Wikipedia talk:Manual of Style/Dates and numbers/Archive 152

Is there a policy that supports conversions of full articles like this [4] Eldizzino (talk) 18:44, 10 July 2015 (UTC)

You may want to read

WP:DATERET, which supports that dates should be retained in articles. If you took the time, you would have noticed that another editor had changed the dates that very morning. JOJ ^Hutton

19:33, 10 July 2015 (UTC)

You are stating the untrue [5] Eldizzino (talk) 20:06, 10 July 2015 (UTC)

DATERET allows for users to normalize dates to the set format for the article; in the example there, the bulk of the article is using DMY dates (particularly in references) so conversion of a handful of MDY to DMY is fine. Particularly as one should not use MDY in prose and DMY in refs, or vice versa. --MASEM (t) 20:22, 10 July 2015 (UTC)

Well it was in response to the same user making This edit on another article that was made the same morning. And in both cases

WP:DATERET still applies because in both articles, MDY was the first date format introduced. Furthermore, in the case of Lego Batman 2: DC Super Heroes, it was determined, through consensus at The Beatles: Rock Band, that the subject determines the date format, not the publisher and the date MDY date format is consistent with the original Lego Batman: The Videogame. Now we can't have it both ways now. Either its the subject or the publisher that determines the date formats, but we can't pick and choose so that they are all DMY.--JOJ ^Hutton

20:24, 10 July 2015 (UTC)

Take it to to the article's talk page guys. —sroc 💬 12:53, 11 July 2015 (UTC)

And avoid fixation on "first major contributor". It is not required to stick with the FMC's choice; consensus can overturn that anywhere. A long process of gradual change moving an article away from the FMC's choice over time, without incident, until way too late in the game someone (who isn't even the FMC) wants to vent about it after the fact, is a consensus that's already changed, especially if there are other reasons to prefer the evolved format, almost always because it agrees with the ENGVAR in use. Basically, a DATERET dispute really should never arise and be taken seriously unless the article is in Canadian English. If it's US, use US format, if it's a US-military article, use intl. format. If it's anything else in the world, use intl. format. (Unless there's some weird case I missed. Does the military of Botswana use ISO format, and do we

GAF? Heh.)  — SMcCandlish ☺ ☏ ¢

 ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  16:16, 20 July 2015 (UTC)

The article involved (Lego Batman 2: DC Super Heroes) was created 30 November 2011 by User:WikiEditor44. However, the initial edit did not establish DMY or MDY usage, subsequent edits were mixed, there is no notably major contributor, and there is nothing on the talk page about which format to use.

A MDY date was introduced into the text on 5 April 2012 in this edit this edit by WikiEditor44. Subsequent edits added dates in both formats, without comment.

On 7 June 2012 this edit by User:X201, with the summary "date formats per WP:MOSNUM by script as per developer nationality", changed the single date in the text and various accessdates, etc., to DMY, and added the "use dmy dates" template.

On 6 Dec. 2012 this edit by

WP:MOSNUM

by script", made recently added dates consistently DMY.

At 11:00 9 July 2015 X201 similarly revised three inconsistent dates with this edit, citing as authority "date formats per MOS:DATEFORMAT by script". At 14:40

WP:DATERET

"

So how should these guidelines be applied? I can I can see various points here that lead to different ends. As overall there has been little strife in this article (very few non-vandalism reverts, no apparent edit-warring, no arguments on the talk page), any strife that should develop here will reflect on these guidelines and the editors here. ~ J. Johnson (JJ) (talk) 00:08, 12 July 2015 (UTC)

Date normalization should be done, and in the case of Batman, it was clear at the time of the latest edit that the majority format was DMY, so converting the exceptional cases to that is reasonable. However, DATERET allows discussion for consensus building to argue a different format if one believes the current one is not correct, and that's what should be done here. For video games we have set a rule of thumb to base the dates on the developer's national ties, but there are cases for clear exceptions (The Beatles: Rock Band being one, since the importance of the Beatles in a video game far outweighed the importance of the developer) --MASEM (t) 20:52, 12 July 2015 (UTC)

A small but possibly key detail: was it really clear as to the established date format? I would take the presence of the 'use dmy' template as a strong indication, but I seem to recall some protest that these do not so indicate. ~ J. Johnson (JJ) (talk) 22:06, 12 July 2015 (UTC)

Keeping in mind that the refernences prior to the chance were pretty much in dmy , that definitely means dates in prose must be dmy too (you can't use ISO in prose, and mdy conflicts with dmy between prose and reflists). --MASEM (t) 22:19, 12 July 2015 (UTC)

I don't know that dates in references control dates in the text. But regardless of that, my question is: what constitutes "clear"? Is an editor supposed to count up all the instances either way? Does a template have any weight on this? ~ J. Johnson (JJ) (talk) 23:23, 12 July 2015 (UTC)

Fixing dates to normalize to the more common format means that it should be readily obvious - without exact counting - what the predominate date format has been used. If a visual scan of the article's prose and references don't easily show a near uniform date format, then it is probably best to check on the talk page. I would safely say that the change in question did have proper allowance based on the near majority of dmy sources. And yes, the template should have significant weight on the assumption it was added in good faith. And the only reason in this case that reference format drive the prose is that it was definitely the case that the ref date format was near uniformly dmy, and since mdy in prose and dmy in reference is unallowable, that forced the prose to be in dmy too. --MASEM (t) 23:34, 12 July 2015 (UTC)

MOS:NUM#Consistency, right above DATETIES AND DATERET) lists three categories (dates in article body text, publication dates, and access and archive dates) within which dates "should all use the same format". But I don't see that it says all three categories should be the same. It says that it is permissible to "normalize" access/archive dates to the text, but does not suggest that the contrary is true. So I would say that the format of access dates does not drive the format in the text. I agree with you that the dmy template should have significant weight, but I suspect many editors would disagree. As to "clear", it does seem to be one of those "it's obvious" matters. ~ J. Johnson (JJ) (talk

) 20:34, 13 July 2015 (UTC)

The second point of Publication dates spells this out: "provided the day and month elements are in the same order as in dates in the article body". That is if you have prose as DMY, you cannot use MDY in citations, but you can use DMY, an abbreviated DMY, or YYYY-MM-DD. While this suggests that prose should drive the citation dates, it's basically the easiest language to state that you can't mix DMY and MDY in prose and references. And since that article before the edit was a mix of DMY and MDY between refs and body (which is not acceptable), it was a appropriate choice to go with what was the fewest number to fix, being moving MDY to DMY. --MASEM (t) 23:16, 14 July 2015 (UTC)

Well, I'm not convinced on that interpretation. But for this discussion I think we could accept that converting to DMY was appropriate, on the basis of retaining an existing usage per DATERET. But what of JOJ's argument that the initial MDY date in the text set the format, with the implication that the subsequent history is immaterial? ~ J. Johnson (JJ) (talk) 20:18, 15 July 2015 (UTC)

Now, there, [6] this edit does appear to be where X201 unilaterally change date formats based on the developer's nationality, which should be something checked in before doing on the talk page (though in the VG project, we do state that date formats should follow the nationality of the developer which Travelers Tales is UK based. But again, this isn't as strong a national tie as, say, if it was the developer's page themselves). --03:27, 16 July 2015 (UTC) ← Masem

Which is to show that X201 wasn't arbitrary, but proceeded on the basis of some kind of established guidance. This raises an interesting question regarding an intermediate level of guidance worked out by a community of editors greater than those working on a particular article, but less than the whole WP community. (Such as on a project level.) Even if such is allowed (and I am not against it), there is a significant problem: it's not in the MOS. So how would some passing passing editor not familiar with that particular guidance know about it? How should the MOS bear on such intermediate level guidance? ~ J. Johnson (JJ) (talk) 21:15, 17 July 2015 (UTC)

As no one else cares to explain how the date guidelines should be applied in this case, I will hazard an opinion. The specific guideline relevant here appears to be [MOS:DATERET]], which starts by saying "If an article has evolved using predominantly one format, the whole article should conform to it ...." Jojhutton, who made the questioned edit, also cites DATERET, and argues two points in this discussion (above): 1) that the article in question is not consistent with a closely related article (implying that consistency should apply to related articles), and 2) that "MDY was the first date format introduced".

