2. Regardless, the MoS is descriptive. As noted below, it describes current practices (as determined both by analyzing those that prevail and by discussing possibilities and deciding what works best).

Meanwhile, a style manual like the AP Styleguide or the Chicago Manual of Style hands down decisions from above. From a practical standpoint, the distinction is that publications following such a manual can legitimately decide to apply a particular convention "because the guide says so" (no analysis required). Some occasionally opt to deviate, but this is optional and relatively uncommon.

Conversely, if someone objects to an entry from Wikipedia's MoS (on the basis that it either doesn't accurately describe current practice or a belief that said practice ought to be modified), we have discussions such as this one. And my point is that within such discussions, there's little need to debate an entry's disputed meaning (because we're here to determine what makes the most sense and describe that in the MoS, not to blindly abide by whatever already has been written there). —David Levy 05:09, 16 December 2010 (UTC)

The MOS, as all guidelines, is supposed to describe the best current practice on Wikipedia. It does fail at this sometimes, though. A. di M. (talk) 15:12, 16 December 2010 (UTC)

Agreed. —David Levy 16:03, 16 December 2010 (UTC)

3. In all honesty, I don’t know what you mean by this.

4b. Logical punctuation is undoubtedly the work of a few disgruntled Brits . . . though it’s not my normal style, it’s at least logically defensible, and most people other than Amurricans use it.

4c. I guess it depends on why they might be preferable (i.e., they work better or they’re easier to enter).

If the MoS is descriptive rather than prescriptive, then editors can do as they damn well please, and it falls upon those who maintain the MoS to describe what the various miscreants are doing . . .

“And if its wording had been clarified at the time”. If you lived here, you’d be home by now. For good or for ill, it reads as it reads.

I think I’ve addressed the purpose the current wording might serve. JeffConrad (talk) 17:17, 15 December 2010 (UTC)

3. Unless I've misunderstood (in which case I apologize), you've represented that there two possibilities:

A) Curly quotation marks are inherently clearer and more readable than straight quotation marks are.

B) For hundreds of years, type designers and publishers have been "wrong" to use them.

4b. I don't know the background behind the decision to adopt logical punctuation across the board, but it's an example of our general preference for inter-article consistency. It isn't a typical example, of course, as we usually draw the line where conventions vary among English varieties. That we've even crossed that line in this instance illustrates the extent to which inter-article consistency is valued.

4c. No, I haven't claimed that all of the practices described arise spontaneously. We frequently engage in discussion to determine what conventions to adopt (as we're doing now). Prevailing preferences aren't sacrosanct, but they must be taken into account. Otherwise, a handful of editors could unfairly impose their preferences on the entire community (and even now, this sometimes is alleged).

It's entirely possible, of course, for a majority of people to be wrong about something (in which case it's sensible to correct them). But almost no one has suggested that straight quotation marks are incorrect. So to ignore the community's overwhelming preference (including among editors unfamiliar with the MoS) is unreasonable.

I asked you to explain how the interpretation "x is recommended and y is not recommended, but feel free to do whatever you please" makes sense from a practical standpoint, and you replied that you "can't answer the question other to say advice is advice." (In other words, as long as the words make linguistic sense, it doesn't matter if the message conveyed is utterly pointless.) —David Levy 19:58, 15 December 2010 (UTC)

Let’s see if I can’t make a better attempt at discerning the purpose. Let’s look just at the text we propose the to change:

The exclusive use of straight quotation marks and apostrophes (see preceding section) is recommended. They are easier to type in reliably, and to edit. Mixed use interferes with some searches, such as those using the browser's search facility (a search for Alzheimer's disease could fail to find Alzheimer’s disease and vice versa). Furthermore, HTML elements (such as <ref name="xxx"/>) may not always work if curly quotation marks are used.

Whenever quotation marks or apostrophes appear in article titles, make a redirect from the same title but using the alternative glyphs.

Let’s take the key elements individually:

1. “They are easier to type in reliably”. If an editor is willing to make the effort and capable of doing so reliably, this presumably is not an issue.
2. “and to edit” I honestly don’t know what this means. Perhaps it simply suggests that inconsistencies will arise from subsequent edits; if an editor is willing to watch the article and maintain consistency, this also is not an issue.
3. “Mixed use interferes with some searches”. If the reference is to the effect of intra-article inconsistency on a search within an article (and I’m not certain that it is), this probably is correct, especially for a search for an apostrophe. It’s much less of an issue if an editor maintains intra-article consistency. I’ll concede that even with consistent use of a typographical apostrophe, a search may be more difficult for a user who does not know how to enter the characters into the browser’s search box, but I wonder how many intra-article searches really involve quotes or apostrophes. Again, if we really think such searches are common, we should ban as many non-ASCII characters as is practicable.
4. “HTML elements (such as <ref name="xxx"/>) may not always work if curly quotation marks are used”. No question about the intent here, and no disagreement. As nearly as I can tell, typographical quotes simply do not work. Though I can’t seem to find an explicit ban in the HTML 4.01 specification, the only character mentioned for quoting attribute values is U+0022 (").
5. “Whenever quotation marks or apostrophes appear in article titles, make a redirect from the same title but using the alternative glyphs”. Because the type of “quotation marks” isn’t given, this is somewhat ambiguous, but several re-reads suggest that it means “if you do use typographical quotes or apostrophes in a article title, make a redirect from the same title using the typewriter characters.” It’s also possible, though less plausible, to interpret this as calling for a redirect from a title using typographical characters to the article title that uses the typewriter characters.

In sum, the most plausible interpretation is to the effect of “we recommend that you not use typographical quotes, because it’s easy to get them wrong, but if you must do so, create redirects to facilitate searches using the typewriter characters”. In any event, it’s almost impossible to interpret it as a ban unless the words have different meanings from those in any dictionary I’ve ever read. JeffConrad (talk) 09:02, 15 December 2010 (UTC)

1. Are you suggesting that most editors are willing to make the effort to generate curly quotation marks and capable of doing so reliably?

2. It means, I think, that subsequent editors are likely to experience difficulty matching the curly quotation marks (whether adding new quotation marks or retyping the preexisting ones when performing major revisions that lead to their removal from the edit window).

I've expressed my objections to the assumption that "an editor is willing to watch the article and maintain consistency."

3. From my perspective, this is one of the weaker arguments against the use of curly quotation marks.

5. This hasn't been followed particularly well, but we're supposed to redirect from plausible search terms (including variants containing whatever similar characters aren't used in the article's title). This has absolutely no bearing on which characters we favor, which this text needn't address. (Whether we use straight quotation marks exclusively, use curly quotation marks exclusively or use a combination of the two, the principle will apply.) —David Levy 12:33, 15 December 2010 (UTC)

5b. I agree that consensus may mean more than agreement among those discussing something, but disagree that it is necessarily determined but disagree that it equates to predominant practice.

As regards “correct and proper”, as I mentioned in 4c, this has been alleged but unsupported. Not everyone agrees. And it appears that we probably shall not.

I agree that a wiki is written primarily by persons to whom curly quotation marks are relatively inaccessible, but as many of us have stated many, many times, this is true for any non-ASCII character, so the argument just doesn’t hold water. Many editors are unfamiliar with many aspects of proper style for WP (or any other formal publication), and will continue to be so. As I’ve said repeatedly, the consequent cleanup makes quotes seem trivial.

The issue of paragraph indention is completely irrelevant. The non-indented style probably predominates on the web because prior to CSS, it was impossible without multiple nonbreaking spaces. At present the decision to indent or not indent is handled by a simple change in CSS—essentially the same way it would be done with printed media, though the formatting software would usually be different. And to further the point, the default style for body text in MS word is block style, just as we use here.

There certainly are some significant differences between printed and online media, such as pagination vs. galley display that we use here, but typography and formatting don’t fall into this category. For example, depending on the editing application used to create a document, the result can be sent to a printer, made into a web page, or a PDF file; the typography or formatting of the output doesn’t change because of the destination medium unless the author goes to great lengths to make it so. Now I would not normally use MS word for a web page, so perhaps there is some issue of the editing application, but it’s hardly one related to the appropriate presentation for particular media. JeffConrad (talk) 07:20, 15 December 2010 (UTC)

I haven't claimed that consensus "equates to predominant practice." My point is that we must consider more than the personal opinions of the users participating in this discussion.

I address the "correct and proper" issue above.

I've explained my (and others') perceived distinctions between quotation marks and non-ASCII characters (such as dashes) "many, many times" (including in the message to which you're responding).

In another message, I acknowledged that "no matter what style[s] we select, some cleanup of inconsistent usage within articles is unavoidable." I then explained that "the allowance of curly quotation marks (either exclusively or optionally) can only cause the need to soar."

While it's merely an example, my point regarding paragraph indentation (and our lack thereof) is that we obviously don't aspire to emulate all of the basic style conventions widely followed in print media.

I don't assert that straight quotation marks inherently provide a more "appropriate presentation" online. The only presentational advantages that I cite pertain to matters of consistency and navigation.

Most of the advantages to using straight quotation marks throughout the encyclopedia directly relate to the article's creation and maintenance (which, of course, affect presentation).

In other words, I don't claim that a finished product appearing online is worse if it consistently contains curly quotation marks instead of straight ones. A major element of the equation is the fact that Wikipedia and its articles never become finished products. —David Levy 12:33, 15 December 2010 (UTC)

Many of us agree that personal opinions are largely unpersuasive. I again defer to the overwhelming preponderance of widely used style guides.

I should think it obvious that “correct and proper” is a matter of opinion.

Again, I’ve seen nothing at all convincing about why quotation marks merit special treatment.

With regard to indention, this has nothing to do with the alleged difference between print and online media.

So in essence, the advantage of straight quotes is that they’re easier to enter? JeffConrad (talk) 17:17, 15 December 2010 (UTC)

1. Others defer to the overwhelming preponderance of websites (including those whose print versions follow those style guides). I prefer to focus primarily upon the proposals' direct and indirect effects on the encyclopedia and its editors.

2. Indeed, "correct and proper" is a matter of opinion. And straight quotation are regarded as correct and proper by a vast, vast majority of people (including those who regard curly quotation marks as stylistically superior).

An example of something obviously not correct and proper would be substitution of the letter "K" for the letter "C" (as in "The kat krossed the kobblestone road."). Conversely, →"← (excluding the arrows) is a legitimate quotation mark.

3. No one has opined that quotation marks "merit special treatment." We advocate that they be treated in the manner overwhelmingly prevalent on the Internet.

We've thoroughly explained our perceived distinctions between curly quotation marks and other non-ASCII characters. You're entitled, of course, to remain unconvinced, but please don't imply that no case has been made. We've agreed to disagree on this point, yes?

4. I don't understand your statement about paragraph indentation. Are you suggesting that such a style element is not employed across most print media and omitted from most online equivalents?

5. Various advantages of straight quotation marks (including some with which I don't entirely agree) have been cited ad nauseam. Most (but not all) of the arguments that I regard as strongest can be traced back to the fact that straight quotation marks are easier to enter (resulting in a situation in which problems will arise if curly quotation marks are permitted in general prose). I hope that I needn't elaborate yet again. —David Levy 19:58, 15 December 2010 (UTC)

1. No disputing common web use; whether common practice here is deference is possibly debatable. A glance at web typography also shows that it’s pretty sloppy on all other non-ASCII characters; what has struck me from the beginning is that WP goes with the common typewriter quotes yet is so fussy about the other special characters. That’s simply schizoid; were there not this disparity, I probably not have started the discussion.

2. “[S]traight quotation [marks] are regarded as correct and proper by a vast, vast majority of people”. How do we know this? Based on many years’ experience with non-professional authors writing for print, online, or whatever, I’d say most don’t know the difference and never even give it a thought. I’ll concede that most people I’ve worked with have been far less fussy about typography than I am (which you’ve no doubt already guessed), but this is the first venue in which the lowest common denominator is almost mandated (and I guess we agree to disagree about the mandate). If use makes something “correct and proper”, your statement is probably correct. It gets back to the question of descriptive vs. prescriptive style (going far beyond typography), and there are proponents of each approach. Merriam-Webster might be an example of the former, and American Heritage an example of the latter.

3. The “special treatment” is as I mentioned in 1 above; with quotes, we pay homage to a character created solely for a device that has hardly seen any use for 20 years, yet we are sticklers for proscribing other characters that device could not properly generate. Quotes are no more difficult to enter or maintain than any other non-ASCII character, yet some here would have us believe that use of typographical quotes would cause the United States east of California to fall into the ocean.

4. You mentioned the lack of paragraph indention as another example of a difference between print and online media (or at least WP. My comment was simply that the difference in paragraph styles has nothing to to with print vs. online media, so it’s a red herring. I still ask the question: what is the substantive difference (if any) between web and print publishing that relates to the issue in this discussion?

5. Supposed advantages, including reducing global warming, have indeed been cited ad nauseam. Much of my objection here has been that so many of the arguments are simply insubstantial. As I’ve tried to make clear, I readily concede some advantages, and suspect you and I agree on most of the substantive ones; we disagree on whether they’re dispositive.

5a. Straight quotes are definitely easier to type. But the same is true for many other typewriter conventions that we disallow, which is why I still have a hard time buying this. Perhaps it is simply that quotes occur more frequently in many articles (which I don’t think anyone has mentioned, though so much has been said that I could have missed it). This isn’t always the case, though; in many technical articles, other special characters occur more frequently (and are far more difficult to get right).

5b. Allowing two types of quotes would create inter-article (and possibly intra-article) inconsistencies. Because of other far more significant inconsistencies, I find the inter-article inconsistency unpersuasive. Intra-article inconsistency would of course violate one of the most basic guidelines of the MoS. It has been suggested that mandated intra-article consistency or not, editors would use straight quotes in articles with curly quotes. I again say this assumes facts not in evidence; I’d burn that bridge if we come to it. That said, I’m sure some editors would do so, just as they do with all manner of other things that other editors must correct if the article is to follow MoS mandates, let alone intra-article consistency.

5c. The presence of two different types of quotes would hinder searches. This may be valid, especially if a single article had two different types of quotes, but I suppose searches for the curly marks could be more difficult for some users. Safari and Chrome for the straight marks find the curly ones as well; Firefox, IE8, and Opera do not. Again, I don’t know if this is a real issue or simply a manufactured argument, but I’ll concede the possibility of the former.

5d. Most of the other proffered advantages for straight quotes are either ILIKEIT or are simply off the wall.

6. This is at least the third time we’ve run through almost the same arguments, and it’s obvious that we agree to disagree. If we at least agree on the points which we disagree, perhaps we could use this as the start of a summary (and place it at the end where readers can find it); obviously, the other side of my rebuttals would need to be included. Perhaps we could make two tables, pro and con, with rebuttals to each, as was done in at least one of the past discussions. In fairness, we might need to include even the arguments we both consider insubstantial.

We have beaten the horse until there’s arguably not enough for a dentist to identify. I don’t think continuing the beating until morale improves is likely to accomplish anything, so I suggest we not do so. Another horse will come along soon enough; I’ll probably let it pass through undisturbed. JeffConrad (talk) 05:25, 16 December 2010 (UTC)

Summarizing the argument, should it remain unresolved, would be highly desirable, mandatory IMO. However, after a day off, I have a couple of other suggestions to consider:

• The MoS’s General principles should ideally guide us here; given that they do not seem to be able to, would anyone care to propose a change to them such that they can?
• A thought experiment: remove the controversial text from the MoS for a time; what do we think would happen? Might we learn something that would help to guide us?

