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

Possessives still need clarifying

(outdent) for the time being i propose removing the crossed-out bits below - there's no sense in recommending Option 3 over the other options when it's considerably less clear than the others; and the other crossed-out bits are simply confusing:

Add either 's or just an apostrophe, according to how the possessive is most often pronounced:

• More easily decided: Jan Hus's life, Sam Hodges' son, Morris's works, the bus's old route.
• Less easily decided: James's house (or James' house), Brahms's music (or Brahms' music), Vilnius's location (or Vilnius' location), Dickens's novels (or Dickens' novels).

Practice must be consistent within an article; the third practice, endorsed in some form by most style guides, is recommended. Where there is disagreement over a pronunciation, the choice should be discussed and then that possessive adopted consistently in an article. Possessives of certain classical and biblical names have traditional pronunciations which may take precedence: Moses' leadership, Jesus' answer, Xerxes' expeditions, but Zeus's anger. In rare cases where such discussion yields no solution, rewording may be an option (The location of Vilnius is strategically important).

Option 3 still needs a lot of clarification and finetuning, but those crossed-out bits seem (for assorted reasons) especially unhelpful and i feel they shouldn't be hanging around in public while the section is still under construction. can i go ahead and remove them, while we continue to try to reach consensus on the option and wording? Sssoul (talk) 15:05, 3 September 2009 (UTC)

the silence is rather ambiguous! so how about something like this:

[options 1 & 2 as they are currently described]

3) Add either 's or just an apostrophe, according to how the possessive is pronounced:

• Add only an apostrophe if the possessive is pronounced the same way as the non-possessive name: Sam Hodges' son, Moses' leadership;
• Add 's if the possessive has an additional /ɪz/ at the end: Jan Hus's life, Morris's works.
• Some possessives have two possible pronunciations: James's house or James' house, Brahms's music or Brahms' music, Vilnius's location or Vilnius' location, Dickens's novels or Dickens' novels.

Whichever of the above three options is chosen, it must be applied consistently within an article. When using the third option, if there is disagreement over the pronunciation of a possessive, the choice should be discussed and then that possessive adopted consistently in an article. (In some cases – particularly possessives of inanimate objects – rewording may be an option: the location of Vilnius, the moons of Mars.)

the third option still needs to be clarified considerably, but that would be better than what's out there now.

and can we please discuss replacing this very open-ended Option 3 with the OUP style mentioned in a now-archived discussion? ie: just apostrophe when the last syllable of the name is pronounced like the verb is; apostrophe-s in all other cases - it has the major advantage of being easy to explain clearly and to apply. Sssoul (talk) 15:54, 4 September 2009 (UTC)

That is so obviously better than what we had now that I've been bold and implemented it (except that I didn't remove the mention of classical names). --___A. di M. 16:13, 4 September 2009 (UTC)

thanks, A di M - i fidgeted with it a bit more out there: "Moses' leadership" is used as an example of "no-extra-syllable" pronunciation, so it doesn't make sense to use it again as an example of "traditional" pronunciation; and i toned down the "traditional" pronunciation part a little because i know too many people who pronounce those possessives differently than presented. something still needs to be added about how to resolve disagreements about pronunciation when using Option 3 ... unless we change Option 3 to the much clearer OUP recommendation (just apostrophe when the last syllable of the name is pronounced like the verb is; apostrophe-s in all other cases). Sssoul (talk) 16:39, 4 September 2009 (UTC)

(outdent) Option 3 still needs work so that it provides clear advice to editors who come to the MoS in search of help with doubts or disputes; it's not helpful at all to just leave it at "discuss it and decide". so when the pronunciation of a possessive is debated or in doubt, what should the MoS recommend?

• a] use the OUP recommendation (just apostrophe when the last syllable of the name is pronounced like the verb is; apostrophe-s in all other cases)?
• b] leave pronunciation out of it and use either Option 1 or Option 2 throughout the article instead?
• c] stick with the probable pronunciation of earlier editors: if the article already includes the possessive of a noun ending in a single pronounced s, use that; if it's not the same possessive as the one you're in doubt/dispute over, maybe you can guess what that earlier editor's pronunciation of the possessive currently in question might be; and if there are no such possessives in the article yet, go back to a or b above, or see what articles on related subjects use, or just flip a coin, or ... ???

maybe it's clear that i think a or b will be more useful advice than c. any other ideas? Sssoul (talk) 15:38, 12 September 2009 (UTC)

Option C works for English dialects and date formats in non-region-specific articles, for the choice between spaced en dashes and unspaced em dashes, for the choice between italic d or upright d for the differential, and for anything else; besides, it has been endorsed by the ArbCom umpteen times. Why wouldn't it work for possessives which can be pronounced in two different ways? --___A. di M. 17:47, 12 September 2009 (UTC)

(outdent) because the pronunciation of possessives varies a lot more than those other issues do. a previous editor left Moses' right hand in an article. now i'm adding something about a new analysis by someone surnamed Charles, and if i find myself in a disagreement with another editor over how to write the possessive of that surname, no one has any way of guessing how that previous editor would pronounce the possessive of Charles. and no matter how we settle it for Charles, the same problem will come up when the next editor needs to add something about the even newer analysis that Richards came up with. Sssoul (talk) 18:52, 12 September 2009 (UTC)

or Davis, Harms or Dos Passos. the pronunciation-based option still needs finetuning to cover cases where there's doubt or dispute. again: the "when in doubt, follow the first editor" principle won't work here, because there's no way to guess what a previous editor's pronunciation was. does it make sense to adapt that principle to "when the pronunciation is in doubt or dispute, follow the spelling/punctuation of the first major contributor to use the possessive of a name ending in a single s"? (and isn't that tantamount to recommending Option 1 or Option 2 instead of a pronunciation-based style? Sssoul (talk) 06:14, 21 September 2009 (UTC)

Huh? If the possessive of Charles does not occur in the article until now, then you are the first editor adding it, so it's your choice. What you're saying is akin to saying that if a editor writes an article in British English but uses no verb which can end with either -ise or -ize in BrE, then if you want to use such a verb you have to guess what spelling the original editor would use. --___A. di M. 20:00, 21 September 2009 (UTC)

Concur here. That's silly. Users should just use common sense. If the article is written in BrE, then the contributor should not use a system that is incompatible with or incorrect in BrE, but other than that, leave them their freedom. Darkfrog24 (talk) 20:22, 21 September 2009 (UTC)

A di M, re "If the possessive of Charles does not occur in the article until now, then you are the first editor adding it, so it's your choice": but the question at hand is what to do when there's doubt or a dispute over the pronunciation, and i don't see how "it's your choice" helps, any more than "just discuss it and decide".

and Darkfrog, does ENGVAR have something to do with the question of how to settle disputes over the pronunciation of possessives? i don't see the connection.

as for common sense: that varies as much as the pronunciation of the possessive of Charles. 8) my common sense, for example, tells me that using apostrophe-s in all cases saves an awful lot of bother; but that if the Wikipedia MoS really wants to include a pronunciation-based option, the clear-cut OUP principle mentioned above would be a lot more functional for Wikipedia than the very vague Option 3 we have now. Sssoul (talk) 06:55, 22 September 2009 (UTC)

For ENGVAR, please see A. di M.'s comment. Where an article is already written in one form of English but does not yet use any possessives, then the next editor should select some form of possessive that is not incompatible with that form of English him or herself, not try to guess what the first major contributor would have preferred. Darkfrog24 (talk) 13:56, 22 September 2009 (UTC)

smile: all right, never mind – the spelling/punctuation of possessives isn't an ENGVAR issue, and i don't think A di M meant that it is (s/he just used -ise/ize as an example of something s/he views as similarly non-problematic). meanwhile, i do grasp that no one other than me sees any need to clarify the existing non-advice. oh well – i tried. 8) Sssoul (talk) 14:32, 22 September 2009 (UTC)

Use of "The" mid-sentence

I find the example that currently refers to a generic "the" confusing. A definite article cannot be generic. A noun phrase can be a proper name or it can be "generic", but not the definite artice itself. If the noun phrase is "generic", of course, there is no question of capitalization; and the United Kingdom is not generic. The noun phrase "the united kingdom" can, of course, be used generically: Britain is not the united kingdom that it used to be. I think what we need to say is that, other than in titles, the definite article is conventionally not capitalized, even when it is (or could be regarded as) part of a proper name: the United Kingdom, the Hebrides, the President of the United States. We should perhaps also state somewhere that the names of newspapers (such as the Guardian are not normally treated as titles in the same way as book titles (such as The Origin of the Species).

However there are a few conventional exceptions to this rule. I suggest we actually list the known exceptions, since there appears to be no general rule. "The Hague" is already given as an exception. Other candidates might be The Times and The Beatles. We should perhaps also discuss the exceptions to the exception, such as the Hague ruling, the Times article, and the Beatles song, where "the" is neither capitalized nor repeated. --Boson (talk) 15:47, 5 September 2009 (UTC)

See several threads in Talk:The Bronx for extended discussion of whether The Bronx should be so capitalized mid-sentence (consensus, which varies from that of many New Yorkers: no), combined with discussion of whether The is part of the Bronx's name (consensus: yes, except for the County of Bronx and the postal address Bronx, NY). ¶ By the way, I think that if a newspaper includes "The" in its name on the front page and in its masthead (something different), then "The" should probably should be both capitalized and italicised. It should certainly be italicised in "the Chronicle article". —— Shakescene (talk) 17:09, 5 September 2009 (UTC)

I see that The Oxford Guide to Style (like other incarnations of Hart's Rules?) says:

For periodicals, capitalize The only for one-word titles: The Times, The Spectator (but today's Times, the Spectator article); the New Yorker, the Yearbook of English Studies.

I actually thought the OUP only capitalized The Times, not the Guardian, as seems to be confirmed here and here, although the Guardian masthead has The Guardian. The Guardian style guide has:

newspaper titles: the Guardian, the Observer, the New York Times, etc, . . .

The Times style guide has:

newspapers and journals; use italics for titles and make sure to use The in the title whenever appropriate. . . . In the UK and Ireland with The in the masthead: . . . The Times, The Sunday Times, The Sun; The Guardian. . . .

Chicago (14th ed.) has:

When newspapers and periodicals are mentioned in the text, an initial The, omitted in note citations, is set in roman type and, unless it begins a sentence, is lowercased.

. So practice varies quite a lot.--Boson (talk) 20:24, 5 September 2009 (UTC)

I was just saying (as an aside) what I think, as opposed to whatever consensus might exist, but there's certainly a lot of variation, in practice as well as in formal style-guides. Newspapers didn't like to use italics for various reasons, but they often capitalized The in newspaper's names (in roman type) within the body text. Now italics aren't so difficult, but many newspapers have dropped "The". The Guardian itself seems to be of several minds about what to call itself, partly because different departments or different generations have different feelings about its origins as The Manchester Guardian [and Manchester Evening News]; on the one hand you have The Manchester Guardian Weekly and on the other, "guardian.uk" (describing the website). I find it much easier to see that (for example) "the Observer" refers to the newspaper if it's italicised or The is capitalised, or preferably both, because the Observer (or the Guardian or the Appeal), even capitalised, could refer to something else. —— Shakescene (talk) 21:04, 5 September 2009 (UTC)

On the grounds of accuracy, I think my personal preference for newspapers would be the same as yours, i.e. capitalize (and italicize) The if it is the name printed on the front page and in the masthead, but I haven't thought through all the consequences; compliance might be a problem if it requires knowledge of the masthead. I would also omit "The" when the name is used attributively (. . . the Times article). I am still wondering about consistent rules for foreign language periodicals; I would personally tend to write according to a report in Der Spiegel but . . . in spite of the Spiegel's reputation for getting its facts right; on the other hand I would probably write . . . in spite of Le Monde's reputation for getting its facts right, and the use of italics followed by a roman apostrophe is a bit of a problem for Wiki markup (and possibly for legibility). Do we actualy have a rule on the use of a roman apostrophe following an italicized name. --Boson (talk) 10:47, 6 September 2009 (UTC)

I wouldn't consider the article in "the United Kingdom" or in "the President of the United States" as part of the proper name, but I would in "The Origin of the Species" and "The Hague". So I would do something like that:

The definite article is not normally capitalized in the middle of a sentence, except when it is part of a proper name. There are idiomatic exceptions both ways: for example, titles of newspapers and other periodicals, especially those where several words follow the article and those of periodicals not published in the UK or Ireland, are usually treated as if the article were not part of the name, even if it is.) Common usage should be followed on a case-by-case basis. As usual, it is a good idea to consult the sources of the article.

Incorrect

He had read an article about The United Kingdom.

Correct

He had read an article about the United Kingdom. (The name is "United Kingdom".)

Incorrect

There are two seaside resorts in the Hague.

Correct

There are two seaside resorts in The Hague. (The name is "The Hague".)

Incorrect

Homer wrote The Odyssey.

Correct

Homer wrote the Odyssey. (The title is "Odyssey".)

Incorrect

Tolkien wrote the Lord of the Rings.

Correct

Tolkien wrote The Lord of the Rings. (The title is "The Lord of the Rings".)

Incorrect

He had read an article in The New York Times.

Correct

He had read an article in the New York Times.

I think "the Beatles" is a special case: "Beatle" is often used as a countable noun to refer to the members of the band (whereas no-one would refer to Jim Morrison as a Door), so there's nothing to stop you from claiming that when writing "the Beatles" you are simply using the noun for the members of the band, especially if you use a plural verb, as usual in British English. --___A. di M. 11:11, 6 September 2009 (UTC)

But can it be succinctly put? In about 20% of the words? Tony (talk) 13:55, 6 September 2009 (UTC)

... sorry, but who decided the title of the newspaper isn't The New York Times??

and in my experience editing music articles, on Wikipedia "The Beatles" isn't an exception; the "the" in band names gets capitalized mid-sentence more often than not: "a member of The Doors/a founder of The Rolling Stones/the drummer in The Clash" - it's not what i do in my normal writing, but it's one of those Wikipedia Style Things i've gotten used to. (and by the way, A di M, the use of a plural verb after "(the) Beatles" isn't British English - it's internationally normal. i think you're thinking of the use of a plural verb in some situations after a singular collective noun like the band are ready to go - that is indeed way more British than American.) Sssoul (talk) 15:42, 6 September 2009 (UTC)

Taking a look at the leads of some articles in WP:FA#Music about American bands, I see "Alice in Chains is an American rock band ...", "Audioslave was an American hard rock supergroup ...", "Megadeth is an American heavy metal band ...", "Metallica /məˈtælɨkə/ is an American heavy metal band ...", "Nine Inch Nails is an American industrial rock project ...", etc. I guess the reason for the plural verb for "the Beatles" is the same one (you wouldn't refer to Trent Reznor as a Nine Inch Nail). British (and Irish) English appears to use the plural consistently: "Radiohead are an English alternative rock band ...", "Motörhead are a British rock band ...", "U2 are a rock band formed in Dublin, Ireland." (BTW, the argument about "the Beatles" was the other way round: there are people writing it with a lower-case "the"[1][2], but you wouldn't want to do that with "The Doors". --___A. di M. 16:09, 6 September 2009 (UTC)

So how does one treat Bob Dylan's former backup, The Band?. I wouldn't write "... the Band are well-known ..." or "...the Band are a well-known band ..." —— Shakescene (talk) 17:07, 6 September 2009 (UTC)

That would depend not on whether the band in question had an American or British affiliation, but whether the article did. Since the Band is American, its article would probably be written in American English, so we'd say "the Band is/was," but if the Band got a mention in any article that, for whatever reason, was written in British English, we'd say "the Band are/were."

Using a plural verb on a singular noun strikes me as off and wrong, but hey, so does spelling "color" with a U. Darkfrog24 (talk) 18:42, 6 September 2009 (UTC)

(edit conflict) The article uses the singular when referring to the band as a whole ("The Band was a rock music group ..."), which is fine in American English (I assume Canadian English does the same); but I'm not sure I'd write "the Band" with a small tee as the article does. --___A. di M. 18:46, 6 September 2009 (UTC)

• OK, it's worth inserting this, as has been done, and well done for spotting the issue; but the addition is 136 words long! It can be done in 41 words, in addition explicitly capturing the account-for-usage issue ("the UK", but "The Hague"). Tony (talk) 04:04, 7 September 2009 (UTC)

sorry about the singular/plural digression - it's not the case that British English always uses plural verbs after singular collective nouns, and Nine Inch Nails is an interesting case - but that's not the question here. all i meant to point out was:

• if Wikipedia is adopting a standard not to midsentence-capitalize the "the" in band names, it's fine with me, but it will be a significant change from current practice; and
  
• why is "The New York Times" suddenly deemed incorrect, when the name of the paper includes the article? Sssoul (talk) 06:52, 7 September 2009 (UTC)

I'm going to second Triple S here, the The is fully part of the name of The New York Times, and should be capitalized when the full name is used. When the abbreviated form "the Times" is used, it can be lowercase, as it's not part of the actual, full title there, but it should be capitalized for the full titleoknazevad (talk) 02:15, 9 September 2009 (UTC)

• From the boxed text above: isn't the name—grammatically—"the United Kingdom", not "United Kingdom", at least as a noun, even though we drop the "the" in short texts and by convention in other contexts (Liverpool, UK), and as an adjective ("her UK passport"). So there's a difference, is there, between a grammatical unit as title, and a title title ("the" vs "The), maybe? I'm thinking through ... "the The Hague City Council" ... weird. I'm lost. Tony (talk) 11:04, 8 September 2009 (UTC)

Sorry to interrupt, but first of all I'd like to say that I haven't read this whole arguement, but in the ESL business we haven't been able to come up with a better summary of the rule than this:

Capitalize "the" if it is an integral part of the name.

(The "mid-sentence" part is covered in the rule about starting sentences, but you might like to keep it.)

For example: They call him The Hulk. but You know, the Hulk Hogan from the new reality show is a very different Hulk Hogan from that of my childhood.

I would definitely caution against trying to spell out in so many words what exactly would constitute "an integral part". Believe me, it's not a simple matter - take it from me, I've had to look it up and prepare lesson plans for groups of Slavic and Asian language students, and it takes a long time to master and then you really never get down to real rules, only general guidelines with an unsatisfying number of exceptions (yes for newspapers, no for rivers, ad nauseam.)

Another reason not to try to spell out "an integral part" is the fact that it's unnecessary. You're not teaching ESL here, and you have to assume that the authors of text on English Wikipedia know English or they'd be at the Chinese one or something. People who know about The Beatles or Talking Heads are responsible to know if the definite article is an integral part of the name or not in a particular context. Let them check the sources or argue it out on the proper talk pages if there's some doubt.

I hope this helps! Chrisrus (talk) 04:44, 9 September 2009 (UTC)

'Capitalize "the" if it is an integral part of the name and let individual articles sort out the application of that for themselves' sounds right to me, but it's probably worth adding a caveat that the titles of Wikipedia articles are not a reliable guide. (as Tony points out, it's the [or The] United Kingdom, even though the title of the Wikipedia article omits the article.) Sssoul (talk) 04:58, 9 September 2009 (UTC)

One good rule of thumb for determining whether something is an integral part of a name is whether or not "the" would be set off from the body text along with the rest of the name. It's not The New York Times or the New York Times. It's The New York Times.

Of course, it doesn't work all the time. Also, I'm not sure if "integral part of the name" necessitates that the definite article must be categorized. We almost never refer to the Hulk as just Hulk, but the "the" is uncapitalized. Darkfrog24 (talk) 13:32, 9 September 2009 (UTC)

Well, maybe The Hulk is not always written with a capital T on the definate article, but it absolutely should be, because that's his name: The Hulk. Why would you say that The Hulk is not written with a capital T? It is too! Chrisrus (talk) 06:35, 13 September 2009 (UTC)

That is one of the things that was being discussed: whether Wikipedia should always capitalize "the" when it is part of a name; and if not, what criteria to apply. Different publishers observe different conventions. Otherwise everybody would write "The Argentine", "The Hebrides", "The Guardian", "The United Kingdom", etc. --Boson (talk) 08:37, 13 September 2009 (UTC)

That's what we're talking about. Just because an article is part of the name doesn't mean that it's a capitalized part of the name. If you pick up a copy of Marvel Zombies 2 (delightful, by the way), you will see the characters refer to Hulk as "the Hulk." Darkfrog24 (talk) 13:37, 13 September 2009 (UTC)

As Darkfrog24 rightly says, "the" is not an integral part of the proper name when referring to the Hulk. It is with the formal title The New York Times or The Hague, but not with the Hulk, the Joker, the Spirit, or the Phantom (the character, although the comic strip is formally titled The Phantom). Please see these blue-linked examples. Consistency is important in order to keep all of us on the same page and avoid the encyclopedia looking sloppy.-- Tenebrae (talk) 14:46, 13 September 2009 (UTC)

Although "integral part of the name" sounds intuitively correct in some undefinable way, I don't know if it is much use. One could put it round the other way and judge if "the" is an integral part of the name based on whether you would capitalize it. "The" is an integral part of the name, in the sense that it is always included, even in names where "the" is not normally capitalized in running text. The exception would seem to be that the is always omitted when the name is used attributively (so you don't write several The Hague tribunals). I am also not sure that we can leave it up to the sources, since different publications have different house styles, which they apparently use consistently. Wikipedia apparently uses The Guardian, and this seems to be confirmed by The Oxford Guide to Style but the newspaper in question refers to itself as the Guardian. I also have a feeling that "the" might be capitalized in things like lists, e.g.

• The Federal Republic of Germany
• The United States of America
• other sovereign states

This is probably not very helpful, but the rule seems to be that in running text the, as the first word of a name or title, is normally not capitalized, except for some (or all) names of newspapers and other specific individual exceptions (not including all bands and fictional characters). I am tempted to draw a line between names and titles. --Boson (talk) 16:40, 13 September 2009 (UTC)

I think it's time to consult some more guides on this. If Oxford, MLA and Chicago all agree that "the" should not be capitalized unless it either leads a sentence or is part of a specific title such as "The Hague," then perhaps we should replace this wording with, "Do not capitalize 'the' mid-sentence unless it is part of a specific title." Darkfrog24 (talk) 18:12, 13 September 2009 (UTC)

re: "I am tempted to draw a line between names and titles" - so is The Guardian (and other such newpapers' appellations) a name or a title?

and re: "unless it ... is part of a specific title name such as 'The Hague'" - that leads directly back to the very same question you find problematic: how does one determine whether or not it's part of a name - why The Hague but not The Hulk, The Everly Brothers, The UK and The Mail on Sunday? Sssoul (talk) 06:20, 15 September 2009 (UTC)

Here is a test you can apply to determine whether The is part of the name of some entity. Suppose you have an occurrence mid-sentence like an article in The New York Times – or should that be an article in the New York Times? Just replace the name by the name of an analogous entity that does not start with The, for instance Le Monde. What do you get? This: an article in Le Monde. This shows that in this example The is part of the name and should be capitalized.

However, even if The is part of the full name of some entity, it is not always part of the name when used before the name without The. Take as an example an occurrence mid-sentence like a member of the New York Times staff. Or should that be a member of The New York Times staff? You can apply the same test. Replace the name by Le Monde as before, and you get: a member of the Le Monde staff. You can see that in this example the word the is mandatory; so in this example it is the standard grammatical definite article and not part of the name. Therefore it is also a member of the New York Times staff. You might possibly use a member of the The New York Times staff, but that sounds strange and is unusual. Or you could use a member of The New York Times's staff, just like a member of Le Monde's staff.  --Lambiam 07:23, 15 September 2009 (UTC)

This is why I think we need to consult the style guides again. I begin to recall that we are supposed to write "the New York Times" for newspapers whose names are in English. I remember that exceptions are made for Le Monde and its ilk because English-speaking writers should not always be expected to know which words translate to "the" and which do not. Darkfrog24 (talk) 13:19, 15 September 2009 (UTC)

I think the test of replacing it with a name that doesn't have "the" works to determine if the is part of a proper name, but that does not necessarily mean that "the" is capitalized. For instance:

• Washington is a town in America.
• London is a town in the United Kingdom (but Hollywood is a district in Los Angeles).

Using the Le Monde analogy, one should then write a member of the The Guardian staff, but one doesn't when the name is used attributively, perhaps partly in order to avoid repetition of the. I suspect that titles of books (as opposed to names of newspapers) are not normally used attributively (so you can write a typical Times viewpoint but you should not write a typical Origin of the Species viewpoint.

