Wikipedia talk:Manual of Style (dates and numbers)/Archive 122

From Wikipedia, the free encyclopedia
Jump to: navigation, search

Poll: autoformatting and date linking

This is to let people know that there is only a day or so left on a poll. The poll is an attempt to end years of argument about autoformatting which has also led to a dispute about date linking. Your votes are welcome at: Wikipedia:Date formatting and linking poll. Regards Lightmouse (talk) 09:11, 11 April 2009 (UTC)

Editprotected request - add shortcut

{{editprotected}} Please add {{shortcut|WP:MOSBD}} to the Dates of birth and death section. This section is often referred to, and the shortcut would be convenient. The redirect at WP:MOSBD has already been created. MANdARAX  XAЯAbИAM 21:18, 12 April 2009 (UTC)

Done. --CapitalR (talk) 21:49, 12 April 2009 (UTC)

April 20 v 20 April

I would like to make clear somewhere in the guideline that British date formatting can be either April 20 or 20 April. When writing a date using only numbers, Brits always write the day first (20/04/2009), but when using the word for a month, April 20 is acceptable, as is 20 April.

The reason I'd like to clarify this is that I often see British editors arrive to change from April 20 to 20 April, often with edit summaries that imply they think the MoS requires it. It's usually done inconsistently; and I've seen editors change image names that contain dates (by mistake when they're changing other things), so that the images are no longer visible. Cleaning up after it can be time-consuming. Even though the MoS makes clear that people shouldn't be imposing style changes on stable articles, this misunderstanding about dates seems very persistent.

Does anyone mind if I clarify it to emphasize that either format is acceptable — when the word for the month is used, and not just numbers — whether in a British-related article or anywhere else? I'm posting below a sample of British and Irish newspapers that use the month/day format on their front pages, and in their articles.

England:

  • The Times [1]
  • The Sunday Times [2]
  • Daily Mail [3]
  • Mail on Sunday [4]

Scotland:

  • The Herald [5]
  • Sunday Herald [6]

Wales:

  • Swansea Evening Post [7]
  • Western Telegraph [8]

Northern Ireland:

  • Ulster Herald [9]

Republic of Ireland:

  • The Irish Times [10]
  • The Irish Independent [11]

SlimVirgin talk|contribs 15:10, 20 April 2009 (UTC)

i have the impression that quite a lot of people would mind the change you're proposing; i think people make the kind of edits you're talking about because they prefer DMY format, not just "because the MOS says so". the fact that some UK publications use the MDY format is not universally persuasive: -ize is widely used in UK publications too, but lots of Wikipedia editors still routinely change it to -ise in articles in UK English. Sssoul (talk) 17:32, 20 April 2009 (UTC)
While I understand this argument and it is internally consistent, I would be loath to see the long-standing consensus that US articles (other than military ones) use MMM-DD-YYYY, UK/Commonwealth ones use DD-MMM-YYYY, and articles on both or on neither use whatever the first major contributor used. While this is not a perfect situation, it has worked for a long time and produces articles which are at least consistent. A bit like WP:ENGVAR, it is a situation where something most people accept is imperfect is nonetheless a hard-won compromise that everyone can just about work with. Of course, if there is a strong consensus that we should change this style guide then I will accept it. --John (talk) 17:33, 20 April 2009 (UTC)
There is considerable evidence that Anzac usage is more artificially consistent than the UK, although in the opposite direction from the schoolroom consistency of American - partly as a reaction to American influence. In addition, April 20 was a standard British usage as recently as the Second World War.
There is no long-standing consensus (the present text is a few months old); there is a handful of half-educated dogmatists. What we should say is "English varies; learn to live with it" - but this would put the bullies out of work.
I would support text saying that either variety is acceptable, if we included WP:ENGVAR's cautions against edit-warring. Septentrionalis PMAnderson 17:59, 20 April 2009 (UTC)
Pmanderson, here is a revision from December 2007 with the language we are discussing in place as of 15 months ago. In Wiki-time, that is a very long time ago, and certainly more than "a few months". Slim, don't spoil your decent argument by inflating the numbers; the Sunday Herald and the Herald are sister papers, as are The Times and The Sunday Times, and also the Mail and the Mail on Sunday. Each pair undoubtedly uses the same style guide. Each therefore only counts as one example. Although I am not disputing this aspect of your argument, exaggerating like this doesn't strengthen it. --John (talk) 18:30, 20 April 2009 (UTC)
i agree that it undermines the argument to list those pairs of "sister papers" as if they were separate entities. Sssoul (talk) 19:31, 20 April 2009 (UTC)
  • The present wording derives from the archives marked D6 through D9 at the top of the page - last August and September. In the process, we observed that newspaper usage for all English speaking countries is inconsistent. Septentrionalis PMAnderson 21:08, 20 April 2009 (UTC)
My apologies if it looked like exaggeration; I just listed them as I came across them. I called up a list of newspapers in the different countries, and there did seem to be quite a few that used April 20, and not 20 April, many more than I've mentioned here. The Encyclopaedia Britannica also uses the April 20 format e.g. its article on Tony Blair says: "Tony Blair, prime minister of United Kingdom ... born May 6, 1953, Edinburgh, Scot."
The important point is that both formats are acceptable in the UK and Ireland, so there's no reason for us to ask editors to prefer one over the other. SlimVirgin talk|contribs 01:32, 21 April 2009 (UTC)
Other than a long-standing consensus. Let's turn this around; given that either format is intelligible to anyone, anywhere, what advantages would accrue to us by changing the MoS guidance? All newspapers have a style guide to ensure consistency within and between articles. Why would we not have? --John (talk) 01:42, 21 April 2009 (UTC)
Because it's being done on the grounds that 20 April is British, and therefore must be used in Britain-related articles. But it's not British. April 20 is British too. What forcing the 20 April format on people has done is lead to inconsistency within the encyclopedia in general (because most articles do use April 20 format), and inconsistency within articles, as people arrive to change some, but invariably not all, date formats, thinking the MoS requires it. If consistency across the project matters, we should remove the "20 April is British" advice, not least because it's factually incorrect.
John, you wrote above that the current advice is the result of a hard-won compromise. What were the issues that this was a compromise for, and do you remember roughly where/when that discussion took place? SlimVirgin talk|contribs 03:56, 21 April 2009 (UTC)
A few problems in all the above:
  • The Daily Mail example is not entirely correct - update dates in the form "21st April 2009" are seen in articles such as this (the day ordinal notwithstanding).
  • Encyclopædia Britannica, Inc. has been American since the early 20th Century; it is absurd to suggest this is relevant for UK standards.
  • Why the excessive focus on newspapers, and particularly their websites, which may be subject to inconsistent software settings and not necessarily what appears in print? Besides, WP:NOTNEWS.
  • Government consistently shows international/dmy date formats (apart from slight differences as day ordinals); academic usage also seems overwhelmingly international/dmy. Those are more compelling sources for style guidance than newspapers.
  • "... most articles do use April 20 format"? Wonder how fast it took to go through all 4,584,963 articles to confirm this? Even if so, factors such as ignorance of the MoS or systemic bias may come into play here.
Dl2000 (talk) 04:28, 21 April 2009 (UTC)
  • Testing Slim Virgin's claim is not difficult. As I write, 9648 articles use the string "April 20"; the comparable figure for "20 April" is 5763. This is about what one would expect, given that most native speakers of English are from countries where "April 20" is customary usage or those where custom is divided. It may be that other dates would give different results; but even "November 5"/"5 November" is 9465/5165.
  • When MOS disagrees with what WP actually does, ignorance of MOS is one possible explanation; but MOS' ignorance of Wikipedia is far more likely - and MOS' ignorance of the English language is always possible. Our guidelines are supposed to describe what editors in general do agree on; not what some small coterie thinks they ought to agree on. Septentrionalis PMAnderson 02:33, 22 April 2009 (UTC)
  • Well why didn’t you say so earlier, PMA? I shouldn’t look to MOS for guidance since it embodies the collective wisdom of a “small coterie” that has fouled it all up. Apparently, you posses all the writing wisdom and uniquely know what is best for Wikipedia. From hereon, I’ll skip the silly step of reading MOS and just ask you what is best. Greg L (talk) 03:29, 22 April 2009 (UTC)
    • You could do worse; but you would do better still to consider what Wikipedians at large actually agree on, and limit this waste of electrons to that. Septentrionalis PMAnderson 05:11, 22 April 2009 (UTC)
  • Formatting inconsistencies across WP are rife: there are articles which have a mix of dmy, mdy, ISO, and other variants. I hope to see a universal date format across the whole of WP which does not rely on autoformatting. The reason we got into autoformatting (and the dates case currently before Arbcom) is that we couldn't agree on a default style. I also believe that '5 November 2008', with one awkward comma less, is more elegant than 'November 5, 2008', but would be prepared to align to mdy dates if uniformity could be achieved. However, each time there is a discussion, it almost invariably goes back to "WP:ENGVAR works well enough." Well, I disagree - it's a total shambles. We have some US articles using international (dmy) style dates and UK/Oz articles which use mdy style dates/

    For WP not to have adopted a single format (or language variant) is seriously unprofessional. Admittedly, WP is not "professional" in the true sense, but that should not stop us taking a serious look at our style guide. It is important to note, as SV points out, that each publication has only one definitive style of dates and one language variant. Being the most widely visited information site on the Internet, WP is one of the few sites where there is a notable absence of prevailing style policy. I don't think the [decentralisation] suggestion is the right way to go. Ohconfucius (talk) 04:55, 22 April 2009 (UTC)

There is long-standing consensus against this nonsense; in particular, we are agreed not to favour any national form of English against any other.
Even if this were not true, making the bog of unsourced, semi-literate, tendentiousness that causes Wikipedia to be an unreliable source "professional" by tidying links and -our endings would be a joke, if it were not consumer fraud. Septentrionalis PMAnderson 05:11, 22 April 2009 (UTC)
  • "bog of unsourced, semi-literate, tendentiousness...". Yep, just like you said, WP is a joke. A little bit of order no doubt offends the anarchist in you... What's wrong with trying to make it a little less of one? Ohconfucius (talk) 06:39, 22 April 2009 (UTC)
  • I believe my position to be little different from Brion Vibber, so I would appreciate it if I was not quoted out of context in an unrelated discussion. Thank you for your attention. Ohconfucius (talk) 01:52, 23 April 2009 (UTC)

The best written English is almost always a reflection of the best spoken English (which is why Shakespeare, the 1611 Authorised/King James Version of the Bible, and the Book of Common Prayer, all written to be to be spoken aloud, are still such powerful influences—even among non-Christians and dramaphobes—four hundred years after they were written, despite major changes in vocabulary and sentence structure.) And most of us don't say "April twenty" or "Twenty April"; we say "April twentieth" and "the twentieth of April", partly because "Twenty" is a cardinal number and "twentieth" is an ordinal (modifying the unspoken "day of"). And in ordinary speech, most of us (I think) use both forms, "April twentieth" and "the twentieth of April", almost every week, if not every day or every hour. At times, we can even use both forms in the same sentence. If the technical reasons related to date-autoformatting, and the managing-editor's concern for saving space, no longer apply, then there's no reason for Wikipedia to disallow April 20th and 20th April, either, especially when they add to the clarity, flow, or style of a sentence or paragraph. ("Lincoln was shot on the 14th of April, 1865, and died on the 15th.") A foolish consistency is the hobgoblin of little minds... (Ralph Waldo Emerson) —— Shakescene (talk) 06:51, 22 April 2009 (UTC)

A reasonable argument for more flexibility; is there dispute - other than on the grounds above that there must be only one standard, as in Highlander? Those who wish to demote WP:ENGVAR should argue for that elsewhere. Septentrionalis PMAnderson 14:16, 22 April 2009 (UTC)
  • The role of any encyclopedia is to educate readers with minimal confusion and ensure they are well prepared to absorb information in their studies elsewhere on the subject. Our English-language version, en.Wikipedia, is read by a world-wide audience but fully one-quarter is our American audience. There is no single date format “April 20, 1985” or “20 April 1985” which is “better” than another and there is no reason to even try to standardize on one format project-wide.

    Keep it simple. And I certainly wouldn’t try to intertwine date formatting to the dialect of English some volunteer editor used for an article; that is a separate matter. I would propose we simply make the date format as natural as possible for the likely readership. Nothing more.

    Allow me to illustrate my point via example: Articles not closely associated with American topics such as Italy, Austria, Basilica di Saccargia and Kilogram have a pronounced non-American readership. Regardless of the dialect of English used by the editor of that article, we should be thinking foremost about our readers. So in articles on general or European subjects (articles clearly not associated with the U.S.), we would simply use Euro-style dates.

    Conversely, for articles on, or closely associated with American topics, such as Spokane River Centennial Trail; Coeur d'Alene, Idaho; American Revolution; and New York Yankees, they have a preponderance of American readers and the date format that is most natural for those readers is the American-style date.

    It is an utterly trivial matter to simply use the date format that is most natural for the likely readership. I would propose this simple guideline:

For articles on, or strongly associated with, the following countries and territories: The United States, American Samoa, Guam, Northern Mariana Islands, Puerto Rico, U.S. Virgin Islands, Wake Island, Republic of the Marshall Islands, Federated States of Micronesia, and Republic of Palau; editors should use the U.S.-style date format (“February 2, 2008”), otherwise, editors should use the international date format (“2 February 2008”) in articles.

The virtue of this guideline is it doesn’t require that editors go research what is an “American territory”; all the criteria they need is right there in the article. The guideline has a simple test: “what is the subject of the article about?”, which equates to “will it tend to have a preponderance of an American readership? If so, use American-style dates. Otherwise, use Euro-style dates. I see no reason to make it any more complex than this. Another virtue of this approach is we no longer need to discern who should be considered the “first major contributor” and “what dialect of English he/she used.” That doesn’t really matter; it’s not about us (editors), it’s about our readership and doing our best to ensure the articles read as naturally and well as possible.

I couldn’t care less if some Italian editor somewhere happen to have used British-dialect English and Euro-style dates when he first expanded Boston Red Socks from a stub. All any editor should have to do, if they encountered a mix of date styles in that article is consider (for all of about one second) what the subject is about and then (quickly) settle on the format most appropriate for the subject matter. Simple. And, yes, there will be an occasional grey-area article, like Gulf war: is that “strongly associated with America?” I think editors can work in good faith to resolve these issues if or when they arise. Greg L (talk) 19:39, 22 April 2009 (UTC)

There is always an easy solution to every human problem - neat, plausible and wrong. - Henry Louis Mencken. This is one of them; it is pure conjecture who reads what article. If it is true that Italians read Kilogram more than Americans (unproven; why should they? They know what a kg is.) and that they are annoyed by April 20 (unlikely), then they will change it; that's what collaborative editing is all about. Septentrionalis PMAnderson 20:10, 22 April 2009 (UTC)
  • Quoting you PMA: …it is pure conjecture who reads what article, well, it may be *pretty* to think as you do, but your logic ignores common sense and the laws of probabilities. It is obvious that articles like Spokane River Centennial Trail; Coeur d'Alene, Idaho; American Revolution; M1 Abrams; New York Yankees; Battle of Midway; and Boston Red Socks are all going to have a much heavier American readership than would articles not closely associated with America such as Basilica di Saccargia and atom. Your and Locke’s endless arguments about how you want to have your articles your way is tedious beyond measure. We want to explore solutions that don’t involve more ANIs and ArbComs and would like to leave the wikidrama behind. Some simple sensible arguments out of you would be most welcome. Greg L (talk) 22:23, 22 April 2009 (UTC)
    • It is reasonably likely that articles about places in Idaho will have been written by Americans, and use April 20. It is plausible, but no more, that they will be read by Americans (but there are foreign tourists, too; some from Canada). But so what? if those likelihoods pan out, they will have April 20 already, and we need not legislate. Septentrionalis PMAnderson 23:05, 22 April 2009 (UTC)
The proposal by Greg L would lead to the rather absurd ruling that day-month format should be used in an article related to Tijuana, Mexico, because it is not a US possession. Who would have more interest in such an article, a resident of California, Sussex, or New South Wales? Chris the speller (talk) 03:24, 23 April 2009 (UTC)

In addition to Chris' remark (quite sound), Greg's proposal would require the use of April 20 on purely American articles. Of the eight samples he names, six do use it without any requirement here; one does not mention any day. This is what one would expect, without any rule, and shows we don't need one. Battle of Midway, however, uses 20 April consistently; this appears to be a recent development among the American military, but is not something we should (or can) suppress without wider discussion.

But not only does it work poorly for American articles, it would work very badly for

Where do you get the idea that "the convention of 20 April has only come in since the Second World War" I have been doing a lot of work on Napoleonic treaties and they often use day month formats see Wikisource:Treaty of Paris (1815), which is based on Foreign Office and Hansard sources translating from French original sources. I am also reading a book "The Civil War in Worcestershire, 1642-1646, and the Scotch invasion by 1651", written in 1905 the author uses day month format in the text. The author also quotes Civil War primary sources and they seem to use day month. e.g. page 133 "Wednesday, the 12th of June, 1644, His Majesty marched out of Worcester...". (The quote is available via Google books in a number of sources). --PBS (talk) 22:45, 23 April 2009 (UTC) Just had a look at Wikisource:Treaty of Paris (1783) there they use "this third day of September, in the year of our Lord" PBS
12th of June (which the present text of MOSNUM forbids, and should not) is long-established. 12 June (which MOSNUM permits) is new; the contrast can be seen in Orwell's letters, for example; he dated June 12 and sometimes June 12th, but the notes, even in the 1968 edition of his collected works, use 12 June 1940. Septentrionalis PMAnderson 02:37, 24 April 2009 (UTC)
I see what you mean--however Hansard was using 20 November (Class C) in 1816 (along with other formats)--but just the two formats "20 April" and "April 20" was easily justified when we were defaulting to auto-formatting dates, as those were the ones the software supported, the justification after we abandoned that was (if my memory serves me well I have not looked in the archive) that this is the modern professional way. Personally I have no strong opinion on it, and would be happy with a laissez-faire policy on this issue, only prohibiting date formats that are confusing to readers e.g. neither dd/mm or mm/dd as in 9/11. --PBS (talk) 05:59, 24 April 2009 (UTC)

I support SV on this change it "to emphasize that either format is acceptable [in British articles] — when the word for the month is used, and not just numbers" with the usual proviso not to change from one majority usage to the other, without agreement on the talk page of the article. --PBS (talk) 22:19, 23 April 2009 (UTC)

Thanks, PBS. I can't see the harm in allowing 20th April or April 20th too, in addition to 20 April and April 20. All these variations are used in the UK and Ireland. The provision that styles shouldn't be changed without consensus is long-established, so there's no reason to suppose that giving people the option to use the date format of their choice will cause additional problems. On the contrary, it will hopefully prevent people from seeking to impose one style on otherwise stable articles. SlimVirgin talk|contribs 11:09, 24 April 2009 (UTC)
Are you actually in favor of allowing "20 April", "April 20", "20th April", "April 20th", "the 20th April", "April the 20th", "the 20th of April" all in the same article? If so that would make WP look awfully sloppy and unprofessional. If you mean to allow any one of those formats in one article, there will be edit wars. Oh, yes, edit wars, not to mention how tedious it will be for an editor to research who used what format first in an article that needs to be cleaned up. Chris the speller (talk) 15:56, 24 April 2009 (UTC)
She didn't say in the same article. Whether it would be unprofessional depends on the circumstances anyway; many military history articles are constructed around a sequence of days, and those should be consistent; but an article with two dates, one at each end, can use different formats for them, and nobody will notice. (20th of April is now usually employed for euphony or emphasis, and if it works, should be left alone.)
If you find an article that is noticeably inconsistent, post a notice on talk asking which format is preferable. If no one cares, no one cares, and you can pick one. Septentrionalis PMAnderson 17:59, 24 April 2009 (UTC)

Comment: SlimVirgin requested (through a note on my talk page) for the "Dates" section to be unprotected for editing. However, there does not appear to be any sort of consensus developing yet regarding changes to date formats, especially given the long-standing nature of the current guideline. Is there actually something ready for updating the MoS? Otherwise, I am inclined to leave the section protected for now. --Ckatzchatspy 18:05, 24 April 2009 (UTC)

I think a consensus is forming, CK. What I would like to do is make the edit, self-revert, and show people the diff to see if anyone objects. Is it normal practice to keep parts of guidelines long-term protected like this, or has there been an ongoing dispute that I'm not aware of? SlimVirgin talk|contribs 18:10, 24 April 2009 (UTC)
I see it was done because of the date-linking issue. [12] The edits I'd like to make are not related to date linking. SlimVirgin talk|contribs 18:15, 24 April 2009 (UTC)
Whoa! PMA and SlimVirgin want to deep-six the guidelines on date formats, and suddenly that's a consensus? "see if anyone objects" ... I strenuously object. Chris the speller (talk) 00:00, 25 April 2009 (UTC)
On what grounds?
I should repeat that Slim Virgin does not appear to be advocating within-article inconsistency; I think it often unimportant, as a minor part of this reform. Septentrionalis PMAnderson 00:41, 25 April 2009 (UTC)
(ec) i object too, on the grounds that editing the guidelines just to "see if anyone objects" is needless, since we're already discussing the issue.
what exactly is being proposed - is the "reform" supposed to affect only articles written in British English? or would it be more like any article, regardless of "national ties", can use either DMY or MDY format as long as the month is spelled out and "internal article consistency" is maintained? or is the proposal to allow other formats as well, and if so which ones? Sssoul (talk) 07:06, 25 April 2009 (UTC)
My proposal is to change the current MoS position that British English should be written as 20 April. My suggestion is that we make clear that both 20 April and April 20 are used in the UK, and that the MoS has no preference, except that articles be internally consistent. This applies only to when the month is written out. SlimVirgin talk|contribs 09:42, 25 April 2009 (UTC)

<--Two things. No, it's not normal that the guideline be locked on such a semi-permanent basis, but I don't see the point of changing it "to see if anyone objected" because we know what the status is; the issue could need an RfC to sort out. Ohconfucius (talk) 10:53, 25 April 2009 (UTC)

How much guidance is really necessary, anyway?