Wrapped in Grey (talk) 06:37, 16 December 2010 (UTC)

I assume that proponents of curly quotation marks would begin replacing straight ones throughout their favorite articles. Opponents would revert, triggering prolonged edit wars.

Meanwhile, a vast majority of editors, oblivious to this conflict, would continue typing straight quotation marks (just as they always have and always will). —David Levy 06:45, 16 December 2010 (UTC)

The idea was to look beyond the immediate and obvious possibilities. For example, edit-wars, should they occur, might be resolved in ways not yet considered here. However, presupposing your answer as ‘not necessarily’, I’d be more interested in your thoughts on the first bullet point: you stated that my interpretation of one of the MoS principles was different to yours; my point is that they should ideally be worded such that they are not open to interpretation. — Wrapped in Grey (talk) 08:31, 16 December 2010 (UTC)

Perhaps the Follow the sources passage could explicitly indicate that it's intended to apply to people and things mentioned in high-quality articles, not to the general style conventions employed by the publications in which they happen to appear (or the print/online versions thereof). I suspect that "such as the treatment of proper names" was intended to imply this, but it could be spelled out more clearly.

Regarding general style conventions, we could note that all high-quality sources can be considered (irrespective of the extent to which we've cited their articles as factual sources), with attention paid to the publications' similarities to and dissimilarities from Wikipedia when determining individual style conventions' applicability thereto.

But I see a more pressing need to revise/expand the Internal consistency passage to indicate that inter-article consistency generally is desirable (though not necessarily mandatory) where no English variety issues exist. I suggest that the subheading be renamed Consistency (preferred) or a separate passage be split off. —David Levy 16:03, 16 December 2010 (UTC)

This isn’t all that crazy. I assure you that I would not go changing quotes, because there simply are too many far more important things to do. I’m pretty scrupulous about honoring existing style unless something is clearly wrong with it; I do this more from common sense and common courtesy than for compliance with the MoS. If a change were made (and based on this discussion, none of the proposals would fly as written), there should be a requirement to honor the existing style unless there is none or if significant numbers of quotes are added to an article with only a couple (which in a sense says there is arguably no existing style. Disruption, for its own sake or otherwise, was the last thing I had in mind when I had the poor judgment to reopen this discussion. I and several others had suggested that changing to curly quotes should not be objectionable, but such a policy is clearly untenable at present. If it were found that people really didn’t dislike curly quotes, the issue could be revisited (as it surely will be anyway . . .). JeffConrad (talk) 08:58, 16 December 2010 (UTC)

I've taken the liberty of extending your numbering (thereby enabling me to address the individual points without repeating what you wrote). If you object, feel free to revert. —David Levy 06:45, 16 December 2010 (UTC)

I now realize why you did it; here losing the paragraph indention causes no harm, so it’s no issue. I’d still prefer in the future that formatting be left intact; I deal with the numbering by looking at a preview or by viewing the current version in a new tab or new window. I concede this doesn’t help with unordered lists. JeffConrad (talk) 08:58, 16 December 2010 (UTC)

1. The logic behind this "schizoid" approach has been thoroughly explained (and of course, you're welcome to disagree with it). Please see point 4c in my above 16:09, 14 December 2010 (UTC) response to you.

2. Okay, you appear to have misunderstood what we mean by "correct and proper." We aren't suggesting that one style is correct and proper and the other is incorrect and improper. The point is that both "this" and “this” are legitimate, widely accepted English-language constructs.

3. See 1.

4. I still don't understand your counterpoint regarding paragraph indentation.

The substantive difference is that straight quotation marks are common (and therefore no source of shame or embarrassment) in online media.

5. Likewise, I regard the purported advantages of Wikipedia using curly quotation marks as insubstantial. As I've noted, the discussion's participants (including you and me) are highly biased and unqualified to formally gauge consensus (or the lack thereof).

5a. "That quotes occur more frequently in many articles" was mentioned in a reply to you, written by JohnBlackburne, 13:10, 15 December 2010 (UTC). This is one distinction of several. (See 1.)

5b. See 1.

5c. I lack first-hand knowledge of the issue on which this argument is based.

5d. In my opinion, "ILIKEIT" (or perhaps "TRADITIONALMEDIALIKEIT") describes the main argument in favor of curly quotation marks.

6. I don't object to such an attempt. —David Levy 06:45, 16 December 2010 (UTC)

1, 3. I can’t find a 4c above that seems to address this; can you give me a text string that will find it?

2. No dispute on “widely used”; any judgmental term is a matter of opinion. Incidentally, I’ve never said that straight quotes were incorrect, though by reference to numerous style guides, I’ve probably implied it.

4. I’ve simply said that indention or lack thereof has nothing to do with a purported difference between print and online media; it’s a personal preference in either one.

5a. It looks as if he did, if one includes apostrophes (which probably is reasonable). Though I don’t agree with all of his assertions, I agree that the apostrophe probably occurs more commonly in most articles than most other non-ASCII characters. Though I agree that a conversion program (were one needed) that would correctly handle nested quotations and apostrophes would be nontrivial, I doubt that it would be as difficult as suggested.

5b. I don’t see how this relates to 1, but it might be more obvious if I could find the 4c you cite.

5c. A search for Alzheimer's in an article that had it as Alzheimer’s using a browser other than Safari or Chrome could fail. I’m not sure this is a real issue, but concede the possibility.

Assuming that the competing browsers fall in line in this respect, which behaviour will prevail I wonder, the smart search or the dumb search? Also, note that merely typing ‘a’, ‘l’, ‘z’ in a modern browser is sufficient to find Alzheimer’s on this page.

5d. Preference is of course one element, but so is readability. As I mentioned, the examples I gave above are so short that the differences are less obvious than they might be if they extended for several paragraphs, and again, the differences are less obvious in Arial than some other typefaces, especially any serif face. But basing a decision solely on the Arial face (which I assume is the default for most users) is like the program that tracks exactly seventeen salesmen.

The readability issue is not simply something I invented; “traditional” media and type designers aren’t blockheads, so their preference is likely more than capricious. Much has been studied on the consistency of patterns in type, with everything I’ve read finding that disruptions in the patterns impede the process. It’s much the same as with the most readable line measure and leading; some have been found to work better than others (yeah, I know we deviate from traditional practice in much online publishing—and it shows). I no longer have access to the half-dozen or so books on typography that I’ve read, and I haven’t reviewed the topic in 20 years, so I can’t say whether thinking has changed. For example, at one time, it was thought that the lack of serifs to guide the eye made sans-serif running text, and many guides accordingly recommended reserving it for short passages such as titles and headings. I still share that opinion, but I don’t know if the folks who study such things still hold that view. Here is perhaps one difference between print and electronic media (“online” is irrelevant): on a low-resolution screen, such as a “smart” phone, a sans-serif face often displays better because it is simpler. Aside from that, however, I just don’t see the asserted distinction between “traditional” and “online” (or whatever) media; in particular, except for pagination, why should the appearance of material in rendered HTML be different from material in a PDF document when they’re both viewed on the same display? JeffConrad (talk) 08:58, 16 December 2010 (UTC)

1, 3. Here's a text string: From a practical standpoint, another material difference

4. I disagree. Paragraph indentation is widely regarded as a standard style convention in the relevant types of print media and typically (though not always) omitted from the equivalent online publications. This deviation likely stemmed from technical limitations, but the end result is a new style convention.

5b. Indeed, it relates to my previous explanation of why the resultant workload would far exceed the current one.

5d. As you allude to above, much has been written on the differences between text printed on paper and text appearing on computer screens (i.e. that unlike paper, sans-serif typefaces generally are more legible on computer screens).

I don't know whether anyone has formally studied the use of quotation marks in electronic media, but it seems entirely possible that the same principle applies. I certainly don't assert that it does, but I regard it as unreasonable to assume the contrary.

I will note that the straight quotation marks appear clearer on both my laptop computer's screen (running Mozilla Firefox with the default text settings) and my smartphone's screen (running the stock Android browser). Unfortunately, I have no interlaced display with which to perform a comparison.

I know with absolute certainty that straight quotation have been widely adopted online, including by highly reputable publications utilizing curly quotation marks in print for many years. To me, this suggests the absence of a problem. —David Levy 16:03, 16 December 2010 (UTC)

We are directed to follow the article sources, most of which are printed and use typographic quotes. If (for some reason that seems to fall outside the MoS guidelines) we feel the need to look at on-line publishers then we should look at those with similar content, e.g. Britannica (which uses typographic quotes). Bear in mind that Wikipedia users may want to print articles or incorporate them in other printed material. News sites, on the other hand, largely have “disposable” content—they are not a good model for Wikipedia. — Wrapped in Grey (talk) 18:19, 14 December 2010 (UTC)

I don't know whether we cite more print sources than online sources, but that's irrelevant. The principle to which you refer is intended to apply to the things about which we write (e.g. the spelling and capitalization used for someone or something's name), not the general typographical styles that happen to be employed by these publications. (Otherwise, shouldn't we indent our articles' paragraphs?) Such broader analysis is reasonable, but only when examining the most similar high-quality sources available (i.e. other websites).

Britannica is similar to Wikipedia in the respect that it's an encyclopedia. Other websites are similar in different respects. Still other elements are shared only with other wikis.

In my opinion, a more relevant criterion than "disposable vs. non-disposable" is "static vs. dynamic." Practices compatible with a "typeset-it-and-forget-it" medium don't necessarily work as well here. —David Levy 19:28, 14 December 2010 (UTC)

I’m not sure why the proposals that we have had so far have been so tentative—merely allowing typographic quotation characters instead of just categorizing them, with em-dash etc., as preferred. A ‘half-way house’ proposal (e.g. per Eng. var.) requires more management which in this case is unnecessary: unlike Eng. var., use of typewriter characters was the result of a limitation, not a preference, and, as demonstrated by the successful use of other typographic characters in Wikipedia, that limitation no longer exists. Jeff stated above “we have a problem”; indeed, this facet of the current MoS has just burnt 100k+ of discussion—maintaining the status quo is not an option. — Wrapped in Grey (talk) 08:23, 14 December 2010 (UTC)

Because both forms of quotation marks are correct and proper. "--" is not a real em dash but " is a real quotation mark. Banning straight quotes would be just as bad as banning curly ones. Darkfrog24 (talk) 11:32, 14 December 2010 (UTC)

Not according to that rather nice encyclopaedia Wikipedia: " is a double prime mark; it “should not be confused with the double quotation mark” and its use as such is a typewriter convention. — Wrapped in Grey (talk) 12:02, 14 December 2010 (UTC)

Not quite so. The double prime mark is ″ (U+2033 DOUBLE PRIME = seconds, inches); " is U+0022 QUOTATION MARK. The Unicode standard says of the latter: ‘neutral (vertical), used as opening or closing quotation mark’; ‘preferred characters in English for paired quotation marks are U+201C LEFT DOUBLE QUOTATION MARK & U+201D RIGHT DOUBLE QUOTATION MARK’. A. di M. (talk) 12:19, 14 December 2010 (UTC)

As I hoped was clear, I was sourcing information from WP articles including the one on Typewriter conventions; typewriters know nothing of U+xxxx. Here’s a direct quotation: “typewriter practices ... include ... the use of prime marks (or "dumb quotes") as quotation marks”. — Wrapped in Grey (talk) 14:00, 14 December 2010 (UTC)

Even without knowing anything about U+xxxx, the ambidexterous quotation marks found on typewriters AFAICT are nearly always vertical, whereas the prime signs used in all typeset mathematical texts I recall ever seeing are always slanted. (I think they originated as italicized sans-serif Roman numerals, as some people, when hand-writing, after f′, f′′ and f′′′ will go on with f^IV and f^V.) A. di M. (talk) 16:56, 14 December 2010 (UTC)

Yes, this is my experience too, so maybe the Typewriter article should be updated with a better description of the " character; however, the point being made—that " is not “a real quotation mark”—still stands (the citation referenced in the typewriter article in fact refers to them as “dumb” quotes and describes their use as a “crime”).

The straight double quote (U+0022) is unquestionably in common use; whether it’s “correct and proper” is a matter of opinion, though I suppose if enough people use it, it’s tough to consider it “incorrect” or “improper”. The double prime mark throws even many typographers, but it’s exactly as A. di M. states. If you want to get it right, it’s important to keep these characters straight. Most folks don’t; it would really be nice if the character links below the edit box included the character names in the titles (“tooltips”). It’s easy to confuse a double prime with a straight double quote (He said, "1° 2′ 3″ "; cf. He said, “1° 2′ 3″ ”) which is perhaps an argument not to use both in the same article; ditto for single prime (U+2032) and straight single quote (U+0027). Incidentally, for another example of a quality source (arguably, a very high quality source) that uses typographical quotes for online material, see The Chicago Manual of Style Online. Who’d have thought? JeffConrad (talk) 12:48, 14 December 2010 (UTC)

1. While I've expressed a clear preference for the use of straight quotation marks, I agree that a complete changeover to curly ones would be preferable to the needless inconsistency proposed. Let's not transform every article into a quotation mark battleground.

2. Why is maintaining the status quo not an option? —David Levy 16:09, 14 December 2010 (UTC)

Because this thread is now 130k bytes long, still with no conclusion and we do want a repeat performance. The justification in the current MoS is disputed (reliability of entry of these non-ASCII characters opposed to other, preferred non-ASCII characters); the intention of the current MoS is unclear (recommended does not mean obligatory). If nothing else, these aspects of the MoS need to be improved.

BTW, here’s an even easier way to enter the commonly-used non-ASCIIs (on an English keyboard). — Wrapped in Grey (talk) 17:59, 14 December 2010 (UTC)

Thanks for clarifying. I certainly don't oppose clarifying the MoS entry (irrespective of the style determinations made). —David Levy 19:28, 14 December 2010 (UTC)

Note: There are three options:

1. To require all articles to use curly quotation marks. There is plainly no consensus for that; many of the objections above are to that option; I would strongly oppose it myself - and I don't see very much support for it.
2. To require straight quotation marks for all articles. I think the persistence and recurrence of this thread demonstrates that this is not consensus.
3. To require neither. Both of the above require non-consensus writing in the guideline; this doesn't. David Levy's position seems to me contrary to the first section of this page: we require consistency within an article, not necessarily across Wikipedia. Since we cannot agree on consistency across Wikipedia, we must fall back on consistency within an article.

Accordingly, I have edited minimally, removing exclusivity (on the grounds that it does not reflect consensus) and requiring consistency within articles. The arguments for straight quotes are still there; somebody who wants curly ones should feel free to respond. Septentrionalis PMAnderson 19:24, 14 December 2010 (UTC)

As I noted above, "consensus" means more than "agreement among persons discussing something." Among editors of this wiki (few of whom will ever see this page), straight quotation marks overwhelmingly predominate.

We don't necessarily require consistency across Wikipedia, but we prefer it when there are no English variety issues and a particular method is widely favored. —David Levy 19:36, 14 December 2010 (UTC)

But that rules out consensus on position #1 above; what is your evidence that there is consensus to prohibit curly quotes, which is what position #2 says? I don't believe most editors care. Without such evidence, edits like this which impose such a prohibition are "edits which do not reflect consensus" - and we should strongly consider dispute resolution about such edits. Septentrionalis PMAnderson 19:47, 14 December 2010 (UTC)

In the meantime, I stand by my position: these pages could be useful, but as they stand, they are a waste of time, and marginally harmful to the encyclopedia. There is policy about rules like that. Septentrionalis PMAnderson 19:47, 14 December 2010 (UTC)

1. I believe that the arguments against the use of curly quotation marks are far stronger than those for their use.

However, I'm hopelessly biased. Involved editors (including you and me) shouldn't formally evaluate consensus.