Perhaps the rule (with exceptions like The Hague) is that in running text the is not capitalized in names but is capitalized in titles, and that the names of newspapers are sometimes treated as titles (like books).--Boson (talk) 18:50, 15 September 2009 (UTC)

(outdent) ... i love going around in circles! it's similar to the situation with possessives, above: other styleguides recommend all kinds of different things, and meanwhile Wikipedia needs a principle that's clear and easy to explain and to apply. "capitalize 'The' when it's an integral part of the name" is not what i do in my own off-Wiki writing, but it is clearer and easier than anything else that's been proposed so far (as long as a caveat is included that the titles of Wikipedia articles do not reliably reflect whether "the" is part of a name - cf United Kingdom). Sssoul (talk) 08:43, 16 September 2009 (UTC)

What's clearer and easier is: "Capitalize the 't' of 'the' when it's at the start of a title or subtitle; not elsewhere." ¶ I dropped my wristwatch into The Thames; last week I flew across The Pacific; you should take that to The United Nations; I wonder how The United States will respond; will there ever be another André The Giant; is there a man on The Moon? All pretty silly if you ask me. When I was a kid I liked the Beatles and the Rolling Stones but I wasn't so fond of the Who: all perfectly fine. This nervous capitalization of "T" looks to me like some kind of overcompensation; it reminds me of "They invited my husband and I". -- Hoary (talk) 09:00, 16 September 2009 (UTC)

yep, i'd go along with that too, as long as we can specify succinctly what's meant by titles/subtitles as opposed to names (and whether or not The Hague is an exception). Sssoul (talk) 09:12, 16 September 2009 (UTC)

When I first read that comment I had no idea what you meant. Name, the Thames; title, The Listener. But it occurs to me that you might mean something like "the Duke of Edinburgh". All right then: "Capitalize the 't' of 'the' when it's at the start of the title or subtitle of a creative work or publication; not elsewhere." I'm perfectly happy to write "the Hague", but the last time I said that hereabouts I got a bollocking from the regulars, so OK we add a third sentence: There are a handful of exceptions and either list them or point people to the refreshingly unpompous Guardian style guide. ([http://www.guardian.co.uk/styleguide/t This is good, though it oddly seems to prescribe "The Thames" etc.) -- Hoary (talk) 09:32, 16 September 2009 (UTC)

re "oddly seems to prescribe "The Thames" etc": damn, so we're going around in circles again? that can be fun for a while, but i vote for not recommending stuff that contradicts what we're trying to propose.

what i meant was "title of a creative work or publication" (does it need to include "website or broadcast" too?); i think it's necessary to be explicit about that, because someone above calls "The Hague" a title, and there was some back-and-forth about whether newspapers have titles or names. but the point about not meaning honorific titles needs to be clarified as well, lest we get The King of Spain, The Queen of Soul, etc. Sssoul (talk) 09:43, 16 September 2009 (UTC)

So don't recommend it. Here you go: Capitalize the 't' of 'the' when it's at the start of the title or subtitle of a creative work or publication; not elsewhere. If the result seems odd, look up the particular term at The Guardian's style guide (available online here) and follow its advice. -- Hoary (talk) 09:56, 16 September 2009 (UTC)

So for these purposes, it's a title if in italics or quotes? I read the Guardian, but I read The Guardian and I read "The Guardian"? Tony (talk) 10:28, 16 September 2009 (UTC)

My personal preference would be to allow either "I read The Guardian" or "I read the Guardian": insistence on the former seems a bit pedantic. But insistence on it is likely to be most people will want hereabouts, so fine. "I routinely read The Guardian", "I thrice read The Mezzanine" (novel by Nicholson Baker), "I twice watched The Magnificent Ambersons", "I read 'The Size of Thoughts'" (essay by Baker within a book so titled), "I chuckled at 'The Admirable Crichton'" (chapter within The Exotics by Morris Bishop), "I turned my nose up at 'The Adventures of Id'" (poem within A Bowl of Bishop; need I specify the author?). All so simple, even a Wikipedia contributor can understand it. Oh all right, pedant-fodder for y'all: last week I read The Adventures of Huckleberry Finn, as it was a photographic reprint of the first British edition, which has the definite article to which some Twainologists loudly object. -- Hoary (talk) 10:44, 16 September 2009 (UTC)

I am starting to think that this is a can of worms, and that a centralised style guide might be better to advise editors to go with the consensus in the English-language sources, to be consistent within an article, and to use common sense to avoid what might look awkward. I've just been at an article that has "The Beatles, The Crickets, The this and that", then "the Crickets" further down. Tony (talk) 12:49, 16 September 2009 (UTC)

When capitalizing the T mid-sentence would be incorrect, the MoS should not recommend it, regardless of what most people want or seem to be doing. I like most of Hoary's text, but I don't know if we should endorse one external style guide in particular. Finding a consensus within style guides might be good. Darkfrog24 (talk) 13:07, 16 September 2009 (UTC)

finding something that be easy to explain and apply on Wikipedia would be excellent. Hoary's proposal seems to be coming close, except i still don't understand recommending the Guardian styleguide if it recommends "The Thames, etc", which is contrary to Hoary's basic proposal.

meanwhile, Tony's remark puzzles me. sure, there are articles where editors mistakenly use various capitalization styles, because (as the discussion here shows) in the outside world practices vary a lot, and that includes sources and styleguides. that doesn't mean Wikipedia's styleguide should avoid recommending anything! Sssoul (talk) 13:19, 16 September 2009 (UTC)

Maybe you're right, Sssoul. But I'm finding this one too hard to resolve in my head. Tony (talk) 13:42, 16 September 2009 (UTC)

• How about this:

Inter-sentence capitalization of "The" in proper nouns

• When the word "the" is formally part of a proper noun; e.g., The New York Times and The Hague, capitalize the word "The"; e.g., After long deliberations, The Hague issued its decision on….
• When "the" is to be used as a definite article (to indicate a particular one or ones) immediately preceding a proper noun that begins with "The", use "The" only once and capitalize; e.g., Scott Martens’ article first broke in The New York Times after… and not …first broke in the The New York Times after… and not …first broke in the New York Times after…

Greg L (talk) 16:09, 16 September 2009 (UTC)

Despite my earlier beliefs, I am pretty sure that it is "the New York Times" even though "the" could be said to be part of the name. Darkfrog24 (talk) 16:28, 16 September 2009 (UTC)

• Just go to their Web site: The New York Times, all in their logotype. The *newspaper* is “The New York Times”. The *company* is “The New York Times Company”. Greg L (talk) 16:50, 16 September 2009 (UTC)

Yes, but that doesn't mean that the definite article should be capitalized all the time. Lots of companies use different capitalization in logos. And, this is more visible in their science articles, NYT doesn't always get it right, especially not in their bloggier articles. On issues of grammar and capitalization, we should consult the most reputable style guides. Darkfrog24 (talk) 17:34, 16 September 2009 (UTC)

• The Associated Press’ on-line manual of style has an “Ask the Editor Questions” section. There is a thread that deals fifteen times with The New York Times (either of the two links here). The editors’ answers there seem authoritative and clear enough to me on this issue. Greg L (talk) 17:10, 17 September 2009 (UTC)

The Chicago Manual of Style also says not to capitalize "the" in "names of periodicals." Darkfrog24 (talk) 17:49, 17 September 2009 (UTC)

• Well, the options are A) do nothing, B) flip a coin, C) choose what looks most natural, D) have MOS remain silent, E) or argue about this until the heat death of the universe. I prefer the A.P. method myself. I think it looks most natural by far. I think the A.P. properly decided that “The New York Times” should be treated the same as “The Hague.” Perhaps you don’t. I’m done here. Goodbye. Greg L (talk) 22:44, 17 September 2009 (UTC)

I'm okay with option A. The MoS's current take on the matter advises looking at things on a case-by-case basis. That looks like it'll do for now. Darkfrog24 (talk) 23:04, 17 September 2009 (UTC)

Question

I have a question about two of the image requirements, which I often see raised at FAC.

1. Avoid sandwiching text between two images that face each other.

2. Do not place left-aligned images directly below a subsection-level heading (=== or lower), as this sometimes disconnects the heading from the text that follows it.

I'm someone who likes to do both: I like sandwiching, and I often use left-aligned images directly under a subsection heading. I often use them when the image (if it's a person) on the left is looking directly at the image on the right, and vice versa. When the text in between is about some issue these people are having with each other, it means the article visually represents the text.

I've never seen left alignment under a subhead, or text sandwiching, cause a problem (unless there are other templates or quote formats in the way). Are there still solid reasons for having these requirements in the MoS? SlimVirgin ^{talk|contribs} 12:49, 13 September 2009 (UTC)

I fully agree with Slim on the left alignment under a heading (why is a No. 2 heading OK but not a No. 3 heading, anyway?). I do have a problem with sandwiched text, although I concede that it might occasionally work thematically, and partly depends on the text size a reader's browser is set at, and indeed the image sizes forced by the editors or the default size for thumbnails chosen by a WPian. The trouble is that most editors are not aware of the need to be cautious when it comes to images placed horizontally adjacent—to try to make it work or to use alternative placements it it does not. Perhaps we might consider expressing the guideline in a way that cautions (e.g., "usually") but does not forbid. Tony (talk) 13:11, 13 September 2009 (UTC)

Apparently there is an accessibility issue with putting left-aligned images under subsection headers. I'll ask at WT:ACCESS for clarification. Dabomb87 (talk) 13:31, 13 September 2009 (UTC)

Cautioning without forbidding would be great. SlimVirgin ^{talk|contribs} 13:57, 13 September 2009 (UTC)

I think "sandwiching" would look real crappy when the images aren't very narrow and the browser window isn't very wide. Some users (e.g. of netbooks, smart phones, etc.) might have to read text in a column with a width of two or three words. But I think that it'd be more useful to just say something amounting to "when adding many images in the same section, narrow your browser window to about 800 pixels or so and make sure that the layout still doesn't look crappy that way", than to give a number of rules which, in absence of a rationale, sound like someone pulled them out of their ass.--___A. di M. 14:22, 13 September 2009 (UTC)

A, can you advise how to narrow the browser to 800 pixels i.e. how we would know when it was narrowed to that degree? Sorry if that's a stupid question. SlimVirgin ^{talk|contribs} 14:34, 13 September 2009 (UTC)

For example, you could open or create a picture of that width, and narrow the window to that it fits in the picture horizontally. --___A. di M. 14:44, 13 September 2009 (UTC)

Okay, I see, thank you. :) SlimVirgin ^{talk|contribs} 15:41, 13 September 2009 (UTC)

I have nothing to add to what I already said at this discussion. Also see Eubulides' comment there. I don't know how sandwiching of images would affect users, because I exclusively use a screen reader. Graham87 15:43, 13 September 2009 (UTC)

Thanks, Graham. To judge by the discussion you linked to, there seems to be no good reason to have this requirement in the MoS. I'm wondering if it's one of those things that has acquired a life of its own over the years, so everyone assumes there's something to it. SlimVirgin ^{talk|contribs} 15:46, 13 September 2009 (UTC)

• I agree that there doesn't seem to be any WP:ACCESSIBILITY reason for the requirement. I suspect (without really knowing) that it is an aesthetic preference, as some editors might object to a layout that looks like this (the Xs are an image):

Subsection header [edit] XXXXXXXXXXXXXXXXXXXXXX The first sentence XXXXXXXXXXXXXXXXXXXXXX of the subsection XXXXXXXXXXXXXXXXXXXXXX starts here and then XXXXXXXXXXXXXXXXXXXXXX there's more text that eventually flows under the image.

because there's a wide visual gap between the subsection header and the first sentence. In contrast, a top-level section looks like this:

Section header________________________[edit] XXXXXXXXXXXXXXXXXXXXXX The first sentence XXXXXXXXXXXXXXXXXXXXXX of the section XXXXXXXXXXXXXXXXXXXXXX starts here and then XXXXXXXXXXXXXXXXXXXXXX there's more text that eventually flows under the image.

and the horizontal line draws the eye from the section header to the first sentence. If my hypothesis is correct, this stylistic preference, while perhaps being a reasonable preference for the default skin, is to my mind too weak to be a guidance that should appear in the MoS.

• In light of the above discussion, I propose that we delete the bullet 'Do not place left-aligned images directly below a subsection-level heading (=== or lower), as this sometimes disconnects the heading from the text that follows it. This can often be avoided by shifting left-aligned images down a paragraph or two.'
• For the sandwiching suggestion, I propose removing 'Avoid sandwiching text between two images that face each other.', and capturing the more-general idea by prepending the sentence 'Images should be laid out so that they work well with browser windows as narrow as 800 pixels and as wide as 2000 pixels.' to the bullet 'See this tutorial for how to group images and avoid "stack-ups".'

Eubulides (talk)

I'd be fine with that. SlimVirgin ^{talk|contribs} 17:42, 13 September 2009 (UTC)

I support this change. Don't forget to update this on all style guidelines. Dabomb87 (talk) 00:42, 14 September 2009 (UTC)

Images can work in very strange ways with section and subsection headers, so there might indeed be a reason, but I'd like to see it articulated somewhere. On the other hand, there can be a definite problem with right-aligned images and tables, because not only the information box, but various medallions hang down from the right (see, e.g. New York City mayoral election, 1917) and can push right-aligned images, maps and tables down in very awkward ways, besides threatening severe sandwiching problems with left-aligned images. Caught between the Devil and the Deep Blue Sea or Scylla and Charybdis. —— Shakescene (talk) 05:43, 14 September 2009 (UTC)

• Has the guideline been changed yet? Could we have the new and the proposed texts here, first? Tony (talk) 12:08, 14 September 2009 (UTC)

  I haven't changed it yet, to give time for discussion. The new and proposed texts are here; please see the two words "propose" earlier in this thread (which I've just now boldfaced). Eubulides (talk) 16:21, 14 September 2009 (UTC)

The proposal is to:

1. remove "Do not place left-aligned images directly below a subsection-level heading (=== or lower), as this sometimes disconnects the heading from the text that follows it. This can often be avoided by shifting left-aligned images down a paragraph or two."

I'm going to go ahead and do that, as there seems to be consensus, and:

2. remove "Avoid sandwiching text between two images that face each other," and replace it with, "Images should be laid out so that they work well with browser windows as narrow as 800 pixels and as wide as 2000 pixels." Then I didn't understand the next part of Eubulides proposal above, which was to prepend the above to the bullet "See this tutorial for how to group images and avoid 'stack-ups'." SlimVirgin ^{talk|contribs} 15:59, 15 September 2009 (UTC)

I followed up in #Sandwiching text below. Eubulides (talk) 23:34, 16 September 2009 (UTC)

Sandwiching text

Here's proposal 2 in more detail (inserted text, deleted text):

• Images should be inside the section they belong to (after the heading and after any links to other articles), and not above the heading.
• Avoid sandwiching text between two images that face each other.
• Use captions to explain the relevance of the image to the article (see Captions, below).
• Images should be laid out so that they work well with browser windows as narrow as 800 pixels and as wide as 2000 pixels. See this tutorial for how to group images and avoid "stack-ups".

Eubulides (talk) 19:54, 15 September 2009 (UTC)

• I welcome the removal of the left-facing bit, but don't understand how sandwiching works with image preferences. I have mine set as 300px (as any sane registered user surely does) and sandwiching of non-forced images always looks awful on my 1024 x 768 - you get 2 to 3 words per line. Or can I now go round saying anything I think doesn't "work well" on my screen is contrary to the MoS? It seems rather subjective. Johnbod (talk) 05:46, 16 September 2009 (UTC)

I'm traveling and haven't been able to keep up with all of these discussions; when all is set, would someone summarize the changes to WT:FAC so reviewers will be aware? It escapes me why this (no left-aligned images under third-level headings) was part of WP:ACCESS for so long if it's not an accessibility issue, particularly since I worked so hard to make sure FAs complied with accessibility ... but ... it is what it is. SandyGeorgia (Talk) 23:00, 16 September 2009 (UTC)

On further thought, the proposed 800–2000px requirement may be too weird. We can't reasonably expect editors to test lots of browsers and window sizes. And the builtin parts of the Wikipedia skins handle 2000px windows so poorly that there's little point to trying to optimize for such display (a phrase involving "lipstick" and "pig" comes to mind ...). Eubulides (talk) 23:34, 16 September 2009 (UTC)

• Oppose. My old desktop is only 800px, and any sort of sandwiching looks terrible. Even on my wide laptop I don't like the look of sandwiched images at all. They should be staggered or limited. Sandwiching also looks bad on my iPhone. Saying 800 to 2000px is not always understood and is too specific when all sandwiching looks bad. Reywas92^Talk 00:18, 17 September 2009 (UTC)
• Oppose. My general usage screen measures a health 1440px, with my screen generally around 1000-1200px wide, and sandwiching still looks bad. Image stacking and sandwiching just doesn't look great (neither does left aligned images breaking up section headings, either, in my opinion.) It basically adds visual clutter and decreases the image afforded to any one image, as a result they all mean a lot less. I think simply saying "Try to optimize image flow for a variety of screen resolutions" would be the best wording. Der Wohltemperierte Fuchs ^(talk) 00:47, 17 September 2009 (UTC)
• Oppose per my comments 3 or so up, & D. Fuchs & Reywas. Johnbod (talk) 04:23, 17 September 2009 (UTC)
• Qualified oppose Sandwiching generally looks bad even in a 1440px-wide window, because it cramps the text. Placing images side-by-side, e.g. in a table, is equally bad most of the time. In both cases the exception is that two small images may not cramp the text too much.
  Re "Images should be inside the section they belong to (after the heading and after any links to other articles), and not above the heading", I agree with "not above the heading", but disagree with "after any links to other articles" - on wider windows, paragraphs are less tall, this makes image layout more difficult, and placing images alongside links to other articles may reduce gaps ({{clear}}) added at ends of sections to prevent images from overspilling into the next section.
  "browser windows as narrow as 800 pixels and as wide as 2000 pixels" is totally unrealistic - I'm starting a new section on width below. --Philcha (talk) 06:03, 17 September 2009 (UTC)

Concern over upright= method

This may be used as an alternative to the pixel method ("upright=1.5" as opposed to "270px"). In this case, it's 1.5 times whatever a logged-in user has preferenced for default image sizes (1.5 times 180px). I find this to be most unsatisfactory. Here's why:

1. Editors have somehow been encouraged not to pixel-size thumbnails up from the default, which in many cases is tiny at 180px.
2. Because many images are sized by this tiny default, some WPians choose a larger default (say, 300px).
3. In a circular pattern, if editors upsize their images using the multiplier method, 300px, for example, becomes a gigantic 450px for that logged-on user (but a more reasonable 270px for all othersand our readers, for whom the multiplier is based on 180px).
4. Yet a logged-on, preferenced user who has chosen a default of 300px for thumbnails might object if she sees an image forced to, say, 250px, which overrides their higher default setting.

I must say that I'd be happy to end the multiplier method and stick with resizing through the specification of actual pixels. This is by far the more common method: what we see is what we know our readers will see. Editors, IMO, should be encouraged to resize images gradually over the next few years using the px method. Tony (talk) 03:13, 17 September 2009 (UTC)

• The absolute-pixel method has a real problem. Suppose a reader prefers large images and specifies 300px thumbnails. Suppose also that an editor wants the image to be somewhat larger than usual and specifies 200px to make it so. (This is a common scenario; it happens to me often.) Now, the reader and the editor both want the image to be larger than usual; but the result of their efforts is to make it smaller than usual for the reader: it is 200px rather than the default 300px.
• On a high-resolution screen, 450px is not gigantic. It's fairly small. It's relatively common these days to have desktop screens that are 1920 pixels wide, and 450px is less than a quarter of the way across these screens. Many editors with big screens set default thumbnail size to be 300px because they don't like Wikipedia's tiny default size. Wikipedia articles should work well on these popular large displays. Absolute pixel widths generate articles that scale relatively poorly to these high-resolution monitors.
• To some extent this problem has always been with us: after all, there have always been some readers with higher-resolution screens, and some with lower. But the problem is getting worse now, due to the divergence between desktops (where 2 megapixels is now quite common) and netbooks (where 600 kilopixels is fairly standard). It's likely that the variance between monitor sizes will get even greater in the future. Furthermore we should keep in mind the problems of visually-impaired (but not blind) readers, who have to deal with far fewer than 600 kilopixels effective. Wikipedia articles should aim to display well on the broad class of displays in current use; we should not attempt to retreat back to the circa-2000 world where specifying fixed-size images was good enough.
• I'd rather not rely on a solution that requires editors to modify all the sizes of all the images by hand, each several times, over a period of several years. Aside from the fact that this scheme assumes an obsolescent world where monitor resolutions don't vary so much, there has got to be something better than that. (Just think of all those wasted editor hours! not to mention edit wars!)

Eubulides (talk) 04:05, 17 September 2009 (UTC)

• The upright= method also has at least one real problem, that a)for several years editors have mostly been working with the other method, and b)I don't think the change was discussed nearly widely enough. I normally think I keep up with image discussions llike this, but this one I never saw - perhaps I was on holiday, like most of the rest of the world. The solution to the problem with pixel size, which at one point was the guideline, is to avoid fixing pixels below 300 in all normal circumstances. This worked as well as anything can, given the range of kit people use WP on. If the MOS stays as it is, how long will it be before all articles are changed over? In the meantime preferences have to be reset every time an article using the current policy is encountered (happily still very rarely). It is surely the new policy that will cause thousands of wasted editor hours. I think it should be debated more widely, & I would support reversing it. Johnbod (talk) 04:23, 17 September 2009 (UTC)

• It is true that the px method has been around longer, and that it's easier to explain than |upright=. These are its main advantages.
• The change was discussed extensively and gained consensus; I'd guess it consumed about 8,000 words and two weeks. It might help to read that thread to avoid overrehashing now.
• "at one point was the guideline, is to avoid fixing pixels below 300 in all normal circumstances" That advice was amusing, as it dueled with Wikipedia:Layout, which until quite recently said the exact opposite, namely that images should never be resized above 300px. Given the obvious guideline confusion and lack of consensus about sizes of images in Wikipedia, I have become pretty wary of having guidelines recommending specific pixel sizes. Whatever particular limits are suggested are often quite wrong for real-world cases. I don't even particularly agree with the 500×400px limit currently in the manual of style (which disagrees with the 550px limit in the picture tutorial, but never mind....), and would favor removing all specific pixel limits.
• "preferences have to be reset every time an article using the current policy is encountered" Can you please give an example of why preferences have to be reset with |upright=? What article has this problem, what preferences are needed to cause the problem, and what's the screen and browser-window size? In contrast, I can vouch for the fact that when I run Firefox in a window 1500 pixels wide, and tell it to use a 20-point font rather than the default 12-point, and tell Wikipedia to use 300px thumbnails rather than the default 180px (these are all reasonable settings for a mildly visually-impaired reader), the upright=-using sections of the picture tutorial look pretty good, whereas the px-using sections look absurdly out of whack: that fixed-size flag of Scotland looks like some Scotland-hater has been editing the page in an attempt to the flag look like a postage stamp with a pompously-oversized caption.

Eubulides (talk) 05:27, 17 September 2009 (UTC)

The only editor whose work I see who has implemented the new policy is User:Ealdgyth. Her Gregorian mission looked ridiculous to me after she had done so], and still looks pretty bad now after I protested & she adjusted. In an article with more images, like any art article, it would be just impossible. Also Gilbert Foliot and others of her FAs. I rather doubt the Layout guideline said "never" to go above 300px, though with these things changing all the time, who knows. There was a time when the guidelines said that fixing should generally not be below, and not much above, 300px, which was sensible, and worked. Re Tony below - I would be interested in a figure for those with size prefs set; I for one have set pic size but not date prefs. Judging from the number of comments and protests about overlapping etc pics that one gets, which must come from those either with tiny screens and/or big pref sizes set, I think the number must be way over 100. Looking at the thread you linked to, I cannot agree with your characteristization at all - in fact I see I commented myself, mostly favourably, right at the end, and just before going on holiday, but I had no idea the proposal would lead to A4 sized images I can't even fit on a single screen. The debate was very diffuse, involving I think fewer than 10 editors in total, and it is very hard to see who is agreeing with what. I know that is the MoS way of doing things, but I don't think it is enough for a change of this nature. I doubt most editors are aware of the change yet. Johnbod (talk) 16:16, 17 September 2009 (UTC)

I agree that Gregorian mission's images are too large; this is true even in the default 180px resolution, and the problem becomes worse at a 300px default, even with font size grown to 20 point (a reasonable value for the mildly visually impaired). In centered images it's OK to use large values of |upright=, such as the |upright=2.2 used in Gregorian mission, but 2.2 is ordinarily too much for floating thumbnails. Perhaps it would help if we changed the "2.2" to a "1.4" in the MoS example? or adjusted some of the other text? Eubulides (talk) 16:51, 17 September 2009 (UTC)

At 180px they look fine to me, except perhaps for the last one, which is a bit big. There aren't very many of them. Some adjustment of the text would help, I'm sure, but the basic problem remains if you have large prefs, & it will be years before enough articles go over to the new way to make it worthwhile reducing preferences. And what happens if/when the default is increased from 180px? Every comment I see seems to support this, but it would mean any images adjusted to the new way would be likely to need resetting, surely? Frankly I don't think this has been thought through. Johnbod (talk) 18:03, 17 September 2009 (UTC)

• They're too large even for the default reader. For example, currently File:Gregory I - Antiphonary of Hartker of Sankt Gallen.jpg is |upright=1.8, or 320px for the default reader. On a low-end monitor that the 180px default is designed for, pixels are about 0.3 mm apart, which makes the image currently about 9.6 cm wide. But the image is too simple to justify such a large size. If you take a look at how a good book does the same thing (say, page 194 of Spielvogel 2008, ISBN 049550288X, a lavishly illustrated large-format textbook, pretty much the extreme case for how big the image should be in a real encyclopedia), the image is only about 7.5 cm wide. To match Spielvogel on the default low-end monitor the image should be shrunk to at most |upright=1.4 (250px default), and I'd suggest even a bit less.
• Regardless of whether one agrees with the previous bullet, there's a more general problem. An article with image sizes close to the limit of acceptability (and Gregorian mission clearly is that) is more likely to adversely affect users with unusual needs or preferences. Perhaps we should add something like that to the MoS advice?