All right, then (although I won't argue to remove edit-protection or to replace Talk-Consensus-Edit with Bold-Revert-Discuss for dates in this Manual), I'll be the one to advocate against required date-format consistency within an article, because it's just not necessary, it serves no function and its absence does little harm. I expect that dates within a single section of an article will often harmonise/harmonize naturally; but in normal writing, I occasionally use more than one format even within a single paragraph. Only when it breaks the flow, or makes comparisons or relations between dates difficult or jerky, should inconsistency be discouraged. An analogy would be the rule of thumb that usually one should avoid writing "five oranges and 23 grapes"; similarly one should generally avoid "proposed on March 17, passed on May 28th and signed on 5 June". Such faulty parallelisms in style usually work themselves out in the normal process of editing and self-editing; I see little reason that date formats need be a special exception. And anything reasonable that reduces the huge, unmanageable and intrusive Manual of Style to something easier for the average editor to learn, grasp and remember allows everyone to concentrate needed attention on genuine sources of ambiguity, obscurity, mistake or confusion. —— Shakescene (talk) 07:01, 25 April 2009 (UTC)

Concur on all points, especially the last sentence. Septentrionalis PMAnderson 20:25, 25 April 2009 (UTC)
Baloney. Dates should be consistent unless in quotations. That's like saying it's OK to write both "honour" and "honor" in the same article. It's not; it's bad editing and bad form. Pick one format, stick with it. It's not written in a 1024-bit encrypted chinese dialect. Dates are not so special in that they should be exempted from guidelines that apply to everything else. Headbomb {ταλκκοντριβς – WP Physics} 21:06, 25 April 2009 (UTC)
Dates aren't so special that they need a special format: they're like the other ordinary words which every native English speaker and writer reshuffles all the time. Every reader should be able to recognise/recognize any of the common forms, and it's quite common for an ordinary reader to use more than one format, while spelling "hono[u]r" or "recogni[s/z]e" in a uniform way. What's really so jarring about seeing 20 April in one paragraph and May 5th in another? There can be faulty parallelisms with ordinary words, too, but they need no special rules because such problems are handled in the normal process of editing. Does it matter if I begin one sentence with a date , put another date in the middle of second sentence and end a third sentence with a different date? Do articles about May Day really need to put the first day of May into a single order or form?—— Shakescene (talk) 04:36, 26 April 2009 (UTC)
  • This is essentially free-for-all, and will result in some unholy mess, let alone edit warring. Who is it that keeps on saying MOSNUM is only a guideline? Surely, that has to be good enough for those who are pushing this? Ohconfucius (talk) 05:38, 26 April 2009 (UTC)
    • It has already resulted in an unholy mess, with Ohconfucius as chief mess-maker. We should acknowledge that anyone can edit implies that we will be inconsistent on such points. OC is welcome to set up a Wikipedia fork where this is not true, and the Manual of Style is enforced by any means necessary; but this is not it. Septentrionalis PMAnderson 17:26, 26 April 2009 (UTC)

All good stylebooks pick a date format and stick to it. Can anybody name one that doesn't? (By the way, if the person who used the word baloney above has an extra slice, I would like to have it to put between two halves of a biscuit/muffin for my lunch/supper.) Yours, GeorgeLouis (talk) 07:13, 26 April 2009 (UTC)

I think the purpose of the MOS should be minimizing the potential confusion and astonishment for the reader. A passage such as proposed on March 17, passed on May 28th and signed on 5 June definitely doesn't minimize it. As for having 20 April in a paragraph and May 5th in another, that might astonish a very small but nonzero number of readers; whereas this isn't a very serious problem, the cost of solving it by using the same format consistently in the article is close to zero (unlike, for example, avoiding the singular they, which would require us to arbitrarily pick a gender which can be seen as sexist, use awkward phrasing such as "he/she", or reword a sentence entirely). By the same token, if people seldom refer to Richard Feynman by his initials or to J. R. R. Tolkien by his first name, then it makes little sense to impose consistency and require that we should write either R. P. Feynman and J. R. R. Tolkien, or Richard Feynman and John Tolkien (this would be even truer if the names weren't in the same sentence), as either choice wouldn't be any less astonishing than just spelling names the way they are usually spelt, i.e. Richard Feynman and J. R. R. Tolkien. Nevertheless I have seen an "oppose" comment on a FAC based on something like that, where the names were in two different footnotes!
As for ordinal dates, I think the "full" form (the 27th of April/April the 27th) looks slightly old-fashioned nowadays, and it could be slightly confusing/astonishing to read them in a recently written article; as for the "semi-short" form (27th April/April 27th), I don't think it is any worse than the "short" form (27 April/April 27), although I think that mixing them in the same article (except in quotations) might be slightly confusing/astonishing.
As for PMA's argument about avoiding anachronistic language, it is an important principle but not the only one: whereas I wouldn't change "Persia" to "Iran" (or "Burma" to "Myanmar") in a sentence about something which happened before 1935 (or 1989), methinks thou agreest not that the "William Shakespeare" article hath to be written in 16th & 17th Century English. So there is a boundary line somewhere, but I'm not sure about which side of it writing 20 April in "George Orwell" would be on. --A. di M. (formerly Army1987) — Deeds, not words. 12:20, 27 April 2009 (UTC)
I would put the point about anachronism as advice; a writer should think about whether 27th of April is excessively florid or formal for his context. If he decides it's fine, then fine; there's no reason for MOS to review his decision. (If somebody objects, then they should discuss it, like other fine points of style.) Septentrionalis PMAnderson 14:41, 27 April 2009 (UTC)
Do we need a universal rule to decide between writing "Congress's intent", "Congressional intent" and "the intent of Congress"? Do we need to keep the same phrasing in different paragraphs? Or do we just let context and the normal kind of editing process decide? Although (living in the U.S. for half a century) I tend to put the day after the month in my own dates, I think using an ordinal, just like adding a comma, helps separate the elements for the eye in dates like "April 11, 1171", as well as reflecting spoken usage better. But while I might prefer that, I don't want to start a campaign to require it in all U.S.-style dates. —— Shakescene (talk) 21:56, 27 April 2009 (UTC)
Do we need a universal rule to decide between writing "Congress's intent", "Congressional intent" and "the intent of Congress"? Not by me; who says otherwise? Septentrionalis PMAnderson 01:32, 28 April 2009 (UTC)
So, then... (bada-boom!), do we need a uniform rule to decide between writing "April 27th", "the 27th of April" and "27 April"? —— Shakescene (talk) 08:02, 28 April 2009 (UTC)

"huge, unmanageable and intrusive Manual of Style". Schakescene, now we know you're a bedfellow of Anderson. He will stop at nothing to re-ignite his "let's dilute MoS" campaign, which has been going on for three years. If you resist, you're labelled a "bully" (only last week, I see) and "chief mess-maker", a personal attack by Anderson three days ago, above (any more and it will be a matter for ANI). And dragging in this concept of how poorly done by "the average editor" is by these pesky guidelines is the last, desperate resort of those who prefer that the site dispense with the authority, cohesion and readability of within-article consistency (and on other levels whole-site consistency).

"Average editors" do not write well and are likely to be pleased to have advice. My writing improved significantly after I regularly consulted MoS. On the dates, consistency within the main text is essential, and no, you never want to read a sentence that mixes them up, not just "generally". Equating grammatical choices such as the "Congress" example above with this concocted argument for editorial freedom over messing with date formats is nonsense. Tony (talk) 03:08, 29 April 2009 (UTC)

No, actually, I've never communicated with him. However, Shakescene, these outbursts of conspiracy theory are often Tony's last recourse before attempting rational conversation with threats to the "authority" he so dearly prizes; accusing new voices of being part of the conspiracy is a way to ignore them, and if it doesn't work, he actually discusses English (sometimes). Please don't give up. Septentrionalis PMAnderson 21:22, 29 April 2009 (UTC)
As for the substance, Tony now contends that the choice between "the intent of Congress" and "Congress's intent" is grammatical choice and the choice between "the 17th of April" and "April 17" is not. No citation for this absurdity is provided; but I would be interested to see if Tony can find one.
Yes, some editors do write so badly that even MOS, in its present unfortunate condition, can help them (I wouldn't boast of it, were I one of them); but clear, concise, and accurate advice would still be better. Septentrionalis PMAnderson 21:32, 29 April 2009 (UTC)

WP:ENGVAR and Indian number format (lakh/crore)

Per WP:ENGVAR, (1) "The English Wikipedia has no preference for any major national variety of the language"; (2) "An article on a topic that has strong ties to a particular English-speaking nation uses the appropriate variety of English for that nation". Now, this article says to use commas every three positions in writing large numbers, e.g. 1,234,567. However, Indian English, the convention is to use two positions for higher than one thousand, e.g. 12,34,567. I see a contradiction here -- WP:ENGVAR suggests that the Indian format is acceptable, especially on Indian articles; WP:MOSNUM here suggests that it is not acceptable. This contradiction needs to be resolved. I would sughttp://en.wikipedia.org/skins-1.5/common/images/button_math.pnggest, that it should be resolved by amending WP:MOSNUM to permit Indian number format when permitted by WP:ENGVAR. --SJK (talk) 05:39, 18 April 2009 (UTC)

I think this may be one of the rare cases where the national dialects of English are not mutually intelligible. That should be done with extreme caution, and only with the rupee - and even so, there will be misguided emendations if we don't have comments explaining the situation. Septentrionalis PMAnderson 22:01, 18 April 2009 (UTC)
Note that, nowadays, Indian English is much less of a major national variety than commonly believed: there are about 178,598 native English speakers (0.021% of the population) and about 90 million people able to speak English (less than 10%).
I'm a non-Asian British immigrant to the U.S., so I'm not being nationalistic, but 90 million is still a huge pool of theoretically-potential readers of Wikipedia, one-and-a-half times that of the entire U.K., and about three times that of Canada, California or Australasia. I don't know if that 90 million includes English-speakers among the several hundred million people in other parts of formerly British or British-influenced South Asia (Pakistan, Bangladesh/East Bengal, Ceylon/Sri Lanka, Burma/Myanmar, etc.) who may—or may not—also think in lakhs & crores. (And considering the relative number and length of articles here compared to Wikipedia in any South Asian language, we should expect even weak readers of English to come here sooner or later for many items.) So there's nothing necessarily odd or excessively fussy about treating lakhs and crores with the same attention that we give to furlongs, ounces, Imperial pints or U.S.-customary tons. [That means, of course, that we should try to avoid confusion both ways, without making prejudicial assumptions. I'm not any more in favour of confusing non-Asians than of disregarding what would make the most intuitive sense to South Asians.] —— Shakescene (talk) 04:57, 20 April 2009 (UTC)

I tend to agree with Anderson. Lakh and crore are not English so the number formatting based on them is not English either and would likely lead to confusion. JIMp talk·cont 11:24, 19 April 2009 (UTC)

Having travelled to India I was surprised with the words Lakh (a hundred thousand) and crore (ten million), but I soon found out what they meant. While doing the Helambu Trek in Nepal, I noticed a couple of adolescent boys coming towards me. It was obvious that they were up to something and when they got close to me they began acting the part of beggars, exaggeratedly asking me for a rupiah. I acted the beggar back to them, saying "Lakh rupiah!" They immediately fell about laughing.
Lakh and crore are so much a part of Indian English that I think that they should be accepted. However, as they are not much known outside the Indian subcontinent, a translation is needed. So, if we were reporting a sentence like this from The Hindu "The Food Ministry on Friday decided to release 2.5 lakh tonnes of sugar each for the month of April and May and one lakh tonne for June, the government said in a statement." [[13]] we would need to add a translation: "The Food Ministry on Friday decided to release 2.5 lakh (250,000) tonnes of sugar each for the month of April and May and one lakh (100,000) tonne for June, the government said in a statement." That's all that really needs to be done - unless you also need a conversion into short tons for American readers. Michael Glass (talk) 14:15, 20 April 2009 (UTC)
Definitely need conversion if used at all. Tony (talk) 02:53, 29 April 2009 (UTC)
I came across the crore when trying to mediate a dispute at Hogenakkal Falls. After a little research, I felt that the best solution was "The Japan Bank for International Cooperation has agreed to fund the Rs 1,340-crore project." That enables newcomers to understand the term, and also matches the text in the cited source, where we do not have the option of adding wikilinks. SHEFFIELDSTEELTALK 13:45, 30 April 2009 (UTC)

"Should" versus "must"

Many discussions on MOS pages revolve around questions of "should" vs. "must". It seems to me that unnecessary drama is introduced partly because we do not formally recognize these distinctions in the language of MOS documents, or in our discussions. For example, in the prior "April 20 v 20 April" discussion, I have the sense of broad common ground on the idea that uniformity of formats is generally a good thing, but there is divergence on whether exceptions are allowable (are they "jarring" or not?). In technical literature, the terms "may", "should" and "must" have strict meanings. Surely there are cases where it is valuable for a MOS document to provide style advice that does not constitute a hard and fast rule. Perhaps the following nit issue on date formatting is an example: If {{OldStyleDate}} is used in a biography infobox, is our guidance that all dates in the infobox use them? Should that have the status of "advice/suggestion/guidance" that contributors should use but may diverge from, or a style rule that must be used and that MOSNUM Cops should unleash their bots on? What I am suggesting here is that MOS formally employ the terms "should" and "must" (respectively) two distinguish between the two cases, and that proposals make explicit distinctions between the two. I propose the following 3 key words to be denoted in italics in MOS documents:

  • Must is an absolute requirement. Bots and MOS brownshirts may get out their jack boots and enforce without exception.
  • Should There may exist valid reasons in particular circumstances to ignore this particular item, but the full implications must be understood and carefully weighed before choosing a different course.
  • May is an item that is truly optional. This is not just for advisory material, but is a shield for minority practices used in some domains. That is, notice is made that articles are explicitly permitted to adopt a convention and that any ramifications of this eventuality must be taken into consideration by others. For instance, bot writers are not permitted to stomp on such items as a MOS error, since the MOS language permits such an option. Other articles may opt out for any reason- for example in cases where the guidance does not suit their domain.

-J JMesserly (talk) 20:26, 27 April 2009 (UTC)

MoS pages are guidelines (see Wikipedia:Requests for arbitration/jguk 2#Manual of Style), so words expressing absolute requirements such as "must", "shall", "always" and "never" do not belong in them. The only exceptions I can think of this are when either
  • something is required for technical reasons, and would otherwise render incorrectly (e.g. The italics markup must be outside the link markup, or the link will not work), or
  • there might potentially be legal reasons to do something (e.g. The quotations must be precise and exactly as in the source—doing otherwise might be a copyright violation, I guess, but IANAL; feel free to find a better example if this isn't good enough).
--A. di M. (formerly Army1987) — Deeds, not words. 21:09, 27 April 2009 (UTC)
  • Possibly also genuinely universally accepted pieces of grammar, on the order of All sentences end with punctuation. We state relatively few of these, largely because they don't become problems. Septentrionalis PMAnderson 21:38, 27 April 2009 (UTC)

I don't like uniformity for its own sake, or the huge multiplication of over-precise rules, but there is another area where I think general norms are very much needed. That's where there's a possibility of unreadability for medical or technical reasons (WP:ACCESS), or the possibility of confusion, error, obscurity or ambiguity for linguistic reasons. Often what is perfectly good (or even preferred) usage in ordinary writing or print presents some difficulty for sight-impaired or non-Anglophone readers. In fact I think this is the best justification for a Manual of Style. —— Shakescene (talk) 22:14, 27 April 2009 (UTC)

Do we need to copy WP:ACCESS or do we need to link to it? Septentrionalis PMAnderson 22:49, 27 April 2009 (UTC)
Many people have the view that there are items in MOS documents that do indeed have teeth and are not optional style guidelines that can be ignored. Even from A di M and Andersson what I hear being said is that Must as I defined it in certain rare instances is useful, so the point seems to be that Must should be used rarely, not that we should not use precise language to make it clear what is and is not being agreed to by the community. Note that my proposal on what these terms mean is not the key point of my post- those definitions come from common practice in RFCs which may well be too rigid for WP's needs. If the definitions should have more latitude, then that is separate matter entirely. The question is, shouldn't precise language me used so it is clear what the MOS states?-J JMesserly (talk) 23:52, 27 April 2009 (UTC)
But look at WP:ACCESS, however; it intentionally uses should consistently, and I suspect it gets better results that way. Septentrionalis PMAnderson 01:30, 28 April 2009 (UTC)
The proposal was not for a single keyword. Internet RFCs typically employ three or four keywords that have well defined meanings, whose use runs the range from optional advisory notes to proscriptive statements. These internet standards forming bodies use more than one for a reason. You are suggesting departing from this norm and that only one term be used, but have not described why this is more desirable. -J JMesserly (talk) 16:05, 28 April 2009 (UTC)
I didn't mean that we oughtn't distinguish absolute requirements from mere suggestions. I meant that there ought not be absolute requirements in the MoS (except in a very limited number of cases, for which I don't object to using the word must). --A. di M. (formerly Army1987) — Deeds, not words. 12:31, 29 April 2009 (UTC)

Should we split the more imperative parts from the more precatory or optional?

I'm beginning to feel more strongly that a smaller Manual of Style or Standard Wikipedia Format, limited to no more than 2 pages (as opposed to the present dozen or two-dozen separate pages), which can be easily read, learned, remembered and assimilated, should be segregated out from hints, suggestions and indications of the conventions and usage that are generally practised, preferred or found useful on Wikipedia. WP:Cheatsheet and WP:Accessibility are examples of the kind of thing I'm thinking of, but on a rather more comprehensive scale. Avoiding "9/6" because it's unclear whether it means September 6th or June 9th, indicating your calendar or thermometer when unclear, offering metric/customary/Imperial unit conversions whenever practical and observing the more-specific parts of WP:Accessibility would be examples of things (upon which we all agree) that should go into the smaller Manual. Other date formats, English/U.S. spelling (except for the microscopic fraction that presents genuine uncertainty), hyphenation, paragraphing, hectares vs km², internal consistency, parallel phrasing, etc. are the kinds of things that could go into a more general and discursive document ("Guide to Better Style" or "Common Wikipedia Usage") that could be broken into several different topic pages. —— Shakescene (talk) 06:32, 28 April 2009 (UTC)

Hey, would you guys like to split off onto a separate page for those who want to abolish taxes, centralised government, and a cohesive approach to language and formatting on the world's foremost information site, to let off steam? This page is for improving the guidelines, not diluting or dismantling them. No, our editors do not want advice fragmented into several places, just to promote an agenda for minimising the influence of that advice. Tony (talk) 03:17, 29 April 2009 (UTC)
    • Tony is confusing his urge to be Emperor of Wikipedia with having widespread approval for it, again; our editors, here, means the usual half-dozen bullies, few of whom edit. He has, however, managed a fairly good hyperbole; MOS is no more cohesive than it is literate or accurate. Septentrionalis PMAnderson 21:18, 29 April 2009 (UTC)
A two page long MOS? To deal with a topic as huge as style??? The CMOS (for example) is a thousand page brick and they have the advantage of being able to pick one format and exclude all others! Headbomb {ταλκκοντριβς – WP Physics} 05:31, 29 April 2009 (UTC)
Shakescene is on to something, reflecting the anguish which I imagine many feel when confronted with WP's labyrinthine style manual. Let us not scoff at the suggestion, nor at the person involved, I might add. GeorgeLouis (talk) 06:40, 29 April 2009 (UTC)
I'm not opposed to building something we'd called the Beginner's MOS or a MOS for dummies (or something to that effect), but slashing the entire MOS down to two page is simply ridiculous, unfeasible and undesirable. Headbomb {ταλκκοντριβς – WP Physics} 07:39, 29 April 2009 (UTC)
It might be possible to write a "Beginner's guide to numbers and dates", with links to the fuller sections here. It could emerge out of advice here from editors who know MOSNUM and its application (and non-application) in articles as to the parts on which editors most commonly need advice. A problem might be that it would need to be updated every so often to match changes to the full guide. A beginners' guide should also be clearly marked as not definitive. Another way of disseminating the skills and knowledge that are necessarily embodied in this style guide is through show-and-tell tutorials, although only some parts of MOSNUM are suitable for this. An example is here, dealing with just one aspect of MoS main page, the use of hyphens and dashes (although Anderson promptly panned it as "pure invention"). I can add to this late May to August (minus a few weeks here and there). Perhaps we need both approaches. Tony (talk) 11:07, 29 April 2009 (UTC)
Summary style. WP:MOS should contain beginner's guide to number and dates, beginner's guide to capital letters, beginner's guide to acronyms and abbreviations, etc.; MOS:NUM should be a more advanced guide to numbers and dates, MOS:CAPS should be a more advanced guide to capital letters, MOS:ABBR should be a more advanced guide to acronyms and abbreviations, and so on. (And, while I agree that the MOS couldn't be two pages long, it could definitely much shorter than it is. A statement such as

Use discretion when it comes to using scientific and engineering notation. Not all values need to be written in it (e.g., do not write the house was 1.25×102 y old, but rather the house was 125 years old in 2008 or simply the house was built in 1883).

is a textbook example of WP:BEANS, and there are many others.) --A. di M. (formerly Army1987) — Deeds, not words. 12:31, 29 April 2009 (UTC)
And how did that happen? Some rule-monger (and I'm not even checking to the diffs to see which of our persistent idiots it was) added to MOS always use scientific notation (or read it into language otherwise intended). This is a corrective. Septentrionalis PMAnderson 21:10, 29 April 2009 (UTC)
I see. I guess it was added to prevent lawyers from interpreting the previous sentence (When using either scientific or engineering notation in articles, consistency is preferred (e.g., do not write A 2.23×102 m2 region covered by 234.0×106 grains of sand)., which was clearly intended to mean "Do not mix scientific notation and engineering notation") as if it meant "Use either scientific notation or engineering notation for all values in the article". I agree that normative (or, in this case, pseudo-normative) language should be unambiguous; but the best way to achieve that IMO isn't instruction creep, it's calling a spade a spade—heck, if you freakin' mean "Avoid mixing scientific notation and engineering notation in the same article", freakin' write Avoid mixing scientific notation and engineering notation in the same article. What's so difficult with that? --A. di M. (formerly Army1987) — Deeds, not words. 10:46, 30 April 2009 (UTC)
What is so difficult with Avoid mixing scientific notation and engineering notation in the same article? It would require writing concisely and clearly, and needs actual thought to apply - although not much -, instead of mechanical checking. Septentrionalis PMAnderson 16:55, 30 April 2009 (UTC)
(Ahem.) So if the reason why the MoS doesn't say this is that (unlike I believed until now) I am so good a writer than what seemed to me the most obvious way of stating that is actually so, er..., brilliant that nobody had been able to come up with it until now, then, unless someone objects in the next 24 hours, I'm going to remove the "do not write the house was 1.25×102 y old" platitude and reword the previous bullet. --A. di M. (formerly Army1987) — Deeds, not words. 09:20, 1 May 2009 (UTC)

(unindent)Coincidently, I've started to write my own guide to writing (User:Headbomb/Writing). The section [[User:Headbomb/Writing#..._when_improving_style] might be of relevance.Headbomb {ταλκκοντριβς  – WP Physics} 19:38, 29 April 2009 (UTC)

    • One trusts you will avoid discussing adverbs; Coincidentally, please. Septentrionalis PMAnderson 21:18, 29 April 2009 (UTC)
Mere typos should not be an excuse for scoffing: We all make misteaks. Your freind, GeorgeLouis (talk) 00:15, 30 April 2009 (UTC)
Oh noes, a spelling misteak! The hue-manatee! Sack-riff-ice a verge-inn two sum dei-tea! Headbomb {ταλκκοντριβς – WP Physics} 03:31, 30 April 2009 (UTC)
If Headbomb's first piece of advice were not the (highly dubious) counsel to always use a spell-checker, it would not be worth remark. But it is. Septentrionalis PMAnderson 16:50, 30 April 2009 (UTC)
This is a talk page, not the Toronto Star's front page. Headbomb {ταλκκοντριβς – WP Physics} 16:56, 30 April 2009 (UTC)
FWIW, my online English–Italian dictionary lists has "coincidentally, coincidently" as a headword, and the definition doesn't mention any difference between the two.
  • Two pages sounds about right; it will cover most of the things that need to be said (two sentences from WP:ENGVAR belong in that list); and someone other than the participants in this discussion may actually read the two pages. Septentrionalis PMAnderson 21:05, 29 April 2009 (UTC)
    Even having one page (WP:MOS) which will be read by everyone and several subpages (MOS:NUM, MOS:CAPS, [[MOS:MATH], ...) which will be read by pedants would be a decent solution. But if either WP:MOS or any of its subpages is more than 32 kilobytes, that's a clear symptom of instruction creep. --A. di M. (formerly Army1987) — Deeds, not words. 10:46, 30 April 2009 (UTC)

Let's start tackling the issue. Unless someone objects in the next 24 hours, I'm going to replace:

  • When using either scientific or engineering notation in articles, consistency is preferred (e.g., do not write A 2.23×102 m2 region covered by 234.0×106 grains of sand).
  • Use discretion when it comes to using scientific and engineering notation. Not all values need to be written in it (e.g., do not write the house was 1.25×102 y old, but rather the house was 125 years old in 2008 or simply the house was built in 1883).
  • Sometimes it is useful to compare values with the same power of 10 (often in tables) and scientific or engineering notation might not be appropriate.


with:

  • It is preferred to avoid mixing scientific notation and engineering notation in the same article (e.g., do not write A 2.23×102 m2 region covered by 234.0×106 grains of sand).
    • Sometimes it is useful to compare values with the same power of 10 (often in tables) and scientific or engineering notation might not be appropriate.


