Wikipedia talk:Manual of Style/Dates and numbers/Archive 134

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

STRONGNAT for Canada

"For the US this is month before day; for most others it is day before month. Articles related to Canada may use either format consistently." What's the reference for this? Check every major Canadian media outlet (CBC,CTV,Global,National Post,Globe&Mail,Sun,CPAC,ad nauseum) and you'll find MDY used by all of them, not DMY. Ditto for the major political party websites. Everyone from the Prime Minister to my digital TV guide seems to use MDY, it's far and beyond the most widely used format in Canada. French-speaking Canada seems to be the only ones that really uses DMY, and since this is the ENGLISH version of Wikipedia that's not a reason to use the odd format. --TheTruthiness (talk) 18:45, 8 April 2011 (UTC)

Well, the English front page for H.M. Government in Canada seems to be inconsistent (and thus un-Wikipedian):

26 March 2011
The 41st federal general election will be held on Monday, May 2, 2011. Canadian citizens at least 18 years old on election day are eligible to vote.
[ More ]

18 March 2011
Prime Minister Stephen Harper made the following remarks on the situation in Libya

On the other hand, the English front pages of Statistics Canada and the Parliament of Canada , as well as the body of Canadian Hansard (e.g. here), seem to be pretty consistently MDY.
So as a non-Canadian, I'm not completely sure that it would be OK to change the first major contributor's DMY style to MDY under WP:ENGVAR, but maybe the tide has shifted so heavily to MDY that Canadian articles should be treated like U.S. ones. —— Shakescene (talk) 20:33, 8 April 2011 (UTC)
Long-standing consensus is that Canadian topics are treated as WP:DATERET, not formally establishing either the US (MDY) or universal/international (DMY) date standards. For a reliable source on this, see The Canadian Style], section 5.14 Dates which indicates validity of either format and explains some inconsistency in Canadian government usage e.g. Environment Canada's weather uses DMY/universal (with at least one instance of ISO 8601) formatting. A discussion on the Canadian boards last year (and likely well in the past) generated plenty of non-consensus to change things; best to leave that can of worms unopened. Dl2000 (talk) 03:54, 9 April 2011 (UTC)
Canadian post office uses dmy. But let me tell you from my VAST experience of gnoming date formats at WP, that Canada-related articles overwhelmingly use mdy format. Indeed, I really notice as an oddity articles that use dmy. I see many francophone-related article on the en.WP using mdy too. But I presume no one is proposing to change the existing guide line, which gives the option. Tony (talk) 06:44, 9 April 2011 (UTC)
I have attempted to gain consensus to use an American, mdy long date format, but pretty much any article that [[user:Miesianiacal] has worked on will have a European, dmy format. I started a catalogue at Wikipedia talk:Manual of Style (Canada-related articles)#List of date formats used. You'll see that not all institutions use American format. Feel free to expand the list, but I don't believe that you'll gain consensus for a change soon. --Walter Görlitz (talk) 07:06, 9 April 2011 (UTC)
The Canadian military usage standard is DMY and has been since the 1960s or earlier. I don't see any reason to change the existing MOS, which captures the usage in Canada very accurately. - Ahunt (talk) 12:11, 9 April 2011 (UTC)
It isn't actually specific to the Canadian military, it's a NATO standard. Roger (talk) 12:23, 9 April 2011 (UTC)
I've been involved with this issue before, and it's very similar to the BC vs. BCE issue, where one editor will use one format and another will revert it to the other format, with an edit war ensuing. The Canadian Style guide that Dl2000 cited actually recommends "August 17" but not "17 August" for a specific month and day without a year, which is strange, so if we were to go by that guide we should prefer the US style for Canada-related articles. But I know we'll never get consensus either way (heck, the government can't even pick one over the other), so it's going to stay as-is. — CIS (talk | stalk) 00:45, 11 April 2011 (UTC)
Furthermore, at the bottom of we have: Date Modified: 2011-03-30 -- a format widely used by gov't webpages. Also in wide-spread use in Canada: 2011-MAR-30. It is somewhat non-global & exclusionary that WP:MOS does not accept YMD, an internationally used format. Btw, saying DMY is "universal format" is a misnomer, as is saying YMD is ISO 8601. --JimWae (talk) 20:06, 17 April 2011 (UTC)
That's just a webpage software thing, no official gov releases use year first. Plus we're taking more about when just month/day is mentioned. --TheTruthiness (talk) 06:19, 18 April 2011 (UTC)

These, and many more, are year first:

--JimWae (talk) 06:42, 18 April 2011 (UTC)

The Canadian Style Guide (mentioned above), also covers YYYY-MM-DD & mentions it may be appropriate for bilingual documents - increasing its STRONGNAT claim a bit there --JimWae (talk) 08:59, 18 April 2011 (UTC)

  • I'll just repeat a post I made earlier, above: "while dd mmm yyyy date format is by no means universally British – the mmm dd, yyyy format is used by some important British newspapers, the consensus is that British articles should universally use dd mmm yyyy, 'despite the sources'. This is the Wikipedia Manual of Style, it may be influenced by other external style guidelines, but remains 'ours'. "

What do we really mean by format consistency?

What do the presently assembled make of Dreaming of You (album) – a GA nominee? On the surface, the article complies with MOSNUM... but is this what we mean? --Ohconfucius ¡digame! 04:27, 21 March 2011 (UTC)

A two-minute scan through looking mainly at numbers reveals: "1994-1995" should be "1994–1995" according to WP:ENDASH. "December, 24 1995" and "January, 8 1996" should be "December 24, 1995" and "January 8, 1996" (WP:DATESNO doesn't explicitly rule out "December, 24 1995" but I would think any basic book on punctuation would, and none of DATESNO's examples look that way. Consistency isn't the issue; the article's other dates are punctuated normally.) "#2", "#21", and several occurrences of "#1" should be "No. 2", "No. 21", and "No. 1" according to WP:NUMBERSIGN. I don't see how consistency is an issue, so what did you have in mind? Art LaPella (talk) 04:53, 21 March 2011 (UTC)
  • Apologies for my lack of precision. On my haste, I neglected to mention I was referring to the use of mdy dates in the body of the article and dmy dates in the refs section. --Ohconfucius ¡digame! 05:16, 21 March 2011 (UTC)
The guideline states that "Dates in article body text should all have the same format." and "Dates in article references should all have the same format." This does not prescribe the same format for both in the same article. Chris the speller (talk) 15:17, 21 March 2011 (UTC)
  • Well yeah, ALP didn't notice it, but such an arrangement sure looks odd to me. So I just wanted to be sure that's what we mean to allow. --Ohconfucius ¡digame! 04:07, 22 March 2011 (UTC)
I didn't say I liked it, or that it looks right to me, and it sure messes up programs I had developed to make dates uniform within an article, but that's what it says, so I gotta live with it. Happy editing! Chris the speller (talk) 13:25, 22 March 2011 (UTC)
  • So the editors decided to use Mdy in prose and dMy in references. Both are consistent among themselves and comply with MOS. I don't see a problem with this article. —  HELLKNOWZ  ▎TALK 13:35, 22 March 2011 (UTC)

If MOSNUM is allowing mdy & dmy in the one article, it should be changed. That's rediculous. This exception was obviously put in place to allow telephone number date format for refs (also silly but hey ...). JIMp talk·cont 15:56, 24 March 2011 (UTC)

Also to allow "14 September 2004" in the text and "14 Sep 2004" in the refs, or "September 14, 2004" in the text and "Sept. 14, 2004" in the refs. --A. di M. (talk) 18:07, 27 March 2011 (UTC)
I say one format throughout. To do otherwise is regarded as amateurish in any other reputable publication house. Tony (talk) 02:31, 23 April 2011 (UTC)

Recent edit to BC/AD etc

This edit seems to turn the whole thing into wishy-washy. Has there been agreement to change the guideline on this point? I have reverted it for the time being, pending discussion here. Tony (talk) 04:05, 20 April 2011 (UTC)

See #Before Christ/In the Year of our Lord, some sections up. If Tony had actually bothered to look at the edit history, he would have seen that the text he restored was a completely novel proposal, 35 minutes old; if he had bothered to read the guideline, he would have seen that "a good reason" is sufficient justification for consensus to change the notation.
However, other readers may be equally careless, so I will make the connexion more explicit, while removing the unfortunate "scheme". To half the anglophone world, that is a quite negative word. Septentrionalis PMAnderson 04:31, 20 April 2011 (UTC)
While I was away from home there have been some "quite negative word"s from several participants. The most recent such word is "careless". Art LaPella (talk) 05:28, 20 April 2011 (UTC)
Moot in any case; the text before Herostratus' edit was restored before I could edit. Can you suggest a better, less negative, word for reading one bullet-point and ignoring the next? Septentrionalis PMAnderson 05:39, 20 April 2011 (UTC)
"reading one bullet-point and ignoring the next" is more negative than "careless", but since I'm not politically correct, I recognize there could be an appropriate time for either expression (I didn't find the bullet-point you refer to). That time would be when you have built a solid case, perhaps at a Wikipedia:Requests for comment/User conduct, that no good faith explanation exists. Without presenting convincing evidence, one wonders why you didn't say "other readers may make the same mistake" and "reading one bullet-point and missing the next". Art LaPella (talk) 00:33, 21 April 2011 (UTC)
The next bullet point in this (now replaced) text, which said and says that BC may be changed to AD (or the other way) for good reason. But this is now moot. Septentrionalis PMAnderson 00:54, 21 April 2011 (UTC)
Thanks, Art: valuable contributors are likely to be turned off when posts are personalised; I ask that avoiding negative references to other editors be a priority. On the edit in question, I don't know or care which substantive system is chosen, but changing "that scheme should generally be used in the article" to "there may well be a good reason to use it in the article" changes the advice into vague opinion and is not useful to editors who are seeking advice. Tony (talk) 17:19, 20 April 2011 (UTC)
Shorter version of the preceding post: "I don't care which system is mechanically imposed, but one of them ought to be." Pernicious nonsense; I am deeply disappointed in its author, of whom I expect better: If someone comes seeking advice, give them advice, not marching orders. Septentrionalis PMAnderson 20:59, 20 April 2011 (UTC)

Anyway, to return to the point in question... the edit spoken of was based on earlier edit, which was removing a prescription to use BCE/CE for articles relating to Judaic studies. I feel that this is 1) too specific, and 2) sounded as if it could be interpreted as special pleading. The larger point that the person was trying to make was "use the scheme that most sources use." The problem is, I don't know if this is something that we want to prescribe or not. It seems a reasonable case could be made either way. I myself would not be inclined to prescribe this, as it seems like unnecessary rule-making, something always to be on guard for, and just another thing for people to fight over ("I count 11 sources using BCE and only nine using BC, so I have replaced BC with BCE in the article!" "Well I have added 3 more sources using BC, so I have reverted this!" etc.). At any rate I have removed the material until we figure out what we are trying to say here and and if we do indeed want to say it. Herostratus (talk) 01:57, 21 April 2011 (UTC)

What we generally want is "use what the overwhelming majority of sources use; if there isn't an overwhelming majority, stay with what's there." This treats the study of Philo, say, just like we treat the various English-speaking nations: Australian Defence Force, but United States Department of Defense. Similarly, if the writers on Philo generally say he was born 20 BCE, so should we. (I doubt the condition is true; but there may people for whom it is; Akiva? Clement of Rome?) Septentrionalis PMAnderson 02:13, 21 April 2011 (UTC)
Septentrionalis has also removed the rule from the Judaism Wikiproject, discussed here. Other groups, especially Jehovah's Witnesses, also prefer CE/BCE; see Common era#Usage. If they don't restore the rule at the Wikiproject, then I don't know why we would keep it here. Art LaPella (talk) 05:24, 21 April 2011 (UTC)
I think projects are well within their role to specify which notation they prefer to use for new articles. I think it would even be within their role to try to build consensus regarding existing articles for which they have sole purview. --JimWae (talk) 22:14, 21 April 2011 (UTC)
Wikiprojects don't own articles; they don't have sole purview over anything. What are we going to do when one Wikiproject says one thing, another Wikiproject says another, and the user who actually writes the damned article couldn't care less what bureaucrats say? Shouldn't the guy who actually writes the new article be free to choose? Why should we have special rules for wikiprojects, aren't all articles part of the same wiki? Flamarande (talk) 23:24, 21 April 2011 (UTC)
Who said "purview" had to mean "ownership"?--JimWae (talk) 23:45, 21 April 2011 (UTC)
It doesn't has to mean ownership but apparently that's one of its meanings: "Purview = Scope or range of interest or control" [1]. Flamarande (talk) 00:20, 22 April 2011 (UTC)
Whatever. Manual of Style guidelines are expected to have exceptions in unusual situations, and therefore we have separate Manuals of Style for subtopics like Judaism, Islam, chemistry, or music. See Wikipedia:Manual of Style, first paragraph. Do we agree that unless there is a BCE/CE rule at the Judaism Wikiproject, then there is no reason to have such a rule here? Obviously you don't have to be Jewish or anything else to change a Judaism Wikiproject rule. Art LaPella (talk) 01:34, 22 April 2011 (UTC)
Well, a problem is that WikiProject members might have a narrower view than the general population. I am in a discussion now regarding some rules for transliteration of Russian names, and naturally the people who wrote those rules are mainly bilingual (and probably a large proportion of native Russian speakers) so things that seem unexeceptionable to them don't necessarily meet the needs of the general readership. Herostratus (talk) 07:12, 24 April 2011 (UTC)

Before Christ/In the year of Our Lord

I've made a bold edit to deal with the inherent Christian POV in dating things BC and AD. Please discuss. David in DC (talk) 02:04, 2 April 2011 (UTC)

I reverted. Historians these days often finesse the issue by reading BC as "before the Christian era", and by giving just a number for the year rather than following it with AD. Unfortunately, I can only say this is what I've seen and heard; I don't have a scholarly quote for you to back this up. I can say that there's plenty of support in current style guides for both systems. We've had several discussions on this subject on this page before, the last was a few months ago. I'm not opposed to a change in the wording (perhaps including that BC is read by some as "before the Christian era", if we can find support for that ... I've heard it often enough in lectures), and I share your twinge of guilt, but this is very well-trodden territory, both on and off Wikipedia. - Dank (push to talk) 02:30, 2 April 2011 (UTC)
I understand. Thank you for your explanation. I'll be looking for some reliable sources about why "finess[ing] the issue by reading BC as 'before the Christian era'" remains exclusionary. I'll note, though, that the way historians work around the issue is not self-evident to a reader who is reading an encyclopedia with his/her eyes, even if it is somewhat evident if the historian is reading aloud. And even if the historian is reading aloud, before the Common Era would be less exclusionary. David in DC (talk) 13:58, 2 April 2011 (UTC)
Here's some food for thought on the topic. David in DC (talk) 20:26, 2 April 2011 (UTC)
Here's an editorial from a Concord, New Hampshire newspaper that makes clear the position of some Christian fundamentalists who oppose BCE/CE. Quite rightly, they see it as a move away from a Christian orientation and toward a secular one. I agree. It's just that I think that's a good thing. Or in wiki-speak, I think it moves away from a Christian POV and toward WP:NPOV. David in DC (talk) 20:37, 2 April 2011 (UTC)
Christian parents in Missouri's Rockwood County schools sure saw moving from BC/AD to BCE/CE the way I do. Check out this article. Again, They favor a Christian POV, where I'm arguing for a non-sectarian one.David in DC (talk) 21:01, 2 April 2011 (UTC)
I didn't know the whole xtian religion would die because Wikipedias uses BCE/CE. I'll have to inform my fellow Atheists. I agree, this isn't an xtian encyclopedia, and it is more NPOV for the 5 billion people on the planet who don't think much of xtianity. OrangeMarlin Talk• Contributions 21:32, 2 April 2011 (UTC)
This is not a question of your particular (snotty) attitude, what some people in some backwoods town in East Nowhere think, or any other point-making exercise. Its about accepted usage. Let's follow that. Herostratus (talk) 23:11, 2 April 2011 (UTC)
Or perhaps we should focus on edits rather than editors. Snotty? WP:NPA. Accepted usage? Clearly not accepted by at least two editors, and gazillions of non-Christians. Let's follow wikipedia's norms and discuss whether consensus might change. Hereabouts accepted usage is called consensus. And hereabouts, consensus can change. Disagreeing without being disagreeable is just about the only true marker of adulthood, dontcha think? David in DC (talk) 02:15, 3 April 2011 (UTC)
I'm against using anything other than "BC". I'll suggest that about half of those "5 billion" come from China and India, and they're free to use whatever system they prefer on their version of WP (and so are the 90%+ of the other half). The use of "BCE" is favored in academia, but we don't write for an academic audience; instead, we write for the average audience that have no trouble in understanding the meaning of the succinct and long-accepted "BC". I'm loathe to put the four ~~~~ characters at the end of this post because the risk of offending billions by introducing the connotation of why we use "2011" this year is just too awful to contemplate.  GFHandel.   03:32, 3 April 2011 (UTC)

¶ There have been strong arguments both ways, from Christian, non-Christian and non-believing editors alike, all the way back to Anno Domini MMIV (Christian Era). The current rough consensus, as I understand it, perhaps mistakenly, is to skip both AD & CE in modern contexts (e.g. Handel, Newton, the Renaissance or the American Revolution), to use CE/BCE in specialised contexts (such as Biblical archaeology, Islam or Jewish topics), and to use AD and BC elsewhere, certainly not for religious reasons, but because CE/BCE is as yet puzzling and unfamiliar to the general English-speaking readership of Wikipedia (concentrated in North America, the British Isles and Australasia). For a guide to some of the dozens of earlier discussions (which a glance at the Archive box above led me to), see:

For some of the later discussions (since 2008), see the 68 search items I found by asking the Archive Box above for MOSNUM discussions here and the 20 others I found by asking the Manual of Style's archive box here. While stability is greatly to be desired, there's certainly no law against more healthy, civil discussion of an inherently thorny and contentious topic (see Wikipedia:Consensus#Consensus can change), but respect for the work of earlier editors suggests at least browsing through earlier discussions. You may find some strong facts and arguments for your own current position, as well as against it; you may even modify your thoughts in the light of reason if you read the older discussions with an open mind. —— Shakescene (talk) 05:06, 3 April 2011 (UTC)