Eubulides (talk) 18:36, 17 September 2009 (UTC)

(1) Eubulides, our readers come first, not the tiny fraction of WPians who set their thumbnail-size preference (I'd say about 100 active WPians, compared with tens of millions of readers every day). (2) There is nothing magical about 300px. (3) 450px may well crowd out the adjacent text. (4) Every image inserted into our articles needs thought as to size: how much detail, how important, how it impacts on text and surrounding images. This one-size-fits-all default needs to be discouraged, unless 180px just happens to be right in a particular situation; in my experience, that is very rare. Tony (talk) 06:04, 17 September 2009 (UTC)

I agree with pretty much everything in the previous comment (except for the "100"; any way to find out what the number really is?). However, nothing in the comment argues for px or against |upright=; both are methods for avoiding the one-size-fits-all default, and the two methods are equivalent for the vast majority of online readers who don't log in or don't change their defaults. Eubulides (talk) 14:37, 17 September 2009 (UTC)

Here's a wild idea: Can't we just bump up the default logged-out image value to at least 200px? I mean, it's been 180 as long as I can remember (since 2006ish?) and I'm sure we can justify increasing it slightly. Der Wohltemperierte Fuchs ^(talk) 14:47, 17 September 2009 (UTC)

David, that's an excellent idea, and would at least lessen the squint factor for our readers in all of the images that haven't been sized by editors (220 would be better, but it might be harder to convince nervous developers: worth a try, though). But it doesn't release us from the need to individually assess each image in this respect. I do it for every article I gnome. The waste, the waste, in popular music articles: I've seen remarkable free images that we can use without guilt, and they are so small they may as well not be there. Eubilides, there is a way of determining the number of preferenced users: we did it for date preferences and found a remarkably low proportion of currently active users who had preferenced date format. I can't recall who did it. I'll ask around. Tony (talk) 15:29, 17 September 2009 (UTC)

Yes, this keeps coming up & gaining support; it should really be pushed for. Johnbod (talk) 16:30, 17 September 2009 (UTC)

Window / content width

An earlier sub-section proposed that layouts (incl images) shoudl work equally well in "browser windows as narrow as 800 pixels and as wide as 2000 pixels", which IMO is unrealistic. My widescreen LCD monitor is 1440px wide, while I ran my previous 17" CRT at 800px (it was as dim as my eyesight). At 1440px a text para / section is only about half as tall as at 800px, and this causes major problems with the height of images - often necessitating gaps ({{clear}}) added at ends of sections to prevent images from overspilling into the next section. Layout problems become even more difficult if we cater for 600px, as in some notebooks.

Here's a wild idea - reduce the max width of the content of WP pages, including the side menu, to 800px. Before you ROFL, try considering the advantages:

• I've read several times that very long text lines are harder to read. My own experience is that widescreens are great for tabular apps (email, spreadsheets, IDEs) but no advantage for text-based apps (web, WP). I'll hunt down some refs if we get serious discussion about this idea.
• Reducing the range of widths will facilitate layout of images, tables, etc. - leaving editors more time to focus on content.
• Fewer ugly gaps ({{clear}}) added at ends of sections.

Fortunately there's an easy way for editors to set their browser windows to the right width to view how their content looks at 800px or lower. Use any old image editor (even MS Paint) to create an image of the required width, then use the brower's Restore function (in the control menu, click the browser's icon) to un-maximise the window, then drag it so it's the same width as the image width. Then you can scrap the image is it's just a ruler.

Those who find this too funny for words, please drop a few coins in my hat on the way out. --Philcha (talk) 06:23, 17 September 2009 (UTC)

It would certainly make it much easier to consistently format our articles, but I don't think this is ever going to happen :) Der Wohltemperierte Fuchs ^(talk) 14:47, 17 September 2009 (UTC)

It's not as wild an idea as it sounds. Very long text lines are hard to read; but it's not because of the pixels, it's because of the number of characters. If we could put into the stylesheets a mechanism whereby lines contain no more than (say) 90 characters, and text would be multicolumn in wider windows, that could be a win. It would mean that Wikipedia articles would be displayed multicolumn in very-wide browsers, but that would improve readability considerably there. See, for example, Ling & Schaik 2007 (doi:10.1016/j.displa.2007.04.003). Perhaps I should set up my hat besides yours? Eubulides (talk) 14:55, 17 September 2009 (UTC)

The 2-col idea is interesting. On my 1440px monitor the left menu looks about 160px. So, allowing for margins, that would leave 2 text cols of about 600px. Issues to be considered:

• Window widths < 1440 but > 800. Ideally Javascript should detect widow width and switch between 2-col and 1-col, e.g. 2-col when maximised and 1-col when restored.
• Would there be 1 set of cols for the whole article or one for each (sub-)section?
• How to handle big graphs and tables in a 2-col layout of about 600pc per col. I've seen some panoramic photos, and in zoology and paleontology articles cladograms can be pretty wide. Assuming that the alignment of a File is output as a containing DIV tag with a float value in its style, in theory Javascript can change the float value from left / right to none if appropriate.This where we'd need input from WP's techies, in termso f the block structure of an article, the CSS, and browser comnpatibility - for example last time I looked (years ago) Opera was well behind the others in ability to manipulate CSS styles.

Of course this kind of trickery would not make layout any easier for editors. --Philcha (talk) 18:48, 17 September 2009 (UTC)

WP:ACCESS

I have been on various computers and readers that have problems with left aligned images under level three (or further) subheadings. It is not a preference concern but an actual concern about major errors that develop. Ottava Rima (talk) 19:31, 17 September 2009 (UTC)

Can you please give an example? Which article is it, what browser and display settings were you using, and what did the result look like? If it wasn't the issue discussed (with a semi-graphical example) in #Question above, then what was it? Eubulides (talk) 20:12, 17 September 2009 (UTC)

Determining which sources need access dates

I'm confused about the convention in applying access dates to references. I've been told they should be used in {{cite journal}} references if the journal is wholly online. However, I notice in the References section for Electron -- a top importance article which passed FAC last month -- that access dates are also applied for journals that are very much available in print, like Science. Examining references to Science in the Electron article doesn't help me much: access dates are lacking for most old articles and present for newer ones, but the underlying rule remains unclear. If someone could shed some light on the guidelines for which types of journal citations warrant access dates, I would appreciate it very much. (For some background information, see the discussion from which this query stems here.) Emw2012 (talk) 20:27, 13 September 2009 (UTC)

The rule of thumb I have always held to is that an accessdate should be used for any resource which can be expected to change over time, and where the changing versions cannot be otherwise identified. So, while different versions of a single book are differentiated by oclc identifiers, different versions of a webpage are not; the latter requires an accessdate, the former does not.  Skomorokh  20:31, 13 September 2009 (UTC)

That's what I've understood as the purpose of access dates. In that logic, it seems adding access dates to {{cite journal}} references is unnecessary: to my knowledge academic journal articles do not change after publication. Unfortunately your answer still leaves me wondering why access dates in FA's are being used as I described in my previous post. I'm current bringing Homologous recombination through a GA review and concerns around this issue -- on whether journal citations warrant access dates (and if so, which journals) -- remain unresolved. Emw2012 (talk) 23:16, 13 September 2009 (UTC)

My $0.02, if the source is an article from a print journal, no access date is needed, even if it is read in an on-line library, as the print version is the master source, which doesn't change once published. In other words, the print version is the cited source, which doesn't depend on what date it was read. oknazevad (talk) 05:15, 14 September 2009 (UTC)

Some online versions differ from the print version. This can be the case with newspapers particularly, where there are different editions and updates. Ty 01:21, 17 September 2009 (UTC)

For {{citenews}} and {{citeweb}}, I can understand and agree with that, but the original question was about {{citejournal}}, particularly online versions of print journals. Since scientific journals are subject to peer review, to change anything in the online version (other than typoo fixes) would be considered dishonest, so they are far more faithful at reproducing the print versions. Either way, due to the peer reviewed nature of a (print) scientific journal, citations should be written crediting the print version, not an online library. oknazevad (talk) 02:35, 17 September 2009 (UTC)

How about a peer-reviewed journal that does not put out a print version? Whereas access dates are used for mutable sources and even wholly online journals (e.g. PLoS Biology, PLoS Genetics and the other PLoS journals) are not changed after their articles are pushed online, wouldn't access dates be unnecessary here? Also, with regard to referencing the print version of a journal article and not the online library (i.e., version?) -- how would this be differ from the current practice of filling out references? Would it imply not providing URLs to the article: no DOIs, no PMIDs, no PMCIDs? Emw2012 (talk) 17:26, 20 September 2009 (UTC)

There's typically no need for access dates for archival journals like PLoS Biology, as their URLs are stable and the articles do not change once published. This is independent of whether these journals publish print versions. One rule of thumb is that if a source has a DOI, it's not likely to change. The rule isn't infallible (some journals move URLs around; and some archival journals lack DOIs) but it's a reasonable first cut. Eubulides (talk) 18:02, 20 September 2009 (UTC)

I've always wondered why we require access dates for anything. If we source something to an unstable website, and the supporting material disappears, and then someone challenges the material saying it's unsourced, a new source will have to be found. It's not clear to me how having the access date of the old source helps. SlimVirgin ^{talk|contribs} 01:05, 17 September 2009 (UTC)

To be blunt, accessdates are essential to unstable website sourcing because of the wayback machine.  Skomorokh  01:07, 17 September 2009 (UTC)

Yes, it can be very helpful knowing which date to start searching through the Internet Archive, when the same URL can sometimes have a few years worth of pages stored. Ty 01:23, 17 September 2009 (UTC)

It has never taken me more than a few seconds to find a version on the Internet Archive that supports the material, assuming the URL has been archived at all. Also, if you're going to use as a source a website version that someone decided to remove (i.e. they not only took the website down, they removed that particular material before doing so), it really wouldn't be a good idea to rely on it as a source.

Can anyone give an example of an access date having been essential, or even just helpful, in tracking down a reliable source that someone has used? SlimVirgin ^{talk|contribs} 21:17, 20 September 2009 (UTC)

(outdent) I agree with Eubulides that scholarly journals don't generally need an access date. In addition, on medical articles, we tend to only supply the URL when the full article text is freely available without subscription or extortion but I don't know what the convention is on other sciences. Therefore, we have have the position where an editor read the material online (as most do these days) but does not provide a link. I don't know of a case that answers SlimVirgin's point but note that other citation rules, such as the Uniform Requirements (Vancouver style)[3] require "updated 2001 Aug 23; cited 2002 Aug 12" both date and access-date attributes. Colin°^Talk 22:00, 20 September 2009 (UTC)

# in British English

I haven't found any discussion of this so far on American and British English differences or Wikipedia:Manual of Style so wanted to bring up this topic.

At Number sign#Usage in the UK and Talk:Number sign#Use_in_UK.3F we've determined that there's aren't any reliable sources for the use of # as "number" in British English. While investigating this, we found a number of Wikipedia pages to do with British musicians that refer to "reaching #1 in the UK album charts" etc. Should these be replaced by "No. 1" and a note made in the Manual of Style? The only source we've found says # is used very rarely in British English [4] but the # usage does seem very common on Wikipedia even though we can't find a relaible source

Examples are:

• Calvin Harris
• McFly
• Leon Jackson

Alexd (talk) 21:17, 13 September 2009 (UTC)

Hmm, really I wouldn't mind if we deprecated this usage for all articles, regardless of dialect. It strikes me as insufficiently formal for an encyclopedia. I say this as a speaker of American English who certainly has no trouble interpreting the number sign.

Not sure about the No. usage either. Maybe write out number, or else reword. --Trovatore (talk) 21:21, 13 September 2009 (UTC)

I didn't understand "#" when I came over to the States in 1961 at age 11. I can't see it being appropriate (outside quotations) in the body text, but there might be tables where compression might make its use expeditious, although in that case there should be some explanation for non-Americans. I can't see much use for "No." in body text either (unless it's in a quotation or part of a name such as Chanel No. 5), but there are places where it's the clearest and most efficient notation in tables, Info Boxes, etc. And remember that widespread though "No." is, many readers are non-English-speakers who sometimes would use a different abbreviation (that would confuse Anglophones), such as "Nr." —— Shakescene (talk) 05:29, 14 September 2009 (UTC)

Regardless of whether "#" or "No." would be too informal, I concur that ENGVAR would apply here. Use the encyclopedic term most common and appropriate in British English. Darkfrog24 (talk) 05:32, 14 September 2009 (UTC)

I've looked at the addition. Thank the blessed lord we can now get rid of those hedgehog #s that litter popular music articles. The appropriate WikiProjects need to be informed. However, I'm not sure I like "number 1" without the "N", I guess because I'm not used to it. It is standard? Also, I'm unsure that "No. 1" should be outlawed from running prose.

Another issue is that I think we should allow the abbreviated forms in tables and infoboxes (not the very informal № 1, but at least No. 1 and #1). Tony (talk) 12:05, 14 September 2009 (UTC)

If "#" could cause confusion, then there's really no need to keep it, as "No." is widely enough recognized in most forms of English as to be suitable for all.

I also don't think we need to get rid of "No." from running text, though, as it is an abrreviation most similar to "Mr." or "Ms.", which are rarely spelled out, even in formal writing.

And as a note to Shakescene, I don't think we need to worry about what abbreviations are used by speakers of other languages, as this is the English Wikipedia. I understand not wanting to confuse non-native speakers who might be used to other abbreviations, but I don't believe it unreasonable to expect someone to learn the standard abbreviations as part of learning the language.oknazevad (talk) 19:21, 14 September 2009 (UTC)

Agreed. We should not deliberately make things difficult for non-native English speakers, but using ordinary English abbreviations does not constitute making things difficult. Darkfrog24 (talk) 21:14, 14 September 2009 (UTC)

• I've amended the guidance to allow No. based on the above discussion. Now, can you guys work out if, since we can't use #, we should use number or issue when discussing comics and magazines? And should the terms be capitalised or not? Hiding T 13:46, 23 September 2009 (UTC)

• Actually, I think we can resolve the issue and number problem looking at a number of featured articles, issue seems to be fine, issue number can work, and I doubt there's a problem with issue No., so we're good. Hiding T 13:57, 23 September 2009 (UTC)

Religious movements and capitalization

I am trying to bring some uniformity to articles on religion. I need a consensus on capitalzation of religious movements. Most of the articles on religion go crazy on capitalization, probably out of respect for the particular sect. We have agreed that Pentecostal and its adjectives should be capitalized the same as Methodist. But there are many movements-oneness, higher life, evangelical, fundamental, holiness, etc. The manual discourages rampant capitalization. What shall I do with all these religious movements? We may run out of capital letters! R/T-รัก-ไทย (talk) 01:05, 14 September 2009 (UTC)

This topic did come up recently, although I can't remember the location or the outcome, if any. Sometimes a word like reformed or evangelical or orthodox is descriptive or qualifying and should not be capitalized (in my strictly personal opinion) to avoid ambiguity ("X's views were strictly orthodox in this regard" or "the Maryknolls are an evangelical order" or "the reformed Anglican diocese proved far more effective"), since the capitalized Evangelical often distinguishes Lutheran churches, the capitalized Reformed often refers to Calvinist, Zwinglian, Congregational or Presbyterian churches, and Orthodox is contrasted from Roman Catholic (and Protestant) among Christians, and from Conservative, Reformed and Reconstructionist Judaism among Jews. —— Shakescene (talk) 05:19, 14 September 2009 (UTC)

Just as a note, though, while it is true that (capital E) Evangelical = Lutheran historically, and remains such in European contexts, in American usage, the distinction is no longer common, as many churches (especially non-affiliated ones) use it as part of their names even if they have no specific Lutheran background. oknazevad (talk) 17:51, 16 September 2009 (UTC)

I'd have to agree, just as the "Episcopal" in African Methodist Episcopal (A.M.E.) and African Methodist Episcopal Zion (A.M.E.Z.) is part of the name, indicating that these churches have bishops, not that they're an integral part of the Anglican Communion or the Episcopalian Church. If Evangelical or Orthodox or Episcopal or Reformed is part of the church's or congregation's title, capitalize it. If it's descriptive, it's usually better not to ("the A.M.E.Z. Church has an episcopal structure.") But of course the more difficult question that was posed is how one treats descriptions of a church's adherents or activities as opposed to the institution itself. —— Shakescene (talk) 06:18, 17 September 2009 (UTC)

Typographic-style punctuation marks

The issue of typographic quotation marks has come up a number of times on this talk, and I looked through the archives but didn't find much clarification... A while ago, I came upon an editor who prefers typographic punctuation marks such as ’, “, and ”. He did not take kindly to my recommendation of following wp:punct, or my later quoting of wp:style:

"The Arbitration Committee has ruled that editors should not change an article from one guideline-defined style to another without a substantial reason unrelated to mere choice of style, and that revert-warring over optional styles is unacceptable."

His exact response was: "You should learn to understand the difference between 'forbidden' and 'not recommended'. As long as “ and ” are not forbidden, I’ll continue to use them." And, indeed he has, including continuing to make edits solely to modify articles to utilize his preferred style. To what extent is the preference to avoid typographic-style punctuation marks enforced, if it is enforced at all? user:J aka justen (talk) 10:16, 14 September 2009 (UTC)

You are right and this other editor is wrong. Where is this? Tony (talk) 12:06, 14 September 2009 (UTC)

Concur. If this were the difference between spaced en dashes and unspaced em dashes, then there would be no reason to change it; here, the MOS explicitly prefers one over the other, and IDONTLIKEIT is not a good reason to ignore the guideline. Dabomb87 (talk) 12:43, 14 September 2009 (UTC)

Originally at Haiku (operating system). More recently at Akonadi, KDE 4, Arch Linux, and, I imagine, elsewhere. The "discussion" with the user, if you can call it that, took place here. user:J aka justen (talk) 13:17, 14 September 2009 (UTC)

You’re kidding, right? Mere two users decide how I should write my articles without even involving me? Yeah, right.... The “Insert” drop-down menu still includes typographic quotation marks, so they are obviously meant to be used. Neither of you two seems to be an admin. So on which authority do you think you have over me? Remove the quotation marks from the “Insert” drop-down menu and I shut up. As long as they are there, a real admin wants users to use them. --KAMiKAZOW (talk) 14:23, 15 September 2009 (UTC)

The three folks who have discussed the issue here didn't decide how you should write your articles; the countless editors who help develop the style guide determine what preferences should be applied across this project. The Arbitration Committee for this Wikipedia has decided, as well, that those style guidelines shouldn't be ignored by any one editor based on their own preferences. Features that are or are not enabled in the MediaWiki software, such as including the typographic marks in the insert section, do not trump consensus. user:J aka justen (talk) 14:57, 15 September 2009 (UTC)

KAMiKAZOW: You don't write your articles. You contribute to building an encyclopedia, subject to consensus and governed by policies and guidelines. See WP:OWN. —Finell (Talk) 15:53, 15 September 2009 (UTC)

Two (or three) people are not a consensus (especially if you discuss matters about me over my head) and PUNCT clearly states “not recommended” as opposed to “forbidden”. If in the future a real democratic decision is made, I’ll happily follow it. I don’t follow a handful random people who try to impose their taste on other people.

MediWiki is FOSS. Some admin should remove the “Insert” drop-down menu if it opposes some alleged consensus. --KAMiKAZOW (talk) 14:36, 16 September 2009 (UTC)

The style guidelines were developed by consensus. The Arbitration Committee has indicated that ignoring that consensus based on your own preference is unacceptable. The three editors attempting to convince you of this here did not singularly develop that consensus, we're simply three people who are attempting to convey it to you at this point in time. user:J aka justen (talk) 15:09, 16 September 2009 (UTC)

Changing punctuation back and forth is a Bad Thing™. I agree that curly quotes would be better, but policy is policy. I will continue to advocate for a change in the policy, but I will obey it in the meantime… —Wulf (talk) 05:20, 17 September 2009 (UTC)

A recommendation is still not a ban. Using `` and ´´ is banned. Using “ and ” is not. --KAMiKAZOW (talk) 15:33, 17 September 2009 (UTC)

Linking units on first occurrence

Why, in a scientific topic, must "kilometre" always be linked on first occurence? Tony (talk) 09:18, 15 September 2009 (UTC)

Because, if you do not give a conversion to miles, readers who don't know how long a kilometre is (and there are hundreds of millions of them) will have to look the conversion factor up. And converting all metric measures to US customary or imperial in a scientific article is as pointless as converting all yards to metres in articles about American football. (For everyday measures such as the kilometre, adding a footnote stating that "1 km ≈ 0.621 mi" would be a valid alternative, but for more technical units such as the pascal, I'd always link the unit rather than (or as well as) giving a conversion to pounds-force per squared inch.) --___A. di M. 09:28, 15 September 2009 (UTC)

Wikipedia (not a reliable source, I know) tells us that Use of the mile as a unit of measurement is largely confined to the United States and the United Kingdom where it remains customary. I believe that every Brit who is not mentally impaired and is between infancy and dotage is acquainted with kilometres, with the exception of those whose poor command of English (probably not their fault) or extreme sloth or bloody-minded incuriosity would anyway render them unlikely to want to read up on scientific subjects. As for the Youessians, they have land borders with two nations that use kilometres. My reading of Jonathan Kozol tells me that knowledge is distributed very unevenly across the US, but I'm surprised to hear that there are hundreds of millions who don't know what a kilometre is. Are you sure of this? -- Hoary (talk) 10:42, 15 September 2009 (UTC)

I'm curious as to how a UK or US reader who has trouble imagining how long a km is would be assisted by kilometre, which opens:

The kilometre (American spelling: kilometer), symbol km is a unit of length in the metric system, equal to one thousand metres and is therefore exactly equal to the distance travelled by light in free space in 1⁄ 299,792.458 of a second.

It is the conventionally used measurement unit for expressing distances between geographical places in countries which use the metric system. While it is defined exactly as 1000 m, it equals roughly a ten minutes' walk [excuse the grammar].

Slang terms for kilometre include click (sometimes spelled klick or klik) and kay (or k).

Then it goes on about pronunciation.

Ten minutes' walk is the closest we get, but how far would a seven-year-old schoolgirl or an 80-year-old man who has consulted the scientific article walk in that time?

What I'm concerned about is the assumption that linking, and linking in every article, is regarded as useful per se. Tony (talk) 11:27, 15 September 2009 (UTC)

Just look how bold I've been! What I wonder about is the location of these hundreds of millions of people who don't know what a kilometre is and yet want to read up on science. I've a hunch that Stephen Colbert could give a convincing explanation. -- Hoary (talk) 11:43, 15 September 2009 (UTC)

Tweaked it further. (Can a native speaker of British English confirm my suspect that it uses the plural in "0.6214 miles"?) The assumption that linking in every article is regarded as useful per se is wrong, but the assumption that, when an article is bad, hiding it away by removing non-useless links to it is a valid solution isn't 100% correct, either... (BTW, in the case of "kilometre" I agree that the link isn't vital, but I think the piece of instruction we're discussing about is more intended to deal with such units as the joule, the pascal, the kelvin, and similar, when used in the discussion of experimental results; writing "It lies in a tunnel 27 kilometres (17 mi) in circumference" with a conversion and no links in the article about the LHC is perfectly reasonable, IMO.) --___A. di M. 15:02, 16 September 2009 (UTC)

Punctuation, inside or outside

The MoS advocates placing punctuation outside quotation marks. I'd like to change that, because it's advocating something that most Wikipedians don't do.

• Inside (see placement of period/full stop): John said, "I hope the period's inside, not out."
• Outside: John said, "Nope, wrong again".

The MoS currently says:

On Wikipedia, place all punctuation marks inside the quotation marks if they are part of the quoted material and outside if they are not. This practice is referred to as logical quotation. It is used by Wikipedia both because of the principle of minimal change, and also because the method is less prone to misquotation, ambiguity, and the introduction of errors in subsequent editing.

Most North American editors place punctuation inside. I don't know what Australia/New Zealand does. Some British editors place it outside, but I'm British and I've always placed it inside. Most publications place it inside, including those outside north America. Just to pull one example up at random from today, The Times of London quoting the prime minister, "Growth is the best antidote to debt." [5]

I frequently find articles I write being changed from inside to outside, especially if they're connected to the UK, but even when they're not. I feel that this violates the spirit of the MoS, which is not to go around making style changes for no good reason. But so long as the punctuation advice is in the MoS, it's bound to happen.

Instead of recommending one over the other, can we not simply describe both practices, say the article must be internally consistent, and that we should stick to the style of the first major contributor? SlimVirgin ^{talk|contribs} 16:19, 15 September 2009 (UTC)

Logical punctuation seems more, well, logical. The place where it makes a difference would be where inserting punctuation inside the qutation marks suggests an ambiguous or misleading interpretation that was not part of the original quotation. For example, if a person says "I was not hurt so much as deeply offended", we are misleading the reader if we report that the person said "I was not hurt." Inserting the period in that location gives the misleading impression that this was the end of the sentence. Conversely, suppose an artist is asked who influenced him in his life responds "There is only one -- my dear mother." But when asked whose art he emulates, mentions Klee and Giacometti. For us to say he credits as influences his "dear mother," Klee, and Giacometti incorrectly suggests that his mother was part of a list. I know these examples are a little forced but I do run into this in editing. In any case, if we do decide to leave the matter entirely up to editors' discretion I don't think we should have a "first major contributor" rule - that's too rigid. Perhaps we just say it's up to the editors, note that people should not be revising entire articles to shift style choices, and leave it up to them after that. Wikidemon (talk) 16:32, 15 September 2009 (UTC)

I'm not sure I really follow your examples, WD. How could inside/outside punctuation affect the meaning of, "I was not hurt so much as deeply offended"?

But the point here is not what we personally like or don't like, but what most people do. Most Wikipedians use inside punctuation, and so do most publications (north American and otherwise) that I am aware of. Given that it's a preference issue, I think we should recommend internal consistency, and deferring to the first major contributor. SlimVirgin ^{talk|contribs} 17:17, 15 September 2009 (UTC)

In my example, the person quoted was hurt. Without the period the statement refines and makes precise the nature of the offense taken. With the period it looks like a simple denial. Again, this may not be the best example. One would also best distinguish between problems caused by American-style punctuation from the more general issue of taking quotations out of context. My overall point is that punctuation or lack thereof sometimes changes the meaning of a statement. By blurring the distinction over whether the punctuation is that of the speaker's versos that of the the editor's, inserting or removing punctuation within a quotation runs the risk of misleading the reader and changing the meaning of the statement. It's hard to know from my personal experience which version is common. I read the logical punctuation section early in my stay here and began to follow it, and because I work on lots of stub and start articles I tend to be the only major contributor to most of them. I would think we want to look at most A / GA / featured articles for guidance, not all the new and messy ones. I agree that consistency within articles is important, although I think the "first major contributor" rule is unduly formulaic where a more flexible common sense courtesy approach might apply more generally. I do like rules where they work, though. Are there examples where "first major contributor" has been codified in any other matters of stylistic choice? Wikidemon (talk) 18:41, 15 September 2009 (UTC)

Regarding your last point, yes, where there are style conflicts, the first major contributor's approach should be respected, per WP:CITE, a couple of ArbCom rulings, and per the MoS itself (#1.2 Stability of articles). But adding a courtesy clause to the punctuation section would be fine with me too. What I would like to avoid is people turning up only to change punctuation.