It is 270 bytes less; it doesn't solve the problem of the bloat in the MoS but it's a good start. --A. di M. (formerly Army1987) — Deeds, not words. 09:20, 1 May 2009 (UTC)

I would still retain the non-example part of second bullet.Headbomb {ταλκκοντριβς – WP Physics} 16:39, 1 May 2009 (UTC)
I think there is a special template for examples, rather than putting them in italics. This avoids violating the advice given by BIPM and NIST to always write unit symbols in upright type. Also, engineering notation should be wikilinked. --Jc3s5h (talk) 17:24, 1 May 2009 (UTC)
{{xt}} produces the green examples now common. Beginning It is clearer to avoid...in the same context saves a couple more bytes, and. more important, gives a reason. Septentrionalis PMAnderson 00:57, 2 May 2009 (UTC)
(Reply to Headbomb) OK, I've retained it, but it is really necessary, or even useful? IMO, it is so obvious that anyone smart enough to understand scientific notation is also smart enough to realize that themself (or do they actually say "I'll be coming there in one-point-five times ten-to-the-first minutes" in real life?). It might even be taken as an insult to the reader's intelligence. --A. di M. (formerly Army1987) — Deeds, not words. 09:46, 2 May 2009 (UTC)

clarification

¶ Let me clarify what I was proposing. I think the nearly-mandatory part, which could be (for working purposes here) be called a Book (or Page) of Rules which everyone here agrees should almost always be followed, could be confined to one or two pages (or if you prefer, articles) of some indeterminate length below 100 kB (if it's hard for some users to load, then it defeats part of its purpose).

Those are things we all agree on, the things that are most important (e.g. something that's understandable by a non-Anglophone or sight-impaired person, as opposed to conventions about paragraphing, spelling or capitalisation/capitalization), and the things that a new editor most needs to be able to learn, remember and use.

We can still argue about how mandatory, precatory, advisory or optional the other parts need to be. When I see two dozen separate pages/articles on everything from dates to flags, on the one hand, they're very useful as a guide to style or general conventions, but on the other, the collection considered as a whole seems absurd and unmanageable as a battery of enforceable rules.

§ Although it might be quite natural to infer that Septentrionalis and I talk together about these issues because our views so often coincide, in fact I've never communicated directly with him/her outside open comments on pages like this. (Check our respective talk pages.) It's certainly nice to feel a little less lonely in my opinions, but I never knew that she/he was waging a three-year campaign (I've only been on Wikipedia for one.)
There's just a sincere difference in philosophy between editors who all want a better Wikipedia, rather than any kind of conspiracy on either side. I just feel that uniformity on a parallel with that attempted by The New York Times, the Clarendon Press, Microsoft or the Encyclopaedia Britannica is not only futile, unrealistic and unrealizable/unrealiseable, but that the disadvantages of such a degree of uniformity in a project like this (which is after all unlike any attempted before) generally outweigh its advantages. If the editors of a particular article think its contents box and headers are more readable with more capital letters (e.g. History and Geography, Theory and Practice), then they shouldn't be defeated by case-insensitive software and rules. Too much variation can be confusing, but some diversity serves the useful purpose of letting readers know how different are the sources from which Wikipedia springs. Too little consistency leaves a risk of letting Wikipedia resemble a random collection of pages, but the costs of imposing the kind of uniformity that requires setting a rule for every conceivable variation (hectares vs km², acres vs square miles vs square yards, April 27 vs April 27th vs 27 April) and thus having the truly intimidating quantity of material, discussion and archives involved so far seems too great to me, in terms of time, effort, mystification and most importantly good will (the date auto-formatting wars have been a veritable festival of bad temper and ill will). And I should be spending more time on creating, correcting and editing actual content than on pages like this.
But others honestly see the relative advantages and disadvantages differently. —— Shakescene (talk) 07:04, 2 May 2009 (UTC)
All I have done is noted the recurrence of the same complaints on these pages, which have repeatedly dismissed by the same voices. There is no conspiracy.
We are not OUP, nor are we the NYT. Both of those have paid staffs, and neither permits anyone to edit. Septentrionalis PMAnderson 21:54, 2 May 2009 (UTC)

"Over" versus "more than"

I came to the MOS for guidance on whether something should be said to have "over 75,000" supporters or "more than 75,000" supporters. Which is appropriate for Wikipedia? Either I overlooked an applicable guideline or none yet exists. Thanks. Benccc (talk) 20:03, 2 May 2009 (UTC)

Personally I prefer writing "more than" because it somehow seems more quickly understandable, but I'd say "over" is fine. Just my opinion and I haven't checked MoS, I doubt there is anything in there.--Goodmorningworld (talk) 21:45, 2 May 2009 (UTC)
I much prefer "more than", just as I prefer "the population is less than 300,000" to "the population is under 300,000". I would always change these when copy-editing, but "over" is not incorrect unless it's ambiguous, which is the case only sometimes ("more than" is never ambiguous, I think). Tony (talk) 09:53, 3 May 2009 (UTC)
Either is appropriate, but more than is IMO superior and clearer in meaning.Headbomb {ταλκκοντριβς – WP Physics} 00:58, 4 May 2009 (UTC)
Another option is "in excess of". It is a matter of individual choice as to which to use, as grammatically, all are correct. None are ambiguous unless the context is unclear. wjematherbigissue 21:11, 4 May 2009 (UTC)
There should be no guidance on this because competent choice depends on word flow, comparative sentence lengths etc. and is extremely hard to codify. --Hans Adler (talk) 23:30, 5 May 2009 (UTC)
Guidance doesn't preclude individuality. Rather, guidance gives a starting point for those who aren't sure (as was the case for the original poster). (For the record, I prefer "more than" to "over".)  HWV258  00:29, 6 May 2009 (UTC)
  • It's not something I'd put into MOSNUM or MoS main; but it's something I'd change if copy-editing, mostly. It's a low-level issue. Tony (talk) 09:41, 6 May 2009 (UTC)

Music date task force

I have attempted to set up a Music date task force in order to change instances of "present" to "onward" among other things in music related articles. Dark Prime (talk) 21:07, 4 May 2009 (UTC)

What? Tony (talk) 09:42, 6 May 2009 (UTC)

Merge talk

WP:MOSNUM has a tag saying…

It has been suggested that the section Symbols for variables in formulas from the page Manual of Style (text formatting) be merged into this page or section. (Discuss)

The “Discuss” link connects to here but I see no discussion thread dedicated to the subject. So I’ll weigh in with my 2¢

  • Oppose “variables and formulas” would seem to me to naturally be part of “numbers” (“dates and numbers”). This is where I’d look for the information. Greg L (talk) 22:30, 5 May 2009 (UTC)
  • I'd rather merge that section into WP:MOSMATH#Choice of type style, it seems more relevant there. --A. di M. (formerly Army1987) — Deeds, not words. 23:08, 5 May 2009 (UTC)
    • With a nice pointer link in MOSNUM? I had never heard of MOSMATH. As long as there are convenient links in all the obvious places. Greg L (talk) 05:05, 6 May 2009 (UTC)
      There is a list of MoS subpages on the right side of the lead section of each one of them, but if you didn't notice it until now, maybe links should be added in more prominent places, too. --A. di M. (formerly Army1987) — Deeds, not words. 09:16, 6 May 2009 (UTC)

Suggested edit to currency section

It currently reads: "Currency abbreviations that come before the number are [...] spaced if alphabetic (R 75)." For consistency with the section at the top about non-breaking spaces, why not recommend an nbsp explicitly here? So it could perhaps read: "Currency abbreviations that come before the number are [...] spaced, with a non-breaking space, if alphabetic (R&nbsp;75)." People coming here to check the situation with currencies, because they've written "R75" and are wondering if it would look better spaced and if the MoS has anything to say on the matter, are unlikely to read the nbsp section too, so it seems sensible to spell this point out briefly. Likewise for any other sections of the MoS where nbsp recommendations come into play and are not mentioned there already. Thanks. 79.64.170.147 (talk) 19:45, 6 May 2009 (UTC)

Sounds like a good idea to me. I think the same approach could apply to other symbols or abbreviations that are sometimes but not consistently associated with other punctuation such as dashes, hyphens, underlines, stops and commas. Users might look at the symbol or abbreviation in the Manual but not at the sometimes-associated punctuation. —— Shakescene (talk) 22:54, 7 May 2009 (UTC)
That sounds like a good idea. I'd suggest explicitly linking to WP:NBSP (or whatever it becomes, if the merge goes through), rather than only making the recommendation, though. That way, users might plausibly consider other options like {{nowrap}}. Also, should this be a full space or a thin space (& thinsp;)? TheFeds 07:47, 8 May 2009 (UTC)
It makes a lot of sense. Tony (talk) 11:49, 8 May 2009 (UTC)

Roman numerals to represent… century of birth?

Anyone seen anything like this? There are others too. I'm guessing it would be best to just say "(died 1760)" if we don't even a good guess as to the birth year. — CharlotteWebb 22:27, 9 May 2009 (UTC)

I've never seen it before. My guess is it's standard in Polish and whoever wrote this didn't suspect that there are some languages in which this is not done. Michael Hardy (talk) 23:05, 9 May 2009 (UTC)
O rly. Well there plenty of others too. The point is I'm wondering if the MOS should specifically deprecate this, and what would generally be best when the year is unknown (question marks, for example, are fairly popular too). — CharlotteWebb 23:36, 9 May 2009 (UTC)

If it's standard in Polish and not in English, then it shouldn't be done here, since this is English Wikipedia. Michael Hardy (talk) 00:45, 10 May 2009 (UTC)

No kidding. If I thought it was "standard" I wouldn't have brought it up. Once again what I'm asking is:
  1. Whether the MOS should specifically deprecate using Roman numerals for centuries (or writing years in base-16 or any other unforeseeable quirk).
  2. Whether I am correct to assume "(died 1760)" is the preferred form when the birth year isn't present.
CharlotteWebb 01:13, 10 May 2009 (UTC)
I agree with your raising this issue, Charlotte: thanks. It looks obstructive to English-speakers, who are quite unused to RNs used as such. I'd support a "should not be used without good reason" clause in the MOS. Tony (talk) 16:13, 10 May 2009 (UTC)
XVIIIth century would be a different question; it is English, but usually would be improved by conversion to eighteenth.Septentrionalis PMAnderson 21:33, 10 May 2009 (UTC)

(I've also never seen it.) As for the MOS, I hardly think it needs to be mentioned; if it's just something that Poles might occasionally write, then it will be rare and those who would do it will change after being told that it's not done in English, with or without MOS. As for the naming question, which this also is, wouldn't avoiding the years altogether, as we usually try to do, be the best solution? Stanisław Potocki (voivode of Poznań) would for instance be close to pl:Stanisław Potocki (wojewoda poznański) (in which article, incidentally, there's not a Roman numeral in sight; the birth year is "ur. 1698"). —JAOTC 21:58, 10 May 2009 (UTC)

Deprecate means disapprove earnestly or avoid the future use of (in computer jargon), for those who don't know. Yours for clarity and with all the best wishes for Mother's Day in the USA, GeorgeLouis (talk) 22:44, 10 May 2009 (UTC)

Conflict between MOSNUM, Template:convert and Template:RailGauge

In October last year, a user edited the MOSNUM section on Large Numbers to specify that only numbers greater than 10,000 should have comma separators, with an edit summary of "as per talk page". I can't find any discussion myself, though I may have overlooked it. Since then, this guideline page has been in conflict with the {{Convert}} template, which uses a comma separator for >=1,000 as seen in the following examples: 1,000 kilograms (2,200 lb), 1,000 metres (3,300 ft), 1,000 hectares (2,500 acres), etc.

Would it be possible to have either this page, or the template, edited so that this contradiction no longer exists? I don't have any concern as to which standard is adopted, although others may have stronger opinions. I've also posted at the talk page of the template to alert people to this. Regards, --DeLarge (talk) 12:04, 9 May 2009 (UTC)

I think it was me who originally proposed the wording which Physchim62 adopted. (I didn't follow the discussion for a couple of days, and I found that wording adopted on the page.) I personally prefer not to use commas for numbers less than 10,000, but it shouldn't be forbidden. On the other hand, four-digit "serial" numbers such as years and page numbers don't have commas (five-digit ones do, see 10,000 BC). I'm thinking of something like "In large numbers (i.e., in numbers greater than or equal to 10,000, and optionally for numbers between 1,000 and 10,000 which are not serial numbers such as years or page numbers) ..." but that's too wordy; can anyone find a more "slender" wording for this? —Preceding unsigned comment added by A. di M. (talkcontribs) 12:51, 9 May 2009 (UTC)
I think it would help to state something like, serial numbers or other numbers that are not normally broken up by a space, comma or period, should not be modified from their original form. Clearly whatever the guide is here, some numbers should not be formatted. Vegaswikian (talk) 18:02, 9 May 2009 (UTC)
I admit to being extremely confused as to why there should be any visual difference between four digit and five digit numbers...both have more than three digits, which I've always understood was the cutoff point for no commas. Shouldn't there be consistency in display...that is, if there are more than four digits in a number, commas should always be present every third digit? Of course, as Vegaswikian mentions, there may be times when any formatting is inappropriate, so that can be dealt with as well. Aside from that, I see the above edit introduced the allowance for spaces rather than commas. This might be worth discussing as well; personally, I think it should either be explicitly allowed, not allowed, or formatting specifically made dependent on rules such as WP:ENGVAR to avoid it showing up where confusion may occur (people from the U.S. have no experience with gaps in the place of commas, for instance, and it may be vice-versa elsewhere). Huntster (t@c) 18:11, 9 May 2009 (UTC)
It says "in scientific articles". A person who couldn't understand the numbers in "Planck units" because of the gaps between the digits would likely be unable to make head or tail of the article at all, even if commas were used. --A. di M. (formerly Army1987) — Deeds, not words. 10:13, 10 May 2009 (UTC)
It shouldn't matter what type of article it is used in, it should be consistently applied throughout. I don't understand what possible good reason there could be for having no indicator, comma or otherwise, in actual four digit numbers. This sets them off from dates, model numbers, etc, as CharlotteWebb mentions below. The indicator provides for less ambiguity, which is always a good thing. Huntster (t@c) 16:02, 10 May 2009 (UTC)
The "in scientific articles" refers to the use of gaps instead of commas, not to the delimitation of numbers less than ten thousands. --A. di M. (formerly Army1987) — Deeds, not words. 17:19, 10 May 2009 (UTC)

I believe the MOS already hints at this when it says:

Years are normally expressed in digits. Avoid inserting the words the year before the digits (1995, not the year 1995), unless the meaning would otherwise be unclear.

My conclusion is that a four digit number with no comma is generally assumed to be a year, if not a model number (e.g. 9500) or something else that is implicitly a proper noun, rather than a measurement of units (which functions more as an adjective). One (or at least I) would write: "2005 was 1,000 years later than 1005". — CharlotteWebb 22:20, 9 May 2009 (UTC)

I would avoid the comma as superfluous; I would also use no comma in "There are 5280 feet in a mile." There must be constructions which would be ambiguous if the words the year were omitted, but the closest I can come is "The length of the foot varied across Europe in 1000." Septentrionalis PMAnderson 21:30, 10 May 2009 (UTC)
Add The Rail gauge templates template:RailGauge 1, 1,435 mm (4 ft 8 12 in), 4 ft 8 12 in (1,435 mm) etc. etc. to this delemna. Peter Horn 01:35, 12 May 2009 (UTC)

Archive Box additions 121 & 122

Hi, I just added 121 and 122, the most recent (or current) archives to the Archive Box at the top of this page. But I'm not sure about the parallel Date archives which would (if appropriate) be D 14, D 15, and perhaps also D 16. Could someone who understands the archiving process better make the necessary adjustments? Thanks. —— Shakescene (talk) 09:27, 11 May 2009 (UTC)

Dynamic Dates syntax

I've always assumed, perhaps incorrectly, that those here who say the current Dynamic Dates syntax is "too complex" or "too complicated" meant the square brackets. Is it reasonable to assume that syntax such as {{#formatdate:December 3, 2009}} would be considered by those here to be even "more complicated" than [[December 3]], [[2009]]? I preferred using square brackets because it seemed simple and easier for a new editor to learn than some template-style invocation, and also because it avoided the tens of thousands of edits that would need to be made to change articles from the old style date format system to any new system. But my core question remains: would #formatdate be considered more complicated, as complicated or less complicated than the current square brackets system? —Locke Coletc 11:08, 2 May 2009 (UTC)