The first point is interesting, but not supported in these guidelines. And both here and elsewhere in the MOS (e.g.,

The second point is implicitly a claim on the third part of DATERET, which starts: "Where an article has shown no clear sign of which format is used ..." [emphasis added]. However, this is inapplicable, as the article was using "predominately one formt" immediately prior to Jojhutton's edit, and has been so, including use of the dmy template, since 2012. DATERET expressly states: Retain exsting format, unless certain exceptions apply, which is not alleged here.

In response to the question asked: there are no policies regarding date format particularly. But the MOS - a guideline - does NOT support conversions of date format where the existing usage is predominately consistent. The specific edit cited here appears to be unsupported, and therefore may properly be reverted.

Pinging the editors involved (X201, Jojhutton, and Eldizzino) for their comments. ~ J. Johnson (JJ) (talk) 22:24, 19 July 2015 (UTC)

First off, please accept my apologies for not contributing to this discussion; major off-wiki tasks/problems have claimed my time. I suppose explaining the rationale for the edits would be a good place to start - although Masem is very close to the full explanation. The things running through my head when I made the change were: A scan of the article where, DMY looked like the dominant form, the WP:VG guideline about developer nationality (Traveller's Tales as mentioned above). I also considered the fact that the product the game is based on is designed by Lego, and the Lego company is well known for its ties to its home nation of Denmark (

MOS:TIES

again), Batman is obviously American, but I'm sure most people will regard Lego Batman as a Lego product, rather than a Batman one. My edit summaries should have been vastly more informative than they were, I hold my hands up to that charge, but the edits were made with the best of intentions about what I thought was the right decision for the article.

My addition to the discussion is that the date guidelines need to nudge editors a little more into going down a particular route. To explain, DATERET reads as a bit Land Grab-y, it explains that it can be changed with discussion, but the weight of the guideline is with the founder(s) of the article. I can understand that, it needs a guideline of some sort and its a decent enough choice until the developers decide bring Wiki software up to date and allow user preference for date formats. What I think its lacking though is the "use common sense element", telling users to think about what is appropriate for an article, rather than just use their own default date format. - X201 (talk) 09:05, 20 July 2015 (UTC)

The superhero article is resolved by using the date format expected by the article's ENGVAR.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  16:16, 20 July 2015 (UTC)

"Common sense" is commonly invoked, but suffers from everyone having different views on just what it is. The whole purpose of a guideline is to guide our sense of what is proper, and to provide a basis for resolving conflicts. ~ J. Johnson (JJ) (talk) 22:53, 20 July 2015 (UTC)

There being no further discussion, either here or at the article's talk page, I have reverted the last date-format change there per DATERET. ~ J. Johnson (JJ) (talk) 20:31, 28 July 2015 (UTC)

I just realized that Canada fixes most EVGVAR vs. DATEVAR disputes. It's quite simple once you see it:

• For encyclopedic-writing purposes, there is no difference between American English and Canadian English that cannot be worked around with some minor rewording occasionally ("washroom" vs. "restroom", etc.), unless the first major contributor insists on Briticized spellings ("colour", etc.). And even then consensus on the talk page could overturn that (but it wouldn't be necessary to, as we'll see). The vocabulary is about 99.99% shared.
• Canadian English accepts either DMY or MDY date format.
• Many topics have a natural DATEVAR due to national ties (e.g. a Russian movie), but no natural ENGVAR because of lack of national ties to the language.
• If written in anything other than North American English, international date format applies by default. No issue.
• If written in North American English using Commonwealth spelling, this is Canadian, and intl. format dates are permitted.
• If written in North American English using non-Commonwealth spelling, this is either American or Canadian English. Canadian permits intl. date format, so we simply declare it Canadian, and here – the dispute sticking point – there suddenly is no issue any more.

It works the other direction, too with a bit more work: Any Commonwealth English article is not much different from a Canadian-with-British-spelling one (there are more vocabulary tweaks that could be needed, but nothing insurmountable). And cases of Commonwealth English but US date format are rare if they even come up at all. Is there any place that conventionally uses US date format but no longer has strong US ties, and might have some articles already written about them in non-North American English? Blind guesses might be Libera or the Philippines.

This would eliminate virtually all fights about DATEVAR vs. ENGVAR. It's also evidence that ENGVAR and DATEVAR are

I prefer to blame Canada. --Izno (talk) 17:35, 20 July 2015 (UTC)

America, fuck yeah? LOL  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  17:41, 20 July 2015 (UTC)

Doesn't this just shift whatever problem we have one stop over? The variety of English applied to a topic is often random according to whatever editors have had a whack at it. If some article has been through a GA candidate process, some practical soul will have gone through and undiscrepancied the text, but by and large if I select a random article that doesn't get a lot of interest, it looks exactly like the result of random editors with random writing styles adding a sentence or two at random times. Wikipedia's prose style rarely sparkles.

Perhaps I'm misunderstanding your solution to a nebulous problem, but it seems to me that you are proposing that allowable date formats are linked to variety of English, so instead of arguing over whether an international-format date applies to (say) an article on a French painter, we will argue over what variety of English the article is written in, and have some more rules to puzzle newbies with. Just as an aside, French painter articles come in a wide variety of styles. --Pete (talk) 22:18, 20 July 2015 (UTC)

Editors do want to link the DATEVAR to the ENGVAR, and they're conceptually closely related. The mind rebels at the notion that dialectal language use is of such paramount importance that we must have rules about it that enforce within-the-article consistency and not changing the style already used in the article, yet also permit a date format change that directly conflicts with the dialect used in every other way in the same article. Walk the logic tree I laid out. If an intractable debate about this arises at a particular article, the way to short-circuit it is to just have consensus (which can override the first major contributor) declare that the ENGVAR that page uses is Canadian, which is compatible with both Commonwealth and US spelling (one or the other at the article in question, not both at once, of course), and both of the typical date formats used in articles. The only adjustments needed will be rare cases of vocabulary incompatibility, easily resolved by simply rewording.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  11:17, 22 July 2015 (UTC)

Ah, I see. Reminds me of my time in the military, where we would often find ourselves doing things that were at the same time quite logical but also patently ridiculous. Taking a shower in full uniform with weapons, for example. --Pete (talk) 16:22, 22 July 2015 (UTC)

I'm confused. Are you implying that "color" is an acceptable spelling in Canadian English? It isn't. RGloucester — ☎ 22:38, 20 July 2015 (UTC)

with which you have no actual familiarity is not a productive endeavor on Wikipedia, and greatly reduces the willingness of other editors to take you seriously each time it happens, which is frequently. I actually lived in Canada only a few years ago, so I know for a fact you're wrong in implying "color" is not acceptable in Canadian English. Have you ever even been there? But all you had to do was Google this, or consult any reliable source, like the The Oxford Canadian Dictionary, or even just read our well-enough sourced article Canadian English.  — SMcCandlish ☺ ☏ ¢

 ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  11:07, 22 July 2015 (UTC)

I missed this. Yes, I've "been there", and even lived there briefly. "Color" has not been an acceptable spelling in standard Canadian written English. Sure, a few newspapers may have used such Americanised spellings decades ago, but none do so now. RGloucester — ☎ 23:03, 28 July 2015 (UTC)

Canadian English has always vacillated between American and British spelling but in recent years a definite standard for "Canadian" English has been developing, and it is tending to be closer to British spelling (or more accurately, Commonwealth spelling) than to American spelling. Thus "colour" has been winning out over "color" and is almost universal now. A few decades ago a lot of Canadian newspapers used American spelling because it saved the trouble of rewriting articles from American news sources, but today editors can change the spelling with the click of a mouse, so almost all media use "colour" instead of "color". It's not British spelling, though, and nobody uses "tyre" instead of "tire" in Canada. The MDY versus DMY dispute remains unresolved. Canadians use either indiscriminately, and I personally use YMD because that is also a Canadian "standard". Canadian banks give you a choice of any of the three on your preprinted cheques (not checks). RockyMtnGuy (talk) 12:32, 31 July 2015 (UTC)