Regarding how many articles use what, it would be difficult to judge by FAs, because when something is submitted for FA, it's invariably changed from inside to outside punctuation, particularly if it concerns anything British. I suppose all we can do is keep an eye open, and perhaps try some random articles.

I still don't get your "hurt" example. Sorry if I'm being dense. :) SlimVirgin ^{talk|contribs} 19:55, 15 September 2009 (UTC)

(outdent) i'm real sorry to see this issue rear its head again so soon - SlimVirgin, do you need a link to the very recent and lengthy discussion where consensus was established once again for the so-called "logical" style? i disagree completely that that style is "something that most Wikipedians don't do" - all the articles i frequent use it, and i'm all for it. Sssoul (talk) 18:00, 15 September 2009 (UTC)

Yes, please, that link would be very helpful. I don't know how we can judge for sure what most articles use. All I know is that I've made 80,000 edits over five years, and the overwhelming majority of articles I edit use inside punctuation. But then so do most publications in general, both in North America and Europe, so I'm a bit mystified as to how outside punctuation managed to get into the MoS as a rule.

That's why I'm suggesting we simply describe the different forms, and allow editors to choose for themselves. SlimVirgin ^{talk|contribs} 18:04, 15 September 2009 (UTC)

I don't see why the examples above are relevant to a proposal to change the MoS. All the examples conform to the current MoS. The first example given in this thread, which is of "inside" punctuation:

John said, "I hope the period's inside, not out."

exactly mimicks the style given in the first "correct" example in WP:LQ:

Arthur said, "The situation is deplorable and unacceptable."

Since "inside" punctuation is just fine in all the examples given in this thread, no reason has been given to change the MoS. Eubulides (talk) 20:02, 15 September 2009 (UTC)

true, the question is not "inside or outside" - it's "is the punctuation part of the quote or not". the most recent discussion of this is in Archive 111 of this page.

re "I don't know how we can judge for sure what most articles use": smile: so please don't use the argument that the style you prefer is "what most Wikipedians/people do". Sssoul (talk) 20:10, 15 September 2009 (UTC)

This is a job for ENGVAR. Articles that use American spelling should also use American punctuation and articles that use British English should use British punctuation.

The main argument in favor of logical/technical punctuation is the idea that putting commas and periods inside the quotation marks can cause confusion. However, this doesn't happen in practice. In American English, it is understood that periods and commas may be changed as part of the quotation process. What is the point of solving "problems caused by American punctuation" if it doesn't cause any problems?

American punctuation already makes exceptions for those few cases in which it might cause confusion. For example, we'd say, that the song performed by Jefferson Airplane is called "White Rabbit," but we'd say that to put a long dash on Wikipedia, type in "—". No one is actually going to think that the comma is part of the name of the song, so there is no reason to put the comma outside in any article written in American English. However, the character-by-character instructions could be misunderstood, so American English makes an exception. The system is fine.

With regard to quoting dialogue, it is understood that commas and periods may be changed in American English. Wikidemon's example is misleading not because it uses correct American punctuation but because the second half of the sentence is chopped off. Look at it this way: 1. He said, "I was not hurt." 2. "I was not hurt," he said. They are equally misleading, in both cases because the words "so much as deeply offended" are not included. (Please note that 3. "I was not hurt," he said, "so much as deeply offended," is not misleading at all, despite the fact that I have added a comma and changed a period to a comma as well.)

HOWEVER I do not feel that we should impose American punctuation on British English articles. That would be an insult to British English Wikipedians, as bad as forcing Americans to spell "center" with a "t-r-e." Darkfrog24 (talk) 20:34, 15 September 2009 (UTC)

I agree with everything you say, Darkfrog, except the last sentence. I'm British, and was educated in the UK, and I'm constantly bemused by an MoS that tells me British English doesn't say September 15, when I do, or British English doesn't write "organized," when I do, or British English doesn't use inside punctuation, when I do. These things are a matter of preference and of individual style guides that publishers use. SlimVirgin ^{talk|contribs} 20:41, 15 September 2009 (UTC)

Then whatever form of English is American and not British, that is the form that should not be imposed upon British English articles. If "organize" and tucked-in commas are acceptable in correct British English, then they should be acceptable on British English Wikipedia articles unless there is some serious reason not to. (We use double quotes instead of single because of the limitations of search features.) However, I personally would want to see confirmation of this in at least one reliable British style guide. Darkfrog24 (talk) 20:48, 15 September 2009 (UTC)

Sssoul, thanks for the link to the Archive 11 discussion. I've not read it yet, but the fact that it has to start with a caution about feelings running high suggests there's no consensus for this advice to be in the MoS. Why do we need it? What's wrong with articles simply being internally consistent? SlimVirgin ^{talk|contribs} 20:45, 15 September 2009 (UTC)

SlimVirgin, when you read the archived discussion you'll see that the emotions were on the part of two or three editors. i know you know that just because someone has strong feelings that doesn't mean they're right, or that they have consensus. Sssoul (talk) 08:16, 16 September 2009 (UTC)

It seemed to me that almost everyone who participated in that discussion had strong feelings about it. Let's not mischaracterize people. In this case, the people who had the strong feelings toward keeping American style banned were the ones who had consensus. Darkfrog24 (talk) 13:10, 16 September 2009 (UTC)

1. It's not that there's no consensus so much as that there are lots of people who like technical quotation a lot more than American quotation and believe, incorrectly in my assessment, that American quotation causes problems. This particular part of the guideline gets challenged a couple of times each year. The discussion in archive 111 resulted in the current wording but didn't result in any change to the rule itself. 2. We don't need it. Darkfrog24 (talk) 20:48, 15 September 2009 (UTC)

I think American should not be imposed on British, and British should not be imposed on American, but also that people shouldn't be telling other people what is and isn't British or American, and the preference of Mr X should not be imposed on Ms Y, and so on. :) In other words, let's go for internal consistency and otherwise not advise. If it's being challenged regularly by established editors, it really shouldn't stay as it is. SlimVirgin ^{talk|contribs} 21:06, 15 September 2009 (UTC)

NOT AGAIN, please. We just went through this a couple of months ago, at length. This is one long-stable guideline in the MOS. Some people, mostly in the U.S., who are more familiar with the "trailing periods and commas always inside" style, as I am, object when they see the MOS guideline or examples of it, which I don't. (For what it's worth, I use and prefer the punctuation inside style in all my own writing; it is the convention where I write, and it is the convention in most printed material in the U.S.) The reasons why it was adopted for Wikipedia are explained in the MOS section, and have nothing to do with U.S. versus UK (i.e., ENGVAR). Despite MANY lengthy arguments, the result is always the same: no consensus to change the guideline. So I repeat, NOT AGAIN, please. Thank you. —Finell (Talk) 21:20, 15 September 2009 (UTC)

Just because this part of the guideline has been discussed before doesn't mean that new people like SV shouldn't state their opinions and no one should badger or bully them into shutting up before they've had their say, so please no one start. SlimVirgin has provided a new look at this issue.

Finell, if you are tired of this discussion, then, if you want, I can message you on your talk page if it looks like we're about to change the guideline. That way, you won't have to bother watching the discussion and you will get your $0.02/vote/contribution to consensus in regardless. Would you like me to do this? Darkfrog24 (talk) 21:43, 15 September 2009 (UTC)

Darkfrog, please. I haven't badgered or bullied anyone, and I resent that characterization. I expressed weariness at going through this again (I've been through it a couple times before you arrived on this page, and then the last time, when you were the lead protagonist) and the wish that we wouldn't have to so. —Finell (Talk) 22:05, 15 September 2009 (UTC)

I am not accusing you, Finell. You will see that I said, "...before they've had their say, so please no one start." There was a lot of that sort of thing last time and I want to head it off. On another note, I can appreciate that you're tired of talking about it, but the bottom line is that you do not have to. There is no reason to tell SlimVirgin not to start a new discussion because you are under no obligation to participate if you don't want to. SV's comments put no burden of any kind on you. My offer is serious, by the way. Just message me if you want to take me up on it, but I won't bring it up again. Darkfrog24 (talk) 22:21, 15 September 2009 (UTC)

If anything, the fact that this part of the guideline is questioned so often should be a big clue that it may not be ideal.

Either the current system is British, in which case it's inappropriate for use in American English articles or it is logical/technical, in which case it's unnecessary throughout almost all of Wikipedia. Either way, it should not be imposed upon American English articles in which it is incorrect and unnecessary.

SlimVirgin, I wouldn't go that far. Even if 300,000 people make a mistake, it's still a mistake. We shouldn't force one group of people to forego their own traditions and use a foreign set of rules, but neither should we allow Mssrs. X and Y to make things up willy-nilly. It's not about what most people are doing. It's about what's correct. Darkfrog24 (talk) 21:43, 15 September 2009 (UTC)

(ec) Hang on, Finell, if it keeps being raised, there is no consensus for it. It isn't explained at all in the MoS, which says, "[Logical quotation] ... is used by Wikipedia both because of the principle of minimal change, and also because the method is less prone to misquotation, ambiguity, and the introduction of errors in subsequent editing."

1. What is "the principle of minimal change"?

2. There is no ambiguity, or misquotation issue. If you think there is, can you give a clear example?

3. What is meant by "the introduction of errors in subsequent editing"?

Finell, what do you see as the advantage of recommending this, rather than recommending internal consistency and leaving it at that? Because that would finally put the issue to rest, which is what you seem to want. SlimVirgin ^{talk|contribs} 21:46, 15 September 2009 (UTC)

(interjecting, after edit conflict, since SlimVirgin asked me specifically)

I'll try to answer your questions briefly. However, if you want to see every possible argument that can possbily be uttered on the subject, please consult the archives of this talk page (and I don't just mean the last go-round).

1. From Wikipedia:Manual of Style#Quotations, under the sub-heading Minimal change: "Preserve the original text, spelling, and punctuation." Suppose a Wikipedia page legitmately quotes part of that in the following passage: The MOS succinctly prohibits altering the spelling of quoted text: "Preserve the original text [and] spelling". The period follows the closing quotation mark because it is not part of the original text. The so-called typographical or American style would render the same passage as follows: The MOS succinctly prohibits altering the spelling of quoted text: "Preserve the original text [and] spelling." The period, although not part of the quoted material, is placed inside the closing quotation mark because that is more aesthetically pleasing (the aesthetic difference is more pronounced in formally printed work, using a seriffed, proportional typeface and so-called typographical quotation marks). But, this version of the page violates the principle that it is stating: it inserts a punction mark that was not part of the quoted text.

2. Suppose source material reads: The judge demanded, "Bring the prisoner to the bench, and button your jacket before you address the court. The prosecutor silently, very slowly, fastened the four buttons of his jacket, one by one. Then he escorted the prisoner, handcuffed, to the bench, the judge towering over both of them. A Wikipedia article says, following the MOS guideline: The judge said, "Bring the prisoner to the bench". Prosecutor Stone complied. The reader, and also a subsequent editor, knows that the quotation is not a complete sentence because the period is outside. Using the other system, the article reads: The judge said, "Bring the prisoner to the bench." Prosecutor Stone complied. Is the quotation a sentence or a fragment?

3. If Wikipedia switched to the other system, a subsequent editor might erroneously revise this material as follows: The climax of the courtroom drama begins with this sentence: "Bring the prisoner to the bench." The prosecutor complies, but the judge's scolding, uninterrupted, fill the next eight pages. (The editor added the scolding bit by combining the next sentence in the imaginary article.) The article misrepresents the fragment as a sentence.

The reasons stated in the guideline persuaded me, contrary to my aesthetic prevference and the punctuation that I learned, that the guideline is the better choice for Wikipedia. In my opinion, the underlying objection of most objecting editors really reduces to WP:IDONTLIKEIT, most likely because WP:THATISNTHOWILERNEDIT. I disagree with the proposition that, "if it keeps being raised, there is no consensus for it". Consensus does not imply unanimity. Some Wikipedians (and I sincerely do not not include SlimVirgin in this category) will argue about anything. —Finell (Talk) 00:46, 16 September 2009 (UTC)

One could just as easily say that WP:IDONTLIKEIT is how this policy got here in the first place. The question of not being able to tell whether a quotation is only part of a sentence is not a real problem. That is why almost every academic discipline in the U.S., no matter how strict, uses U.S. punctuation. Again, the problem in the examples provided is that the information itself is left out, not that the punctuation causes confusion. Darkfrog24 (talk) 01:08, 16 September 2009 (UTC)

Then I did not explain properly. This does count as consensus even though not everyone agrees. Whenever a Wikipedia guideline is to be changed, there must be consensus for the change. It's sort of like the burden of proof being on the prosecution. If the two sides can't agree, then the change doesn't happen. It is more a case of people who want to include American punctuation being outnumbered. The burden of forming a new consensus is on the people who want to make the change.

1. The principle of minimal change is the idea that, in direct quotations, the text from the source should be changed as little as possible. Some Wikipedians believe that American punctuation violates this, but it does not.

2. Correct.

3. Some people believe that if one editor uses American punctuation, and then six months later another editor rewords the passage, that this will introduce errors. However, this is a risk regardless of which style of punctuation is used. The editor must be conscientious and look at the original text. Period.

The bottom line is that in order to know exactly how the original text was punctuated, one must look at the original text. This is just as true of British and logical/technical styles as it is of American style. Darkfrog24 (talk) 22:01, 15 September 2009 (UTC)

My main concern

Just to clarify, I don't mind how most articles tend to be punctuated. It's not a big deal. I do mind editors arriving at articles I'm writing, and changing the punctuation, which means I have to try to remember to write differently, or else the article will end up inconsistent. Or I have to change it all back again. I think that's discourteous, and it violates CITE and the MoS itself. So I am proposing one of two things:

1. We change that part of the MoS to describe the two different punctuation methods, and let people choose which to use, without the MoS recommending one or the other (this would be my preference); or

2. We add a note to that section reminding people not to change the style in stable articles i.e. a courtesy provision.

I don't mind which we do. But I would like to see an end to people imposing this on articles they're otherwise not involved in writing. SlimVirgin ^{talk|contribs} 21:56, 15 September 2009 (UTC)

See, most of us do think that punctuation is a big deal. I support 1. with the exception that I would have the punctuation tied to the spelling style. It doesn't matter who's involved and who isn't. It's about what's correct. If I'm reading an article out of the blue and I see a typo, then it's perfectly all right for me to jump in and fix it. Darkfrog24 (talk) 22:01, 15 September 2009 (UTC)

Yes, if it's a typo, but there is no right or wrong here, just preference. I don't like to have spent months writing an article, and have someone arrive for three minutes to change my punctuation. It's very annoying. :) Especially because it means I have to keep remembering to write in the same way from that point on, and I never can remember.

I really would not want to see this tied to spelling style. That reinforces, "This is British and therefore must be written this way," even if you, as a British editor, have never done it. I am tired of feeling disenfranchised by that absolutism. :) SlimVirgin ^{talk|contribs} 22:17, 15 September 2009 (UTC)

See, people shouldn't be making style changes for no reason regardless of whether they've been involved in the article before or not. It would have been just as wrong for that person to come in and change one acceptable style to another even if he or she had made a major contribution to that article in the past. That's what I'm talking about.

I happen to be an American, actually, just a rather stuffy one. Darkfrog24 (talk) 22:27, 15 September 2009 (UTC)

(outdent) for the record, the point about not changing styles in stable articles only applies when the MoS accepts more than one style. so unless consensus is demonstrated for adopting another punctuation style in addition to the so-called "logical style", that kind of "courtesy provision" wouldn't be appropriate. Sssoul (talk) 08:16, 16 September 2009 (UTC)

Since this issue keeps coming up, and since a consensus hasn't been reached on this page, would it be worthwhile to open a community RfC? While I understand the reasons why the guidance is there, I suspect that this is one of the most widely-ignored instructions in the MOS, and I doubt that most readers even recognize its significance. It may be useful to get more opinions. Karanacs (talk) 01:07, 17 September 2009 (UTC)

Common misconception: There was a consensus according to Wikipedia's operational definition of the term, just not a universal one. I don't know that we need an RfC on this one. Yes, this part of the MoS is frequently challenged, and yes, I feel that that should be a big fat clue, but challenging and changing things is part of the Wikipedia consensus process. New people come in with new ideas, then the issue is discussed again and then the community either forms a new consensus or keeps the old one. This particular issue may bring out strong feelings, but this is how it's supposed to work. Darkfrog24 (talk) 04:39, 17 September 2009 (UTC)

In other words, yes I think that this rule should be changed, but not because there wasn't a consensus for it in the first place. There was. Darkfrog24 (talk) 04:40, 17 September 2009 (UTC)

I see no consensus for change. Tony (talk) 04:51, 17 September 2009 (UTC)

A relevant observation (out of the previous round of discussions ad nauseum on logical quotation): Logical quotation isn't a matter of style, it's a matter of content. Regional spelling variations don't change the information content of articles, but the British and American quotation "styles" convey less information about the source than does logical quotation. That's essentially the definition of logical quotation, that it consistently uses placement of final punctuation inside or outside the quotation marks to convey information about the source, rather than subordinating conveyance of information to regional stylistic convention. --Pi zero (talk) 13:35, 17 September 2009 (UTC)

(1) Your statement would be true if Wikipedia used logical quotation consistently, but that's not the case and unlikely to become the case in the near term. When you see "This means war." on Wikipedia you get no information about the source. That might be logical quotation or US style. The only way to know is to look at the original source.

(2) Adjusting US-style quoted text to logical quotation is quite difficult and requires access to all the original sources. This seems like a large burden of work to place on copyeditors. Assuming that at the very least a large minority of contributions continue to be made using the common US method, do we have enough editors interested in doing this research to accomplish the work of fixing those articles? If not, Wikipedia is unlikely ever to adhere substantially to this aspect of the MOS and it may be more useful to adjust the MOS to reflect a perhaps less ideal, but more achieveable standard. Christopher Parham (talk) 14:17, 17 September 2009 (UTC)

Consensus in favour of logical quotation was reached long ago. Though challenged from time to time there has never been consensus to change this guideline. I doubt there ever will be. JIMp _talk·cont 14:27, 17 September 2009 (UTC)

My position on this is the same as my position on all other style issues: if an article is internally consistent, and if it uses styles recognized by mainstream style guides, and if it is stable, and if the writers of the article are happy with it, then editors shouldn't arrive at the article simply to change from one style to another citing the MoS. That is, in effect, what the MoS itself says at section 1.2: Stability of articles.

Does anyone mind if I add a very brief point about that to the section about punctuation? SlimVirgin ^{talk|contribs} 16:03, 17 September 2009 (UTC)

"Does anyone mind ... ?" yes. as noted above: section 1.2 refers to situations when there's more than one guideline-defined style. so doesn't apply to this issue unless/until consensus is demonstrated for the MoS to recommend more than one punctuation style. Sssoul (talk) 17:35, 17 September 2009 (UTC)

While Wikipedia's rules do not require that changes, even major changes (such as this would be), be discussed on the talk page ahead of time, that's probably best for this case. If you made that change, then someone would just change it back. In other words, I don't mind, but in this case it is necessary to establish a consensus for the change before that change is made. I personally believe that Wikipedia should not forbid editors to use correct styles so long as they are consistent or unless there is a very clear and unambiguous reason not to (case in point: use of single quotes can mess with search features, so Wikipedia prefers double). I would support such a change, though I would prefer to tie it to ENGVAR. Darkfrog24 (talk) 17:43, 17 September 2009 (UTC)

To be clear, nobody is forbidding users from employing US style quotes. But I agree with your general view. Modifying articles that already consistently employ a widely used and understood style is probably not a good use of time in any case. Especially in this one where such modification is likely to be unusually burdensome. Christopher Parham (talk) 19:05, 17 September 2009 (UTC)

Actually, Parham, the MoS does ban the use of American-style punctuation. Darkfrog24 (talk)

No, the MOS says that American-style punctuation does not represent Wikipedia's best style; it doesn't have the authority to "ban" anything. Quality content contributions using American-style punctuation, or otherwise ignoring MOS formatting details, are generally encouraged. (e.g. the premise of SV's comment is that she is writing an article using American style quotes and nobody in this discussion has suggested she should stop doing so.) That's part of the problem with the current wording: regardless of what MOS says, we will have many people adding content using American style punctuation, because they are accustomed to that style having grown up in an environment where it is overwhelmingly standard. It is then very difficult to convert that text to conform to the guideline. This is quite different from say, the dash rules where an article using hyphens takes seconds to reform using automated tools.Christopher Parham (talk) 22:06, 17 September 2009 (UTC)

I really wish that were how it worked, but it isn't. I got called up for an AN/I for tucking in stray commas in articles that already used American punctuation as their prevailing style. If that's not a ban, then I don't know what is. Ergo, we may call them guidelines, but they're rules. Darkfrog24 (talk) 22:36, 17 September 2009 (UTC)

Tucking in stray commas (i.e., making purely stylistic changes to existing text) is just about the opposite of the "quality content contributions" I was talking about. You were not rebuked for use of American style quotes but for altering existing formatting from conforming to non-conforming, as I understand it. I can assure you that I have written quite a few articles, including FAs, using American-style quotations exclusively, and had no issues with it. Christopher Parham (talk) 15:40, 18 September 2009 (UTC)

No Chris, I was tucking in the strays while making other changes. These changes would count as gnoming rather than content, fixing awkward phrases and the like. The point that I was trying to make, though, was that if the MoS did not constitute a ban on American punctuation, then no one would have objected to my making such changes, as they improved each article's internal consistency. Darkfrog24 (talk) 19:15, 18 September 2009 (UTC)

re: "Does anyone mind". Yes. If your position is "don't change what is stable/reasonable/not-broken" on all style issues, then why are you arguing for a specific reminder/get-out-clause for this style issue. We can't have our guideline pages littered with "use common sense", "don't be disruptive", etc. reminders next to each of the slightly contentious issues. Colin°^Talk 20:18, 17 September 2009 (UTC)

I understand that consensus on this was reach a long time ago. I question whether the consensus has changed, and whether the opinions of the editors who frequently comment here accurately reflect a Wiki-wide consensus on this issue. As Christopher Parnum pointed out, many, many editors do not use logical quotation, and it is not easy to tell when looking at an article which practice was intended. Karanacs (talk) 16:09, 17 September 2009 (UTC)

I don't frequently comment here, but I support the logical quotation system for the simple reason that it doesn't alter the quotation. That many editors are ignorant of this guideline, or that the style in-use is hard to spot, doesn't change the reasons for preferring this guideline, which is nearly as old as the MoS itself. I'm also worried about the tone of WP:OWNership about folk fiddling with "articles I'm writing". Colin°^Talk 20:18, 17 September 2009 (UTC)

That would hold weight, except American style doesn't alter the quotation either. Darkfrog24 (talk) 21:10, 17 September 2009 (UTC)

Yes it does, and that is the point. So-called typographic or American punctuation puts ending commas and periods in the quotation when the quoted text did not have them. And this has always (or at least for a very long time) been explained in the MOS as the reason for the guideline. Finell (Talk) 17:04, 18 September 2009 (UTC)

It does not alter or disrespect the quotation any more than the quotation marks themselves do. The terminal period or comma is understood to be part of the process. Darkfrog24 (talk) 19:17, 18 September 2009 (UTC)

Never understood

Re Finell: I have long wondered about this idea of "not changing the punctuation". If I take the examples in the present MOS too literally, I will go from:

Arthur said, "The situation is deplorable and unacceptable." (The period is part of the quoted text.)

Arthur said that the situation was "deplorable". (The period is not part of the quoted text.)

to

Arthur said that the situation was "unacceptable." (The period is part of the quoted text.)

But that is certainly not the intention of the MOS.

Now we all know that if I wanted to cut off the first quotation after "deplorable", I would need an ellipsis:

Arthur said, "The situation is deplorable ...".

But the principle explicated by the present MOS suggests that the ellipsis can be omitted, leading to

Arthur said, "The situation is deplorable". (The period is not part of the quoted text.)

That is clearly not what we want. So what conclusions do I draw when I reflect on best practice?

(1) If you quote a single word, you will probably ignore any puncuation in the original text anyway.

(2) If you quote a phrase from the original, you will need an ellipsis if you cut off a sentence.

In either case, there is no possibility of confusion with the American system tat is not also present in the current MOS recommendations. — Carl (CBM · talk) 18:00, 18 September 2009 (UTC)

Finell is not on trial here. Surely you mean to address the entire board so that anyone may answer. I agree that the logical/technical style has no application in the quotation of words as words. (The word "gender," unlike "sex," does not have lascivious connotations.) Darkfrog24 (talk) 19:31, 18 September 2009 (UTC)

Actually, a change in the guideline (made in connection for the last dispute, if I recall correctly) is responsible for the problem you raise. Given the contentiousness of this guideline, I decided not to fix it. I believe the guideline previously said to include the end punctuation if it is part of the sense of the quoted text (please forgive me for not taking the time to look up the diff). I construed that to mean that you do not include a period (full stop) just because it follows the last word of the quotation (e.g., if you are only quoting a word or phrase at the end of the sentence); this is the practice that I still follow under the current guideline. Further, the punctuation guideline does not override the use of ellipses (which, like square brackets, are understood to indicate something omitted or changed). If the logical quotation system is followed correctly, there are fewer opportunities for ambiguity, and therefore errors by subsequent editors, and even those can be eliminated with careful writing—such as by quoting more or less). The second example from the MOS eliminates ambiguity and the need for ellipses by paraphrasing and quoting less. And I do hope that a citation, with the page number, will follow the quotation! Finell (Talk) 19:40, 18 September 2009 (UTC)

Thing thing that I don't see is how the American system actually introduces ambiguity. For example, consider the sentence

Jones: All cars are fast, but some are faster than others.