Thank you. I have read a fair number of the more recent discussions, and a smattering of the older ones, if a latter one linked back to an earlier one. I agree with you that civility is critical, and am troubled by the tone this converstion has taken on, so quickly. I don't think I've been incivil, but I've obviously offended a couple of folks above. I regret that and I apologize.
I will continue posting views from reliable sources, and civil argument, on this topic. It is my hope that consensus will change. We stand on the shoulders of giants, so I agree that prior editors' work should be respected. It's why I hung the WP:BOLD tag on my original edit. I sort of expected it to be reverted, but I had some hope that views in favor of non-sectarian dating might have already evolved. I still think it's the style of dating that best accords with WP:NPOV. But sometimes I've believed as many as six impossible things before breakfast, so I'm happy to hear civil disagreement any time. It helps all of us edit better. David in DC (talk) 13:22, 3 April 2011 (UTC)
Be aware that editors become weary of opposing perennial proposals. If a few editors are perceived as trying to get their way by outlasting the weary majority, reactions to any changes to the guideline by one of the few editors may be rather abrupt. Jc3s5h (talk) 14:08, 3 April 2011 (UTC)
That sounds like a good topic for an essay. WP:WEARY? — Arthur Rubin (talk) 14:19, 3 April 2011 (UTC)
Concur re essay. I'd read it happily.
But who are these few editors you write of, Jc? Assuming good faith, why would they be perceived as "...trying to get their way by outlasting the weary majority." Were I such a weary editor, I'd try awfully hard not to let my weariness trump my assumption of good faith.
As for me, I'm not trying to outlast anyone, nor will I try again to edit the guideline without strong evidence of a new consensus. I'm not trying to game the system. I'm not trying to take advantage of anyone's weariness. I'm not "a few" anything, except maybe a few tacos short of a full lunch combo.
I'm simply a single, longtime editor who finds the current consensus wildly at odds with WP:NPOV. The terms "Our Lord" and "Christ" exclude me, and any other English-speaking non-Christians reading or editing en.wikipedia. I do not doubt the existence of a historical Jesus. But I do not see him as Christ (which means Messiah) nor as my Lord. That's a Christian point of view. A non-sectarian point of view, it seems obvious to me, would not use explicitly Christian terminology.
I see, understand, and accept that it's not yet sufficiently obvious to my colleagues in editing this great collaborative project to warrant a change. Some day, I expect it will be. Fortunately, we are not writing, or editing, on any sort of deadline. I'm happy to take the advice about the perceptions of weary editors, but I sure hope it doesn't extend to continued dialogue about the topic on this talk page. Sometimes, proposals are perennial because they are like the proverbial bad penny, turning up over and over again. More often, I suspect, they're perrenial because they are seen to have merit by successive cohorts of editors. Please grant the possibility that this one is more like the latter example than the former. Thanks. David in DC (talk) 20:49, 3 April 2011 (UTC)
I have no belief in a higher or outside Power or Being myself, and can often be upset at the assumption that everyone must have such a belief. But since the calendar in common use is in fact dated very roughly from the ministry of Jesus of Nazareth (whom most people, regardless of their views of His divinity or His anointing with the chrism, know best as Jesus Christ), I just find it far easier to use the familiar terms A.D. and B.C. While some atheists and agnostics may prefer B.C. and B.C.E., and they're not inherently loaded, most of those who use the terms seem to be devout Jews or Muslims, which would give (however unjustifiably) the appearance of a different sort of bias. On the other hand, while understandable, it's not justified for Christians to see the use of B.C. and B.C.E. to be part of some kind of deliberately offensive anti-Christian campaign on Wikipedia, any more than the use of American spelling in some articles and British in others is some kind of Anglophile or Anglophobe campaign. —— Shakescene (talk) 21:24, 3 April 2011 (UTC)
I don't think this discussion has progressed to the point of a few editors trying to outlast a majority of editors who follow the development of style guidelines, and I think the time to mention such a possibility is before it happens. That said, I observe that outside of Wikipedia, changes in core parts of the language related to numbers and measurement tend to take a decade or more (just look at metrication in the USA). It would be nice to have some sort of tracking mechanism for these discussions, so an editor new to the topic could be referred to a list of when a perennial proposal was last discussed. I don't know when AD/CE was last discussed, but considering the decade or more it would probably take for such a change to be commonplace in English, I see no need to discuss it here every few months. Jc3s5h (talk) 14:48, 4 April 2011 (UTC)
BC and AD do not represent a Christian POV any more or less than BCE and CE represent a Eurocentric and Western POV. How does one objectively determine which of these supposed POVs is more exclusive or divisively biased than the other? By your logic, we should need to either (1) refer to all notable terms within Wikipedia of a potentially sectarian/religiously-connotated nature like Thursday (Thor's Day), by a generic, neutered term like "Common Day 4", or (2) use the Islamic year-numbering system within Islam-related articles, the Hebrew year-numbering system within Judaism-related articles, etc. Just because Common Era happens to be a notable term doesn't make it NPOV. Google hits for AD/BC far outweigh those for BCE/CE, so technically BC/AD should be our standard, but I understand how insane that would drive most supposed "atheists" on here. As an atheist myself, I never could understand how fellow atheists or skeptics could be "offended by" or completely insulted by any reference to BC/AD because of its "pro-Christian bias", all the while using the Roman/Norse pagan days of the week and months without concern about any "pagan bias".
The interesting thing is — if we get right down to semantics — "BC" and "AD" only represent two letters each of the alphabet, so they are in no way representative of a Christian POV through the abbreviated form in which they are used on Wikipedia. Even if you grant that BC is an acronym for "Before Christ", the term Christ only directly means "messiah", not specifically Jesus of Nazareth. You would have to further understand and research the context to even realize that "Christ" here is referring to Jesus of Nazareth. Similarly, with AD, the phrase "In the year of [Our/the] Lord" does not specify who or what the "Lord" is. You would again need to research the matter to realize that it is a shortened form of "In the year of [Our/the] Lord Jesus Christ", and so "AD" in itself does not even reference Jesus of Nazareth himself, much less any claimed supernatural component.
So, when you strip this issue down to its core, BC and AD are nothing but letters, and the words they represent do not even specifically address the Christian deity by name. BCE and CE, however, represent what has been arbitrarily labeled a "common era" that all people are expected to accept and use like any other commodity, even though the era is clearly a Christian system centered on the birth of the Christian deity. I think that's a whole lot more offensive and biased than the original acronyms ever were. If we had just left well enough alone, the terms BC/AD would be as religiously charged to people as Tuesday and Wednesday are. That being said, I am in no way advocating that we stop using BCE/CE on Wikipedia. Both systems should be used and the current guideline should remain. What irks me is when people pull this tired and poorly-thought out PC argument about why BC/AD are biased and BCE/CE are neutral. — CIS (talk | stalk) 14:14, 4 April 2011 (UTC)
The vehemence here is startling. As are the contortions of logic necessary to conclude that C doesn't stand, explicitly, for Christ and AD doesn't stand, explicitly, for "In the Year of Our Lord" or that Lord, in this context, does not connote divinity. "Tired and poorly thought-out" are an interesting mix of projecting Jc's "weariness" onto the thinking of a BCE/CE proponent and then indicating a defect in the proposer's thinking process. The latter is a personal attack.
The hoary accusation of "Political Correctness" has now become so ubiquitous as a criticism of any change as to render the imprecation without much sting. It's a convenient rubric --- more a slogan than an argument with any intellectual heft behind it.
I agree that just because "Common Era" is notable does not render it NPOV. However, it's notability does seem to contradict the notion that BCE/CE is obscure and hard to understand. It's well understood by "...some people in some backwoods town in East Nowhere...." But Herostratus counsels that we ought not care about what they think. So which is it, that BCE/CE isn't understandable by commoners who aren't among "the majority of editors who follow the development of style guidelines", or that BCE/CE is understood quite well by the people in East Podunk, but that what they understand is irrelevant.
In my opinion, the analogy to Thor's Day or Odin's Day is a weak one. Thursday and Wednesday have an etymological tie to Thor and Odin, but they do not frame history from a Norse or Pagan viewpoint. BC and AD do frame history from a Christian viewpoint.
"Google hits for AD/BC far outweigh those for BCE/CE, so technically BC/AD should be our standard..." is a peculiar assertion. Imagine carrying that logic to the phrase "miserable failure" a couple of years ago. Please guide me to any policy-based rationale for using google hits as a measure of what en.wikipedia's standards should be. Google hits are generally decried and treated with derision when suggested as the standard for notability. Is it different for issues of style?
As I started, so shall I end. What's the vehemence and vitriol all about? It's at odds with collaboritive editing. It smacks of WP:OWN and even WP:WALLEDGARDEN. If you're weary, for goodness' sake, rest and be rejuvenated. I'll not try to edit the guideline without evidence of a changed consensus. It's obvious that's not demonstrated here. Smacking me around like a pinata for daring even to bring up the subject is disproportionate. It seems remarkably defensive. Defensiveness rarely accompanies argument in favor of something the advocate sees as well-settled and self-evident. Much more often, it accompanies an argument in favor of a status quo the advocate sees as threatened by progress. David in DC (talk) 18:07, 4 April 2011 (UTC)
Actually the point about "Thor's Day" etc. is spot on. It's only not a winning argument if "BC and AD do frame history from a Christian viewpoint" and BCE/ACD don't. But they do. The beginning of the "Common Era" is not a random year; it's the traditional birth year of Jesus. Changing the terminology from BC to BCE alters the system very little. As far as I'm concerned, the BC/AD - BCE/ACE system is annoying for two reasons: 1) it's silly to have the year "1" occur right in the middle of history (and also makes some calculations harder) and 2) it's Eurocentric. I'd support a movement to pick some random year say 10,000 years ago as the year "1" and date the "Common Era" from that. But absent that, it's just messing around with semantics and doesn't really make a difference either way. If you find the "C" annoying and the "CE" not, even though they mean the same thing, that's your choice and your privilege, but most people don't much care. Maybe just pretend that "BC" stands for "Before Commontime" and "AD" stands for "After Dat" or something, or whatever, but I'd recommend just letting it go. Herostratus (talk) 00:17, 5 April 2011 (UTC)
I think it is oversimplifying the debate to suggest that it is Christians versus the rest on BC vs BCE. In fact, many Christians find BC/AD problematical because it appears that Jesus was born in about 6 or 7 BC, so the calendar isn't even accurate. As this is so, some Christians have also argued that it is best to refer to it as the Common Era (CE) or before the Common Era (BCE). A second problem is that some non-Christians find the term objectionable. Of course, BC/AD is traditional, but that does not stop the terms from being problematical for a growing number of Christians and non-Christians alike. Michael Glass (talk) 02:45, 5 April 2011 (UTC)
If you read through my comments here, you'll see that I'm an atheist and I am against BCE/CE and support BC/AD. So there are people from all sides of the table with all types of viewpoints. I've made my case here, and I think it's a reasonable argument. I don't find BC/AD any more objectionable than Wednesday or January. It's all the same mythology to me, and I find it ridiculous to single out one bit of mythology for neutralization because it happens to be that of the world's current largest religious group. — CIS (talk | stalk) 03:50, 5 April 2011 (UTC)
In case I didn't make it clear bafore, I'm another atheist (an agnositical, non-evangelical one since I think such questions are at bottom undecidable in this world) who supports AD & BC over the still-not-widely-known CE & BCE in most contexts, but I don't think that there's anything wrong or irrational with those of any religious opinion who support either CE/BCE or AD/BC. —— Shakescene (talk) 04:14, 5 April 2011 (UTC)
Right. And given that both CE/BCE and AD/BC are inherently biased, I think practicality should be the determining factor for which is best to use. In that respect, BC and AD are the clear winners in my view. BC and AD both have two letters each, and they sound distinctly different from each other in speech. In contrast, CE has two letters but BCE has three, and when speaking, the "E" in CE and BCE can also tend to seamlessly blend with the "C" unless you intentionally pause in speech between the "C" and "E". This can make the listener think you've actually said BC instead of BCE anyway, making the entire effort laughably pointless. In addition, many often mistake "ACE" as the proper acronym for CE (to equate the three digits of BCE), when in fact this is not the case. There's also the further confusion and debate over what the "C" is meant to stand for. On Wikipedia we promote it foremost as "Common", but most external sources seem to ascribe the term "Christian", and "Current" is yet another interpretation. And all this for what? To avoid the Christian implications of a Christian dating system? Please. CE proponents should consider supporting the Holocene Era instead. — CIS (talk | stalk) 05:34, 5 April 2011 (UTC)
Just for the record, I don't think that BCE and CE are biased; in fact they're probably more objective and value- or belief-neutral (since they imply no assumption either way about the divine or messianic status of Jesus of Nazareth); and in some relatively-limited contexts (which might include some broad fields like archaeology) perhaps more appropriate, just as giving the Jewish (Masoretic), Islamic, Roman (A.U.C.), French Revolutionary or Julian year is sometimes appropriate. So while I understand the upset that some Christians periodically express at seeing BCE or CE, and wish to offer no one unnecessary offence, their complaints of religious (or anti-religious) bias are not in fact justified. My preference is based purely on how familiar current general readers of Wikipedia are with (B)CE and less importantly their rarity in non-specialist speech. —— Shakescene (talk) 01:29, 7 April 2011 (UTC)

Uncle. David in DC (talk) 00:19, 6 April 2011 (UTC) ──────────────────────────────────────────────────────────────────────────────────────────────────── I did a Google search on "siege Jerusalem 70" on the assumption that most of the editors on that site would be Jewish rather than Christian. A significant number of entries (probably 40%) chose to use AD rather than CE. I would therefore suggest that the issue of using BC,AD or BCE,CE be treated in the same way as the choice of UK or US English - the first editor of any article has the perogative to make the choice and others follow suit. Martinvl (talk) 11:29, 11 April 2011 (UTC)

"...others follow suit". If I'm honest, I'm not overly fond of that either. I've never understood why the first one sets the standards? I'm hoping that you will permit "BCE" to later become "BC" based on consensus forming for an article?  GFHandel.   00:00, 12 April 2011 (UTC)
Well, I think the idea being WP:ENGVAR ("the first editor of any article has the prerogative to make the choice and others follow suit") is just to avoid sterile edit warring. And the same principle applies de facto to other things that aren't spelled out in a MOS - whether to use dates or just years in the parenthesized vital information at the beginning of biographical articles, for instance. Generally, I think that's good. One thing is, there is always a cost to making a rule that must be followed by all, that cost being the loss of editors. If, for instance, we were to make the decision "Well, we must have a rule regarding English vs American spelling, for the sake of consistency", we would gain in consistency, but we would lose a lot of English editors (if we decided that American spelling must be used) or a lot of American editors (if we decided that English spelling must be used) who would just say Well if I'm required to use the spelling from across the pond I'm simply going to find another hobby. I think the same thing comes into play (to a smaller extent) any time we proscribe that a particular style must be used, when this is contentiouss. Herostratus (talk) 11:23, 12 April 2011 (UTC)
Actually, Engvar is a much simpler idea: Leave it alone, unless there is consensus to change. After consensus, stay with the new consensus. The "first contributor" business is an effort to deal with the case where there has never been consensus. Septentrionalis PMAnderson 15:44, 18 April 2011 (UTC)

To everybody interrested in the issue: Articles about 'Jewish subjects' should preferably use BCE/CE per Wikipedia:WikiProject_Judaism/Manual_Of_Style#Gregorian-Calendar_Dates. Just don't blame me - I didn't wrote this official policy. Flamarande (talk) 20:49, 18 April 2011 (UTC)

Should we offer conversions (without proselytising)?

Since I'm one who strongly concurs in Wikipedia's policy of always offering conversions between metric/ISO and non-metric units — as in 1 mile (1.6 km) or 100 degrees Celsius (212° F) — and since my opinions on AD/BC/E are governed by how well they're understood by others, I think it might be worth adding a suggestion to Wikipedia: Manual of Style (dates and numbers)#Year numbering systems for a similar explication of years, at least for the first appearance of CE and BCE in an article which has many occurrences of CE and BCE; if there are only a few appearances of CE and BCE, maybe put (AD) or (BC) after each one. Similarly, although I'm far less sure of this, put (CE) and (BCE) after at least the first occurrence of AD and BC. In the latter conversion I'm not sure what would be the best style, for example AD 70 (CE) or 70 AD (CE). 753 BC(E) or 753 BC (BCE). —— Shakescene (talk) 05:45, 20 April 2011 (UTC)

Or we could just use negative years for BC(E) and positive years for (AD|CE). Boom.
In all seriousness: policy should be for articles on religious topics to use the appropriate convention for that religion (eg. Christian articles = BC/AD; any-religion-other-than-Christian = BCE/CE as needed), for non-religious topics to use the conversion template(s) as proposed above by Shakescene if necessary. For example, I don't need to be told that Adolf Hitler, to pick a random historical personage, was born in 1889 AD/CE; just 1889 will suffice.
The number of appearances of BC(E) or AD/CE should probably also be limited. Once we know Person X is born in 700 BC(E), we don't need to be told that this event happened in 690 BC(E) and then this event in 675 BC(E) and so on. The only time it should be mentioned again is if there is possibly confusion -- eg. events that happen very close to year 0. --Mûĸĸâĸûĸâĸû (blah?) 06:20, 24 April 2011 (UTC)
Well, something like "753 BC(E)" would be useful for a reader who was all like "BC? What is that? Oh, it's BCE, I get it!". Is this a problem that readers have? I don't suppose so. Maybe in future it will be. Also: there is no such thing as a "Christian article" and so forth. There are only "Wikipedia articles". We can't/don't/shouldn't assume that readers of articles on a given faith are themselves adherents of that faith, and even if we did we shouldn't worry too much about their sensibilities. Herostratus (talk) 07:06, 24 April 2011 (UTC)
No... but by that same logic we shouldn't assume that readers of British topics will be British and therefore should use British spelling. American English on all British topics! Not done. Just the same way American topics use American-style dates (eg. April 24, 2011) whereas an article about, say, a Polish topic will use Polish-style dates (24 April 2011). Christian topics, (eg, the page on Jesus Christ or a pope, etc) should use Christian-styled era specifiers (BC/AD) while non-Christian topics should use the others if necessary. --Mûĸĸâĸûĸâĸû (blah?) 19:39, 24 April 2011 (UTC)
Well, variations between English-speaking nations is a special case. I would say that MOS:TIES is mostly nonsense and I certainly don't pay any attention to it, and when I create an article about a British entity I certainly don't write "favour" and "colour" and "bonnet" (of a car) and what have you because those words are not in my vocabulary. Am I supposed to keep a dictionary of British spellings and usage at hand for when I create articles about British entities? I hope not, because I don't and don't plan to. (Now, it is true that de facto most articles about British entities have been written (or started) by British persons and therefore have British spellings which are (properly) retained per WP:ENGVAR, and vice versa for articles on American subjects.)
But as to Polish usage, are you serious? This is the English Wikipedia. I have written many subjects on Russian subjects, and I use American dating (because I'm American -- if I was British I would use British dating I suppose) and don't pay any mind what system they use in Russia. If the Russians used an idiosyncratic Russian system -- DD/YYYY/MM say, or dating years from the founding of Kiev -- should we follow that? Of course not. If I choose to start an article on a "Christian topic" -- whatever that is -- I'm willing to be constrained by a general "For all topics, use what the great majority of sources use, if applicable" if this is insisted upon, although I wouldn't advocate such a rule. But I am very much not willing to be constrained by a general "Christian topic? Use BC/AD" rule. And I would find such a rule a worth -2 Annoyance Points. And we want to be careful: Every time we constrain how an editor may use his own judgment in expressing a thought, we lose X editors where X may not assumed to be zero. We should only constrain editors where there is specific benefit that we can point to that offsets the inevitable commensurate loss in Annoyance Points. Herostratus (talk) 22:39, 24 April 2011 (UTC)
In Europe dates are written day month year. 24 April 2011. 24-04-2011. Actually I think that's how it is in most of the rest of the world as well. I just happened to use Polish as an example; I could equally well have said English or Canadian or Spanish or French or anything else. This is what I was attempting to point out. MOS:DATES, this article in fact, goes on for some time on this topic.
I'm thinking that this whole BC/AD vs BCE/CE thing should be lumped in with "Strong national ties to a topic" -- eg. WP:STRONGNAT. And I'm not saying that you should constrain yourself when starting a Christian topic, say; I'm saying, in case of edit warring or disputes, go with AD/BC on Christian topics and BCE/CE for others. Just like if you were to start an article about a clearly British topic and write it in American English, and someone came by and converted it to British English, there's no problem; but if there's an edit war that starts over it the British English variation will win. --Mûĸĸâĸûĸâĸû (blah?) 03:39, 25 April 2011 (UTC)
To be honest, I had never heard of WP:STRONGNAT and [[MOS:TIES before all this, don't think much of them, and don't intend under any circumstances to start spelling "favor" as "favour" when I start an article (respecting the existing spelling used in an existing article is another matter entirely). Dates doesn't matter to me since I always spell out the month (and so should you, and everyone) to avoid the confusion over whether 10/12/2001 means December 10 or October 12. (However, I'm certainly not going to start writing "7 July, 1998" instead of "July 7, 1998" when creating articles about Russia since there is no good reason to do that).
"in case of... disputes, go with AD/BC on Christian topics and BCE/CE for others..." -- this sounds like you are suggesting that all articles that are not, specifically, about Christian topics should use BCE/CE (if someone disputes an existing BC/AD scheme, which anyone could on any article). This is not a good a idea and is not going to fly anyway... didn't we just go over this just recently? Just let the original writer of the article decide, and respect whatever she used. Herostratus (talk) 04:50, 25 April 2011 (UTC)

Changes in Format consistency section

Hello. Sorry for not posting this immediately. (Technical problem). One hour ago, I changed this sentence:

These requirements apply to dates in general prose and reference citations, but not to dates in quotations or titles.


These requirements do not apply to dates in quotations or titles.

I think you know why, right? Because at the top of the section, we have:

Dates in article body text should all have the same format

Surely, article body doesn't just mean prose, right? It includes image captions, tables, lists, infobox and everything else except references.

Is everyone okay with this? Fleet Command (talk) 09:52, 21 April 2011 (UTC)

Fully agree, that's what was agreed on. However currently user:Tony1 is systematically eliminating all yyyy-mm-dd style dates from references against consensus. That should be blocked and reverted. −Woodstone (talk) 17:51, 21 April 2011 (UTC)
No, he should be constrained to act by consensus; he believes he is acting in the best interests of the encyclopedia, is always an honest man, and often a learned one; but he reminds me of what Franklin said about John Adams. Septentrionalis PMAnderson 22:43, 21 April 2011 (UTC)
Woodstone, why didn't you say something to me? I had no idea there was any objection. It's not a core issue for me. Tony (talk) 02:28, 23 April 2011 (UTC)
Why should refs be any exception? JIMp talk·cont 00:32, 25 April 2011 (UTC)
To make things easier for editors. Two words: "Habit" and "scripts". Fleet Command (talk) 05:17, 25 April 2011 (UTC)
The assertion that it is easier for editors is really quite debatable; if, however, readers are placed first, we are doing them no favours at all as undoubtedly they will as confused with "2011-04-25" as "2 Mebibytes". -- Ohconfucius ¡digame! 12:02, 25 April 2011 (UTC)

Tables should not be required to have the same date format as prose, due to the need for compactness in tables. Jc3s5h (talk) 00:48, 25 April 2011 (UTC)

No, it's best to use three-letter abbreviations in this case but is there any need to resort to stuff that looks like code? JIMp talk·cont 03:26, 25 April 2011 (UTC)
You are right, Jc3s5h; but right now, it is exactly as you like: This manual of style right now allows tables to use YYYY-MM-DD whenever there is a need for conciseness. (YYYY-MM-DD is easier to maintain for long tables, since it is natively sortable.) In time, tables that are not long (i.e. those which do not have a need for conciseness) should use the same style as the rest of the article. {{dts}} and {{sort}} can be used in these cases. Fleet Command (talk) 05:14, 25 April 2011 (UTC)
It really is odd that people should complain about making the date formats in a reference section consistent with those in the main text of the article. Tony (talk) 14:05, 25 April 2011 (UTC)
Some people seem to like having some yyyy-mm-dd dates in the reference section, never mind the rest of the dates in that section are a mangle of different formats. --Ohconfucius ¡digame! 14:40, 25 April 2011 (UTC)

Very odd ... here are the arguments for yyyy-mm-dd.

  • yyyy-mm-dd is concise ...
... but "25 Apr 2011" is only one character longer than "2011-04-25" (for "Apr 25, 2011" you've only got to add a comma).
  • yyyy-mm-dd is sortable ...
... but {{dts}} and/or {{sort}} take care of that.
  • yyyy-mm-dd is easy for robots to read ...
... but aren't we writing for humans?

Keep all dates in an article in the same format where "25 Apr 2011" is considered essentially the same format as "25 April 2011" just abbreviated (and similarly for mdy).

As far as I can see yyyy-mm-dd in refs is a hangover from the dark old days of date linking wherein the citation templates had to be given this and it was assumed that humans would have their prefs set such that it would we rendered readable. This assumtion was proven false, date linking was depreciated, the templates were adjusted but the yyyy-mm-dd input remained & is now spat out raw, not auto formatted. Monkey see monkey do is now the go. Editors now see this yyyy-mm-dd hangover in refs & assume that this is how refs are meant to look on WP. Editors conform. No, Tony is acting in the best interests of the encyclopedia by turning these telephone numbers into dates readable by humans. Consensus? Recall the citation templates & how it was assumed that this would be done automatically anyhow; that's your consensus. JIMp talk·cont 00:18, 26 April 2011 (UTC)

I'm pretty sure that you're wrong that there's a consensus that changing date formats in references is a good idea. I think there was an RFC about this a while ago and that position was rejected. Quale (talk) 01:41, 26 April 2011 (UTC)
(e.c.) Anyone got a link to it (and any earlier discussions)? To me, this is worth having another go at, for three reasons: (1) to achieve consistency with the main text of the article; (2) most readers, I'm assuming, find the gobbledy all-numeric dates hard to parse: I have to pause each time to "translate" them into English; and (3) the position of month and day are reversed between US (and Canadian?) usage, and that for the rest of the anglophone world, which adds to the confusion. I do believe ISO dates are for specialists, which doesn't go well for our readership. Tony (talk) 05:16, 26 April 2011 (UTC)

You are all correct, at least to some extents if not completely.