I know. Doesn't affect what I'm saying here, though: If people are having cognitive dissonance about an apparent conflict between the ENGVAR and DATEVAR being used, a Canadian English designation provides an "escape valve". Since we accept any non-pidgin English dialect as valid, it doesn't even matter that this might eventually shift some spellings and vocabulary in the article.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  21:22, 2 August 2015 (UTC)

Here, I raised an issue of using the convert template for metric caliber identification & it was suggested the MoS governs. (I imagine the template will need to be adjusted to permit the change, if approved, too.) So, can editors who need this "special case" get a template able to cope with it? FYI, posted the issue on the Firearms & Milhist Talk pages, with links here. TREKphiler ^{any time you're ready, Uhura} 22:56, 21 July 2015 (UTC)

Can you restate that in more concrete terms? I can't quite peer into your mind. :-)  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  12:32, 22 July 2015 (UTC)

As explained at the first link, this is about not repeating the unit when a X is used. For example listing the caliber "9 × 25 mm (0.35 × 0.98 in)" [proposed] instead of "9 mm × 25 mm (0.35 in × 0.98 in)" [current]. -Fnlayson (talk) 19:10, 22 July 2015 (UTC)

Perhaps this would be better raised at the template talk page? I'm not sure if we have a guideline on this precise application. --Pete (talk) 22:36, 22 July 2015 (UTC)

"Perhaps this would be better raised at the template talk page?" I went there first & was told it was an MOS compliance issue... If it's not... (And Fnlayson has me right: I want to be rid of the first use of the unit, or limit to one after only. No spaces would be good, too.) TREKphiler ^{any time you're ready, Uhura} 23:28, 23 July 2015 (UTC)

It will be a template issue if there's a change, but yes, you are right. It is an MoS issue, and the usage mentioned above seems to be specifically ruled out. As the usage is commonplace in military/firearms circles and our own articles (5.56×45mm NATO and 7.62×51mm NATO and 7.62×39mm, for example), I think we should allow it and modify the wording appropriately. Are there any objections? --Pete (talk) 23:48, 23 July 2015 (UTC)

Two separate issues

1. MOS should be recommending "9 × 25 mm (0.35 × 0.98 in)" not "9 mm × 25 mm (0.35 in × 0.98 in)", except in cases where for some reason some ambiguity or confusion could result. Using "9 mm × 25 mm (0.35 in × 0.98 in)" is obnoxiously redundant and treats our readers like morons.
2. "5.56×45mm NATO" is a name, that happens to be based on a measurement, but is not itself a measurement; it is thus correct without the spaces if the overwhelming majority of reliable sources (not just specialist sources) write it that way, which appears to be the case. I.e., a correct sentence to illustrate style not how to write a good encyclopedia sentence would be "The 5.56×45mm NATO ammunition measures 5.56 × 45 mm."
3. These measurements (not as embedded in the caliber name!) should be given with the {{Convert}} template.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  03:27, 26 July 2015 (UTC)

Tread carefully re 5.56 × 45 mm vs. 5.56 mm × 45 mm. There's been (surprise!) some very hot debate on this in the last 6-18 months -- can't remember exactly, but I know it's been since I first became active here in early 2014, and all I can find is

this (which nonetheless gives you a taste). EEng (talk

) 21:51, 2 August 2015 (UTC) Sorry I wimped out at /Linking after all that fuss, but life's too short.

This issue has come up many times, most recently in May 2015. The crux is whether we are talking about the name of the ammunition, in which case it should be whatever the name is, or about the dimensions of said ammunition, in which case IMO it should '5.56 mm x 45 mm'. Tread carefully is good advice. Dondervogel 2 (talk) 22:23, 2 August 2015 (UTC)

While in general C notation seems fine, in some specific cases I believe other notation should be used. Articles about other languages that have an appropriate notation should use that notation. Personally, I have never found anyone confused by the OS/360 assembler X'hexdigit' notation, but if it could be confusing an article should explain the notation early on. Verilog and VHDL each have their own hex and binary notation, as do some other languages. But I agree, in general discussion, without a specific language in context, C notation is a fine choice. Gah4 (talk) 20:42, 21 July 2015 (UTC)

Are you proposing that a sentence be added that, in discussion of a particular computer language, the language's own convention should be used? And where does this leave the mathematical joke "Halloween = Christmas" (that is, 31OCT = 25DEC)? — Arthur Rubin (talk) 22:05, 21 July 2015 (UTC)

I think the proposal is that without a specific language in context, C notation is a fine choice. All the best: Rich Farmbrough, 23:14, 3 August 2015 (UTC).

 •   Osmund (fl. 760–772) ... [already did this edit there – possibly wrong? ]
 •   Aethelwalh (fl. c. 660 – 685) ...

Seems "760–772" should be spaced (or this example dropped). I'm not sure anyone flourishes this exactly.. or vice versa "fl. c. 660–685"? Unclear if spaced for circa is meant for the space after "c. " only or for the years. There is a rule about both years requiring c. then "fl. c. 660 – c. 685".

Anyway was looking into if this in British Antarctic Survey should be spaced (can't change/not sure how with timeline template):
1973–May 1987 – Richard Laws[7][8] [..]
2007–May 2012 – Nick Owens[11][12]
November 2012–September 2013

Also seems "FIDS/BAS/" has an extra slash.. comp.arch (talk) 15:53, 22 July 2015 (UTC)

Yes, when one end of a year range begins with a prefix such as fl. or c., the en dash must be spaced. Among other things, this emphasizes to the reader that the prefix on the first item in the range does not also apply to the second item. This is really just a special case of the rule that if either end of a range includes a space, the en dash must be spaced, which requires

1973 – May 1987

November 2012 – September 2013

—Anomalocaris (talk) 14:39, 24 July 2015 (UTC)

Generally fl applies to the range, not one end of it. Therefore the first endpoint doesn't contain a space, and therefore there's no spaced ndash (unless one or the other endpoint has e.g. a month, or c. -- though not sure c. really makes sense with fl.). EEng (talk) 23:10, 24 July 2015 (UTC)

I agree, and I have crossed out part of my previous comment. —Anomalocaris (talk) 07:01, 25 July 2015 (UTC)

Note, for this reason, you would not have (fl. c. 660 – 685) as there needs to be a space after the fl. before the starting point in the range; thus: (fl.  c. 660 – 685) (assuming the end point was not approximate, otherside it should have c. 685). —sroc 💬 03:41, 26 July 2015 (UTC)

My take:

Format	Meaning & notes
660–685	between those exact years, inclusive
c. 660 – 686	660-ish to exactly 686
660 – c. 686	660 exactly to 686-ish
c. 660 – c. 686	660-ish to 686-ish
c. 660–670	around that decade; while essentially shorthand for c. 660 – c. 670, more concise yet as c. 665 in many contexts; possibly confusing in general. Deprecation suggested (i.e., merge into next example).
c. 660–685	Deprecated – uncertain meaning; instead use c. 660 – 686 or c. 660 – c. 686, as needed
fl. 660–685	between at least these years, both certain
fl. c. 660 – 685	between 660-ish and 685 for certain; probably better explained in words (easy to misinterpret as fl. c. 660 – c. 685, as well as possibly confusing in general)
fl. 660 – c. 685	between 660 for certain and 685-ish; probably better explained in words (possibly confusing in general)
fl. c. 660 – c. 685	between 660-ish and 685-ish; rare usage, probably better explained in words (possibly confusing in general)
fl. c. 660–670	somewhere in that decade probably; while essentially shorthand for fl. c. 660 – c. 670, more concise yet as fl. ca. 665 in many contexts; probably better explained in words (possibly confusing in general). Deprecation suggested (i.e., merge into next example).
fl. c. 660–685	Deprecated – uncertain meaning; instead, use fl. c. 660 – 685, or fl. c. 660 – c. 685, as needed