2. You believe that the MoS (and this entry in particular) wastes time? That's how I feel about an unnecessary style introduction that virtually guarantees an increased cleanup need for the remainder of the affected articles' existence. —David Levy 20:03, 14 December 2010 (UTC)

You've claimed this "increased cleanup cost" several times. You've never explained what it might be. What we have now is a clean-up cost hunting down all the EEEvil curly quotes and getting rid of them - and an explanation cost justifying this to everybody who likes them. Removing "exclusively" would remove all this cost. What cost would replace it? Septentrionalis PMAnderson 20:17, 14 December 2010 (UTC)

I've elaborated several times. Most of our editors, including those unfamiliar with the MoS, use straight quotation marks (and will continue to do so, no matter what). A page written with curly quotation marks is likely to frequently receive straight quotation marks (to the extent that an advocate of curly quotation marks has proposed routine monitoring by "shepherding editors").

No matter what style[s] we select, some cleanup of inconsistent usage within articles is unavoidable. But the allowance of curly quotation marks (either exclusively or optionally) can only cause the need to soar. —David Levy 20:47, 14 December 2010 (UTC)

“Shepherding editors” already exist; they clean-up all sorts of MoS, grammar, and other problems left by less-experienced editors. Cleaning-up w.r.t. to this single aspect of the MoS is of little significance in that realm. On the point made about (not) paragraph indenting, the information that is currently entered as article text is sufficient to allow paragraphs to be formatted either way (indented or as currently, with leading). It would be possible to allow selection between the two styles as a user preference. Similarly, display of quotation characters as typographic or typewriter could be made a user preference but probably only if the former were entered in the article text. — Wrapped in Grey (talk) 23:45, 14 December 2010 (UTC)

I certainly don't mean to belittle editors who take the time to monitor articles and address problems that arise. Greg L, in his use of the terms "shepherding authors" and "shepherding editors," has proposed a somewhat different concept in which what constitutes appropriate behavior on the part of general editors depends upon whether a "shepherding author/editor" has staked a claim on a particular article (which others would be left to somehow divine, unless "This article's content is controlled by User:_____" tags were to be placed on talk pages).

Regarding paragraph indentation, I was referring to our default layout (which a vast, vast majority of visitors will see). If our goal were to emulate print media, we surely would indent paragraphs (possibly excepting the first, depending on the style used) by default.

I think that the user script idea is an excellent one, and while rendering straight quotation marks as curly ones would require fancier code, I don't see why it shouldn't be feasible. —David Levy 00:19, 15 December 2010 (UTC)

How the hell would such “fancier code” tell whether the ' in 'cause is an apostrophe (in which case it should be rendered as ’cause) or an opening quotation mark (in which case it should be ‘cause? A. di M. (talk) 15:07, 15 December 2010 (UTC)

By "fancier," I meant "capable of analyzing context instead of blindly converting x to y when encountered." I had in mind a need to detect a mark's location (to determine whether it opens or closes a quotation). The issue that you've cited is much trickier, I'd imagine. But I'm not a programmer, so please take these largely off-topic comments with a grain of salt. —David Levy 15:25, 15 December 2010 (UTC)

Some more thoughts: the choice of a handful of sites to use curly quotes is not a random choice but one that ties into an overall, very coherent design that includes choice of (serif) font, use of colour, text sizes, and many other choices. WP simply cannot do that as through user preferences and customised stylesheets users can specify most of these things. I.e. the fact that the NYT uses curly quotes and it makes sense for them does not mean it makes sense here.

We are nothing like print: there are so many things you see in print you don't see here, like paragraph indentation, pages and page numbers (which you also see on some sites), and many things you see here you don't see in print. And there is no single standard for print: just pick up three newspapers in this country, the Guardian, Telegraph and Sun say, and compare them.

In theory there could be a user preference, as there is on the Chinese WP for Simplified or Traditional characters, but it would be a massive amount of work, modifying every article by hand, and rewriting the software to handle it, so much so it will never happen.--JohnBlackburne^words_deeds 01:06, 15 December 2010 (UTC)

Please stop performing contested edits to an MoS entry that's currently the subject of a discussion in which you're involved. I wish to modify it too, but let's try to resolve the dispute, perhaps with the assistance of uninvolved parties. —David Levy 19:49, 14 December 2010 (UTC)

It has been discussed for weeks. There is no consensus for the present wording. Restoring it is editing which does not reflect consensus. If this section is still here in the new year, I will consider what dispute resolution is adequate. Septentrionalis PMAnderson 20:17, 14 December 2010 (UTC)

I don't see how why you think there is consensus against the current version. In the discussions on the various alternatives suggested the consensus, simply going by the number of oppose statements, is clearly against any change. There are few of these in this last sub-section, perhaps as editors having expressed their view once do not feel the need to restate their opinions on essentially the same issue. Nor is this some irrational prejudice by editors: the main reasons are in fact given in the article, though we have come up with some more in this discussion such as it breaking links because of a mismatch in styles between two articles.--JohnBlackburne^words_deeds 20:29, 14 December 2010 (UTC)

What I said is that there is no consensus for the present text - repeatedly. There are three states: consensus for something, consensus against something, and no consensus either way; policies and guidelines should require the first - and anything which does not is an essay. There is no consensus for any prohibition.

Statements against a requirement of curly quotes (like most of the opposes here) affect whether there is consensus for #1 above, the prohibition of straight quotes - which there isn't; they cannot affect whether there is consensus for anything else, unless they express an opinion on something else. Septentrionalis PMAnderson 20:47, 14 December 2010 (UTC)

Is there a particular reason why you regard yourself as uniquely qualified to gauge consensus (or lack thereof) in a discussion of which you're an active participant? —David Levy 20:52, 14 December 2010 (UTC)

There is consensus, or at least whenever anyone has put a change to a poll it has been convincingly opposed. That is consensus for the current version, unless I am seriously misjudging the intent of the participating editors.--JohnBlackburne^words_deeds 21:24, 14 December 2010 (UTC)

Again, involved parties (including you and me) shouldn't attempt to evaluate consensus in a binding manner. —David Levy 20:47, 14 December 2010 (UTC)

Thank you for summarizing exactly what you and John Blackburne are doing (you two are very involved parties and revert-warring to bind the text to your preferred version). Septentrionalis PMAnderson 00:41, 15 December 2010 (UTC)

As noted in the summary, my one edit served only to restore the discussion pointer that you added (removed by John via his second reversion). —David Levy 01:00, 15 December 2010 (UTC)

Please address this. If you made a mistake, that's fine, but I want to resolve any confusion. —David Levy 02:57, 15 December 2010 (UTC)

But on the merits: Let us review the bidding. Several editors began this by opposing the exact position that straight quotes should be required; none of them have changed their mind. Your precise position, that we ought to require straight quotes, and that sources are irrelevant, seems to be upheld solely by the two of you. This is consensus? Let's have specifics. Septentrionalis PMAnderson 00:41, 15 December 2010 (UTC)

Come again? What is the significance of the editors not changing their minds? Who said that sources are irrelevant? And what gives you the idea that John and I are alone in our opinions? I read the above several times, and I honestly don't know what you're asking me to defend. —David Levy 01:00, 15 December 2010 (UTC)

Start with the proposition that even a majority of those discussing this support exclusive use of straight quotes; we can get to what supermajority would be consensus in the presence of active dissent later. As far as I can tell, nobody but you two and perhaps Headbomb has actually endorsed that position; several editors have weighed in opposing exclusive use of curly quotes which nobody suggested (until you claimed it as your second choice), but that's a different question. Septentrionalis PMAnderson 01:32, 15 December 2010 (UTC)

Firstly, when did I cite the claim that "a majority of those discussing this support exclusive use of straight quotes"? I've explicitly stated that "'consensus' means more than 'agreement among persons discussing something'" and noted that I regard the relative strength of the opponents' arguments (in my highly biased assessment) as a key factor. David Levy 02:57, 15 December 2010 (UTC)

If you do not claim majority, then either you claim

• A minority is consensus,or
• You do not have consensus.

The first is novel policy, the second concedes that the minority language should not stand. (It should not be replaced by other minority language, it should simply be removed.) Whether you have even a majority follows; it is necessary but not sufficient. Septentrionalis PMAnderson 03:06, 15 December 2010 (UTC)

As I wrote above, I haven't cited such a claim. A simple vote count yields a sizable majority opposed to the proposals, but that isn't how consensus is gauged. Numbers aren't irrelevant, but it's the arguments' strength that matters most. I've opined that the opponents' arguments are significantly stronger overall (while simultaneously noting that I, like you, am hopelessly biased and unfit to formally gauge consensus). —David Levy 04:17, 15 December 2010 (UTC)

Secondly, from what do you derive the idea that all opponents apart from John, Headbomb and me have strictly opposed "exclusive use of curly quotes"? Edokter, 86.173.169.220, Cybercobra, Woodstone, Adabow, Mclay1, Arthur Rubin, SlimVirgin, David Fuchs, Johnuniq, Jc3s5h, SarekOfVulcan, Oknazevad and Trovatore appear to oppose any use of curly quotation marks (possibly apart from special exceptions) in the encyclopedia. Some cited a desire to maintain intra-article consistency and concern that the acceptance of two styles would trigger edit wars. —David Levy 02:57, 15 December 2010 (UTC)

Clarification, I vehemently oppose any and all use of curly quotes, except when mandated by context (like when discussing curly quotes themslves). Headbomb {talk / contribs / physics / books} 20:03, 15 December 2010 (UTC)

Incredible. Edokter's reply is here; it is a vivid and clear argument that he doesn't want to use curly quotes (and I don't think he should have to). It says nothing about the point at issue: that other editors should be compelled not to; most of this list of "supporters" is equally fantastic. Is Mediation acceptable to you? This is beyond the reach of reason. Septentrionalis PMAnderson 03:19, 15 December 2010 (UTC)

1. I'm utterly baffled by your interpretation of Edokter's comments. Edokter cited legibility issues, "breaking Wiki markup and search" and "break[ing] links to article titles and section headings with quotation marks." How do you interpret these concerns as applicable strictly to Edokter's editing habits?

2. I'm 100% open to mediation, if you believe that it would help. But I honestly don't understand why you've taken such a confrontational, accusatory approach (from my perspective, at least). I'm sincerely attempting to participate in civil, constructive discourse, but you seem to perceive my actions as outrageous and offensive. Even my reinsertion of your discussion pointer has been inexplicably labeled "revert-warring to bind the text to [my] preferred version." Are you having a bad day? —David Levy 04:17, 15 December 2010 (UTC)

I agree that the cited references are printed, but let’s face it—they aren’t likely to give the books away by putting them on the web. I keep seeing “we’re not like print”, but I’ve not been shown why this is a convincing argument against typographical quotes. I also see it mentioned that the examples with typographical quotes are on pages with serif type; since I said this myself at least a half dozen times before anyone else mentioned it, I find it unsurprising. Is the intent somehow to say that while typographical quotes are inappropriate for sans-serif type? If so, why? Or is it simply that the differences are harder to see with sans-serif type (especially the Arial that is the default for many users)? If so, what is the problem? Typographical quotes would display properly for those who specified serif font, and would look essentially the same as straight ones for the majority of people who don’t.

I don’t buy the example of the lack of paragraph indention as supporting anything; what in the world has that got to do with this? Not all printed sources indent paragraphs, and many web pages do. One of the worst things to happen to the web was the commingling of content and presentation, which ultimately led to the development of CSS. In fact, the HTML 4.01 strict DTD disallows most presentational elements and many purely presentational attributes. In addition to eliminating a lot of messy code, this development recognizes that material may be repurposed and that the presentation may change, and such a change is easily managed when content and presentation are kept separate. When the content is viewed only through the lens of a particular presentation, and entered accordingly, the change may be more difficult. With straight quotes, a change as simple one to a serif typeface (which some users do) makes for ugly presentation. Though we’re far from agreement on anything that would preclude such a problem, proscribing something that would partially avoid it doesn’t make much sense.

For reasons I cannot comprehend, there is some obsession with quotes, the likes of which I’ve never encountered elsewhere. As several of us have said repeatedly, were we to treat all non-ASCII characters (some of which have no keyboard shortcuts) the same as we do quotes, I might find some of the arguments somewhat more convincing. But we don’t and I don’t. I shall repeat yet again: though we have cited many well-regarded style guides that show only typographical quotes, or in some cases specifically urge against straight quotes, I have yet to see one that urges against typographical quotes. In fact, every style guide that I have, save one (CMoS 16, which deprecates the use of straight quotes as “unsophisticated”) treats straight quotes as if they did not exist.

David, you said, “the circumstances under which a wiki is written justify the use of straight quotation marks”, suggesting that a wiki is somehow different from other media. I’m not quite sure what you mean by this; is it that typographical quotes are to difficult to enter? Or is it something else?

I am perhaps more sympathetic to David Levy’s concern about subsequent editors inserting straight quotes into articles with typographical quotes. While I think this assumes facts not in evidence, it’s probably not an unreasonable concern. But new editors make edits all the time that conflict with articles, typographically, stylistically, in use of terminology, and especially in the style, formatting, and quality of references, yet we don’t seem to bat an eyelash. I simply cannot understand this; compared to most other inconsistencies I encounter, keeping quotation marks consistent is a trivial undertaking. Perhaps it depends on the type of article: for one with many quotations and that changes frequently, maintenance would require more effort, but for an article with few quotes or that was fairly stable, little would be required. As I mentioned, editors need to fix other inconsistencies introduced by new editors all the time. A trivial example: many times I have had to insert nonbreaking spaces between values and units, only to have the same editors almost immediately make the same mistake in another edit. Less trivial to clean up are edits that ignore the prevailing reference style; in some cases, cleanup is a nightmare, especially if the changes have been made at various points in the article. Though either action clearly violates MoS, we don’t chastise such editors and seem to take less vitriolic objection.