Quale, the only point that you are wrong in is the word "Change". No one said "change". "Difference" yes, but not "change". And yes, once the references section of an article is predominantly or uniformly using one date style, no one can change it. And yes, there was three arbitration committee cases regarding it. (I don't know about RFC.)

As for the rest of you, yes, I am also a pro of banishing yyyy-mm-dd. But "what I prefer" right now is in contrast with "what the policy allows". If you guys think there are a lot of people in favor of banishing yyyy-mm-dd, we can propose it (in village pump?) to make official. (But I don't think so.)

Fleet Command (talk) 09:44, 26 April 2011 (UTC)
Wikipedia:Mosnum/proposal on YYYY-MM-DD numerical dates —Preceding unsigned comment added by (talk) 11:34, 26 April 2011 (UTC)
Hmmm... Thought so. Fleet Command (talk) 18:01, 26 April 2011 (UTC)
It was rejected pretty decisively by WP standards, roughly 2-to-1 opposed. The failed proposal was favored by some of the loudest voices here and at WT:MOS, but a large number of editors who don't normally write here made their views known and vetoed it. A year and a half is certainly enough of a wait to make it reasonable to try again. If outside participation is minimized this time, the loud voices may be able to get the result they desire. Quale (talk) 04:35, 27 April 2011 (UTC)

Substantial change in MOSNUM policy

This paragraph was cut from MOSNUM:

  • When the source uses one set of units, generally put that one first; if editors cannot agree, put the source's units first. If they are not first, this should be stated in the citation.

I think it is important that it remain.

  • It is important that if the source units are not put first that it be stated in the citation. This is simply good practice.
  • Putting the source unit first actually applies in most cases: generally SI for general articles and most of the world and generally American customary measures for the United States. That leaves only UK articles in the anomalous position where there are likely to be disputes about which unit should appear first.
  • The other provisions set out a means of solving disputes about which unit to put first and that is valuable. Michael Glass (talk) 14:55, 18 April 2011 (UTC)
I would suggest that it is perfectly obvious that we shouldn't have a policy that directly contradicts itself, which is what you propose here. There are many good reasons why source-based units are inappropriate, and you set out no particularly good reason why this should remain.
You claim that source-based units approximate common local usage. In fact, in many cases, they don't. And even if they did, it does not follow that we tell should users to use the specific source (regardless of whether it actually matches with local usage) when we can just tell them to apply local usage. You claim that it's useful in dispute resolution - but if people need dispute resolution we already have mechanisms in place: editors are more, not less likely to need dispute resolution when we have rules that directly contradict one another. And the point about putting measures in the source can perfectly well be put in the section regarding sourcing units.
There are also many reasons why source-based units are problematic. These have been repeatedly pointed out to you, and include (and there's plenty more of this in the archives):
  • The fact that significant inconsistency, both within and between articles, may result from sticking to whatever source people happen to find, rather than basing on local usage or clear guidelines.
  • The fact that sticking to particular sources is not, in general, an adequate measure of local usage in cases where there are strong national ties to an article.
  • The fact that source-based units are appallingly open to abuse by editors (including you) who have a habit of picking the source for the units and not vice versa. Pfainuk talk 17:22, 18 April 2011 (UTC)

No, there is a much better case for source-based units: that's what is actually verifiable. Saying that A is about 20 km away from B has a reasonable error of several kilometers; saying that A is about 12 miles away from B has a reasonable error of about one kilometer. Septentrionalis PMAnderson 17:31, 18 April 2011 (UTC)

Which would seem to imply that you need to deal with such situations appropriately, a point already made by MOS:CONVERSIONS. This does not imply that we need to randomly start giving distances in kilometres in US-related articles - or distances in miles in France-related articles - just because that's what our source happens to do. Pfainuk talk 17:41, 18 April 2011 (UTC)
I completely agree. Source based units has been repeatedly rejected for the simple reason it leads to articles that are inconsistent. We have a MOS for a reason, for articles to be consistent. Wee Curry Monster talk 17:45, 18 April 2011 (UTC)
It's a balance of considerations. This removal should not stand. Septentrionalis PMAnderson 18:24, 18 April 2011 (UTC)
Furthermore, you should make up your mind: if this edit summary is correct, and the advice was moved, this guideline already upholds it; if not, why claim a falsehood? Septentrionalis PMAnderson 18:30, 18 April 2011 (UTC)
I genuinely believed it had been moved, it seems I was wrong. It was an error not a deliberate falsehood and I would appreciate an apology for that remark. However, I do believe the Pfainuk's edit was valid, it clarifies the policy and the comment on source units remains in the policy. If you look above there is a substantial discussion on source based units and the proposal was rejected. Wee Curry Monster talk 18:43, 18 April 2011 (UTC)
Because you thought it was true is a perfectly reasonable answer. Septentrionalis PMAnderson 05:58, 19 April 2011 (UTC)
Indeed. As it has been on more occasions than I care to count in the archives. Saying that we should be using miles on France-related articles or kilometres on US-related articles just because some people aren't able to choose the correct precision for their conversions is not sensible - a sledgehammer being used to crack a nut. Nor is it sensible to have a manual of style that directly contradicts itself within the space of two lines, which is what you're calling for here. Pfainuk talk 19:07, 18 April 2011 (UTC)
You misunderstand the reasons for the objection. The verifiable assertion is what the source says, including the implicit precision of what it says. No conversion does a great job of retaining that information, and {{convert}} is mindless. Now if we have sources in both systems, fine; but often we don't. Septentrionalis PMAnderson 05:58, 19 April 2011 (UTC)
I don't have a problem with asking people to note the source measure or quote the source in the reference - this would seem to be sufficient for verifiability. But as I noted above this can easily be added in the section where we require that all measurements be sourced - without sacrificing consistency or running roughshod over WP:TIES, and without creating any contradiction or lack of clarity in what the rules are (as the current wording does). Pfainuk talk 20:56, 19 April 2011 (UTC)

No-one has commented on the deletion of this provision:

  • {I]f {the source units] are not put first, this should be stated in the citation.

This, as I said, is simply good practice. This provision should go back into the policy.

Also, I have yet to see anyone come up with an actual example where using locally based sources would cause a significant inconsistency in any article. Michael Glass (talk) 22:12, 18 April 2011 (UTC)

  • As there are editors working to ensure that conversions/((tl|convert}} are being provided for all units of measure, there would seem little need to further burden the citation. --Ohconfucius ¡digame! 02:29, 19 April 2011 (UTC)
    Does {{convert}} indicate which unit the source used? If not, what has this to do with anything? Septentrionalis PMAnderson 03:16, 19 April 2011 (UTC)
Why should a measurement be any different to any other piece of sourced information, particularly when the link is there to be clicked on for verification purposes? We do not go quoting every sourced statement in every citation, and this should held to the same standard unless there was a particularly good reason to do so: for example, if it was somehow in dispute. --Ohconfucius ¡digame! 01:59, 20 April 2011 (UTC)
I think it is better to let people know what the original measurement was. The convert function can come up with anomalous results when the original number ends in a zero or two. Take this conversion: 4,700 square miles (12,000 km2) The figure given is rounded, and it ends up very nearly 143 square km short! This is equivalent to just over 81% of the area of Weddell Island, the third largest island in the group. The note The source uses metric units. is used quite frequently in Falkland Islands articles to indicate that the figures given in the text are derived figures. Michael Glass (talk)
That's no anomaly. The figure is meant to be rounded (in accordance with the guidelines on this very page). "4,700 square miles (12,173 km2)" would be a case of false precision. But I agre that it's important to let readers know what the original measurement was. JIMp talk·cont 05:22, 19 April 2011 (UTC)
One of the dangers of conversion was illustrated in the reporting of the fire under the M1 in London a few days ago. The severity of the fire meant that the fire service declared an exclusion zone of 200 metres and closed the motorway. This was subsequently reported as being a 200 yard exclusion zone. It then became 200 yards (183 metres). The next report must have shown it as 183 metres, then 180 metres and the last report that I saw in the British press reported 180 yards. This shows a need to discover what units the primary souerce used and that information to guage the reliability of secondary sources. Martinvl (talk) 06:10, 19 April 2011 (UTC)
Which is why we call on our editors to provide sources for all measurements. If there is any doubt, the reader can always go to the original source and find out what the units it used were.
I think it's worth noting that in the case you cite, this would come under the nominal or defined units section of the policy - which based on the guideline should always be in the units that the measure is nominal or defined in. The sort of pure source-based approach that Michael advocates might well see us put it in yards, but I don't think any other approach would. Pfainuk talk 20:56, 19 April 2011 (UTC)
But the same thing could happen if the "width of destruction", say, had been measured at 200 m. The only thing deterring it would be that the exact width might not be so often repeated. Septentrionalis PMAnderson 21:20, 19 April 2011 (UTC)
If the source, which uses metric, relies on figures derived from imperial, then the problem surely is with the source. I don't quite see the point of the 4700 sq mi conversion – surely it's a fallacy to justify your sort of conversion accuracy saying that the rounded figure of 4,700 square miles (12,000 km2) ends up very nearly 143 square km short and equivalent to 81% of the area of Weddell Island? These are clearly not of the same order of magnitude, and are strictly uncomparable except to give an idea of that order of magnitude. We certainly do not go around saying 'the Earth is 93,000,000 miles (149,668,992 km) away from the sun', as opposed to 93,000,000 miles (150,000,000 km), and try and justify not having a rounding error because it would represent 26 times the diameter of the Earth (12,742 km)! --Ohconfucius ¡digame! 02:19, 20 April 2011 (UTC)

The main point is to find the best source. If the best source gives information in metric or Imperial, go with it. However, the distance of the Earth from the sun neatly demonstrates the need for using good sources. The Earth goes round the sun in an ellipse, so the distance varies see [2]. As for the difference between the area given for the Falkland Islands in square miles vs square km, it is unsafe to presume that just because one figure ends in two zeros that it is the original figure.

I should also note that I have never advocated a "pure source-based approach", only that it is worthwhile general rule. No-one has so far come up with any concrete example to show that any article would be hurt by such an approach whereas the Falklands Units approach has led to numerous anomalies. Michael Glass (talk) 07:25, 21 April 2011 (UTC)

  • Take an idiom in a foreign language, translate it into your own, then getting someone else to translate it back into the first, or into a third language, you will end up with something incomprehensible or anomalous. This is analogous to what we you are trying to demonstrate. Trouble then is it's next to impossible to know what "correct" data are if even official data are subject to such conversion or rounding errors--Ohconfucius ¡digame! 09:12, 21 April 2011 (UTC)
  • I think you know the example I used was to illustrate the potential rounding error in the conversion, and not to digress into astrophysics. ;-) --Ohconfucius ¡digame! 09:50, 21 April 2011 (UTC)

I agree. We must go for the best sources of information and beware of rounding errors. Michael Glass (talk) 08:00, 22 April 2011 (UTC)

But that does not mean that we should pretend that the sources of information that some might consider the best are not necessarily prone to rounding errors. If you find a precise measurement in square kilometres that gives an exact round number in square miles, then particularly when the subject is related to a country that sees significant use of imperial measures, we cannot take it for granted that this is not an overprecise rounding. The fact that a figure is given precisely by a source in metric units does not mean that it does not suffer from rounding errors - even if it is only given in metric units.
I did a bit of searching on this particular case, and there are plenty of sources that use imperial measures first and plenty that use metric measures first. There are also sources that use imperial measures only and sources that use metric measures only. But when you ignore the units, you find almost all sources supply one or two of the following values: 12000 square kilometres, 4700 square miles and 12200 square kilometres. There are sources for all three that qualify them with words such as "approximately", "around" or "just over", and there are sources that convert 4,700 square miles to the nearest square kilometre. But I see no reason to assume that this, taken together, implies that the area is exactly 4,700 square miles. Pfainuk talk 10:11, 22 April 2011 (UTC)

I agree that it is sometimes quite difficult to work out what is the original figure. No doubt the original measurements of the Falkland Islands were made in Imperial measurements, but British maps have been made using metric measures for decades. This link shows metric maps for Lesotho (1978) and Tristan da Cunha (1980) so one would expect that similar maps were made of the Falkland Islands at about the same time. I think that all we can do is to go on is the information we have. If it's metric, I think we should go with that, but if it is Imperial then we need to use that. However, if better information comes up, we should use that. Michael Glass (talk) 11:20, 25 April 2011 (UTC)

This would seem to require significant inconsistency between units - if, for example, the source for one measure is imperial and another later in the sentence is metric.
I do not, of course, accept your contention that the use of metric units on stamps issued by post-independence Lesotho and Tristan da Cunha is evidence that maps of the Falkland Islands would have been measured in metric units at that time; nor do I accept that the fact that they maps have been measured in metric units is necessarily evidence that the most accurate figure is the one that is a round number of square miles instead of the ones that are round numbers of square kilometres. Pfainuk talk 16:18, 1 May 2011 (UTC)
For the record, the 3rd series of 1:50000 maps of the Falkland Islands were made in 1986 - one of the consequences of this survey was that the Argentines had given the name "Mount Independence" to the wrong mountain - they intended to give it to the highest mountain on the Archpelago, but the older maps were wrong. Since UK Ordinance Survey maps were converted to metric in the 1970's, it is almost certain that the 1986 survey of the Falklands was done in metric units. (See here). Martinvl (talk) 20:23, 1 May 2011 (UTC)

rare date-time strings

Does anyone have any recommendations/suggestions as to how one ought to format date time strings so that they can be easily parsed by human and machine alike? I have encountered a problem at 2011 AFL season, where we have instances such as 'Friday, May 6 8:10 pm'. It strikes me that we have a missing separator at the very least. Should we put the time in parentheses, or precede it with a semicolon, for example? Or any other ideas? --Ohconfucius ¡digame! 01:39, 27 April 2011 (UTC)

I always write "Friday, May 6, 8:10 PM". I'm taught to separate date from time. Fleet Command (talk) 04:13, 27 April 2011 (UTC)
Heck, we have no guideline to prevent this? The MoS already says to avoid jumbling up numerals: "thirty-six 6.4-inch rifled guns; not 36 6.4-inch rifled guns". I'm sure MOSNUM says the same or similar. By analogy, here, a comma at the very least, please, as Fleet Command says. Other opinions? Tony (talk) 15:53, 29 April 2011 (UTC)

The term 'mil'

The term 'mil' is ambiguous even when we know it means length:

  • It's sometimes used to mean 'mm'
  • It's sometimes used to mean 0.001 inch. This has a synonym 'thou'.

An example of it being used on wikipedia is:

Can anybody describe the usage of the units 'thou' and 'mil' and how separate they are? Lightmouse (talk) 20:36, 18 April 2011 (UTC)

It is also has military usage for milliradian. 6400 mil = 360°. Its use in the example you gave is decidedly ambiguous and is one of those examples where the source is inadequate and should be clarified. What units are used in the original source? Wee Curry Monster talk 20:42, 18 April 2011 (UTC)
It's also short for millilitre. JIMp talk·cont 05:18, 19 April 2011 (UTC)
I think it's usage as short for “millimetre” is quite informal and I wouldn't expect it to be used in an encyclopedia (and anyway, there should be a conversion to inches, so it's obvious what it means). As for the thousandth of inch, it should be accompanied by a conversion to squared millimetres or squared microns, so it's obvious what it means. ― A. di M.plédréachtaí 10:44, 19 April 2011 (UTC)
The "thousandth of an inch" definition for "mil" is the first in the Free Dictionary.[3] It used to be more common in the US and it's still used to measure films like mylar or Teslin (material). However most machine shops moved to metric a decade or two ago. It may be unfamiliar in other countries so it'd help to give a metric conversion.   Will Beback  talk  12:27, 19 April 2011 (UTC)
Interesting. The first definition I'd have come up with would have been an overcolloquial abbreviation for "millilitre". The second would be a thousand in the sense of "per mil". And the third a unit of currency, the equivalent of a "cent" where currencies are divided into a thousand. Wikipedia has articles about "mil" distances ranging from a thousandth of an inch to ten kilometres. While it may be the most appropriate unit to the context, it is one of those cases where linking the unit and providing a conversion is a good idea. Pfainuk talk 21:31, 19 April 2011 (UTC)

I have been researching the history of microprocessors and in the 1970’s integrated circuit dies were measured in mils (0.001 inches). In 1975 a typical silicon wafer was 3 inches (75 mm) in diameter. These were the most common units in technical papers and trade magazines. In the 1980s millimeters became standard. See my Motorola 6800 article. If you listen to oral histories of the semiconductor industry, the design engineers can recall the chip size in mils of every chip they worked on. -- SWTPC6800 (talk) 15:13, 19 April 2011 (UTC)

Where I used to work, we used the mil measurement often. Most of the time it was to measure the thickness of a layer of paint with an elcometer and to measure the "depth of cure" of ultraviolet (UV) adhesives. It was also the common measurement on our latex gloves and trash-bags. Never used the term thou and using the term mil for millimeter/milliliter on wikipedia would be like using 'clicks' for kilometers. :) —MJCdetroit (yak) 22:37, 19 April 2011 (UTC)
I've often seen "mil", but I'd never heard of "Thou (length)" until this discussion. Maybe that article should be moved to "Mil (length)".   Will Beback  talk  22:56, 19 April 2011 (UTC)
Seems that it might be an American thing, in the UK we use mil to exclusively mean millimetre. Our thickness gauges are "thou" (imperial) and "mil" (metric). I have never heard inches/1000 referred to as mil in all my time as an engineer, since 1980, nor even in discussions with pre-metric engineers that used Whitworth etc. BSW/AF/BSF/BFC Chaosdruid (talk) 15:03, 25 April 2011 (UTC)

Mil and kcmil should always be used when talking about American wire gauge, and PCB trace widths. Gigs (talk) 17:56, 2 May 2011 (UTC)

It never ceases to amaze me how much specialist knowledge is required to get the "correct" non-metric unit for any domain, region, and time-period. The instance that started this debate was:

What is the metric conversion for that? Lightmouse (talk) 18:41, 2 May 2011 (UTC)

mil^2 = 0.00064516 mm^2 Gigs (talk) 18:47, 2 May 2011 (UTC)

Good, that's how I would have calculated it. So 50,600 square mils is 36.1 square millimetres? Lightmouse (talk) 18:57, 2 May 2011 (UTC)

Yes, as some have already commented "mil" is American for "thou". A millimetre should never be written "mil" though this shortening is often heard in speech in the UK. Best to simply use inches and mm. Thus 1 "mil" (thou) is .001 inch (.0254 mm).

Davidlooser (talk) 15:01, 12 May 2011 (UTC)

WP:WikiProject_Trains going to violate WP:UNIT

WP Trains going to use units without space and force US/UK/imperial on almost all categories, including modern Russian broad gauge. TellSI (talk) 12:59, 29 April 2011 (UTC)

Could all editors please examine the on-going discussions at Wikipedia talk:WikiProject Trains before jumping to any conclusions over the above claims made by User:TellSI. This user's account has been recently created (today!) and appears to be intended for pushing a particular POV.
The discussions at WPT result from some heavy-handed POV-pushing category changes by an editor who was discovered to be a known sockpuppet of a banned user (on several WP sites). The views of User:TellSI (and his username) appear to be remarkably similar to the user who created the mayhem that WPT is trying now to address, and who was again banned as a result. -- EdJogg (talk) 14:58, 29 April 2011 (UTC)
I'd like to get the opinion inter alia of User:Lightmouse, who IMO is our greatest expert on units and conversions. Tony (talk) 15:48, 29 April 2011 (UTC)
Despite the claims above, nobody at TWP is trying to force anything. Ideas have been put forward, were discussed, have been refined a bit and are still under discussion. A consensus seems to be emerging. Lightmouse (and any other editor) is welcome to make constructive comments at WT:TWP. Mjroots (talk) 16:03, 29 April 2011 (UTC)
User:Vmenkov is of other opinion. The TWP "consensus" is fringe group consensus, mostly of UK/US based editors, violating MOS with not using the space between number and quantity. TellSI (talk) 17:32, 29 April 2011 (UTC)
Note, TellSI is a confirmed sockpuppet of a banned user. Da.squirrels (talk) 17:49, 29 April 2011 (UTC)

I understand this issue has been resolved. Lightmouse (talk) 12:02, 30 April 2011 (UTC)

Yes, it has, thanks. The socks of a banned editor have been blocked. We decided at TWP not to use spaces before "ft" and "in", but keep them for mm gauge categories. MOS applies more to articles than it does categories, doesn't it? Mjroots (talk) 11:12, 2 May 2011 (UTC)

Could you provide links to illustrate what you mean please? Lightmouse (talk) 11:16, 2 May 2011 (UTC)

See for example Category:5ft 6in gauge railways and Category:600 mm gauge railways. The discussion is at Wikipedia talk:WikiProject Trains#Specifics and Wikipedia talk:WikiProject Trains#Straw poll on format of categories (just search that page for the word "space" to find the relevant parts). Thryduulf (talk) 11:39, 2 May 2011 (UTC)
That's utterly and completely unacceptable for TWP to devitate from the MOS in this regards. Headbomb {talk / contribs / physics / books} 13:27, 21 May 2011 (UTC)

Adding/editing units

I've been adding/editing units of measure for many years now. I've recently been running a bot (approved as Lightbot5) to add conversions to feet and miles, plus their squares or cubes. The current scope was deliberately limited but it is now time to address the limitations. In the following examples:

  • It is 20 feet tall and 15 feet 6 inches wide
  • he is taller than his opponent at 6 feet tall compared with 5 feet 10 inches
  • she drove 10 miles at 120 mph
  • she is only 5 ft tall and weighs 85 pounds

it will convert the former but not the latter in each case.