{{#formatdate:December 3, 2009}} is eight characters more than [[December 3]], [[2009]]. On the other hand, if that parser function were just called "d", {{#d:December 3, 2009}} would be one character less than [[December 3]], [[2009]].
Anyway, since 59% of those editors who care about this (enough to be bothered to add a non-"Neutral" vote to the last RFC) oppose autoformatting in principle, I can't see the point of discussing which syntax is more or less "complicated". --A. di M. (formerly Army1987) — Deeds, not words. 12:09, 2 May 2009 (UTC)
Well, there's one other reason to retain the linking syntax: if we made a Preference for turning links on or off, it would make far more sense to use the link syntax than something else. As for character usage, is that the final decider of "complexity" for those here? I preferred the linking syntax because it is already used for other non-link activities (categories, images). If moving to something like {{#d:Month Day, Year}} resolves that aspect of the dispute, I'd be fine with it, but obviously more input is needed.
As for this discussion, it started because of some comments made at the Proposed Decision for the arbitration case. As an aside, 59% is not (and should never be) considered consensus, not when such a large minority opposes. We don't vote here, and disputes are not resolved from a straight reading of a poll (unless it's nearly unanimous). And for what it's worth, the RFC prior to that indicated a majority support for "some form" of auto formatting (Dynamic Dates), so it's clear to me the community can go either way on this. That, to me, is as "no consensus" as it gets. —Locke Coletc 12:21, 2 May 2009 (UTC)
Um ... I don't think so. Where's the link to this claim of majority consensus for some form of DA? I can recall only RFCs that said no thanks very much. I can't believe you're still banging on about this, as though there were a problem to solve in the first place. <sigh> Tony (talk) 12:25, 2 May 2009 (UTC)
Tony you know well enough where it is. But here, again is the link. Please note that I started this discussion because it seemed at least one participant at the Proposed Decision welcomed discussion aimed at finding a real consensus on this. I note again that you're back at it with your usual antics already, feigning ignorance of prior discussions that you were involved with... *sigh* —Locke Coletc 12:39, 2 May 2009 (UTC)
"[S]uch a large minority oppose[d]" in that RfC, too (if it was actually a minority—I won't bother to count, but it's approximately two screenfuls of supports and two of oppositions). BTW, per WP:CCC only the most recent one is relevant. --A. di M. (formerly Army1987) — Deeds, not words. 12:57, 2 May 2009 (UTC)
And per WP:CON such a large amount of support in the most recent RFC cannot simply be ignored. Besides, another reason for this discussion is that many of those who opposed Dynamic Dates may have done so because of the syntax. If the syntax issue were resolved (via compromise here) then perhaps there would be more support for it. Polls do not end discussions. —Locke Coletc 13:04, 2 May 2009 (UTC)
FWIW, the #formatdate function can be called from a template, so say if it were {{d}}, it would appear as {{d|May 2, 2009}}, thus reducing the number of characters by 3 compared to bracket DA. --MASEM (t) 12:30, 2 May 2009 (UTC)
So that's still seven more characters than just leaving it alone. Tony (talk) 12:37, 2 May 2009 (UTC)
And here comes the "my way or the highway" response. Thank you for at least being formulaic in all this. —Locke Coletc 12:40, 2 May 2009 (UTC)
So it's strictly complicated by how many characters it is? I'm honestly interested, because I always thought the people saying it was complicated were saying that because of the the supposed learning curve of knowing to write dates in brackets (and so it logically follows, to me anyways, they would have an even harder time finding out about something like {{d|date}} or {{#formatdate:date}}). —Locke Coletc 12:39, 2 May 2009 (UTC)
The real problem is the imposition (without consensus) of an (unworkable) layer—on top of simplicity. The real point is that "[[December, 3]] [[2009]]" will always be more complex than "December, 3 2009". The atrocious square bracket formatting was introduced without community consensus and should be consigned to the dustbin as soon as possible. As an aside to the community, please don't be fooled by the seemingly simple reduction of "{{#d:December 3, 2009}}—the point being that as soon as that syntactical door is opened, a plethora of syntax-finesses will expand the complexity, whilst providing nebulous functional "improvements". We all saw many such suggestions during the debate.  HWV258  21:39, 2 May 2009 (UTC)
False; the community interested in such things was smaller in 2003, but autoformatting was a consensus compromise (one editor standing out); whether it was the best compromise is another question - and my answer is that it wasn't. Septentrionalis PMAnderson 21:59, 2 May 2009 (UTC)
Could you provide a link to the RfC that provided the wider WP editing community an opportunity to find consensus for the issue in 2003? I (and I'm sure many others) would like to find out why a majority of support for the atrocious square bracket system disappeared in the intervening years.  HWV258  22:17, 2 May 2009 (UTC)
Back then there wasn't such a large community that an RFC would have been totally necessary... as for why support has disappeared, probably because people think if they agree with you and your lot that maybe they'll stop hearing the same cries of "sea of blue" and "it's broken". —Locke Coletc 22:22, 2 May 2009 (UTC)
"...as for why support has disappeared..."—gracious of you to acknowledge this—thanks. Needless to say, I disagree with you over why "support" has disappeared (see the multitude of reasons given in the previous months; not to mention the "sea" of voting comments in the various RfCs).  HWV258  22:39, 2 May 2009 (UTC)
Lovely, give an inch, take a mile. You know what I meant; the fact remains, there is still a large amount of support for Dynamic Dates, enough that I think it's safe to call it "no consensus". This leaves us in the position of needing to find something the community will find more palatable, and I think that's fixing the problems with the system and resolving the issues of linking and formatting being joined. Will you help in finding a way forward that will bring the community together, instead of being divisive? —Locke Coletc 15:22, 3 May 2009 (UTC)
"Divisive"? How ridiculous, but I guess if you tell yourself such things often enough then you must eventually start to believe them. (To drag it away from your customary personalisation of the debate,) you continue to miss the point that is obvious to everyone else—namely that there is trivial support for the current system of auto-formatting. You persist in mixing-up "current system" with "a system". The current system should never have been implemented, and should now be dismantled in order to provide a "clean sheet". If, as you believe, some sort of system of auto-formatting/linking is needed then feel free to work as hard as you like in order to specify that—I'm not stopping you. Perhaps you could spend some time in completing the grid that I started (here). Good luck with that.  HWV258  23:05, 3 May 2009 (UTC)
I take it then that you refuse to try and compromise on this, and instead insist on having it your way over the cries of others? Please answer, will you participate in good faith on reaching a workable compromise, or do you intend to join Tony in trying to force your way? —Locke Coletc 22:22, 2 May 2009 (UTC)
Not up to me—the community has spoken. For the record, I have done more than my share in the attempt to reach a compromise (see my post here). Unfortunately for you, the community wasn't that keen to develop an alternative when push came to shove—when they had the chance. Too little, too late now—this is not the time to rearrange the nails and crooked bits of wood on the current "system". The current faulty and tragic system of linking and auto-formatting must now be dismantled (in line with community support). If others wish to specify an alternative system that can be put before the community for a RfC in order to determine real consensus on the specifics of a proposed system, then I'll be happy to assist in the judgement of that. I'm not going to prolong this (definitely against community desires now) in order to try and specify some sort of alternative—something that even the proponents of date auto-formatting and/or linking have failed miserably to do over the previous six months.  HWV258  22:53, 2 May 2009 (UTC)
Care to point me to the community consensus for dismantling Dynamic Dates? I've looked at all the RFCs and I don't see it. I see the community voicing concerns with the current system, as-is, but many of those concerns can be easily resolved. As to being tied to linking, that too can be resolved. So I ask again, will you participate in an effort to reach a compromise on Dynamic Dates, or is it your intent to follow Tony1's lead and cry foul at any attempt to reach something that really is consensus? —Locke Coletc 15:22, 3 May 2009 (UTC)
"...but many of those concerns can be easily resolved"—shakes head in wonder. If there is one thing that has been clearly demonstrated over the previous six months is that there is no magic or easy solution for fixing the current system of date linking/auto-formatting. If you feel it is so easy, please specify exactly how it should work so that the community can get a clear idea of what you (and others?) have in mind. Your other points have been covered many times and are not worth wasting further wiki-ink in replying.  HWV258  23:59, 3 May 2009 (UTC)
I don't recall you offering to fix the problems inherent in Dynamic Dates. In fact, once the RfC started, you and CKatz appeared to head off into general principles in the RfC rather than specific fixes, and later embracing Werdna's {{#formatnum}}. I'm still not seeing the answer and the general support you said was there. Anyway, the conversation has now moved on - nobody is talking about switching off Dynamic Dates anymore, so why can we not move on too? Ohconfucius (talk) 04:00, 4 May 2009 (UTC)
  • I think HWV would be keen to help if he thought there was a problem to fix. I appreciate (even admire) your dedication, and am regarding it as in good faith; but I don't think there's the support you want. You are upset—and apparently angry with me in particular. I'm sorry that this is the case. You have a lot of talent, Locke Cole, and I think we'd all be delighted if you steered it towards other ways of improving the encyclopedia. Why won't you let us benefit from your talent? Tony (talk) 15:56, 3 May 2009 (UTC)

Actually, it appears that linking the day of the year and the year in full dates was already customary before autoformatting was first introduced (for some reason which is beyond me), and it's still customary in many other Wikipedias (e.g. the Italian one) where autoformatting was never enabled (as there is only one date format in common use in Italian, and in most languages other than English). See e.g. /vote: a vote about which format to use, with no mention of unlinked dates, and only one option supporting autoformatting. For examples of its use, see [14], [15], [16], etc. They just picked a syntax which happened to be already in use around dates. I think the first proposal of autoformatting was in /Archive 1#Date markup. --A. di M. (formerly Army1987) — Deeds, not words. 11:20, 3 May 2009 (UTC)

I do have to point out once again that one of the compelling reasons advanced for immediate automated delinking of all dates was that it was just too damn easy for other editors to adopt the same format they already saw in articles, and thus date-linking was bound to persist. I can't square this with any suggestions that the square-bracket method was in any way complicated. It was just plain simple, and the suggested use of a leading colon in the link to force blueing was also the same (though perhaps exactly converse) to using a leading colon to explicitly insert a [[Category:category link]] rather than a category itself. The rest of the discussion was down to viewpoints and jigging up the software. Franamax (talk) 01:10, 4 May 2009 (UTC)

  • One thing that is clear is that the community thinks it should be a (very) rare date that is linked in regular editorial content. So let’s assume that for the purposes of this discussion, linking is wholly separated from autoformatting. Let’s examine then, a fundamental issue about autoformatting: it is intended to give special benefits (a custom date format) for a very, very select group of individuals:
  1. They are registered editors…
  2. who have logged in and stayed logged-in for the last re-logged in every 30 days ,
  3. who have set their date preferences to something other than “No preference”.
In the mean time, everyone else, particularly I.P. users, who are probably—literally—99.99% of our readership, just get a plain-text, default format. The community clearly stated in the Date autoformatting RfC that it adheres to the fundamental principle that all editors should see the same article content as regular I.P. users. The latest RfC on autoformatting by Ryan doesn’t invalidate that earlier RfC; it supplements it. Taking all four RfCs in total (Tony’s, Masem’s, Greg’s, and Ryan’s), and integrating all their teachings makes it clear that there is insufficient community support to warrant wasting one’s time by continually pitching these ideas.

I imagine the parties who have been, uhmm… *promoting* the idea of autoformatting might well be enjoined by ArbCom for one whole year from continuing to persist at this. Were I such an advocate, I’d be hanging really low right now. Greg L (talk) 02:01, 4 May 2009 (UTC)

I can't get the "who have logged in and stayed logged-in for the last 30 days" (why? don't you get preferences as soon as you create an account?); for the rest you're correct. "Masem’s" above links to the wrong place, but I won't bother to search for where that RfC is actually archived. --A. di M. (formerly Army1987) — Deeds, not words. 09:17, 4 May 2009 (UTC)
  • When you log in, you can check a check-box that is labeled with “Remember me (up to 30 days)”. Doesn’t that mean that we all have to re-register? Doesn’t that mean that inactive editors seldom notice that their username no longer appears at top (and wouldn’t even care)? That’s the way I’ve always figured it. Nevertheless, my wording was misleading. I’ve revised it to better reflect what I meant. Greg L (talk) 19:45, 4 May 2009 (UTC)
Greg please stop citing your RFC. Barely anyone participated in it (and with good reason), it certainly doesn't represent the view of the wider community. You're also getting ahead of yourself: as you can see from the section title, this section is attempting (and so far failing) to discuss the issues with the syntax that were raised. One step at a time, yes? It should be clear from my prior comments that I support attempting to resolve the issues you have with anonymous/IP users being unable to benefit from the auto formatting system, so let's just clear this one small hurdle first, then continue trying to parse out the other issues. —Locke Coletc 11:26, 6 May 2009 (UTC)

I've also been a vocal opponent of autoformatting, mostly because of the added complexity it adds for new users. Most new users don't understand that they have to put special syntax around dates. They usually figure out pretty quickly that you need special syntax to create wikilinks. In my opinion, telling those users that yes, you use the same double brackets around dates, but they don't make links, they do something else, is going to be highly confusing. Something like formatdate makes it much more clear what the purpose of the syntax is (that is good!), but it is yet another piece of syntax that we are requiring users to learn (that is bad!), and (the other half of my objection) only to provide a limited advantage to a very, very small population (why?). For the vast majority of readers, and likely editors, that syntax, whatever it may be, is useless. Even if we make the default be the same for everyone, most people (judging by the % of users with preferences set) don't care at all. In my opinion, then, any special syntax for date autoformatting is making things overly and unnecessarily complex. Karanacs (talk) 20:37, 4 May 2009 (UTC)

I couldn’t have said it better Karanacs. Moreover, integrating all the teachings of the four RfCs makes it clear that yours is a common sentiment of the Wikipedian community. Greg L (talk) 00:19, 5 May 2009 (UTC)
The reason for using the brackets is that, it was my hope, we could provide a preference for linking dates (choices being "always", "sometimes" (default) and "never"), which would address the concerns of those of us who actually like date links. In that regard, using square brackets makes sense because the dates may be linked and auto formatted (or any combination of one or the other). The reason for trying this is that it's clear we don't have support to entirely ditch auto formatting (something like 57% opposed in the last RFC, to 43% supporting), but perhaps some compromise system where we address as many of the concerns as we reasonably can and resolve as many of the connected issues as we can will find a consensus (again, a compromise; nobody gets their way exactly as they want it, but we all get a better encyclopedia for our trouble). —Locke Coletc 11:23, 6 May 2009 (UTC)
Just why you'd bother perplexes me. Not worried about color/our? Tony (talk) 04:11, 16 May 2009 (UTC)
No, not at all. —Locke Coletc 05:26, 16 May 2009 (UTC)

And so it goes, as usual

Rather than a useful discussion where opinions are respected and a compromise is considered, we have the usual suspects appearing, claiming nothing here need to be done, and carrying on as if things will go their way. This is not how Wikipedia should work, ever. —Locke Coletc 02:49, 9 May 2009 (UTC)

Jargon

It would be nice if some of the more difficult parts of this page were simplified. What, for example, is a significand? And why should we care? Yours very sincerely, GeorgeLouis (talk)

The mantissa (see the link on significand, or logarithm: the 1.2345 in 1.2345 * 109. What would you call it? Septentrionalis PMAnderson 05:14, 13 May 2009 (UTC)
The 1.2345 is the signifcand, 109 (or simply 9) is the exponent, assuming of course common logs (i.e. log10) where 10 is the base, and not Naperian logarithms or binary logarithms or whatever. I think the whole problem here is to remember who are we writing for? I think a technical article can handle saying it's 100 parsecs or whatever, wheras a more general argument needs to spell it out a bit more.
I still have my slide rule and can do multiplication and division quicker on that than on a n electronic calculator. You can do lots of clever tricks on them. SimonTrew (talk) 16:29, 13 May 2009 (UTC)
Remember, this is not requiring that we use the word significand; it is stating a technical limitation on the template (which is unlikely to matter, but should be documented somewhere). Septentrionalis PMAnderson 18:27, 13 May 2009 (UTC)

"The mantissa (see the link on significand, or logarithm: the 1.2345 in 1.2345 * 109. What would you call it?" Well, I would call it jargon that is not comprehensible to most readers of this page. Sincerely, still your friend, GeorgeLouis (talk) 20:50, 13 May 2009 (UTC)

  • Why the concern over “significand”? It doesn’t matter. When I went to school, I learned it as the “mantissa”. But when I first started working on the {val} template and had to write about it here and document it, I found that the mathematical community apparently got more specific over the last few decades (like no longer calling it “atomic weight” but “atomic mass” instead) and now call it “significand.” In fact, mantissa now just goes to a disambiguation page, so I’m not about to call it that. I don’t even understand the basis of your objection; the first use of the term on WP:MOSNUM is linked so it’s easy enough to learn about this if one doesn’t recognize the term. What’s good enough for our readers ought to be good enough for us. This is, after all, an encyclopedia; we’re all here to learn. Greg L (talk) 03:13, 14 May 2009 (UTC)
  • If you don't know what a significand is, chances are you don't need to know what it is. The term is linked for those who don't know what it means. Never heard the term mantissa before, but then again my native tongue is French, so that may be why. Headbomb {ταλκκοντριβς – WP Physics} 03:27, 14 May 2009 (UTC)
Yeah we used to call it mantissa when I was growing up. I agree it doesn't matter, except perhaps that the standard texbook formula for quadratic equations, the "m" stands for mantissa (I forget where this was raised but someone asked it not so long ago, why choose "m"). It's the same with the names of axes I guess, they were the abscissa and I forget the other now we just call them x and y. (and z.... etc).
I am not sure being Francophone would have much to do with it, probably as Greg says just what field you're in. The symbols example for Boolean operators vary between computing, electronics, and pure maths, rather unnecessarily really but we're kinda stuck with it.
I think the point we're all pretty much agreeing on here is trying to think "who is our reader?" Of course anyone can read any article but if they come to this page what are they likely to be looking for and how can we help them? At the risk of going overlinking, I think it's best kept because someone looking up this article is probably not as such interested in what a significand (or mantissa or exponent) is, they are editing an article and want to know the preferred way to express it. They are, I would hazard, editors of articles rather than general readers. As such they just want to see how to format a number, not to learn what a significand is-- if they wanted that they could have typed "significand" into their search box.
That's rather a muddled argument. SimonTrew (talk) 15:13, 15 May 2009 (UTC)
Ah I see the point I was aiming for. That is: when you're editing an article the more help you can get the better, and really you're concentrating on the article not these niceties (depending on the state of the article of course). So, the more links etc the meta-articles like MOS and stuff can give, the better, you want to be able to find what you're looking for very quickly. Providing it doesn't seem like Blackpool illuminations I'd thus say link, link link. One of the hardest things I find as a fairly regular editor is simply to find out how to do something: there is kinda a bias that you can only find out how to do it if you already know how to do it.
This example in itself is of course trivial. templates are the worst. If you don't know they exist it's pretty much a game of roulette to try to find one; and even if you do they may have entirely inadequate documentation. The convert template, as a counter-example, is by nature a but tricky to use, but it's usable partly because fairly well (though not perfectly) documented. SimonTrew (talk) 15:21, 15 May 2009 (UTC)

Is there an existing template for FX conversions?

If not, I plan on creating one. Please check out this discussion at the template talk page.    7   talk Δ |   05:43, 14 May 2009 (UTC)

BC/AD, BCE/CE

After considerable discussion, the MOS wording was changed to to the version that I have copyedited into MOSNUM. For information on the discussion go to [17] Michael Glass (talk) 22:50, 8 May 2009 (UTC)

I've added in this bullet point: Common sense should be used to avoid possible offence (e.g., In Jewish, Islamic or Buddhist articles). Refer to Common Era for information on the usage. (Taken verbatim from wording proposed by Michael Glass on Talk page referenced above.) It is necessary to have this in order to avoid an atrocious situation such as on the German Wikipedia, where policy is to use the † symbol for date of death, even in articles on Jewish subjects.--Goodmorningworld (talk) 23:13, 8 May 2009 (UTC)

Good move. Michael Glass (talk) 00:07, 9 May 2009 (UTC)

Can somebody please put a link to the Section in question? Thank you. Sincerely, GeorgeLouis (talk) 00:16, 9 May 2009 (UTC)
Here you go.--Goodmorningworld (talk) 00:24, 9 May 2009 (UTC)
Makes sense. GeorgeLouis (talk) 00:28, 9 May 2009 (UTC)

I have reverted this addition because it is badly written (we do not have "Jewish" articles), and it is not our goal to avoid giving offence (see WP:NOTCENSORED). This would invite endless edit wars about whether the one or the other system should be used, and would set an odd precedent.  Sandstein  20:29, 9 May 2009 (UTC)

Nonsense. The point of my insertion was to prevent edit wars, not invite them - which is exactly what Sandstein's revert accomplishes. The remark about "Jewish articles" is incomprehensible unless intended as a straw man argument of the lowest order (no one here said anything about "Jewish articles"). (Side note: admin Sandstein improperly removed rollback rights from me a short while ago, we will see if this grows into a pattern of harrassment on his part.)--Goodmorningworld (talk) 07:36, 11 May 2009 (UTC)
What I meant by that remark is that we have articles about Jewish subjects but not "Jewish articles" (as in your proposed addition); our articles are not human beings and have no religion or ethnicity.  Sandstein  19:13, 12 May 2009 (UTC)
Just wanted to add that I completely back up Sandstein in this matter. Wikipedia is not censored and does not give guidelines to avoiding offense. The only reason both BC and BCE are acceptable is because they're both widely notable. Would you also suggest Wikipedia use the Islamic calendar alongside the Christian calendar on each article to avoid offending those who follow solely the Islamic calendar?. — `CRAZY`(lN)`SANE` 05:24, 15 May 2009 (UTC)
See my modified proposal here, together with my acknowledgement that life is too short to wage a campaign against the obstreperous obstructionists.--Goodmorningworld (talk) 09:38, 16 May 2009 (UTC)
MoS is written in US English. Is it "offense"? Tony (talk) 16:15, 10 May 2009 (UTC)

Units of measure

They are not units of measurement. That is wordiness. One makes a measurement and records a measure. Measurement is wrong, just wrong. I started the discussion at Template:Convert talk page but I think best to bring it here; so far I have one supporter (and no objector). —Preceding unsigned comment added by SimonTrew (talkcontribs) 17:13, 13 May 2009

I don't object to changing the phrase "units of measurement" to "units of measure". However, no one is in charge of the English language, and if you go around with the attitude that some commonly used words or phrases are "wrong, just wrong" expect to get reverted a lot. --Jc3s5h (talk) 18:00, 13 May 2009 (UTC)

My built-in Apple dictionary states "measurement |ˈme zh ərmənt| noun the action of measuring something : accurate measurement is essential | a telescope with which precise measurements can be made. • the size, length, or amount of something, as established by measuring : his inseam measurement. • a unit or system of measuring : a hand is a measurement used for measuring horses."

or see the definition at Measurement: Measurement is the process of assigning a number to an attribute (or phenomenon) according to a rule or set of rules. The term can also be used to refer to the result obtained after performing the process.

Yours, GeorgeLouis (talk) 20:56, 13 May 2009 (UTC)

I expect most (if not all) dictionaries would tell you the same. Measurement can mean either the process or the result. Using either is certainly not "wrong", it is simply a matter of personal preference. wjematherbigissue 21:32, 13 May 2009 (UTC)
  • The term “units of measure” is the proper phrase. Greg L (talk) 03:15, 14 May 2009 (UTC)
Just to come back on this, I don't go round editing things because I find them "wrong, just wrong". I was simply expressing my opinion on this talk page. Since WP:MOSNUM has to make standards, let's make them the simplest they can be.
I don't see how this differs from deciding about whether commas and spaces are used in numbers, how dates should go, and so forth. All style guides are going to differ, their aim is to find an internal consistency. My perhaps rather blunt statement was so settle on saying "unit of measure" not "unit of measurement".
The guys over on the convert template do great work. Some of us who use those templates find problems and they sort them out very quickly. I use that template as much as possible because it ensures consistency. That's what we're going for here isn't it? SimonTrew (talk) 14:56, 15 May 2009 (UTC)
  • I don’t understand what the exact issue is with templates or not. I don’t see how that affects the fact that “unit(s) of measure” is the commonly accepted, proper expression, not “unit(s) of measurement.” Greg L (talk) 01:33, 17 May 2009 (UTC)
Because people who write the documentation for it have to know which phrase to use. SimonTrew (talk) 01:56, 17 May 2009 (UTC)
Oh yeah I forgot. Wikipedia has no documentation. SimonTrew (talk) 01:59, 17 May 2009 (UTC)
The relevant article is titled units of measurement, so I would say for the sake of consistency, that should be the phrase to use. Incidentally, units of measure redirects to a disambiguation page, since that phrase covers more than just measurements. wjematherbigissue 05:27, 17 May 2009 (UTC)
I doubt unit of measure is a clear choice; the OED does not cite it, although they do cite units of measurement. Septentrionalis PMAnderson 17:44, 17 May 2009 (UTC)

High-precision numbers

The new WP:MOSNUM#High-precision numbers section begins "Numbers with more than four digits after the decimal point". But, if such numbers do exist in an article, and visual gaps* are used to group the digits into groups of three, should we do the same to all numbers in the article, including large integers?

*I use the phrase "visual gaps" to indicate that when viewed with a browser, there appears to be a gap between certain digits, no matter whether there is an actual space character there or not. This avoids discussing which template is used to create the effect. --Jc3s5h (talk) 16:01, 15 May 2009 (UTC)

  • Short answer: yes. On Wikipedia, consistency within a given article is a valued attribute. As a general principle, numbers with more than four digits to the right of the decimal marker should be delimited to the right with narrow gaps. So for two reasons (consistency and ease-of-parsing), yes; all the numbers should be delimited in the article. Using the {{val}} template is best for a variety of reasons.