I don’t sense much consensus of any kind, much as stated in the register for the MoS. Conceptually, I completely agree with A. di M.’s reminder that “Consensus is ultimately determined by the quality of the arguments given for and against an issue, as viewed through the lens of Wikipedia policy, not by a simple counted majority”, but would have no idea how to apply here. Those of us who support some sort of change find most of the arguments against it insubstantial, and would not be surprised if those who oppose feel likewise. We continue to rehash the same arguments, seemingly to no end. I wonder if it would be worth at least attempting to summarize the points on which we apparently agree to disagree. JeffConrad (talk)

I sincerely appreciate your effort to understand everyone's views, Jeff, but most of the points/questions that you've raised already have been addressed/answered. In particular, I've cited material differences between Wikipedia and print media, expounded on the "paragraph indentation" example, compared the resultant inter-article inconsistency problem to that which currently exists, and explicitly clarified what I meant by "the circumstances under which a wiki is written justify the use of straight quotation marks." (Please search the page for that text.) —David Levy 04:17, 15 December 2010 (UTC)

With your latest replies, I agree that most of the points have been answered. Whether they have been addressed is another matter, but it of course depends on what is meant by “addressed”. I still await any substantive response to why quotes should be treated differently from many other non-ASCII characters with regard to ease of entry, displayability, or article readability. But perhaps that is just another point on which the two camps agree to disagree. I think think it would be worth trying to see if we could at least agree on the points on which we agree to disagree. So much has been said that I’m not even sure I remember everything that I’ve said. JeffConrad (talk) 09:27, 15 December 2010 (UTC)

I've explained my (and others') perceived distinctions between quotation marks and non-ASCII characters (such as dashes) in great detail. I personally make no claims regarding differences in displayability or article readability (though others disagree). And yes, I've also explicitly agreed to disagree on these points.  :) —David Levy 12:33, 15 December 2010 (UTC)

Quotes are different from non-ASCII characters for two main reasons. First with e.g. dashes to use a hyphen in place of an n-dash or minus is simply wrong. That means it's simple to devise a policy for dashes – use the right ones – and changes made to fix problems with quotes are uncontroversial, so they can even be done by bots. Conversely there is nothing wrong with straight quotes or curly quotes, they are just different stylistic choices. Second quotes are far more common, appearing in almost every paragraph, not just around quotations but used for emphasis, for contraction and to indicate possessives. These two differences mean changing the policy will involve much more work, because of the number of paragraphs affected and the relative complexity of the issue, so they are substantially different from other non-ASCII characters in this regard.--JohnBlackburne^words_deeds 13:10, 15 December 2010 (UTC)

make default

How about if we just make something so the default is curly? I'm not even joking. I bet more than 50% of the time, when you're using quote marks, you're using them for a quote. So just have them display as quotes. It's just some code, right? Then put the fancy toolbard button on there for when you want to keep the straight ones. P.s. I confess to only reading about a 1/4 of the discussion. Not a troll, honest.TCO (talk) 15:50, 15 December 2010 (UTC)

Do you want the default to be “ or ” ? They look the same on my preview, but I copied a left curly and a right curly from WP:PUNC. It's usually easy for a human to distinguish the inside of quotes from the outside, but not nearly so easy for software, so one misplaced quote marks makes all the opening quote marks look like closing quote marks and vice versa. My AWB software tries to identify quotes so it can avoid making corrections that would be made if it weren't a quote, but it gets confused about a third of the time by things like 7" records, or by extra quote marks left by accident (usually in a citation template). The opposite would presumably be easier: change software so curly quotes appear as straight. Art LaPella (talk) 02:39, 16 December 2010 (UTC)

I donno the details but MS Word seems to have figured this out for at least 10 years and logistically it is not that tricky. Sure you will still have an occasional glitch, but so what. a lot of the manual stuff glitches all the time, too. TCO (talk) 02:47, 16 December 2010 (UTC)

For the last 10 years MS Word has turned 'cause into ‘cause regardless of whether it’s the beginning of a quotation or a contraction for because, in which latter case ’cause would be correct. It has also turned 2" into 2” regardless of whether it’s the end of a quotation or a measurement in inches or arcseconds, in which latter case 2″ would be correct. A. di M. (talk) 15:05, 16 December 2010 (UTC)

It might be possible to have a “smart” quote feature that did the conversion on the first preview, but care would need to be taken to avoid problems like Art mentioned, and care would need to be taken not to alter quotes in markup. Conceivably, the feature could be optional, or could be run manually via a button on the editor toolbar by editors who prefer to use typographical quotes; I think we’d need a much better hit rate than 67%, though. I’m surprised the rate Art mentions is so low; I’ve written scripts that added markup to ASCII files, and probably had about a 95% success rate. Of course, I did not need to deal with markup, so perhaps I would have had a lower success rate with marked-up WP source. It’s not difficult to prevent one misplaced quote from messing everything up, but it’s also tough to catch everything, such as Art’s example with the “inch marks”. As Art mentions, the reverse process is easy. JeffConrad (talk) 03:49, 16 December 2010 (UTC)

Close RfC

At this point all that happens is people writing essays about what people really meant and their personal bewilderment that essays are being written about what people really meant. The ship has sailed, let's close this RfC. Consensus is pretty clear against the use of curly quotes.

Original poll (Oppose to support 8:4)

For

• A. di M.
• Wrapped in Gray
• JeffConrad
• Darkfrog

Oppose

• Edokter
• JohnBlackburne
• IP
• Headbomb
• CybraCobra
• Artur Rubin
• SlimVirgin
• David Levy

The Arguments for are "Since we allow endashes, why not curly quotes? I like curly quotes. They are typographically superior and do no harm.". The argument against are "We have consistency now, this will break it, and will bring countless edit wars, and make articles hell to maintain. Straight quotes are a proper quotation mark, thus the parallel with double hyphens (which is improper) and emdashes do not apply. The typographical edge they have over straight quote is established offline, but not online. I hate curly quotes."

The following polls and subpolls are basically reformulation of the same things in the hopes of changing things, but which failed to do any better than the original, and debate on technical/software things. While I'm in the oppose camp myself, I don't think I'm in the wrong when I'm saying that the opposition for this is pretty intense, much more so that for their support, and that none of pro-"curly quotes" arguments even addressed any of the oppose's concerns. When taken the ILIKEITE/IDONTLIKE IT arguments from the pot, their remains very little of them for the "pro" side, and lots of them for the "anti" side. So please let's close this down, consensus is pretty clear that curly quotes do not have consensus for use on Wikipedia, and that allowing them is not a good idea. Let's not escalate this to 20 archives or debate on this like we do with the IEC prefixes, or have bans and blocks resulting from ARBCOM cases like with the date delinking fiasco. Headbomb {talk / contribs / physics / books} 07:38, 16 December 2010 (UTC)

I somewhat disagree with your summary, especially the comments that the “pro” arguments are essentially empty, and that the concerns of those in opposition have been addressed. I agree that with 2–1 against, there certainly is little support for any of the changes specifically proposed. That said, there have been (and still are) many who think the current wording does not reflect consensus, as the register for the project page states. So perhaps the RfC has run its course, but I’d like to see the summary better reflect the points of contention. JeffConrad (talk) 09:08, 16 December 2010 (UTC)

I should add that we probably should mention your earlier comment about citation templates. I don’t use them except under duress, but many editors do, and as presently coded, they would not work well with typographical quotes. Like so many other things, I don’t think it’s dispositive, but unlike some of the other arguments, it is a valid issue. I would even agree that it really has not been addressed. JeffConrad (talk) 23:01, 16 December 2010 (UTC)

Yes, the register should be amended to take into account recent developments. Headbomb {talk / contribs / physics / books} 12:08, 16 December 2010 (UTC)

I somewhat disagree with the summary and the disregard of WP policy: Consensus is determined not by a simple counted majority. — Wrapped in Grey (talk) 09:27, 16 December 2010 (UTC)

If you had bothered reading my post, you'd have seen that it was not a "simple head count". Headbomb {talk / contribs / physics / books} 12:07, 16 December 2010 (UTC)

“polls should always be regarded as structured discussions rather than voting” — Wrapped in Grey (talk) 12:27, 16 December 2010 (UTC)

No, it was not a "simple head count", it was a quite creative head count. Where is my name, for instance?—Emil J. 15:18, 16 December 2010 (UTC)

Some arguments are missing, as well as names. For example, the fact that most reliable sources including on-line ones such as The Times, The New York Times and Britannica use them. This doesn't look like “consensus against” to me, it looks like “no consensus”. (Now, what happens in this case? “The status quo stays” or “the MOS should be silent”? We should sort that out once and for all.) A. di M. (talk) 15:28, 16 December 2010 (UTC)

As I noted earlier most sources don't, or at least there are far more that don't than those three. But even if many more sources using curly quotes could be found I don't see how that is an argument for using them here. We do not know the cost to them of doing this (which I suspect is considerable for a daily newspaper), and there are many things we do which are different from the NYT, Times, etc.--JohnBlackburne^words_deeds 15:51, 16 December 2010 (UTC)

Indeed, I noted that professional sources (both print and online) publish completed texts, which are standardized via editorial hierarchies. We lack such a process, and our articles are never finalized. —David Levy 16:03, 16 December 2010 (UTC)

To clarify, are you suggesting that most reliable online sources use curly quotation marks? —David Levy 16:03, 16 December 2010 (UTC)

David, I can’t speak for A. di M., but I think most of us readily concede that that most web pages use straight quotes (as well as marginal typography in general). I’m not sure the lowest common denominator is the best guide, however. We’ve noted some quality sources that do care about typography, CMoS being one good example. They’re only one site, but they’re heavy hitters. JeffConrad (talk) 23:18, 16 December 2010 (UTC)

"Lowest common denominator" is loaded language. One could just as easily refer to websites using curly quotation marks as "pretentious," but I prefer to focus on substantive issues.

(To be clear, I'm not actually asserting that websites using curly quotation marks are pretentious.) —David Levy 23:40, 16 December 2010 (UTC)

It's not an issue of "most web pages" (many of which only use straight quotes because the author didn't know the difference at all), it's an issue of "most reliable sources". Of course you won't find many curly quotes in Facebook walls, but we don't want to use it's as its or b4 as before because they're so common out there. A. di M. (talk) 03:31, 17 December 2010 (UTC)

Straw man. We're referring to reliable, high-quality online sources (most of which use straight quotation marks). "Facebook walls" are irrelevant to the discussion. —David Levy 03:50, 17 December 2010 (UTC)

The MoS directs us to “high-quality sources”, not “high-quality online sources”; if you wish to base arguments on the latter then you must first change, through consensus, the MoS accordingly. — Wrapped in Grey (talk) 06:35, 17 December 2010 (UTC)

Firstly, you're responding out of context. The above exchange can be summarized as follows:

• A. di M apparently claims that most reliable online sources use curly quotation marks.
• JohnBlackburne replies that this is incorrect and I inquire as to whether it was A. di M's intended statement.
• JeffConrad acknowledges that "most web pages" (written without qualification) use straight quotation marks.
• A. di M. reenters the thread with a dismissive reference to "Facebook walls," misleadingly implying that this is the sort of web content that opponents seek to cite.
• I reply to note that we've been discussing "reliable, high-quality online sources" (which is true regardless of whether other high-quality sources should be considered in this context), so "Facebook walls" are irrelevant.

Secondly, you're disregarding previous discussion (pertaining to a principle's [in]applicability to this situation), to which you haven't replied. I'm quite disappointed, as you appeared to express genuine interest in my interpretation (upon which I elaborated at your request). —David Levy 09:35, 17 December 2010 (UTC)

(ec) Among reliable online sources, only Britannica, The New York Times, and possibly The Times use curly quotes. — Arthur Rubin (talk) 04:08, 17 December 2010 (UTC)

What evidence do you have for such a sweeping statement? (And does that mean that this is not a reliable source?) A. di M. (talk) 11:46, 17 December 2010 (UTC)

John, I have no idea about the cost, but quite honestly, it’s not a question I’d even think to ask, because it’s something I’ve done for 25 years and I never even think about it—I just do it. It’s slightly more effort, but it just isn’t the Herculean task that many seem to envision. I use Windoze, so from the keyboard, it’s Alt + four keystrokes. For other web pages, I use the HTML entities; for sure, it’s a bit more work than typing a typewriter quote, but at least I don’t have bad typography to detract from I’ve written. I’ll concede that many people probably don’t even notice. JeffConrad (talk) 23:18, 16 December 2010 (UTC)

Right, you aren't considering the cost. It comes naturally to you, so it's difficult for you to see what all of the fuss is about. But most Wikipedia editors don't even know how to type curly quotation marks, let alone possess the inclination to do so.

Your belief that straight quotation marks constitute "bad typography" is one that few people hold in this context. —David Levy 23:40, 16 December 2010 (UTC)

The cost for a newspaper is the dedicated staff of copy-editors whose role is to turn raw copy into newspaper-ready content. Part of their job will be to enforce the house style, so making straight quotes curly or vice-versa as needed. WP will not need such staff (we don't have a schedule like a newspaper when the work must be done by) but the work will be done by someone at some point. It will be potentially be more work as a newspaper, which knows in advance what its style is, can enforce it on at least office based writers by putting MS Word on all machines set to use the correct style (of quotes and everything else), but we have no way of doing this.--JohnBlackburne^words_deeds 00:46, 17 December 2010 (UTC)

David, I’ve been using typographical quotation marks for a long time, so I suppose now it’s now almost habit, but wouldn’t say that it “came naturally”—I had to train myself to do it. Quite honestly, until this discussion began, I had never used the keyboard shortcuts for WP; I had always worried that the values entered would be Code page 1252 rather than Unicode, but the actual characters seem indeed to be Unicode. I never knew the shortcuts for quotes because MS Word takes care of them automatically; I tried them a few times here and they seem to work fine (for other than apostrophe, I find the character links faster). I will concede that I sometimes tend to assume that because I have no problem doing something that no one else should have a problem doing likewise, and perhaps this isn’t always reasonable. But I also think it’s sometimes more work to discuss why something can’t be done than to simply do it.

As regards “bad typography”, I guess it depends on which people and what is meant by “this context” (WP or web publishing in general). I don’t think I’ve ever seen a source that addressed the issue that held otherwise (most guides ignore the issue). So I’ll stand by my citation of CMoS 6.112. Now the importance of good typography is arguably debatable. “LCD” was a bit tongue in cheek, but only partially. Typography becomes pretentious when used as Mrs. Malaprop might employ it. For crying out loud: typographical quotes are essentially inverted commas—should they not match? JeffConrad (talk) 02:02, 17 December 2010 (UTC)

1. No one has claimed that the proposed actions "can't be done." We've cited reasons why we believe that they would do more harm than good.

2. By "this context," I meant web publishing in general. I've addressed the matter of style guides elsewhere in this discussion. —David Levy 03:50, 17 December 2010 (UTC)

The comment addressed effort, not harm, and was in response to suggested great difficulties, and simply said that the difficulties are as much imagined as real. I suppose I should have concluded with, “it’s sometimes more work to discuss why something is so difficult than to simply do it.” JeffConrad (talk) 04:57, 17 December 2010 (UTC)

We aren't merely arguing that the proposed changes would be difficult. —David Levy 05:45, 17 December 2010 (UTC)

I fully realize that. However, it would sure make the discussion easier to follow if we kept the various issues separate. We were discussing the difficulty of entering the special characters (which somehow comes naturally to me), and I was simply suggesting that to a large extent, some are making a mountain of a molehill. Of course there could be ramifications of the difficulty of character entry: editors just won’t do it in articles with curly quotes, leading to inconsistencies, massive cleanup efforts, and perhaps a triple-dip recession. I think the discussion stays more productive if we focus on the individual issues (as well as their interrelationships). JeffConrad (talk) 06:50, 17 December 2010 (UTC)

I respectfully disagree with your belief that we've exaggerated the issue.

Some of your messages seem to imply that such character entry is relatively easy to learn. I don't doubt that this is true, but we aren't in a position to teach it to most editors (who will never see the MoS and never even think of doing anything other than typing the straight quotation marks that they've used throughout most of their lives).

You want us to isolate this matter, but it directly relates to others. Quite simply, many of us believe that there is very little to be gained from such changes. So even if one views the aforementioned difficulties as minor, we see insufficient justification. We regard straight quotation marks as appropriate for use in our encyclopedia, see no need to deviate from them and foresee net disadvantages to doing so. —David Levy 09:35, 17 December 2010 (UTC)

I guess I can’t comment on ease of learning for anyone but myself, but I yet again make this observation: it’s no more difficult to learn for quotes than for any other special character. If the “learning is too difficult” theory holds, we’re limited to characters on the the typical QWERTY keyboard, which probably describes the bulk of web content, especially for nonprofessionals. Again, I just don’t see the problem. When someone mentioned the keyboard shortcuts for quotes, I decided to try the one for the apostrophe on Windoze, Alt-0146 (numeric keypad), and lo and behold, I got ’, apparently real Unicode. For practice, I guess I could try it again: ’1’2’3’4’5’6’7’8’9’10 ... five minutes ago, I couldn’t even spel xpurt—now I are one! I don’t mean to make light of this, but I must be missing the difficulty.