Hope that helps. PS: We should be permitting "ca." as well as "c."; they're both equally acceptable in usage guides, and as attested in reliable sources, while "ca." is easier to parse as "circa" instead of as something else., especially in running prose (though spelling it out in that case might better).  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  06:40, 27 July 2015 (UTC)

Maybe we can integrate your table into the guidelines -- many are confused by this, at least the rare times they encounter it. Perhaps I'll mae a go at it after I get started over at /Linking. EEng (talk) 16:23, 27 July 2015 (UTC)

Well, we'd have to use better wording. My "-ish" stuff is an informalism.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  15:18, 29 July 2015 (UTC)

I urge not using c. 660–670 because the reader has no way of knowing that the unspaced en dash is intended to bind the c. to the whole range; in many cases the c. is intended to bind only to the start of the range. It's just sloppy. —Anomalocaris (talk) 06:40, 28 July 2015 (UTC)

I could get behind that. I wasn't recommending using it, just documenting it as meaning something when people do use it.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  15:18, 29 July 2015 (UTC)

In terms of lifespan fl. 670-790 is distinct from fl. 670 died 790 . Also we may write fl. 670 - c. 790 or fl. 670 died c. 790 : in the first case we have a document attesting she was alive which is dated around 790, in the second one stating she died, that is dated around 790 (for example - it may be that the document is precisely dated but that it refers to an event which isn't). All the best: Rich Farmbrough, 23:23, 3 August 2015 (UTC).

MOS:DATEFORMAT

In the "Acceptable date formats" table, it says that YYYY-MM-DD is an acceptable format for brevity. But then it says "No equivalent for general use" for the expanded format.

Why exactly can't we use YYYY MMM DD or YYYY MMMM DD? Seems perfectly reasonable to me... – SarahTehCat (talk) 20:53, 2 August 2015 (UTC)

YYYY-MM-DD is very widely used ISO standard.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  21:05, 2 August 2015 (UTC)

MOS:BADDATEFORMAT

I've two questions about this section:

1. Why are the formats YYYY MMM and YYYY MMMM not allowed? Seems a bit half-assing it, to be honest, if you allow year-month-day but not year-month, especially since the reverse, year-day, is allowed. Just wondering.

2. I would hardly blame anyone for refusing to use this format, but just out of curiousity, if you can't do YYYY MMM or YYYY MMMM, is the format YYYY-'M'MM allowed? After all, this is one of the accepted forms within the ISO 8601 standard.

I assume it is not allowed, but I would like to know for sure, nevertheless. :) – SarahTehCat (talk) 20:53, 2 August 2015 (UTC)

1. Same answer as above. The spaced version isn't a standardized format, it's just weird and easily misinterpreted.
2. Huh? That seems to be an argument for 2015-'August'08 instead of August 2015. I don't find anything at that article suggesting anything like this. MOS doesn't, and can't allow every possible ISO 8601 quirk, anyway. Many of them are not parseable except by experts, and were intended for machine reading. We would never use 2015-08-02T15:57:44Z or 2015-W31-7 (and those are comparatively human-readable, given some of the even more compressed constructions). And some of them just do not work at all in our context, no matter how formatted, e.g. 'one may write ... "19" to refer to the century from 1900 to 1999 inclusive.' On WP, as in all other publications intended for human reading, a year value of "19" means
   19 CE not the 20th century.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  21:16, 2 August 2015 (UTC)

   PS: I'll quote User:EEng two threads above this: 'ISO 8601 is a data exchange standard and has nothing to do with everyday writing'.  — SMcCandlish ☺ ☏ ¢

    ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  21:19, 2 August 2015 (UTC)

I keep repeating this and being ignored, but I'll repeat it again. The YYYY-MM-DD date format is widely used in written communications in Canada, as also are the European DD/MM/YYYY and American MM/DD/YYYY formats. I've worked for a number of companies that mandated YYYY-DD-MM as their standard for all company documents. It solved the fundamental ambiguity on whether 05/08/2015 is May 8 or August 5 - t's always bad to show up in the wrong month for an important event, and some people have done that. Mandating the YYYY-MM-DD format also avoided annoying arguments like these on Wikipedia between British and American expatriates over which date format to use, and is compatible with Quebec law, which often makes it illegal to use English words like "August" in workplaces. YYYY-MM-DD is also the standard date format in Chinese, and Chinese have been the largest group of immigrants to Canada in recent years. (Scary fact: there are more people in China who speak English than there are people in the US who speak English). The same companies often mandated YYYY-MMM-DD as their standard for short dates and YYYY MMMM DD as their standard for long dates. Thus the date could be written as 2015-08-05, 2015-Aug-05, or 2015 August 5. RockyMtnGuy (talk) 18:54, 5 August 2015 (UTC)

You aren't being ignored. You are simply not getting traction for your viewpoints and refuse to drop the stick. This is a good example, since pretty much the one date format NOT being discussed in this subsection is YYYY-MM-DD. And, to rehash the previous objection to your arguments - even if YYYY-MM-DD is commonly used in technical documentation (citation needed) and particularly in areas that require machine readable consistency, Wikipedia is written for humans. And humans don't use "2015-08-05" in regular conversation or reading. Resolute 19:12, 5 August 2015 (UTC)

Yes. yyyy-mm-dd, with its leading year, is a special format used where sorting of dates is a consideration (including filing of coreespondence). It's also an all-numeric format, and a special case in being the only numeric format allowed here. In formal writing (i.e., WP article text), the year is generally trailing, and the issue between DMY/MDY (where "M" is non-numeric) is the order of the day and month. ~ J. Johnson (JJ) (talk) 20:46, 5 August 2015 (UTC)

Yes, humans don't use "2015-08-05" in regular conversation or reading - if you exclude the Chinese, of which there are 1.4 billion, from the human race. What you are really saying is that "Americans and British don't use 2015-08-05 and they are the only people who matter." But Americans and British don't agree on which format to use, which is why many Canadians use "2015-08-05" instead. It is the Mercutio A plague on both your houses approach.

To quote from the Wikipedia article Date and time notation in Canada:

While the Canadian Standards Association has adopted ISO 8601 as CSA Z234.5:1989, its use is not mandated in every situation. Thus in Canada three date and time formats are in common use. According to the Canadian Payments Association, which regulates cheques, the big endian ISO 8601 YYYYMMDD is preferred, but MMDDYYYY or DDMMYYYY may be used, and cheques must include date indicators showing which format is being used. The federal government tends to use the big endian format, but some federal forms, such as a commercial cargo manifest, offer a blank line with no guidance. Passport applications and tax returns use YYYY-MM-DD. Government of Canada regulations for expiry dates on foods mandate YYMMDD, MMDD, and DDMMYY. English language newspapers use MDY (MMM[M] D, YYYY). In Quebec a variation of DDMMYYYY is used.

I often liken it to screws. Americans usually use Phillips screws. Canadians prefer Robertson screws, which are much superior screws. The only reason Americans use Phillips screws is that Henry Ford didn't want to pay to use Robertson's patents, and Phillips more or less gave his patents away for free. As a result I am continually taking out the screws supplied with American products, often with a screw extractor because they have been damaged due to their poor design, and throwing them away, substituting better Robertson screws. Of course, car manufacturers now use Torx screws because you can't get away with screwing cars together with inferior screws any more, but you still have to put up with crappy Phillips screws in other American consumer products, and buy a set of torx screwdrivers to work on your car. Should Canadians be forced to use inferior screws just because Americans do? Of course not. Same thing with date formats. RockyMtnGuy (talk) 03:55, 6 August 2015 (UTC)

I like YYYY-MM-DD as a format and use it often. Every time I make a subdirectory to hold a day's photos, for example. The Bigendian format sorts automatically, making my directories easy to navigate. In a perfect world, we would all use this or a similar format and there would be no disputes.