Apparently one argument for "logical punctuation" is that

(1) According to Jones, "All cars are fast."

is bad. Yes, it is bad, but

(2) According to Jones, "All cars are fast".

is equally bad. Both require an ellipsis, because both cut off the quote in mid-sentence. I am not a frequent participant in these discussions, and so it may be that there are better examples that actually show ambiguity reduced by correct use of logical quotation over correct use of American quotation. But replacing (1) with (2) is hardly in improvement in clarity and accuracy. — Carl (CBM · talk) 21:16, 18 September 2009 (UTC)

In case someone asks, I am referring to what the Chicago Manual calls the "rigorous method"; see e.g. section 11.65. — Carl (CBM · talk) 21:31, 18 September 2009 (UTC)

Anyone? I'd like to understand this issue, since the present MOS language has always seemed like a distinction without a difference. — Carl (CBM · talk) 13:13, 19 September 2009 (UTC)

(outdent) the examples in the MoS don't make the point very well, i agree. maybe it becomes clearer when one considers that the so-called "American style" wants all end punctuation inside the quote marks, not just periods and commas, and it wants them there within titles as well as quotations:

• Who wrote "You Can't Always Get What You Want?"
  

is an example where the so-called "American style" is misleading in a way that so-called "logical style" is not:

• Who wrote "You Can't Always Get What You Want"?
  

there's not as much potential for ambiguity when periods and commas are involved, but it is more accurate not to misrepresent titles by making it look as if they end with punctuation that they don't in fact include. Sssoul (talk) 05:59, 21 September 2009 (UTC)

That is not a correct assessment of American-style punctuation. Here is an example verbatim from section 6.9 of the Chicago Manual:

Which of Shakespeare’s characters said, “All the world’s a stage”?

Indeed, the rule is:

6.9: Colons, semicolons, question marks, and exclamation points  Unlike periods and commas, these all follow closing quotation marks unless a question mark or an exclamation point belongs within the quoted matter.

So it is only with periods and commas that punctuation might be moved inside the quotation marks, unless that punctuation is part of the quoted material. — Carl (CBM · talk) 10:56, 21 September 2009 (UTC)

okay, i'm glad to hear it and i'll strike my misconception (which matches the common misconception that either "British style" or "logical style" always puts the punctuation outside the quote marks). what's left is: i agree that it's difficult to think up an ambiguous example with just periods or commas, but it is still more accurate not to misrepresent titles (or quotes) as including punctuation when they don't. Sssoul (talk) 12:10, 21 September 2009 (UTC)

I'm asking specifically about quotations from prose, since this is the justification the MOS gives for the logical quotation system. The issue of periods in the titles of artistic works is a much less significant issue than the alleged loss of accuracy in quotations. — Carl (CBM · talk) 13:29, 21 September 2009 (UTC)

smile: there are different points of view - to me the proper representation of titles is a much more significant issue. but anyway: i'm not sure what you mean by "the alleged loss of accuracy in quotations" - are you talking about my allegation above, that it's more accurate to represent titles (and quotes) without additional punctuation, or something in the MoS? Sssoul (talk) 14:21, 21 September 2009 (UTC)

The MOS itself says that logical quotation "is less prone to misquotation, ambiguity, and the introduction of errors in subsequent editing." I am asking for an actual example of that in which both the American-style quotation and the logical-style one are properly formatted, including ellipses for omitted material as required by WP:MOSQUOTE. — Carl (CBM · talk) 14:26, 21 September 2009 (UTC)

• It is not "the American system". Many publications and sites outside North America use internal. Wikipedia simply has sound reasons for being very fussy about leaving quotations be, where possible. Tony (talk) 14:00, 21 September 2009 (UTC)
  • I am asking, what are those sound reasons? There are claims about accuracy of quoted material being improved by logical quotation, but the examples in the MOS do not appear to support such claims. — Carl (CBM · talk) 14:26, 21 September 2009 (UTC)

I understand that not many publications outside the U.S. use American punctuation, but many of them use American English as well and that doesn't make it any less American English. (Hey, if I have to call Wikipedia's system "logical" for people to know what I mean, then calling the American style "American" isn't too much of a jump.) Darkfrog24 (talk) 19:08, 21 September 2009 (UTC)

(ec) here's what i reckon the MoS means by proneness to ambiguity and misquotation:

• She stated: "I have no objection to anything he said." if that's punctuated in the so-called American-style, the reader can't tell whether or not that was her full statement, whether the editor who punctuated it has read WP:MOSQUOTE, used ellipsis properly, etc.; but if it's in the so-called logical style, the reader can rest assured that that was indeed the end of her sentence.

and when i recycle that same quote in my own publication, which uses so-called logical style, the so-called American-style is more prone to misquotation because i have to guess whether or not that was the end of her statement. (smile: i know i know – obviously i should look her statement up elsewhere instead of relying on Wikipedia! but there are loads of journalists who don't bother have time for that.) Sssoul (talk) 15:12, 21 September 2009 (UTC)

Actually, in American style (e.g. "rigorous" Chicago style), you do know that it was the end of her statement, because otherwise there would have to be an ellipsis. You can't use an example where someone didn't follow American style as an example of problems with American style, because that agument applies equally well to logical style: how do I know the author followed logical style correctly, just because the MOS requires it? So we have to assume that the writer did follow the appropriate style as it requires. I am becoming more and more convinced that there is no example where correctly-written quotes actually have a problem, as I will explain below. — Carl (CBM · talk) 15:24, 21 September 2009 (UTC)

(ec) smile: somehow i had a feeling you were going to reject that example of ambiguity, CBM - but it's what i believe the MoS means anyway. (i could point out that using so-called logical style does imply that someone has at least looked at the MoS, since it's what the MoS recommends - but ... go ahead with the explanation you want to give us.) Sssoul (talk) 15:34, 21 September 2009 (UTC)

After thinking about this more and looking at the MOS, I am becoming more and more convinced there is no example involving quotations (not titles of works) where the American/logical style distinction makes a difference. Here is why:

(1) For short quotes that could not be confused with a whole sentence, I have been assured that we ignore the punctuation of the original anyway. Thus we write,

She said he was "boring".

rather than

She said he was "boring,".

(2) For longer quotes, WP:MOSQUOTE already requires an ellipsis for omitted material, as does the "rigorous style" in the Chicago Manual. Thus we already cannot write,

She said, "I am coming to the ball".

because either that quote is a whole sentence, in which logical style would say to use

She said, "I am coming to the ball."

or the quote is not a whole sentence, in which case MOSQUOTE requires

She said, "I am coming to the ball...".

This seems to only leave the issue of titles of works. Now, I don't really care one bit about which style we pick on Wikipedia, but we should be more upfront that the issue is one of style rather than somehow an issue of accuracy with quotations. — Carl (CBM · talk) 15:32, 21 September 2009 (UTC)

Can we all concede that taking a one-sentence quotation such as "I was not hurt but rather deeply offended" or "I have no objection to anything he said" and cutting it down to "I was not hurt" or "I have no objection" would constitute a misquotation regardless of which system of punctuation is used? Neither logical/technical nor British nor American punctuation would be able to turn "I was not hurt" into something that could reasonably imply the original quotation's true meaning.

So far, I haven't seen any sound reasons for banning American punctuation, only imaginary reasons. Darkfrog24 (talk) 19:08, 21 September 2009 (UTC)

Left-side images under section headings—reversion?

I thought there was consensus above, but someone has just reverted the removal of this point:

• Do not place left-aligned images directly below a subsection-level heading (=== or lower), as this sometimes disconnects the heading from the text that follows it. This can often be avoided by shifting left-aligned images down a paragraph or two.

It has been confirmed that there are no accessibility issues. Why can't it be removed? Tony (talk) 15:39, 17 September 2009 (UTC)

Doesn't it look funky, though? - Peregrine Fisher (talk) (contribs) 15:48, 17 September 2009 (UTC)

Funky is good. :) SlimVirgin ^{talk|contribs} 15:56, 17 September 2009 (UTC)

Disconnecting the header from its text is not purely an accessibility issue, nor is it the only consideration. Personally, I find it looks horrible and unprofessional and makes an article more difficult to read, particularly on lower resolutions. I reverted the removal, however, as I did not see any good consensus for removing it, as the discussion above involved only a small number of editors who, from what I gathered, were not entirely sure why this was a guideline in the first place other than presuming it is an accessibility issue. Considering how long it has been a guideline, I think it more appropriate to have a larger discussion before just removing it. -- Collectonian (talk · contribs) 16:02, 17 September 2009 (UTC)

Can you please give an example of where it looks horrible and unprofessional, in an actual Wikipedia article? (An old version of an article would be fine.) Certainly there was a consensus in #Question above, but of course we can always come up with a new consensus based on more-informed discussion. Eubulides (talk) 16:06, 17 September 2009 (UTC)

(ec) You don't object when images are placed on the left under a regular header (==), do you? That also disconnects the header from its text. SlimVirgin ^{talk|contribs} 16:07, 17 September 2009 (UTC)

Actually, I do...images directly under any header should really be right aligned, however the wording currently only notes level 2 headers (why, I do not know) -- Collectonian (talk · contribs) 16:38, 17 September 2009 (UTC)

Time to advertise at VP and Centralized Discussion? Tony (talk) 16:48, 17 September 2009 (UTC)

I think general discussion, possibly VP & centralised as Tony suggests. I don't like left-aligned images in most cases - they look like a more "artistic" layuot, but can cause difficulties in readability - indents lost on bullets and quotes; forces to look for the next line as it doesn't start at the normal location (x-coordinate). --Philcha (talk) 16:55, 17 September 2009 (UTC)

(ec) We could do that, though I think very few people would agree that all images under all headers must be right-aligned. I wonder if we should add a general caution about "one size fits all" in the images section. It might preempt a lot of these debates in future. SlimVirgin ^{talk|contribs} 16:58, 17 September 2009 (UTC)

Throw it up on cent. Who knows what the hoi polloi reallly thinks. You can take off my entry on adding non-money ads to articles, if you want (it's mostly dead). - Peregrine Fisher (talk) (contribs) 19:26, 17 September 2009 (UTC)

• To the opposite, I can confirm that there is an accessibility issue. Ottava Rima (talk) 19:32, 17 September 2009 (UTC)

• Could you please describe that issue? I suggest putting the details in #WP:ACCESS above, to avoid duplication of threads. Thanks. Eubulides (talk) 20:12, 17 September 2009 (UTC)

• It has been a while so I don't remember what page. It was just when the issue started about WP:ACCESS. I believe it was a blackberry or something similar. The image on the left caused the text to be broken up in a way I believe comes from the left alignment of text. All I remember is that words were cut in an odd manner mid-word or that the text was in a long gap from the heading. I am sure there are online readers that could be found. Ottava Rima (talk) 20:42, 17 September 2009 (UTC)

• It'd be helpful to have a precise example that we can look at. For example, it's possible that there was a bug in the Blackberry software, which has since been fixed. Do you have a Blackberry now? Can you reproduce the problem, and describe what it looks like? Eubulides (talk) 20:58, 17 September 2009 (UTC)

• My cell phone is 10 years old, so no. Anyway, I am sure there are people who use various readers that may be able to answer. Ottava Rima (talk) 21:57, 17 September 2009 (UTC)

• Well, I asked at WT:ACCESS and there seemed to be no indication that it might be a problem. Graham87 (talk · contribs), who uses a screen reader, said that he did not find any problems. See also User talk:Dabomb87/Archive 13#Postman's Park image placement. Dabomb87 (talk) 22:27, 18 September 2009 (UTC)

• If it is going to be raised centrally, the "upright multiple" issue should also be raised. Personally, if there is no technical issue I strongly favour being able to use left-aligned images for right-facing pics and for variety. Johnbod (talk) 20:09, 17 September 2009 (UTC)

• I like the ability to use left-aligned images under sub-heads when the layout design within the article is clearly improved. Seems natural and logical especially if there isn't enough text to bury it two paragraphs down...Modernist (talk) 20:51, 17 September 2009 (UTC)

I don't think that reading above anyone has brought this up, but the reason that the MoS only specifies level three subheads and below is because level two's have the horizontal line underneath them, therefor it's not as much of an issue for text breaks. Der Wohltemperierte Fuchs ^(talk) 12:03, 18 September 2009 (UTC)

Yes, that matter was discussed in the thread that resulted in removing the requirement; see #Question above. I see now that the requirement was reinstalled with no comment here, which is unfortunate: there is certainly no consensus to reinstall it. Eubulides (talk) 15:17, 18 September 2009 (UTC)

625 instances in featured articles

In my sandbox I generated a list of all 625 instances I found in featured articles of left-aligned images under 2nd-level (or lower) headings. Most of them look just fine. Here are some examples:

• Anarcho-capitalism #Nineteenth century individualist anarchism in the United States. I see no problems with this layout. There is no "breaking up" between the header and the text.
• Charles Darwin #Descent of Man, sexual selection, and botany. No problem here either; there isn't any breakup.
• Gregory of Nazianzus #Second Ecumenical Council and retirement to Arianzum. Nothing wrong here either.
• History of evolutionary thought #Christian philosophy and the great chain of being. Likewise.
• Rudolf Vrba #Broadcast of the report and the end of deportations. Likewise.
• Uranium #Military. This is just under a top-level header, which supplies any visual cues needed.

Some instances do look bad, but they're in the minority and the badness is typically due to a combination of features, not simply due to left-aligned images under 2nd-level headers. Given that they're commonly used in high-quality articles, and (when properly used) don't seem to be causing any real problems, it is mystifying why the prohibition was reinstated without comment here, despite the earlier consensus to remove it. Eubulides (talk) 22:40, 18 September 2009 (UTC)

That's a very helpful list, thank you. SlimVirgin ^{talk|contribs} 22:46, 18 September 2009 (UTC)

Could you speak more about when it is a problem? I know I've had problems with images on the left, but most look OK. I can't remember what was wrong with the ones that didn't work. - Peregrine Fisher (talk) (contribs) 02:33, 19 September 2009 (UTC)

Yeah, given all the controversy over this I was pretty amazed to see that the vast majority of those 625 examples are OK. I did find some questionable ones, though. With a wide screen and a small font, the left-aligned image at the start of Gliding #Bungee launch obtrudes into the following top-level section, causing its header to appear to the right of the image; this problem is relatively minor. The serious problems I found almost all involved sandwiched text, e.g., Uranium#The effect of pH, West Wycombe Park #Ethos, Voting system #Rated voting methods; but sandwiched text is really a different problem. Eubulides (talk) 03:40, 19 September 2009 (UTC)

I see. So it's a problem when sections are too small, or have too many images (barring some new info about screen readers or whatever). Then a rule probably isn't needed, and commons sense can be used on an individual basis. - Peregrine Fisher (talk) (contribs) 03:54, 19 September 2009 (UTC)

Yes, that's it. So I take it that it's OK with you to remove the unnecessary rule by undoing your recent change? Eubulides (talk) 19:43, 21 September 2009 (UTC)

So in other words we say to avoid breaking the alignment of headers and stacking images, proof the alignment on a variety of screen sizes if possible. Pretty simple. Der Wohltemperierte Fuchs ^(talk) 13:21, 19 September 2009 (UTC)

• As a matter of longer-term strategy, is WP really going to try to serve miniaturised portable platforms as well as the standard monitor sizes on desktop and mobile computers? It seems that it will never be possible to please both, and that if someone wants to read a WP article on a relatively tiny mobile phone or Blackberry screen, it is they who should expect to make compromises, not those who user monitors on a scale that is more conducive to the display of text and images in an encyclopedic register. Tony (talk) 13:34, 21 September 2009 (UTC)
  • Tiny matter screens such as Blackberries are typically fed via software that ignores all our size suggestions, so we shouldn't worry about them when specifying sizes. Netbooks are a different matter: size hints are generally obeyed on netbooks. Some of them are as narrow as 800px but generally they're 1024×600 or larger; these screens should be catered to. Eubulides (talk) 14:51, 21 September 2009 (UTC)
  • Furthermore, you can have a 800-pixels-wide window in a 2048-pixels-wide screen, but not vice versa; therefore, a layout looking gawky in an 800-pixel window but fine in a 2048-pixel one is a more serious problem than one looking fine on 800 px and gawky on 2048 px. --___A. di M. 15:42, 21 September 2009 (UTC)
    • All the above implies we should just concentrate on being ok on screens of 800 & up (maybe we do already), for which a 180 default is unnecessarily small. Let's push on this. Johnbod (talk) 21:04, 21 September 2009 (UTC)

Proposed addition to MOS:HEAD

Hey all, I'm dealing with a Mediation Cabal case where a user who is the primary contributor to an article flat-out refuses to have sections in the article, and its not just a simple WP:OWN case. So, I was looking through the policies and MOS for something saying articles should have sections, and to my surprise I couldn't find anything, despite the huge convention (All GAs and FAs have them, and pretty much anything above a Start class). There's such a vast precedent for it that its a de facto guideline, so I propose the following text to go at the top of MOS:HEAD:

• Articles longer than a stub should be broken into sections by sub-topic for readability.

I think that with this line, we can formalize what has been the convention on Wikipedia for years. As with the rest of the MOS, there are exceptions, which is why this is a guideline and not a policy. The Wordsmith_{(formerly known as Firestorm)}^Communicate 03:30, 19 September 2009 (UTC)

The article is Eurymedon vase, I'm guessing. Firstly, the use of headers really cannot and should not be proscribed because, unlike dashes and non-breaking spaces, there's no hard-and-fast, mechanical rule that be applied here, nor is there an established rule as far as I know (although some WikiProjects have drawn up rough organizational guidelines for certain types of article, such as biographies). Articles' contents and structures vary widely by topic, and even within similar subjects (compare today's FA Economy of the Han Dynasty to Economy of the United States). A similar issue arose at WP:LEAD about the length of leads; some editors wanted to institute a character-count guideline for leads, which would be more burden than help for many of the same reasons here (compare writing a lead about a 7th-century monk about whom there is only infrequent mention in standard texts of the time to that of a technical, basic subject such as Fungus). While section headers are definitely necessary in most well-developed articles, they aren't always helpful—a barebones lead followed by a few sections consisting of no more than two or three sentences would IMO disrupt the flow more than organize the article for the reader's benefit. In a nutshell, this issue is better worked out through common sense and editorial consensus than by a rule in the Manual of Style. Dabomb87 (talk) 04:02, 19 September 2009 (UTC)

Dabomb is right. Sections are a great idea most of the time. They keep articles neat and organized. However, we shouldn't explicitly require them in every case. Also, we might call them guidelines, but they're rules in practice. The idea that articles should have sections is treated as a guideline now. That, I think, is what we want. Ironically, it means we must not call it a guideline. Darkfrog24 (talk) 04:32, 19 September 2009 (UTC)

Just looking at Eurymedon vase, I think part of the problem of flow could be cured by writing a short lead that's different from the first paragraph, to let ignoramuses like me see the context faster, and why this is in fact an interesting subject. Then the table of contents wouldn't break up the flow of argument and exposition so badly (where the second paragraph refers to definitions in the first), and the body of the text could fall under some general second-order heading with less-obtrusive third-order subheads. But just like the length of articles and the number of paragraphs in a lede, the number of headings in an article should be guidance, which, as Darkfrog said, ironically means we can't call it a guideline. This, incidentally, tends to reinforce my desire to separate a short manual of more prescriptive and proscriptive rules (for things like ambiguity and accessibility) from general guidance and suggestions about style and common practice. —— Shakescene (talk) 04:47, 19 September 2009 (UTC)

Articles should always have sections beyond stub class. - Peregrine Fisher (talk) (contribs) 05:37, 19 September 2009 (UTC)

Adding section headers implies creating a lead; so I would tend to base any recommendation on WP:LEAD. A lead is supposed to introduce and summarize the article; so, ideally, the person adding the headers would also write a summary of the article for the lead, rather than just breaking an article into sections with no real lead. I can understand a certain reluctance to add a lead to an article if the whole article is shorter than the lead of a typical featured article, unless there is an obvious intent to create structure before adding more material. --Boson (talk) 10:47, 19 September 2009 (UTC)

If there's no section, then the lead is coterminous with the article; if it's too long for a lead, it should be split in sections. --___A. di M. 10:36, 20 September 2009 (UTC)

I agree. That would constitute good article management. But do you feel that it would be best for this to be stated outright in the guideline? (I don't.) Darkfrog24 (talk) 12:22, 20 September 2009 (UTC)

Maybe not in MoS, but perhaps somewhere in WP:Layout. --___A. di M. 12:31, 20 September 2009 (UTC)

Yes, good idea, ADM. And in WP:LEAD? Tony (talk) 13:55, 21 September 2009 (UTC)

I'm opposed to mandatory sections. An article having three paragraphs, like Eurymedon vase does not lead headings. The first paragraph there is the WP:LEAD. Pcap ping 20:11, 21 September 2009 (UTC) By the way, if you are still considering this, someone thought sections were appropriate because ... the number of words in each was "meeting some recommendations on a help page. The contents of each section is obviously irrelevant as long the the number of words is meeting the specification. ROTFLMAO. Pcap ping 20:32, 21 September 2009 (UTC)

Please do not misrepresent my position and present my comments out of context. My point there was as a direct answer to the incorrect assertion made by the opposing party that our argument was that 'style preferences should be settled by consensus rather than precedent', where he had already been explicit in defining 'precedent' in this matter as meaning his edits and preferred style, going so far as to say he had a right to veto any and all changes and that he would consider any edits he disagreed with as vandalism (and indeed went on to revert them as such). I was refering to pertinent information regarding guidelines for section size to show that there is indeed precedent that supported our position and also that one specific part of his argument, that the text of the article was inherrently not of sufficient length to warrant sections, was not supported by this information. This was not the basis, let alone the entire basis, of my argument in any event, but merely supporting evidence.Number36 (talk) 00:14, 24 September 2009 (UTC)

Standard table background colors

Many tables on Wikipedia contain repetitive, scalar kinds of information. Some people made templates for this task such as {{yes}} which usually transclude a cell background color and a default text. These are widely used as far as I can see. I’d like to pose the question whether the colors and sets shouldn’t be standardized more and then moved to MediaWiki:Common.css as much as possible. Please see also the collective Talk page. — Christoph Päper 17:16, 19 September 2009 (UTC)

Summary done

I have finished my summary of sections 3 and 4 of WP:MOSNUM, intended to replace sections 10 and 11 of WP:MOS. It is found at User:A. di M./MOSNUM. If no-one objects in one week, I'm going to do the replacement. Meanwhile, feel free to tweak it, copy-editing it, adding important stuff which I left out, removing or trimming not-so-important stuff which I kept in, etc. --___A. di M. 22:54, 20 September 2009 (UTC)

It's a pity that so much work has gone into producing two shortened forms of the same sections simultaneously, as though they're in competition. What would have been much more useful is for the non-MoS parts of MOSNUM—that is, the specialist stuff—to be (1) properly defluffed; and (2) disentangled from the material that is covered here at MoS main page. That would be a prelude to cleaning out of MOSNUM all of the duplicated stuff here. Is it possible that we might collaborate on cleaning up MOSNUM? Tony (talk) 13:22, 21 September 2009 (UTC)

What concerns me most is that I think the current coverage of MoS main page in these areas should stay (in trimmed form), and MOSNUM should delete them. MOSNUM is best concentrating on the substantial part of its content that is for specialist use. Tony (talk) 00:58, 24 September 2009 (UTC)

This is to say that the "Numbers as figures or words" section of MOSNUM should only contain:

• Using figures at the start of sentence risks the period being read as a decimal point or abbreviation mark.
• Do not use Nineteen forty five and 1950 were important elections for the Labour Party, but rather The elections of 1945 and 1950 were important for the Labour Party.
• Do not use the seventh of January or twelve forty-five p.m. or Two thousand eight was the year that ....
• Historical references such as Fifth of November are treated as proper names.
• Decimal representations containing a decimal point are not spelled out (1.00, 3.14159).
• Numbers in mathematical formulae are never spelled out (3 < π < 22/7, not three < π < 22 sevenths).
• Do not use spelled-out numbers before symbols for units of measurement: write five minutes, 5 minutes, or 5 min, but not five min.
• Quasi-continuous quantities are normally stated in figures, even when the value is a small positive integer: 9 mm, The option price fell to 5 within three hours after the announcement.
• When expressing large approximate quantities, it is preferable to write them spelled out, or partly in figures and part as a spelled‑out named number; e.g., one hundred thousand troops may be preferable to 100,000 troops when the size of the force is not known exactly; write Japan has the world's tenth largest population, with about 128 million people (as it is just an approximation to a number likely to be anywhere between 127,500,000 and 128,500,000), but The movie grossed $28,106,731 on its opening day (the exact quantity).
• When both a figure and spelled-out named number are used in a quantity, it is useful to use a non-breaking space, as in 128 million or 128{{nbsp}}million to prevent a line break from occurring between them.
• Sometimes figures and words may carry different meanings, for example Every number except one implies that there is one exception (we don't know which), while Every number except 1 means that the specific number 1 is the exception.

That is, there are rationales and examples of rules which aren't stated because the rules themselves are at MOS; also, the general rule is not stated because it's at MOS but particular cases are. It means that someone trying to get all the rules of when to spell out numbers and when to use figures will have to jump back and forth between pages, or have two windows each with one page. That doesn't sound right to me. --___A. di M. 10:21, 24 September 2009 (UTC)

Sorry, is this your version, the current MoS version or the current MOSNUM version? I have a number of micro-problems with it. Tony (talk) 10:58, 24 September 2009 (UTC)

Section headings/article titles - avoid superfluous ending punctuation?