The other simple option is to use the character links below the edit box. If it’s not obvious what the characters links do (and I’m not convinced it is), we need to fix that, or the feature is useless. The current implementation of single quote is admittedly awkward, adding only pairs, so that one must be deleted, and a simple backspace won’t do it. A while back, there were links for individual right and left quotes; a link for a single right quote (apostrophe) should be added. The behavior with wikEd is even stranger: attempting to insert a pair of quotes before the s in quotes gives ‘quotes’. But these are minor software glitches that should be addressed anyway.

There also is another simple thing we could do; a user recently asked how to enter a dash, and was given keyboard shortcuts. It was suggested that this be included in the MoS, and there seemed to be no objection to this (I’m not sure the MoS is the right place, but wherever this might be included should be easy for a new editor to find). I suggested that we go even further and include a short table of the most common non-ASCII characters (like this). I haven’t included the Mac keystrokes because I don’t have a Mac, so someone who does would need to do it. Please don’t link to this as a reference because I don’t regard it as stable or permanent. JeffConrad (talk) 11:12, 17 December 2010 (UTC)

Now I suppose the argument again is that because apostrophes are more frequent than other special characters, we should worry about editors entering them incorrectly and not worry about the other characters. At least to me, a most strange line of reasoning.

I did not suggest that we isolate anything, but simply when we’re talking about difficulty of entry (or whatever) we focus on that, and then move to possible ramifications, perhaps also discussing how one might follow from the other. In other words,

• Issue one.
• Issue two.
• Issue three.

rather than IssueOneIssueThreeIssueTwo.

If most editors will never see the MoS (which may well be the case), then all manner of things that violate MoS will be added to articles. Many of these (e.g., a naked URL for a reference to Joe’s Blog) are far more egregious and more difficult to fix than a quote or two. It also seems that what is urged is essentially prior restraint; perhaps we should act before we’re faced with a mushroom cloud, but I don't see this possibility as quite so imminent as do others. If most editors will never see the MoS, I also assume they could not reasonably be chastised for using curly quotes. No disagreement that many see no benefit of changing, but there also are quite a few who are of the opposite opinion, and accordingly disagree with the summary that has been proposed.

I think it’s obvious that we both are preaching to the unconvertible here, and I think we should just admit that we are not going to agree on some of these things, and that continually beating it to death will do nothing but waste even more time. To many of us, there clearly is no consensus; I’m honestly not what we do with that. JeffConrad (talk) 10:57, 17 December 2010 (UTC)

1. The material differences between curly quotation marks and other non-ASCII characters have been discussed at great length. You're entitled to disagree, but continually expressing bewilderment (as though no one has even addressed the issue) is extremely frustrating and invites nothing other than pointless reiteration.

2. Again, we are not in a position to teach most users how to enter special characters. Typical editors will never even notice our advice; they'll simply type as they normally do. This means that they'll use straight quotation marks. Those of us who regard straight quotation marks as appropriate see the proposed changes as unnecessary and view any increased burden (referring primarily to the resultant cleanup, not to the education that few editors will ever receive) as unjustified.

And yes, relatively few editors will enter dashes either. But as previously noted, because their approximations vary wildly (e.g. one hyphen, two hyphens, surrounded by spaces, not surrounded by spaces), this is an element that we need to clean up anyway (to maintain intra-article consistency, at the very least). Conversely, almost every Wikipedia editor uses a particular style of quotation marks, so sticking to it means that few instances will require modification.

Indeed, editors shouldn't be chastised for writing in a manner inconsistent with the MoS (which most have never seen). But under the current setup, this rarely occurs in regard to the quotation mark style used (because almost all editors independently use the recommended style). The allowance of curly quotation marks in general prose would cause accidental MoS violations to frequently arise in affected articles (with the introduction of straight quotation marks creating intra-article inconsistency). And this doesn't even touch on the inevitable edit wars between users preferring conflicting styles and wanting their articles to contain them. —David Levy 11:58, 17 December 2010 (UTC)

The edit-war Armageddon could easily be addressed by following current general policy: unless the prevailing style of an article is seriously wrong, follow it unless there is consensus to change. Non issue.

As to the other issues, I find your responses every bit as frustrating as you apparently find mine. For the most part, we agree to disagree. Nothing I say is likely to change your mind, and nothing you say is likely to change mine, so we should close this RfC as suggested. The only remaining task is to arrive at a summary that represents a reasonable consensus of where we stand, where we differ, issues that we consider significant, perhaps issues we consider fairly minor, and perhaps whether there is any interest in a revised proposal. JeffConrad (talk) 02:10, 18 December 2010 (UTC)

1. No, it is not a non-issue. The principle that you describe often fails to prevent edit wars from occurring. And as you can see, whether there is "consensus" to do something can be disputed.
Additionally, these proposals lack such a stipulation (and the latter two actually encourage "shepherding authors" to replace straight quotation marks with curly ones, at which point the style would instantly become sacrosanct).
And this has no bearing on the use of straight quotation marks (in articles containing curly ones) by editors unaware of the issue.
2. I don't find your replies frustrating because you disagree with me and won't change your mind. The frustrating responses are the ones in which you imply that no one has even addressed a particular issue. It's perfectly reasonable to express disagreement with our arguments, but please don't ignore them.
3. If nothing else, at least we agree on what needs to be done on this page. (-: —David Levy 15:11, 18 December 2010 (UTC)

1. I agree that the principle doesn’t always prevent edit wars, but if we accept that as controlling, we accept that people will edit war regardless of policy, and we might as well not have the MoS. I have previously suggested that the wording be modified to call for observing the prevailing style absent consensus to change. Again, not a material issue.

2. There is a difference between addressed and having made a response. We obviously disagree on what constitutes “addressed”.

3. I haven’t ignored any arguments; I obviously consider many of them insubstantial at best. JeffConrad (talk) 23:53, 18 December 2010 (UTC)

1. It is a material issue. You appear to refer to a hypothetical scenario in which the principle fails to prevent edit wars because it's simply disregarded by users who don't want to follow it. That certainly does occur, so let's address this first.

Your point appears to be that such editors will disregard the MoS (and edit war) anyway. That's true, but almost all of our editors (whether edit warriors or not) use straight quotation marks, so under the current setup, the opportunity to edit war over quotation marks seldom arises. Conversely, if we permit the small minority of users preferring curly quotation marks to create and maintain articles containing them, editors among the vast majority using straight quotation marks (some of whom are edit warriors) will come across these articles and seek to fix them (in addition to those who introduce straight quotation marks without even recognizing the disparity).

But edit wars also stem from causes other than intentional disregard of policies and guidelines. Another likely scenario is one in which someone simply isn't aware of our style conventions. Yes, a lack of familiarity with the MoS is equally likely to exist no matter what conventions we adopt, but again, almost all of our editors independently use straight quotation marks (so their lack of familiarity with the MoS will become a relevant issue only if we start expecting them to use curly ones).

Still another scenario is one in which there is disagreement regarding which style was used in the article's first non-stub revision. What constitutes a "stub" (and precisely when a small article ceases to be one) is debatable, and it's possible for an article's style to be inconsistent when the threshold is crossed. But the "non-stub" criterion is important, as the principle otherwise would encourage users to create essentially empty pages as a means of "squatting" on future articles (i.e. locking in their preferred English varieties and other style elements).

Yet another scenario is one in which two or more articles with differing styles are merged. What then? Should the oldest article's style prevail? How about the longest article's style? Perhaps we should go by whichever article has received the most hits. Or maybe we should favor the article ranking highest on a quality scale. No, I suppose that we should retain the style of the article occupying the merged article's title. But wait, we're merging in multiple articles that all contain a different style, so wouldn't it make more sense to stick with that? And what if we're merging all of these articles into a brand new title?

Do you see what I mean? "Respect the original style" seems like a simple concept until one thinks through the complications that can (and do) arise.

Conversely, when we maintain a consistent style, the ambiguities vanish.

2. In this context, "address" means "to deal with or discuss." No one is demanding that you agree with us or believe that a satisfactory resolution has been reached, but it's extremely frustrating when you reiterate the same concerns without acknowledging that we've even responded, let alone attempting to refute the arguments with which you disagree (which, to someone reading your messages, apparently don't exist).

3. I, too, regard some of the arguments (on both sides) as insubstantial, but I don't pretend that they haven't been made (and express bewilderment regarding the positions that they're intended to defend, as though no reasons at all have been provided). —David Levy 02:15, 19 December 2010 (UTC)

John, I understand what you’re saying, but for many online versions of major newspapers, I see precious little in the way of copy editing. Perhaps it’s the result of cutbacks, perhaps it’s the result of time pressures. And in any event, it would seem to cost just as much to enforce a house style of typewriter quotes. JeffConrad (talk) 02:02, 17 December 2010 (UTC)

At a professional publication, that's true. At Wikipedia, it's false (simply because most of our editors use straight quotation marks). —David Levy 03:50, 17 December 2010 (UTC)

Um, my point here was that I am unconvinced that organizations that use typographical quotes incur great costs in doing so. JeffConrad (talk) 04:57, 17 December 2010 (UTC)

I'm expressing agreement with your point that professional publications likely incur similar expenses either way (which actually undermines your "cutbacks" theory) and noting that this doesn't apply to Wikipedia. —David Levy 05:45, 17 December 2010 (UTC)

The arguments for are better than the arguments against. It has been established that Wikipedia requires consistency across articles, not across all of Wikipedia. Darkfrog24 (talk) 15:46, 16 December 2010 (UTC)

To my mind, as a non-involved editor, the arguments for and against have been essentially equal in quality... On the issue of whether to keep or remove the material, I would say that there is "no consensus" either way. No ideas on how to resolve that except by counting heads. Blueboar (talk) 15:58, 16 December 2010 (UTC)

In my opinion, the arguments against the use of curly quotation marks are far stronger than the arguments for their use. Of course, both of us (Darkfrog24 and David Levy) are hopelessly biased.

No one has asserted that consistency across all of Wikipedia is required. But it's very much preferred unless there is a good reason to incorporate inconsistency (such as differences among English varieties). "We believe that our way looks better" (scare quotes) has not been accepted as a good reason to invite inconsistency. —David Levy 16:03, 16 December 2010 (UTC)

There are lots of things in which WP requires inter- but not intra-article consistency, many of which don't correlate strongly with BrE/AmE difference: whether to use unspaced em dashes or spaced en dashes, whether to separate fields in citations with commas or periods, whether to use BC/AD or BCE/CE, whether the possessive of James is James' or James's, whether or not to use the serial comma, whether to use d or d for the differential, whether to denote vectors by boldface or by arrows above, whether the metric signature of spacetime is (+++−) or (+−−−) or (−+++) or (−−−+), yadda yadda yadda. A. di M. (talk) 18:37, 16 December 2010 (UTC)

From one of my messages within the discussion, timestamped 07:57, 13 December 2010 (UTC):

I agree that Wikipedia already contains avoidable inconsistencies. We should be working to eliminate them, not introducing new ones.

The inconsistencies that you cite have led to countless arguments/edit wars and a great deal of wasted time and effort. I can't imagine why anyone (irrespective of personal preference) would want to add quotation marks to the above list.

As I noted in a message timestamped 16:09, 14 December 2010 (UTC), "while I've expressed a clear preference for the use of straight quotation marks, I agree [with the user to whom I responded] that a complete changeover to curly ones would be preferable to the needless inconsistency proposed. Let's not transform every article into a quotation mark battleground."

Incidentally, a material difference between quotation marks and the above examples is that our editors (including those unfamiliar with the MoS) overwhelmingly favor a particular style of quotation marks (rendering their uniform adoption easier to implement and maintain than any alternative would be). —David Levy 19:06, 16 December 2010 (UTC)

I prefer straight quotes but support permitting curly quotes. That's hopeless bias? Darkfrog24 (talk) 19:13, 16 December 2010 (UTC)

I'm referring to bias toward a particular discussion outcome, not toward a particular quotation mark style. Anyone who favors a particular discussion outcome (including you and me) lacks the objectivity needed to fairly gauge consensus or the lack thereof. —David Levy 19:18, 16 December 2010 (UTC)

I disagree. It is perfectly possible for a human creature to fairly judge such things independently of his or her own interests and preferences. However, to establish whether or not consensus exists in this particular case, we would first have to hammer out exactly what consensus is. I don't think anyone's ever done that here. Darkfrog24 (talk) 22:41, 16 December 2010 (UTC)

I don't assert that it's impossible for someone with a vested interest in the discussion's outcome to analyze it in a dispassionate manner. In my mind, that's what I've done. I assume that you believe the same of your analysis, but you and I have arrived at opposite conclusions.

My point is that while it's perfectly reasonable for us to express our views regarding the various arguments, they mustn't be relied upon as impartial. —David Levy 23:40, 16 December 2010 (UTC)

Would the esteemed editors consider taking this overall discussion on quotation marks to a separate public page? Thanks. Student7 (talk) 00:40, 17 December 2010 (UTC)

Let's examine this pragmatically. (quotes)

A great deal has been written and debated, with many conflicting ideas and principles tossed back and forth (to the extent that the discussion has become muddy and difficult to follow). So I think that it might be helpful to take a step back and break down the individual points of contention:

1. The first thing to decide is whether we regard straight quotation marks as "incorrect." It's quite clear that there's no consensus for that.

2. Having established that, we must similarly address curly quotation marks. It's equally obvious that we don't regard them as "incorrect" either.

3. So now we must decide on what style (or styles) to permit within the encyclopedia's general prose.

3a. Does the fact that neither style is incorrect mean that we must accept both? No, the MoS contains many instances in which one legitimate style has been selected over one or more others. Inter-article consistency isn't always feasible, but it's generally preferred. However, there also are instances in which we have accepted multiple styles, so it's reasonable to consider doing so.

3b. What are our options? Our options are as follows:

• Use curly quotation marks exclusively (apart from common-sense exceptions, of course). Most of us agree that this is untenable.
• Allow "shepherding authors" to replace straight quotation marks with curly ones and monitor the affected articles to maintain consistency. This falls dangerously close to WP:OWN territory and relies upon the assumption that curly quotation marks are preferable to straight ones (for which there clearly isn't consensus).
• Allow the original non-stub editor's selection to stand. This principle often is misunderstood. It's intended to serve as a last resort ("when all else fails") in situations in which we have nothing else to go by (e.g. when the article's subject doesn't dictate the use of a particular English variety). It doesn't mean that we must respect every convention preferred by a substantial segment of users. We routinely replace the original editor's otherwise valid style when it doesn't comply with the MoS.
• Use straight quotation marks exclusively (apart from common-sense exceptions, of course). How does this compare to the previous option? Well, it requires us to edit articles to remove a style that some of our editors prefer (curly quotation marks). But said editors compose a small minority. Almost all of our editors use straight quotation marks (and will continue to do so no matter what we decide here). Therefore, the previous option entails considerably more work; it requires us to replace far more errant quotation marks (straight ones introduced to articles written with curly ones, as opposed to the comparatively tiny number of curly ones introduced encyclopedia-wide). The previous option also would lead to edit wars stemming from ignorance of the principle and disputes regarding which style was used in the first non-stub versions of articles (as well as the infamous "well, yeah, the initial change shouldn't have been made, but that was a long time ago and the article has been stable since then" scenario).