However, we at Wikipedia do not yet have the full control over making the world perfect. We must work with what we have, and the sad fact is that in everyday English text numeric format dates jar, especially in YYYY-MM-DD sequence. It would be like reading aloud except that everytime you encountered a comma, instead of pausing, you clapped your hands. If we went to YYYY-MM-DD in article text, there would be an uproar. People would say we had lost our grip.

Personally, I think the US Middle-endian date format sucks. But it's what people use in the US (mostly), and it's what our American editors feel happiest using as they write our articles. I think that the current situation, where US subjects use US dates, British subjects use international dates, Canadians and every other subject use either works well to minimise disruption. We might be able to jigger our system around a little, but any major departure from what we have would piss off our readers and editors alike. We're an encyclopaedia, presenting information to those who seek it. We aren't a religion, dogmatically enforcing our views and foibles on a helpless world. Yet. --Pete (talk) 05:04, 6 August 2015 (UTC)

I object to calling the DMY format the "international" date format. It is really just the conventional European date format, with former European colonies in South America and Australia following former colonial standards. China, Japan, and Korea use the YMD format instead. So, in reality, the world has three formats: the American MDY format, the European DMY format, and the Asian YMD format. The ISO 8601 standard is compatible with the Asian date format, so it is what is used by a large portion of the world's population. Since it is set by the International Standards Organization (ISO), the global reality is that the International date format is YMD, not DMY. RockyMtnGuy (talk) 19:01, 7 August 2015 (UTC)

I take your point on ISO being an international format. However. China, Japan, and Korea use a different way of presenting text. It's not a phonographic system such as we and many other cultures use–Indian use of Devanagari, for example–where each character represents a specific sound, more or less. We use Roman characters and an English alphabet, and we don't present dates in text as number strings. In the USA and one or two other places, it is August 8, 2015. In the rest of the world, it is 8 August 2015. In some places, where they use other ways of writing, it is 2015-8-8. But we need not bother about those other places, because we are not presenting our encyclopaedia's text in those characters. I call DMY "international format" because that's what it is, when compared to US format. If you ae arguing that we should use ISO dates in our English-language encyclopaedia because the Chinese use it when writing a different language in a different character set, then I hear what you are saying and regard it as irrelevant to our work here. --Pete (talk) 19:29, 7 August 2015 (UTC)

"Yes, humans don't use "2015-08-05" in regular conversation or reading - if you exclude the Chinese, of which there are 1.4 billion ..." We do exclude them from consideration here, because this is the English language Wikipedia, not the Chinese language Wikipedia. —sroc 💬 14:09, 6 August 2015 (UTC)

And more importantly, RockyMtnGuy, please stop claiming Canadians use that date format on a regular basis in normal writing, because I can tell you as a fellow Canadian, you are completely wrong. Resolute 20:37, 6 August 2015 (UTC)

What you are really saying is that Canadians use other date formats in informal writing. Business writing is a different issue. I know Canadians use YYYY-MM-DD on cheques because I was Treasurer of a not-for-profit organization for 8 years. It was the preferred format for the Canadian Payments Association so it was what I used. I know it is used in business writing because I was a Business Analyst working and/or consulting for several companies, and many of them mandated the YYYY-MM-DD format for all written documents. I know Canadian governments, particularly the ones I dealt with, often used it on forms, although governments are mostly consistent in their inconsistency. And, whenever I designed computer systems, I used YYYY-MM-DD. It saved my employers and clients a large fortune when the Y2K imbroglio hit, because everything I designed from the start of my career was Y2K compliant, from 1970 forward. Being non-compliant with standards and catering to everyone's personal whims costs companies money, and ISO 8601 is clear, unambiguous and foolproof, unlike DMY and MDY. The main obstacle was Microsoft, whose programmers didn't even know what ISO 8601 was. Some organizations came close to suing them over their date handing in Y2K. Some organizations I worked for actually did sue them, but over other things they screwed up. RockyMtnGuy (talk) 18:30, 7 August 2015 (UTC)

What format is used on cheques, on forms, in computer systems, is immaterial here. We're talking about writing dates in text. If you could provide links to some samples of contemporary Canadian prose with dates in iSO format, that would be helpful here. --Pete (talk) 19:29, 7 August 2015 (UTC)

Yeah. None of the cheques I was paid with when I lived in Canada were in YYYY-MM-DD form. That's something that a large business entity would do, at the request of their bank, their own account dept, or (most often) because a payroll company did it that way. It would be done that way for automated data processing reasons, in other words, not because it's an everyday Canadian human preference, not even in business writing. When I lived in Toronto, the preference, for dates everywhere (on signs, letters, TV listings, whatever) was somewhat toward "6 August 2015", but "August 6, 2015" was also encountered (among more wordy constructions like "August 6th, 2015", "the 6th of August, 2015", etc., in journalism and other less clipped contexts). In compressed form, people did all kinds of things, but it was pretty rare, in my memory anyway, to encounter "06-08-2015" or anything like that, because of the innate ambiguity. People would almost always use something clear like "6 Aug 2015" or "6-Aug-2015" or "Aug. 6 2015" or whatever suited their preferences, but not all-numbers, in any order, including ISO order except among the programmers I was working with (who lean toward that format worldwide, just as they also often put slashes through their zeroes in even in handwriting). There seemed to be an uneasy, default assumption in TO that "06-08-2015" or "6/8/2015" meant an August date not an American June date (probably because a majority of e-mail, blog, webboard, and other software does it this way, worldwide, when not using YYYY-MM-DD format); but urban, south-border Canada does so much business with urban north-border US that people genuinely seemed to avoid this kind of date format in everyday life. I wasn't in Montreal, the Martimes, the Prairies, Vancouver, the Yukon, etc., so I can't generalize about Canada, only report what I noticed in the GTA in the 2000s.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  01:30, 9 August 2015 (UTC)

There is a template called {{

1. Is "age" the correct syntax for US style date?
2. If so should an change be made to the template to implement a switch from "age" to "aged" depending on the date format chosen?

-- PBS (talk) 13:17, 3 August 2015 (UTC)

I've long counseled that templates shouldn't "enforce MOS" by forcing the output to conform to MOS' current ideas of e.g. that this date format goes with that choice of age vs. aged. Consensus on such things can change, and it's intolerable that what should be a technical tool (the template) becomes a defacto encystment of an old decision which makes it difficult to change that decision at a later time.

In particular, as to "If we can agree on 1. above, then let's do 2.", I guarantee that's a bad idea‍—‌rank-and-file editors HATE that kind of silent, involuntary mass change based on a discussion among a relatively few editors. This is true even if the change is a default that can be overridden.

Unless someone thinks age is always "wrong", we can implement the following change right now:

make age/aged simply another parameter, with the default being the only current possibility i.e. aged.

This will change nothing in current invocations, but editors will have it available. Maybe (maybe) someday there it will be appropriate for MOS to say something about which of age/aged to use, when. EEng (talk) 20:19, 3 August 2015 (UTC)

I think "age" is always wrong, and I'm in the US. One might say "died August 4, 2015, at age 77." — Arthur Rubin (talk) 21:24, 3 August 2015 (UTC)

Thank goodness for that! One less linguistic division. All the best: Rich Farmbrough, 23:25, 3 August 2015 (UTC).