Back in October, I started the process to extend the scope to cover situations such as the above. The bot approval group has a high workload and is looking at a lot of other requests. Given the length of time that's passed and the domain knowledge that exists here, I'm asking for more editor eyes on the requests starting at Wikipedia:Bots/Requests for approval/Lightbot 6 upwards. The most help can be provided by commenting at:

In all cases, the functionality was approved as Lightbots3 but approval expired. Rather than spending more time debating theory, I think it's time for actual edits, even if only as a trial (in either the Lightmouse or the Lightbot account) with the 'save' button pressed for each one. Regards Lightmouse (talk) 17:40, 30 April 2011 (UTC)

Lightmouse, I'm sure editors at MOSNUM are interested in this endeavour; I certainly am. Would it be possible to occasionally report to us the progress of your script/bot operation? Specifically, trends and patterns you observe, difficulties, editors' reactions, if any? Tony (talk) 13:30, 18 May 2011 (UTC)

Date formats when month name is used

I have recently had an article in which I consistently used the dating format: "November 29th 1911" corrected to "November 29 1911" as, apparently, the former is "incorrect" according to the Wikipedia style manual. But this "corrected" version is American, not British, usage. I then got told that correct British usage would be "29 November 1911". This seems to be based on an American misunderstanding, just because we write "29/11/1911" does not mean that we necessarily put the day first when the name of the month is written. Even then to be correct British usage it should be "29th November 1911".

I cannot see why my original format - November 29th 1911 - should be regarded as "incorrect" by Wikipedia, its not ambiguous and here its the most commonly used of any of the varients refered to above. I regard its correction to "November 29" as an unwarranted and unwelcome American imposition.

Davidlooser (talk) 15:29, 12 May 2011 (UTC)

The corrected version (which I hope included a comma: "November 29, 1911") is Wikipedia usage, not American. November 29th, 1911, is perfectly workable in American usage too. So this isn't an American imposition. -- JHunterJ (talk) 15:54, 12 May 2011 (UTC)
Well... [4][5] A. di M.plédréachtaí 16:10, 12 May 2011 (UTC)
David, Manual of Style does not agree with using ordinal suffixes in dates, so if you wish to do it according to MOS, take your pick: It is either November 29, 1911 or 29 November 1911. Now, if you don't wish to do it according to MOS, that's another matter entirely. If you wish to propose to change the MOS, I guess the right place to do so is a place called Wikipedia:Village pump. But you may wait for more input here. Fleet Command (talk) 16:11, 12 May 2011 (UTC)
David, the manual of style serves to introduce some uniformity/professionalism into the look of our articles. We use the terms 'American' and 'International' date formats by convention mainly for convenience although we accept there is usage of both dmy and mdy formats on either side of the pond. FYI, the date formats in 'your' article were in no way consistent, as you implied. you had:
  • several instances of dd month yyyy (e.g. 29 November 1911)
  • an instance of dd mmm (abbreviated months) yyyy (e.g. 10 Dec 1972)
  • an instance of ordinal dd month yyyy (e.g. 15th July 2009)
  • several instances of month dd, yyyy (e.g. November 29, 1911)
  • an instance of ordinal month dd yyyy without commas (e.g. September 29th 1972)
  • an instance of ordinal mmm dd yyyy (abbreviated months) without commas (e.g. Dec 10th 1972)
  • several instance of month dd yyyy with commas (e.g. November 29, 1981
I have now made them all MOSNUM compliant, except for the abbreviations. --Ohconfucius ¡digame! 17:50, 12 May 2011 (UTC)

From my talk page:

Date Formats

You changed the date formats on the Electric Palace article quoting MOSNUM as your authority, yet it clearly states:

"Edit warring over optional styles (such as 14 February and February 14) is unacceptable. If an article has been stable in a given style, it should not be converted without a style-independent reason. Where in doubt, defer to the style used by the first major contributor"

So please leave it alone.

Davidlooser (talk) 21:23, 12 May 2011 (UTC)

  • Thank you for quoting WP:RETAIN. That's just it... I demonstrated clearly there wasn't stability in any given style – that none could be discerned, therefore the above clause doesn't apply. What you are arguing for appears to be something called WP:OWN and WP:ILIKEIT. Now if you go and put it into a style that conforms with MOSNUM, whether that may be mmm dd, yyyy or dd mmm yyyy, then fine. If you insist on resisting, like you seem to be, then you are probably guilty of disruptive editing. --Ohconfucius ¡digame! 23:09, 12 May 2011 (UTC)

--Ohconfucius ¡digame! 23:09, 12 May 2011 (UTC)

I believe the "optional styles" clause (MOS:NUM lead) has previously been used to argue that the rest of the Manual doesn't matter, so perhaps it should be clarified. Art LaPella (talk) 02:03, 13 May 2011 (UTC)

Well OK, it looks like I am defeated - for now. I would like to point out that every date *I* wrote had a consistent format. But I didn't write all of the article and I didn't "correct" anyone else's date formats.

I do sincerely beleive that the ban on the use of "th" after the number in a date is wrong-headed; far from making the text look "professional" as someone claimed above, it makes it look sloppy as though written by someone who couldn't care less about presentation.

Clearly I need to take this matter further, when I've manaaged to work out how to get through the labarynthine layers of the Wikipedia system.

Davidlooser (talk) 16:16, 14 May 2011 (UTC)

The Wikipedia is pretty forgiving about date formats and allows several (and is pretty unusual in that regard, compared to other encyclopedias). Which is fine, we don't need to have consistency as a hobgoblin (except within articles). However, this doesn't mean that "anything goes", and there is a proscription on using ordinal numbers for dates regardless of the format used. This seems like a reasonable editorial standard to me. Herostratus (talk) 17:20, 14 May 2011 (UTC)
If you believe that the ban on "th" in a date is wrong-headed, then you have reached the correct part of the labyrinth in which to convince us to change it. Quoting from style books is more likely to convince us than simply asserting that the opposite style is sloppy. Art LaPella (talk) 00:07, 15 May 2011 (UTC)
David, there is no such a thing as "the labarynthine layers of the Wikipedia system". Wikipedia is very simple: Everything works through consensus. But the fact is that everyone in the Wikipedia is in favor of not using the nonsensical ordinal suffixes. (Or at least they have been so at time of writing this MOS.) I don't care what their reasons were because I think this a color of the bike shed issue. However, if you wish to propose their use, I told you how to do it.
My only advice is to stop playing a martyr and using sentimental statements like "looks like I am defeated" and "the labarynthine layers of the Wikipedia system". Every vandal who has been banned for using the seven dirty words have once used these statements and so we are immunized to them. If you have a very good logical reason that the suffixes should be put back, then you can consider yourself the winner. Fleet Command (talk) 07:27, 15 May 2011 (UTC)
I like ordinal suffixes for all kinds of reasons, if not all the time everywhere, and I wouldn't impose them on others. Why the prohibition on ordinals (or how far into the distant past) came about is a little beyond me. Of course they could have interfered with date autoformatting at some point, but that's (happily) distant history. And Wikipedia looks incredibly labyrinthine, opaque and forbidding to almost every newcomer, more so than perhaps it is. On the other hand, consensus means that discussions here can go on at enormous length (e.g. on hyphens and dashes, which defeated me completely when I was trying to write a non-redirected link to Nicholas Browne-Wilkinson, Baron Browne-Wilkinson for the Lord Browne disambiguation page. No combination of hyphens and en-dashes seems to work.) And there's only 100+ archives to thumb through if you wonder how & why some consensus once came about. But, please, however you write a date, always put either a comma or a month between the day and the year. —— Shakescene (talk) 11:46, 15 May 2011 (UTC)
What about just copying and pasting the title (what I normally do for titles longer than a couple of words, even without dashes)? :-) A. di M.plédréachtaí 16:23, 17 May 2011 (UTC)
Shakescene, are you under the impression that there has not been a clear and verifiable shift away from ordinal suffixes for decades and decades? Is there also a deep trend towards brevity and efficiency in expression in the language? (Why use one character when five will do? as a friend of mine says.) BTW, I hope never to see date and year numbers jostle each other in the mdy format without a comma. Please. Tony (talk) 13:26, 18 May 2011 (UTC)

Capital "c" calories

MOSNUM should give some sensible advice on the capitalisation of the kilogram calorie. There is a convention that suggests we distinguish large and small calories by capitalisation. It isn't widely followed in the World out there. Nor is it widely followed in the world in here. Half-hearted adherence to the convention is likely to engender more confusion than it solves. We should either adopt it or drop it. Consensus against the convention is already evident on hundreds (maybe thousands) of WP pages.

A situation whereby some pages follow the convention and others do not is likely to cause confusion. If a reader sees the food calorie capitalised on some pages, he or she will probably come to expect that WP follows the convention mentioned above. When the reader sees the unit uncapitalised on another page he or she might well think that this is the gram calorie we're now talking about. Capitalisation is a means of distinguishing the big & the small just doesn't work unless you're consistant about it.

The convention is a dud. "A practice advocated by some, and not widely followed." the article Capitonym calls it. Yes, there are a few people who are all for the idea but, for the most part, the world means "kilogram calorie" when it says "calorie". Note how the convention breaks down completely when the word would be capitalised anyway (like in a proper noun e.g. 100 Calorie Pack Oreo).

No, people aren't just forgetting the "kilo-" part of the real "kilocalorie" that scientists use. The kilogram calorie came first. There once was a time when the gram calorie was used in science. That time is in the past, scientist have switched to SI (i.e. joules). On nutrition-related pages we know "calorie" means the big one. The unit hardly ever appears on any non-nutrition-related pages.

Normal practice is currently reflected on WP. I have searched about two and a half thousand WP pages for capital "c" calories. I found a couple of dozen pages which follow the odd convention. That's only about 1%. Perhaps we want to discount instances where the word "calorie" is used as a (somewhat-dubious-in-terms-of-encyclopædic-tone ... but that's another question) synonym for "food energy". Then there are a few pages talking about the unit itself. There are even a couple of pages using the gram calorie. Discount these and we still only have about 2% of instances in which this strange convention is followed.

If common practice is a good indicator of consensus, then consesnsus is to treat the "calorie" as any other unit name.

Unit names, even those named after people, are common nouns. They are not capitalized when written in full, except where common nouns take a capital.

Let's, however, make it clear that "calorie" is no exception to the above guideline. Let's bring the couple of dozen articles currently deviating for normal practice in line. Let's reduce the potential for confusion by explicitely rejecting the rarely followed convention.

P.S. Note this proposal has to do with the word "calorie" vs "Calorie" not the symbols cal vs Cal (distinctions made by capitalisation of symbols is standard practice, e.g. "M" for "mega-" vs "m" for "milli-"). JIMp talk·cont 06:59, 16 May 2011 (UTC)

My stance hasn't changed since Wikipedia talk:Manual of Style (dates and numbers)/Archive 133#Calories: calorie and kilocalorie in the context of nutrition are in practice synonymous, so there's no need to disambiguate, by capital letters or otherwise. (I'd add to that that advertisement normally uses calorie when they want it to sound small, e.g. “this cheese only contains 100 calories per serving”, and kilocalorie when they want it to sound big, e.g. “this will help you burn 500 more kilocalories a day”.) A. di M.plédréachtaí 11:11, 16 May 2011 (UTC)
There will be a new generation of people that will not know calories at all. Since Wikipedia addresses the wider public, it should make the change to Joules (actually kJ) and add calories in brackets where appropriate. Calories will still be used in science for a long time, because scientists love their jargon and are usually the last to change to new terminology. This, however, does not change the fact that calories are a legacy unit and a thing of the past. It only means that we need a good page on the legacy calorie unit that discusses all the confusing uses and jargon discussed above (There are about 8 different calories mentioned at NIST, way too confusing for the typical WP user). Kehrli (talk) 12:40, 22 May 2011 (UTC)
And that new generation will be at least 50 years from now. I'm pretty sure that most people in their twenties in the Northern Hemisphere can roughly estimate how many kilocalories their meal contains but have no idea of the equivalent figure in kilojoules, at least not off the top of their heads. And the various definitions differ by less than 0.2% from one another, which is probably less than the average difference between two random packages of the same food, so that's no big deal at all. A. di M.plédréachtaí 16:48, 22 May 2011 (UTC)
Kehrli, I'm afraid I have to disagree. It is the general public who are clinging to this dated unit. Scientists have joined the 21st century. But the question is whether we should capitalise the word when we're talking about the kilogram version. What's your stance on this? JIMp talk·cont 00:07, 25 May 2011 (UTC)

I'd use "calorie" to mean 4186.8 joules as is common in public use, and I'd give this "calorie" its capitalization as appropriate to a common noun. When someone instead means 4.1868 joules then the someone should use "joules" (or "small calories"). —Quantling (talk | contribs) 19:49, 25 May 2011 (UTC)


Colleagues, perhaps I missed some previous discussion about this and similar article titles. MOSNUM (and ISO) say to gap the multiplication sign and the value–unit. There are other similar article titles. Within the article, I started changing it, given that in some places it was already gapped. But I gave up pending your advice. Tony (talk) 16:59, 18 May 2011 (UTC)

Well, MOSNUM and ISO also say that the unit should be repeated (5.7 mm × 28 mm) for that matter; anyway, have you asked on the article's talk page or to relevant WikiProjects? A. di M.plédréachtaí 08:17, 19 May 2011 (UTC)

On a somewhat related note, Wikipedia:MOSNUM#Decimal points notes that MOSNUM need not be applied to common terms like .22 caliber:

Quote: Numbers between −1 and +1 require a leading zero (0.02, not .02); exceptions are sporting performance averages (.430 batting average) and commonly used terms such as .22 caliber.

I would have figured that the same reasoning would also apply to a cartridge name like 5.7×28mm‎; the unspaced style is not only the most commonly used style for that cartridge, but it is also the most commonly used style for essentially all cartridges that are named in that manner (e.g. 9×19mm Parabellum, 5.56×45mm NATO). As for my personal opinion on the issue, I definitely prefer the unspaced style over the spaced style (let alone repeating the units, as 5.7 mm × 28 mm). I don't think this is a case where MOSNUM needs to be applied to such an extent. ROG5728 (talk) 09:59, 19 May 2011 (UTC)

Yeah, if cartridges are usually named that way, I'd consider it to be a quasi-proper name and leave it that way. (This is also why they are allowed to sell e.g. 3½-inch floppy disks in France despite the inch not being a legal unit there.) A. di M.plédréachtaí 12:24, 19 May 2011 (UTC)
OK, I changed a couple to spaced before I came here to ask; but that's only because I found other examples in the main text in which exactly the same expression is spaced. Can we confirm that this is the way to do it (other opinions?), and then make it consistent throughout? Tony (talk) 12:38, 19 May 2011 (UTC) PS And are you sure it's not done somewhere else with an ex rather than a multiplicaiton sign? I find the whole expression jammed up together like that very uncomfortable to read. (3.5-inch disks in France ... irrelevant, I think). Tony (talk) 12:43, 19 May 2011 (UTC) PS and here, upper part of a google search, there's 5.7 x 25mm, 5.7x28mm, and 5.56 x 45mm, all on the one page. No × as in our article. I really fail to see why WP should replicate the mess found outside (actually, bits of mess), when we have a house style. People are sloppy; we should not be. Tony (talk) 12:49, 19 May 2011 (UTC) Oh, and just above it, the ugly and hard-to-read 5.7X28mm. Tony (talk) 12:50, 19 May 2011 (UTC)
Are we not people? A. di M.plédréachtaí 15:05, 19 May 2011 (UTC)
We might or might not be; but it's our readers who count, not editors. One reason WP is so successful is that it speaks with a voice that is (fairly) unified. It's this extraordinary combination of "anyone can edit" and the guidance of pillars, policies, and guidelines. When you see a WP article—especially a well-written one—it is not sloppy. Tony (talk) 15:20, 19 May 2011 (UTC)

The multiplication sign (5.7×28mm) is another matter; that is a change that was recently introduced and it is being discussed in detail here. The multiplication sign is occasionally used (e.g. this article), but the original unspaced style (5.7x28mm) is by far the most common style used for that cartridge (and any cartridge named in such a manner). ROG5728 (talk) 00:22, 20 May 2011 (UTC)

If some sources use formatting that adheres to the widely respected ISO rules, and our own house style, and that of many other style guides, why do we need to race to the bottom? May I ask: does anyone have access to respected sources on this matter? Again, the sources are all over the place on this usage, and don't even treat "5.7x28mm" as a name. Tony (talk) 14:08, 20 May 2011 (UTC)

The space issue comes up from time to time. Here is a previous example with Olympic sports:

The arguments were very similar to this case:

    • some people wanted to rely on consensus within the Wikipedia specialism
    • some people wanted to copy the style of specialist publications
    • some people wanted to clarify anomalies with the scope.

I'm not sure if that exception still applies but most people (inside and outside WP) don't notice or care about spaces. That's why inconsistent spacing arises. I did a quick check and found inconsistent spacing NATO, national military organisations, manufacturers, and Wikipedia. It isn't regarded as a feature of the name like the spelling, it's just a style choice made by each publisher with or without a styleguide. If it is important to our readers that our spacing style contains an exception, then we need to ensure that the scope boundary is logical. It seems logical to me that a weapon name (in full or symbol form, metric or non-metric), and a fired-object name (in full or symbol form, metric or non-metric) should have the same style. Lightmouse (talk) 18:55, 20 May 2011 (UTC)

I don't agree with the Olympic spacing, and I don't see why we should be slaves to it. On "5.7×28mm‎", apart from the ISO and MOSNUM, there are two problems: first, it breaches the naming policy in not be recognisable and not being a noun or noun phrase (this unholy clump of characters seems to be an adjective, or if not, it's an ellided noun phrase that lacks the "head", the vital part: 5.7×28mm‎ what? Second, it is an indegistible clump of difficult characters which is why the ISO said to space them in the first place. Tony (talk) 03:55, 21 May 2011 (UTC)
The "unholy clump of characters" is appropriately named according to cartridge naming conventions for metric calibers. The unspaced version is also consistent with common usage, as well as with both of the manufacturer's official websites for the product (FN Herstal and FNH USA), along with the manufacturer's print literature. Furthermore, as I demonstrated earlier, guidelines at WP:MOSNUM#Decimal points clearly indicate that MOSNUM need not be applied to such an extent when dealing with a commonly used naming style like this one. Whether or not the end result is "indigestible" or not is a matter of personal opinion. I happen to think the unspaced version is easier to read; when I'm reading a sentence and I come upon a space, I subconsciously expect it to precede a new word, not a jumbled series of spaced characters. ROG5728 (talk) 05:37, 21 May 2011 (UTC)
I see the Company's website uses exes, not the multiplication sign used by some sources and required by the ISO and MOSNUM. Yet the WP article name uses the multiplication sign. Do you mean we should pick and choose in a random fashion which bits of our house style we apply, and which bits we don't? To hazard a guess, I'd say the multiplication sign was chosen because the ex makes the "unholy clump" even less decipherable than it already is. The other issue is: 5.7 × 28 mm‎ what? Condoms? WP:TITLE says noun phrases are expected in article titles, not adjectives; we had the same issue with "Two-party-preferred" a while back—now Two-party-preferred vote. Tony (talk) 08:42, 21 May 2011 (UTC)
Syntactically Jack Daniel's is a headless noun phrase too, but I don't think anyone would propose moving that to Jack Daniel's whiskey. (OK, that's a bit more familiar than ammunitions, but still...) A. di M.plédréachtaí 17:06, 22 May 2011 (UTC)

No, what I said applies to the multiplication sign, as well. I think it is an unnecessarily detailed application of MOSNUM / ISO, and the line I quoted from MOSNUM would seem to agree (.22 caliber should not be titled 0.22 caliber, even though the latter would normally be correct). I did not initiate or support the recent change to the multiplication sign. To my knowledge, a single WP:MILHIST editor was responsible for that change and all other editors currently discussing the style at WT:MILHIST strongly oppose it. As for 5.7x28mm, it is no less a noun than FN P90 or FN Five-seven (the two parent weapons); 5.7x28mm is a thing, a cartridge. Its name happens to double as a description of the cartridge's dimensions only because cartridge naming conventions are set up like that. ROG5728 (talk) 09:12, 21 May 2011 (UTC)

This debate seems to have wide scope. Do we need consistency between ammunition names and weapon names? Lightmouse (talk) 11:22, 21 May 2011 (UTC)
A good question. ROG? Tony (talk) 12:47, 21 May 2011 (UTC)

Consistency in what aspect, specifically? We can't expect perfect consistency between ammunition names and weapon names. A metric caliber name like 5.7x28mm invariably consists of a difficult clump of characters, by convention (see also: 9x19mm Parabellum, 5.56x45mm NATO, etc). Weapon names are easy to deal with, because they are not named according to any such convention. Either way, though, 5.7x28mm is a noun just like the weapon names are nouns; the 5.7x28mm noun happens to also be descriptive of the cartridge's dimensions (in the manner of an adjective, somewhat) because cartridge naming conventions are set up like that. ROG5728 (talk) 23:16, 21 May 2011 (UTC)