    Caveats: Sometimes numbers are expressed to high precision as a near-decorative effect to demonstrate how an irrational number (like pi) can be really long; there is no expectation that readers actually parse the value nor appreciate distinctions past, say, the fifth digit after the decimal point. Also, a not-too-uncommon failing Wikipedians fall victim to is to express numbers with greater precision than is truly necessary to convey the point, leading editors to not really care if it is easy for readers to parse and understand a number or not. So…

    On the assumptions that 1) the number is not decorative-to-impress, and 2) is expressed at a precision suitable for conveying germane information, it is therefore important for readers to be able to easily parse the value and it should be delimited—commas to the left of the decimal point, narrow gaps to the right. And, yes, that would include “large integers”—that doesn’t change the fact that there is (supposedly) important information to the right of the decimal point that requires parsing. Greg L (talk) 01:29, 17 May 2009 (UTC)

I see that {{val|12345|06789}} is displayed as 12345.06789. This format is unique to Wikipedia; no other publication mixes two methods of delimiting a number in the same article, never mind the same number. Although MOSNUM implicitly accepts this format by mentioning the Val temlate, nowhere does it explicitly say this format is acceptable. --Jc3s5h (talk) 01:50, 17 May 2009 (UTC)
Well in that case don't take it to 5 DPS or sig figs? Is that not simpler? I'm missing something here.
  • Yeah, it seems you’ve made up your mind and are going to do what you wanna do, Jc3s5h. Nothing is going to change the fact that large numbers with four or five or more digits to the left of the decimal marker must be delimited with commas here on Wikipedia in our general-interest articles; that’s simply the way it’s done; we use the American system to the left of the decimal marker, not BIPM, not Swedish, 1, Swedish 2, or whatever. So if you think it somehow best serves the interests of our readers to have really, really long, non-broken stretches of digits to the right, it strikes me that you are going to do it your way. Happy editing. Greg L (talk) 02:05, 17 May 2009 (UTC)
  • My reading is that since the MOSNUM allows gap delimiters in appropriate types of articles, an article that takes advantage of this possibility should use gap delimiters on both the left and right of the decimal. The Val template should be modified accordingly. Since that is what every other publication that allows gap delimiters does, the MOSNUM should be read to follow customs that have some kind of precedent, rather than being read to require an unprecedented format. --Jc3s5h (talk) 02:20, 17 May 2009 (UTC)
  • You’ve arrived late to the discussion. The {{val}} template was no accident. It was the product of lengthy (very lengthy) discussions both here on WT:MOSNUM and at WT:MOS. The template is locked for a reason. If you don’t like the mix of commas and thin gaps, and if you are in a scientific article, you can always use scientific notation, such as 5.7395739×10−17 m, which avoids the issue altogether. Greg L (talk) 02:26, 17 May 2009 (UTC)
Yes, I agree with Greg L. I had made the assumption that val and convert had been very carefully considered. I believe the ISO standards do specify spaces in the decimal parts of numbers, though in the UK that is rarely actually used. The difficulty then becomes, write it to an international standard, or write it the way people actually use it? That's a very difficult one to answer and as Greg says has no doubt gone through much debate before.
The convert templates sometimes produce bizarre results, and they get fixed very quickly. Some people like me use them not because they are simple (they are a bit hard really) but because we know that then the article is future-proof against changes and it gives future editors, and translators into other languages, an idea of the actual measure being said. I struggles for example when translating from French when it says "about 10 kilometres from X or Y" do I put that in miles or leave in kilometres? That kind of thing, if not resolved by using a template, at least gives some future editor an idea of what was meant.
I had some troubles at Electric car with knowing what precision was actually implied when someone says "200 mph maximum speed", like exactly 200? About 200? It is that kind of thing where putting an exact figure from a reliable source, and then knocking it down to a sensible precision for the reader, is very valuable (pardon the pun).
Best wishes SimonTrew (talk) 14:50, 17 May 2009 (UTC)
SimonTrew wrote "I believe the ISO standards do specify spaces in the decimal parts of numbers...." That is not correct. Several international recommendations including the brochure published by the BIPM require that only thin spaces be used to group digits, "neither dots nor commas are inserted in the spaces between groups of three." This applies both left and right of the decimal point. I infer this was intended as a way to avoid the problem that in some places, a period is used as a decimal point, while in other places a comma is so used. --Jc3s5h (talk) 15:07, 17 May 2009 (UTC)
  • (←outdent). The BIPM also recommends that a space be placed between a number and the percent symbol; e.g., 75 % and not 75% (5.3.7 Stating values of dimensionless quantities, or quantities of dimension one). Notwithstanding that some editors think the sun rises and sets on standards organizations like the BIPM and the IEC, Wikipedia wisely eschews such advise (they are “recommendations”, not laws etched by God into stone) and we follow the practices that cause the least confusion with our readership. So, like Encyclopedia Britannica, we use commas. There are a number of workarounds to your “predicament”.. Greg L (talk) 22:45, 17 May 2009 (UTC)
It would indeed be impossible (and probably undesireable)to get editors to follow every BIPM, IEC, and ISO practice, such as always using symbols for SI units rather than spelling them out. However, when the most popular English-speaking customs just aren't up to the job in certain articles, is it better to switch the whole article over to a less-common custom that gets the job done, or is it better to make up a format that no other publication uses? I'd go with the less-common format over the made-up format. --Jc3s5h (talk) 22:58, 17 May 2009 (UTC)
That makes sense and would seem to be WP:COMMON. I think the main thing is to consider your readership. As Greg L implied, in articles for general purpose readership it would be absurd to use bizarre notations. In a very technical article it may be the most appropriate. Certainly to start writing in phonecian sexagesimal or something would be a bit absurd. SimonTrew (talk) 17:28, 18 May 2009 (UTC)

Feb.

{{editprotected}}

Abbreviations such as Feb are used only where space is extremely limited, such as in tables and infoboxes. Please add footnotes; I assumed this was obviously covered, but AWB programming seems to differ . Septentrionalis PMAnderson 17:48, 17 May 2009 (UTC)
This page is only semiprotected so please make the edit yourself. (It's only the dates section which is fully protected.) — Martin (MSGJ · talk) 09:23, 18 May 2009 (UTC)
I have reverted the change. PMA proposed adding "footnotes", but did not propose dropping "extremely". In the past, a number of editors in this forum have reaffirmed that date formats in footnotes should be consistent with the existing guideline. Space in footnotes is not "extremely limited" or even "limited". Shortening an occasional footnote that takes 2 or 3 lines to 1 or 2 lines by abbreviating "March" to "Mar" does not greatly improve WP (it's no improvement at all IMO), but having different guidelines for date formats in footnotes is a detriment. If there is a rare case where the footnotes really require such abbreviating, and would improve the article, (and I can't begin to imagine what that might be), then just ignore the guideline and improve the article. There's no good in encouraging 6 formats ('6 Feb', '6 Feb.', '6 February', 'Feb 6', 'Feb. 6', 'February 6') where we currently get by fine with 2 formats. If this change stood, crews and/or bots would soon be going through all articles mashing the months. Chris the speller (talk) 17:31, 18 May 2009 (UTC)
I just realized it would be even worse than that; 'September' would have 10 possible formats, since it is often abbreviated as either 'Sep' or "Sept'. Chris the speller (talk) 17:37, 18 May 2009 (UTC)
Fine, we can put back extremely, if it matters to people. I took it out primarily because infoboxes aren't extremely limited to begin with.
  • If this change stood, crews and/or bots would soon be going through all articles mashing the months. No, this is exactly what I am trying to avoid; if we permit all of Sep, Sept., and September in footnotes, there is no justification for any bots on the subject.
  • Abbreviating months in footnotes has two minor, but real, advantages. It permits copying some citations in exactly the format preferred by the journal in question; it will sometimes permit a footnote to go on one line instead of two, making a list of footnotes shorter and easier to read. Septentrionalis PMAnderson 18:30, 18 May 2009 (UTC)
I have restored the original language again. PMA is the only editor who asserts that space is "extremely limited" in footnotes. We do not owe allegiance to any journals in the matter of date formatting. Abbreviating a month may sometimes allow a short paragraph in the body of an article to remain on one line instead of flowing onto a second, but that's not a worthy goal either. We can't assume how others see the footnotes, since they use various browsers and settings. Chris the speller (talk) 18:47, 18 May 2009 (UTC)

(unindent) New editors often learn by example. If they happen to examine articles that contain no full dates in the body of the article, but dates with abbreviated months in the footnotes, they might conclude they should abbreviate months everwhere. --Jc3s5h (talk) 17:47, 18 May 2009 (UTC)

This seems far fetched. How many articles mention no month in the text, have footnotes, and cite journal articles in them (the principal reason why anybody would want a month in a footnote). How many newbies will begin exactly in that small subset? how many of them will have so small a degree of fluency in English that they will not already know that "Journal of Irreproducible Results (Apr. 1984) pp.1345-2658" is a condensed form of English, not to be lightly adopted in article text.
It would make as much sense to ban pp. so that newbies won't write "the last four pp. of the novel" in plot summaries. But I fear I've violated WP:BEANS. Septentrionalis PMAnderson 18:45, 18 May 2009 (UTC)
I can think of no reason whatsoever to abbreviate months in footnotes. There is no space limitation. Wrapping depends on browser settings more than anything. The formatting preferences of the cited journal are irrelevant. As Chris points out, should there ever be legitimate cause to abbreviate, then simply ignore the MOS guidelines. wjematherbigissue 18:44, 18 May 2009 (UTC)
The longer the footnote, the more browsers it will wrap on. Saving four spaces by using "Dec." will therefore help some readers, and improve accuracy. (Again, I would not make "Dec." mandatory - whether there is any case for it depends on the journal, and the length of the footnote.) Septentrionalis PMAnderson 18:58, 18 May 2009 (UTC)
It would be nice to have a MOS that didn't have to be ignored; is that beyond hope? Septentrionalis PMAnderson 18:51, 18 May 2009 (UTC)
Apparently. Pity, we could actually use one. But, clearly, Emerson was right. Septentrionalis PMAnderson 18:58, 18 May 2009 (UTC)

Degrees

This manual says:

e.g., write 5° 24′ 21.12″ N for coordinates, 90° for an angle, but 18 °C for a temperature

Are there authoritative sources independent of Wikipedia that prescribe this usage? I'd have written 18° C. Michael Hardy (talk) 03:13, 2 May 2009 (UTC)

BIPM, NIST, all (or nearly all) scientific journals, if I recall correctly. I'd have to look them up to be sure, but I've never seen 18° C anywhere.Headbomb {ταλκκοντριβς – WP Physics} 03:32, 2 May 2009 (UTC)

I've found this.

The numerical value always precedes the unit, and a space is always used to separate the unit from the number. Thus the value of the quantity is the product of the number and the unit, the space being regarded as a multiplication sign (just as a space between units implies multiplication). The only exceptions to this rule are for the unit symbols for degree, minute, and second for plane angle, °, ', and ", respectively, for which no space is left between the numerical value and the unit symbol.



This rule means that the symbol °C for the degree Celsius is preceded by a space when one expresses values of Celsius temperature t.
BIPM

JIMp talk·cont 08:24, 2 May 2009 (UTC)

But the reasons why the angle should be an exception (and also the superscript apostrophes for feet, inches, minutes, and seconds) would seem to apply to temperatures too: It's the same symbol, and it's a superscript, and it's not written out in letters but is instead a sort of punctuation mark.
If one says the temperature is 68 degrees, without specifying Celsius, Fahrenheit, etc., would you write 68 ° or 68°? Michael Hardy (talk) 02:58, 3 May 2009 (UTC)
I wouldn't write a thing until I know wheter they mean 68 °F or 68 °C. 68° is an angle. Whoever writes 68° (or 68 °) to mean either should sent to WP:Speedy De-existenciation. Headbomb {ταλκκοντριβς – WP Physics} 00:53, 4 May 2009 (UTC)

I suppose it doesn't seem that way to the BIPM. Like it or not 68 °C appears to be the standard convention. Is altering such conventions what Wikipedia should be doing? JIMp talk·cont 08:49, 3 May 2009 (UTC)

  • I wish I would have noticed this thread sooner. According to the BIPM’s SI brochure: Subsection 5.3.3, Formatting the value of a quantity, a space is always used to separate the unit symbol from the numeric value. Notable exceptions are the unit symbols for degree, minute, and second for plane angle, °, ′, and ″ (e.g., a latitude of 47° 38′ 8.8″). One properly writes It is cool outside as it is 18 °C. It is also proper to write Turn right 90°. Greg L (talk) 06:53, 14 May 2009 (UTC)
Interestingly, the Canadian government style manual does not follow the BIPM lead in this case. Its recommendation is not to use any spaces, i.e. "18°C". I believe the reason is that the space is redundant since the meaning is clear whether it is used or not.RockyMtnGuy (talk) 17:17, 19 May 2009 (UTC)
I would offer the observation that using a space or a non-breaking space to convey meaning begs for errors, particularly in the context of line wrapping while editing the wikitext. Many editors have no clue what " & n b s p ; " stands for and some will happily delete it or replace it with a tap of the spacebar. This simple act shouldn't break the meaning of the article. Better to have {{convert|18|C|F}} generate 18 °C (64 °F) and let the template carry the freight.LeadSongDog come howl 18:07, 19 May 2009 (UTC)

Ages?

I know Associated Press style is different than Wikipedia style, but what is the rule for ages and numbers? A.P. says always a numeral figure, but I can't seem to find it on this page. Fdssdf (talk) 23:05, 17 May 2009 (UTC)

The section Numbers as figures and words. Our basic rule is spelling out numbers less than ten, but other ages less than 100 come in a grey area. It may be that ages are another exception, of which we already have a dozen. Septentrionalis PMAnderson 01:36, 18 May 2009 (UTC)
This is one of the sections which could be drastically simplified, with a combination of Follow the sources, and a rule of thumb to use figures for large or precise numbers. (Largeness is subject dependent: 13 battleships is a large number, thirteen houses is not.) Septentrionalis PMAnderson 14:01, 18 May 2009 (UTC)
If you own thirteen houses it is. JIMp talk·cont 13:06, 23 May 2009 (UTC)
WP:MOS (not MOSNUM) says "Mathematical quantities, measurements, stock prices, etc., are normally stated in figures." I don't think it's obvious whether that "etc." includes ages, so I guess you could write either 9 years or nine years and they would both be fine. (But avoid writing "John was 17 years old and James was nine", the difference in format makes it take at least half a second longer to compare ages without any good reason—an arbitrary boundary at 10 is no good reason IMO.) --A. di M. (formerly Army1987) — Deeds, not words. 19:20, 18 May 2009 (UTC)
I wrote it; it doesn't. Stock prices and so forth are quasi-continuous quantities; ages are normally integers. If this section is not simplified, the etc, should be recast; but I am leaving this page, for reasons explained at WT:MOS#What's wrong with MOS. Septentrionalis PMAnderson 15:10, 19 May 2009 (UTC)
I'm sure PMA didn't literally mean that "ages are normally integers". Perhaps "personal ages are normally stated as integers"? It certainly isn't appropriate for geological or cosmological ages. In any case personal age is too often used in a perishable context that requires an "as of 2009" clause. In such cases it is usually better given as a birth year or birth date to avoid growing stale. LeadSongDog come howl 16:08, 19 May 2009 (UTC)
Another alternative is a template which will do the subtraction. Septentrionalis PMAnderson 18:09, 19 May 2009 (UTC)
Are personal ages normally stated as integers? Small children tend to be very particular to tell you they are six and a half, and so on. SimonTrew (talk) 16:28, 19 May 2009 (UTC)
But we have very few articles about small children; even John Stuart Mill says he learned Greek "at the age of three", not a fractional value. Septentrionalis PMAnderson 18:09, 19 May 2009 (UTC)
Good point. Mozart wrote his first piece "at the age of three years" or whatever. I just didn't want to exclude it unnecessarily. SimonTrew (talk) 23:02, 19 May 2009 (UTC)

Protection

How long is the dates section going to stay protected? I say unprotect and block whoever tries to force something through with editwarring. --Apoc2400 (talk) 15:40, 23 May 2009 (UTC)

April 20 v 20 April

I would like to make clear somewhere in the guideline that British date formatting can be either April 20 or 20 April. When writing a date using only numbers, Brits always write the day first (20/04/2009), but when using the word for a month, April 20 is acceptable, as is 20 April.

The reason I'd like to clarify this is that I often see British editors arrive to change from April 20 to 20 April, often with edit summaries that imply they think the MoS requires it. It's usually done inconsistently; and I've seen editors change image names that contain dates (by mistake when they're changing other things), so that the images are no longer visible. Cleaning up after it can be time-consuming. Even though the MoS makes clear that people shouldn't be imposing style changes on stable articles, this misunderstanding about dates seems very persistent.

Does anyone mind if I clarify it to emphasize that either format is acceptable — when the word for the month is used, and not just numbers — whether in a British-related article or anywhere else? I'm posting below a sample of British and Irish newspapers that use the month/day format on their front pages, and in their articles.

England:

Scotland:

Wales:

  • Swansea Evening Post [24]
  • Western Telegraph [25]

Northern Ireland:

  • Ulster Herald [26]

Republic of Ireland:

  • The Irish Times [27]
  • The Irish Independent [28]

SlimVirgin talk|contribs 15:10, 20 April 2009 (UTC)

i have the impression that quite a lot of people would mind the change you're proposing; i think people make the kind of edits you're talking about because they prefer DMY format, not just "because the MOS says so". the fact that some UK publications use the MDY format is not universally persuasive: -ize is widely used in UK publications too, but lots of Wikipedia editors still routinely change it to -ise in articles in UK English. Sssoul (talk) 17:32, 20 April 2009 (UTC)
While I understand this argument and it is internally consistent, I would be loath to see the long-standing consensus that US articles (other than military ones) use MMM-DD-YYYY, UK/Commonwealth ones use DD-MMM-YYYY, and articles on both or on neither use whatever the first major contributor used. While this is not a perfect situation, it has worked for a long time and produces articles which are at least consistent. A bit like WP:ENGVAR, it is a situation where something most people accept is imperfect is nonetheless a hard-won compromise that everyone can just about work with. Of course, if there is a strong consensus that we should change this style guide then I will accept it. --John (talk) 17:33, 20 April 2009 (UTC)
There is considerable evidence that Anzac usage is more artificially consistent than the UK, although in the opposite direction from the schoolroom consistency of American - partly as a reaction to American influence. In addition, April 20 was a standard British usage as recently as the Second World War.
There is no long-standing consensus (the present text is a few months old); there is a handful of half-educated dogmatists. What we should say is "English varies; learn to live with it" - but this would put the bullies out of work.
I would support text saying that either variety is acceptable, if we included WP:ENGVAR's cautions against edit-warring. Septentrionalis PMAnderson 17:59, 20 April 2009 (UTC)
Pmanderson, here is a revision from December 2007 with the language we are discussing in place as of 15 months ago. In Wiki-time, that is a very long time ago, and certainly more than "a few months". Slim, don't spoil your decent argument by inflating the numbers; the Sunday Herald and the Herald are sister papers, as are The Times and The Sunday Times, and also the Mail and the Mail on Sunday. Each pair undoubtedly uses the same style guide. Each therefore only counts as one example. Although I am not disputing this aspect of your argument, exaggerating like this doesn't strengthen it. --John (talk) 18:30, 20 April 2009 (UTC)
i agree that it undermines the argument to list those pairs of "sister papers" as if they were separate entities. Sssoul (talk) 19:31, 20 April 2009 (UTC)
  • The present wording derives from the archives marked D6 through D9 at the top of the page - last August and September. In the process, we observed that newspaper usage for all English speaking countries is inconsistent. Septentrionalis PMAnderson 21:08, 20 April 2009 (UTC)
My apologies if it looked like exaggeration; I just listed them as I came across them. I called up a list of newspapers in the different countries, and there did seem to be quite a few that used April 20, and not 20 April, many more than I've mentioned here. The Encyclopaedia Britannica also uses the April 20 format e.g. its article on Tony Blair says: "Tony Blair, prime minister of United Kingdom ... born May 6, 1953, Edinburgh, Scot."
The important point is that both formats are acceptable in the UK and Ireland, so there's no reason for us to ask editors to prefer one over the other. SlimVirgin talk|contribs 01:32, 21 April 2009 (UTC)
Other than a long-standing consensus. Let's turn this around; given that either format is intelligible to anyone, anywhere, what advantages would accrue to us by changing the MoS guidance? All newspapers have a style guide to ensure consistency within and between articles. Why would we not have? --John (talk) 01:42, 21 April 2009 (UTC)
Because it's being done on the grounds that 20 April is British, and therefore must be used in Britain-related articles. But it's not British. April 20 is British too. What forcing the 20 April format on people has done is lead to inconsistency within the encyclopedia in general (because most articles do use April 20 format), and inconsistency within articles, as people arrive to change some, but invariably not all, date formats, thinking the MoS requires it. If consistency across the project matters, we should remove the "20 April is British" advice, not least because it's factually incorrect.
John, you wrote above that the current advice is the result of a hard-won compromise. What were the issues that this was a compromise for, and do you remember roughly where/when that discussion took place? SlimVirgin talk|contribs 03:56, 21 April 2009 (UTC)
A few problems in all the above:
  • The Daily Mail example is not entirely correct - update dates in the form "21st April 2009" are seen in articles such as this (the day ordinal notwithstanding).
  • Encyclopædia Britannica, Inc. has been American since the early 20th Century; it is absurd to suggest this is relevant for UK standards.
  • Why the excessive focus on newspapers, and particularly their websites, which may be subject to inconsistent software settings and not necessarily what appears in print? Besides, WP:NOTNEWS.
  • Government consistently shows international/dmy date formats (apart from slight differences as day ordinals); academic usage also seems overwhelmingly international/dmy. Those are more compelling sources for style guidance than newspapers.
  • "... most articles do use April 20 format"? Wonder how fast it took to go through all 4,584,963 articles to confirm this? Even if so, factors such as ignorance of the MoS or systemic bias may come into play here.
Dl2000 (talk) 04:28, 21 April 2009 (UTC)
  • Testing Slim Virgin's claim is not difficult. As I write, 9648 articles use the string "April 20"; the comparable figure for "20 April" is 5763. This is about what one would expect, given that most native speakers of English are from countries where "April 20" is customary usage or those where custom is divided. It may be that other dates would give different results; but even "November 5"/"5 November" is 9465/5165.
  • When MOS disagrees with what WP actually does, ignorance of MOS is one possible explanation; but MOS' ignorance of Wikipedia is far more likely - and MOS' ignorance of the English language is always possible. Our guidelines are supposed to describe what editors in general do agree on; not what some small coterie thinks they ought to agree on. Septentrionalis PMAnderson 02:33, 22 April 2009 (UTC)
  • Well why didn’t you say so earlier, PMA? I shouldn’t look to MOS for guidance since it embodies the collective wisdom of a “small coterie” that has fouled it all up. Apparently, you posses all the writing wisdom and uniquely know what is best for Wikipedia. From hereon, I’ll skip the silly step of reading MOS and just ask you what is best. Greg L (talk) 03:29, 22 April 2009 (UTC)
    • You could do worse; but you would do better still to consider what Wikipedians at large actually agree on, and limit this waste of electrons to that. Septentrionalis PMAnderson 05:11, 22 April 2009 (UTC)
  • Formatting inconsistencies across WP are rife: there are articles which have a mix of dmy, mdy, ISO, and other variants. I hope to see a universal date format across the whole of WP which does not rely on autoformatting. The reason we got into autoformatting (and the dates case currently before Arbcom) is that we couldn't agree on a default style. I also believe that '5 November 2008', with one awkward comma less, is more elegant than 'November 5, 2008', but would be prepared to align to mdy dates if uniformity could be achieved. However, each time there is a discussion, it almost invariably goes back to "WP:ENGVAR works well enough." Well, I disagree - it's a total shambles. We have some US articles using international (dmy) style dates and UK/Oz articles which use mdy style dates/