Should there be some statement discouraging the use of punctuation such as colons or full stops/periods as the last character of a heading? Or perhaps even generally applicable to article titles? For example, the h3 subheadings under Mutya Buena#History use colons after a date range (which seems appropriate, when further words follow) but the headings also terminate with colons (which seems superfluous and inappropriate, a practice that does not seem to be normal in WP). Appropriate exceptions could be quotations or punctuation that is integral to proper names. Dl2000 (talk) 01:11, 21 September 2009 (UTC)

I'm not aware of any relevant provision in the MoS, but colons or another punctuation at the end of article names and sections should almost never exist. If there isn't a statement in the MoS that's applicable to this, I think one should be added. Emw2012 (talk) 01:41, 21 September 2009 (UTC)

Amongst

Despite the repeated reverts, I'd like to see a reference that amongst is "not widely accepted" as this MOS claims. Do not point me to List of English words with disputed usage. It's not covered there. As I pointed out it my initial edit, all that article says is that between is disputed when used instead of among/amongst, not that amongst is disputed, as this MOS claims. If anything, between should be mentioned in the MOS when used to mean among/amongst. Pcap ping 06:59, 21 September 2009 (UTC)

For what it's worth, here's what the 3rd edition of Fowler's says about it. First, in Burchfield's collections among is about ten times as common as amongst. Second, "There is no demonstrable difference of sense or function between the two, and the distribution is puzzling except that amongst seems to be somewhat less common in AmE than in BrE." Given this, I don't see why the MoS should prohibit amongst; it is less common than among but I see no sense in which it is "not widely accepted". Certainly amongst/among is in a different category from whilst/while, as whilst is not used in American English whereas amongst is. Eubulides (talk) 07:41, 21 September 2009 (UTC)

I only noticed that section of the MoS today because of the reverting. I think it should be removed. I often use whilst, amongst, overly (overly should be avoided?), and in writing I would sometimes use thusly. I take the point about "straining for formality" suggesting an insecure grasp of English, but that's when the words are over-used or used inappropriately; if we were to advise against using all the words Wikipedians use wrongly when they think they're being formal, we'd decimate the vocabulary. But these words, like the rest of the language, are fine when used correctly. SlimVirgin ^{talk|contribs} 07:56, 21 September 2009 (UTC)

(ec) Thanks for finding that detailed discussion. In the mainstream dictionary links given in List of English words with disputed usage amongst is given as variation of among, but with no discussion as to its use or acceptance (the only discussion was the disputed use of between instead of among/amongst). So, I assumed amongst itself is not disputed (contrary to what our MOS says). It looks like your reference confirms that in more explicit terms, even though among is more common. Since the MOS couldn't possibly give all examples in the "List of..." article, and that amongst isn't even disputed there, I think it's a really bad example to give in the MOS. Pcap ping 08:02, 21 September 2009 (UTC)

While, but not whilst, we're on the subject of forbidden words, apparently the MoS also doesn't allow the use of "didn't". I've had someone arrive at an FA candidate of mine to remove several instances of it because, it seems, the MoS says no. And not only remove it, but replace it throughout with "did not," leaving the sentences awkward and clunky. The people writing the MoS ought to bear in mind that there are lots of editors out there who hang on its every word, and indeed on its every forbidden word, so that what starts life as a helpful tip is interpreted as an unalterable and unquestioned fact about the universe that must be applied rigidly to all texts without exception or mercy.

If there is indeed a section advising against "didn't," does anyone mind if I remove it? SlimVirgin ^{talk|contribs} 08:06, 21 September 2009 (UTC)

I think contractions in general should be avoided. Encyclopedic writing is just about the highest register that exists; informality that would be perfectly acceptable in an academic journal or textbook (I was just noticing yesterday that Kunen's well-regarded set theory text has a section called How free are we to monkey with the powers of regular cardinals?) is jarring here.

My American ear reacts violently to amongst, and while I personally have no objection to overly, I'm perfectly willing to let it go if the A-word goes as well. --Trovatore (talk) 08:12, 21 September 2009 (UTC)

This should be about published canons of English, not some personal preferences of the few editors that watch this page. I suspect you'd object to British spelling like colour etc. But, that's explicitly allowed in this MOS. Pcap ping 08:20, 21 September 2009 (UTC)

I don't particularly like the colour spelling, but I can live with it just fine. Amongst on the other hand is really really jarring, sort of ultra-British, completely out of bounds on this side of the pond. I would be OK with it in articles on strongly British topics, I suppose, but not in ones that just happen to use British spelling because it's established in that article. After all, among is just fine in British English as well. --Trovatore (talk) 08:51, 21 September 2009 (UTC)

Philip Roth uses it, arguably America's greatest living writer. SlimVirgin ^{talk|contribs} 08:58, 21 September 2009 (UTC)

And so did Benjamin Franklin and Mark Twain. [6], [7]. Granted, they're amongst the dead. Pcap ping 09:11, 21 September 2009 (UTC)

(outdent) re "I think [this section] should be removed": would it be useful simply to provide the links so that people who get into disputes over these terms know where to look? maybe the section title needs changing as well – WP:Words to avoid isn't really "contested vocabulary", at least not in the same sense as List of English words with disputed usage. Sssoul (talk) 08:18, 21 September 2009 (UTC)

Our MoS has to take into account that the people applying it are not professional editors. If I'm writing for a magazine with a style guide that doesn't like "didn't," and my article has several of them, one of two things will happen: either my note in the text to leave "didn't" will be respected, or the editor will rewrite the sentences to remove it. But he will rewrite them. He won't simply plonk in a "did not" wherever there was a "didn't."

Because our MoS is applied by people who aren't editors, we can't invariably afford the luxury of having rigid rules as other style guides do, because we can guess that they won't be applied well. We are therefore far better not getting into issues such as when it's appropriate to use contractions, because it sometimes is, and it sometimes isn't, and the only way to tell, really, is to develop a feel for it. And even then, people will disagree. But this "highest register" thing is a recipe for precisely the kind of writing that was mentioned earlier, the "straining for formality" that suggests an insecure grasp of English. SlimVirgin ^{talk|contribs} 08:42, 21 September 2009 (UTC)

Well, the right solution there is not to take the MoS that seriously, which by a happy coincidence is my opinion already. I'll take your word for it that there are situations (other than, obviously, direct quotes) where someone might defensibly use a contraction in WP; I can't think of any, but then I can't think of everything. But in the range of those things the MoS can usefully do, I think pointing out that contractions are at least usually a bad idea, is clearly among the things it should do. --Trovatore (talk) 08:59, 21 September 2009 (UTC)

The problem is that it seems to be impossible to get people to apply the MoS with common sense. Either it's ignored entirely, or it's applied rigidly—and the people who ignore it entirely tend to do so because they've had a run-in with its rigidity. Given that this is a long-term problem, and is getting worse, it has to be addressed at the guideline level, and the only way to do that is to withhold the ammunition. SlimVirgin ^{talk|contribs} 09:03, 21 September 2009 (UTC)

But that contractions should be avoided, at least in most cases, is one of the best pieces of advice the MoS gives. That's precisely the sort of thing it should do. If you take that out, then what shouldn't we take out on the sort of grounds you're discussing? --Trovatore (talk) 09:15, 21 September 2009 (UTC)

I would take out anything that suggests there's a recipe for good writing. There are times when "didn't" and "amongst" are absolutely wrong, and there are times when they're the only words that fit. I would make that sort of thing clear. And I would strengthen the prohibition on going around changing from one style to another, or removing forbidden words, just because the MoS has expressed a view. A less intrusive MoS would be a more respected one. SlimVirgin ^{talk|contribs} 10:46, 21 September 2009 (UTC)

Now, I really can't imagine a case where amongst is "the only word that fits". Can you please give an example? --Trovatore (talk) 19:57, 21 September 2009 (UTC)

• Whether MoS says so or not, my view is that "amongst" and "whilst" (and many occurrences of "upon") should be discouraged by copy-editors: these forms have become distinctly old-fashioned, have a rather formal ring about them, and go against what seems (to me) a general rule of thumb that if it adds no meaning, don't use it (like most occurrences of "in order to"). Tony (talk) 13:27, 21 September 2009 (UTC)

Instead of banning this word or that word, we should advise against "straining for formality." That would allow editors to handle these matters on a case-by-case basis. In the case of "amongst," it's certainly not going to confuse anyone. Sometimes it's awkward and sometimes it's not. While it is more common in British English than American English, it's not incorrect in either one, so this isn't as clear-cut as other ENGVAR issues.

With regard to the fact that not everyone working on Wikipedia has professional-quality editing skills, however, I do not think that we should codify this in the MoS. Of course not everyone contributing to Wikipedia will do so on a professional level, but the whole idea is that then other editors who are more skilled can go in and polish up what they have added. So no, the MoS should not explicitly permit inappropriate styles solely because not everyone knows about them.

I would love to deal with the "people don't treat it like a guideline" issue, but that deserves a separate section and a new conversation. Darkfrog24 (talk) 14:00, 21 September 2009 (UTC)

I too am not in love with this these usually unnecessary formalized variations of common English words, but I don't think the MoS should ban them. However, whenever I review an article for a content-review process, I almost always recommend that the words are changed to their simpler forms. Dabomb87 (talk) 23:34, 21 September 2009 (UTC)

Since this topic has been seriously side-tracked by the discussion in the subsection below, let me try to summarize the main positions:

1. no real-world evidence has been provided that amongst is "not widely accepted" (personal opinions notwithstanding)
2. some editors still feel amongst should be forbidden because it's excessively formal

If (2) has consensus, can someone at least edit the guideline and provide a non-misleading reason for banning amongst? Pcap ping 18:09, 22 September 2009 (UTC)

It's not "excessively formal". It's agressively British. Since among works in British too, amongst should be avoided. Yes, this is a personal opinion, but it's one I happen to be right about. --Trovatore (talk) 18:16, 22 September 2009 (UTC)

Are you seriously proposing we write "don't use because it's aggressively British" in the MOS? Pcap ping 18:23, 22 September 2009 (UTC)

Not in those exact words. I am saying that this is the commonality among those words: amongst is gratuitously British, overly is gratuitously American, so ideally neither should be used in articles without strong natural ties. By gratuitously I mean that there are perfectly workable substitutes in both dialects, that can easily be swapped in without affecting the meaning, and that do not have a negative effect on readers from another dialect. --Trovatore (talk) 18:29, 22 September 2009 (UTC)

Fowler's contradicts the claim that amongst is aggressively British. Please the second comment in this thread. Eubulides (talk) 18:41, 22 September 2009 (UTC)

At the same time Fowler's does say that amongst is less common. So, if we still want to forbid it, perhaps say something like "When a word has several variants with the same meaning, use the most common one." Pcap ping 18:46, 22 September 2009 (UTC)

Fowler is from 1926. The other examples of American usages were from the 19th century, excepting Philip Roth, for whom I suspect it's an individual idosyncracy. Really, amongst is an extreme outlier in contermporary American usage. To me it seems that there is no reason to use it except to emphasize one's Britishosity; that's what I mean by "gratuitous".

As to "forbidding" it: Let's keep in mind that the MoS has no authority to "forbid" anything. What it can do is point out that the word is likely to be received negatively by American readers, as it is possible that some Commonwealth editors don't know that — I didn't know about overly until I saw it in that list. --Trovatore (talk) 19:33, 22 September 2009 (UTC)

That's exactly what we're talking about under "MoS's role as a guideline." Perhaps the MoS isn't supposed to be taken as a ban on this word or that, but it is in practice. We have to take that into account on the "amongst" issue. Darkfrog24 (talk) 20:00, 22 September 2009 (UTC)

No, we should fix the problem. --Trovatore (talk) 20:07, 22 September 2009 (UTC)

Fowler's quoted by Eubulides is the New Fowler's. It also gives multiple examples of well-known authors using it in 1970s and 1980s. It won't bother listing them here. Pcap ping 19:46, 22 September 2009 (UTC)

Replying to Trovatore's comment of 18:29: that also applies with verbs in "-ise", but we also have article titles such as characterisation which is an article having nothing to do with the UK, but AFAIK no-one ever proposed to deprecate -ise verbs on WP. --___A. di M. 12:47, 23 September 2009 (UTC)

I think there's actually a very faint difference in connotation or style between "among" and "amongst", because I might use "amongst" in some contexts and "among" in most others. A tabloid (even an American one) might scream "The Terrorists Amongst Us", and the difference from "the terrorists among us" to me is that the latter suggests that the terrorists are one of us, while the former suggests more of an alien incursion. This kind of thing might fit in a general discussion of style as found in something like Fowler or Eric Partridge or William Safire, but it has no place in a prescriptive style-sheet like the AP's and should certainly not be in a Manual of Style that's used by 'bots and typo-patrollers. —— Shakescene (talk) 13:52, 23 September 2009 (UTC)

Data

The Cambridge Guide to English Usage has much more informative data:

“ amid, amidst, among or amongst

These four prepositions share much the same grammatical functions these days, but differ somewhat in their regional distribution and their applications. Overall the shorter forms (amid and especially among) are much more frequent than the longer ones, as the relative percentages show in both British and American databases:

BNC CCAE
amid 3.8% 6.2%
amidst 1.7% 0.4%
among 78.8% 93.1%
amongst 15.7% 0.3%

”

So, Trovatore was right about amongst being very British :-) But this kind of details are over the top for MOS. I think a general recommendation to use the common variants of words with the same meaning is sufficient... Pcap ping 10:11, 23 September 2009 (UTC)

The less common versions like amid(st) have their own idiomatic usage however, e.g. amid(st) speculation but not among(st) speculation (there's longer table in the Guide, I won't copy it here). Please read the whole entry in the Guide before jumping to banning any words explicitly. Pcap ping 10:27, 23 September 2009 (UTC)

Merriam Webster's Dictionary of English Usage has this to say:

“ Among, amongst. Most of the commentators who mention these words note that amongst is less common but both are correct. Our evidence confirms this; it also shows amongst a bit more common in British usage than American. The few commentators that who call amongst quaint or overrefined are off target. Here is a selection of mostly recent and unquaint examples [about 7 examples from 1970-1990 works.] ”

Not that quantitative, but the same idea. Pcap ping 12:49, 23 September 2009 (UTC)

That seems to be the point here. This may be an ENGVAR issue, but it's not a straight ENGVAR issue. In American English, "centre" is wrong. In British English "center" is wrong. However, while "amongst" is certainly more common in British English, it is not incorrect in either variation. We should not ban it. Darkfrog24 (talk) 13:43, 23 September 2009 (UTC)

(ec) By the way, unlike most dictionaries, this M-W usage dictionary is readable on google books; the Cambridge Guide isn't. Pcap ping 13:46, 23 September 2009 (UTC)

Overly

Since this was raised above as an example in the other (that is, overly American) direction, this is what The Cambridge Guide says:

“ This adverb meaning "excessively" seems to have originated in the US in C18. It has steadily gained ground in the UK, by the note in the Oxford Dictionary (1989), and by its use in everyday writing in almost 150 BNC texts. [Examples:]

• the demise of the overly modest bathing costume
• not being overly concerned about the future

Despite suggestions that overly is an unnecessary word--which could be replaced by "too or "excessively"--it fills the niche between them in terms of bulk. Overly is equally useful in noun and verb phrases, as shown in those examples, whereas "too" is awkward in the noun phrase, and "excessively" rather an overkill in either. One other alternative is to use the adverbial prefix over-, creating compounds such as overmodest, overconcerned, although this works better with the second type (adjective / past participle) than the first (a simple adjective).

”

So, this is even less clear cut than amongst. All these discussion only serve to show that edicts banning one word or the other in this MOS are fairly misinformed... Pcap ping 13:06, 23 September 2009 (UTC)

M-W Usage Dictionary has this entry:

“ Bache 1869 and Ayres 1881 succinctly insulted contemporaries who used this word, calling them vulgar and unschooled. Times have changed: modern critics merely insult the word itself. Follett 1966, for example, claims that overly is useless superfluous, and unharmonious and should be replaced by the prefix over-. Bryson 1984 adds that "when this becomes overinelegant ..., the alternative is to find another adverb: 'excessively' or 'unnecessarily' or even the admirably concise 'too'." You may not want to go to such lengths to avoid overly, and some modern commentators (Evans 1957, for example) would agree that there is no need to. We concur. In fact, in some cases none of the alternatives sound as good as overly.

• ... a not overly clean tunic
• ... the author of various not overly inspired books
• make his otherwise overly comfortable life more interesting

[I've omitted the original source citations]

Even when over-, too, excessively, or unnecessarily would be a successful substitute, the fact that a synonym is available does not mean that you have to use it. [6 examples follow]

Overly is not as commonly used in England as in the U.S., but according to the OED Supplement it is gaining ground there.

”

So, I'm going to remove overly from the short list of banned words. Pcap ping 13:33, 23 September 2009 (UTC)

Support. "Overly" is not incorrect English. Rather than banning it, we should simply allow users to correct it where it is awkward. Darkfrog24 (talk) 13:55, 23 September 2009 (UTC)

MoS's role as a guideline

(Conversation has been sectioned off from "Amongst" as of 9-21-2009. Darkfrog24 (talk) 01:52, 22 September 2009 (UTC)

(reply to Darkfrog) That is the key issue about the MoS, in my view. If people would remember it's a guideline, which means it's advisory, it would be fine to have all this detailed information. But it's being applied as though it's policy, and applied rigidly, regardless of context. That makes it unpopular, which means it ends up being ignored even when it would be a good idea to heed it. I would like to see a section making very clear that this is a guideline, but written carefully in a way that won't undermine it. SlimVirgin ^{talk|contribs} 23:39, 21 September 2009 (UTC)

SlimVirgin, can you clarify what you mean by "advisory"? This road sign has an "advised" speed limit, but many drivers can and do ignore that speed recommendation (myself included) anytime they want. By the same token, are you saying that editors should only follow the MoS when they feel like it (i.e., if they just don't like what is says, they can ignore it without further reason)? I shudder to think what would happen if editors took that view toward WP:RS and WP:SPAM, and I don't think that's what you meant.

If you in fact meant that the MOS is a guideline that should generally be followed except for those times (not too frequent, but not unheard of) that there's a good reason (i.e. more than IDONTLIKEIT and in compliance with editorial consensus) to ignore it, then I'm right with you.

Now, with regard to the specific proposal, I agree that we don't need to spell out every word that should not be used; after all, we've been working for the past few weeks to condense the MOS by reducing redundancy and these unnecessarily specific points. Dabomb87 (talk) 00:58, 22 September 2009 (UTC)

Could it be that the MoS is a major but silent agent for minimising edit wars? Tony (talk) 01:16, 22 September 2009 (UTC)

To my mind, "guideline" means "here are some helpful but not obligatory pieces of advice" in ordinary speech. In my opinion, on Wikipedia, "guideline" should mean, "Do this unless you can get a consensus that there's a good reason not to." For example, the guideline might say that the mid-sentence "the" should not be capitalized, but the fans of the band "the Beatles" decided that it should be in the case of the name of the band (and let's assume for the sake of this example that their reasons were sound and logical), so articles on the Beatles ignore the guideline and spell the name "The Beatles."

Tony also makes an excellent point. The MoS might not have been designed to serve as a tiebreaker in the prevention of edit wars, but it probably does serve that role. Anything we do should preserve that.

SV, we might spell out a paragraph saying that this is a guideline and not hard-and-fast rules, but would people take it at face value? Right now, the culture of Wikipedia is to treat the guideline as rules. We could, of course, just use that by spelling out the exact circumstances under which the guideline should or shouldn't be followed...

We should acknowledge this as one of our options: We could always just change the name and refer to the MoS as rules openly. That, at least, might get people treating other guidelines as guidelines. Darkfrog24 (talk) 01:52, 22 September 2009 (UTC)

The MoS has made a major contribution to improving the standard of articles over the past few years: I see no reason to weaken its role. Tony (talk) 01:58, 22 September 2009 (UTC)

The MoS is a guideline. Most editors who are not too new to Wikipedia roughly know what a guideline is, both in absolute terms and relative to policies—on one hand—and to essays—on the other hand, though some are clearly more influential than others. Adding a section explicitly saying to take anything the MOS says with a grain of salt would probably further undermine the MOS's role, as well as add to the bloat we've been fighting so hard to keep out. The authority of the guideline is clearly stated in the banner at the top, and #General principles is enought to keep most style arguments that aren't decided by MOS at bay. Dabomb87 (talk) 02:12, 22 September 2009 (UTC)

Dabomb, when we say guidelines are advisory, it means editors don't have to follow them. The policy on sourcing is V, not RS. People are welcome to ignore RS, except that, most of the time when I've looked at it recently, it says the same as V, and really shouldn't exist for that reason, but that's a separate issue. Policies are regarded as mandatory, but guidelines not.

Tony, I've seen the MoS cause and resolve edit wars. It's hard to know what it does most. I do know that, whenever an MoS-related issue has come before AN/I or ArbCom, editors are advised not to arrive at articles only to make style changes, as the MoS itself as always said or implied, going back to 2002: "Writers are not expected to follow all these rules ...", or 2004 quoting the Chicago Manual of Style: "Rules and regulations such as these ... are meant for the average case, and must be applied with a certain degree of elasticity." Then in 2005 or 2006, we had a couple of ArbCom cases where the principle was upheld.

It's that elasticity that is being eroded. It is being applied as though it's policy. SlimVirgin ^{talk|contribs} 02:24, 22 September 2009 (UTC)

I'm not saying that guidelines should be treated as policies. I know that guidelines, including the MOS, are not to be enforced with an iron rod, and that flexibility and allowance for exceptions are vital principles of a guideline. What I am saying is that on the other side of the spectrum, editors sometimes (intentionally) don't follow guidelines beacause they simply don't like what it's telling them to do without providing a specific reason, common sense or otherwise, to ignore it. Since guidelines reflect (or at least are supposed to reflect) consensus, they generally should be followed. Dabomb87 (talk) 02:46, 22 September 2009 (UTC)

I will add that if users are trying to wield MOS and other style/content guidelines as policy, then the problem might not rest with the MOS or any one guideline so much as a flaw in the way guidelines are viewed, and their relationship with policies. Dabomb87 (talk) 02:53, 22 September 2009 (UTC)

(ec) Guidelines often don't reflect consensus, though, because too many people edit them, and not enough people keep watch over them. It's much harder to change a policy—they're more stable and there are more eyes on them, which makes it safer to assume there's consensus.

Most people have no idea what the MoS says. I'm one of those editors who frequently ignores it, either because I don't agree with some part of it, or because it's not clear that there's any consensus for it. The business about images needing to be thumbs, and not being allowed to be on the left under third-level headings, are two recent cases in point. Wikignomes have been going around enforcing these for the last couple of years, and edit warring if anyone reverted them, yet there was no consensus for them (as you could see by the fact that they were widely ignored), and no good reason to have them in the guideline in the first place.

Every time a wikignome does something like, he undermines the MoS. A section advocating more flexibility would strengthen the MoS (if carefully worded), not weaken it. Think of it as a high-rise building in a strong wind. We need it to bend. SlimVirgin ^{talk|contribs} 03:00, 22 September 2009 (UTC)

I agree with SlimVirgin. Too many editors have defined their purpose on this wiki as the enforcement of the MOS. Furthermore, these editors constantly want to expand the set of rules that justifies their actions. Just look in the section above titled #Proposed addition to MOS:HEAD, which is motivated by a WP:LAME edit war over adding section headings to a three-paragraph article (Eurymedon vase, drama still going strong). Pcap ping 06:00, 22 September 2009 (UTC)

Adding headers is being equated with vandalism. :) SlimVirgin ^{talk|contribs} 06:34, 22 September 2009 (UTC)

I agree with SlimVirgin's original comment and the subsequent posts that emphasise that MOS is a guideline, not a policy. Perhaps we need to spell out some criteria or higher-level objectives by which to judge whether MOS should be applied in full in specific cases.

I suggest as criteria: whether full application of MOS would benefit ordinary readers (as opposed to style mavens); and whether from the ordinary reader's point of view the effort would better spent on other aspects of the same article, or on other articles. --Philcha (talk) 06:14, 22 September 2009 (UTC)

I don't know about that, Philcha. In general, ordinary readers benefit from precise and correct styles, even if they cannot recognize them. A non-expert, upon viewing up a crummily edited page, can tell that there's something amateurish and sloppy about it even if he or she can't pick out all the specific mistakes. Also, consider that just because people are told to stop doing one thing doesn't mean that they switch to other parts of the article. I'm not an expert on, say, Zoroastrianism, but if I happen to be reading an article on that subject, I might duck in to correct some punctuation even if I wouldn't presume to change the content. If I were not to polish the style, then I wouldn't go off and spend my time scrubbing content; I'd do nothing to the article at all. Darkfrog24 (talk) 13:51, 22 September 2009 (UTC)

Avoid instruction creep has always been an important principle on WP. It advises

"For proposed new instructions, instruction creep can be avoided if all of the following hold:

"1. There is a good indication of an actual problem (as opposed to a hypothetical or a perceived problem).

"2. The proposed instructions truly solve this problem (as opposed to treating symptoms or making symbolic gestures).

"3. The instructions have few or no undesirable side effects (such as false positives, overcomplexity, or unnecessary prohibitions)."

And then an important point: "Policies and guidelines exist to document accepted practice, rather than actually dictate such. If on such a page an instruction appears which does not accurately reflect commonly accepted practice, and then this instruction gains the consensus of editors who happen to participate at the talk page in question, then the process has failed."

In other words, there is a strong descriptive element to policies and guidelines, not only prescriptive. This is one of the MoS's key problems. There are parts of it that describe only what the editors of the MoS, often going back many years, wanted people to do. Forcing editors to use only thumbnails of images, for example, was widely ignored by good editors. It had to be ignored because it made articles look silly, but it sat here for years, all efforts to get rid of it were fought, and wikignomes routinely edit warred to remove fixed image sizes. It wasn't until Tony noticed it that we finally got it out of the MoS. I'm guessing there's a lot more material like that still in it, and there are routine attempts to add even more. SlimVirgin ^{talk|contribs} 06:27, 22 September 2009 (UTC)

←"Policies and guidelines exist to document accepted practice, rather than actually dictate such." But they do end up dictating: WP:CIVIL, WP:NFC, WP:NC and many more. If there is consensus at an article to breach an aspect of the MoS, those who want to follow that course need to provide compelling reasons when another editor wants to bring the article into compliance with the MoS. Otherwise, I say to them: go to WT:MoS and argue it out there. It is, after all, part of the boiling pot that feeds into the MoS (as you pointed out WRT the section on images).