I just don't get it. The article opens with "The 5.7×28mm cartridge is a small-caliber, high-velocity cartridge created by ..."; yet you reverted someone who yesterday who made the article title understandable to everyone. Your preference seems to be for an informal, in-house abbreviation. It does not do as an article title, where the context is not yet provided. The writer(s) clearly felt the context wasn't even provided for the first sentence. Should we now go about closing the ISO/MOSNUM gap between value and unit in hundreds of articles, like 85 mm air defense gun M1939 (52-K)? Tony (talk) 03:14, 22 May 2011 (UTC)
There is no reason to give a complete description of the subject in the title. As WP:TITLE notes, article titles should be as concise as possible, and in this case there are also no similar articles that would call for disambiguation of 5.7x28mm. 5.7x28mm is a noun, so 5.7x28mm cartridge is an overly precise title. As soon as the reader commences reading the article, he/she knows exactly what 5.7x28mm is. He/she can't necessarily know that by the title alone, but the same could be said of just about any Wikipedia article. If you were to read the title (and only the title) of Wikipedia's current featured article, God Hates Us All, and the subject matter was foreign to you, would you know from the title that the subject is a studio album? Definitely not; it could be a book or any other such thing.
The article you linked covers an artillery gun, not a small arms cartridge. Furthermore (and most importantly), "air defense gun" is actually a translation of the full Russian name; it is actually the full name of the subject, as opposed to being a descriptor that was added by Wikipedia editors. Currently, the titles of every cartridge article on Wikipedia do NOT use such a high degree of specificity, nor do they space the value and unit in the title, so your proposal is the one that would actually require changing hundreds and hundreds of cartridge articles like 9x19mm Parabellum, 9x18mm Makarov, 5.56x45mm NATO, etc. The same is true of hundreds and hundreds of WP gun articles: FN P90 is not titled FN P90 submachine gun, FN Five-seven is not titled FN Five-seven pistol, etc, etc. As for the 5.7x28mm title "seeming empty" (as someone else put it), one change that would possibly be appropriate to remedy that would be to title it FN 5.7x28mm, since FN is the designer and manufacturer, and continues to be closely associated with the cartridge. ROG5728 (talk) 04:56, 22 May 2011 (UTC)

FN ... is a whole lot better: then it's clearly a noun phrase, as required, not an adjective. Tony (talk) 05:35, 22 May 2011 (UTC)

I went ahead and moved the article to FN 5.7×28mm. That is technically the full name of the cartridge (see official product website), and the change should simultaneously remedy the problem of trying to be a bit more descriptive in the title without being unnecessarily wordy or redundant. ROG5728 (talk) 07:36, 22 May 2011 (UTC)
It looks to me like the product website actually says "FN 5.7x28mm" instead, which I think matches the way it is commonly spelled. Certainly × can be found in a few sources but it seems to be a decidedly minority usage in cartridge names in my limited experience with the subject. You mentioned this briefly earlier. Quale (talk) 03:57, 24 May 2011 (UTC)

Correct, the ex is definitely more commonly used than the multiplication sign; someone recently applied that change without consensus. WT:MILHIST has been trying to resolve that issue. ROG5728 (talk) 04:39, 24 May 2011 (UTC)

cm vs mm

Some opinions here about an issue raised at BAG would be useful. I personally want to know what stance to take. Tony (talk) 08:02, 22 May 2011 (UTC)

I've opened a new discussion about this at Template talk:Convert#Default target unit for inch conversions. A. di M.plédréachtaí 20:06, 23 May 2011 (UTC)

The topic was raised at a Lightbot discussion page. But note that Lightbot has no dependency on the convert template default for inches. It will operate within a different set of constraints. Lightmouse (talk) 20:46, 23 May 2011 (UTC)

Typography of the ordinal suffix for a number indicated by a variable

If I want to refer to the last of k objects, do I write "kth object", "k-th object", or what? Does the answer depend upon whether this falls within a sentence vs. within a formula? What about the penultimate object; e.g., is it written "(k-1)-th object"? —Quantling (talk | contribs) 19:04, 24 May 2011 (UTC)

See Wikipedia talk:Manual of Style (dates and numbers)/Archive 133#superscripts on ordinals with a variable in them. I'd go with “k-th object” and “(k − 1)-th object” within a sentence. (An example of it “within a formula”?) Anyway, that's more in the scope of WP:MOSMATH than of WP:MOSNUM, I think. (Now, the best part is to decide whether it's (n+1)-th or (n+1)-st.) :-) A. di M.plédréachtaí 19:41, 24 May 2011 (UTC)

LT and ST symbols for long ton and short ton

25-May-2011: The MOSNUM needs to show "LT" as the symbol for "long ton" and "ST" for "short ton". Because the abbreviation "LT" has been used for "long ton" for more than 100 years, people have been finding it in thousands of source documents. This usage requires a change to WP:MOSNUM, and for now, "WP:RWA" ("Reality Wins Again"), so we need to determine the risk of allowing "LT" & "ST" to remain in articles. See 1916 book "A Community Arithmetic" (at Google Books webpage link ACA-p108): which states, "1 long ton or gross ton (LT)". I am unsure why "LT" and "ST" had been omitted in the past. -Wikid77 20:52, 25 May 2011 (UTC)

As I recall "LT" and "ST" were given the thumbs down because they aren't used ... reality won again (or was meant to have). If this was based on a misreading of reality, by all means change MOSNUM, but we really need to be sure what the facts are. One source using "LT" doesn't go that far in supporting the argument. JIMp talk·cont 00:59, 26 May 2011 (UTC)
  • Okay, for any doubting Thomas, the following list of 23 sources, below, shows books in each decade, for the past 124 years, using "LT" to indicate long tons:

Those are some published books which show an abbreviation for "long ton" as a measurement for more than 124 years. I could list a dozen more books; however, I think the use of "LT" to indicate "long ton" has been established since, at least, the year 1887, with continual use through 2011. -Wikid77 (talk) 04:53, 26 May 2011 (UTC)

Is the long ton in use any more? As far as I am aware, all measurements of that order of magnitude uses tonnes (or metric tonnes), the various types of ton have the same order of magnitude, UK English will refer to the "long ton" when dealing with historic matters, but the "tonne" when dealing with contemporary matters, while US English articles will usually deal with the "short ton" and sometimes the "metric tonne". May I therefore put forward the proposition that articles should be consistent in their use of long tons, short tons and tonnes and that at the first mention, the type of ton be clarified as follows: "(long) ton", "(short) ton", "tonne" or "(metric) ton". This is consistent with the style guide used in the Economist. Martinvl (talk) 06:47, 26 May 2011 (UTC)
But see for example London Underground F Stock where LT is used for long tons in the infobox. Thryduulf (talk) 08:26, 26 May 2011 (UTC)
As this artcile refers to a British train the specification and source document was probably in long tons. I would have no problem if the info box used the word "tons" with a reference attached to the first instance that stated something like "Long tons (2240 lbs/1.019 tonnes/1.12 short tons)" . This would remove some of the clutter. IMO, quoting both tons and tonnes in this case is adding trivia. This arguemnt applies only to tons/tonnes, not to kg/lbs or to m/ft. Martinvl (talk) 08:34, 26 May 2011 (UTC)

Consistency in unit conversion target units

As of the current revision of the Ridge and furrow article[6], heights are quoted in inches in two places. In the lead section they are converted to centimetres ("up to 24 inches (61 cm) tall") and in the "Origin" section they are converted to metres ("a height difference of 18 to 24 in (0.5 to 0.6 m)").

I think it's important that articles are internally consistent with which units they are converting to and from (for numbers in the same range), but I can't find mention of this issue in MOS:NUM?

I've left a talk page question about the specific article, as I don't know whether there is a convention for one unit over the other in the relevant field (no pun intended). For reference {{convert}} defaults to millimetres when given 24 inches as the input (24 inches (610 mm)) but given the context (archaeology / historic agriculture) I think that millimetres to be an unlikely choice of unit and would imply too great a precision. Thryduulf (talk) 01:40, 26 May 2011 (UTC)

Fixing MOSNUM to allow LT/ST symbols

27-May-2011: I have been discussing the WP:MOSNUM unit table, to show "LT" as the symbol for long ton and "ST" for short ton. See discussion:

WT:Manual_of_Style_(dates_and_numbers)#LT and ST symbols for long ton and short ton

Because the abbreviation "LT" has been used for "long ton" for more than 120 years, people have been finding it in many source documents. This usage required a change to WP:MOSNUM. The following extensive list of 23 sources, below, shows books in each decade, for the past 124 years, using "LT" to indicate long tons:

Those are some published books which show an abbreviation for "long ton" as a measurement for more than 124 years. I could list a dozen more books; however, I think the use of "LT" to indicate "long ton" has been established since, at least, the year 1887, with continual use through 2011. Based on historical use of "LT" then "ST" naturally follows. -Wikid77 (talk) Wikid77 (talk) 12:58, 27 May 2011 (UTC)

I reverted after I consulted a number of mainstream publications including the Oxford Dictionary of Weights, Measures and Units and a [US Census publication] both of which declined to list abbreviations or symbols for long and short tons. Moreover, writing as a British computer programmer, the text "LT" means "less than" to me, while "ST" is an uppercase variant for "steradian". If you want to reinstarte, please cite an authority such as ISO or NIST rather than a cookery book. Martinvl (talk) 13:23, 27 May 2011 (UTC)
Surely we should be responding to 124 years of actual use rather. including by such organisations as the USGS [], the NIST [7], American Heritage Dictionary, Collins English Dictionary and Random House Dictionary [8], World Coal Organisation [9], rather than what some arbitrary authority doesn't say? Thryduulf (talk) 14:31, 27 May 2011 (UTC)
Firstly, I have never seen LT or ST in use to represent "short tons" or "long tons". I consulted my copy of the "Concise Oxford Dictionary" (1960 vintage which has 14 pages of abbreviations). It showed "st" (lower case) as being an abbreviation for "stone", but made no metion of "short tons".
It is possible that "LT" and "ST" are in common use in the US, but they are certainly not in common use in the UK. If that is the case, then they have no place as being a default abbreviation in Wikipeidia (which is meant to accomodate English-speaking readers from all countries). If an author wishes to use them, then I suggest that he refers to the notes on the use of the words "lakh" and "crore" and use the philosophy. Martinvl (talk) 15:57, 27 May 2011 (UTC)

The current guideline is based (at least partially) on this discussion. JIMp talk·cont 16:08, 27 May 2011 (UTC)

Quantities of bytes and bits

I wonder if it should be added in this section that despite the two systems being ambiguous, most of the time this isn't important, although I guess it will become more important over time as the error builds up with each larger prefix. This seems to be the way most articles are written, but this section of the MoS seems to suggest conversions/clarifications should always be present. -- (talk) 14:40, 27 May 2011 (UTC)

Year ranges

I see a problem with the current year ranges guidance that we're giving here. The current text in question is as follows:

Year ranges, like all ranges, are normally separated by an en dash, not a hyphen or slash: 2005–06 is a two-year range, whereas 2005/06 is a period of twelve months or less, such as a sports season or a financial year. A closing CE or AD year is normally written with two digits (1881–86) unless it is in a different century from that of the opening year, in which case the full closing year is given (1881–1986). A closing BCE or BC year is given in full (2590–2550 BCE). While one era signifier at the end of a date range requires an unspaced en dash (12–5 BC), a spaced en dash is required when a signifier is used after the opening and closing years (5 BC – AD 29). Ranges expressed using prepositions (from 1881 to 1886 or between 1881 and 1886) should not use en dashes (not from 1881–1886 or between 1881–1886).

I see a couple of issues with this point. First, where does the 2005/06 format come from? I don't recall ever seeing that used, let alone recommended.

Second, and more immediately important, is the "unless it is in a different century from that of the opening year" portion. The example given is a good one, in that we should obviously be including the century portion of both years when there are a number of years in both centuries. So, the example given (1881–1986) is an excellent example. The problem is when the date range is within a single 12 month period. This is most often seen in sports related articles, especially for seasons that took place over the winter of 1999 and into 2000. Some editors have taken the current guidance here and converted all instances of 1999-00 to instead say 1999-2000. These edits have been reverted in many instances, for what appears to me to at least be a reasonable rational. Within sports related articles, where what is being referred to is clearly and unambiguously the leagues season, using 1999-00 is preferred by many because it is more succinct and is consistent with the date range used for the preceding and following seasons. Therefore, I'm planning on adding a caveat to the text above about sports seasons, but I wanted to start a discussion about it first.
— V = IR (Talk • Contribs) 14:00, 3 June 2011 (UTC)

"2005/06" is widely used both in real life and in articles to describe a season (for example, a sports league season, budget season or a university semester). I think 1999-00 is acceptable since there is little danger of confusion. 1881-00 would be different thing. Nanobear (talk) 14:10, 3 June 2011 (UTC)
1999–00? That's weird. I've never seen it before. Why is the "single 12-month period" relevant? Tony (talk) 14:11, 3 June 2011 (UTC)
Not a sports fan, eh? (or do they do something different down under?) You can easily see examples of these date ranges given for seasons in thousands of Soccer, (American) Football, Basketball, and Hockey articles (to name just a few). The usual convention is to give the full year for the year that the season started in, and then simply the decadal for the year that the season ended in. People naturally understand that, for example, The 2011-12 season refers to the team sports leagues' season that runs from some point in the year 2011 to some point in the year 2012. There's really no danger of confusion here, so it's normal to use the abbreviated 2011-12 form (This is probably driven mostly by newspapers, since there's a cost per character within the industry. I know that you're aware of that issue, Tony. However, whether or not it is due to the fact that newspaper coverage of sports events are so widespread, the fact is that the sports community has overwhelmingly accepted the convention here). Anyway, I'm generalizing it somewhat to include the possibility of use outside of sports topics. Maybe that's a mistake though, and this should be limited to sports related date ranges?
— V = IR (Talk • Contribs) 14:42, 3 June 2011 (UTC)

I'm trying to come up with a good way to word this though. I'm thinking that adding something to the end of the current sentence would be most appropriate, so:

  • A closing CE or AD year is normally written with two digits (1881–86) unless it is in a different century from that of the opening year, in which case the full closing year is given (1881–1986). The different century rule does not apply in cases where the year range only covers two calender years however, such as the range from 1999 to 2000, in which case the closing year can be written with two digits (1999–00).

the problem is, that seems overly verbose to me.
— V = IR (Talk • Contribs) 16:44, 3 June 2011 (UTC)

Ultimately, it comes down to context. In an article, especially one related to sports, progression from 1998–99 to 1999–00 to 2000–01 would be understood, i.e.: [10], [11]. So I definitely support modifying this MOS to that end. In my experience, 2010–11 is usually the preferred format for a season crossing two calendar years. 2010/11 is something I usually see in non-English sources. Resolute 01:26, 4 June 2011 (UTC)


A closing CE or AD year is normally written with two digits (1881–86) unless it is in a different century from that of the opening year, in which case the full closing year is given (1881–1986). Exceptions may be made where the change in century would be understood in context, such as a sports season, where 1999–00 would be expected to mean 1999–2000

? Resolute 01:32, 4 June 2011 (UTC)

Should not an exception be made out of necessity, rather than to suit a personal preference? Usage across 1999 to 2000 is varied. I've seen 1999-00, 1999-2000 and 99-2000. Now, people often write lazily. Example, 2nd instead of second. This is what passes for writing in sports media and social media. I think that the Wikipedia MOS tries to aspire to writing style above the shorthand, no? That said, my concern there would be in the prose and article titles. Let's not go for a 'lazy style.' Let's use a good style as much as possible. Aspire to good writing. There could be a clarification that you could use 1999-00 in tables, where the preceding and following seasons are shown, so that there is information as to what is meant. I think it would be wrong to use 1999-00 without a link or surrounding context, no? ʘ alaney2k ʘ (talk) 04:01, 4 June 2011 (UTC)
I can agree with the idea that we shouldn't write lazily within Wikipedia, but that's not what we're talking about here in my opinion. The use of SSSS-EE year formatting is a common convention that is used, which isn't driven by laziness. Within the context where it's used it's easily understood what the meaning is, which makes it's use desirable. Numbers tend to break up the flow of running text which makes anything that shortens the numbers which need to be given, while retaining meaning, desirable. Within tables it's use helps as well, since it shortens the width required for the row.
It seems to me that we've been clear that the 1999-00 (and 1899-00, mostly for baseball and soccer) convention is limited in use to those areas where it is unambiguous. Both Resolute and I were clear that the surrounding context had to be present, I think.
— V = IR (Talk • Contribs) 11:39, 4 June 2011 (UTC)
I think we agree that it is okay in tables. I don't think it is appropriate simply because it is a sports article. ʘ alaney2k ʘ (talk) 23:16, 5 June 2011 (UTC)
I agree with the idea that it's not limited, or because, the information is in a sports article, but realistically that's where these cases are going to be seen the most. I think. I don't really have anything to back that up, but it seems correct to me based on my own experience... I'd love to see examples of other areas that this could affect though. If we need to adjust things, it would be nice to do that now.
— V = IR (Talk • Contribs) 23:21, 5 June 2011 (UTC)
Frankly, I've never seen this convention, and it doesn't look good to me. Better just to drop the century indicator ('99–'00) if pressed for space in a table. To those unfamiliar with the convention, it's just confusing; eventually they'll decide that for want of a likely alternative, it must mean xx99-xy00 [or x999-y000], but they still won't be completely sure. There are all kinds of non-athletic years that cross 31 December/January 1, such as academic, ecclesiastical, legislative and fiscal (financial) years, as well as some military campaigns, some annual reference books, and of course meteorological winters (or in the Antipodes, summers). I wouldn't want to see "the South African War (1899-'02)", "the winter of 1799-'00" or "the 1999-'00 budget".—— Shakescene (talk) 11:12, 6 June 2011 (UTC)
If you read what's actually being said here though, you'll see that 1899-'02 is specifically discouraged (the text has always stated that: unless it is in a different century from that of the opening year, in which case the full closing year is given (1881–1986).). Would you find During the 1999-00 school year, the students... to be confusing or ambiguous, at all? (good call on academic, ecclesiastical, legislative and fiscal years, by the way).
— V = IR (Talk • Contribs) 18:58, 6 June 2011 (UTC)
PS: The MoS has discouraged the use of an apostrophy for the removed century (2011-12 not 2011-'12) for a long, long time, as far as I can tell.
— V = IR (Talk • Contribs) 18:58, 6 June 2011 (UTC)

Simply saying 'exceptions' is not acceptable. That is why I reverted the change. It should be made explicit to the point where it achieves consensus. Note that the sentence for ranges of less than two years is supposed to be 2005/06 and not 2005-06. We should clear this up too. ʘ alaney2k ʘ (talk) 16:51, 10 June 2011 (UTC)

What does "explicit to the point where it achieves consensus" mean, exactly? Anyway, I can see how "Exceptions may be made" could be problematic, in that it's not very imperative, but keep in mind that this page is for users not computer programs or AWB instructions. Since we're talking about it though, is there demonstrable consensus anywhere for changing existing uses of "1999-00" into "1999-2000"? And can you explain your reasons why for your assertion about "ranges of less than two years is supposed to be", please?
— V = IR (Talk • Contribs) 16:57, 10 June 2011 (UTC)
See the sentence "Year ranges, like all ranges, are normally separated by an en dash, not a hyphen or slash: 2005–06 is a two-year range, whereas 2005/06 is a period of twelve months or less, such as a sports season or a financial year.". -All- of the article pages are title 1999-2000 'whatever', so this affects a lot of current usage to just give exceptions. I would propose adding the modifier, 'where the context of the preceding season and following season is given, for example in tables.' ʘ alaney2k ʘ (talk) 17:08, 10 June 2011 (UTC)

Spacing of en dashes in ranges

There is a discussion at Wikipedia talk:Manual of Style/dash drafting/discussion#Spacing of en dashes in ranges. Ozob (talk) 10:48, 13 June 2011 (UTC)

What date format should be used by bots for article references which don't match the date format templates?

Hypothetical article has {{use dmy dates}} that asks to use dmy — "8 October 2011". The two questions below are exclusively about references/citations.

The article contains a citation {{cite web |url= |title=Better Example |accessdate=2010-11-13}}. Question #1: should a bot adding |archivedate= to this citation use "2011-10-08" or "8 October 2011"?