    For WP not to have adopted a single format (or language variant) is seriously unprofessional. Admittedly, WP is not "professional" in the true sense, but that should not stop us taking a serious look at our style guide. It is important to note, as SV points out, that each publication has only one definitive style of dates and one language variant. Being the most widely visited information site on the Internet, WP is one of the few sites where there is a notable absence of prevailing style policy. I don't think the [decentralisation] suggestion is the right way to go. Ohconfucius (talk) 04:55, 22 April 2009 (UTC)

There is long-standing consensus against this nonsense; in particular, we are agreed not to favour any national form of English against any other.
Even if this were not true, making the bog of unsourced, semi-literate, tendentiousness that causes Wikipedia to be an unreliable source "professional" by tidying links and -our endings would be a joke, if it were not consumer fraud. Septentrionalis PMAnderson 05:11, 22 April 2009 (UTC)
  • "bog of unsourced, semi-literate, tendentiousness...". Yep, just like you said, WP is a joke. A little bit of order no doubt offends the anarchist in you... What's wrong with trying to make it a little less of one? Ohconfucius (talk) 06:39, 22 April 2009 (UTC)
  • I believe my position to be little different from Brion Vibber, so I would appreciate it if I was not quoted out of context in an unrelated discussion. Thank you for your attention. Ohconfucius (talk) 01:52, 23 April 2009 (UTC)

The best written English is almost always a reflection of the best spoken English (which is why Shakespeare, the 1611 Authorised/King James Version of the Bible, and the Book of Common Prayer, all written to be to be spoken aloud, are still such powerful influences—even among non-Christians and dramaphobes—four hundred years after they were written, despite major changes in vocabulary and sentence structure.) And most of us don't say "April twenty" or "Twenty April"; we say "April twentieth" and "the twentieth of April", partly because "Twenty" is a cardinal number and "twentieth" is an ordinal (modifying the unspoken "day of"). And in ordinary speech, most of us (I think) use both forms, "April twentieth" and "the twentieth of April", almost every week, if not every day or every hour. At times, we can even use both forms in the same sentence. If the technical reasons related to date-autoformatting, and the managing-editor's concern for saving space, no longer apply, then there's no reason for Wikipedia to disallow April 20th and 20th April, either, especially when they add to the clarity, flow, or style of a sentence or paragraph. ("Lincoln was shot on the 14th of April, 1865, and died on the 15th.") A foolish consistency is the hobgoblin of little minds... (Ralph Waldo Emerson) —— Shakescene (talk) 06:51, 22 April 2009 (UTC)

A reasonable argument for more flexibility; is there dispute - other than on the grounds above that there must be only one standard, as in Highlander? Those who wish to demote WP:ENGVAR should argue for that elsewhere. Septentrionalis PMAnderson 14:16, 22 April 2009 (UTC)
  • The role of any encyclopedia is to educate readers with minimal confusion and ensure they are well prepared to absorb information in their studies elsewhere on the subject. Our English-language version, en.Wikipedia, is read by a world-wide audience but fully one-quarter is our American audience. There is no single date format “April 20, 1985” or “20 April 1985” which is “better” than another and there is no reason to even try to standardize on one format project-wide.

    Keep it simple. And I certainly wouldn’t try to intertwine date formatting to the dialect of English some volunteer editor used for an article; that is a separate matter. I would propose we simply make the date format as natural as possible for the likely readership. Nothing more.

    Allow me to illustrate my point via example: Articles not closely associated with American topics such as Italy, Austria, Basilica di Saccargia and Kilogram have a pronounced non-American readership. Regardless of the dialect of English used by the editor of that article, we should be thinking foremost about our readers. So in articles on general or European subjects (articles clearly not associated with the U.S.), we would simply use Euro-style dates.

    Conversely, for articles on, or closely associated with American topics, such as Spokane River Centennial Trail; Coeur d'Alene, Idaho; American Revolution; and New York Yankees, they have a preponderance of American readers and the date format that is most natural for those readers is the American-style date.

    It is an utterly trivial matter to simply use the date format that is most natural for the likely readership. I would propose this simple guideline:

For articles on, or strongly associated with, the following countries and territories: The United States, American Samoa, Guam, Northern Mariana Islands, Puerto Rico, U.S. Virgin Islands, Wake Island, Republic of the Marshall Islands, Federated States of Micronesia, and Republic of Palau; editors should use the U.S.-style date format (“February 2, 2008”), otherwise, editors should use the international date format (“2 February 2008”) in articles.

The virtue of this guideline is it doesn’t require that editors go research what is an “American territory”; all the criteria they need is right there in the article. The guideline has a simple test: “what is the subject of the article about?”, which equates to “will it tend to have a preponderance of an American readership? If so, use American-style dates. Otherwise, use Euro-style dates. I see no reason to make it any more complex than this. Another virtue of this approach is we no longer need to discern who should be considered the “first major contributor” and “what dialect of English he/she used.” That doesn’t really matter; it’s not about us (editors), it’s about our readership and doing our best to ensure the articles read as naturally and well as possible.

I couldn’t care less if some Italian editor somewhere happen to have used British-dialect English and Euro-style dates when he first expanded Boston Red Socks from a stub. All any editor should have to do, if they encountered a mix of date styles in that article is consider (for all of about one second) what the subject is about and then (quickly) settle on the format most appropriate for the subject matter. Simple. And, yes, there will be an occasional grey-area article, like Gulf war: is that “strongly associated with America?” I think editors can work in good faith to resolve these issues if or when they arise. Greg L (talk) 19:39, 22 April 2009 (UTC)

There is always an easy solution to every human problem - neat, plausible and wrong. - Henry Louis Mencken. This is one of them; it is pure conjecture who reads what article. If it is true that Italians read Kilogram more than Americans (unproven; why should they? They know what a kg is.) and that they are annoyed by April 20 (unlikely), then they will change it; that's what collaborative editing is all about. Septentrionalis PMAnderson 20:10, 22 April 2009 (UTC)
  • Quoting you PMA: …it is pure conjecture who reads what article, well, it may be *pretty* to think as you do, but your logic ignores common sense and the laws of probabilities. It is obvious that articles like Spokane River Centennial Trail; Coeur d'Alene, Idaho; American Revolution; M1 Abrams; New York Yankees; Battle of Midway; and Boston Red Socks are all going to have a much heavier American readership than would articles not closely associated with America such as Basilica di Saccargia and atom. Your and Locke’s endless arguments about how you want to have your articles your way is tedious beyond measure. We want to explore solutions that don’t involve more ANIs and ArbComs and would like to leave the wikidrama behind. Some simple sensible arguments out of you would be most welcome. Greg L (talk) 22:23, 22 April 2009 (UTC)
    • It is reasonably likely that articles about places in Idaho will have been written by Americans, and use April 20. It is plausible, but no more, that they will be read by Americans (but there are foreign tourists, too; some from Canada). But so what? if those likelihoods pan out, they will have April 20 already, and we need not legislate. Septentrionalis PMAnderson 23:05, 22 April 2009 (UTC)
The proposal by Greg L would lead to the rather absurd ruling that day-month format should be used in an article related to Tijuana, Mexico, because it is not a US possession. Who would have more interest in such an article, a resident of California, Sussex, or New South Wales? Chris the speller (talk) 03:24, 23 April 2009 (UTC)

In addition to Chris' remark (quite sound), Greg's proposal would require the use of April 20 on purely American articles. Of the eight samples he names, six do use it without any requirement here; one does not mention any day. This is what one would expect, without any rule, and shows we don't need one. Battle of Midway, however, uses 20 April consistently; this appears to be a recent development among the American military, but is not something we should (or can) suppress without wider discussion.

But not only does it work poorly for American articles, it would work very badly for

Where do you get the idea that "the convention of 20 April has only come in since the Second World War" I have been doing a lot of work on Napoleonic treaties and they often use day month formats see Wikisource:Treaty of Paris (1815), which is based on Foreign Office and Hansard sources translating from French original sources. I am also reading a book "The Civil War in Worcestershire, 1642-1646, and the Scotch invasion by 1651", written in 1905 the author uses day month format in the text. The author also quotes Civil War primary sources and they seem to use day month. e.g. page 133 "Wednesday, the 12th of June, 1644, His Majesty marched out of Worcester...". (The quote is available via Google books in a number of sources). --PBS (talk) 22:45, 23 April 2009 (UTC) Just had a look at Wikisource:Treaty of Paris (1783) there they use "this third day of September, in the year of our Lord" PBS
12th of June (which the present text of MOSNUM forbids, and should not) is long-established. 12 June (which MOSNUM permits) is new; the contrast can be seen in Orwell's letters, for example; he dated June 12 and sometimes June 12th, but the notes, even in the 1968 edition of his collected works, use 12 June 1940. Septentrionalis PMAnderson 02:37, 24 April 2009 (UTC)
I see what you mean--however Hansard was using 20 November (Class C) in 1816 (along with other formats)--but just the two formats "20 April" and "April 20" was easily justified when we were defaulting to auto-formatting dates, as those were the ones the software supported, the justification after we abandoned that was (if my memory serves me well I have not looked in the archive) that this is the modern professional way. Personally I have no strong opinion on it, and would be happy with a laissez-faire policy on this issue, only prohibiting date formats that are confusing to readers e.g. neither dd/mm or mm/dd as in 9/11. --PBS (talk) 05:59, 24 April 2009 (UTC)

I support SV on this change it "to emphasize that either format is acceptable [in British articles] — when the word for the month is used, and not just numbers" with the usual proviso not to change from one majority usage to the other, without agreement on the talk page of the article. --PBS (talk) 22:19, 23 April 2009 (UTC)

Thanks, PBS. I can't see the harm in allowing 20th April or April 20th too, in addition to 20 April and April 20. All these variations are used in the UK and Ireland. The provision that styles shouldn't be changed without consensus is long-established, so there's no reason to suppose that giving people the option to use the date format of their choice will cause additional problems. On the contrary, it will hopefully prevent people from seeking to impose one style on otherwise stable articles. SlimVirgin talk|contribs 11:09, 24 April 2009 (UTC)
Are you actually in favor of allowing "20 April", "April 20", "20th April", "April 20th", "the 20th April", "April the 20th", "the 20th of April" all in the same article? If so that would make WP look awfully sloppy and unprofessional. If you mean to allow any one of those formats in one article, there will be edit wars. Oh, yes, edit wars, not to mention how tedious it will be for an editor to research who used what format first in an article that needs to be cleaned up. Chris the speller (talk) 15:56, 24 April 2009 (UTC)
She didn't say in the same article. Whether it would be unprofessional depends on the circumstances anyway; many military history articles are constructed around a sequence of days, and those should be consistent; but an article with two dates, one at each end, can use different formats for them, and nobody will notice. (20th of April is now usually employed for euphony or emphasis, and if it works, should be left alone.)
If you find an article that is noticeably inconsistent, post a notice on talk asking which format is preferable. If no one cares, no one cares, and you can pick one. Septentrionalis PMAnderson 17:59, 24 April 2009 (UTC)

Comment: SlimVirgin requested (through a note on my talk page) for the "Dates" section to be unprotected for editing. However, there does not appear to be any sort of consensus developing yet regarding changes to date formats, especially given the long-standing nature of the current guideline. Is there actually something ready for updating the MoS? Otherwise, I am inclined to leave the section protected for now. --Ckatzchatspy 18:05, 24 April 2009 (UTC)

I think a consensus is forming, CK. What I would like to do is make the edit, self-revert, and show people the diff to see if anyone objects. Is it normal practice to keep parts of guidelines long-term protected like this, or has there been an ongoing dispute that I'm not aware of? SlimVirgin talk|contribs 18:10, 24 April 2009 (UTC)
I see it was done because of the date-linking issue. [29] The edits I'd like to make are not related to date linking. SlimVirgin talk|contribs 18:15, 24 April 2009 (UTC)
Whoa! PMA and SlimVirgin want to deep-six the guidelines on date formats, and suddenly that's a consensus? "see if anyone objects" ... I strenuously object. Chris the speller (talk) 00:00, 25 April 2009 (UTC)
On what grounds?
I should repeat that Slim Virgin does not appear to be advocating within-article inconsistency; I think it often unimportant, as a minor part of this reform. Septentrionalis PMAnderson 00:41, 25 April 2009 (UTC)
(ec) i object too, on the grounds that editing the guidelines just to "see if anyone objects" is needless, since we're already discussing the issue.
what exactly is being proposed - is the "reform" supposed to affect only articles written in British English? or would it be more like any article, regardless of "national ties", can use either DMY or MDY format as long as the month is spelled out and "internal article consistency" is maintained? or is the proposal to allow other formats as well, and if so which ones? Sssoul (talk) 07:06, 25 April 2009 (UTC)
My proposal is to change the current MoS position that British English should be written as 20 April. My suggestion is that we make clear that both 20 April and April 20 are used in the UK, and that the MoS has no preference, except that articles be internally consistent. This applies only to when the month is written out. SlimVirgin talk|contribs 09:42, 25 April 2009 (UTC)

<--Two things. No, it's not normal that the guideline be locked on such a semi-permanent basis, but I don't see the point of changing it "to see if anyone objected" because we know what the status is; the issue could need an RfC to sort out. Ohconfucius (talk) 10:53, 25 April 2009 (UTC)

How much guidance is really necessary, anyway?

All right, then (although I won't argue to remove edit-protection or to replace Talk-Consensus-Edit with Bold-Revert-Discuss for dates in this Manual), I'll be the one to advocate against required date-format consistency within an article, because it's just not necessary, it serves no function and its absence does little harm. I expect that dates within a single section of an article will often harmonise/harmonize naturally; but in normal writing, I occasionally use more than one format even within a single paragraph. Only when it breaks the flow, or makes comparisons or relations between dates difficult or jerky, should inconsistency be discouraged. An analogy would be the rule of thumb that usually one should avoid writing "five oranges and 23 grapes"; similarly one should generally avoid "proposed on March 17, passed on May 28th and signed on 5 June". Such faulty parallelisms in style usually work themselves out in the normal process of editing and self-editing; I see little reason that date formats need be a special exception. And anything reasonable that reduces the huge, unmanageable and intrusive Manual of Style to something easier for the average editor to learn, grasp and remember allows everyone to concentrate needed attention on genuine sources of ambiguity, obscurity, mistake or confusion. —— Shakescene (talk) 07:01, 25 April 2009 (UTC)

Concur on all points, especially the last sentence. Septentrionalis PMAnderson 20:25, 25 April 2009 (UTC)
Baloney. Dates should be consistent unless in quotations. That's like saying it's OK to write both "honour" and "honor" in the same article. It's not; it's bad editing and bad form. Pick one format, stick with it. It's not written in a 1024-bit encrypted chinese dialect. Dates are not so special in that they should be exempted from guidelines that apply to everything else. Headbomb {ταλκκοντριβς – WP Physics} 21:06, 25 April 2009 (UTC)
Dates aren't so special that they need a special format: they're like the other ordinary words which every native English speaker and writer reshuffles all the time. Every reader should be able to recognise/recognize any of the common forms, and it's quite common for an ordinary reader to use more than one format, while spelling "hono[u]r" or "recogni[s/z]e" in a uniform way. What's really so jarring about seeing 20 April in one paragraph and May 5th in another? There can be faulty parallelisms with ordinary words, too, but they need no special rules because such problems are handled in the normal process of editing. Does it matter if I begin one sentence with a date , put another date in the middle of second sentence and end a third sentence with a different date? Do articles about May Day really need to put the first day of May into a single order or form?—— Shakescene (talk) 04:36, 26 April 2009 (UTC)
  • This is essentially free-for-all, and will result in some unholy mess, let alone edit warring. Who is it that keeps on saying MOSNUM is only a guideline? Surely, that has to be good enough for those who are pushing this? Ohconfucius (talk) 05:38, 26 April 2009 (UTC)
    • It has already resulted in an unholy mess, with Ohconfucius as chief mess-maker. We should acknowledge that anyone can edit implies that we will be inconsistent on such points. OC is welcome to set up a Wikipedia fork where this is not true, and the Manual of Style is enforced by any means necessary; but this is not it. Septentrionalis PMAnderson 17:26, 26 April 2009 (UTC)

All good stylebooks pick a date format and stick to it. Can anybody name one that doesn't? (By the way, if the person who used the word baloney above has an extra slice, I would like to have it to put between two halves of a biscuit/muffin for my lunch/supper.) Yours, GeorgeLouis (talk) 07:13, 26 April 2009 (UTC)

I think the purpose of the MOS should be minimizing the potential confusion and astonishment for the reader. A passage such as proposed on March 17, passed on May 28th and signed on 5 June definitely doesn't minimize it. As for having 20 April in a paragraph and May 5th in another, that might astonish a very small but nonzero number of readers; whereas this isn't a very serious problem, the cost of solving it by using the same format consistently in the article is close to zero (unlike, for example, avoiding the singular they, which would require us to arbitrarily pick a gender which can be seen as sexist, use awkward phrasing such as "he/she", or reword a sentence entirely). By the same token, if people seldom refer to Richard Feynman by his initials or to J. R. R. Tolkien by his first name, then it makes little sense to impose consistency and require that we should write either R. P. Feynman and J. R. R. Tolkien, or Richard Feynman and John Tolkien (this would be even truer if the names weren't in the same sentence), as either choice wouldn't be any less astonishing than just spelling names the way they are usually spelt, i.e. Richard Feynman and J. R. R. Tolkien. Nevertheless I have seen an "oppose" comment on a FAC based on something like that, where the names were in two different footnotes!
As for ordinal dates, I think the "full" form (the 27th of April/April the 27th) looks slightly old-fashioned nowadays, and it could be slightly confusing/astonishing to read them in a recently written article; as for the "semi-short" form (27th April/April 27th), I don't think it is any worse than the "short" form (27 April/April 27), although I think that mixing them in the same article (except in quotations) might be slightly confusing/astonishing.
As for PMA's argument about avoiding anachronistic language, it is an important principle but not the only one: whereas I wouldn't change "Persia" to "Iran" (or "Burma" to "Myanmar") in a sentence about something which happened before 1935 (or 1989), methinks thou agreest not that the "William Shakespeare" article hath to be written in 16th & 17th Century English. So there is a boundary line somewhere, but I'm not sure about which side of it writing 20 April in "George Orwell" would be on. --A. di M. (formerly Army1987) — Deeds, not words. 12:20, 27 April 2009 (UTC)
I would put the point about anachronism as advice; a writer should think about whether 27th of April is excessively florid or formal for his context. If he decides it's fine, then fine; there's no reason for MOS to review his decision. (If somebody objects, then they should discuss it, like other fine points of style.) Septentrionalis PMAnderson 14:41, 27 April 2009 (UTC)

Suggested change

I've reopened this, as we didn't seem to reach a conclusion. I'd like to make a tentative edit the page to see whether there is agreement. To clarify, I'd like to add that either April 20 or 20 April is acceptable in British English, when the word for the month is written out (as opposed to written in numerals). This is factually correct, so I can't see the problem with it. Any thoughts? SlimVirgin talk|contribs 23:47, 10 May 2009 (UTC)

Fine by me, but then I supported this before. Septentrionalis PMAnderson 00:04, 11 May 2009 (UTC)
I objected before, and I object again. Allowing a mishmash of varied date formats makes WP look sloppy and anything but professional, and makes it less easily readable. The current guidelines have been working fairly well for a long time, and I don't think we need to rock the boat. Loosening the guidelines too much would also lead to perpetual flipping back and forth, unintentional edit wars and intentional edit wars. No conclusion was reached because no consensus was reached, because not much excitement was generated. My guess is that because few editors jumped on board this bandwagon, most of the editors who are happy with the current guideline failed to take the new suggestion seriously, as they don't see this going anywhere. Of course, my guess might be wrong, but this doesn't seem like something most editors want. Chris the speller (talk) 00:55, 11 May 2009 (UTC)
It wouldn't be a mishmash, Chris. It would allow people to write April 20 or 20 April, so long as they're consistent within articles. What would be the harm in that? SlimVirgin talk|contribs 01:04, 11 May 2009 (UTC)

(outdent) i object again as well, on the same grounds as before: making "a tentative edit the page to see whether there is agreement" doesn't make sense when we're already discussing the proposed change; and "we didn't seem to reach a conclusion" before because there isn't any consensus that this change should be made. i grasp your point about some UK newspapers' styleguides favouring MDY format, but that doesn't mean Wikipedia can't opt for DMY format in articles written in UK English if that compromise works well, which it seems to. Sssoul (talk) 05:55, 11 May 2009 (UTC)

I also object. A manual of style needs to make recommendations, or else it isn't worth having. --John (talk) 05:25, 13 May 2009 (UTC)
  • Minimal rocking of the boat is important on this particular issue; a lot of water has gone under the bridge on this one. Yes, consistency within the article is paramount. And, personally (*IMO*) I’d much prefer that we not look to which dialect the editor used as some sort of guide to which date format ought to be used. How our articles read should not be about who wrote it and really ought to be about what the article is about. Since the majority of our readership is non-American, I think it makes perfect sense to use “4 April” since this is the date format used by most English-speaking people and reads most naturally. And I say this as an American. However, fully one-quarter of our readership—on average—is American. I’m quite sure that American-related topics like New York Giants enjoy a predominantly American readership. Accordingly, such an American-branded article should, in my opinion, use the “April 4” format Americans are accustomed to. This is particularly important for Americans since they aren’t as worldly as Europeans and are largely unaccustomed to other date formats—much the same way that readers in Europe are accustomed to two, or even three ways to delimit numbers but Americans know only one. Greg L (talk) 02:01, 17 May 2009 (UTC)