We must have a degree of centralised formatting and style—it provides cohesion, authority, even branding for WP. This is much of why the wiki mechanism works: freedom and constraint working alongside each other—sometimes in dynamic tension. When a reader chooses a WP article in a google search, they unconsciously expect to find those branding attributes. It's just like every other reputable publisher: a house style is vital to the message and the brand. It has become part of WP's business model, its public image, and we should put our energy into helping the style guides to evolve, not to removing or weakening their authority. Moreover, there is much anecdotal evidence that editors want to follow guidance. Show me an article riven with MoS breaches and a sibling article that is MoS-compliant and you'll see a much better product (if not, change MoS, don't disregard it).

When I read my Scientific American each month, I would be upset to find discrepancies in the use of images, headings, certain punctuation. I trust those surface details and they partly confirm my admiration for the deeper meanings. It is the dressing. Tony (talk) 07:47, 22 September 2009 (UTC)

I agree with a lot of what you say. But I think you overlook a few key issues:

(1) It is virtually impossible to change the MoS on certain points, because a small number of editors simply won't allow it. I'm therefore not allowed to write "organize," because some people think it's not British, when it is. I'm not allowed to write September 22 for the same reason—and that's also perfectly standard in the UK. I'm not allowed to place punctuation inside quotes, even though that's as British as ham and egg for breakfast—it's so-called logical punctuation that's the newcomer.

A group of professional editors of the kind who wrote the Scientific American style guide are not making decisions based on mistaken notions of what is and isn't British. I'm happy to go along with a guideline that is rational, even if I disagree with it, but it's infuriating to see so much of it based on misunderstandings.

(2) Other style guides don't have our "descriptive, as well as prescriptive" rule. That's peculiar to Wikipedia, and people did not agree to let the MoS be an exception. We are supposed to prescribe whatever is best practice on Wikipedia, and that's what good editors do as a matter of fact. Best practice is not what good editors would do if they were doing what the MoS editors wanted them to do back in 2004.

(3) Scientific American is a professional publication. Everyone is trained, everyone is paid, and there are professional copy editors, who are flexible when they need to be. In my own writing offwiki, I have never been dictated to about style issues the way I have been onwiki. SlimVirgin ^{talk|contribs} 08:03, 22 September 2009 (UTC)

(Pohta, in a Bishonean mood, looks at this talk page and can't help but nod towards "it's infuriating to see so much of it based on misunderstandings.") Pcap ping 09:09, 22 September 2009 (UTC)

I just read my previous post again, and I must apologize for sounding so shrill. :) It's just that some of this is really quite frustrating. SlimVirgin ^{talk|contribs} 09:17, 22 September 2009 (UTC)

I tend to disagree with you about the impact of the MOS on Wikipedia's cohesion and branding. When people arrive at a site and see our page design and logo in the corner, they recognize that they are on Wikipedia. They probably also recognize they've arrived here when they find video game trivia in articles about staid historical figures. Nobody, however, associates logical quotation or some other technical point of punctuation with our brand and it wouldn't be harmed if articles assumed varied postures on these issues.

These rules exist, in my view, largely divorced from any calcuation of their necessity or usefulness in the context of Wikipedia. The logical quotation rule is an excellent example because it makes sense at WT:MOS - editors here feel it's a good idea so why not make it a rule? cheers all around - but little sense in the context of actually reading Wikipedia - nobody has any idea what quotation rule any given instance follows, so there's no information to be gained. The current structure of MOS encourages a conflation of two very different questions: (1) what would be an ideal house style if I were starting a publication from the ground up? and (2) what guidelines best serve Wikipedia and its readers? Christopher Parham (talk) 15:08, 22 September 2009 (UTC)

It is true that guidelines are not policy. That does not mean that guidelines may be freely ignored. Both policies and guidelines are to be applied using common sense. An editor should have a good reason, beyond personal preference, for deviating from a guideline in a particular instance. According to Wikipedia's policy page on Policies and guidelines, "Wikipedia policies and guidelines are developed by the community to describe best practice, clarify principles, resolve conflicts, and otherwise further our goal of creating a free, reliable encyclopedia ..." As for their role, "Guidelines are primarily advisory. They advise on how to prevent or avoid causing problems, and on how to apply and execute policy under specific circumstances." However, guidelines are enforceable: "If an editor violates the community standards described in policies and guidelines, other editors can persuade the person to adhere to acceptable norms of conduct, over time resorting to more forceful means, such as administrator and steward actions."

Most publications strive for consistency of style and presentation, and adopt style manuals as a means to that end. Unlike Wikipedia, at most publications, most editors have no say about the publication's style manual; a small group of editors known to be good "stylists" prescribe the styles that all editors must follow. Further, while the editors who write those style "guides" try to compile the specific usages that they believe are best suited to their particular publication, in instances where alternative usages might be equally appropriate, they pick one for the sake of consistency. Wikipedians who chafe at our MOS should realize that it allows individual editors far more freedom than any other style guide on this planet. Most style guides either prescribe or prohibit the so-called serial comma, for example; our MOS leaves the choice up to the editors (but, appropriately, requires consistency within a single article). Further, those editors who do not wish to master the entire MOS should be thrilled that some of us are so deranged that we will do this grunt work for them. Finell (Talk) 16:36, 22 September 2009 (UTC)

I couldn't disagree more with your last sentence. To the extent that any content contributor bothers to follow any aspect of the MOS, that's a kindness those who care about this shouldn't expect. Indeed, if you're copyediting an article, you should be thrilled that an author has released his original, legally protected content under a license that allows you to fiddle with it. The article was in all likelihood perfectly clear and understandable, and equally valuable to any reader, before it was brought into compliance with the MOS. And to your first point, we would never block or sanction someone for not following the MOS (at least not any of its technical points) w/r/t original content they were writing. (People making edits, especially small ones, to existing content are a different story.) That would simply violate our values. Perhaps this indicates that "guideline" is not quite the right status for MOS; it's not really a standard we expect all editors generally to adhere to, but rather a standard we expect the encyclopedia to aspire to as a whole and don't want anyone to actively subvert. Christopher Parham (talk) 18:19, 22 September 2009 (UTC)

Guys, this is no time to get all content vs. gnoming. Both contributors and editors improve the Wikipedia reader experience and both roles should be celebrated. Finell, Chris P, if you would both join me for a brief sing-along... 19:58, 22 September 2009 (UTC)

Whoa! Christopher, where in the world did you get the idea that I don't value content? Or even that I value style (in the sense we are using that term here) as much as content? It is a given that content is king. Good editing improves the quality of content by improving its readability and communicative power, but editing can't do anything without quality content. Further, I value content editing (the kind that, say, Encyclopaedia Britannica staff editors to with the content submitted contributing "editors", who are actually authors) to improve organization, grammar, syntax, word choice, concision, etc., over style-guide conformity. But copy editing (in its limited sense) and style-guide conformity still contribute to improving the quality of Wikipedia. But I repeat, content is king. I am not aware of any editors who think otherwise. Finell (Talk) 21:31, 22 September 2009 (UTC)

I can appreciate what Parham is saying, but the content of the MoS isn't the issue at hand. The problem isn't that we have an MoS in the first place—that's a good thing. The issue is that people treat the MoS as if it were rules, not as if it were a guideline. What we should be trying to establish is 1. whether or not this is a problem 2. if so, what outcome would be best and 3. how to accomplish said outcome. Darkfrog24 (talk) 17:51, 22 September 2009 (UTC)

Finell, your quoting that policy page above is a good example of why unstable policies or guidelines need to be ignored. That page has been under attack for a few months by a couple of editors to want to minimize the mandatory nature of the policies. To do this, they've been editing that page to equate policies with guidelines. That's the only reason there's anything about people possibly being blocked for violating a guideline. In fact, no one would be blocked for that, especially not for violating the MoS. At countless AN and AN/I discussions, and a few times at ArbCom, the principle is always upheld that no one should be edit warring to impose style changes on stable and internally consistent articles.

As Christopher said, there are a few style issues that affect our content in the sense of providing us with a strong corporate identity: page layout being the most obvious. But most of the MoS revolves around personal preferences that otherwise no one notices. Those are the issues I'm arguing should not be imposed on articles over objections. Because otherwise what we're saying is that the personal preferences of people who regularly edit the MoS matter more than the personal preferences of the writers who may disagree. SlimVirgin ^{talk|contribs} 21:50, 22 September 2009 (UTC)

Is this all about internal vs external punctuation? Is it born mainly of the frustration of those who "don't like it"—that they've never been able to gain consensus to change the external rule? It seems to me that the whole of the MoS is now being denigrated because of this. Tony (talk) 03:38, 23 September 2009 (UTC)

It doesn't really have anything to do with punctuation (at least not from my perspective). It has to do with instruction creep, and the way the MoS is enforced. It was never intended to be enforced rigorously like this. The point of it was to offer advice about some common errors, find compromises where possible, and offer options where not possible. Now we have a situation where it's causing a fair bit of ill feeling or it's being ignored entirely, neither of which is good for it. What I would like to see is a very professional style guide that is to some extent descriptive of what editors already do, and which is gently enforced. SlimVirgin ^{talk|contribs} 03:56, 23 September 2009 (UTC)

As there is no logical or practical reason to ban American punctuation, I would have to agree that problem seems to be that there are people who just don't like it and have imposed their preferences on the rest of us. But no, that's not why I split the section off.

It's come up again with "amongst," but it's happened before. Someone requests that a certain thing be added to the MoS, but we find that it wouldn't be best for us to do it because Wikipedians would take it too much to heart. For example, it would probably be okay to have an actual in-practice guideline advising Wikipedians to avoid words like "amongst." We've had people say, "Oh it would be all right to add this because the MoS is only a guideline." However, if we say "don't use 'amongst'" on the MoS, then people will treat it as a hard-and-fast rule, removing the word even in situations in which it would be desirable to have it remain. The result is that we don't get to add stuff like that to the MoS.

The MoS is described as a guideline, but people treat it like a set of concrete rules. I split off the conversation so we could evaluate 1. whether or not this is a problem (we could of course just keep things as they are, leaving things like "avoid 'amongst'" unwritten), 2. what the ideal role of the MoS should be and 3. how we would effect it.

As for descriptive/prescriptive, the MoS's job is to give instructions. It should be written in the imperative. The issue is how fiercely those instructions should be implemented. Darkfrog24 (talk) 04:07, 23 September 2009 (UTC)

Constructing an article, or individual parts, that defy the basic instructions of the MoS are doomed to failure. Browsers like Firefox increasingly have pluggable editors that automatically analyze Wikipedia articles and recommend spelling and style changes that can be enacted with one-button while reading or editing. The MoS does not necessarily have to be enforced rigorously but attempting to maintain styles against the MoS is increasingly an attempt to hold back the tides. Miami33139 (talk) 04:14, 23 September 2009 (UTC)

Ignoring rubbish prescription like we had for overly until a few minutes ago is simply common sense. I fail to see what "pluggable editors" have to do with it. Maybe you plug one into your brain/Firefox and it writes "correct" English according to this MOS?! Like I wrote above a couple of times, a shocking amount of material in the MOS is based on misconceptions not supported by scholarly research or other widely used style guides. Take a look at the section on punctuation in quotes for another example. (#Never understood). Pcap ping 13:59, 23 September 2009 (UTC)

Actually, the existence/use of automated tools for wikignoming are an argument for being conservative with the prescriptions here. See below. Pcap ping 14:34, 23 September 2009 (UTC)

I think User:Shakescene raised a really good point above, namely that we need to carefully consider the demographics of our audience in these guidelines. You can fully expect some teenager to write a script replacing a "banned" word in all articles. I've seen editors auto-link words ignoring context for example. We already had some holy wikiwars involving automated tools over MOS issues I consider irrelevant, e.g. date linking. So, we need to be conservative with the MOS prescriptions because there's a real chance of causing WP:LAME incidents rather than preventing them. Pcap ping 14:20, 23 September 2009 (UTC)

I don't think I agree with a single thing in the previous post. Tony (talk) 14:33, 23 September 2009 (UTC)

I think you're confusing User:Tony1/How to improve your writing, which is a respectable enterprise, with this MOS, which is an equally respectable, but different enterprise. Pcap ping 14:39, 23 September 2009 (UTC)

What Pcap seems to be describing is that we accept that the MoS is treated as rules and therefore walk carefully when adding or removing instructions. This is one way to deal with it. But if we're going to do that, I think we should acknowledge it openly. Darkfrog24 (talk) 15:12, 23 September 2009 (UTC)

Some of these grammar/vocabulary prescriptions might be appropriate at MOS:BETTER, although I don't see why that page needs to be a guideline at all. What makes it different from this page? It addresses no discernible subtopic but rehashes some of the stuff here. Pcap ping 17:23, 23 September 2009 (UTC)

(1) "Amongst terrorists" is hedgehog language. (2) I'm not convinced that there's anything wrong with the provision of a small amount of assistance for readers in the fields of grammar and word usage. Where is the line drawn if advice against the use of old-fashioned words such as this (with letters redundant against their modern counterparts) is expunged?Tony (talk) 01:02, 24 September 2009 (UTC)

Proposed rule

Based on the above discussion I suggest the following:

When a word has multiple variants with the same meaning but different degrees of use in major varieties of English, prefer the word variant globally accepted. For instance, opt for among instead of amongst (the latter is seldom used in American English) and thus instead of thusly (the latter is seldom used in British English).

Comments? Pcap ping 09:36, 24 September 2009 (UTC)

P.S.: I know that thusly is considered "hypercorrect" and is used mostly for ironic or comic purposes even in AE, but I don't want to make this rule longer than it strictly needs to be. Other rules (should) deal with the degree of (in)formality acceptable in Wikipedia articles. Pcap ping 09:45, 24 September 2009 (UTC)

Oppose. While "among" will often be preferable to "amongst," the latter isn't incorrect and we should not ban it. We already permit editors to correct awkward or overly formal turns of phrase. That already covers "amongst." Darkfrog24 (talk) 12:04, 24 September 2009 (UTC)

Thusly

Since this may cause a dispute as well: both Fowler's 3rd edition and The Cambridge Guide discuss thusly only briefly, and stress its Americanness and its (original) ironic/comic purpose. M-W Usage Dictionary has nearly a page-long entry, in which they point out that even in some non-ironic instances thusly is more suitable than thus, and recommend that one should not automatically replace thusly with thus in all circumstances, but suggest that "in this way" or "as follows" should also be considered as replacements depending on context. Pcap ping 10:02, 24 September 2009 (UTC)

"Custom" TOCs

Are there any guidelines on custom TOCs, like this [8]? I think they are a bad idea as breaking the automatic link between the article headings and the TOC means it's liable to stop working if someone changes the former without updating the latter. The Red Hat of Pat Ferrick ^t 11:16, 21 September 2009 (UTC)

Such things should not be used. Articles that are sufficiently long should use the regular Mediawiki TOC and not the NOTOC keyword. The example provided has a TOC that looks exactly like an infobox. There are many benefits to having all articles use the same format for TOCs. — Carl (CBM · talk) 11:20, 21 September 2009 (UTC)

Thanks. I posted a notice on the talk page of the editor who (re)added it. The Red Hat of Pat Ferrick ^t 11:24, 21 September 2009 (UTC)

Just to be complete, here are some of the reasons I have in mind for using the standard TOC:

• Readers are used to looking in that place for the TOC, and so putting it somewhere else makes navigation slower. If every article puts the TOC in a different place, we have to start searching for them. This example makes the TOC look like a navigational template pointing to other articles.
• Having all the TOCs use the standard system allows us to format them all uniformly. If some TOCs use a different system, changes we make to the standard system will not be reflected in those TOCs
  • Moreover, we already have a system in which the various parts of the TOC can be styled by CSS and manipulated by scripts. Hand-made TOCs such as this [9] do not have the appropriate element IDs and CSS styles. We should not expect users to maintain these manually; the automatic TOC includes them without any user effort.
• Manually-created TOCs have to be updated every time the sections are changed, rearranged, or retitled. This is a maintenance problem because few editors expect that, when they retitle a section, they must also edit the TOC manually.

— Carl (CBM · talk) 11:49, 21 September 2009 (UTC)

Well, the typographic standards for TOCs are pretty sucky on Wikipedia (To me, they are the 2nd most annoying thing next to the incredibly long lines [on >1024px-width screens], which negatively impact readability). The Macau article looks way better with the custom TOC. Having said that, there are indeed good maintainability reasons to use an auto-generated TOC. Is there a way to style the auto-generated TOC on a page independently of the global CSS settings? Pcap ping 12:37, 21 September 2009 (UTC)

It seems strange to make each page have a different TOC; really the appearance of the TOC should depend on the skin, not on the article. The solution would be to have an RFC to choose a design, saying how we want all tables of contents to look. If there were agreement on a redesign, then we could implement that in Mediawiki without much effort. — Carl (CBM · talk) 13:42, 21 September 2009 (UTC)

We do have formatting templates like {{TOCright}} and custom TOC templates like {{AlphanumericTOC}}. But we shouldn't be creating anything that looks different, like in the Macau example. --NE2 16:04, 21 September 2009 (UTC)

Most of the uses of TOCright in articles should be removed; unless there is a left-aligned image or similar, the TOC should just be in the default position. For pages in other namespaces, it makes less difference. Many of the uses are disambiguation pages, where there is particularly no reason to float the TOC anywhere. — Carl (CBM · talk) 17:49, 21 September 2009 (UTC)

Silently correct typos in quotes?

"Trivial spelling or typographical errors should be silently corrected (for example, correct ommission to omission, harasssment to harassment—unless the slip is textually important)." Why? Isn't that what [sic] is for? Dabomb87 (talk) 03:13, 22 September 2009 (UTC)

[sic] is only for misspellings in printed text written by the original author/speaker. If the quote is taken from a newspaper report of something that was originally spoken, then one presumes that it was the newspaper's writer/copyeditor/typesetter that made the error, not The original speaker that is being quoted. In such cases, it's usually appropriate to just fix the mistake, unless the point is to illustrate the mistake in the first place. oknazevad (talk) 03:33, 22 September 2009 (UTC)

Thanks. Might this be clarified in the MOS? Dabomb87 (talk) 03:51, 22 September 2009 (UTC)

I'd be very wary of "correcting" any but the most blatent typos unless one is absolutely confident of one's command of the style of English being used. Meddling people are always "correcting" period quotes or UK English & have to be reverted. Johnbod (talk) 04:36, 22 September 2009 (UTC)

I concur with Dabomb. The fact that this is only supposed to be used in cases in which the source can be expected to have misspelled another entity's spoken words should be spelled out. In this case, using [sic] would suggest that the speaker, not the newspaper etc., made the error. Darkfrog24 (talk) 05:28, 22 September 2009 (UTC)

i think that's why the guideline refers specifically to typographical errors (which can slip into quotes of written statements as well as spoken ones); but yeah: clarifying that [sic] is used to indicate errors made by the original writer/speaker is probably worthwhile. Sssoul (talk) 05:46, 22 September 2009 (UTC)

This may be quite irrelevant, but I think that [sic] doesn't always indicate error; it can show uncharacteristic words coming from an unexpected source or in an unusual context when the words are either intended or doubtful. "X said that the world will blow up in 3 seconds [sic] ", "Y has declared that our greatest enemy is strawberries [sic] " If those are the words X or Y said, then of course it would be wrong to correct them silently.—— Shakescene (talk) 06:58, 22 September 2009 (UTC)

The article about [sic] says that it can be used in writing "to indicate that an incorrect or unusual spelling, phrase, punctuation, and/or other preceding quoted material has been reproduced verbatim from the quoted original and is not a transcription error."

-- Wavelength (talk) 07:11, 22 September 2009 (UTC)

FWIW, I recently came across an important use of "sic" in a published article about the Manson family murders. They wrote "Healter Skelter" at one crime scene, inspired by the Beatles' song "Helter Skelter". That's a famous spelling error. It'd be a problem if readers thought it was a typo or if helpful editors "fixed" the quotation. But I agree that trivial erros should simply be corrected silently.   Will Beback  talk  07:54, 22 September 2009 (UTC)

"[Sic]" considered harmful

Let's leave the MoS alone. "[Sic]" should be used only when a typo (or other weird construct) is important. Using "[sic]" to mark unimportant typos is pedantry so distracting that it harms the encyclopedia. It's typically better to correct the typo silently. Suppose we had an article about Kosovo that said: 'The BBC reported that the Belgrade Agreement "was not acceptible [sic] since it allowed only for the autonomy of Kosovo within Serbia"' (citing http://news.bbc.co.uk/2/hi/europe/195006.stm). Aagh! Just fix the spelling and omit the "[sic]": the article's supposed to be about Kosovo, not about the BBC's low-quality copyediting, and the "[sic]" is distracting the reader from the goal. Eubulides (talk) 07:36, 22 September 2009 (UTC)

Avoid [sic], per Eubulides. --Philcha (talk) 07:41, 22 September 2009 (UTC)

Agree with Eubulides. It's standard practice in journalism quietly to fix quotations where there's been an irrelevant typo or grammatical error. To focus on it is not only distracting, it risks making the source look stupid, and it makes the writer who's added "sic" look as though he's mocking the source. SlimVirgin ^{talk|contribs} 07:44, 22 September 2009 (UTC)

What SlimVirgin said. Hesperian 07:49, 22 September 2009 (UTC)

I didn't consider it harmful at all, I was just confused as to why MOS said that. Now I know :) Dabomb87 (talk) 12:47, 22 September 2009 (UTC)

Isn't it better to put the corrected letters in brackets, like it was done here? Pcap ping 09:33, 22 September 2009 (UTC)

Perhaps, but it's visually distracting. Looking at Eubulides's example, you would have 'The BBC reported that the Belgrade Agreement "was not accept[a]ble since it allowed only for the autonomy of Kosovo within Serbia"'. Dabomb87 (talk) 12:47, 22 September 2009 (UTC)

That particular edit is a bad example because it's not really a typo - the editor just changing the author's word choice to his own. Had it been a typo, I think silently correcting it would have been better. Christopher Parham (talk) 13:15, 22 September 2009 (UTC)

Indeed. In American English "ensure" can also be spelt "insure". And it cannot be a typo, as the E and I keys are nowhere near each other on the keyboard. I'd "silently correct" only if I were sure beyond reasonable doubt that it's a typo; here, I'm sure "almost" beyond reasonable doubt that it isn't. --___A. di M. 13:22, 22 September 2009 (UTC)

• So here's an example where I think [sic] needs to remain: on Okami#Wii Version, there's a quote that includes the line "As we are NOT at that point in the process yet, we are loathe [sic] to even mention any potential changes or enhancements for fear of disappointing the fans/media." There have been arguments on the talk page on whether the word should be "loathe" (to hate) or "loath" (unwilling to do something), as both readings "work" for this, and does change the intent of the statement. I can't see this being assured this is a typo so this case I would think that [sic] needs to remain. --MASEM (t) 13:30, 22 September 2009 (UTC)

At User talk:Noetica#Complications with ly, Noetica has said:

Generally, I would add "[sic]" only:

• where the reader would otherwise be uncertain whether the peculiarity was original or introduced; and
• where the peculiarity is noteworthy in some relevant way, and is not signalled by any other means for the reader's attention.

and: "The best editing is like invisible mending. Ars est celare artem." -- Wavelength (talk) 15:22, 22 September 2009 (UTC)

Per my example in the previous section, I think a third reason to use "sic" is if it was used by a secondary source for the quotation. If our source thought that "sic" was necessary and appropriate then we should retain it rather than making the correction ourselves.   Will Beback  talk  20:26, 22 September 2009 (UTC)

Will Beback, according to my understanding, what Noetica said refers to two conditions conjunctively, as indicated by the word and. (The word or would indicate two conditions disjunctively.) In other words, if both conditions are met simultaneously, then Noetica would add "[sic]". The first condition in isolation is covered by the principle of correcting silently unless there is a reason to do otherwise. Therefore, Noetica mentioned one compound reason and the reason which you would add is a second reason.

-- Wavelength (talk) 22:09, 23 September 2009 (UTC)

• I agree with Slim. Tony (talk) 08:51, 23 September 2009 (UTC)

Let's end the MoS/MOSNUM duplication

Here. Tony (talk) 08:50, 23 September 2009 (UTC)

Addendum: Time magazine, linked to from the current Signpost, asks questions about what has stunted the growth of WPians. Inter alia, it says: "for your edits to stick, you've got to learn to cite the complex laws of Wikipedia ... The foundation has been working to address some of these issues; for example, it is improving the site's antiquated, often incomprehensible editing interface."