The article also contains a citation {{cite web |url= |title=Example page |accessdate=14 Sep 2010}}. Question #2: should a bot adding |archivedate= to this citation use "8 October 2011" or "8 Oct 2011"? —  HELLKNOWZ  ▎TALK 10:04, 8 June 2011 (UTC)

Those damn date format templates are a real problem... argh! Anyway, to answer your actual question, please use something other than ISO style dates. So: "8 October 2011", or "October 8, 2011". Abbreviated months are just unwanted as far as I can tell, so I don't think that changing them would be seen as problematic by anyone. Now that I think about it though, ISO dates do have their supporters for various reasons (which I personally think are crap, but whatever). That being the case, changing them by bot is probably a bad idea... except making all dates consistent is a good thing, if you can determine what would do that (which, in my experience, bears little or no relation to the date template that someone has happened to slap in to an article). If most reference dates are in ISO format, then they should probably all be in ISO format, regardless of what formats dates in the prose happen to use.
— V = IR (Talk • Contribs) 13:11, 8 June 2011 (UTC)
I'd say, if it says to use dmy dates, let it use them. Similarly, if it says to use mdy dates, use them. I notice, though, that there exists {{use ymd dates}}; why, when, according to this page, they "are uncommon in English prose, and should not be used within sentences."? JIMp talk·cont 03:40, 9 June 2011 (UTC)
I've found the answer to my own question. The template may be called {{use xyz dates}}, but the doc makes it clearer: "xyz dates are used". Ichiro Suzuki, for eaxmple, is tagged with {{use mdy dates}} and {{use ymd dates}}, both are used for different purposes (e.g. mdy in the text & ymd in {{persondata}}). JIMp talk·cont 06:21, 9 June 2011 (UTC)
Personendata is usually for machine use only and is invisible by default. It seems that its date field is not format sensitive, but the advice (per the documentation) is to follow the MOS. There's no point of transcluding another template that is inconsequential, if you ask me. --Ohconfucius ¡digame! 06:38, 9 June 2011 (UTC)
I abhor ISO date style. I say stick to {{use xxx dates}} recommendation. Fleet Command (talk) 05:30, 9 June 2011 (UTC)
The answer is, of course, use the dominant date format present in citations. Headbomb {talk / contribs / physics / books} 16:11, 9 June 2011 (UTC)
Now, hang on a second, does "present in citations" mean in the news/magazine/book source itself, of the date that the editor used in the citation on our page? I've never been real clear on this point, and I get the feeling that others have the same problem.
— V = IR (Talk • Contribs) 16:29, 9 June 2011 (UTC)
Present in whatever appears in our Reference section. Headbomb {talk / contribs / physics / books} 23:51, 9 June 2011 (UTC)
Our page, I'd say. Consistancy is important. I agree with Headbomb ... almost. There are some formats which we don't want. If the ref section has, for example, "9/6/11", we shouldn't follow suit. JIMp talk·cont 23:43, 9 June 2011 (UTC)
Well obviously that's between allowed formats. 26 Mar 2001; 2001-03-26, 26 March 2001; March 26, 2001; etc... Headbomb {talk / contribs / physics / books} 23:51, 9 June 2011 (UTC)
@Jimp @Headbomb So ignore {{use xxx dates}} templates? —  HELLKNOWZ  ▎TALK 06:40, 10 June 2011 (UTC)
If an article has {{use dmy dates}}, I'd say you should use the indicated format, although you wouldn't necessarily be wrong to use yyyy-mm-dd format. You certainly shouldn't use mdy. Please note that Rjwilmsi's bot does just that. --Ohconfucius ¡digame! 06:51, 10 June 2011 (UTC)
  • Geeks and technical people: why is it that normal readers are expected to know what those telephone-number gobbledies mean? Even I have to stop on each ISO date and parse it consciously. It's made worse by the fact that normal people learn another way of numericalising dates, depending on where they live (8/6/70, which is June 8, 1970 or August 6, 1970 ... isn't this a complicating factor in thrusting yet another all-numbers system at the poor readers? (70-08-06, or is it 70-06-08? I have to stop and think carefully, and those zeros look like they're made for filling out a computerised form.) What was the original fascination with putting ISO dates just in ref lists, when article text, thank god, has normal-speak dates? Who complains about normal, spelled-out dates in ref lists ... what is the disadvantage in using them consistently throughout an article? Why do we put up with inconsistency within article in this respect but no other (aside from quotations, etc)? I put it to you that the month should be spelled out for ease of reading, given our massive and widely strewn readership. Tony (talk) 07:12, 10 June 2011 (UTC)
  • This is slightly off-topic, but I would be perfectly happy only using spelled out dates for my bot. But this is not what WP:CITEVAR and {{use ymd dates}} advise. Hence my question on record that shows I did consult others. Currently, I follow the date format templates; and if none are found, whatever allowed style is already used in the citation. —  HELLKNOWZ  ▎TALK 12:13, 10 June 2011 (UTC)
  • What you suggested is fine. I don't think CITEVAR or {{use ymd dates}} advise any more that MOSNUM or {{use dmy dates}} do. If the task is to be done by bot, then I believe it should follow the format defined by the template, or yyyy-mm-dd if all or most of the refs section are in that fortmat. --Ohconfucius ¡digame! 01:24, 11 June 2011 (UTC)
  • I'm not actually defending ISO dates (I hate them myself), but the reasons that I've come across for their use are because they are compact (which seems to be a concern within references, for some reason), and... well, because they are understandable to those who use them regularly, apparently. There may be other reasons, but those are what I notice.
    — V = IR (Talk • Contribs) 12:55, 10 June 2011 (UTC)
  • Ohms, I thought of compactness and dismissed it because phone-number dates are not used in infoboxes, where space is really at a premium. As you hint, it's hard to see that space is an issue in ref lists. I can see that some tables might benefit from their compactness, that's all. I wonder when we're going to find a systemic cure for this unexplained disease. I really don't think your average reader understands them. Tony (talk) 14:20, 10 June 2011 (UTC)
When they're used in tables it's even worse because you tend to just get a wall of numbers and very little information successfully communicated. See for example this list. Even worse (because of the extra dimension of numbers) is this one - and I would suggest that it would be no clearer if they switched the orders of the numbers around to yyyy-mm-dd. Pfainuk talk 19:16, 16 June 2011 (UTC)
  • Well, you know, "perception is reality". Regardless of whether or not you and I think that space is at a premium in references, other people certainly do. I think that part of that is driven by the fact that references are inserted inline, so they break up the flow of the source text. I think that gives editors a feeling that they need to make the text within the reference tags as short as possible.
    — V = IR (Talk • Contribs) 14:37, 10 June 2011 (UTC)
  • I wouldn't say ignore the templates. If the templates don't agree with the style used, something's amiss.
  • I second Tony questioning re phone# dates in refs. I see no reason not to simply follow the format in the article body.
  • I don't buy the spurious argument for compactness. Just use a three-letter abbreviation for the month.
JIMp talk·cont 22:28, 10 June 2011 (UTC)
The ymd format is standard in astronomy, in both tables and prose. See, for example, the Five Millennium Canon of Solar Eclipses by NASA. It is not ISO format: The month is always spelled out or abbreviated to three letters, never numbered; and dates before 1582 are always in the Julian calendar, not the Gregorian calendar. — Joe Kress (talk) 01:10, 11 June 2011 (UTC)

A template for physical constants?

You are invited to join the discussion at WT:MEASURE#A template for physical constants?. A. di M.plédréachtaí 14:31, 16 June 2011 (UTC) (Using {{pls}})

Conversion from Hijri calendar to Gregorian calendar

Usually there have been several attempts to edit sourced Gregorian dates at various Islam/Arab related articles.
Editors who change the dates (or convert it from Hijri calendar to Gregorian calendar) should note following points:

  • Apparently change of date is being done on basis of some date conversion tool(s),
    • Most of the date conversion tools are not accurate (try to convert a recent date which you know and you will get a difference of day or two, this difference tends to scale up when we talk about dates as old as 1500 years)
    • Hijri dates tend to be different at different parts of world at same point of time (e.g. if Id-ul-fitr falls on Monday in Hejaz, in Persia it will be on Tuesday and in Hind it will be Wednesday and so on) most convertor tools don't take such considerations in view (& believe me b'coz I myself am a developer and its too difficult to make a convertor tool having such minute consideration and preciseness and it will require great effort and resource to make such tool and very few people will give it for free)
  • The converted dates availabale in sources like the Encyclopedia Britannica are the work of professional date experts & are quite precise but they also don't claim it to be cent-per-cent accurate because in Hijri calendar actual celestial date and perceived / accepted date may be different.

Considering above points please change dates only when you have sufficient proof and not on basis of any date conversion tool.

A word of caution should be added regarding the conversion of Islamic dates (especially those before 10 AH) into Western calendar dates. All modern conversion tables (and programs on the internet) assume that the ancient Arabian calendar (as observed in Mecca) was identical with the present-day Islamic calendar and do not account for the intercalary months which (up to 10 AH) were inserted every two or three years in order to keep the calendar in step with the seasons. As we know very little about the regulation of the lunar calendar in Mecca before 10 AH it is pointless to argue which Western calendar date best corresponds with Ali's birth (the same b.t.w. also applies to the claimed dates of Muhammad's birth).

Also, the start of year till 60-61 AH was not from the month of Muharram but from the month of Rabi-al-awwal. And this fact is also not considered in most of the modern conversion tools.

What is your source for this claim? AstroLynx (talk) 11:27, 11 June 2011 (UTC)

See also the discussion of this problem at, Islamic_calendar#History, Edits related to conversion from Hijri calendar to Gregorian calendar, Islamic-Western Calendar Converter & Martyrdom of Imam Husayn and the Muslim and Jewish Calendars.

Can we have some guidelines related to this?
--Sayed Mohammad Faiz Haidertcs 11:57, 10 June 2011 (UTC)

This is more appropriate for Wikipedia:Manual of Style (Islam-related articles) and its talk page. — Joe Kress (talk) 01:53, 11 June 2011 (UTC)
There are many historical events that has mentioned by Hijri calender in the Muslim world which does not neccesarily related to Islam related articles. These articles includes birth dates of renowned people, date of battels, etc. --Seyyed(t-c) 04:29, 11 June 2011 (UTC)
I agree here, as some of these dates mention historical figures in-general. Some aren't even related to Islamic history. ~ AdvertAdam talk 06:01, 11 June 2011 (UTC)
I've seen many bias conversions, trying to even convert months and days. It's misleading as westerns might think that a certain date is a holiday, and editors will disagree on who's right (while they're all wrong). I suggest using the AH as a standard and note that adding "(approx. #### CE)" should be done carefully. ~ AdvertAdam talk 06:01, 11 June 2011 (UTC)
Great suggestion Adam. It is better to have AH dates as preferred one especially for early Islamic/Arab dates and Julian/Gregorian dates should be included on basis of academic RS e.f Encyclopaedia Britannica (and not on basis of some on-line date converter which may even not convert date of present year correctly).--Sayed Mohammad Faiz Haidertcs 19:10, 11 June 2011 (UTC)
Unfortunately, in some cases the reliable sources disagree on the Gregorian dates for the same Hijri date.--Seyyed(t-c) 06:56, 12 June 2011 (UTC)
Agree (when RS are available), however, a MOS guidelines for conflicts are tough (Sa.vakilian); so we just need to inform in the suggested MOS to be careful, and force "est/approx". Maybe we can suggest a slight estimate when there's conflicts, like (est. mid-sixth century CE). Moreover, conversion--with RS--should only be used when needed, to visualize a historical incidence that needs an estimate (like births, incidence, etc).
Any drafts to get a push here :) ? ~ AdvertAdam talk 10:42, 14 June 2011 (UTC)
If you do, I believe WP:YEAR prefers "c." to "est/approx." Art LaPella (talk) 20:07, 13 June 2011 (UTC)
Yep, looks good. I forgot to sign above... ~ AdvertAdam talk 10:42, 14 June 2011 (UTC)
Don't you add anything to Wikipedia:Manual of Style (dates and numbers).--Seyyed(t-c) 08:20, 18 June 2011 (UTC)

Ton(ne) conversion templates

Tonne conversions
1 tonne = 1000 kg
1 tonne = 1.1023 US (short) tons
1 tonne = 0.9842 UK (long) tons

We currently convert all metric units to imperial and vice-versa. In some cases this can get tedious, particularly for those units where the UK and the US versions are different - tons, gallons, quarts, pints and fluid oz. I propose that editors should have the option of replacing in-line conversions for these units with tables as displayed alongside this posting. This will make for easier reading of articles that have a large number of measurements - for example

"The levels of rock cod taken in the whole of the South Atlantic dropped from 399,700 tonnes (393,400 long tons; 440,600 short tons) in 1969-70, to 101,560 tonnes (99,960 long tons; 111,950 short tons) the following year and 2,740 tonnes (2,700 long tons; 3,020 short tons) in 1971/72"

will become

"The levels of rock cod taken in the whole of the South Atlantic dropped from 399,700 tonnes in 1969-70, to 101,560 tonnes the following year and 2,740 tonnes in 1971/72".

I beleive that the latter is much easier to read and that readers who are really interested in the last decimal place are in such a minority that they can work it out for themselves - the information that they need will be present. I propose a creating templates for tonnes (as shown alongside), US (short) tons, UK (long) tons and later extending the set to include fluid measures. Comments? Martinvl (talk) 18:19, 19 June 2011 (UTC)

The difference in the numbers, particularly with the short tons, is quite significant, and I think expecting people to go around doing non-trivial maths when all they want to do is understand the article probably isn't appropriate for a general audience. In this case, if we're trying to illustrate the decline, I think it would actually probably be better to say that it had declined by X% in 1971-72 rather than just listing the raw figures. The basis for this could be placed in a footnote in all of its converted glory. Pfainuk talk 21:12, 19 June 2011 (UTC)
a "The broggage was 12.56 snugs (14.07 kiloflinks or 13.89 hockles) last Frelig."
b "The broggage was 12.56 snugs last Frelig. ... (P.S. 1 snug = 1.120223 kilofinks = 1.10589 hockles ... you go figure it out)"
year level of rock cod taken in the South Atlantic
tonnes long tons short tons 1969/70 comparison
1969/70 399,700 393,400 440,600 100%
1970/71 101,560 99,960 111,950 25.4%
1971/72 2,740 2,700 3,020 0.69%
Sentence b is much easier to read ... for those who think interms of snugs but not for those who think in terms of kilofinks or hockles.
My take on it is that when we've got so many figures a table is often the best way of presenting the data. The table could be put in the footnotes if need be.
"From 1969/70 to 1971/72 the levels of rock cod taken in the whole of the South Atlantic dropped by 99.3%."
JIMp talk·cont 00:13, 20 June 2011 (UTC)

Default to lower case for non-metric units, list exceptions

The US and UK governments use lower case by default for unit abbreviations. That appears to be the unstated convention on Wikipedia but I do see inconsistencies (e.g. MPH, Mph, mph, PSI, Psi, psi, LB, Lb, lb, FT, Ft, ft). It would be useful to make the convention explicit and only require debate for exceptions.

The 'Unit symbols' section appears to be the right place for this. I suggest something like:

  • Non-metric units should be written in lower case unless otherwise specified in this Manual of Style (dates and numbers).

The 'Specific units' table is the place to list the exceptions. I see it already has a few instances. What do people think? Lightmouse (talk) 17:31, 6 June 2011 (UTC)

Sounds fair unless the list gets out of hand. JIMp talk·cont 18:06, 6 June 2011 (UTC)
I agree.
— V = IR (Talk • Contribs) 18:59, 6 June 2011 (UTC)
Agree. Tony (talk) 10:27, 8 June 2011 (UTC)
I agree too. Headbomb {talk / contribs / physics / books} 16:12, 9 June 2011 (UTC)

Done. See Wikipedia:Manual_of_Style_(dates_and_numbers)#Unit_symbols. Regards Lightmouse (talk) 17:16, 16 June 2011 (UTC)

I have clarified the definition by specifying the SI brouchure (which inlcudes a number of pre-SI and non-SI symbols such as "Ci" for "curie", "R" for "röntgen" etc) or the ISO 8000 document which is aligned with the SI brouchure, but which specified "ft" for the deprecated unit the "foot".Martinvl (talk) 17:45, 16 June 2011 (UTC)
  1. The clause states requires 'ft' to be lower case. Why do I need to go to ISO 8000 to see that 'ft' is lower case?
  2. Please can you provide a link to the text of ISO 8000 so editors can use that clause?
Lightmouse (talk) 18:14, 16 June 2011 (UTC)
Unfortunately ISO 8000 costs loadsamoney, but a few parts of it are available
  1. Part 3 - Space and timehere
  2. Part 4 - Mechanics: here
  3. Part 5 - Thermodynamics: here.
Martinvl (talk) 18:30, 16 June 2011 (UTC)
They say 'Draft Ethiopian Standard'! Lightmouse (talk) 18:35, 16 June 2011 (UTC)
I know, but the ISO 8000 standard can be reverse-engineered from them by replacing the "," decimal separator by a ".". Martinvl (talk) 19:22, 16 June 2011 (UTC)

I deleted the reference to the ISO 8000 standard because editors can't access it. I kept the reference to the SI brochure. I sympathise with the intention to use an international standard but the US and UK government sources should be enough. Does the ISO 8000 mention any unit that isn't mentioned by the SI brochure or those two government links? Lightmouse (talk) 08:19, 17 June 2011 (UTC)

I have reintated my original statement. In particular, the text that I removed prohibits the use of "Ci" for "curie", "R" for "röntgen" as these are non-metric units, even though they appear in the SI Brochure. Martinvl (talk) 08:57, 17 June 2011 (UTC)

Martin, this interaction is frustrating. The proposed text has been on this talk page for a while and I know you watch this page. Your text hasn't been reviewed and instead of engaging with the community here, you are reverting the guide itself.

You drew my attention to the fact that some non-metric units are in the SI brochure so I added that. However both instances of your text are unworkable for readers by your own admission it requires readers to read a document from Ethiopia! And I had to drag that put of you. I share your desire for an international standard for non-metric units but editors can't comply with your text. They can with mine. Can other people are what I mean? Lightmouse (talk) 09:38, 17 June 2011 (UTC)

LM's change was reverted on the basis that editors "can't access" the ISO8000 list. I'm surprised this list isn't on the net somewhere. Tony (talk) 10:28, 17 June 2011 (UTC)
ISO is a private organisation and one has to pay for copies of their publications in the same way that one has to pay for copies of learned journals. Parts of the ISO 80000 standard are available unde rthe guise of Ethiopian Standards - the only material difference being the use of the comma, not the dot as a decimal separator. Martinvl (talk) 14:40, 17 June 2011 (UTC)
May I propose a totally different wording which will, I believe, achieve the same result:
  • The rules that governing the use of upper case and lower case letters for all unit symbols (or abbreviations) shall follow the rules in the SI Brochure - use lower case except where the unit is named after a person. The use of "L" or "l" for litres is an exception to this rule and is discussed in the brouchure.
This will accommodate degrees Fahrenheit (°F), röntgens (R), curies (Ci) and so on. Martinvl (talk) 20:19, 21 June 2011 (UTC)
I had no response to my lapartially copied from the SI brochure, removes the word "metric" but otherwise is consistent with what I believe people want. Martinvl (talk) 18:14, 23 June 2011 (UTC)


  • I’m too busy to address this one at the moment and it would take too long for me to research whether Martinvl’s change to the guideline (∆ edit, here) is sound advise. Normally I would write For non-SI units of measures, you should generally follow the practices of most-reliable RSs used in the field or (preferably) that particular article. His edit appears to be extraordinarily broad and written as if he is a primary authority trying to establish “best practices” for planet earth so as to have perfect cross-project consistency for unit symbols, which is not achievable. I think this change is too radical and am reverting until this is more fully discussed and and the principal of “following the RSs” is properly addressed. I find his change too prescriptive. Greg L (talk) 04:37, 24 June 2011 (UTC)
I find Greg's approach "The consensus appeared to be one in which the SI brouchure was followed for SI/metric (which?) units, but to use lower case otherwise, except where agreed. The cgs unit (metric, but non-SI) of viscosity is the stoke (St). The customary unit of temeprature is the degree Fahrenheit (°F) and for radioactivity is the curie (Ci). I believe that I have encompassed all of these issues. I have therefore reworded my proposal and reinstated it. Before anybody reverts, please ask yourself which RSs are not being addressed. I cannot think of any. Martinvl (talk) 06:51, 24 June 2011 (UTC)
What about the byte? :-) (IOW, trying to encompass all the units in the universe only means we're likely to forget about some of the exceptions. We should say “generally”, with the understanding that if one wants to be 100% sure of what the symbol of – say – the galactic year is, they can just read the article Galactic year. Duh.) A. di M.plédréachtaí 17:34, 24 June 2011 (UTC)

Date unification

I just edited a portion of MOS:DATEUNIFY so that it's actually recommending that people use the same dates formats throughout, which is what I always understood the section to mean anyway. If anyone wants to object, I'd appreciate it if we could discuss the matter here. Thanks
— V = IR (Talk • Contribs) 16:11, 10 June 2011 (UTC)

Alaney2k (talk · contribs) reverted without discussion, so I reverted the text back. I'm sticking to my point here because the history of the section clearly shows that the intent is that all dates use the same format (one for prose and one for references). The talk page archives show the same, as well.
— V = IR (Talk • Contribs) 16:46, 10 June 2011 (UTC)
...and, he's reverted back again. Nice. I'd like to remind people that you're supposed to propose changes to policy before changing the text on the policy page. I'll come back to this tomorrow though, in order to avoid the 3RR.
— V = IR (Talk • Contribs) 17:03, 10 June 2011 (UTC)
I only meant to revert the text in the Years section. I obviously messed up this other part. The Years section needs to be discussed, above. I'll fix this up. ʘ alaney2k ʘ (talk) 17:12, 10 June 2011 (UTC)
On looking it over, I don't think that I changed the dateunify text. What is the error? ʘ alaney2k ʘ (talk) 17:17, 10 June 2011 (UTC)

Ohms, you're the one making the change. Previous discussions have said that date styles in references can have one style for publication dates and a different style for accessdates. Thus the examples with mdy or dmy for the publication date and yyyy-mm-dd for the accessdate are OK. Gimmetoo (talk) 19:15, 10 June 2011 (UTC)