Protection

I'm concerned about that section being protected for so long, and by an admin who's been involved in editing the page. Is there something going on here that I'm not aware of -- that is, any reason normal editing can't resume? SlimVirgin talk|contribs 00:22, 11 May 2009 (UTC)

Note to SlimVirgin: SV, you asked about this a while ago, and I explained it then: the entire page was protected due to edit warring. The move to protect only the "Dates" section arose as a way to free up most of the page while still preventing wars over the dates language. When you asked previously, there was no consensus to unlock the dates text. This has nothing to do with me personally, I just happened to be the one who saw the requests. --Ckatzchatspy 00:58, 11 May 2009 (UTC)

There was no consensus to unlock it among whom, CK? It might be best if this were reviewed by an uninvolved admin -- no disrespect to you, of course, it's just that we're not meant to protect pages we've edited recently. What and when were the edit wars over the date language exactly, because if they were so bad that the page needed long-term protection, then there's no consensus for what's currently on the page. That's perhaps why it would better to allow regular editing to resume, and protect again only if edits wars start up again. SlimVirgin talk|contribs 01:02, 11 May 2009 (UTC)
Please read the previous discussion from when you were seeking to make your change, when I asked if there was consensus to make changes. Also read through Franamax's reply to your post on my talk page. The MoS discussion (where I only acted as an admin, not voicing an opinion either way) did not indicate any such consensus. As for your concerns re: involvement, I'm not - I didn't place the protection on the whole page, any of the seven or so times that it has been fully protected. (The most recent was by GlassCobra, as clearly outlined in the protection log. My action was to unlock most of the page, leaving only the part affected by injunction and edit warring.) --Ckatzchatspy 04:54, 11 May 2009 (UTC)
I think you've shown on several occasions a lack of appreciation of WP:UNINVOLVED (want the diffs?). You appear to have breached this major policy in using admin tools on this page, whether the whole page or the section on which you are a partisan. It is expected in such cases that you ask an uninvolved admin to review the issue rather than act yourself. Tony (talk) 08:36, 11 May 2009 (UTC)
Tony1, this has become annoyingly repetitive - can you please actually review events before you comment on them? I didn't protect the page, I unprotected it (or at least most of it). And guess what - it was at the request of your friend Greg. Furthermore, the solution I proposed - note the word "proposed" - was approved by Greg ("Perfect... That’s a great way restore some sense of normalcy"), A. di M. ("Hallelujah"), and PMAnderson. Now, if you still want to continue with this "spew and hope no-one checks the facts" behaviour, what say I just wash my hands of the matter, revert myself, and leave the entire page the way GlassCobra had set it up - fully protected. Would that satisfy you? --Ckatzchatspy 09:39, 11 May 2009 (UTC)
If this is the case, I apologise. Tony (talk) 10:19, 11 May 2009 (UTC)

International numeric dating

The ISO has proposed the YYYY-MM-DD HH:MM = 24hr clock (or the YYYY/MM/DD) format for international use. The government of Quebec has beeb usung this for years, if not decades, now and the government of Canada is well on it's way of adopting it. Peter Horn 01:51, 12 May 2009 (UTC)

ISO 8601 is unfit for use in Wikipedia because Wikipedia contains many dates before the adoption of the Gregorian calendar, and these dates are usually stated in the Julian calendar. However, ISO 8601 requires that every single date in that format be in the Gregorian calendar or the Proleptic Gregorian calendar. Furthermore, it requires agreement among the parties exchanging data (that is, editors and readers) if using years outside the range 1583 through 9999. No such agreement is in effect among Wikipedia editors and readers. --Jc3s5h (talk) 02:06, 12 May 2009 (UTC)
Wikipedia should meet the average user on his or her own ground, not try to be some kind of Improving Force moving them to more-enlightened ways. And to most non-scientists in the U.S., ISO dates are just gobbledygook. And if you do, as a lay reader, try to figure it out for yourself, you have to decide if the 2nd number is a month or a day, which is why Wikipedia urges fully-spelt out months rather than numeric combinations. [Canadian governments can convince themselves (rightly or wrongly) that they know best for their citizens, and in a bilingual/multilingual country would try to make months clear to non-Francophones and to non-Anglophones.] If you want to encourage something like "1977 May 15", that's something different: it may not be the most familiar format, but it's perfectly clear intuitively (where Julian and Gregorian calendars can't be confused) for years after 999 A.D./C.E. —— Shakescene (talk) 07:36, 12 May 2009 (UTC)
Edit summary said confusing for non-scientific Americans; more likely non-scientific everyone. IMO, WP should stick to date formats that include a spelled-out month; this is not only easier for most readers to comprehend, but unambiguous, since numerical dates of the more non-ISO type (3/4/09) mean different things in different areas (3 April or 4 March). Tony (talk) 08:14, 12 May 2009 (UTC)
I agree. The month should be in writing so that the date is unambiguous. Michael Glass (talk) 11:15, 12 May 2009 (UTC)
Have you (all people here) ever looked upon receipts? They are ISO date-fest, regardless of the country you come from. Saying readers can't and don't understand 2009-04-01 is simply stating something untrue. That being said, ISO dates are currently limited to only a few uses, and I don't see why that should change. Spelled out dates are already the standard. Headbomb {ταλκκοντριβς – WP Physics} 12:14, 15 May 2009 (UTC)
Ah, nope: just got a receipt, and it says "15 May 2009 18:23". Tony (talk) 12:18, 15 May 2009 (UTC)

Speaking of looking at receipts, I'm the treasurer of a Canadian organization with an international membership, and I'm sitting here looking at a stack of cheques (or checks, if you're American). Most of the dates are ISO format. The ISO date format has the overwhelming advantage that it is the only numeric date format that is unambiguous. If you see "07/08/09", you have no idea whether that means "July 8, 2009", or "7 August 2009" unless you know the author's national origin and his personal preferences. If you see "2009-08-07", you know what it means, whether you express it as "7 August 2009" or "August 7, 2009". The latter three mean the same thing and the differences are only cosmetic. If you're writing in English (some of these cheques are in other languages), spelling the month out is the best solution, but the arguments about Julian/Gregorian calenders and readers being too stupid to figure it out are just nitpicking.RockyMtnGuy (talk) 18:16, 19 May 2009 (UTC)

One who understands the ISO 8601 standard, and writes a non-Gregorian date in a publication that professes to use the ISO 8601 standard, is a liar. --Jc3s5h (talk) 18:21, 19 May 2009 (UTC)
We do not profess to use the ISO standard, and we do not have the prior understanding with our readership which the ISO requires before it is used for any date which could be Julian. We should not use any all-numeric format for such dates, whatever this page says. Septentrionalis PMAnderson 18:25, 19 May 2009 (UTC)
I don't think it's a question of readers being "too stupid" to figure it out. I think it's partly a question of undue prominence/distraction. For many purposes the reading of an article is not to find out to the exact day when something happened, or indeed when it happened at all. To kinda distract a reader by having to parse a date-- or even to realise it is a date in order to skip it-- is counterproductive to me. SimonTrew (talk) 08:22, 26 May 2009 (UTC)

Cleanup up poorly formatted dates

First, it is not my intention to start another date autoformatting debate here. Please do not make this a date autoformatting debate.

As a result of recent date formatting and linking polls (here, here, and here), there appears to be a consensus to (eventually) discontinue the use of date links as a means to achieve date autoformatting. (Whether date autoformatting will continue in another form and the degree of date linking to remain is still being debated elsewhere.) While the long term implications of these polls are yet to be determined, it remains a possibility that linked-date based auto-formatting ($wgUseDynamicDates option) will eventually be turned off – not today, not this week, maybe not even this year. If and when this does occur, auto-formatting of linked dates will cease and, these dates be displayed as coded in the wiki-markup.

In the great majority of cases, this will not not a problem, since most dates are coded in the manner consistent with their intended display. Anonymous users will continue to see what they have always seen, and registered users will see what anonymous users see.

However, there are a significant number of cases where the wiki-markup dates are poorly formatted, i.e., they do not have the proper punctuation or spacing. For example, there are 130,000+ pages with dates of the form "[[month day]] [[year]]" (missing comma) and 30,000+ pages with "[[day month]], [[year]]" (unexpected comma), and another 13,000+ pages with various other poorly formatted dates (no space or comma, comma with no space, two or more spaces, etc.). Currently, a side effect of date autoformatting is that even when no date preference is selected, most of these cases are still adjusted to have the appropriate punctuation. For example, both "[[25 May]][[2009]]" and "[[25 May]] , [[2009]]" will display the same as "[[25 May]] [[2009]]" (25 May2009 and 25 May , [[2009]). Many of the pages containing these issues are documented here.

Date auto-formatting also recognizes ISO 8601 like dates for the form "[[2009-05-25]]" and "[[2009]]-[[05-25]]" and will reformat them as if they were coded as "[[2009]]-[[May 25|05-25]]", adding the explicit link targets (2009-05-25 and 2009-05-25).

I propose identifying and correcting the formatting for existing poorly formatted dates, so that they contain the correct punctuation as written thus reducing the dependence on date autoformatting. The corrected dates would also follow the "what you see is what you get" principle, which I consider to be good practice. (An editor need not wonder why "[[25 May]], [[2009]]" displays as "25 May 2009" or "[[May25]][[2009]]" displays as "May 25, 2009".) For ISO 8601 like dates, adjusting the markup so that the presentation is the same, but month-day links would continue to work even if the wgUseDynamicDates option is turned off.

Details:

  1. Linked day-month-year style dates will be updated as needed so as to have a single space between the day-month and year parts.
  2. Linked month-day-year style dates will be updated as needed so as to have a comma+space between the month-day and year parts.
  3. Linked year-month-day style dates will be updated as needed so as to have a single+space between the month-day and year parts.
  4. Linked year-day-month style dates will be evaluated and handled manually. (This is a rare and unusual format.)
  5. Linked ISO 8601 like year-month-day will be split into separately linked year and linked month-day parts.
  6. The linked month-day parts of ISO 8601 like dates will be fixed to correctly reference the date pages. (This will either be done with a piped link, or via redirects from pages named mm-dd. The latter is my preferred approach, but this is open to discussion.)
  7. The current appearance of updated dates will remain unchanged and will continue to be autoformatted as long as wgUseDynamicDates remains enabled.
  8. Date links will continue to function as before.
  9. This proposal does not involve linking or delinking dates. Linked dates will remain linked. (
  10. This proposal does not attempt to resolve inconsistent date usage within articles.
  11. This proposal does not involve dates not currently recognized by the current date autoformatting function. (Example: "[[Sept 1]][[2009]]" which currently displays as "Sept 12009" (abbreviation not recognized).
  12. This proposal does not attempt to fix bad dates that are not currently fixed by the current date autoformatting function. (Example: "[[MAY 25]][[2009]]" which currently displays as "MAY 252009" (redlink due to all caps).
  13. Dates in comma separated lists will be handled manually, if there is any ambiguity in the content.
  14. Year-comma-month-day and Year-comma-day-month will be handled manually, due to possible special intended use of comma. (See 1840s in association football for examples.)

I am currently developing bot to make these corrections, but would like to get the opinions consensus of those who frequent this forum before I proceed. There have been prior related discussions here. I am purposefully limiting the scope of this proposed change in an attempt to avoid related controversial issues.

Finally, it is not my intention to start another date autoformatting debate here. My intention is to reduce our dependence on date auto-formatting so that our options remain open. Please do not make this a date autoformatting debate. Thank you.

-- Tcncv (talk) 06:02, 25 May 2009 (UTC)

Comments

  • Please see this discussion, and the associated data here. You will notice that the data is from a March 2009 dump, and is probably incomplete (for the dmy formatting errors). Ohconfucius (talk) 07:57, 25 May 2009 (UTC)
  • Thank you for the links. I had not seen that side discussion. I am aware of the "list of articles" page and included a link to it above. I have done similar scans and others on the March 2009 dump, which is where I came up with the numbers above. I am presently working off the May dumps. As for date delinking and replacement autoformatting discussions, I am attempting to steer clear of those topics and focus on the technical issues relating to the $wgUseDynamicDates option. Date delinking and potential autoformatting replacement discussions and resolution can proceed in parallel with this proposed cleanup. -- Tcncv (talk) 12:20, 25 May 2009 (UTC)

Chronological order in Wikipedia lists?

Does Wikipedia have a policy on the chronological ordering of years and dates? I am a bit troubled by the inconsistent listing of years/dates, sometimes the most recent can be found on the top of the list, sometimes on the bottom. I've searched high and low in the Manual of Style and other areas of the Help section without finding the answer, so I'd be very grateful if someone could enlighten me. Thanks. Wameya (talk) 09:43, 17 May 2009 (UTC)

Oldest to newest. See Wikipedia:Manual of Style (lists of works). Garion96 (talk) 09:49, 17 May 2009 (UTC)
Thanks, Garion96. The page you referred me to is titled "Manual of Style (lists of works)" and warns against "possible exceptions". Does the oldest to newest rule apply also to events (e.g. sporting events, political events, etc.)? If so, I would say the title is slightly misleading. Sorry to rant about this, but I really found it hard to locate the policy, which perhaps is partly a cause of the inconsistencies.Wameya (talk) 10:04, 17 May 2009 (UTC)
Actually, I've never seen that being enforced in general (perhaps for list of work), but there are plenty of lists that goes from newest to oldest. IMO there is a de facto freedom of choice on this subject. Use whichever makes more sense to you. Personally, I sort references from newest to oldest, and "events" from oldest to newest, but that's just me. Headbomb {ταλκκοντριβς – WP Physics} 13:19, 17 May 2009 (UTC)
I agree with Headbomb. Some go up some go down. I will reorder the lists if things are out of order (and the same for things out of alpha order, typically happens when people carelessly just throw additions at the end), but for some articles it makes more sense to put them newest first. I think here the rule of WP:COMMON overrides WP:MOS. A list battles in a war would best be put oldest to newest, but a list of a football team's results would probably be better newest first. As always the thing to keep in mind is what a reader likely to be looking for. SimonTrew (talk) 14:55, 17 May 2009 (UTC)
Actually there's a very important exception to this. References are listed in the order they appear in the article, not in chronological order. SimonTrew (talk) 18:45, 17 May 2009 (UTC)

When I started work a year ago on New York City mayoral elections, the previous editors had already gone from 2009 backwards to 1929, election by election, which makes sense in this case because more readers (rightly or wrongly) are likely to have interest in Michael Bloomberg (Mayor since 2002) and Rudolph Giuliani (1994-2001) than in Robert A. Van Wyck (1898-1901) or Seth Low (1902-3). This has to be balanced against some confusion from the backward chronology. But in creating several summary tables, I started with 1897 or 1834 and worked forwards in time. I think this is an area where general guidance and advice would be very helpful, but any hard-and-fast rule (or anything which by some unstoppable natural process of Wikinevitability evolves in a virtual rule) would be out of place. —— Shakescene (talk) 20:45, 18 May 2009 (UTC)

I think we understand each other. What would be the general guidance? Suggestion: "Put lists of dates in chronological order. Sort them from oldest to newest unless it makes more sense to sort them the other way round". That's clumsy but just a starter. SimonTrew (talk) 21:45, 18 May 2009 (UTC)

But when does it make more sense? In my POV it never makes sense to use newest to oldest. A disadvantage of common sense, not everyone's common sense is alike. :) Garion96 (talk) 21:47, 18 May 2009 (UTC)

And surely it's not hard to look at a list and realise it goes up or down in chronology? SimonTrew (talk) 21:48, 18 May 2009 (UTC)

Oh that, no I think that should be possible.... I wasn't talking about that. Garion96 (talk) 21:51, 18 May 2009 (UTC)

Garion it makes more sense for exactly the point User:Shakescene made: the reader. They are more likely to want to read about the current or previous mayor than the one from the century ago, and mutatis mutandis for football results or whatever. Aren't we here editing to make things the best we can for readers? SimonTrew (talk) 21:51, 18 May 2009 (UTC)

I am also a reader and I don't like newest to oldest. Even when I compare it to other sources, the only one I know which uses newest to oldest is IMDB. Garion96 (talk) 21:54, 18 May 2009 (UTC)

In the case of long lists with commentaries or details, you're really balancing two inconveniences: unnatural order vs time to reach what interests the reader most (or first). The article or listing of New York City mayoral elections is quite long, so it makes sense to put what interests the largest number of likely readers, and about which we have the widest range of reliable information, closest to the top. (The election of 1897 is interesting enough to click or scroll down for, but it won't attract so many initial readers.) However in my overall tables which take up one screen (or at most 2-3), chronological order is more convenient and best for showing changes over time. I'd think the same considerations would apply to something like a summary of all the FIFA World Cup Finals since 1930. Guidance to an editor for these kinds of question can be quite helpful, but a single rule (or even set of rules) wouldn't fit. —— Shakescene (talk) 00:31, 19 May 2009 (UTC)

I don't know if this is of relevance but I am just throwing this in while I think of it. If you look at UK railway station articles (e.g. Dullingham I am deliberately choosing one I know and is short) the current information is listed before the previous information. That just makes common sense, I think. I kinda agree with Garion that common sense is not as common as all that i.e. others have a different common sense, but anyone looking at that article can find where the route runs first without scrolling down. As I say I deliberately chose a very short one but I am just using it as an example, this pattern is followed in all railway station articles for the UK (no doubt there are some exceptions because of oddities or errors but this is the rule of thumb). I just mention it as another example in case it is useful, and probably have not put it very well. SimonTrew (talk) 16:24, 19 May 2009 (UTC)

Since pitching my question, I've been following the discussion sitting on my hands lest I should declare my personal preference (in certain cases) for newest to oldest lists. Of course, my preference is as irrelevant in this as any other’s personal preference, be it an editor or a reader. What I make of your exchange, however, is that a WP should perhaps be flexible enough that the contributor make a decision on the matter (thanks for the NYC mayor and FIFA WC Finals examples, Shakescene, I think you make a good point there). Now back to my initial frustration, is this already implicit in the Manual, or is there a need for formal clarification on the matter? (I’m a Freshman editor, sorry if I’m not up to speed on protocol here.)Wameya (talk) 22:09, 19 May 2009 (UTC)

¶ Let me say in defence of Garion96's point, that I don't actually like the way that New York City mayoral elections is organized because I have to keep thinking backwards. However, the newest-first order (2009-1929) which I found when I first saw it seemed most practical for most likely readers. If Garion's interested in collaborating on a demonstrative reconstruction (on one of my sandbox pages or one of his or hers) of the mayoral election page with oldest-first (starting in 1897), I'd be very interested to see how it would look. —— Shakescene (talk) 20:06, 21 May 2009 (UTC)

¶ Perhaps in the case of those chronological tables where this is feasible, there might be language recommending sortable tables for just the reasons adduced above. Newest-first is good for some things, but oldest-first is more natural and easier to follow mentally through time. Where the editors can offer the reader a choice of orders to suit his or her needs, we might suggest doing so. —— Shakescene (talk) 05:24, 23 May 2009 (UTC)

I'd thought of that, but does sorting work for dates? SimonTrew (talk) 11:06, 26 May 2009 (UTC)
Sorting works fine on dates in yyyy-mm-dd format, which is allowed in tables by the WP styleguide. −Woodstone (talk) 11:21, 26 May 2009 (UTC)


Hi. Just a note that you can have a hidden sort value in a table. For example, instead of a table such as:

{|class="wikitable sortable"
||Number
||Date
|-
||1
||4 Jan 2009
|-
||2
||30 Jan 2009
|}

which produces:

Number Date
1 4 Jan 2009
2 30 Jan 2009

you can have:

{|class="wikitable sortable"
||Number
||Date
|-
||1
||{{Hs|2009-01-04}}4 Jan 2009
|-
||2
||{{Hs|2009-01-30}}30 Jan 2009
|}

which produces:

Number Date
1 2009-01-044 Jan 2009
2 2009-01-3030 Jan 2009

In the second case, the Date column sort behaves as expected (try sorting both date columns in ascending order to see the difference). For an example of this technique used big-time, have a look at List of compositions by George Frideric Handel. Cheers.  HWV258  22:56, 26 May 2009 (UTC)

The {{Hs}} template appears to be a recent duplication of functionality available in the {{sort}} template. Templates {{dts}} for dates, {{nts}} for numbers, and {{sortname}} are also commonly used templates for sorting. I suggest we not promote duplicate functionality. -- Tcncv (talk) 04:28, 28 May 2009 (UTC)

Singular or plural for values from –1 to 1? (template:convert)

This relates to a question I posed on the Template:Convert talk page. Which one (assuming only one) is proper: "0.0074 acres" or "0.0074 acre"? Based on my my first-grade teacher's rule that the plural is for values greater than one (also less than -1, I presume), then singular is proper. However, because all such decimal examples start with 0, does the "zero takes the plural" standard govern? My example is based on the phrase "containing 0.0074 acres more or less" from Calvin Coolidge's proclamation #1745 (1925 September 5) establishing Father Millet Cross National Monument. However, the current U.S. Government Printing Office Style Mannual gives an example of "0.25 inch" (Chapter 12.9d). — Eoghanacht talk 01:37, 22 May 2009 (UTC)

Question two: Dropping the assumption that either one or the other is proper, could they both be correct, is it a dialect thing?
Question three: Should we draw a distinction between fractions and decimals?
My view is that fractions between +1 and −1 should be singular (being read as "half a metre", "a quarter of an ounce", etc.) but that decimals are a different beast altogether. I would say that "0.5 kilograms" is not incorrect but that this need not imply that "0.5 kilogram" is incorrect. Perhaps it's a dialect thing or something up to personal choice. JIMp talk·cont 02:28, 22 May 2009 (UTC)
In my opinion decimals are always plural and fractions between 0 and 1 always singular; "the ingot weighed zero point two five kilogram" is definitely incorrect to my ears, whereas "the ingot weighed one quarter kilogram" is fine (and 3/4 kg would be pronounced "three quarters of a kilogram" and thus be singular as well). It could well be a cultural distinction, given User:Wavelength's suggestion of "zero point seven five of a kilogram" on the other thread. Chris Cunningham (not at work) - talk 08:24, 22 May 2009 (UTC)
Yeah, it does seem to be a dialect thing maybe. I would not say "one quarter kilogram" but "a quarter OF a kilogram (but would write, for abbr=on, 1/4 kilograms). "kilogram" might edge it here because one can always just manually stick the "s" on the end but can't take it away.
I would say though decimals should be plural, "0.75 kilograms". I think what I have written here is "BBC English" if you see what I mean.
There's difficulties maybe with time and other measure e.g. "a quarter hour" or "a quarterpounder" but I am not sure what kinda grammatical rules those come under. The first, meaning 15 or 45 minutes past the hour (i.e. a measurement of an absolute point in time, not a duration) is the norm in the UK (though the other would be perfectly understood), similarly "I'll see you at half [past] six" but "it took half an hour to complete". It all starts to look quite complex. The time thing may just be an exceptional case that can be entirely discarded, and similarly a quarterpounder is, to a certain degree, simply a name for a recognised product and not really a measure (though I think on the McDonald's page it is noted that in countries that don't use pounds it may be called other things).
Just throwing in some thoughts for discussion. SimonTrew (talk) 21:56, 22 May 2009 (UTC)