So the bottom line, as I see it, is that it simply is far more practical to consistently maintain the style used by a vast majority of editors (given the fact that we don't regard it as "incorrect") than it is to implement any of the alternatives. Those who prefer curly quotation marks shouldn't interpret this as a slight, just as those who prefer traditional punctuation (not even a minority!) shouldn't regard our exclusive use of logical punctuation as such.

To be clear, I'm aware that the above wording favors my arguments. It isn't intended to serve as a neutral summary, which also would be helpful. —David Levy 16:28, 18 December 2010 (UTC)

Nice summary. I think I was on the curly side from personal preference and because more of "my homies" were on that side. But as I think about it, more and more, I think just being a stern captain of the ship and saying let's use straight makes sense for now. Probably, there will come a time when the wiki engine imrpoves in functionality. But at this point we need to be realistic about the trade-off of an "anyone can edit" encyc. that also is displayed on monitors, versus the SUPERIOR work product that can be put out on professionally typeset book. (Or even on a fixed website, where you pay someone to put it together and it has minimal changes.) Certainly, I think curlies are just FINE in MS Word, and while occasionally annoyed by autocorrect fjunctions of that program, I think they are really an improvement...for typing in Word. But I've basiccally moved to a die is cast, Rubicon is crossed, Gordion knot is sword-whacked stance on this one. Viva la straighties! Die, curlies! TCO (talk) 16:50, 18 December 2010 (UTC)

This is essentially my thoughts on the matter. Curly quotes aren't any more "wrong" or "right" than straight quotes by pretty much any book, but in the final judgement, it makes much more sense to apply the de facto style as de jure. In most cases we want the MoS to codify best common practices, and even proponents of curly quotes have to admit the vast majority of editors use straight quotes. Der Wohltemperierte Fuchs^(talk) 19:09, 18 December 2010 (UTC)

David Levy, I agree that your summary is strongly biased toward your POV. Although I partially agree on some points, and I agree that it reflects the POV of the majority of those who have commented here. I don’t think it reflects consensus. We’ve beaten the points on which we disagree to death, so I won’t repeat them. A few comments specific to the summary:

1. Although I (and presumably some of the proponents) might not insist that straight quotes are “incorrect”, few who are familiar with typography consider straight quotes good typography (do a Google Books search for typography "quotation marks" incorrect or something similar; you won’t find many who say that straight quotes are bad typography. I may concede that few people know (or care) much about typography.

2. Certainly most web pages use straight quotes; they also violate many other long-accepted principles of typography. I just don’t buy the argument that “everyone does it” as persuasive; recall the USA Today poll showing that 70% of Americans surveyed thought Saddam Hussein was involved in the 9/11 attacks.

3. If we accept that most editors will do as they damn well please regardless of what we say in the MoS, we don’t need the MoS.

4. You omitted one obvious possibility: actually follow the principle of following the prevailing non-stub style “if it reasonably complies with the MoS”. JeffConrad (talk) 00:36, 19 December 2010 (UTC)

1. Like you, I don't wish to continue rehashing the relevant arguments. But I'll note that I broke down the issues for a reason.

If straight quotation marks constitute "bad" typography, we shouldn't allow them within the encyclopedia's general prose at all (apart from common-sense exceptions, of course). Few have even suggested that.

This means that the encyclopedia will retain straight quotation marks. Therefore, when deciding whether to also include curly ones, the issue of whether straight ones are "bad" is absent from the equation. We need only examine the possible methods of switching from the exclusive use of straight quotation marks to mixed usage, as well as the effects thereof.

2. The argument is not "everyone does it." It's "most reliable, high-quality websites do it." What you perceive as "[violating] ... long-accepted principles of typography" also can be described as "adhering to differing typographical principles in a new medium."

But again, this is irrelevant to a hypothetical scenario in which both types of quotation marks are described by the MoS as acceptable for use within the encyclopedia's general prose.

3. I address this point in my new reply within the Close RfC section (bearing the same timestamp as this message).

4. I don't understand what distinction you're drawing between that and "allow the original non-stub editor's selection to stand." We're discussing the Manual of Style's content, so if we were to take this route, it would be modified accordingly (and the prevailing non-stub quotation mark style inherently would comply with the MoS). —David Levy 02:15, 19 December 2010 (UTC)

1. As I said, I know of no one who knows anything about typography who considers typewriter quotes good typography; if you have credible evidence to the contrary, I’ll be glad to back off on this statement. The issue is more the importance of some accepted principles of good typography.

2. You again refer to “differing typographical principles in a new medium”, which I think has been extracted from the place where they keep the imaginary diseases. I seriously doubt that most web page authors have any principles of typography; sorry if that’s pejorative or arrogant, or both, but I think it’s accurate. I have yet to see anyone demonstrate why the medium makes any difference, much as CMoS indicate. I might object less if it were suggested that the issue is one of the people who contribute to the medium; of course, the same consideration (and perhaps allowance for) unfamiliarity with many other aspects of writing, most of which require far more effort to master.

Agree that if both types of quotes were allowed, this would be of less consequence.

3. I’ll say it yet again: if the MoS is descriptive rather than prescriptive (or proscriptive, as the case may be), any editor is free to do as she pleases. At the very beginning of this discussion, you said precisely the opposite.

4. The distinction is obvious: you cited the example of articles that do not comply with the MoS as subject to changes in style; obviously, stare decisis would not apply in such a case. Unless, of course, the MoS were merely descriptive. JeffConrad (talk) 03:25, 19 December 2010 (UTC)

Just to jump in here, Jeff. Editors are not free to do as they please. I was brought up on AN/I for doing something that the MoS tells us not to do ("proscriptive"), even though I was making articles internally consistent by so doing. It might be nice if the MoS were just a suggestion or a guideline in the Pirates of the Caribbean sense, but in practice it is not. That is why it must be written and edited carefully. Darkfrog24 (talk) 14:13, 19 December 2010 (UTC)

And I'm baffled as to how Jeff read my earlier explanations and believed that I claimed otherwise. —David Levy 17:24, 19 December 2010 (UTC)

By no means to I suggest that the MoS is to be ignored. I simply have no problem distinguishing between recommendations and mandates when words are interpreted according to their common meanings (and if they are not, nothing means anything). As I mentioned in my reply to David Levy below, this is very common practice. JeffConrad (talk) 02:50, 20 December 2010 (UTC)

You haven't explained what practical purpose this particular text serves if interpreted as optional (and you obviously don't intend to). —David Levy 03:33, 20 December 2010 (UTC)

1. How commonly are these principles applied on the websites of reliable, high-quality sources? Do the publishers of newspapers and magazines whose online versions contain straight quotation marks not "[know] anything about typography"?

2. We aren't discussing "most web page[s]." We're discussing reliable, high-quality websites, including those of reputable newspapers and magazines.

Various reasons for the differing style conventions have been discussed here, but we needn't concern ourselves with those (excepting issues affecting our editors, which is a separate matter). The bottom line is that straight quotation marks are widely used on high-quality websites (whether you like it or not).

But as you more or less concede, this is irrelevant to a hypothetical scenario in which both types of quotation marks are described by the MoS as acceptable for use within the encyclopedia's general prose.

3. No, you've misunderstood. The MoS describes the conventions that we've decided to adopt (which are based in large part upon our editors' common practices). When we note that the MoS isn't prescriptive, we mean that we don't follow conventions because they're written there; we write them there because we've decided to follow them.

4. See above. —David Levy 04:54, 19 December 2010 (UTC)

1. Within a publishing organization, I there certainly are people who know about typography, and there probably are far more who do not. I’ve not done a survey, and doubt that you have, either. I cannot speak to the reasons for publishers’ decisions; they may have decided that without the assistance of copy editors (and in many cases, the number of errors strongly suggests that there is little proofing of online content), many writers, often facing deadlines, would have trouble with them (I see many hyphens used for dashes, suggesting this isn’t an unreasonable assumption). If this is the case, it’s not terribly unlike the situation we have here, except for deadline. Or the reasons could be entirely different. A decision to use typewriter quotes would not necessarily suggest that publishers consider typewriter quotes good typography; it is far more likely an acquiescence in the practicality of working under deadline. But quite honestly, we’re both guessing.

I don’t think I ever said that I dislike use of typewriter quotes on web sites of major newspapers.

3. I’ve understood descriptive according to its common meaning. To say that the MoS is partially descriptive and partially descriptive, as you’ve essentially now stated is something different.

4. I stand by what I said; there is a difference between following the established style of a non-stub article when that style complies with the MoS, and changing a style that does not. JeffConrad (talk) 07:34, 19 December 2010 (UTC)

1a. Let's assume that your theory is correct. Do you expect Wikipedia editors to uphold a higher degree of care than professional website editors do?

1b. Please explain why most publishers duplicating their print content online replace curly quotation marks with straight ones.

1c. Again, it doesn't matter why the new conventions arose. The bottom line is that they did. The use of straight quotation marks is an established online practice.

1d. And as you acknowledge, none of this is relevant to a hypothetical scenario in which both types of quotation marks are described by the MoS as acceptable for use within the encyclopedia's general prose.

3. No, I'm not saying anything different. I'll copy and paste what I wrote earlier:

• Most of our rules are descriptive, not prescriptive. The addition of an MoS entry without advance discussion can be perfectly appropriate and uncontroversial, provided that it reflects actual practice. For the same reason, "the rules say so" is not a valid argument. (In other words, this rule has remained because the practice that it documents has remained, not the reverse.) [from a reply to you, 07:57, 13 December 2010 (UTC)]
• Professional style guides, contrary to our MoS, are prescriptive rather than descriptive. They've been slow to address the Internet (including its terminology) and acknowledge its stylistic differences. It made headlines when the AP Stylebook finally switched from "Web site" to "website" (reflecting longstanding common usage).
• As previously noted, our MoS is descriptive, not prescriptive. If it doesn't accurately describe our conventions, it should be revised. And of course, consensus can change (which is why we're having this discussion). Users of varying opinions are providing arguments as to why their preferred outcomes are advantageous. "The rules [don't] say so" is a weak rationale for any of us to advance. [both from a reply to you, 12:33, 15 December 2010 (UTC)]
• [The] MoS is descriptive. As noted below, it describes current practices (as determined both by analyzing those that prevail and by discussing possibilities and deciding what works best).
  Meanwhile, a style manual like the AP Styleguide or the Chicago Manual of Style hands down decisions from above. From a practical standpoint, the distinction is that publications following such a manual can legitimately decide to apply a particular convention "because the guide says so" (no analysis required). Some occasionally opt to deviate, but this is optional and relatively uncommon.
  Conversely, if someone objects to an entry from Wikipedia's MoS (on the basis that it either doesn't accurately describe current practice or a belief that said practice ought to be modified), we have discussions such as this one. And my point is that within such discussions, there's little need to debate an entry's disputed meaning (because we're here to determine what makes the most sense and describe that in the MoS, not to blindly abide by whatever already has been written there). [from a reply to Darkfrog24, 05:09, 16 December 2010 (UTC)]

4. Sorry, I still don't understand what point you're arguing or how it contradicts anything that I've written. At least one (if not both) of us must have misunderstood the other. —David Levy 17:24, 19 December 2010 (UTC)

Let’s not beat it to death any further. JeffConrad (talk) 02:50, 20 December 2010 (UTC)

TCO,

1. I fail to see how there is a material difference between something done with MS word or anything else except that the latter may require slightly greater effort. And of course, some WP editors use MS Word and paste the results into the edit window, in which case it’s a non-issue.

2. Of course our principle is that “anyone can edit”, but I don’t think we mean it quite so literally. For example, I certainly hope this doesn’t include people who know nothing about the subject of an article or who insist on inserting strong POV.

3. (To David Levy as well). Perhaps I’m hopelessly biased because I learned document processing using markup language, but find it hard to believe that people with any significant writing experience lack at least rudimentary computer skills. Many editors here enter other special characters all the time, and it makes no sense to say they somehow cannot do the same with quotes. Again, if we say they will use straight quotes anyway, whether or not they can do otherwise, why tell them to do anything? JeffConrad (talk) 00:36, 19 December 2010 (UTC)

1. Again, you're focusing on the "slightly greater effort," thereby ignoring repeated explanations that we aren't in a position to inform most editors of how to engage in it (let alone expect them to comply).

2. A Wikipedia editor need not possess any prior knowledge of a subject before editing its article. He/she need only cite reliable sources when introducing factual information and refrain from inserting content that we regard as inappropriate (including non-NPOV material). And of course, plenty of editing has no bearing on factual claims.

3a. What is the relevance of "people with any significant writing experience"?

3b. Once again, you're repeating your "How are quotation marks any different from other special characters?" (scare quotes) argument, misleadingly implying that the point has not been addressed.

3c. We aren't telling the users in question to do anything. Most of them will never see the MoS. We're describing what they do and advising the tiny minority of editors concerned with maintaining style conventions to act in kind. —David Levy 02:15, 19 December 2010 (UTC)

1. Again, you say “we aren't in a position to inform most editors of how to engage in it [entering special character]”. Nonsense; we mandate use of special characters in many cases, and to expect any attempt at compliance without some guidance is elitist and just plain nuts. I don’t see why it’s any different from providing help on tables, images, or TeX.

2. A little knowledge can be dangerous, but person who has researched reliable sources arguably has some knowledge of the subject (copying and pasting, which is disallowed, doesn’t count). Realistically, though, I find it astonishing that someone with no knowledge of a subject would think of writing on it, or that such a contribution would be of value. Sorry if this seems arrogant; it strikes me as simple common sense.

3a. Perhaps I didn’t choose the best words; perhaps I might better have said “experience using a computer”, but are there really hordes of people who have used a computer without ever having written anything? It’s possible, of course, that someone who has never written anything could indeed be quite knowledgeable. I think this is an unusual situation, however. My point was that making any worthwhile contribution requires at least rudimentary knowledge of the mechanics, such as providing usable references to reliable sources (let alone formatting them). For many, myself included, this is a continuing learning experience. The effort to learn how to deal with special characters is trivial by comparison.

3b. We disagree on the appropriateness of “scare quotes”, and I take issue with “misleading”. If the main thrust is that quote characters (especially apostrophes) occur more frequently than other characters, then you and I probably don’t disagree (though I wouldn’t mind some quantitative examples). To put this in perspective: it takes a lot more effort to add references than to simply make a few statements; does this mean we should not require the former because it is too difficult for some editors?

3c. Again, if the MoS is descriptive, no one is bound by it. JeffConrad (talk) 03:25, 19 December 2010 (UTC)

1. We don't expect most editors to generate any special characters, including the oft-mentioned dashes. People often type non-MoS-compliant approximations, which we replace. Some have advocated that we take the same approach with curly quotation marks, while others (myself included) have cited material distinctions between dashes and quotation marks. In response to you, I wrote:

And yes, relatively few editors will enter dashes either. But as previously noted, because their approximations vary wildly (e.g. one hyphen, two hyphens, surrounded by spaces, not surrounded by spaces), this is an element that we need to clean up anyway (to maintain intra-article consistency, at the very least). Conversely, almost every Wikipedia editor uses a particular style of quotation marks, so sticking to it means that few instances will require modification.

It also has been noted that dashes appear far less frequently than quotation marks do and are easier to recognize as different. Other distinctions have been noted, but some are irrelevant to the issue discussed above.