  • Let's redraw the line: it's clearly nonsense to permit, although the examples don't specifically say so and some people have interpreted it this way, for dates to be mdy in the body and dmy in the refs section. --Ohconfucius ¡digame! 01:07, 11 June 2011 (UTC)
Clearly. I'm not sure why it isn't clear that permitting ymd dates in the refs when there are normal human style dates in the body. JIMp talk·cont 06:13, 13 June 2011 (UTC)
That last comment doesn't parse. Gimmetoo (talk) 00:31, 14 June 2011 (UTC)
It is missing a verb after "refs", but there are enough clues for me to translate it: "Why do we permit ymd dates in the refs just to please computers, when there are normal dates that humans (not computers) use in the body of the article?" Art LaPella (talk) 01:07, 14 June 2011 (UTC)
We "permit" different styles in different places because that's how the Wiki developed organically, and we MOS:RETAIN the existing style as far as possible4 unless there is a damn good reason for changing it. Also, a yyyy-mm-dd format (presumably what you mean by ymd, which means something else to me) is not "just to please computers", though I wonder why that would matter if it was. Date style discussions here have routinely distinguished body text from ref texts. Gimmetoo (talk) 02:05, 14 June 2011 (UTC)
  • Gimmetoo, "We "permit" different styles in different places because that's how the Wiki developed organically"—That kind of "we do it because we do it because we do it" reasoning is tantamount to saying that nothing should change. I'm sorry if you've adopted a personal policy of opposing change for the sake of it; a wiki's advantage is that it can change, evolve, improve, continually. I'm happy for Ohms to go ahead with this change. Tony (talk) 10:31, 17 June 2011 (UTC)
    • No, it's not "we do it because we do it", and I know you know that. You also know very well the reasons for the yyyy-mm-dd style, and you also know that when a style issue splits the community, the guidelines stay to stick with the existing style. Furthermore, once you make an edit that changes the style of the entire article, just about any subsequent edit will create edit conflicts and make your style change difficult to undo, which makes it a WP:FAITACCOMPLI. Come on. Gimmetoo (talk) 13:23, 17 June 2011 (UTC)
  • I'm disappointed with the continued efforts to phase out the YYYY-MM-DD format (of which this edit seems to be a part), although the majority of editors has supported retaining the format. There are many arguments for retaining (please see the link); it is disruptive to demand that the arguments are repeated every time the latest drive to remove YYYY-MM-DD against consensus starts. Nanobear (talk) 11:47, 17 June 2011 (UTC)
    This is one good reason to make sure date formats are stated unambiguously, with the month identified in words, instead of being written in machine code. --Ohconfucius ¡digame! 15:20, 17 June 2011 (UTC)
    What are you pointing to as a "reason" in that diff? An existing article had a clearly consistent reference style [12] and a new editor added a bunch of references apparently with the same style [13] but made a couple mistakes, including forgetting to close some of the templates with }}, and copy-pasted some of the mistakes a number of times in the article. And your solution is not just to close the templates and fix any other issues with the content and refs added by the new editor, to align with the pre-existing style, but to completely change the existing style of the references in the article [14]? Gimmetoo (talk) 22:19, 17 June 2011 (UTC)
    Have a closer look: they were not consistent. You may have understated the magnitude of the error, as it affected almost half the dates in that refs section. On the surface, it seemed that the problem was chronic, and dated back to as early as "2009-30-03". Yes, you also noted the citation templates were a mess, probably because someone did a search-and-replace carelessly on top of that, so please stop lambasting me for correcting these errors. Of course, you would be excused for not noticing the errors in the 'machine code' dates (or their extent). Who knows which other dates on there were wrong – I only corrected those that were manifestly so? Such errors would be instantly apparent if 'months' were written out in plain words. --Ohconfucius ¡digame! 02:55, 20 June 2011 (UTC)
    Have another look. I linked to this version of 14 June, before a newly-registered editor began editing the article. I don't see any instances of "2009-30-03"; it looks consistent to me, but maybe I'm missing something. The newly-registered editor then added some text and refs, and made some mistakes. No offense intended to the editor, but yes, there is a learning curve with wikipedia. OhC, if you're using this article as an example to advance any argument, then it would appear to me you must be arguing that, because a new editor can make mistakes with yyyy-mm-dd format, it had to be removed on that article, and indirectly, therefore, the format has to be removed on all articles. If so, the same example equally shows a new editor can make mistakes with templates, which should, by analogy, imply an argument that templates need to be removed, too. I wouldn't find either argument persuasive. Gimmetoo (talk) 07:22, 20 June 2011 (UTC)
    I'll advance your 'Have another look' with a 'read my post again'... I noted an incorrect accessdate of 2009-30-03 (in the version I saw before I made my edit), which on the surface implied it was inserted on 30 March 2009 – don't ask me where that editor got it from. I have never visited that article before, and it didn't appear relevant to me when it was actually inserted, I took that in good faith and aligned it. To be fair, I'd say that that is the first time I've seen a date stated that way; I see plenty of slash dates (both American style and international style); the most common error among dates using dashes has to be ones stated like 11-12-2011. Like their slash counterparts, these are often impossible to fix. Yes, I was using these examples as an argument that dates with months written out in longhand are more easily parsed by humans. And by cleanly removing that incorrect date format as well as any similar date formats 'unparseable' by newbies, it significantly reduces the possibility that the editor will make another edit with the same errors. --Ohconfucius ¡digame! 07:42, 20 June 2011 (UTC)

The last RFC was unambiguously clear that the editing community did not want to deprecate or forbid the use of YYYY-MM-DD format. If you want to change this, create a new RFC on it, show that consensus changed, and THEN change the section. Headbomb {talk / contribs / physics / books} 02:06, 20 June 2011 (UTC)

RFC: restructuring of the Manual of Style

Editors may be interested in this RFC, along with the discussion of its implementation:

Should all subsidiary pages of the Manual of Style be made subpages_of WP:MOS?

It's big; and it promises huge improvements. Great if everyone can be involved. NoeticaTea? 00:18, 25 June 2011 (UTC)

Rework of sections needed

I think a rework of sections is needed. We have a section heading 'SI units' with non-SI content. We have a section called 'How to present the units' but spelling and abbreviation guidance is in other sections. We have a section called 'Units and symbols often written incorrectly' but that's a bit of an oddball heading and the content is, or should be, within other sections such as 'Specific units'. The 'Units of measurement' section contains the text 'consult other editors' but it applies to the whole MOS so I think some trimming could be done. I'm not proposing massive changes, just rework on headings and a few clauses targetted at the greater goals of succinct clarity and accessibility. What do people think? Lightmouse (talk) 08:31, 17 June 2011 (UTC)

The current sections are:
  • 4 Units of measurement
  • 4.1 Which units to use
  • 4.1.1 SI standard
  • 4.1.2 Scientific and technical units
  • 4.2 Unit conversions
  • 4.3 Avoiding ambiguities
  • 4.4 Unnecessary vagueness
  • 4.5 Unit names and symbols
  • 4.5.1 Conventions
  • Unit names
  • American spelling
  • Unit symbols
  • 4.5.2 Units and symbols often written incorrectly
  • 4.5.3 Quantities of bytes and bits
  • 4.6 Adopting suggestions from standards bodies
  • 4.7 Specific units
Here are some issues:
  • The section 'SI units' has non-SI content.
  • A section says 'consult other editors' but that advice isn't exclusive to that section
  • There is a section 'How to present the units' but spelling and abbreviation guidance is distributed in other sections.
  • Almost all of the 'Units and symbols often written incorrectly' section would be better in the 'Specific units' section.
  • The 'Which units to use' section is misnamed and mostly describes something like 'Which units come first'.
  • Almost all of the 'Avoiding ambiguities' section would be better in the 'Specific units' section.
  • Almost all of the 'Units and symbols often written incorrectly' section would be better in the 'Specific units' section.
  • The entire 'Unnecessary vagueness' section seems to add little value.
  • The content of 'Conventions' could easily be in other sections
  • In general, the sections overlap. I'm familiar with the guide but I still find it difficult to navigate and the headings don't help me.
We need to make it easier for an editor to find the information. All that's required is a top-down view of how the current information is arranged. The distributed guidance aboud specific units can be moved to the 'Specific units' section. Sections could be revised into:
  • 4 Units of measurement
  • 4.1 Which units come first
  • 4.2 How to write units
  • 4.2.1 Unit abbreviations and symbols
  • 4.2.2 Unit names
  • 4.2.3 Guidelines and style conventions
  • 4.3 Specific units
  • 4.4 Guidance on bits and bytes
Lightmouse (talk) 14:26, 25 June 2011 (UTC)
LM, this sounds like a very good idea, although I haven't examined it in detail. I also think the text of MOSNUM could be rationalised down to about 60% of its current size. Tony (talk) 14:31, 25 June 2011 (UTC)
Does anyone else have comments? Lightmouse (talk) 17:06, 27 June 2011 (UTC)
In general, I'm happy so long as the advice itself doesn't change.
By the "SI units" section I assume you mean the "SI standard" section, which is so-named because it refers to the technical standard - though I agree that this is not entirely clear. The "how to present the units" and "Scientific and technical units" sections are additional guidance on which units come first. The current split is between basic principles and detailed guidelines as to how to apply those principles, so some split is useful, though this is probably not as well done as it could be at the moment. Pfainuk talk 17:21, 27 June 2011 (UTC)


Although, the MOS recommends not using seasons instead of more precise and worldwide dates, it allows them in certain circumstances. If seasons are used, should they be in the format "summer 1960" or "the summer of 1960" or either? McLerristarr | Mclay1 13:44, 17 June 2011 (UTC)

Summer 1960 was ... January, February, or December 1960? Tony (talk) 14:30, 17 June 2011 (UTC)
The night of 21 March was ... from 19:00, 20 March (UTC) to 07:00, 21 March (UTC) or ...? (IOW, I think that the grounds to avoid using summer are no stronger than those to avoid using night. It means the summer/night of whatever place you're talking about, of course.) A. di M.plédréachtaí 16:44, 17 June 2011 (UTC)
It is usually not immediately clear when seasons are used, and it usually is clear when times are used. Tony (talk) 08:29, 18 June 2011 (UTC)
What are you talking about? Are you seriously suggesting that someone might misunderstand After the 2009 winter shutdown, the LHC was restarted (shortly after being told that the LHC is in Geneva, Switzerland) as referring to the Southern Hemisphere winter? A. di M.plédréachtaí 10:32, 18 June 2011 (UTC)
This statement is ambiguous - do you mean December 2009 or January 2009? Unless the preceding paragraph identifies which winter we are talking about, it would be better to write "After the 2009/10 winter shutdown ...". Martinvl (talk) 10:42, 18 June 2011 (UTC)
I've changed it, even though the preceding sentence is about November 2009 and the following one about March 2010 so there was no real ambiguity. A. di M.plédréachtaí 12:13, 18 June 2011 (UTC)
  • By the same token, let's stop all US conversions: metrics only, since they're totally unambiguous. No, I object to talking in terms of winter shutdowns unless "northern-hemisphere winter shutdown" is specified. Tony (talk) 13:02, 18 June 2011 (UTC)
    And I object to talking about 5 p.m. broadcasts unless "5 p.m. ETZ" is specified. What's the difference? A. di M.plédréachtaí 17:38, 18 June 2011 (UTC)
    I don't see how context can fail to be informative, here. For articles about subjects that are clearly related to Europe, the US, Canada, Japan, Russia, etc... it's obvious that season names refer to northern hemisphere seasons. Likewise for articles related to Australia, South Africa, Brazil, etc... Of course, seasons should be avoided for articles without a geographically related component. There's a nuance here that is lost with the simplistic "don't use seasons" stricture. The only effect that being overly prescriptive on issues like this is going to have is that people will simply ignore the MoS.
    — V = IR (Talk • Contribs) 18:00, 18 June 2011 (UTC)
In a common reference style like "Journal of Astronomy, Summer 2005 issue, page 46", a hemisphere speech would be ridiculous in that context. I don't think the guideline's advice against capitalizing seasons applies either, even though that is perhaps the most common reason to mention a season on Wikipedia. Art LaPella (talk) 00:59, 19 June 2011 (UTC)
OK, forget summer. If it was autumn or spring, then there is no ambiguity about which part of the year is being refereed to. In that case, should it be "spring 1960" or "the spring of 1960"? "Mr X bought a cat in spring 1960/the spring of 1960. McLerristarr | Mclay1 05:04, 19 June 2011 (UTC)
For the record, all the seasons are reversed in the Southern Hemisphere. We have summer in January and winter in July, autumn in April and spring in October. As well as this, in many parts of the tropics they have a wet season and a dry season. So the policy of avoiding using the seasons as a marker for a particular time of year is quite appropriate. Michael Glass (talk) 13:21, 19 June 2011 (UTC)
For the record, all the times of the day depend on the time zone. We have noon at 14:00 (UTC), evening at 21:00 (UTC), morning at 13:00 (UTC) and night at 02:00 (UTC). As well as this, in Arctic regions they have the midnight sun and the polar night. A. di M.plédréachtaí 14:40, 19 June 2011 (UTC)
I was born in autumn but my birthday's in spring – I realise that the seasons are reversed with the two hemispheres. What I meant was that with summer or winter, winter 1960 could refer to January/February 1960 or December 1960, whereas with spring or autumn, the whole season is continuous within a single year. McLerristarr | Mclay1 10:34, 25 June 2011 (UTC)
  • For the record, it's really irritating when the WMF talks of its "spring" meeting. When TF is that? Over my dead body are seasons going to be used when a month can be used in a WP article. Tony (talk) 15:27, 19 June 2011 (UTC)
Which is why the guideline says "usually". And I think even that is too strong, considering how often a season is used as a name for a journal issue in a reference. Art LaPella (talk) 19:44, 19 June 2011 (UTC)
What about an event that sources say took place in a US city in the summer of 1973? How do we express the date? -- SWTPC6800 (talk) 20:12, 19 June 2011 (UTC)
If there were riots in Cheyenne in 2005 and they were known as the "Cheyenne Summer riots of 2005", so be it, I have no problem with using the term "Summer of 2005" in that context - it is pretty obvious that they refer the the US, or rather the Cheyenne frame of reference. However, it would be incorrect to refer to a war-time offensive in the Pacific in June/July 1944 as the "Pacific 1944 summer offensive" as the campaign possibly straddled the equator.
BTW, lest anybody for Cheyenne takes offence at this example, there were no riots in Cheyenne - I chose that example because my daughter had a holiday job there that summer. Also, I am guessing about the Pacific offensive, but for purposes of an example, it will do.
In short, no hard and fast rules, just common sense but put yourself into the shoes of a reader from another part of the world when applying common sense. For the record I was brought up in the South Africa and while studying in the UK, I confused a friend by discussing the South African Christmas holidays and in the next breath mentioning the South African summer holidays. The context suggested that I was talking about the same holidays. I was. Martinvl (talk) 21:05, 19 June 2011 (UTC)

"Of course, seasons should be avoided for articles without a geographically related component." That simple statement seems to cover it nicely. However, experience shows that some northern hemisphere authors often fall into the northern-season identifier trap. Writing about global events for a global readership, Wikipedians can do better, with care. "Pij" (talk) 06:49, 2 July 2011 (UTC)

Fraction hyphenation

English compound#Hyphenated compound adjectives says "Fractions as modifiers are hyphenated ... [but] Fractions used as nouns have no hyphens". MOS:NUM#Typography says fractions are hyphenated, without mentioning modifiers or nouns. Is this an unlisted exception, or a contradiction that should be harmonized? Art LaPella (talk) 21:23, 28 June 2011 (UTC)

Looks like a contradiction to me. I can't see any reason to hyphenate fractions as nouns; you can have a dog and you can have two dogs, you can have an apple and you can have two apples, you can have a third and you can have two thirds; there's nothing fancy going on there. Similarly if it's part of an adjective use hyphens as usual. PS ... "five-eighths inches" ... what exactly does this mean ... five eighths of an inch? JIMp talk·cont 23:53, 28 June 2011 (UTC)
I've seen two styles: one tenth of the length of Earth's equator, where tenth is serving as a completely regular noun, and one-tenth the length of Earth's equator (no of), where one-tenth is an external modifier (in CGEL's terminology) akin to e.g. twice or half. I prefer the former, but suspect there might be an ENGVAR difference. A. di M.plédréachtaí 09:46, 1 July 2011 (UTC)
Well...[15][16] The ENGVAR difference between those two forms is the one I expected, but, surprisingly to me, in both AmE and BrE one-tenth of the is more common than either. A. di M.plédréachtaí 09:54, 1 July 2011 (UTC)
OK, but the question that matters to the rest of Wikipedia is: should the contradiction be harmonized? How? Art LaPella (talk) 20:14, 1 July 2011 (UTC)
The article is about English in general and the MOS is about Wikipedia, but still, the article is too categorical in saying “Fractions used as nouns have no hyphens”, and I wouldn't mind the MOS allowing to omit the hyphen in fractions used as nouns. A. di M.plédréachtaí 22:36, 1 July 2011 (UTC)

Linking year ranges

Here's an interesting thing. I was just looking up Asashōryū and read that "From 2004 until 2007, Asashōryū was sumo's sole yokozuna". The beginning and end years were linked to the respective YYYY in sumo pages but aren't we looking at a range not just at it's beginning and end. Link 2004 in sumo and 2007 in sumo but what about 2005 and 2006? They're just as much a part of the range. JIMp talk·cont 16:32, 30 June 2011 (UTC)

I can't think of a good way to phrase this; add them to the See Also if you can't either. Septentrionalis PMAnderson 23:55, 1 July 2011 (UTC)

American spelling

"American spellings of unit names (e.g. meter or liter) should be used on pages written in American English." How does one know that an article is written in American English? Does the first instance of American spelling make the article American English, or could that instance be an inconsistency? Should there also be a statement, "British/international spellings of unit names (e.g. metre or litre) should be used on pages written in British/Australian (etc) English"?

Please be aware that the [Manual of Style (dates and numbers)] itself uses a mix of litre and litre, meter and metre.

When an article is encountered which uses such an inconsistent mix, should it be fixed? If so, who should decide whether it should be fixed to American English, or to British English? "Pij" (talk) 02:35, 2 July 2011 (UTC)

If an article's subject doesn't relate to a specific nation, we usually go by the English variety used in the first consistent non-stub revision to contain English variety-specific spellings/terminology.
Yes, inconsistencies (apart from proper nouns, direct quotations of written material, text pertaining to alternative spellings, etc.) should be fixed in accordance with the above. If every non-stub revision up to that point contains inconsistent spellings/terminology, simply select your preferred English variety (at which point you'll create the first consistent non-stub revision to contain English variety-specific spellings/terminology). —David Levy 03:02, 2 July 2011 (UTC)
Generally, if a topic has a strong tie to a nationality, that variant should be used. i.e. Queen Elizabeth II would use British, Jarome Iginla Canadian, Jon Stewart American, etc. So that would be your first clue. If a topic has no strong national tie, then the format the article was created in is often the default. Generally though, when I see a mixture of spellings where there is no strong national tie, I'll match the most prevalent dialect. Not just for meter/metre, but other differences: honor/honour, defense/defence, etc. Resolute 03:04, 2 July 2011 (UTC)

Thanks for those replies. I have since found the relevant references:

"Pij" (talk) 03:07, 2 July 2011 (UTC)

So I see that the original [Manual of Style (dates and numbers)] 16 Feb 2003 used "meter", but the current version has mostly non-American spelling. I'm tempted to edit the page into British English, but I'm not sure I'm brave enough. "Pij" (talk) 03:39, 2 July 2011 (UTC)

Since the international standard is to use non-American spellings, surely this MoS should use the international variations, purely for the sake of conforming and having a reason to choose a particular style other than "that's what the other guy used". McLerristarr | Mclay1 07:01, 2 July 2011 (UTC)
Searching for the most obvious differences, I can find favours, catalogued, capitalised, unitalicised, but no obvious American-only spelling except in “Geographical coordinates” where meters is used. A. di M.plédréachtaí 10:41, 2 July 2011 (UTC)
Interesting. We have two questions here:
  1. The topic of this page is units. Do we want unit spelling to be consistent, if so do we use 're' or 'er'?
  2. Do we want non-unit terms to be consistent, if so which spelling should be used?
I care more about the former than the latter. Lightmouse (talk) 10:55, 2 July 2011 (UTC)
Yes, they should be consistent. I'd choose the British spelling so that there'd be less work to do, but I don't strongly care either way. (This applies to text in the voice of the MOS; examples could use both styles.) A. di M.plédréachtaí 12:05, 2 July 2011 (UTC)
"That's what the other guy used" is pretty much one of the stated guiding principles of the MOS. Consistency is another. Upon re-reading, I agree that there are few examples of American variation, apart from examples of "meter" and "liter" as noted. So, yes, I agree that consistently British choices would seem the best answer. On the other hand, while the topic is as much about variation as it is about consistency, perhaps it is a place for tolerance of both major forms to be exercised. I would imagine some would object to placing this template
at the top of this page. (Of course, I'm happy to remove it from this discussion if it offends.) "Pij" (talk) 18:49, 2 July 2011 (UTC)

I've changed several instances of meter and liter to metre and litre. It will be interesting to see if they stay changed. "Pij" (talk) 19:50, 2 July 2011 (UTC)

Also found, and for the sake of consistency altered to British spelling: capitalized, categorized, recognize, categorizes,standardize "Pij" (talk) 22:57, 2 July 2011 (UTC)