For some reason I feel that the plural is more "natural" except for phrases which use the indeterminate article right before the noun, e.g. "zero point five kilograms" but "half a kilogram" (probably because it's more-or-less what my native tongue does). OTOH on The Feynman Lectures on Physics we read stuff such as "1.782 × 10−27 gram" and "10−12 second". Might it depend on whether the writer expects the reader to pronounce it as "ten to the minus twelve seconds" or as "a trillionth of a second"? Or is it just another American–British difference? I begin to suspect the latter. --A. di M. (formerly Army1987) — Deeds, not words. 10:23, 23 May 2009 (UTC)

That's what I'm beginning to suspect. JIMp talk·cont 13:07, 23 May 2009 (UTC)
I wonder I'm just muddying the waters, but there may be a parallel with euro currency here. When that currency was introduced it was widely understood that the plural of euro "must" be euro. In fact that is only true for legislative purposes: in other documents people are free to write 10 euro or 10 euros in English. It still seems to be quite up in the air and different publications seem to have different style rules. This is outside the scope of Convert template but just wonder, seems to me MOS would have to make either an arbitrary but consensual decision on the matter, or just stay silent on the matter. Perhaps the same should be done here: there are bigger fish to fry.
Similarly plural of slang "quid" (pound sterling) is always "quid" not "quids". SimonTrew (talk) 07:32, 26 May 2009 (UTC)

From here and the other talk page, there does not seem to be enough concensus to set a rule one way or the other. From a programming perspective for the template, I presume it is easiest just to leave it as-is (plural). — Eoghanacht talk 15:03, 28 May 2009 (UTC)

Singular is much more logical, isn't it. Plural is often used, including by me until I read on MOSNUM back in 2006, when I rewrote the page, that singular was the recommendation. It's the first I've known that it has since been changed to plural. I think it should be only a recommendation or a statement that both are used. Personally, I'd prefer the singular. Tony (talk) 07:37, 29 May 2009 (UTC)

Full-date unlinking bot proposal

Hello. I am working on a proposal for a bot to carefully unlink dates, hopefully clearing main remaining dispute. I would unlink full autoformatting-style dates such as 5 November 1989, while leaving e.g. 1989 and November 5 alone. I would be happy if you would take a look at Wikipedia:Full-date unlinking bot and comment. It is still just a draft, but everyone is invited to point out mistakes and suggest improvements to the proposal. I plan to start an community RFC soon, which will be widely advertised and open to general opinion. --Apoc2400 (talk) 19:31, 29 May 2009 (UTC)

Given that ArbCom has yet to complete its decision on the date issue, and that the injunction against delinking dates is still in place, this seems very premature. --Ckatzchatspy 19:48, 29 May 2009 (UTC)
Premature? Given that we decided to do this almost a year ago, and have repeatedly re-decided to do it since, I think long overdue is a better description...--Kotniski (talk) 19:51, 29 May 2009 (UTC)
The injunction will surely end long before this bot would start. I wish to start the discussion early so that everyone concerned can voice their opinion, even those who might get restricted after the arbcom case ends. Also, ongoing arbcom cases are not supposed to halt all community discussion. --Apoc2400 (talk) 19:56, 29 May 2009 (UTC)
Discussion is fine, of course. However, given that the ArbCom decision will directly affect the conditions under which any 'bot would operate, how do you propose to stage an RfC when these conditions are not yet known? --Ckatzchatspy 20:03, 29 May 2009 (UTC)
I don't see any proposed decision that would prevent the bot I am suggesting so it should not be a problem. Starting the discussion now has the advantage that the arbs can comment if they wish before their decision is set in stone. --Apoc2400 (talk) 20:13, 29 May 2009 (UTC)

As of 22:04, 29 May 2009 (UTC), Arbcom has not yet approved of a proposal that would restrict date delinking in general, only for specific users. Dabomb87 (talk) 22:04, 29 May 2009 (UTC)

Em dashes for open date ranges

This revert summary linked here, but I can't find the exception it refers to. Is it just me? —JAOTC 17:53, 30 May 2009 (UTC)

I agree with you. Everything I see indicates that an en-dash should be used. I looked back at year-old versions of MoS and MoS:DATE and the guidelines are virtually unchanged. The closest I came to a counter example was the sentence-stopped-short example in dash, but even that is an example of interruption, while the open-ended date range is not. -- Tcncv (talk) 06:25, 2 June 2009 (UTC)
Yes, en dashes please; see WP:MOSDASH. Tony (talk) 06:35, 2 June 2009 (UTC)
prob here is essentially that an en dash (which should be half the widthh of an em dash) varies a lot in its display depending on your browser etc. I tend to use em dash because it is much clearer what is meant. And I am a typographer so i actually have done this in hot lead. I disagree with en dash, an em dash is better, but if MOS says en dash i will go with that. 10:49, 6 June 2009 (UTC)
I'll ask SandyGeorgia's advice; I'm pretty sure she knows where this debate was previously located. Tony (talk) 16:08, 6 June 2009 (UTC)
Why is there an endash and a colon? I don't recall where the discussions were, and the archives here are so extensive it would be hard to find them, but I do recall that we settled on an endash rather than emdash. But remove the colon. SandyGeorgia (Talk) 17:25, 6 June 2009 (UTC)

Decimals

8¼ mi or 8.25 mi? I don't see this covered in the manual itself. Thanks for any help. —Anonymous DissidentTalk 13:18, 31 May 2009 (UTC)

I don't know if the manual covers it, but fraction characters like ¼ should not be used because
    • they are hard to read
    • articles change frequently; if it is necessary to add a new fraction to the article that is not avaliable as a character, such as 8 78 mi then the poor editor who adds it will have to go back and change all the existing fractions for consistency (Note this bullet shows how to use the {{Frac}} template.)
--Jc3s5h (talk) 15:20, 31 May 2009 (UTC)
The Symbols inserter below the edit summary box includes halves, thirds, quarters and eighths (½, ⅓, ⅔. ¼, ¾, ⅛, ⅜, ⅝ & ⅞). One generally assumes (rightly or wrongly) that a symbol down there would be one that Wikipedia accepts. As for the substantial question, I'm not sure there's a good answer. 8.25 miles (let alone 8.25 years) is far too formal and precise (and sometimes too pompous) for many purposes, and can indicate far more precision than justified, but 8¼ and 8 ¼ render poorly in many fonts, sizes and systems, while 8 1/4 and 8-1/4 can look ugly and lead to all kinds of ambiguity and uncertainty. —— Shakescene (talk) 00:12, 1 June 2009 (UTC)
I'd use fractions very cautiously. Is the decimal form over-formal? Tony (talk) 07:25, 2 June 2009 (UTC)
At least in principle, "8 14 miles" suggests the measurement was made to within one quarter of a mile, whereas "8.25 miles" suggests it was made to within 0.01 miles or not much worse than that, which is likely (but not necessarily) over-precise. --A. di M. (formerly Army1987) — Deeds, not words. 15:22, 2 June 2009 (UTC)
One exception (which needs just this sort of distinction from everyday working measurements and approximations) would be the dimensions of race tracks, swimming pools, test beds and similar athletic or scientific facilities. They may come out in nice round numbers, but they were consciously (Intelligently?) designed to be both round and precise. Few people are interested in comparing (say) Sir Roger Bannister's time over 1,757 yards with Sir Christopher Chataway's over 1,762. —— Shakescene (talk) 18:32, 2 June 2009 (UTC)

I think fractions are good when something is defined as a fraction, especially if it's thirds. There was a bit of discussion of this at Talk:1950s Topps/Archive 1. --Apoc2400 (talk) 18:39, 2 June 2009 (UTC)

In general, you should use decimals, rather than fractions, unless you have a good and sufficient reason to use fractions. And as for deducing the accuracy of a number, let me say that in 1852 the British Trigonometrical Survey of India measured Mount Everest's elevation as 29,000 feet above sea level. They changed it to 29,002 feet so that no one would think they had rounded it off to the nearest thousand. The official value is currently 29,029 feet. In other words, you can't really tell what the accuracy is unless it is explicitly stated.RockyMtnGuy (talk) 18:46, 2 June 2009 (UTC)
Yes, but misleading the reader into believing that the precision is worse than it actually is is usually a lesser evil than vice versa. If a value is 2.7245(32)×1010, stating "about 27 billion" usually makes more sense than "27,245 million". (As a rule of thumb, I don't state values with errors of more than 1.5 units in the last place unless I'm explicitly stating the error too.)--A. di M. (formerly Army1987) — Deeds, not words. 00:06, 3 June 2009 (UTC)
(Of course, what I stated above about false precision doesn't apply to defined values; they should preferably be stated the way they are stated by whichever standard or ruling defines them, but any way of stating them is OK as long as they're mathematically equivalent. --A. di M. (formerly Army1987) — Deeds, not words. 00:06, 3 June 2009 (UTC))

WP:UNLINKDATES

I still don't see consensus for the copy there; it was not actually discussed in the most recent RfC. I don't see consensus for my revision, either, but it would make more difficult semi-vandalism creating a fait accompli by removing the relevant information from the year article, and then unlinking the year. I'm not intending to imply that these edits have consensus, but I think they should be made.

I don't see a specific finding in any of the RfCs that years of birth and death should not be linked, but there is clearly no consensus that they should be linked, even though I believe Wikipedia would be improved by those links. I see a weak consensus that Saint Patrick's Day should link to March 17, which seems to be contradicted by the text, although "as any other link" seems ambiguous as to whether it means "as any other date-link" (in other words, subject to the rest of the paragraph), or "as to any other link" (disregarding this paragraph, and applying only the general linking guidelines).

The separate question of whether the birth year and death years, or the year of creation of an object (album, song, bridge, whatever) in infoboxes or whether the year of succession should be linked in succession boxes hasn't been brought up yet, and I believe it needs to be. The text is also ambiguous, in my opinion, as to whether 2008 in sports should have month-day links, or links such as [[2005 in sports|2005]]. — Arthur Rubin (talk) 00:43, 27 May 2009 (UTC)

Linking
Month-day linking
Month-day articles (February 24 and 10 July) should not be linked unless their potential content is germane and topical to the subject. Such links should share an important connection with that subject other than that the events occurred on the same date. For example, editors should not link the date (or year) in a sentence such as (from Sydney Opera House): "The Sydney Opera House was made a UNESCO World Heritage Site on 28 June 2007", because little, if any, of the contents of either June 28 or 2007 are germane to either UNESCO, a World Heritage Site, or the Sydney Opera House.

References to commemorative days (Saint Patrick's Day) are treated as for any other link. Intrinsically chronological articles (1789, January, and 1940s) may themselves contain linked chronological items.

Year linking
Year articles (1795, 1955, 2007) should not be linked unless they should contain information that is germane and topical to the subject matter—that is, the events in the year article should share an important connection other than merely that they occurred in the same year. For instance, Timeline of World War II (1942) may be linked to from another article about WWII, and so too may 1787 in science when writing about a particular development on the metric system in that year. However, the years of birth and death of architect Philip C. Johnson should not be linked, because little, if any, of the contents of 1906 and 2005 are germane to either Johnson or to architecture.
Can you at least wait until the bodies have started to cool from the ArbComm date linking debacle? More germane to your assertions, the proposed ArbComm decision is currently tending towards affirming this text: Two RFCs held in December 2008 reaffirmed that date autoformatting is undesirable, and that WP:OVERLINKing of dates is not desirable, however consensus has not been found on when dates should be delinked. While ArbComm may ultimately drop that from the proposed decision, or go with alternate text, it's likely that that will be the starting point of the next round of consensus-seeking. Personally, I don't see birth/death day links as improving the encyclopedia at this time. Studerby (talk) 01:12, 27 May 2009 (UTC)
Agree.  HWV258  01:36, 27 May 2009 (UTC)
See this RfC which resulted in (approx.) 42% for "support" and 58% for "oppose".  HWV258  01:36, 27 May 2009 (UTC)
Arthur, you make some good points. Here are my responses:
  • Labeling the series of semi- and fully automated edits to delink dates "semi-vandalism" is a not a good faith comment. Can you rephrase?
  • I think that if we tried to spell out every instance of when dates should or should not be linked, we'd have editors on both sides of the debate trying to look for loopholes, and a lot of unnecessary discussions would crop up. Of course, my proposal still stands, but the community seems to favor the broader guidelines. I do not support the linking of birth and death dates, but that's not the point I'm making here.
  • I don't mind strengthing the language that supports linking chronological items in holiday and date articles.
  • I do think that the year itself should be linked. Hidden links in those articles are OK, I guess. I'm not sure about linking month-day fragments. Dabomb87 (talk) 01:46, 27 May 2009 (UTC)
  • The community appeared to be a lot less favourable to linking md/dm than the year in the Ryan RfC, but that still does not translate into wanting to link years in general, or even years of birth or death. Ohconfucius (talk) 02:22, 27 May 2009 (UTC)
  • The semi-vandalism I'm talking about is hypothetical; I'm not suggesting that any of the existing parties in the RfCs or RfAr would engage in such a procedure, but I'm trying to reduce the ability to game the system to remove any links which are appropriate. — Arthur Rubin (talk) 07:37, 27 May 2009 (UTC)
  • Arthur, can you explicate here exactly what the insertions of potential and should add? It seems to be a vague epithet, especially to an editor who comes upon it without the background. Tony (talk) 11:00, 28 May 2009 (UTC)
    • (Thanks to whoever boxed the proposal.) What I'm trying to do here (pending a formal agreement as to whether option #1, as written, really has consensus) is: Assuming that 1999 were relevant to a not-obviously-chronological article, what is to prevent a vandal from deleting the relevant material from 1999, and a conscientious editor then removing the link to 1999, per this MOS. Even if the material is restored to 1999, the link will probably be gone indefinitely. Perhaps we should refer to a stable version of the year or month-day page? — Arthur Rubin (talk) 23:51, 28 May 2009 (UTC)

Oh for fuck's sake, not another one of these. Can't you even wait until ARBCOM is done? This is pretty damned near trolling at this point. This also applies to the user below. Headbomb {ταλκκοντριβς – WP Physics} 21:24, 29 May 2009 (UTC)

Nonsense. Arbcom is explicitly not ruling on content of the MOS, and this is a content question. — Arthur Rubin (talk) 23:35, 5 June 2009 (UTC)
arthur my personal opinion is that date linking generally is useless but births and deaths would be very useful to link for on this day kinda pages. Yeah, there is no consensus. I am just adding my opinion here. SimonTrew (talk) 10:44, 6 June 2009 (UTC)
Links should be used to help readers understand the subject they're reading about, not so that readers can access a trivia (we have the search button and the "On this day" section on the Main Page for that). Indeed, if more year articles looked like 1346, the "trivia" and "on this day" argument would become moot. If you read the responses at the latest date linking RfC, many of the opposers to date links were opposed just because of their sole use as an "on this day" article. Dabomb87 (talk) 00:50, 12 June 2009 (UTC)

Problem in "Other date ranges"

I tend not to agree with the line "The form since 1996 should be used in favor of 1996–present in article text and infoboxes." in the "other date ranges" section... particularly in the case of infoboxes.

In most infoboxes the value is set to "Years active:" or some other form thereof, implying that a year range be specified. Something like "since 1996" is not a range per se, as opposed to "1996–present", which implies a range, i.e. "1996 through to the present". Also, I think placing the year first (before the word "since") looks better and provides the needed information more prominently. What are thoughts on this? Has this already been discussed widely? It should be noted that I have no problem with "since 1996" in the article text itself, as certainly "since 1996" flows better in paragraphs.

And just to add, I don't see the MOS's issue with noting "–present" in date ranges. Sure, "the present" is a constantly moving target, but we know that "present" will simply be replaced with the appropriate date/year when the tenure is completed; i.e. "1996–present" will become "1996–2009" (example) when appropriate. — `CRAZY`(lN)`SANE` 20:45, 4 June 2009 (UTC)

I have no objection to "since 2001", etc., and often use it myself, but since there's an inherent imprecision or ambiguity in intended and understood meaning, I don't think it should be positively recommended to the disadvantage of other forms.
The imprecision comes from the fact that it's not always clear whether all or part of "2001" is included in "since 2001" or not. Let me give some hypothetical examples which are, however, neither far-fetched nor patently false to show the ambiguity:
  1. Michael Bloomberg has been Mayor of New York City since 2001.
  2. Michael Bloomberg has been Mayor of New York City since 2002.
  3. Rudolph Giuliani has not been Mayor of New York City since 2001.
  4. Bill Clinton has not been President of the United States since 2001.
  5. Since 1999, the Christian world has been in its third millennium.
  6. Since 2000, the Christian world has been in its third millennium.
  7. Since 2001, the Christian world has been in its third millennium.
The salient underlying (if that's not a mixed metaphor) facts are that:
  1. The Clinton presidency gave way to the Bush presidency at noon on 20 January 2001.
  2. Rudolph Giuliani's term as Mayor of New York (elected in November 1993 and 1997) ran from 1 January 1994 to 31 December 2001.
  3. Michael Bloomberg's first term as Mayor (elected in November 2001 & 2005) began on 1 January 2002.
  4. Some count the third millennium as beginning on 1 January 2000 (the popular version) and others as beginning on 1 January 2001 (the more pedantic version).
—— Shakescene (talk) 02:57, 6 June 2009 (UTC)
Shakes i miss your point a bit here. I can see the 2000/2001 distinction (I am on the pedantic side but still had a bloody good party on 2000 new years) but how come 1999 i don't get what you are arguing there, can't possibly see how it could be 1999.
One of the more interesting things is the change of calendar in 1582 in catholic countries and 1752 in britain and her colonies. you would think for example in gullivers travels when gulliver moved around between those times he would note the difference (he is constantly mentioning his wristwatch) but never does. neither does sam johnson or an other writer from that time that i have read. SimonTrew (talk) 11:10, 6 June 2009 (UTC)
If the second millennium according to overwhelming popular (but not to hyper-pedantic) usage ended on 31 December 1999, then we've been in the third millennium since then (1999). —— Shakescene (talk) 23:03, 6 June 2009 (UTC)
Sometimes "since" is used as a synonym for "after" and sometimes it isn't, which leads to the ambiguity. When I (correctly) write that "every Republican presidential candidate since 1964 has carried Kansas" you can't tell for certain how Kansas voted in 1964. I can also, without contradicting reality, write "every Republican presidential candidate since 1968 has carried Kansas", since in fact Kansas broke stride to vote for Democratic President Lyndon B. Johnson against Senator Barry Goldwater (R) in 1964 (and for the Republican former Vice-President Richard Nixon in 1968). With the exception of 1964, Kansas has voted for every Republican presidential candidate since 1936 (when she voted against her own Republican Governor Alf Landon to re-elect Democratic President Franklin D. Roosevelt), or is that more properly, since 1940? —— Shakescene (talk) 23:03, 6 June 2009 (UTC)
  • Date-ranging to "the present" is bad practice on WP, as acknowledged in the style guides. "The present" is no longer the present next year. Tony (talk) 09:57, 7 June 2009 (UTC)
    But in (almost?) all cases the thing that the range applies to will still be from a certain start point to "the present" at any given future "present" until it comes to an end, at which point neither "since" nor "the present" will be correct and the range will have defined start and end points. For example, as of the timestamp of this comment:
    1. The current monarch of the United Kingdom has been part of the House of Windsor "from 1917 to the present"
    2. The current monarch of the United Kingdom has been part of the House of Windsor "since 1917".
    Both statements are currently equally true, and unless something drastic happens, both will continue to be equally true in 1, minute, 1 week, 1 year, 35 years, etc from now. However, as per Shakescene's point, you cannot determine from statement 2 whether the monarch of the United Kingdom in 1917 was a member of the House of Windsor or not - hence it is ambiguous.
    For something with a forseeable end:
    1. The Czech Republic has held the presidency of the Council of the European Union from January 2009 to the present
    2. The Czech Republic has held the presidency of the Council of the European Union since January 2009
    Both are currently true, and will continue to be true until 23:59 (UTC?) on 30 June 2009. After then though, both statements will be equally incorrect until changed to
    • The Czech Republic held the presidency of the Council of the European Union from January–June 2009
    Thus the only difference is that one statement is more ambiguous than the other - i.e. based soley on statement 1, you cannot definitively answer the question did the Czech Republic hold the presidency in January 2009? Thryduulf (talk) 15:18, 9 June 2009 (UTC)
The "since" version is preferable, don't you think? Tony (talk) 15:42, 9 June 2009 (UTC)
No, exactly the opposite. The "since" version is ambiguous and so I prefer the "to present" form. In prose we should, imho, generally discourage "since" except in situations where both precise dating isn't the overriding concern and it is stylistically the better choice (for example "since the mid 14th Century" is fine). In infoboxes I think the "to present" form should be mandated. Thryduulf (talk) 16:52, 9 June 2009 (UTC)
I know you gave an example above where it does matter, and deserves to be spelled out. But in most cases, it's perfectly clear that "Since 1946" includes 1946, and doesn't start from 1947 ... isn't it? Tony (talk) 17:34, 9 June 2009 (UTC)
I believe that the use of the phrase "Since X" is generally understood to be inclusive of X rather than exclusive. While the "From X to Present" is somewhat less ambiguous as far as that is concerned I do not think the potential for confusion warrants a mandate. Both constructs are equally valid and I seriously doubt that it is a big enough issue to merit a change to the MoS. Shereth 17:42, 9 June 2009 (UTC)
I agree that "since" is preferable, and that there really isn't any reason to change the MOS. Plastikspork (talk) 21:06, 9 June
but "since" can sometimes lay false scent. Since 2001 New Year fell on a Sunday... the reader is temporarily sent the wrong way (and I don't know it if did there must be a calendar on this ne computer but I can't seem to find one!) Personally I always write "because" for that reason UNLESS i am talking about dates. I am not sure if that point is relevant to MOSNUM but perhaps more widely to MOS saying do not use "since" EXCEPT for date ranges. Of course I am putting it very roughlky here I am not suggesting that as an exact wording. SimonTrew (talk) 03:17, 19 June 2009 (UTC)