3a. Again, no one is asserting that learning how to generate special characters is difficult. It simply isn't something that most of our editors will do (or even know that it's expected of them, which it really isn't).

3b. I don't assert that your comments are intentionally misleading, but they are misleading.

That quotation marks occur more frequently than other characters is one of several distinctions. I personally regard the one that I copied and pasted above (which you presumably don't recall reading) as the most significant.

Unsourced claims reduce the encyclopedia's quality. There is neither consensus that straight quotation marks reduce the encyclopedia's quality nor a formal proposal to disallow their use.

3c. See above. —David Levy 04:54, 19 December 2010 (UTC)

1. Be assured I read your earlier response; I didn’t ignore it, but was simply unpersuaded. I would hope reasonable people can disagree without taking offense.

3a. Actually, there have been more than a few comments about the difficulty of entering special characters. I’m not sure it’s even been established that most editors prefer straight quotes; as several others have suggested, many may not know otherwise. I would again guess that neither of us has done a survey.

I certainly found this to be the case when I first got involved with electronic document preparation; prior to the computer, typographical quotes were not an option, so no one even gave thought to using them. After a couple of brief discussions (OK, I strongly encourage my staff to use the proper marks. But let’s be realistic: the basic document editor had to master quotes, dashes, and sometimes trademark symbols and minus and multiplication signs; others had to deal with mathematics, tables, graphs, and so on, to which a few special characters paled by comparison). The key thing is that people simply did it. But this was some time ago, so conditions may now be different in a way that I’ve failed to notice.

As you suggest later (and I largely concur), this issue may be moot. I still see no problem with a Help article or subsection that describes how to use these characters, whether or not we expect most editors to use them. With a bit of demystification, more people just might use them, as I just illustrated.

3b. I don’t see what GF has to do with it. I’ve acknowledged many times that apostrophes are probably more frequent than most other special characters, but I’m not convinced that makes them materially different. It should be obvious that we disagree on this issue, and probably will continue to do so.

I’m not sure what you mean by “unsourced claims”, but I agree there is no consensus for the changes proposed, or that straight quotes reduce quality (there is also no consensus that they do not, though I don’t think the proponents have made a big deal of this). There is no formal proposal to disallow the use of typewriter quotes, although you have repeatedly claimed that the MoS currently does so, and some have suggested (though not formally proposed) that we proscribe them. By repeatedly quoting the register, I have simply maintained that there is no consensus for such a proscription. If we reasonably agree on this, there is no need for further argument. JeffConrad (talk) 07:34, 19 December 2010 (UTC)

1. Again, I'm not offended by your disagreement. I only object to your continual reiteration of points in a manner implying that no counterpoints (whether persuasive or unpersuasive) have been written.

I am interested in why you regard that particular argument as unpersuasive. Do you dispute the assertion that most Wikipedia editors generate a particular style of quotation marks (straight ones)? Do you dispute the assertion that the same editors introduce varying approximations of dashes (e.g. one hyphen, two hyphens, surrounded by spaces, not surrounded by spaces), thereby requiring significant cleanup (to maintain intra-article consistency, at the very least), regardless of what style[s] we adopt?

3a. Indeed, the relative difficulty of entering straight quotation marks is highly relevant, but not in the context that you've continually addressed.

If a standard computer keyboard included curly quotation mark keys (and especially if it also didn't include a straight quotation mark key), the marks' use would be significantly easier and more intuitive. Because that isn't the case, almost all of our editors simply type in the manner to which they're accustomed (i.e. with straight quotation marks).

You've noted (and I don't dispute) that learning how to generate curly quotation marks is not extremely difficult. But most of our editors will never even become aware of the option, let alone see a need to take advantage of it. They'll simply continue typing the familiar, widely used quotation marks that appear on their keyboards.

3b. I linked to WP:AGF to emphasize my assumption that you aren't intentionally ignoring my points. You've made this slightly more difficult by again implying that my argument is based primarily upon frequency of use (in response to a message in which I explicitly stated that I regard a different distinction, copied and pasted above, as more significant).

By "unsourced claims," I'm referring to claims for which reliable sources have not been cited.

You appear to be under the mistaken impression that Wikipedia:Manual of Style/Register is some sort of official document. In actuality, it's an informal page to which an editor added a highly slanted summary (labeling opponents' claims "dubious at best") in May of this year. That text carries no more weight than any comments from this discussion do. —David Levy 17:24, 19 December 2010 (UTC)

1. The intent was not to suggest that no counterpoints have been written.

3a. I don’t know what you mean by the context I’ve continually addressed, but I don’t think continued discussion is worth the time.

We’ve beaten this to death several times, and probably won’t accomplish much by continuing. JeffConrad (talk) 02:50, 20 December 2010 (UTC)

As previously noted, I don't allege that such was your intent.

I'm disappointed by your decision to decline answering my questions. —David Levy 03:33, 20 December 2010 (UTC)

David Fuchs, I would say that anything made de jure that isn’t consensus isn’t de jure. And I would say that after this discussion, the statement “Currently there is no consensus regarding which quotation glyphs to use” that begins the register on quotation marks for the project still holds.

Again, I fear our greatest problem in wrapping this up is, whether or not we admit it, we agree to disagree on the consensus of the discussion. Or even on what “consensus” is. JeffConrad (talk) 00:36, 19 December 2010 (UTC)

The MoS has long recommended against the use of curly quotes. The vast majority of editors use straight quotes. The vast majority of articles use straight quotes. There's no clear consensus that the MoS should be amended above. That's pretty much de facto and de jure. Der Wohltemperierte Fuchs^(talk) 03:06, 19 December 2010 (UTC)

I don’t see how it’s de jure, but completely agree there is no consensus for any of the changes exactly as proposed, so I’m not sure we really disagree. The MoS currently discourages but does not proscribe their use; the main thrust of the proposals would be to put the onus of maintaining intra-article consistency on those editors would would use typographical quotes anyway. Though I originally suggested that changing typewriter quotes to typographical quotes should not be objectionable, it’s clear that such an approach has no support, so I withdrew the suggestion, as I’ve mentioned several times. Were I to submit a new proposal, I would amend it to require that editors defer to the use of straight quotes in a non-stub article. I’m not sure we need to do that, however, because this would seem covered by the general principle of stare decisis for most matters of style (though I’m not sure this was clear prior to this discussion).

So if that’s all on which we disagree, as far as I am concerned, we can close this discussion with the conclusion that there is no consensus for the changes proposed, provided that we don’t attempt to slip any other conclusions into it, except perhaps to confirm that the prevailing quote style for a non-stub article is to be treated the same as any other matter of pure style, and that it is not to be changed without consensus. In particular, the comment in the register would remain as is unless we somehow determine that we have examined any new issues. I would, of course, like to see that this is acceptable to most of the other proponents before signing off on it. JeffConrad (talk) 07:34, 19 December 2010 (UTC)

1. You've yet to explain how your interpretation of the MoS (that it "discourages but does not proscribe" the use of curly quotation marks) makes any sense from a practical standpoint. Again, what would be the logic in conveying "x is recommended and y is not recommended, but feel free to do whatever you please"? Thus far, your response has boiled down to "I don't know, but that's what it says, so that must be what it means."

And of course, the MoS is intended to reflect accepted practices, not the reverse.

2. We've discussed the problems with "[putting] the onus of maintaining intra-article consistency on those editors would would use typographical quotes anyway."

3. If editors would be expected to "defer to the use of straight quotes in a non-stub article," how would Wikipedia benefit (assuming, for the sake of argument, that curly quotation marks are stylistically superior)?

4. We do not have a "general principle of stare decisis for most matters of style." We fall back on such a principle as a last resort (i.e. when there is no other valid reason to favor a particular style). "For most matters of style," we seek encyclopedia-wide consistency. Exceptions (stemming most notably, but not exclusively, from differences among English varieties) arise when no particular style sufficiently predominates among our editors. Straight quotation marks overwhelmingly predominate. They're used by a vast, vast majority of our editors.

5.As noted above, you appear to be under the mistaken impression that Wikipedia:Manual of Style/Register is some sort of official document. In actuality, it's an informal page to which an editor added a highly slanted summary (labeling opponents' claims "dubious at best") in May of this year. It's quite disheartening that you wish to affirm such language. —David Levy 17:24, 19 December 2010 (UTC)

1. I can’t explain it because I can’t read minds, but I suggested a plausible explanation far above. It doesn’t really matter, though because I interpret words according to their common meanings without asking rhetorical questions. Do a Google search for "recommends but does not require" or "encourages but does not require" and you’ll find this is not uncommon practice.

4. The MoS reads as it reads. This is simple common sense as well as a basic principle of good editing.

5. I see it as no more biased than your initial summary, and think the statement of lack of consensus is still correct. As to disheartening, I feel the same way about most of the arguments presented here. But you and I will never agree on this, and at this point, I’m not sure it matters. Let’s move on, as Tony and Student7 have begged us to do. JeffConrad (talk) 02:50, 20 December 2010 (UTC)

1/4. You haven't explained what logical purpose this particular text serves if interpreted as optional. I honestly can't think of one.

And again, the MoS isn't a document to be blindly followed. It's supposed to be based upon accepted practices, not the reverse. "The MoS says so" (not an actual quotation), without regard for whether it makes practical sense or reflects reality, never is an appropriate rationale.

5. My comments were neither intended to serve as an impartial summary nor represented as such (as I explicitly noted when authoring them). Conversely, you've repeatedly cited Wikipedia:Manual of Style/Register as though it's a formal, binding document.

Unfortunately, I must agree that the discussion isn't accomplishing much at this juncture. I never expected to change your mind, but I hoped to improve your understanding of my perspective (and given the time and effort that we mutually invested, I imagine that you possessed a similar aspiration). —David Levy 03:33, 20 December 2010 (UTC)

• WP:TLDR. Guys, there's no way most contributors are going to read this much text. Could I suggest in future that points be made very tightly? On the substantive matter, while I prefer to see curly quotes, I seem to remember there are good technical reasons for using straight quotes, but don't ask me what they were. I've abided by that ever since. Articles would need to be consistent if the option of using curlies were made available. That's all from me. Tony (talk) 14:41, 19 December 2010 (UTC)
  • I agree with Tony on this one. I gave up about a week ago trying to follow this discussion, and almost removed this talk page from my watchlist because of it, even as I was trying to follow other discussions here. As for me, I too prefer typographer's quotes over typewriter quotes, but consistency will need to be key. Imzadi 1979 → 15:19, 19 December 2010 (UTC)

Tony, I agree. I opened this discussion primarily to challenge one statement in the MoS (which has been changed as a result). I regret having taken the discussion any further; previous discussions should have given fair warning of what was to result. Editors agree to disagree on this issue. We should kill it and move on. JeffConrad (talk) 02:50, 20 December 2010 (UTC)

David Levy, I don’t mean to dodge your questions; however:

• We’re making the MoS Talk page almost impossible for others to monitor.
• Accordingly, they’re begging us to stop.
• This discussion has become so complex that it is very difficult to reply. Because of edit conflicts, it took three tries to make the last replies scattered throughout the document. Making individual replies would just make things worse for other users.

There are many issues on which I don’t think we’re going to agree regardless of how much we discuss it, but if we must do so, can we do it via e-mail? I have a feeling the discussion would not be brief, and I don’t want it cluttering up my Talk page, either. If there’s a better approach, I’m open to that as well. JeffConrad (talk) 04:23, 20 December 2010 (UTC)

I sincerely appreciate the above response. If you're most comfortable continuing via e-mail, that's fine by me. —David Levy 04:34, 20 December 2010 (UTC)

I really don’t care, and don’t object to making the discussion available to others (as if anyone would care ...). If there is a way to follow Student7’s request that we continue on some other page, that’s also OK, but I don’t where such a page would be (my Talk isn’t it). JeffConrad (talk) 04:44, 20 December 2010 (UTC)

Well, a common solution is to create a dedicated sub-page (which anyone is free to watch or ignore). —David Levy 06:09, 20 December 2010 (UTC)

That’s fine with me; any location but my user space is OK. In any event, a man with a trombone case just came to my door and suggested that I consider commenting no further here . . . JeffConrad (talk) 06:32, 20 December 2010 (UTC)

Yes, my suggestion did intend that others interested in the topic be able to continue following and participating in the discussion. So the new discussion location would ideally be posted here below with some close out notice. When you have reached a solution, you might post it here for a few days and listen to the anguished comments of those who wish they had followed it more closely!  :) Student7 (talk) 14:02, 22 December 2010 (UTC)

No "hiding" information?

I have just discovered that the policy (MOS:COLLAPSE) prevents concealing a list of people who have run an organization, like the mayors of a town, or the presidents of a university. Yes, I know that Wikipedia is NOTALIST. Some of these people in a list were notable; but most weren't. This was once the only list that was allowed, no other names allowed in place or school articles. Some of these lists were long, and concealing them did allow readers to skip them at will. I guess they will have to be erased which is a shame IMO. A bit of history lost to readers. And often, there is no other list anywhere that is current. Student7 (talk) 21:59, 7 December 2010 (UTC)

Don't delete good content. The sometimes-inconvenience of scrolling past a list is not a reason to delete a list. --Cybercobra (talk) 02:52, 8 December 2010 (UTC)

• I'm not sure this discussion belongs here, but also, you need to define what "good content" is. My inclusionist friends may disagree with me here, but we ought to follow WP:N and WP:NOT#DIRECTORY too. I rarely hesitate to remove list of non-notable people (be they supporting cast, faculty, committee members of an organisation/production) where they are of limited relevance to the central topic (and sometimes even when they are directly relevant, but resolutely fail WP:N) or otherwise constitute a laundry list. To cite an example: we might often have details of the president or chairman of a notable organisation, but where this article goes on to list every "Tom, Dick or harry" on the committee, it crosses the line, IMHO. Collapsing a list may make it less obtrusive, but there really ought to be a good reason to keep same information, collapsed, expanded or otherwise. --Ohconfucius ^¡digame! 03:26, 8 December 2010 (UTC)

I guess you have almost convinced me about lists of mayors, presidents, etc. I still think that some elected positions in a town should be listed. I have to admit, these lists can get pretty long. Someone correctly observed in a small village I was maintaining that the number of "officials" was a rather high proportion of the number of adults living in the village. I had to agree that the list shouldn't be that long. Almost none are notable. Sorry, if this discussion shouldn't be here. It arose because of no COLLAPSEd lists. Student7 (talk) 16:10, 10 December 2010 (UTC)

I'm having a hard time understanding that guidance on collapsing. I don't think I agree with it either, and I don't know the background, but to start with, it's hard to understand. On content: surely collapse is a highly useful device. (even for text). And why would putting a list in a table make it any different than if bulleted (although I guess putting stuff in a table allows getting the needed benefit, while still matching policy.TCO (talk) 16:36, 15 December 2010 (UTC)

Remember, this is one of the items, I had to reverse myself on!  :)

The question would arise: Why am I "hiding" information? In my case above, I was hiding it because most people would not be interested in a long list of non-notable people (people without articles). The other editors are saying (in affect) "Aha! Why am I putting non-notable stuff in an article to begin with?" (I had no good answer).

If these were all notable mayors, people would be tremendously interested because how many times do members of a tiny city get to be notable? It would be of great interest. But my list didn't have that quality.