I don't particularly care about this, I simply advertised it here so that editors who do care (like EEng and AR and RF) can express their view in that debate (it seems futile putting views here, as it will not affect the outcome of that debate). -- PBS (talk) 15:42, 4 August 2015 (UTC)

I'd say it's best not to tie one thing to another on the basis that Americans do x and y whilst normal people do i and j. Hey, Canadians might just do x and j. This kind of thing has plagued {{convert}} for years (e.g. "US"/"U.S." linked with "~re"/"~er" ... somewhat my bad but in spite of my attempts to undo it's hard to stamp out). I'd suggest keeping these separate. Note also that usages change anyhow. Jimp 16:28, 4 August 2015 (UTC)

Age/aged is not an ENGVAR matter, it's purely syntactical; as Arthur points out, "age" can be used after a preposition like "at". But the construction "died age 77" or "died, age 77" is substandard

telegraphic writing, and is properly "aged" not "age", in any major dialect.  — SMcCandlish ☺ ☏ ¢

 ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  01:51, 9 August 2015 (UTC)

 •   1881–86;  1881–92 (not 1881–6;  1881 – 86)

The real trouble however begins if say 1998–2000 needs to be introduced (say another section title). Do we want to mix YYYY-YY and YYYY-YYYY ranges? We can never know what needs to be added later to an article.. comp.arch (talk) 14:20, 27 July 2015 (UTC)

Many (most?) other publications would use 2000–01 rather than 2000–2001 to mean the the year range. I don't know of any English-language prose publication that uses 2000-01 to mean January 2000, although one might see it in databases. Since people manage to figure it out in all those other publications, I think they can figure it out in Wikipedia too. Jc3s5h (talk) 14:29, 27 July 2015 (UTC)

In all countries and contexts, where the ISO 8601 date format is the recommended date format or even mandantory to be used, 2000-01 would be normally interpreted as January 2000, not as 2000-2001, regardless of language. Even more so since abbreviated year numbers are broadly deprecated since the millennium for all those Y2K bugs they caused. The good news is that it is possible to learn from past mistakes. --Matthiaspaul (talk) 15:28, 27 July 2015 (UTC)

Help is at hand. The MoS does not make the two digits mandatory. There is another rule - we use the most common construction. People generally write as they speak - they say, for example, "nineteen sixty-six to seven" and they write 1966 - 7, not 1966 - 67.

Thus 2001 - 2 is perfectly acceptable, and avoids the 2001 - 02 construction which some would think means February 2001. 213.123.194.188 (talk) 15:39, 27 July 2015 (UTC)

Another way round this is to use the slash. You could write 2000/01 for example and I don't think anyone would mistake that for January 2000. 213.123.194.188 (talk) 15:43, 27 July 2015 (UTC)

@Matthiaspaul:, I don't accept your implication that there is an English-speaking country that requires, or even recommends, ISO 8601 in prose. Indeed, English-speaking countries aren't even in the business of telling their inhabitants how to speak or write. Can you prove that such a recommendation from the government of an English-speaking country exists, and if so, that the recommendation carries any weight with the inhabitants? Jc3s5h (talk) 17:47, 27 July 2015 (UTC)

Christ, this ISO crap again??? ISO 8601 is a data exchange standard and has nothing to do with everyday writing. And as Jc has pointed out, no English-speaking country, AFAIK, commands its inhabitants to read, write, and speak in certain ways. EEng (talk) 19:01, 27 July 2015 (UTC)

Thanks (all) for answering. I'm sorry for starting with the minor point. Nobody has addressed my main point ("real trouble"..). [2001–2 wouldn't really help there, 1991–2002, might happen (or just 2001–12), besides, 1991–2 is just as ambiguous as 1991–02 is that case.] comp.arch (talk) 30:13, 27 July 2015 (UTC)

• It simply isn't true that most other publications would use 2000–01; many informal and journalistic sources would do this, especially in compressed spaces, but it introduces a hopeless ambiguity if the number after the dash is under 13, and will be interpreted by many as a month.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  02:02, 9 August 2015 (UTC)

Following the discussion at Wikipedia talk:Manual of Style/Dates and numbers/Archive 152 § Interaction between DATEFORMAT guidelines §§ Essay that petered out, I humbly suggest the following hatnote under the heading for Wikipedia:Manual of Style/Dates and numbers § Chronological items §§ Dates, months and years §§§ Formats:

Formats

See also: Wikipedia:Overview of date formatting guidelines

Thoughts? —sroc 💬 07:20, 17 August 2015 (UTC)

I think a "Hitchhiker's guide to date formats", to help editors tie together the various date-format advice scattered here and there, fills a niche. EEng (talk) 17:44, 17 August 2015 (UTC)

Such a link seems dubious. To the extent your essay is just a restatement of the guideline it adds nothing new, and there is not point for the link. To the extent that it goes beyond a mere restatement and strengthens a particular interpretation, but without acknowledging that strengthening ("clarification"), linking it into the offical guideline amounts to a covert modification of the MOS. ~ J. Johnson (JJ) (talk) 23:25, 17 August 2015 (UTC)

And we-rrrre off! EEng (talk) 04:40, 18 August 2015 (UTC)

I hope not. Sroc asked for thoughts, and I have provided them. I think he will understand that I am not trying to instigate a re-argument of what the proper interpretation should be. He may understand why I think his essay is not the kind of "Hitchhiker's Guide" you contemplate, and why I think a link would be improper. He may think otherwise, in which case we can agree to disagree. He may even have doubts, in which case I might be able to help resolve some aspects. We will see. ~ J. Johnson (JJ) (talk) 22:45, 18 August 2015 (UTC)

Indeed, I totally see JJ's point here in blurring the lines between guideline and essay. The only I reason I brought it up is because others had suggested (in the earlier discussion) incorporating the essay into the guidelines somehow and I thought this might be a simple way to do so, but I'm happy to let sleeping dogs lie. At least I hope the essay serves as a useful point of reference to show my understanding of the interaction between the listed provisions—an understanding shared by others, but not purporting to be a proper guideline—as it may be a useful re-statement to link in discussions from time to time, whether others use it or not. —sroc 💬 13:31, 20 August 2015 (UTC)

Your interpretation of the existing provisions (quite aside from their goodness or usefulness, or how they might be improved) derives from your understanding of their interaction. It might be useful for future discussion to explicitly identify those interactions, and distinguish them from alternative understandings of same. But that is rather a "meta" discussion, and not, I think, useful for anyone going to MOSNUM for guidance. ~ J. Johnson (JJ) (talk) 21:17, 20 August 2015 (UTC)

We should not be linking to essays for guidance; this will (rightly) be interpreted as inappropriate deference to one editor's opinion. If such a page is seen as needed, it should be vetted as an MoS subpage guideline, i.e. proposed as WP:Manual of Style/Date formatting overview or the like.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  23:56, 20 August 2015 (UTC)

• Current wording "A date range may appear in 2005–2010 format if it is a range of sports seasons in an infobox."
• Proposed change "A date range may appear in 2005–2010 format if it is a range of sports seasons in an infobox."

Rationale in a nutshell: The intent is to establish that compact 8-digit year ranges (XXXX-XXXX) are acceptable for a range of sports seasons. In all other respects, these sports ranges should be subject to the same general usage rules as the compact 6-digit year range (XXXX-XX).

Current use in FAs: Rudolf Caracciola (motor sport), Thierry Henry (assoc football), Karmichael Hunt (rugby union), Otto Graham (American football), Joel Selwood (Austrailian rules football), Grey Cup (Canadian football), Michael Jordan (basketball), Jackie Robinson (baseball), Wayne Gretzky (ice hockey)

Background: My bold edit was reverted with the edit summary of "oh, no, no, no. There's WAY too much history to the current wording. I understand what you're trying to do but this will need to be discussed".