Please stop. It is against WP:ENGVAR to remove Anglo-American spelling differences. Septentrionalis PMAnderson 23:22, 2 July 2011 (UTC)
Was that tongue-in-cheek humour? I wouldn't have bothered changing to a single consistent set of spelling, as it doesn't bother me either way, but for the fact that WP:ENGVAR tells me I should. In particular, Consistency within articles and Retaining the existing variety in combination with the above discussion (which identified that the article was not predominantly written in American English). I'd be interested to see where it is forbidden to make the spelling style consistent. "Pij" (talk) 23:41, 2 July 2011 (UTC)
This page is written using American spelling in the main text. Examples can be in any major variety. Tony (talk) 01:48, 3 July 2011 (UTC)
Really? Looking at it before doing any editing, it was clear that it was a mish-mash of British and American, but the balance seemed to tip toward the British. I made no changes before discussing here, and the discussion indicated that British English was the best choice. How do you conclude that the text was American? Would you rather it was reverted back to mish-mash? And, no, I am not referring to examples within the text. Those were clearly in particular styles for the points they were illustrating. You could easily revert the changes back to mish-mash language, but I suppose then you would almost have an obligation to fix the article into one consistent style. I think I can see why nobody bothered to do this before me: No matter which flavo(u)r of consistency one chooses, others will object. So much for the It seems people would prefer contention over consistency. "Pij" (talk) 04:16, 3 July 2011 (UTC)
It seems more like YOU prefer contention, since you kind of sprung this after discussing articles. I agree it should be fixed, but first discuss which way to go, since you started a discussion. It looks to me like there was one "ised" and a bunch of "ized", and you fixed all that toward British due to a possible imbalance between the "meter" and "metre" spellings; but you didn't make it consistent, even in the non-example text. So there's still an opening to discuss which national variety to use before claiming that you've make the first consistent version. Dicklyon (talk) 05:08, 3 July 2011 (UTC)
I went with the way the discussion indicated. The discussion was broader than just meter/metre. The "ized" endings don't bother me in any way, as where I live they are commonly used. I agree I didn't achieve complete consistency, but I worked towards it. That's the objective, right? It's good to see some participation on the discussion, anyway. It would be great if anyone who was inclined to revert changes, or make other changes, would join the talk. That way we might get close to an edit that was created by consensus. "Pij" (talk) 05:28, 3 July 2011 (UTC)
Well, I can see you changed a bunch of "ize" and a "liter" (and left a few "meter" in one paragraph), but I can't see what Britishisms you thought were there already (I don't know how to spot them); maybe you can mention some that outnumber the AmEs? Then we can see how long ago they came in... Dicklyon (talk) 06:51, 3 July 2011 (UTC)
I listed them above (the ones I was able to find, at least): favours, catalogued, capitalised, unitalicised. (Spelled and -ize are used both in AmE and in BrE – though in the latter spelt and -ise also exist, so I didn't count them either as Br'isms or as Am'isms.) A. di M.plédréachtaí 14:53, 3 July 2011 (UTC)
Back in 2009 there was favor and not catalogue. And there were a lot more "ize" compared to the few "ise", even yesterday. If you don't count "ize" as American, you also can't count "metre" as British, since it's a common alternative in AmE. I still don't see it. Dicklyon (talk) 16:49, 3 July 2011 (UTC)
It doesn't look like that to me.[17] A. di M.plédréachtaí 18:36, 3 July 2011 (UTC)
Yes, meter is now almost 10X more common in American than metre. But the best analysis I've seen says, "The present situation is that the spellings of metre and meter are both regarded as correct in the USA (as are litre and liter) and it is quite legal in the USA to use either metre or meter." A lot of us use metre to distinguish the length unit from the measuring device. Dicklyon (talk) 19:41, 3 July 2011 (UTC)
Well, -ise is only 1.5 times as common as -ize in British, so the grounds to call -ize an Americanism are even weaker than those to call metre a Britishism. A. di M.plédréachtaí 20:28, 3 July 2011 (UTC)
OK, let's leave "ize" and "metre", then, and most of the text will be perfectly correct in both versions of English. If you want the rest to be British, even though it had "favor" before "favour", that's OK by me. Dicklyon (talk) 20:47, 3 July 2011 (UTC)
That sounds like a good solution. Keep the "ize" spellings, standardise to "metre/litre", and make a decision about the "...our" words. Can we gain a consensus on Dicklyon's suggestion? "Pij" (talk) 03:21, 4 July 2011 (UTC)
Also, the lead treats Committee as a plural (last sentence starts with They), while IIRC it's treated as singular in AmE. A. di M.plédréachtaí 20:28, 3 July 2011 (UTC)
  • Under the ENGVAR rules, we should go back in the history page to find out what it was originally, and how it's been since. Tony (talk) 06:04, 3 July 2011 (UTC)
    Of course, Tony. is one of the guiding lights, as discussed earlier. "Pij" (talk) 06:14, 3 July 2011 (UTC)
    Only if it has not “evolved sufficiently for it to be clear which variety of English it employs”. In this case, excluding the examples and the “Geographical coordinates” section, I could find no Americanism at all. (My search was very quick so it could have missed some, but it did find several Britishisms.) A. di M.plédréachtaí 15:01, 3 July 2011 (UTC)
    There are several cases of "meter" that are not in examples; this is pretty American. There used to be "favor"; and the many "ize" suggest non-British even if they're not uniquely American (otherwise, why were they changed?). Dicklyon (talk) 17:01, 3 July 2011 (UTC)
    There are exactly three cases of meter not in examples, all of which in the same paragraph (in the "Geographical coordinates" section), compared to thirteen of metre all over the page. A. di M.plédréachtaí 18:36, 3 July 2011 (UTC)
    Yes, I saw that. Metre has been dominant there for many years; I think because it's a spelling that's perfectly normal and acceptable in all English variants. Dicklyon (talk) 19:43, 3 July 2011 (UTC)
    Please stop making things up. Metre is like honour, acceptable in American English, especially for special purposes (wedding invitations, for example), but not normal.Septentrionalis PMAnderson 04:13, 4 July 2011 (UTC)
    When I see "metre" in a non-obvious U.S. context, I assume it's reflecting something about rhythm, since (I think) both "metre" and "meter" are about equally used for a musical beat or poetic form, while Americans almost always write 39.37 inches as "meter", just as in a non-obvious British context, one assumes "meter" has to refer to a measuring device ("Lovely Rita, Meter Maid"), rather than 100 cm. U.S.: Our band teacher used a meter stick to mark the metre. Br: The parking meters are spaced about five metres apart. ¶ So meter/metre is different from seeing the non-ambiguous "centre" or "glamour" as an optional U.S. (and frequent Canadian) spelling. —— Shakescene (talk) 18:32, 4 July 2011 (UTC)
I'm wondering if the page might actually have been fairly consistent as Canadian English. I'm not familiar enough with the details of Canadian to really notice something being written in that form. I suppose, like many others, that I tend to "see" either US American or British spelling. In any case, it seems to me that greater use of during the early stages of development of Wikipedia pages could help prevent slippage, over time, into other written dialects. "Pij" (talk) 03:21, 4 July 2011 (UTC)
My link didn't show up. I meant greater use of the Varieties of English Templates "Pij" (talk) 06:37, 4 July 2011 (UTC)
If the pages appears Canadian, I don't think that's deliberate. It's just a mix of American and British. I don't think it's important really whether a guideline is in one form or another. Although, I think we should at least use the international versions of the unit names. McLerristarr | Mclay1 11:44, 6 July 2011 (UTC)

Excessive/unnecessary unit conversions

The Unit Conversions section says "Generally, conversions to and from metric units ... should be provided" (with 2 exceptions). Some bots (or bot-like editors) have interpreted this to mean that every instance of a unit should have a conversion - even if the article already contains the conversion many times and the unit appears in a proper noun (e.g. see Montana class battleship#Secondary battery which has "5-inch (127 mm) ... gun" 7 times in 3 short paragraphs). Some might say this already falls into the "make a common or linked expression awkward" category, but that doesn't seem strong enough to curb the activities of the bots. Is it worth adding to MOSNUM that it's generally unnecessary to have the same conversion more than once in a section and where the unit appears as part of a linked name (e.g. Three Mile River, 16"/50 caliber Mark 7 gun, Quarter Pounder ...) the conversion is not generally appropriate at all ? DexDor (talk) 21:48, 4 July 2011 (UTC)

I had complained to the LightBot guy, too, about converting BTU as an air-conditioner rating to kJ, since it really means BTU per hour. I'm not convinced that changing it to BTU/h and converting to kW (which he did, instead of replying) respects the conventions in this field. Dicklyon (talk) 22:18, 4 July 2011 (UTC)
If it really means BTU per hour, write BTU per hour. Should we blindly follow confusing conventions? We are writing for a general audience who might not understand the conventions. Writing "cc" for cubic centimetres of engine displacement is one thing but using specalised jargon is something else. Of course, this is actually a different issue to the one originally raised.
I'm all for conversions (to and from metric) but agree that we generally don't need the same conversion repeated in the one section (& often not even in the same article at all). I wonder whether the bot could look out for this. JIMp talk·cont 23:24, 4 July 2011 (UTC)
This is an encyclopedia. We should not assume our readers are aware of the conventions in various fields to use unphysical units (BTU as a measure of power, yard as a measure of volume, etc.). I believe it is appropriate to use physically correct units and jargon be dammed. Jc3s5h (talk) 23:25, 4 July 2011 (UTC)
I don't object to physically-correct units, but if a thing is "100,000 BTU rated" we should say so, and then say what it means. The guy running the bot didn't know what it meant, and adding /h probably didn't make it much clearer, since it doesn't say if that's input power or output power or something else. Dicklyon (talk) 23:28, 4 July 2011 (UTC)
That sounds fair enough. I removed the "/h"s from the text and put a explanation and conversion in the footnotes. JIMp talk·cont 00:42, 5 July 2011 (UTC)
I made a further clarification at Solar air conditioning, since as far as I can tell, the BTU rating is about a heat removal rate at some standard conditions, not a power output per se. Dicklyon (talk) 05:22, 5 July 2011 (UTC)
  • Montana class battleship: I added *one* conversion to a caption. It seems to me that body text and captions are different.
  • Solar air conditioning: I didn't change it to BTU/h instead of replying. I was informed of the issue only 2.5 days ago and since then several editors have been working on the issue. I even asked if we could convert BTU to watts.

I'm glad that DexDor and Dicklyon have brought these issues here. They're generic issues, not specific to one or two edits. Lightmouse (talk) 08:01, 5 July 2011 (UTC)

Okay, I'm owning up to the change to BTU/h. JIMp talk·cont 08:31, 5 July 2011 (UTC)

Sorry, I misread who did what. The "/h" made the units formally correct, but was a very unconventional way to put it, and converting to kW and calling it "power output" compounded the problem. One needs to look into this more... Dicklyon (talk) 15:52, 5 July 2011 (UTC)

Depreciated or deprecated?

There is a category called 'Category:Depreciated conversion templates'. Should that be 'Category:Deprecated conversion templates'? If so, can somebody change it? Lightmouse (talk) 10:27, 6 July 2011 (UTC)

You're right. JIMp talk·cont 12:34, 6 July 2011 (UTC)
Fixed. JIMp talk·cont 12:40, 6 July 2011 (UTC)

Year ranges

At the main WP:MOS page, in the summary of what is here, I read the following:

"A closing CE–AD year is normally written with two digits (1881–86) unless it is in a different century from that of the opening year (1881–1986). The full closing year is acceptable, but abbreviating it to a single digit (1881–6) or three digits (1881–886) is not. A closing BCE–BC year is given in full (2590–2550 BCE)."

I have two questions: (1) What is the rationale for treating closing BCE-BC years differently from CE-AD years? In my experience, not writing the closing CE-AD year in full can give rise to confusion and I now tend to nearly always write it out in full, even if in the same century. (2) Are there earlier discussions about this that consider the case where someone is searching within a text for a year and fails to find (in the above example) the year "1886" because it is written as "86"? This, in my view, is a strong argument for always writing out in full at least once anything that is abbreviated within any single article (in this case, 1886 has been abbreviated as 86). Carcharoth (talk) 00:56, 30 June 2011 (UTC)

I didn't write the rule, but to me "the rule of King XXXX, 1881–86" is less confusing than the seemingly backwards "the rule of Pharaoh XXXX, 1886–81 BC". 1886–1881 BC also seems backwards until you think about it, but at least you know that the century isn't the problem. Art LaPella (talk) 02:02, 30 June 2011 (UTC)
Thanks. That explains the 'write BC in full' rule. Any views on the point that writing 1886 as 86 is a form of abbreviation and can have unforeseen downsides (in terms of search engines, or any searching function, looking for a year in an article), so to avoid this a year should, at a minimum, be written out in full the first time it is used? Carcharoth (talk) 02:20, 30 June 2011 (UTC)
The two-digit rule for closing ranges is common and produces easier-to-read years ranges, which can often be clunky. It is paricularly apposite to tables and infoboxes, where space is short. Tony (talk) 03:44, 30 June 2011 (UTC)
Uh huh. And the other part of what I said that you failed to address? I just had a thought that an "abbreviated year" template could be used that might ensure that the full year shows up for search engines or anything that searches the wiki-text, but anything searching the plain text will fail to 'see' what the abbreviation represents. I'm positive that the Manual of Style explicitly states that all abbreviations must be written out in full the first time they are used in an article - why would this not apply to abbreviated years? Carcharoth (talk) 03:52, 30 June 2011 (UTC)
OK, I'll try again. Yes, MOS:ABBR#Acronyms and initialisms states that acronyms must be written out the first time, but compliance is rare. "unforeseen downsides" are hard to imagine because writing a year as 86 is a predictable feature in normal text; a system that can't handle an article with "86" would be like a system that can't handle the letter "Q". "search engines" sounds like this, so why would you want to look through 87,900 results? "looking for a year in an article" is a better idea, but looking for "86" instead is unlikely to get you more than a couple false positives for something like "page 86" and "ISBN 123456-86". I can't remember ever making such a search. So I don't know how you would justify the instruction creep and the extra article length. Art LaPella (talk) 04:00, 30 June 2011 (UTC)
How often do people search for a year? It must yield hundreds of thousands of results. I wonder what the utility is. And searchers even now search "1891" and "–91", and probably "-91", since some are still some done with hyphens. Generally I'm a little nervous about introducing even more templates into edit-box text; it makes WP editing more exclusive, even if some cluey visitors can sort of work it out. Tony (talk) 04:04, 30 June 2011 (UTC)
Oh, I search for years all the time (through Google and on Wikipedia), but obviously with other search terms associated with the year to narrow down the results (I have little patience for people who assume that I am talking about single-term searches). Obviously I can search for the abbreviated year as well, but it strikes me as leaving information out of articles. If the actual data is "1886", why would we insist on writing it as "86"? If you were giving a page rage, would you abbreviate the page numbers? My point here is that the data within the text of an article should be as accessible as possible to both readers and machines, and presumptions shouldn't be made about how searches are done and whether or not other people do them, merely because you happen not to have had reason to do such a search. I'm currently trying to find some online guidance about this (with arguments for both formats). Are there any useful links available? Carcharoth (talk) 04:28, 30 June 2011 (UTC)
You mean like 1914–18 here? As for abbreviating page numbers, there are lots of them at William Shakespeare#References for instance. Art LaPella (talk) 05:00, 30 June 2011 (UTC)
Fair enough. I'm going to concede the argument at this point. Though I do note that the main MOS page says that using a "full closing year is acceptable", so clearly this is something that can be done on a case-by-case basis if needed. From what I've seen, there are times when using a full closing year is acceptable, so hopefully no-one goes around trying to force consistency on this. To return to the example Tony gave of lists and tables - if you have a list of year ranges and some are XXXX-YY and others are XXXX-YYYY, then that can look odd in a table, so I'd tend to do all one or all the other, and not mix them up. I suspect (and this is anecdotal only) that you will also increasingly find biographical year ranges (i.e. the birth and death years for people) given in full when used formally at the start of an article or after a person name (when providing context). Also, if you look through a biographical dictionary, such as the ODNB, you will see that they give the biographical years ranges in full when referring to someone, even if those years are in the same century. There must be a reason for that (OK, maybe I'm not totally conceding the argument!). Carcharoth (talk) 05:18, 30 June 2011 (UTC)
Interesting point. Probably consistency: a substantial number of lives have to use the full form: John Smith (1896-1980) can't be abbreviated; so if there is a standard format, it has to be XXXX-YYYY. But we routinely include day and month in the first line, at which point the whole question becomes moot. Septentrionalis PMAnderson 22:30, 2 July 2011 (UTC)
I agree that it is likely due to a standard format. The question is not entirely moot, though, as there are cases where it is useful to give context to the mention of a person in an article by providing the birth and death year in brackets (though sometimes the phrase "Xth-century occupation" is enough to give context). The place where it is most common to give birth and death year in brackets in the main text of an article is when mentioning other family members in a biography, where those family members don't have an article. In those cases, I would give the years in full regardless. If just describing a period in someone's career, I would usually shorten the second year of the range, though I might also rephrase things in such a way as to avoid the issue altogether. Overall, shortening the year only seems necessary when space is at a premium, and it rarely is around here. My conclusion is that shortening the final year is OK, but not strictly necessary. Carcharoth (talk) 05:47, 5 July 2011 (UTC)
Yes, MoS does say "full closing year is acceptable", and I don't know of anyone who opposes that clause. I checked several disambiguation pages like John Smith, and all of them use complete years for almost all of the names. Art LaPella (talk) 06:20, 30 June 2011 (UTC)
I much prefer XXXX–YY. This is normal writing style. I think the best solution is to adjust the way you search. JIMp talk·cont 16:32, 30 June 2011 (UTC)
Funny that. I much prefer XXXX–YYYY, as this feels like proper writing style to me. I suppose it is similar to the way some people prefer to write "Street" instead of "St." and "Saint" instead of "St." I guess taste and style varies (among both individuals and publications). Carcharoth (talk) 17:49, 30 June 2011 (UTC)
I agree. The abbreviated form seems informal, lazy, or old-school, from when saving a bit of space on paper was worth something. The potential ambiguity introduced, esp. by improper generalizations to other numbers and date ranges, isn't worth the savings. And the full version seems to be what's most often used in books (and in WP). Dicklyon (talk) 06:07, 5 July 2011 (UTC)
It was not just saving space that was a motivation, but also repetition. Back in the days of handwriting (anyone remember that?) it was excruciating to repeat stuff such as the "000" in thousands and millions, the first two digits in year lists, and so on. The repeat symbol (dunno what it is called) was used in lists to avoid this, and other symbols and abbreviations developed (including those for units of measures). These all transferred over to print, though Tony is right to say that many forms of abbreviation aid reading comprehension, so there is a balance needed. Nowadays, lists can be generated automatically and you can avoid the need to type things out over and over again. The primary consideration should be the balance between readability and making sure all the content is there at the first use (i.e. not excessively abbreviated so that machines can't read it or process it). Carcharoth (talk) 06:16, 5 July 2011 (UTC)
  • Carcharoth, I bet I could sit you in front of a monitor and put you through a bank of fast flash tests to show you could more quickly discern the year range without the redundant numbers. I've no scientific evidence (yet), and would like to demonstrate it. I do concede that being primed for it helps. Tony (talk) 17:55, 30 June 2011 (UTC)
    • For short year ranges (less than 10 years), I have no doubt you are right. For ranges spanning several decades (as is typical for a human lifetime), I think you may find that the full years work better. Compare John Smith (1905-96) to John Smith (1905-1996). The question no-one has answered to my satisfaction is why biographical dictionaries ues the XXXX-YYYY format. The Australian Dictionary of National Biography does it as well (I used the ODNB as an example above). It seems that for biographical lifetime ranges, full years are not uncommon, even if in the same century. Carcharoth (talk) 18:32, 30 June 2011 (UTC)

Back to the original question, 1881–86 is unambiguous (as for whether it's clearer, it's in the eye of the beholder... I like it for years but not for page numbers, but what's wrong with allowing editorial discretion?), but the end of 2590–50 BC could be either in the twenty-sixth or in the first century BC. Only the former interpretation is likely if it's clear the range is contained within an individual's lifespan, but imagine we're talking about a dynasty that reigned in 190–80 BC. A. di M.plédréachtaí 10:00, 1 July 2011 (UTC)

Ah, thanks. I hadn't thought of that. It arises because in BC you are giving a range from a larger to a smaller number, whereas for AD the ranges are smaller to larger numbers. The only other system I can think of where you go from larger to smaller number is again a "before" dating system, the "years before present" system used in geological timescales, but those are approximate, not precise. Saying that a dinosaur species lived from 180-65 million years ago is not likely to be mistaken for 180-165 million years ago, or is it? I suspect that all "mya" (million years ago) and similar figures are written out in full anyway (not sure about 'thousands of years ago' figures). Carcharoth (talk) 22:57, 1 July 2011 (UTC)
There is less reason to abbreviate if there are only three significant digits; similarly I would expect 431–404 BC and 117–138 AD are normally written in full. Septentrionalis PMAnderson 04:28, 4 July 2011 (UTC)
Why? 04:30, 4 July 2011 (UTC)
because 117–38 is only saving one character (and probably one syllable). 1914–18 saves two. Septentrionalis PMAnderson 04:37, 4 July 2011 (UTC)
Which reminds me, I must do a (personal) survey at some point to see how common "1914-18" and "1914-1918" (and the same for WWII) are when used in books as compared to formal documentation and on war memorials. I suspect (though this is only speculation) that formal usage (titles, inscriptions and so forth) has the years in full, while the shortening of 1918 is only done in passing mention in text in publications. For now, looking at book titles and using a Google Books search of the titles, we have about 169,000 hits for the phrase "1914-1918" in book titles, and about 17,300 hits for titles with the phrase "1914-18". For 1939-1945 and 1939-45 the respective hits are 226,000 and 31,900. Not sure if this tells us anything, though (other than the fact that there may be more books on WWII than WWI). Maybe someone could look at common usage for other year ranges for other wars and other periods? Carcharoth (talk) 05:32, 5 July 2011 (UTC)
Yes, for some reason it looks weird to me with three-digit years. Probably it's because I'm not as used to see it as I am with four-digit years. A. di M.plédréachtaí 11:40, 4 July 2011 (UTC)
Yet pp. 186–91 is typical house style for publishers. Tony (talk) 05:52, 5 July 2011 (UTC)
Page ranges are likely treated differently to year ranges. Carcharoth (talk) 06:05, 5 July 2011 (UTC)
I think abbreviating page ranges this way is even more dangerous, in terms of potential for misunderstanding, and most publishers don't do it any more (at least not in their online article listings). You sometimes see things like p. 12-3, and someone comes along and changes it to pp. 12–3 or pp. 12–13, when what was intended was the single third page of chapter 12 or something like that. If we try to enter things right, we won't run into mis-corrections so often, hopefully. Dicklyon (talk) 06:12, 5 July 2011 (UTC)
  • Could we use dashes in examples, so they don't look like phone numbers or computer coding? 12–3 is absolutely unacceptable, as the MoS says. It must be 12–13, minimum of two. And pp. 101–03, of course. I don't see how this can be ambiguous. Tony (talk) 07:25, 5 July 2011 (UTC)
    • "12–3 [meaning pages 12–13] is absolutely unacceptable, as the MoS says." I don't think it says that, unless of course you can show us where it does. Art LaPella (talk) 03:15, 6 July 2011 (UTC)
      • The logical place would be Wikipedia:Manual of Style (footnotes), but I can't see anything there. Maybe raise it on the talk page over there? Carcharoth (talk) 23:18, 6 July 2011 (UTC)
        • Art, that's weird. It always used to be there. Here it is end of 2009: "The full closing year is acceptable, but abbreviating it to a single digit (1881–6) or three digits (1881–886) is not." End of 2010 it had disappeared. Where was the consensus for this? Tony (talk) 15:11, 7 July 2011 (UTC)
          • I thought we were talking about page numbers. That's why I said "[meaning pages 12–13]". The year number guideline has been rephrased at WP:YEAR: "A closing CE or AD year is normally written with two digits (1881–86) unless it is in a different century ..." Where is the consensus for that rephrasing? I don't remember any because it was considered a rewording, not a guideline change. Art LaPella (talk) 22:24, 7 July 2011 (UTC)

Well, I didn't do either; but it is nice to see the "absolutely unacceptable" fiats being swept away. Even Oxford began with "use as few figures as possible"; they did have one small exception, but using actually readable 1s should make it irrelevant.

Was this ever consensus? Is it widely recommended? Is it usage? And, if the answer to all of these is No, why should there be consensus for it now?

Instead, let's add a note that 12–13 and similar forms may be useful for clarity. That's guidance. Septentrionalis PMAnderson 18:00, 7 July 2011 (UTC)