This resonates with the feeling among some editors here, including me, that the MoS main page is too wordy and not sufficiantly easy to navigate around, and that the entire MoS infrastructure needs to be gradually rationalised. I rest my case for creating the trim version of MoS main page and for advocating a saner relationship between MoS main and MOSNUM. Tony (talk) 13:07, 23 September 2009 (UTC)

It does not seem to me that "antiquated, often incomprehensible editing interface" has anything to do with the MoS or how long or short it is. I think they're talking about the edit screen. Is there anything in the rest of the article that makes you think they're talking about the MoS? Darkfrog24 (talk) 13:48, 23 September 2009 (UTC)

No, I am drawing the association. Tony (talk) 14:35, 23 September 2009 (UTC)

The article discusses in general how too many rules and bureaucrats are driving people away, or failing to encourage new people. So I agree with Tony. Anything that can be done to streamline, clarify, summarize, and encourage sensible application of the rules, is heading in the right direction. SlimVirgin ^{talk|contribs} 19:46, 23 September 2009 (UTC)

Oh I can agree with that. It is very hard to get into the Wikibureaucracy. Darkfrog24 (talk) 21:40, 23 September 2009 (UTC)

What if the main MOS page provided just enough information to cover what 90-99% of the editors that typically edit will see (when/how to use wiki-markup for bold and italics, how to use refs, etc.), and then break out the most specific guidelines like MOSNUM where only a fraction of the editors will detail with the specifics? --MASEM (t) 00:52, 24 September 2009 (UTC)

The trimmed version has chopped out 60% of the word length. Additionally, I suppose it could be reorganised to put what editors need first, sequestering the less-needed information until later. But it may not be easy to do that. Tony (talk) 00:55, 24 September 2009 (UTC)

Masem, I don't think that's a good idea. This page might be big, but it's well-organized and easy to use. The rules are easier to find if they're all on one big page. When I was new to Wikipedia, I found it frustrating that I had to click through five and six and ten pages before I could find the information I needed.

That being said, Tony has come up with a shorter version of the MoS and I believe he's still taking comments on it. Care to add your voice? Darkfrog24 (talk) 01:37, 24 September 2009 (UTC)

An idea: markup for bad examples

How's this:

• Titles are generally nouns or noun phrases (Effects of the wild, not About the effects of the wild).

A template could be created specifically to mark up bad examples. --___A. di M. 00:32, 24 September 2009 (UTC)

Looks good: accessibility issues? I suppose they couldn't be bolded as well, could they? On my monitor, which has sharp colours, the supposedly "dark"green colour is not very dark, and the green examples are thus slightly harder to read. The red is much clearer (almost too clear), but will not be distinguishable for the 8% of males who have red–green colour blindness. Tony (talk) 02:24, 24 September 2009 (UTC)

The colour can be tweaked; and I'm not proposing to use it as the sole way of distinguishing good examples from bad ones; as long as each bad example is explicitly marked as such as they are now, doing this is not going to make things worse for anybody. (They could be bolded, but then you mightn't use it for examples which do involve boldface; but anyway they're going to be rare.) I'm also thinking of adding underline to correct examples to better contrast with strike-through in bad examples, such as

• Titles are generally nouns or noun phrases (Effects of the wild, not About the effects of the wild).

What do you think? --___A. di M. 09:55, 24 September 2009 (UTC)

I think it's better than the current system. Would you like to mark up one section as a demo? My concern now is that the struck through text is more difficult to read. However, without underlining or striking through, the addition of the red to the green would look very similar to what colour-blind people see now; that is, it would be no worse. For 96% of readers, it would be better.Tony (talk) 11:01, 24 September 2009 (UTC)

I have no objection to the use of color in this way. It doesn't look as aesthetically pleasing, but it looks like it could do some good. We ought to go without the strikethrough, though. Darkfrog24 (talk) 12:11, 24 September 2009 (UTC)

to me it would make no sense without the strikethrough - the red would make the examples of incorrectness more visible than the examples of correctness, and the colour-blind among us would see no difference at all. (as long as i'm here, i might as well note that i find the use of that other font for examples seriously unattractive; at the same time, that green isn't bright enough to distinguish it well from the surrounding text. so i'd be glad to see changes to the markup of good examples as well as bad examples.) Sssoul (talk) 12:44, 24 September 2009 (UTC)

They would have to be marked with "(Correct)" and "(Incorrect)" in addition to being differently colored, but the color might make a nice shorthand in addition to that. Darkfrog24 (talk) 14:25, 24 September 2009 (UTC)

Yes, I think we can consider the negative/positive context as satisfying the accessibility principle that colour alone should not provide the message. It does appear that people are not entirely satisfied with the green currently used. I must say that it looks to be of slightly lower resolution than the black on my monitor, although that may be an optical illusion. Pity there isn't a darker, sharper green; this is why I've bolded green-coloured text where I've used it in my tutorial exercises. Bold is probably too much here, though. Maybe a less dominating shade of red/brown might work for the negative. What is it now? Darkred?

In producing the trimmer version of the MoS, I was very pleased to have ADM's green template, despite its drawbacks. It avoids a forest of quotation marks or italics, both of which can create problems of formatting and logic. On the other hand, one has to be careful not to create a messy appearance on the page by using a font/colour/formatting that is too sharp or different. Torn between the two, we may be.

And the other quandary I remember when the green was developed ?last year was that whatever you come up with is likely to be dispalyed differently on the various monitors, browsers and platforms. Tony (talk) 14:39, 24 September 2009 (UTC)

You might consider using these symbols: ✓ (CHECK MARK), ✔ (HEAVY CHECK MARK), ✗ (BALLOT X), and ✘ (HEAVY BALLOT X).

Also, you might consider using white text on a black background to indicate incorrect usage.—Wavelength (talk) 14:57, 24 September 2009 (UTC)

That's a sexy idea, although I would use the particular shade of maroon found on the {{xt}} talk page. The new template could be located {{!xt}}. I say thumbs up with underlines and strikethroughs. Headbomb {^ταλκ_{κοντριβς} – WP Physics} 15:00, 24 September 2009 (UTC)

An idea I had is to display the underline/strike-through with a brighter colour than the text itself, maybe like this:

• Titles are generally nouns or noun phrases (✔ Effects of the wild, not ✘ About the effects of the wild).

Unfortunately, then, the strike-through goes in front of the text, making it even harder to read. Does anyone know a workaround to show the text in front of the strike? --___A. di M. 15:48, 24 September 2009 (UTC)

BTW, I think that this would eliminate the need for Georgia typeface, so that we'd not have messy numbers due to "lower-case digits". --___A. di M. 15:50, 24 September 2009 (UTC)

Here's a sandbox. Maybe the font colours might be made a tad darker, and the tick before correct examples could be removed, only keeping the cross before wrong ones. --___A. di M. 16:21, 24 September 2009 (UTC)

• I'm sorry: I just loathe the ticks and crosses aesthetically, both as one reads and as one looks synoptically at the page. Tony (talk) 16:16, 24 September 2009 (UTC)
  • The sandbox looks very overformatted and crowded. I need sunglasses. The idea at the top of this section seemed relatively simple, but has now ballooned. I say this knowing that ADM has put considerable work into the sandbox. Tony (talk) 16:45, 24 September 2009 (UTC)

Just a personal observation, but to me, when text has been struckthrough it has been negated, so striking a statement that something is wrong, indicates that it is not wrong, because the statement itself is wrong... if that makes any sense. Before reading through this section, I assumed that the text in question was part of an initial proposal that had subsequently been discarded (and struck, as you often see on talk pages) as the discussion progressed. I would suggest that, as a reader of the guideline, this part of the MOS is quite clear as it currently stands, and there is potential here to confuse things. Cheers, Miremare 17:08, 24 September 2009 (UTC)

Huh? I had this idea looking at my English grammar book, which also uses strike-through to show incorrect examples. It might look like we're negating that it's incorrect if the "not" were struck as well, but I don't think it can confuse anyone if it isn't... --___A. di M. 17:22, 24 September 2009 (UTC)

I couldn't say whether grammar books it in examples (but I bet there are plenty that don't too), I'm just saying how it appears to me as a reader. But regarding my second point, is there a reason that this needs to be changed from how it currently is? Miremare 23:24, 24 September 2009 (UTC)

• A. di M., I love what you are proposing and think there are many uses for it on Wikipedia. However, I hate that you propose to “change” an existing template. On my talk page, you wrote I've made a proposal to change the behaviour [of {{xt}} ].

  Once a template has been in use for a while, those editors who favor using it A) are typically quite pleased with exactly what they are using, and B) don’t want to see work that looks a certain way changed.

  That a template will be relatively stable is a reasonable expectation of template users. We often lock down our templates since doing so is a *pinky promise* to the community that editors will have stability in their articles. This practice encourages adoption and widespread use of templates. I think it is exceedingly bad form to make radical changes to templates that are already in use; doing so discourages the adoption of new templates. Now…

  If what you are proposing is an extension to the existing {xt} template, where additional pipes yield the results you are proposing, then that is fine; non-piped existing expressions of {xt} won’t be affected. However, if you are proposing to flat-change it’s behavior, then I strenuously object to that way of achieving your ends.

  Again, I’m not sure exactly how you intended /now intend to implement this. However, I would suggest that the best way to get this out there for us to all try and begin using would be to create a new template, perhaps {{xtu}}. The current {xt} template, by virtue of the fact that it changes only the face and color of text, is, IMO, the perfect technique in a MOS where style advise for using “quotes”, italics, bolding, and underlining of text. What you propose is somewhat of a Swiss Army knife that is really a different animal and should properly be a separate template for us to use, not a “change” to an existing template.

  Frankly, since it would be a different (new) template, why don’t you just go ahead and make it? There can’t possibly any harm in your doing so. We can all give it a whirl and even try it in those sections of MOS and MOSNUM where it would be most appropriate. Greg L (talk) 17:45, 24 September 2009 (UTC)

Hmmm... Yeah, making another template rather than changing the existing one would be a better idea. --___A. di M. 18:03, 24 September 2009 (UTC)

OK, I've created {{xtu}} and {{!xt}}. --___A. di M. 18:12, 24 September 2009 (UTC)

Before we go ahead with anything, be advised that the MoS advises against using red and green together because people with red-green color blindness can't see the difference. Darkfrog24 (talk) 17:56, 24 September 2009 (UTC)

As long as we don't rely on colour to show the difference, that's not a problem. Colour-blind readers will be able to tell incorrect examples from correct ones just like they do now. --___A. di M. 18:03, 24 September 2009 (UTC)

• Indeed, that’s OK. Darkfrog24, A. di M. is correct. Color is only assistive here for normal-sighted individuals. The thing to remember is that color alone—particularly red and green—must not be used to convey an important distinction, like “this is good” but “this is bad” or “GOOD/BAD”. This is prohibited under Wikipedia policy and it is never wise to do so. I remember a hydrogen sensor that used a dual-color LED, red and green, to denote “OK” and “leaking hydrogen”. Talk about “stupid.” The test lab’s director actually had red/green color blindness and he’d come to me and ask me to stare at the ceiling and tell him which sensor had triggered.

  A. di M.’s use of color is perfectly fine because the strike through and the big “X” (v.s. underlined and a checkmark) provides all the clues any of us would require, even if we had a old Classic Mac with a black & white screen. Color simply provides yet another quick clue for normal-sighted individuals to quickly distinguish “OK” and “Not OK.” This is similar to the chemistry wash bottles found in wet labs: the isopropanol wash bottle has a blue top, ethanol = orange top, methanol = green top, acetone = red top. There is a big difference between acetone and methanol (red/green). If one is color blind, you read the wording on the bottle. If you have normal color vision, you have both indicators, where color is the quicker one. To this day, whenever I think of “acetone”, I think “red.” It’s the same for cylinders of compressed hydrogen; they come in red cylinders. Oxygen (big difference) cylinders are green. Of course, both are labeled with their contents too. Color is simply assistive; same here. Greg L (talk) 18:11, 24 September 2009 (UTC)

• And even without check marks or underlines, “Effects of the wild, not About the effects of the wild” would not be any worse than “Effects of the wild, not About the effects of the wild” for colour-blind readers, and it'd be better for non-colour-blind ones. --___A. di M. 18:20, 24 September 2009 (UTC)

• Good point. Back to the razor-sharp logical basics, I see. Greg L (talk) 18:28, 24 September 2009 (UTC)

• Let me try this: Write I have been properly trained in the discipline of technical writing. and not I ain’t been learned good English. (looking at it in Preview here…)

  Can you make it so the underlining and strike through doesn’t apply to both the checkmark and the X? It would also be nice if they both left a plain space after the checkmark and X before the example text starts. Greg L (talk) 18:36, 24 September 2009 (UTC)

I do not feel that the strikethrough is a good idea. It makes things harder to read. Darkfrog24 (talk) 19:12, 24 September 2009 (UTC)

Is this maybe just a solution in search of a problem? Are there vast numbers of editors who can't comprehend the current "correct" and "incorrect" indicators throughout the style guide? I don't find the current format difficult to understand or read; am I alone on that? The green, red, symbols, all just seem to make things more difficult to read (especially so, I fear, for those of our editors with visual impairments). I respect the enthusiasm, but I don't see the net return on investment, unless I'm just missing it entirely? user:J aka justen (talk) 19:50, 24 September 2009 (UTC)

• I agree about the strikethru; it’s hard to read. Sometimes it’s quite important to scrutinize and read precisely what it is one is not supposed to do. Without the attendant change in typeface to Georgia, I’m not sure how one would setoff example text; that is, without resorting to italic text or “Putting ‘quoted’ text” in quotes, both of which screw up example text when our advise is about quoting text that itself contains quotations or is concerning italicizing. In my mind, simply writing…

“ Editors should write subsections as nouns, such as Hydro dams and not About hydro dams. ”

…is clear enough. While I see some potential with some of these design element concepts, it’s just not falling into place for me so far. Frankly, the big red s and nice green checkmarks seems to be a trick taken from the “…For Dummies” books, where they might have an icon of a policeman blowing his whistle and holding his baton in the air placed inside a sidebar containing cautionary advise. While sorta nice, I’m not seeing the need here if it compromises being able to read something. Greg L (talk) 20:46, 24 September 2009 (UTC)

Well, maybe you all are right. Doing all of that clutters the page. Re. "is there any problem right now": No, but I have seen strike-through used to mark incorrect examples in a grammar book and thought it was a good idea. (That book uses black text and a grey strike, so I attempted to have darker text than the strike in ✘!xt; unfortunately the strike goes in front of the text, making it even less legible. Does anyone know any way of having the text in front of the strike?) --___A. di M. 09:38, 26 September 2009 (UTC)

BTW, now I've redirected {{xtu}} to {{xt}}, and made {{!xt}} identical to {{xt}} except for the colour and the cross at the beginning. Now the sandbox looks like this. --___A. di M. 09:45, 26 September 2009 (UTC)

I think it should stay as it is for the moment. The green template is a significant improvement on what we had before; let's not ruin it with clutter. Tony (talk) 13:19, 26 September 2009 (UTC)

Internal consistency within the Wikipedia Manual of Style

The section on [and parentheses treats the term "brackets" as the preferred word for these symbols, not "parentheses". "These rules apply to both round brackets ( ( ) ), often called parentheses, and square brackets ( [ ] )." However, the section on [[10]] treats them in the opposite manner: "A pair of commas is often used for parenthetic material, and it interrupts the sentence less than parentheses (brackets) or dashes. Sometimes other punctuation can mask the need for a comma, especially the second in such a pair when parentheses are also used."Eric Kindig 00:53, 25 September 2009 (UTC)

This is similar to the discussion we had about whether to call them "full stops" or "periods." We're going to have to mention both terms in each section so that readers from all major varieties of English will understand. I feel no particular need to make sure that every section uses the same one first, but I have no objection either. Darkfrog24 (talk) 02:30, 25 September 2009 (UTC)

However, many editors are not so quick to see everything in terms of engvar. I really don't want to see unnecessary clutter, and I don't care whether parentheses or brackets, period or full-stop are mainly used, although one conversion somewhere might be ok. One assumes that WP editors have some pre-existing exposure to these terms. Tony (talk) 08:22, 25 September 2009 (UTC)

If the Manual is for the guidance of an average non-literary editor writing about sports or politics or popular music or a difficult technical subject or the history and geography of a particular town, one might presume that he or she has probably seen such words as parentheses. bracket, full stop and period, but that doesn't mean that he or she has the same understanding of their meaning, or even know that others a continent away might put a very different and specific meaning to such words. For those who want Wikipedia to assume some disembodied "encyclopedic tone", varying the order of conversion would indicate an impartial detachment from any one place; for those who want Wikipedia to reflect the diversity of its human editors, varying the order would show that. —— Shakescene (talk) 08:49, 25 September 2009 (UTC)

Needs to be minimised. Certainly not once per section. Tony (talk) 09:35, 25 September 2009 (UTC)

On the contrary, Tony, we must assume that users are not going to read the MoS from beginning to end but rather jump to the sections that they need. A high school graduate from the U.S. will have heard of "brackets" to mean {}, not (). Regardless of whether it's an American/British split, it is a split. We should mention both each time. Darkfrog24 (talk) 14:04, 25 September 2009 (UTC)

• The more length and clutter you insist on, the more you will drive users to the trimmed down version of the MoS. Tony (talk) 14:33, 25 September 2009 (UTC)

I have no objection to that. Let editors gravitate to the MoS that they find more useful. Evidence in practice is better than theoretical arguments in this case. Darkfrog24 (talk) 16:11, 25 September 2009 (UTC)

Thoughts on MOS structure

In reading through the above discussions, I started thinking about what the scope of the MOS should be, and how it reflects and is reflected in the actual usage if English on en.wikipedia. What came to mind is that there's really three types of English used on this encyclopedia.

The first is what I'll call common usage, that is the sort of basic English language usage rules that is inherent in (almost) all articles simply because they are written in English. They turn up in, oh say, 85+% of all articles. New (concientious) editors would likely want to quickly, and in one sitting, familiarize themselves with wikipedia-standard usage before editing.

The second are what I call general usages. These sorts of usages may not be as common, but they aren't tied to a particular area of knowledge, and therefore could turn up in just about any article. They're the sort of more esoteric rules that show a strong, highly-literate grasp of the English language. While their usage would reflect well upon the project, the average editor might not be as familiar with them, and a guide that allows that editor a place to look up how to use them in the wiki-context would be beneficial.

The third are what I call special interest usages. These are strongly tied to a specific area of knowledge, and are likely only to be encountered in articles directly related to those areas (i.e., writting classifiction names on articles about indidivual species, heraldic blazons in an article about a coat-of-arms). Since these areas can get very precise and specific, a separate page for each is best, as it not only gives an editor a specific place to look, it prevents a long technical section from overwhelming a more general-intrest guide.

As there are three tiers of editor's usages, there should be a three-tiered structure to the MOS as a whole.

For implementing this structure, Tony's Beginner's Guide, still under construction, looks to be the sort of read-through-in-one-sitting-to-get-a-general-idea-of-wiki-usage guide that fulfills the needs of the first tier of editor usage.

The current MOS is the look-up-a-less-common-wiki-usage guide. The current MOS shouldn't be trimmed of the content in the Beginner's Guide, though, as it's use of more numerous, and often more subtle, examples provides the greater guidence that an editor in the second tier of usage is likely to find useful.

The current MOS subpages fall into the third tier, fulfilling the role they were originally intended for. The hardest part is making sure that what's on a subpage doesn't conflict with what's on the other two tiers, such as was recently discovered between MOS main and MOSNUM. Vigilance and cross MOS discussion will be necessary.

Thougts?oknazevad (talk) 17:31, 25 September 2009 (UTC)

Would you give us an example of each of your three categories, Oknazevad? Darkfrog24 (talk) 18:05, 25 September 2009 (UTC)

Italics, for example. That titles of long-form creative works should be in italics is first tier, while the italicization of non-anglicizied foriegn loan words is more second tier. Finally, the italicization of binomial classification (as opposed to common name) of a species of animal would fall into the third tier.

Of course, the borders are fuzzy. The binomial classification might turn up in an article on the geographic feature or person the species is named after, but that's relatively uncommon. oknazevad (talk) 18:58, 25 September 2009 (UTC)

"Of course, the borders are fuzzy." That is an understatement. Finell (Talk) 16:40, 26 September 2009 (UTC)

Wikipedia:WikiProject Comics/Style guidance has been marked as part of the Manual of Style

Wikipedia:WikiProject Comics/Style guidance (edit | talk | history | links | watch | logs) has recently been edited to mark it as part of the Manual of Style. This is an automated notice of the change (more information). -- VeblenBot (talk) 02:01, 26 September 2009 (UTC)

OK, this raises the issue that came up earlier this summer with the MoS for French- and France-related articles. How do these fit in, and under what policy or procedure? There are (as I recall from my last count) around 100 pages in the MoS category, plus two dozen subpages of MoS like WP:MOSNUM and WP:ACCESS and nearly a hundred Naming Conventions. And why would "Anime and Manga" be subpages of MoS and this (at least at the moment) not?

By the way, if you look at the document, it is truly forbidding in the amount of detail it specifies in regard to a carefree children's diversion. Those of a certain age might remember Spin and Marty in the Mickey Mouse Club; one of them joked that the Triple R Ranch was named for "Riding, Roping and Reading old comic books." —— Shakescene (talk) 03:28, 26 September 2009 (UTC)

Not only the French—there was yet another one more recent, and I can't recall the name. They're growing like Topsy. We need an RfC to establish a board at WikiProject MOS, charged with auditing these Johnny Come Latelies and accepting or rejecting them. Tony (talk) 03:41, 26 September 2009 (UTC)

More importantly, we need to step back and behold the entire structure as it now exists. I doubt that there are more than fifty people on the entire planet who've read all 73 Naming Conventions, 101 Style Guidelines, 90 Style Guidelines of WikiProjects and dozens of subpages and offshoots. How does it all fit together? How much is reasonable for us to expect people to know about, let alone read or master? What is essential? When is specialized guidance necessary or desirable? What (as Tony implies) are the purpose and minimum standards for a Project Guideline? And looking at the entire corpus of guidelines that appear to govern the first sentence you start typing into Wikipedia, is it any wonder that ordinary non-vandal readers are a little hesitant (even gun-shy) about offering more than a tiny tweak, an external link, a spelling correction or a revised date?

Who here has read all of the above (and remember they're the "Key" policies and guidelines, presumably the essential ones)? Who's read all the sub-pages and all the non-key (but presumably printworthy) guidelines and policies?

As for Tony's specific point, there is this (apparently not quite adequate) wording on the top of Category:Wikipedia style guidelines:

Category:Wikipedia style guidelines is a top style guide category for pages related to standards for design and writing of Wikipedia documents. A page may be considered a style guide if it is intended to help keep the formatting, grammar and style of Wikipedia's articles consistent.
Official guidelines that may be categorized outside of the Manual of Style may be categorized here automatically by using {{Subcat guideline|style guideline}} at the top of the page.
Please post a message at WT:Manual of Style if you add or are thinking of adding a page to this category. We encourage people to write style guidelines on new topics, but group efforts work best. For a short introduction on how to go from proposal to style guideline, see Category:Wikipedia proposals and Wikipedia:Manual of Style.

—— Shakescene (talk) 05:48, 26 September 2009 (UTC)

Yes, we urgently need to correct this; at the moment, a free-for-all do-as-you-please is encouraged. Can anyone imagine the creation of featured articles and lists along the same lines? I note that suddenly, without consensus, comic-related featured nominations have to obey this self-elevated guide line. We have to bite the professional bullet, and we owe it to all who have contributed to the MoS to straighten out the jungle these 60 or so pages have become: otherwise, the community may well regard it with suspicious or even contempt. Apart from this, the wording is woolly and partly irrelevant.

Something like the following change to the wording Shakescene has quoted above would immediately create a meaningful role for WP:MOSCO, a Wikiproject that was started with high intentions of cleaning up the MoS mess, but which has been a ghost town ever since. MOSCO's role may subsequently evolve towards the systematic review of style-guide quality, overlap and consistency—which we desperately need. But first things first. This change in the wording at the style-guide category would be a start:

Category:Wikipedia style guidelines is a top style guide category for pages that set out standards to promote consistency in the formatting and language style of Wikipedia's articles. Pages are accepted as part of the Manual of Style only if a proposed application for MoS status gains consensus at the WikiProject_Manual_of_Style.
MoS pages are categorized as such here automatically by using {{Subcat guideline|style guideline}} at the top of the page.

Please be aware that some editors at outlying pages of the MoS may be sensitive to the notion that MoS main page might be trying to take them over. That is why the neutral page, MOSCO, was created, and why it should be the focal point for coordinating the clean-up.

It may need an RfC, or perhaps we can simply put the proposal at the category talk page and (one hopes) gain consensus there. Advertising the proposal at VP, Centralized discussions, and on the talk pages of all current MoS pages (respectable and shady alike), would help to marshal awareness of this step towards sanity in the MoS community.

Your thoughts? Tony (talk) 13:59, 26 September 2009 (UTC)

User:Emperor added a header at 16:00, 10 December 2008. User:Hiding updated the header at 22:19, 25 September 2009.

Maybe this discussion should involve those editors. -- Wavelength (talk) 16:13, 26 September 2009 (UTC)

(edit conflict) ¶ I think that we need to make a distinction between guidelines and naming conventions that came out of existing WikiProjects and an apparently-large number that arose from other places. In the former, such as the French/France-related guidelines, the Comic Book style guidelines (originally called "editorial guidance"), heraldic blazon, scientific nomenclature and Railroad Naming Conventions, there was (or should have been) at least some consensus among some of the editors of such articles about the need for regularization and convergence, and about the best ways to answer such a need. As Tony says, it's not only foolish for non-experts to trample ignorantly in such fields, but might be seen as disrespectful of the enthusiasts' work and of compromises that may only have been reached by careful work and difficult arguments.

¶ But what about all the other subpages and style guides? Abbreviations, capitalization (with which I often disagree strongly), external links, footnotes, dates & numbers, etc. Where did they come from? How was consensus reached? How do they equate with the specialized guides? (To repeat my earlier query: why is Anime & Manga an official MoS sub-page while Comic Books is not?) If the non-specialist guides are really part of the MoS, how much is reasonable and how much an unattainable degree of perfection that only a 'bot could presume to absorb (very imperfectly, of course, since 'bots can't read)? —— Shakescene (talk) 16:22, 26 September 2009 (UTC)

So the first hole to plug is that which allows unfettered self-promotion to MoS status from now on. If we can get that right, the way is open for MOSCO to systematically and probably rather gradually work its way through the more obvious, easier tasks of quality assessment, status, and rationalisation of those pages that have already slipped through. Tony (talk) 16:50, 26 September 2009 (UTC)