The flip side is, maybe I am trying to conceal someone else's material that I don't like or agree with! That would normally get quick attention though.  :) In other words, there seems to be no good reason to "hide" stuff. Having said all that, it is normal to hide stuff in navigational boxes at the end. They accumulate and are simply too long to display them all to people who aren't normally interested. But that is the only exception I have found. Student7 (talk) 21:39, 17 December 2010 (UTC)

I think there is a lot of benefit of having some click to expand text versus a subpage, at times. See subpages and summaries constantly clashing. Yes, a click to expand could be non-consistent as well, just as sections of an article might. but I think consistency would average higher and also reader appreciates staying in same window for comparisons, etc. this is not to say never have a subpage. also, it's just a feature. like wikilinking or the like. media is changing...

Of course, your list, well...that's not what I mean really. I mean more when articles are getting long, but still splitting into pages will not be best solution.TCO (talk) 23:33, 17 December 2010 (UTC)ho

The browser still has to load the full text of the page including the hidden stuff, even if the user doesn't click "show", so if the problem is that a page is long, {hide} doesn't solve it. A. di M. (talk) 01:26, 18 December 2010 (UTC)

I don't think browser size is what drives the limitation, but, rather, the ergomics of information processing.  ;) TCO (talk) 02:01, 18 December 2010 (UTC)

I too would support at least an RFC on collapsible content. The biggest argument against it is that since pages shouldn't exceed a certain size, the mount of scrolling should be limited. The exception to this argument is long lists (of short items). Therefore I would support an exception allowing collapsibility of long lists of short items. Rich Farmbrough, 02:24, 22 December 2010 (UTC).

Almost every time I see country subpages with summaries, there are marked differences between the two. I suspect having the content on the same page, would help keep things more consistent (yes, you could still ahve a breakdown, just like with section to section, but I think it would be less). All kinds of other parts of the web, like blogs, use click for more info. Readers don't like having to open a seconde window. Click for more info would be sweet. TCO (talk) 02:34, 22 December 2010 (UTC)

Summaries are horrible. Don't know how to keep them in sync. New editors add to summary and neither edit nor even check the forked article. (Sometimes the original editor deliberately omitted the info the new editor is adding). Eventually the summary becomes as big or bigger than the fork! "Summaries" are a another whole topic. Student7 (talk) 13:52, 24 December 2010 (UTC)

unit conversion templates, good or bad

(And please don't tell me not to ask here, to ask on the FA forum, since until I have an FA done, I lack street cred.  ;))

What do y'all think of the use of unit conversion templates as opposed to manually typing the conversion? (good or bad, or even the movement to require them). Sounds dumb, but I find that the conversion templates have occasional bad features. Like when I have a rounded number of meters (200 m) giving me 679 ft or the like, just looks bad. There's an implied vagueness of measurement in the first measurement which I WANT TO KEEP (we are talking about how far away from land a painted turtle nests, NOT the well-surveyed height of Everest.) If I do it manually, I can kind of decide how to tweak the rounding on the second unit. Also, I may try yards vice feet to get around this. But I find the conversions in the first place kinda load the article down. And I am even a bit of a conservative, but in a science article, could see using metric (not in a US state article, though). But anyway, if we are going to have them, the FA requirement to use them, ends up making the text look worse than what I could do on my own, manually. I'm just saying...TCO (talk) 16:35, 18 December 2010 (UTC)

P.s. and just to keep my conservative street cred, kn is so, so unused for knot. I'll bet you can survey style guides, published usage, etc. and find WAY more kt than kn. I've got my "time on the pond" on that unit. TCO (talk) 16:38, 18 December 2010 (UTC)

The {{convert}} template rounds unwisely, and chooses units arbitrarily. It is always better to do unit conversions by hand, if your arithmetic can be trusted. (Not everybody's can, and it's more trouble; that's why there is a template.) Septentrionalis PMAnderson 17:42, 18 December 2010 (UTC)

I see no special merit in putting in the conversion template rather than calculating and typing for yourself, esp if the data is unlikely to change in future edits. However, if you change your edit to {{convert|200|m|ft|sigfig=2}}, or to {{convert|200|m|ft|-1}}, then you'll get 200 metres (660 ft) Kevin McE (talk) 17:52, 18 December 2010 (UTC)

Manual conversion is unquestionably more vandal-prone. If necessary, require the usage to specify the number of significant digits rather than inferring it, but by all means avoid the manual conversion when possible. LeadSongDog come howl! 18:03, 18 December 2010 (UTC)

(edit conflict) There's no move that I'm aware of to mandate the use of the {{convert}} template, as opposed to conversions done by whatever means, and I very much doubt that there ever will be. I pretty much always use the template myself. The issues that PMAnderson alludes to are the template's default behaviour, all of which can be parameterised. Your example of "implied vagueness" can be dealt with like this, for instance {{convert|200|m|ft|sigfig=1}} (giving 200 metres (700 ft)), depending on how much vagueness you're looking for. Malleus Fatuorum 18:13, 18 December 2010 (UTC)

Sig figs field is perfect. We will use and conform (to the policy that is NOT creeping in). Seriously, thanks!TCO (talk) 18:22, 18 December 2010 (UTC)

BTW, in that situation, I would go with 660 as sorta "looking right". I think the reader understands it is a converted quantity, thus the added digit. But 659 looks bad, and 700 is too far "off". Same thing comes up with converting 2 kms to miles, 1.2 just reads better than "1". Thanks again on the sig figs field! TCO (talk) 18:35, 18 December 2010 (UTC)

The convert template is more easily maintained as pointed out by Leadsongdog, because it is relatively vandal resistant. Try to figure out by eyeball whether a change by an IP from horsepower to joules is a correction or vandalism. It's a pain the rest of us don't need. When I find something changed that an editor entered manually, I am forced to use convert to reenter and ensure it was either done correctly the first, or the second time. Why fool around? It's better to do it right the first time. Help out your fellow editors and those of us who "watch/patrol" pages by using the template.

And the initial, well-meaning, trustworthy, reliable editor is still prone to error. Convert will also figure out what to convert to, which is sometimes useful (sometimes not). For example, it converts inches of rain to millimeters. Who would have thunk that the rest of the world would want rain in millimeters? It makes no sense, except in a desert someplace, but I don't have to worry about it. Convert makes the decision for me. At other times it makes more sense to convert a large number of small units into square miles/km, which the template does easily, with no mistakes. Student7 (talk) 21:15, 20 December 2010 (UTC)

In short, {{convert}} is not always useful; it chooses inappropriate targets to convert to; it chooses the wrong number of digits to state; but we have to use it anyway "to help our fellow editor" (especially when they can't check a multiplication - that's what calculators were invented to do). Don't help your fellow editor; help the reader - and if checking calculations is not what you're best at, do feel free to check something else. Septentrionalis PMAnderson 23:29, 20 December 2010 (UTC)

Both the target unit(s) and the rounding can be specified. (I think the defaults in many cases make no sense at all, but that'd belong to Template talk:Convert, not here.) A. di M. (talk) 02:00, 21 December 2010 (UTC)

I think Convert helps the reader in the same way a vandalbot helps the reader. Both prevent the reader from seeing wrong information immediately, not later when someone happens to fix it. Automation also frees up editors to help the reader in other ways. Art LaPella (talk) 02:10, 21 December 2010 (UTC)

The presence of an adjustable sig fig may be all I needed. Also had an interesting glitch where we used it for converting celsius to farenhiet, BUT we were refering to a difference of degrees and of course it converted absoltue temperature. No biggie though. I'm fine with fitting in.TCO (talk) 23:49, 20 December 2010 (UTC)

The temperature difference vs. absolute temperature problem probably affects hundreds of Wikipedia articles; I wonder whether someone could design a clever bot to find them all. But further to your original question, an unmentioned advantage of {{convert}} is that it suppresses inappropriate wrapping without the need for  , which I suspect would be more confusing to new editors than the template. A disadvantage is that it sometimes breaks sorting in tables, as its output gets sorted "alphabetically", by which I mean 2000 would get put in between 10 and 30. Adrian J. Hunter^{(talk•contribs)} 07:02, 21 December 2010 (UTC)

To convert temperature differences, use e.g. increased by 14 °F (8 °C) (though this is one of the cases when if no precision is specified the default is completely absurd). The sorting table is not a disadvantage of the template as the same would also happen when giving the conversions by hand, and in both cases you can write {{sort|0074|{{convert|74|kg|lb|0}}}}/{{sort|0074|74 kilograms (163 lb)}} (assuming the largest value in the table is less than 10,000 kg). A. di M. (talk) 12:54, 21 December 2010 (UTC)

Ah! I didn't know either of those things. Thanks for the tips! Adrian J. Hunter^{(talk•contribs)} 14:05, 21 December 2010 (UTC)

I agree with PMA and others: conversion templates are usually not a good idea. Editors need to exert control over the details. When I copy-edit, sometimes I have to remove them to make the conversions consistent and sensible. On this matter, I do wish WP editors had a readily accessible page for calculating conversions. It seems that the very existence of these templates has made us lazy on this count. Tony (talk) 15:45, 21 December 2010 (UTC)

If you don't like the template, there are websites like http://www.unitconverters.net/ Art LaPella (talk) 22:13, 22 December 2010 (UTC)

Or Google for that matter. A. di M. (talk) 10:00, 23 December 2010 (UTC)

You can control details with {{convert}} if you know how to do that. (You can't control everything, but the conversions for which I don't use it because I can't get it to do exactly what I want are about one in twenty.) A. di M. (talk) 16:02, 21 December 2010 (UTC)

But the effect of our present language has not been to get editors to learn the tools of {{convert}} which make it adjustable; it has been to inspire bots and bot-like edits which go around installing {{convert}} at its default settings whether those are appropriate or not. Septentrionalis PMAnderson 22:48, 21 December 2010 (UTC)

What "present language" are you talking about? Right now the MOS doesn't even mention Template:Convert, so if editors go blindly adding it without even checking whether the defaults make sense, that's only their own fault. A. di M. (talk) 04:31, 22 December 2010 (UTC)

Actually, MOS:CONVERSIONS does mention it: "Category:Conversion templates can be used to convert and format many common units, including {{convert}}, which includes non-breaking spaces." Then there's the usual contradiction at WP:VAMOS#Measurements: "convert by template, as the MOS requires". Art LaPella (talk) 05:42, 22 December 2010 (UTC)

It seems that (as so often happens) we need to make the MoS consistent. It also seems to me that {{convert}} can do pretty much everything we'd like, but that it's frequently misused. That makes me hesitant to require it. On the other hand, it's more maintainable and vandal-proof than manual conversions, so it feels like a better long-term solution. I would support language that recommended {{convert}}, but that strongly advised editors to read the documentation and carefully ensure a good result; if editors are unable or unwilling to use {{convert}} correctly, they should be required to use a manual conversion. Ozob (talk) 13:40, 22 December 2010 (UTC)

Me too. A. di M. (talk) 13:49, 22 December 2010 (UTC)

The Manual of Style is a guide applicable to all Wikipedia articles(?)

The first box on the page says, "This guideline is a part of the English Wikipedia's Manual of Style."

In contrast, the first sentence under the section [General Principles] states,
"The Manual of Style is a guide applicable to all Wikipedia articles."
If this is intended, it is hard to tell where the General Principles end and the "English Wikipedia Manual of Style" section starts.  The next use of the word "English" appears in the subsection "Follow the Sources."

Proposal is to insert the word "English":

The Manual of Style is a guide applicable to all English Wikipedia articles.
RB  66.217.118.31 (talk) 18:16, 18 December 2010 (UTC)

I don't see the need. Every policy and guideline on this site applies to the English Wikipedia. Other language versions have their own policies and guidelines. ---— Gadget850 (Ed) ^talk 21:50, 18 December 2010 (UTC)

Concur. We have a gazillion other places where we just refer to Wikipedia. Gonna be a hell of a lot of Englishes that need to get dropped in otherwise. And when no one is really confused. Only worth clarifying when we are talking inter-wiki or the like.TCO (talk) 22:02, 18 December 2010 (UTC)

Yeah, verily, or there be too many repetitionſ and variationſ of Engliſh. -- SEWilco (talk) 23:13, 18 December 2010 (UTC)

(ec)First of all, for my information, where does the policy say "Every policy and guideline on this site applies to the English Wikipedia"?  Secondly, let me try to respond to the doubt that there is a purpose for this proposal.  My experience was to skim down the entire page and there was no place that actually says that this MOS is for use with English.  I can't tell if the "General Principles" is from an old version written back when English was the only Wikipedia, or if there is some application for this MOS in other languages.  And nowhere does the page say "Use English".  (WP:TALK explicitly says, "Use English".)  Thanks, RB  66.217.117.119 (talk) 23:37, 18 December 2010 (UTC)

I agree, having stated it once at the top, it seems unnecessary to repeat it. Tony (talk) 14:42, 19 December 2010 (UTC)

• I think there is an abundance of context in this case; the other Wikipedias have their own manuals of style. Unfortunately, the language herein kinda creates its own ambiguity. We should let context work to its fullest – I see little need to use the word 'Wikipedia' in the sentence "The Manual of Style is a guide applicable to all Wikipedia articles." --Ohconfucius ^¡digame! 06:20, 21 December 2010 (UTC)

Amending WP:DASH to include a note about figure dashes

Please see figure dash I have changed the section of the MoS redirected from WP:DASH to include a brief not about the use of figure dashes on the Englih Wikipedia. As far as I'm aware, it's only come up twice in naming an article: 867-5309/Jenny and Royal Albert Hall London May 2-3-5-6, 2005. I moved both of those from the version that includes hyphens to one using figure dashes and the consensus was to move back the title of the article, but to use the figure dash in the body text as appropriate. For what it's worth, I am still in favor of using the figure dash in the title as well. —Justin (koavf)❤T☮C☺M☯ 20:54, 24 December 2010 (UTC)

The problem with that is North American phone numbers are defined as using the hyphen, not the figure dash by the standards body that determines these things. The citation is in the archives somewhere from a previous discussion. oknazevad (talk) 21:14, 24 December 2010 (UTC)

Found it here.oknazevad (talk) 21:22, 24 December 2010 (UTC)

The cited standard is: International Telecommunications Union E.123 : Notation for national and international telephone numbers, e-mail addresses and Web addresses. The standard says the space character is the separator. If you use the space character in both national and international numbers then you're compliant with E.123. If you use any other character in an international number then you're non-compliant. You may use another character in national numbers and claim compliance with E.123 if you can refer to some other agreement. I hope that helps. Lightmouse (talk) 16:30, 26 December 2010 (UTC)

That wasn't the one I'm refering to, I was refering to the North American Numbering Plan website, which uses hyphens.oknazevad (talk) 18:30, 26 December 2010 (UTC)

OK, I wasn't sure what you meant. Thanks for clearing that up. 86.171.93.148 (talk) 19:48, 26 December 2010 (UTC)

Sure I appreciate Lightmouse and Oknazevad looking into the technical specs on this matter, but I'm not sure that I see how relevant they are. Publications make arbitrary style decisions--including typographical ones. While technical manuals can (and should) be consulted, they aren't a trump card for how Wikipedia as a community will (or should) decide how to style our articles. Are you suggesting that we follow these specifications or are you simply pointing out what they are for the consideration of anyone else who chooses to engage this discussion? —Justin (koavf)❤T☮C☺M☯ 17:13, 26 December 2010 (UTC)

The issue is that is not a matter of style, it's a matter of technical correctness. These are the correct standards, and we should follow them. The correct usage doesn't take a powder because we don't like the way it looks. Factually incorrecct usage must be fixed.oknazevad (talk) 18:30, 26 December 2010 (UTC)