As seen by the above sampling of Feature Articles from various sports, the 8-digit format is already being used in places where a compact format using en dash is more suitable than using words i.e. "from XXXX to XXXX". Aside from infoboxes, this is already used in cases like section headers or tables. For sports year ranges, the 8-digit year range is not in practice used any differently than if the 6-digit format were chosen. This is consistent with

The previous discussion to add the 8-digit format can be found at Wikipedia talk:Manual of Style/Dates and numbers/Archive 144#Date range redux. As a background, the 8-digit format was preferred in for some sports where a single season straddles two years, and is typically expressed with a 6-digit range. While quite a few of the comments were with respect to infoboxes, the spirit of the discussion was allowing 8-digit sports year ranges where the MOS previously allowed only the six-digit format, which was contrary to actual practice. The discussion was closed with consensus to accept the 8-digit form on February 3, 2014. The MOS was not updated until March 7, 2014, when I added it with the overly restrictive limitation for infoboxes.[7]—Bagumba (talk) 07:37, 28 July 2015 (UTC)

I've modified some text which may or may not have caused confusion (to SMcCandlish below?). Removed text has been struck, while inserted text is in red and underlined: "As a background, the 8-digit format was preferred in for some sports where a single season straddles two years ..."—Bagumba (talk) 08:56, 29 July 2015 (UTC)

Doesn't change my position, just reminds me that some people are treating "generally" as if it meant "always".  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  15:39, 29 July 2015 (UTC) Yes, that red correction help.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  00:13, 21 August 2015 (UTC)

Comments

• Support proposal per nominator's rationale. Rikster2 (talk) 19:54, 28 July 2015 (UTC)
• Partial support with clarifications: This should be used in "infobox-like" compact presentation, such as section headings, lists, tables, and navboxes. Limiting this format to exactly infoboxes was a consensus-assessment error. It's also permissible in running prose, but only in reference to specific, individual seasons, as in the 2014–2015 season, or some favor the 2014–15 season. It should not otherwise be used in running prose, when referring to a span of multiple seasons (use the 2008–2009 through 2013–2014 seasons), or just some date range unconnected to seasons, e.g. played for the team, 2010–2014; that's unencyclopedic
  telegraphic writing, a news style as used in headlines (same goes for a "...2010–14" version); instead, use played for the team from 2010 to 2014". This is surely central to why the incautious change to a blanket statement in favor of eight-digit style was reverted. The purpose of the 2014–2015 line-item is to cover a conventional, specific usage that is used as a shorthand in particular constructions, not to play some "sports wikiprojects are exempt from date formatting rules" favoritism that scraps basic rules of good writing. There are a few other things with regard to date-range formatting that should probably be re-examined and may need tweaking, but I'll address that separately in the #Discussion section below.  — SMcCandlish ☺ ☏ ¢
   ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  21:55, 28 July 2015 (UTC) Clarified. 15:32, 29 July 2015 (UTC)

the eight-digit format should never be used to describe a specific season, as in your example. If specifically referring to a sports season (vs. a range), it should be 6 digits (example "the 1993–94 season"). The 8-digit format is only used to express a range (essentially standing in for prose like "Smith played for the Tigers from 1991 to 1996"). Otherwise I agree with your statements. Rikster2 (talk) 02:17, 29 July 2015 (UTC)

Rikster is correct about referring to individual seasons. Per

WP:DATERANGE: "Periods straddling two different years, including sports seasons, are generally written with the range notation (2005–06)."—Bagumba (talk

) 08:39, 29 July 2015 (UTC)

Rikster is not correct, because "generally" does not mean "always" and thus does not translate, in the inverse, into "never". The six-digit format is problematic, because it's easily misread as YYYY-MM in many situations; this is one reason why it has the "generally" caveat, I'm pretty sure (another is that there are sports in which it is not conventional). We shouldn't be using it at all if you ask me, for the same reason we don't do this with page number ranges (it's pp. 239–272 not pp. 239–72). We should not make the situation even worse by swapping out plain English Smith played for the Tigers from 1991 to 1996 for telegraphic gibberish like Smith played for the Tigers 1991–1996. WP is not a newspaper. Actually, even a newspaper wouldn't do that except in a headline. So, I'll support this for compact presentations (infobox, nav, heading, table, list) only, not running prose unless it's in ref. to a single season.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  15:32, 29 July 2015 (UTC)

I agree with your comment on usage in compact presentations as opposed to running prose. However, that applies to the six-digit format as well, and is not unique to eight-digits. Thus, the proposal was made to strike "in an infobox" from the eight-digit note. No prejudice if the MOS is enhanced separately to generally discuss common practice regarding compact form vs prose to cover both six- or eight-digit formats. If you agree that is can be handled separately, please consider changing your !vote from partial to full support. Thanks.—Bagumba (talk) 15:56, 29 July 2015 (UTC)

"generally" does not mean "always": True, MOS is a guideline, and

WP:GUIDES says "Editors should attempt to follow guidelines". So in that sense, there can always be common sense exemptions.—Bagumba (talk

) 16:01, 29 July 2015 (UTC)

I agree with most of what you wrote and support the proposal, however, to me it does not make sense to restrict the yyyy-yyyy range to sport seasons. yyyy-yyyy is the normal form to express a range of years in compact form (that is, when the long form "from year yyyy to year yyyy" is undesireable to be used for some reason). Therefore we should allow it whenever the abbreviation yyyy-yy is allowed at present.

If you ask me, I would also support to go further and deprecate the yyyy-yy form (except for in citations and where a range needs to be expressed in 7 characters for space reasons). It looks like a leftover from the past century to me. As I wrote earlier, there was a lesson to be learned from the Y2K nightmare, and this is to avoid abbreviated years. In fact, I very rarely see this form being used outside the English Wikipedia now, perhaps once or twice a year, whereas I see the 4-digit year form almost on a daily basis. Most people now seem to stick to 4-digit years.

Generally speaking, the purpose of the MOS is to assist editors in chosing a writing style which is non-ambiguous, easily accessible, consistent (where possible), and logical (hopefully). At present, the section on pure year ranges recommends the form yyyy-yy and then lists a long lists of conditions and exceptions when yyyy-yyyy should be used instead because yyyy-yy causes ambiguity. So, why don't we just swap this around, and recommend the easier to parse form yyyy-yyyy as the default compact representation and specifically allow yyyy-yy in those cases where we know that it does not cause confusion? Sounds way more consistent and logical too me.

--Matthiaspaul (talk) 19:00, 29 July 2015 (UTC)

Can I ask that a new thread be started with your proposal to remove "if it is a range of sports seasons" as well? I don't want to sidetrack from this (hopefully) simpler proposal already here. Thanks.—Bagumba (talk) 19:08, 29 July 2015 (UTC)

Sorry, I didn't want to "hijack" your thread. I just thought the proposal would be related enough to be discussed in tandem in order to reduce the general overhead and free everyone's time for more actual article editing. Anyway, you have my support... --Matthiaspaul (talk) 00:28, 31 July 2015 (UTC)

Discussion

I'd rather see us approach this more programmatically, but need to review the exact present wording before getting into the details of what tweaks may be needed. I'll post a followup here later.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  21:55, 28 July 2015 (UTC)

• Follow-up: I agree we're talking about multiple different things, above, and have attempted to outline all of them:
  1. YYYY–YYYY is the normal way, in good writing, to express a date range where compacting (in a table, etc.) a phrase like from YYYY to YYYY is necessary.
  2. This is not exclusive to sports, but across all topics.
  3. YYYY-YY should be avoided everywhere, because it's sloppy, and it produces confusing results like "2008-09" which automatically means "September 2008" to many readers, and is ambiguous to all others. Same goes for YYYY/YY.
  4. In running prose, the YYYY-YYYY season is a normal way to express a single sport season that spans the Dec. 31 / Jan. 1 year boundary.
  5. This is also true of some other things, e.g. television seasons that span a year (the YYYY-YYYY season again), a serial publication issue that spans a year (the winter 2008–2009 issue), and literal seasons that span a year (a population that dropped precipitously in winter 2008–2009 or ...summer 2008–2009 in the southern hemisphere); it is often not the only way to write it, e.g. the winter of 2008–2009. So, again, we have no reason to limit this to sports specifically.
  6. There are some [potential] exceptions to – formatting, contexts in which / is traditionally used and is the majority use in reliable sources on that topic. Some sports conventionally use the 2014/2015 season or even season 2014/2015, given in compact form (tables etc.) as 2014/2015. Whether MOS wants to "honor" that or not is an open question; an argument can be made that it's totally harmless ("an acceptable alternative style which MOS should not prohibit" is how we usually approach such things), while another can be made that it's the WP:Specialized-style fallacy. I lean toward the former, specifically because it would have to directly conflict with normal usage in a way that confused people, in order to be an SSF, and I don't see any evidence of such confusion. Even the winter of 2008/2009 is common in reliable sources, probably more common than the dash version. I can break out the pile of style guides if necessary; I have a suspicion that a slash may be recommended in place of a dash by many of them, for two consecutive years.
  7. In running prose, willy-nilly use of YYYY-YYYY and YYYY-YY are substandard, informal writing, for sports and all other topics: played for the Tigers 2008–2015.
  8. One sports-journalism usage in particular, which seems to be the genesis of this thread, is also substandard writing: the halfassed-compressed form played for the Tigers in the 2008–2015 seasons. It's gibberish. There is no such conceptual entity the 2008–2015 seasons. They are separate things, and what is meant is from the 2009–2010 through 2014–2015 seasons (or from the 2009/2010 through 2014/2015 seasons, depending on the sport's season naming conventions). If it there's no in-context need to mention seasons specifically, this can be compressed to from 2009 to 2015.
  9. Some conceptual date entities do span more than a single year boundary, as in seeking the 2016–2020 US Presidency.
  10. A side point not discussed here yet, this time, but worth reaffirming is that we should never capitalize as Season (in TV, sports, climate, or anything else), per the general
      WP:RMs
      and other previous discussion demonstrate. An obvious exception would be in titles of published works, e.g. a DVD box-set called The X-Files Season 9.

 — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  21:03, 2 August 2015 (UTC)

@SMcCandlish:: Given your general support for XXXX-XXXX across all domains, can I assume you support the premise of allowing XXXX-XXXX for sports when a compact form is needed, and not exclusively for infoboxes (as it is currently worded)? If so, I would propose to incrementally make the minor change originally proposed here, and start a separate discussion on expanding XXXX-XXXX for use beyond sports, which I imagine might need more discussion. Thanks for your consideration.—Bagumba (talk) 22:10, 2 August 2015 (UTC)

I've already given that, above: 'This should be used in "infobox-like" compact presentation, such as section headings, lists, tables, and navboxes. Limiting this format to exactly infoboxes was a consensus-assessment error.'  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  23:14, 2 August 2015 (UTC)

With regard to #3 I would disagree that 6-digit date format should be avoided "everywhere." It is the most common display for a split-year sports season like basketball or hockey ("the 2009–10 season") and should remain the default for that IMO. So I suppose that means I disagree with #4 and if the proposal is to change MOS to reflect this I would like more discussion on that point. I also don't think this proposal was suggesting #8, that in prose a range use the "–" instead of from/to. I believe the proposal has always been about compact usage and don't believe anyone was suggesting the MOS be changed to allow this. Rikster2 (talk) 23:27, 2 August 2015 (UTC)

Update There seems to be consensus to remove the "in an infobox" text as proposed. Moreover, Matthiaspaul and  SMcCandlish have expressed an interest to expand the XXXX-XXXX format beyond sports, which they are encouraged to continue (possibly under a new thread that doesnt end in "... for sports", so as to involve other interested parties in the wider discussion.—Bagumba (talk) 01:38, 3 August 2015 (UTC)

I am against' encouraging YYYY–YY usage anywhere unless it is done to meet width constraints on tables, but even then it should only be contemplated if there is no elegant alternative. Back in the good old days when Wikipedia was edited by real editors and not quiche eaters, editors used to link all years like this 1900-1905 (notice also the use of dash instead of ndash). Now although times have moved on and we have bots to change dashes to ndashes between such dates, I am against the change of using two digit years, because if people want to search the internet for a date, any pages using two digests formats are likely to be missed. As this is not a paper encyclopaedia, we do not have to save characters like they do in paper encyclopaedias by truncating dates or page range (as it will have a detrimental affects on internet searches). The old wording in this guidance was the typical neutral wording of a consensus compromise:

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 (1881–1986). The full closing year is acceptable, but abbreviating it to a single digit (1881–6) or three digits (1881–886) is not.

It was changed by this edit on 24 September 2010 by user:PL290. to

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).

I think that broke a compromise which exists in fact (it certainly did on the pages I edit which retain all four digests in such cases). AFAICT there was little or no discussion of that change in the archives. The no no examples were reinstated by user:EEng with this edit on 12 January 2014 but without the wording "The full closing year is acceptable" which I think was a mistake.

In the case of sporting seasons or school years (like night time bombing raids during World War II: "night of 17/18 May 1940" (eg

-- PBS (talk) 14:36, 3 August 2015 (UTC)

• I strongly concur with "I am against' encouraging YYYY–YY usage anywhere unless it is done to meet width constraints on tables". It's sloppy
  telegraphic writing / headline style. "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" is wrong. It's fairly (probably decreasingly) common in journalism, is avoided in academic writing, and is something we cannot do with any automated tool, or in any other circumstance where what comes after the hyphen might be under 13, because it produces hopeless ambiguity, and will be interpreted by many readers as a month not an abbreviated year. "The full closing year is acceptable" is an understatement; it's preferable.  — SMcCandlish ☺ ☏ ¢
   ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  01:58, 9 August 2015 (UTC)
  • I was just wondering whether anything had been done about this and was delighted to see this discussion underway. Although I agree entirely with the points made above in support of XXXX-XXXX format, perhaps the strongest argument of all is that the current guideline is practically never followed. The MOS should reflect usage not dictate it wherever possible, and the vast majority of editors are clearly using the full 8-digit format. If this needs to be formalised somehow I hope it is, but otherwise surely there is consensus above to make the change? Frickeg (talk) 12:31, 18 August 2015 (UTC)
    • I am somewhat unclear what is being suggested here, but for the record I am against moving away from using the six-digit date format to denote distinct sport seasons that span a calendar year (like 2013–14 Dallas Mavericks season). This is the overwhelming manner in which these are displayed in reliable sources in the U.S. (as well as other countries) and directly discouraging the practice (in context of course) does not seem like something MOS should do. It seems like Wikipedia should reflect reality. Rikster2 (talk) 18:41, 18 August 2015 (UTC)
      • My comment, for the record, has nothing particularly to do with sports seasons, about which I know essentially nothing and am not qualified to have an opinion, really. It was in support of the 8-digit format being either the preferred format of date ranges generally, or at the very least as an equal option with the 6-digit format and preferred in tables. Frickeg (talk) 02:53, 19 August 2015 (UTC)
        • Ok, good, because if people tried changing single-season articles that span two calendar years to an 8 digit format, you'd have a war on your hands. "2015–16", as the upcoming NHL and NBA seasons will be, is a very common format and is completely expected by the reader. For date ranges exclusive of single seasons, I would generally support (and I wrote the Grey Cup article referenced above) making the eight digit format the preferred style, but not a mandatory one. Resolute 00:38, 21 August 2015 (UTC)

At

Series and sequences

When reliable sources are not consistent in their usage of the designation of things in a series or sequence, generally default to using a spelled-out, leading

roman numerals by the published work to which they pertain, consistently across editions/releases (as in Star Wars Episode IV: A New Hope). Events are usually best distinguished by date (e.g., Harlem riot of 1964 or the 1964 Harlem riot), even if they also have formally numbered names (e.g. Super Bowl XLIX), unless the year is also given or is clear in the context. (See also § Regnal numbers

.)

I think that should cover at least most of what needs to be spelled out, and will consolidate it in one place.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  09:51, 20 August 2015 (UTC)

SMcCandlish, This generally looks good, but I think the recommendation to spell out ordinals should be modified. the Tenth International Conference on Climate Change is fine, but what about the (fictional) presented at the 205th Conference on Species Names. Ordinals over some limit (many use ten, some 99) should not IMO be spelled out, and theis should be mentioend here, I think. DES ^(talk) 18:40, 21 August 2015 (UTC)

Ah yes. We already have a rule about not spelling about bigger numbers, so we should cross-reference it.  — SMcCandlish ☺ ☏ ¢ ≽^ʌⱷ҅_ᴥⱷ^ʌ≼  23:22, 21 August 2015 (UTC)